Developing Country Debt and Economic Performance, Volume 3: Country Studies--Indonesia, Korea, Philippines, Turkey (National Bureau of Economic Research Project Report)

This Page Intentionally Left Blank

Developing Country Debt and Economic Performance

A National Bureau of Economic Research Project Report

Developing Countrv Debt and Economic Performance Volume

3

Edited by

Country StudiesIndonesia, Korea, Philippines, Turkey

Jeffrey D. Sachs and Susan M. Collins

The University of Chicago Press

Chicago and London

JEFFREY D. SACHSis the Galen L. Stone Professor of International Trade at Harvard University and a research associate of the National Bureau of Economic Research. SUSAN M. COLLINS is an associate professor of economics at Harvard University and a faculty research fellow of the National Bureau of Economic Research.

The University of Chicago Press, Chicago 60637 The University of Chicago Press, Ltd., London 0 1989 by the National Bureau of Economic Research All rights reserved. Published 1989 Printed in the United States of America 98 97 96 95 94 93 92 91 90 89 5 4 3 2

1

@The paper used in this publication meets the minimum requirements of the American National Standard for Information Sciences--Permanence of Paper for Printed Library Materials, ANSI 239.48-1984.

Library of Congress Cataloging-in-PublicationData (Revised for Volume 3) Developing country debt and economic performance. (A National Bureau of Economic Research project report) Papers presented at a conference held in Washington, D.C. on Sept. 21-23, 1987. Includes bibliographies and indexes. Contents: v. 1. The international financial system-v.3. Country studies-Indonesia, Korea, Philippines, Turkey. 1. Debts, External-Developing countries-Congresses. 2. Developing countries-Economic conditions-Congresses. 3. International financeCongresses. I. Sachs, Jeffrey. II. Series. HJ8899.D48 15 1989 336.3'435'091724 88-20866 ISBN 0-226-73332-7(v. 1 : alk. paper) 0-226-73335-1(v. 3 : alk. paper)

National Bureau of Economic Research Officers Richard N. Rosett, chairman George T. Conklin, Jr., vice-chairman Martin Feldstein, president and chief executive ojlicer

Geoffrey Carliner, executive director Charles A. Walworth, treasurer Sam Parker, director of finance and administration

Directors at Large John H. Biggs Andrew Brimmer Carl F. Christ George T. Conklin, Jr. Kathleen B. Cooper lean A. Crockett George C. Eads Morton Ehrlich

Martin Feldstein George Hatsopoulos Franklin A. Lindsay Paul W. McCracken Leo Melamed Geoffrey H. Moore Michael H. Moskow James J. O’Leary

Robert T. Parry Peter G. Peterson Robert V. Roosa Richard N. Rosett Bert Seidman Eli Shapiro Harold Shapiro Donald S. Wasserman

Directors by University Appointment Charles H. Berry, Princeton Bill Brainard, Yale James Duesenberry, Harvard Ann F . Friedlaender, Massachusetts Institute of Technology Jonathan Hughes, Northwestern Saul Hymans, Michigan Marjorie McElroy, Duke

James L. Pierce, California, Berkeley Andrew Postlewaite, Pennsylvania Nathan Rosenberg, Stanford James Simler, Minnesota William S. Vickrey, Columbia Burton A. Weisbrod, Wisconsin Arnold Zellner, Chicago

Directors by Appointment of Other Organizations Richard Easterlin, Economic History Association Bruce Gardner, American Agricultural Economics Association Robert S . Hamada, American Finance Association Robert C. Holland, Committee for Economic Development David Kendrick, American Economic Association Eric Kruger, The Conference Board

Ben Laden, National Association of Business Economists Rudolph A. Oswald, American Federation of Labor and Congress of Industrial Organizations Douglas D. Purvis, Canadian Economics Association Dudley Wallace, American Statistical Association Charles A. Walworth, American Institute of Certified Public Accountants

Directors Emeriti Moses Abrarnovitz Emilio G. Collado Solomon Fabricant

Frank W. Fetter Thomas D. Flynn Gottfried Haberler

George B. Roberts Willard L. Thorp

Relation of the Directors to the Work and Publications of the National Bureau of Economic Research I . The object of the National Bureau of Economic Research is to ascertain and to present to the public important economic facts and their interpretation in a scientific and impartial manner. The Board of Directors is charged with the responsibility of ensuring that the work of the National Bureau is carried on in strict conformity with this object. 2. The President of the National Bureau shall submit to the Board of Directors, or to its Executive Committee, for their formal adoption all specific proposals for research to be instituted. 3. No research report shall be published by the National Bureau until the President has sent each member of the Board a notice that a manuscript is recommended for publication and that in the President’s opinion it is suitable for publication in accordance with the principles of the National Bureau. Such notification will include an abstract or summary of the manuscript’s content and a response form for use by those Directors who desire a copy of the manuscript for review. Each manuscript shall contain a summary drawing attention to the nature and treatment of the problem studied, the character of the data and their utilization in the report, and the main conclusions reached. 4. For each manuscript so submitted, a special committee of the Directors (including Directors Emeriti) shall be appointed by majority agreement of the President and Vice Presidents (or by the Executive Committee in case of inability to decide on the part of the President and Vice Presidents), consisting of three Directors selected as nearly as may be one from each general division of the Board. The names of the special manuscript committee shall be stated to each Director when notice of the proposed publication is submitted to him. It shall be the duty of each member of the special manuscript committee to read the manuscript. If each member of the manuscript committee signifies his approval within thirty days of the transmittal of the manuscript, the report may be published. If at the end of that period any member of the manuscript committee withholds his approval, the President shall then notify each member of the Board, requesting approval or disapproval of publication, and thirty days additional shall be granted for this purpose. The manuscript shall then not be published unless at least a majority of the entire Board who shall have voted on the proposal within the time fixed for the receipt of votes shall have approved. 5. No manuscript may be published, though approved by each member of the special manuscript committee, until forty-five days have elapsed from the transmittal of the report in manuscript form. The interval is allowed for the receipt of any memorandum of dissent or reservation, together with a brief statement of his reasons, that any member may wish to express; and such memorandum of dissent or reservation shall be published with the manuscript if he so desires. Publication does not, however, imply that each member of the Board has read the manuscript, or that either members of the Board in general or the special committee have passed on its validity in every detail. 6. Publications of the National Bureau issued for informational purposes concerning the work of the Bureau and its staff, or issued to inform the public of activities of Bureau staff, and volumes issued as a result of various conferences involving the National Bureau shall contain a specific disclaimer noting that such publication has not passed through the normal review procedures required in this resolution. The Executive Committee of the Board is charged with review of all such publications from time to time to ensure that they do not take on the character of formal research reports of the National Bureau, requiring formal Board approval. 7. Unless otherwise determined by the Board or exempted by the terms of paragraph 6, a copy of this resolution shall be printed in each National Bureau publication. (Resolution adopted October 25, 1926. as revised through September 30. 1974)

Contents

Preface

X

Debt, Policy, and Performance: An Introduction Susan M. Collins

1

Book I

Indonesian Economic Policies and Their Relation to External Debt Management Wing Thye Woo and Anwar Nasution

17

1. Introduction and Summary

19

2. Political and Economic Instability, 1950-65 3. The Political Economy Factors in Policymaking 4. The Fiscal System

32

5. Monetary Policy and Financial Structure 6. Exchange Rate Policy 7. External Debt Management 8. Conclusions and Prospects Notes References

Book I1

External Debt and Macroeconomic Performance in South Korea Susan M . Collins and Won-Am Park

56 67

83 96 114 130

140 146

151

1. Introduction

153

2. Historical Background: Economic Development Prior to 1962

159

vii

viii

Contents

3. An Overview of Korea's External Debt

170

4. Three Cycles of Debt Accumulation, 1960-86

182

5. Internal versus External Shocks

206

6. Introduction to Part Two 7. Korea's Rapid Growth

219

8. Savings and Investment

234

9. Exchange Rate, Trade, and Industrial Policy 10. Exchange Rates, Wages, and Productivity 11. Fiscal and Monetary Policy 12. Income Distribution 13. Lesson's from Korea's Experience: A Synthesis Data Appendix Notes References

Book I11

The Marcos Legacy: Economic Policy and Foreign Debt in the Philippines Robert S. Dohner and Ponciano Intal, Jr.

218

249 270 282 300 3 19 330 358 362

37 1

1. Introduction

373

2. Government Expenditure and Revenues

40 1

3. Trade Policy, Industrial Policy, and the Exchange Rate

433

4. Government Interventions and Rent Seeking

460

5. The Philippine Financial System and the Debt Crisis

48 1

6. External Debt and Debt Management

503

7. Debt Crisis and Adjustment

524

8. The Aquino Government and Prospects for the Economy Notes References

558 607

Book IV

615

Debt, Adjustment, and Growth: "hrkey Merih CelGsun and Dani Rodrik

593

PARTI POLICY PHASES AND ADJUSTMENT PATTERNS 1. Turkish Economic Development: An Overview 2. Economic Boom and Debt Crisis. 1973-77

617 629

ix

Contents

3. Crisis Without Adjustment, 1978-79 4. Stabilization and Adjustment Policies, 1980-85 5. Performance and Adjustment Patterns in the 1980s

655 662 679

PARTI1 SELECTED ASPECTSOF DEBTAND ADJUSTMENT 6. External Borrowing, Real Wage Flexibility, and Equilibrium Exchange Rates: A General Equilibrium Analysis 7. Trade Regime and an Anatomy of Export Performance 8. The Public Sector: Fiscal Adjustment and Resource Mobilization 9. External Financial Relations and Debt Management 10. Conclusions and Prospects Appendix A. Political Chronology Appendix B. Statistical Appendix Notes References

702 716 73 1 750 76 1 768 77 1 797 804

Biographies

809

List of Contributors

810

Name Index

81 1

Subject Index

815

Preface

This volume contains four country studies that were prepared as part of a research project by the National Bureau of Economic Research on developing country debt. Studies of Argentina, Bolivia, Brazil, and Mexico have been collected in a separate volume. In addition to the eight country studies, this project includes eight papers that examine other debt crises that occurred before World War 11, the role of the banks during the current crisis, the effect of developed country economies on the debtors, as well as possible solutions to the debt crisis. These papers have also been published in a separate volume. A fourth book contains shorter and less technical summaries of all sixteen papers. The findings of NBER’s debt project were presented at a conference for government officials of lending and debtor countries, economists at international organizations, and representatives of banks and other private firms with interests in the debtor countries. The conference was held in Washington, D.C. from 21-23 September 1987. We would like to thank the Agency for International Development, the Ford Foundation, Mr. David Rockefeller, the Rockefeller Brothers Fund, the Starr Foundation, and the Tinker Foundation for financial support of this work. The success of the project also depended on the efforts of Deborah Mankiw, Yasuko MacDougall, Kirsten Foss Davis, Ilana Hardesty, Robert Allison, and Mark Fitz-Patrick. Jeffrey D . Sachs and Susan M . Collins

X

Debt, Policy, and Performance: An Introduction Susan M. Collins

For the developing countries, the 1970s were a time of growing external indebtedness but strong real growth. Borrowing seemed to be part of a sensible strategy of growth and development. In the 1980s, however, the role of foreign borrowing was much less innocuous. After 1982 the majority of heavily indebted countries found themselves in the midst of a debt crisis which was more severe and more persistent than most observers had predicted. Growth rates were stagnant and often negative. In many cases, per capita incomes in 1987 were below their 1980 levels. Fixed capital formation as a share of income declined precipitously, diminishing prospects for future growth. More troubling, the indicators of debt burden rose in a large number of countries. The debt crisis remains a long way from resolution. At the same time, there have been substantial differences in the experiences of the heavily indebted countries. Some countries announced debt moratoria while others avoided a crisis and countinued to repay their debts on, or ahead of, schedule. Some maintained relatively high growth rates and financial stability during the early and mid-l980s, while others wrestled with exploding inflation. These diverse experiences raise interesting and important questions about the roles of foreign borrowing and macroeconomic policy for small, open economies in an uncertain world environment. How did those countries which navigated the series of external shocks more successfully differ from those which are still struggling to “adjust”? Did they simply borrow more prudently? How are the differences in performance attributable to the The author would like to thank Geoffrey Carliner, Dani Rodrik, and Eliana Cardoso for comments and Donna Zenvitz for editorial suggestions. Of course, the author retains responsibility for the views expressed.

1

2

SusanM.Collins

severity of external shocks as opposed to current and previous policy actions? How did external debt interact with political, social, and economic structure? The NBER country studies (volumes 2 and 3) examine in detail the policy and performance of eight heavily indebted developing countries: Argentina, Bolivia, Brazil, Korea, Indonesia, Mexico, the Philippines, and Turkey. The countries differ in many ways. Indonesia and Bolivia are low-income countries while the rest are middle income. Mexico and Indonesia are oil exporters, Bolivia exports natural gas and some oil, while the other countries are oil importers. Korea, Indonesia, and Turkey had relatively strong macroeconomic performance in the early to mid- 1980s, although foreign borrowing played a central role in their development. Brazil did relatively well during 1982-84, but emerged as a “problem debtor” by 1986. In Argentina, Bolivia, Brazil, and Mexico, price instability interacted with external debt to exacerbate the difficulties of stabilization and structural adjustment. In those countries and in the Philippines, improved external balance came at the expense of domestic investment and growth prospects. Thus, the allocation of domestic resources among investment, consumption, and net transfers abroad to service the debt remains a critical issue. Each study takes a broad perspective, examining the role of debt within the context of macroeconomic policy and performance. While there is no single model or approach, there are common elements throughout. In particular, the focus is on the debtor country’s perspective. Also, each study contains extensive data tables in the text and appendices. Finally, the authors have attempted to integrate social and political factors into their analyses and to emphasize the importance of historical context. The most striking aspect of these country studies is the similarities in their conclusions. It appears that countries which performed well in the 1980s differed from the other countries in three fundamental ways. First, they maintained stable and competitive real exchange rates. Second, they were successful in having a disciplined fiscal policy, containing budget deficits and maintaining a broad tax base. Third, they emphasized investment with incentives for capital accumulation in export sectors. While microeconomic policies, the trade regime, and the severity of external shocks all played a role (especially in the Philippines), these three macroeconomic policies stand out in explaining the range of performances. The next section provides a brief overview of the experience of each country. The third section turns to a cross-country comparison of the role of foreign borrowing and macroeconomic policy in the diverse performances.

External Debt and Macroeconomic Performance Three of the countries, Indonesia, Korea, and Turkey, maintained moderate to high real growth in the early to mid-l980s, with foreign borrowing playing a central role in each experience. In Indonesia, balance of

3

Introduction

payments difficulties erupted in 1966. They resulted in a debt rescheduling and a coordinated, long-term plan of official assistance. Turkey’s external crisis came in 1977, after the first oil price shock but well before repayment difficulties emerged in the other heavily indebted countries. Finally, Korea’s economic crisis began in late 1979 as a result of external shocks and internal political and economic factors. The crisis was relatively short, as performance had improved considerably by 1981. While Korea did not reschedule its debts, it did undertake a substantial shift in economic policies. The Philippines presents a stark contrast. Economic performance deteriorated sharply during the 1980s, and the economy remains far from a path of stable growth. In many respects, the Philippine experience was similar to Mexico’s. Both borrowed to finance capital flight and public investments which did not pay off in long-term growth or foreign exchange earnings. However, Mexico exports oil while the Philippines is an oil importer. In Argentina and Brazil the main economic problems revealed themselves in the form of inflation rather than as pressure on the external balance. However, debt played a central role both in fueling the inflation and in compounding the difficulties of debt reduction. South Korea South Korea is widely heralded as an economic success story. In 1987 real growth was 12 percent, inflation was just 3 percent, and the trade surplus had risen to nearly 9 percent of GNP. However, as Collins and Park emphasize, Korea was not always a high growth surplus country. On the contrary, it faced a severe crisis in 1980. Real output declined by 5 percent, inflation rose to nearly 30 percent, and the trade deficit mushroomed to 9 percent of GNP. Korea borrowed heavily to finance these deficits and by 1981 was the fourth largest debtor country in the world, behind Argentina, Brazil, and Mexico. How then was Korea able to combine external adjustment with real growth and price stability? A poor, war-devastated economy heavily dependent on foreign aid in the 1950s, Korea embarked on a new economic strategy of active export promotion and emerged as a newly industrialized country with growth rates averaging over 9 percent per year in the mid-1960s and 1970s. High rates of investment in exportables and competitive exchange rates, with credible and consistent incentives to exporters, are the key factors behind Korea’s growth. Fiscal policy was used countercyclically, but budget deficits were not allowed to become large. Initially, domestic saving rates were very low, and high rates of investment were financed by extensive foreign borrowing. A remarkable aspect of Korea’s development has been a trend rise in aggregate domestic savings from less than 6 percent of GNP in the early 1960s to over 30 percent by the mid-1980s. Korea’s economic strategy shifted during the early 1970s. Concerned about industrial deepening and building up military strength, policymakers launched the Big Push to develop heavy and chemical industries in 1973.

4

Susan M. Collins

Investment was to be increased and concentrated in these sectors. However, the period coincided with the first oil price shock and slowdown in world growth. Korea decided to pursue the Big Push nonetheless and borrowed heavily to finance the investments. While fiscal policy was expanded to stimulate growth, the exchange rate was devalued and then fixed, and heavy taxes were imposed on petroleum products. World demand recovered during 1976-78, improving Korean export performance, and high growth rates resumed. Savings rose and the current account deficit was eliminated. However, a growing number of economic dislocations became apparent. The real exchange rate began to appreciate, export growth slowed, the current account deteriorated, and inflation surged. The government began to intervene more heavily in domestic markets, controlling prices, restricting imports, and rationing credit. As the other country studies make clear, this is the same pattern that emerged in countries which ran into difficulty. Korea differs in that these distortions were addressed before they became extreme. Policymakers became increasingly concerned about inflation and economic distortions. A new stabilization plan, announced in 1979, called for monetary and fiscal restraint, gradual reductions in price controls, and trade and credit market liberalizations. This policy shift was initiated before domestic authorities were forced to adjust since the oil shock had not yet hit, and foreign lending was still available. Interestingly, the initial program did not call for devaluation despite the fixed exchange rate and resulting appreciation. The exchange rate adjustment came in January 1980 in the midst of the crisis. Still, Korea’s overvaluation was corrected after the real exchange rate had appreciated by 25 percent. In contrast, Argentina experienced a real appreciation of over 80 percent between 1977 and 1980, before an adjustment was undertaken. Despite the severe shocks which hit Korea in 1979 and 1980, a full-scale debt crisis was avoided. However, the assassination of President Park, disastrous agricultural harvests, the rise in oil and commodity prices, higher interest rates, and the slowdown in world demand all contributed to very poor economic performance in 1980. Collins and Park point to four key elements which explain the impressive economic turnaround between 1980 and 1985. First, Korea was given breathing space by foreign creditors for the stabilization and structural adjustment. The devaluation did not generate an immediate export recovery. But fiscal policy was expansionary in 1980-81 to stimulate growth. Imports were contained, not through recession but because of the recovery in agricultural output which reduced food imports and boosted growth. Korea continued to borrow in the early 1980s to finance the (shrinking) current account deficits. The economy did not undertake a structural readjustment at the same time that austerity measures were used to improve external balance. Only in 1982, as output and export growth improved, did a fiscal contraction take place.

5

Introduction

Second, Korean policy was stable and consistent. The real exchange rate varied less than in most developing countries. Also, budget deficits were kept relatively small, averaging 2.7 percent of GNP during 1973-86 and ranging from 1.0 percent to 4.7 percent. Third, Korea persistently maintained high rates of investment throughout the adjustment (29 to 33 percent of GNP). These investments placed Korea in a prime position to take advantage of the revival in world demand. The final element is the dramatic rise in domestic savings. While Collins and Park show that a large part of the increase is associated with Korea’s rapid growth, much of Korean saving behavior remains unexplained. Turkey Turkey is unusual in that it experienced a debt crisis in 1977, before the second oil shock which caused most of the other indebted countries to have repayment difficulties. At the time, it was the most severe debt crisis of the postwar period. Turkey rescheduled its debts and undertook a comprehensive stabilization and liberalization program. By 1982, as the debt crisis was just erupting in most countries, Turkey had reestablished creditworthiness. In explaining the timing of events, Cellsun and Rodrik argue that it was not the 1973 oil shock and the policy response that differentiate Turkey. Instead, their analysis highlights the role of convertible Turkish lira deposits (CTLDs), a scheme for mobilizing short-term foreign loans. The Turkish episode had four phases. In 1963 Turkey launched the first in a series of five-year development plans. The plan embodied an import substitution strategy with emphasis on domestic, especially public sector, savings. The plans were successful in achieving moderate growth and in containing inflation. However, trade restrictions and increasing overvaluation discouraged exports. By the end of the 1960s, growth began to slow and attempts at reorientation, including devaluation and some trade liberalization, were not followed through. Turkey’s debt crisis erupted at the end of the second phase (1972-77). It began with a surge in worker remittances from abroad in 1972-73, which stimulated growth, generated a current account surplus, and removed the impetus for policy refocus. This reluctance to undertake macroeconomic adjustments distinguishes Turkish policy from policies in Korea and Indonesia in the late 1970s. With its large foreign exchange inflows leading to a debt crisis largely through macroeconomic mismanagement, Turkey’s experience parallels on a smaller scale what was to happen later in Mexico. Turkey embarked on an ambitious industrialization plan, concentrating on capital-intensive industries. As in Korea, the period saw rapid debt accumulation, growing price distortions, budget deficits, and exchange rate appreciation. Exports stagnated, and the current account deteriorated, especially during 1975-77 when public spending surged, financed through domestic credit expansion, financed in turn by foreign borrowing. Cellsun

6

Susan M. Collins

and Rodrik argue that the dynamics of the CTLDs, which encouraged short-term private loans from abroad by protecting domestic borrowers from exchange risk, allowed the surge in spending to take place. The loans were converted to domestic currency, expanding the domestic money supply and providing credit to the public sector. As overvaluation increased, the current account deteriorated, foreign exchange reserves fell, investors lost confidence, and the scheme collapsed. During 1978-79, Turkey rescheduled some of its debt and initiated a series of stabilization programs. However, fiscal and exchange rate adjustments were too little, too late. The external balance improved through import compression, but investment and growth declined while inflation accelerated. This period of adjustment parallels the experiences of many other debtor countries. The case studies show no examples of programs that achieved stabilization with growth when the trade balance improvement came from import compression. From 1980 to 1982, Turkey launched a comprehensive stabilization plan. The initial policies were orthodox, export-oriented shock treatments, including devaluation and fiscal retrenchment. Real wages fell sharply, as did private spending, but Turkey managed to combine external adjustment with real growth. The key was surprisingly strong export performance. While real depreciation was important, Cellsun and Rodrik attribute the largest portion to “special factors,” including the strong demand in the Middle East associated with the Iran-Iraq war and hefty export subsidies. They also stress the importance of debt relief, together with substantial new lending from the IMF, the World Bank, and the OECD which reduced the need for import compression. These external factors clearly eased the difficult early stages of Turkey’s adjustment. Indonesia Indonesia’s debt crisis occurred in the mid-l960s, not during the 1970s and 1980s. Woo and Nasution emphasize Indonesia’s earlier political and economic crisis in explaining why macroeconomic performance was relatively strong in the 1980s. Prudent macroeconomic management, in particular fiscal and exchange rate policies, also explain why foreign exchange inflows from oil revenues did not lead to a debt crisis as they did in Mexico and Turkey. When General Soeharto took office in October 1965 following a military coup, Indonesia was in the midst of civil war and economic turmoil. There were growing budget deficits, external debt and inflation, overvalued multiple exchange rates, stagnant economic growth, and a shrinking export sector. One of the government’s first tasks was to stabilize the economy. A generous long-term plan of official assistance, including direct assistance and favorable terms for rescheduling existing debts, was coordinated by Western governments. This assistance enabled the government to raise investment,

7

Introduction

take advantage of a strong resource sector, and stimulate growth. In many respects, Indonesia’s growth strategy in the late 1960s paralleled Korea’s strategy. The government adopted a pro-export orientation and a balanced budget requirement. (Budget deficits could not be financed by domestic credit creation although they could be financed by foreign borrowing.) Woo and Nasution argue that the willingness to use active exchange rate policy to maintain competitiveness, along with incentives for agriculture and light manufacturing, arose from the desire to retain political support in rural areas. In any case, 1966 to 1971 was a period of strong growth in output and exports. The next phase in Indonesia’s macroeconomic history came with the fixed exchange rate during 1971-78. Strong export performance and then growing oil revenues led to a sizable current account surplus. By 1974 Indonesia had regained access to international financial markets. However, the foreign exchange inflows did not trigger aggressive debt accumulation leading to a debt crisis as occurred in Mexico and Turkey. Although the government increased investments and other expenditures and did not cut back on external borrowing as oil revenues rose, debt declined as a share of exports and of GNP. The government’s relatively cautious debt strategy was reinforced by the default of the state-owned oil company in 1975, which had borrowed extensively and was unable to roll over its large short-term debts. As a result, state-owned enterprises were denied direct access to international credit markets. In 1978 Indonesia devalued by 50 percent, marking the return to an active exchange rate management. The striking feature of this devaluation is that it was not triggered by a balance of payments crisis. Economic growth and the external balance were strong. Foreign exchange reserves were abundant. However, the real exchange rate had appreciated, hurting the competitiveness of nonoil exports. Nonoil exports responded strongly to the devaluation. As export receipts grew in 1979-81, public investment was increased further. Indonesia did run into some difficulties in 1982 and 1983 as oil prices slumped and world demand stagnated. Policymakers turned increasingly to quantitative import restrictions, exacerbating microeconomic distortions. Still, macroeconomic policies were the key to explaining Indonesia’s avoidance of a debt crisis. Strong export performance aided by active exchange rate adjustment, prudent debt management with relatively little short-term borrowing, and the favorable terms of their existing debt from the earlier rescheduling were the main elements of Indonesia’s success. The Philippines Philippine macroeconomic performance stands in stark contrast to the performance in the three other non-Latin American countries in the NBER group, and to expected performance in the late 1970s when the Philippines

8

Susan M. Collins

was often grouped with the Asian tigers. In the 1980s, the Philippines became the only Asian country to declare a debt moratorium. Real growth, which had averaged 4.6 percent during 1980-81, averaged -5.2 percent during 1984- 85. In some respects, macroeconomic policies in the Philippines during the late 1970s were not dramatically different from policies in Korea or Indonesia. The real exchange rate appreciated at times during 1970-82, but by at most 15 percent. Korea’s real appreciation of the late 1970s was over 25 percent. Similarly, the Philippine budget deficit rose from 2 percent of GNP in 1978 to 6 percent in 1981. However, Korea’s budget deficit rose from 1 percent of GNP in 1979 to 5 percent in 1981. Both the real appreciation and the budget deficits were small when compared to 80 percent real appreciations and 18 percent budget deficits in Argentina. Why then did Philippine macroeconomic performance deteriorate by SO much? Dohner and Intal point to two factors. First, the magnitude of the external shocks was more severe for the Philippines than for most other developing countries. Unlike Korea, for example, the Philippines suffered a secular terms of trade deterioration. Some structural adjustments were required if the country were to maintain growth rates and a sustainable external balance, even without the two oil price shocks. Second, the political-economic environment under President Marcos was of critical importance. Individual favoritism in loan allocation and misallocation of other resources discouraged private investment, particularly in exportables, and encouraged capital flight. Government intervention became considerably more extensive than in Korea or Indonesia during the 1970s, exacerbating the difficulties of an overvalued exchange rate and growing budget deficits. The Philippine experience had five phases. In the first, from 1966 to 1969, Marcos borrowed heavily to finance a domestic expansion. The result was growing budget and current account deficits, leading to a balance of payments crisis. The external debt was rescheduled in 1970. Like Soeharto, Marcos had early experience with the potential pitfalls of foreign borrowing, but his caution lasted only until the late 1970s. The second period, from 1970 to 1972, included an economic stabilization plan supported by the IMF. Devaluation, together with tight monetary and fiscal policies, resulted in a sharp reduction in real wages and an increasingly violent political situation. However, strong export performance helped to revive economic growth. From 1972 to 1979, Philippine economic performance looked strong. Real growth rose to 6.2 percent per year, nontraditional exports increased, and investment boomed. However, these statistics mask underlying economic difficulties which help to explain the reversal after the second oil shock. In particular, total exports failed to grow as a share of GNP as traditional exports declined. Investments undertaken produced very poor returns and

9

Introduction

labor productivity actually declined. Much of the explanation for the system’s fragility comes from delayed exchange rate adjustments, disincentives for traditional export sectors, and the uncertainties and misallocations inherent in the government’s growing “crony capitalism” which favored certain industries and individuals for political reasons. The situation deteriorated during 1979-82. The Philippines was hit hard by the second oil price shock. However, unlike Korea and Indonesia which had undertaken stabilization measures even before the shock, the Philippines reacted with expansionary policies to counteract the contractionary effects. External borrowing soared to finance public investment projects, some of which never materialized. Increasingly, borrowing was short term. The value of exports fell sharply, budget and current account deficits continued to grow, and capital flight exacerbated the balance of payments difficulties. In October 1983, when reserves were nearly depleted, the Philippines declared a debt moratorium. During 1983-85, harsh austerity measures were undertaken. Inflation was reduced and the current account deficit nearly eliminated, but the output costs of stabilization were very large. Per capita income levels plunged to as low as those in the mid-l970s, and investment fell by more than half. The severe economic situation contributed to the defeat of Marcos in 1986. The new government of Corazon Aquino continues to struggle to revive growth. Many of the remaining problems are microeconomic. However, the case of the Philippines also points to the difficulties in achieving sustained growth in an economy where domestic residents have grown wary of investing their resources at home and where investment has been slashed to improve the trade balance. In contrast, Korea’s ability to reverse negative growth rates quickly is caused in large part by its history of consistent and credible policies and by the persistently high rates of fixed capital formation, even in the midst of the 1980 crisis. Mexico Mexico’s recent economic performance raises the question why a windfall in oil revenues should result in stagflation and a debt crisis. Buffie cites the “sustained bout of extraordinary fiscal indiscipline” as the major factor, and argues that the subsequent stabilization effort, which has contributed to the collapse of investment, is ill conceived. Again, microeconomic distortions are part of the story, but the keys are poor exchange rate and fiscal policies. Buffie’s analysis contrasts Mexico’s post- 1972 performance with the high growth, low inflation period (1958-72). In the earlier period, fiscal expansions were short lived and macro policy was managed with an eye to maintaining a stable exchange rate. Two cycles of expansion were followed by macroeconomic crises. The first was generated by rapid fiscal expansion, financed in large part by central bank credit. Initially the expansion

10

Susan M. Collins

stimulated growth, but then inflation accelerated, the real exchange rate appreciated, the current account worsened, and capital flight erupted, financed by rapid debt accumulation. The result was a balance of payments crisis in 1976. A large devaluation and an IMF program of monetary and fiscal austerity followed, but the down side of the cycle appeared in 1977 as inflation surged while employment declined. Recognition of Mexico’s oil wealth, together with easily available external credits, was used to launch an even larger fiscal expansion in 1978-81. As with the inflow from worker remittances in Turkey, the windfall was used to avoid, not to facilitate, a deeper structural adjustment. Again, the expansion generated strong growth in output and employment and high investment, but was unsustainable. Buffie points to three major flaws in the policy. First, public sector revenues, net of transfers of nonparastatals, declined despite the large oil revenues. The tax base was small and shrinking. Second, despite the dollar earnings from petroleum exports, the current account deficit deteriorated. Trade liberalization along with real appreciation generated an import boom and stagnant nonoil exports. At the same time, expected devaluation and lack of confidence resulted in large outflows of private capital. External debt was rapidly accumulated. A third factor in the program’s unsustainability was the growing concentration in short-term debt. Unfortunately, the debt did not primarily finance investments which could generate foreign exchange earnings and sustain growth. Much of the increase in public outlays went to current expenditure, while many of the investments in state-owned enterprises later proved unsound. As Mexico became unable to service its debts in 1982 and additional foreign loans dried up, the boom again turned to stagflation. Although the fiscal expansion continued and inflation soared, large real devaluations and trade restrictions cut imports, including intermediate inputs. Four years later, in 1986, Mexico’s real output was below its level at the beginning of adjustment. In 1982-84 a severe austerity program was put into place. The current account improved, and there was some reduction in the budget deficit. However, inflation remained high, real GDP fell substantially, and private investment collapsed. The investment decline in Mexico, as in Argentina, was caused by sharp reductions in available credit, together with high prices of capital goods and imported intermediates. There was no progress in reducing the budget deficit, as tax revenues remain stagnant and little effort was made to broaden the tax base. The deficits were financed largely by government bond sales, raising real interest rates, and contracting credit available to the private sector. The collapse of oil prices on the world market in 1986 brought renewed austerity. Again, output declined while inflation jumped. The severe terms of

11

Introduction

trade deterioration compounded the difficulties of Mexico’s stabilization and structural adjustment. Argentina Dornbusch and de Pablo study the Argentine debt crisis from the perspective of the country’s long history of macroeconomic instability and political uncertainty coupled with the particular dynamics of debt, deficits, and inflation. They conclude that policy and economic structure eventually would have led to a debt crisis; the external shocks of the 1970s and 1980s merely brought the crisis on sooner. Again, both fiscal and exchange rate policies were central. A massive overvaluation in the late 1970s created a large external debt. Despite aggressive exchange rate management, the difficulties were compounded in the 1980s as the budget deficit mushroomed out of control. The late 1970s were a period of relative macroeconomic stability for Argentina. Finance Minister Martinez de Hoz had brought inflation down from 2,000 percent in 1975 to 100-200 percent in 1976. Similarly, the budget deficit had been reduced from 15 to 7 percent of GDP. In an attempt to reduce inflation further, Argentina introduced a new exchange rate regime in 1979. The exchange rate depreciation was to be preannounced, and capital markets were opened fully in the hope that international competition would discipline domestic price setters. While inflation did fall, the cumulative real appreciation exceeded 50 percent by the end of 1980. In anticipation of a maxidevaluation, there were massive capital outflows and accumulation of external debt. A series of large devaluations between 1981 and 1982 restored the real exchange rate to its 1976 level. However, the Argentine story differs strikingly from that of Indonesia, Korea, and Turkey, where large devaluations set the stage for rapid growth in exports and GNP, easing the burden of external debt repayment. Instead, Argentina’s economic performance deteriorated substantially during 1981-83. Despite the large exchange rate adjustments, external debt accumulation accelerated, real growth turned negative, and inflation soared. There are three main reasons why the devaluations did not generate export-led growth as had occurred in Turkey and Korea. First, Argentine exports did not respond quickly to the real depreciation as they had in Turkey because of sluggish world demand and a more fragmented export sector. Second, and more importantly, Argentina was not given the “breathing space” of continued capital inflows in the first years of adjustment which had eased the adjustments in Indonesia, Korea, and Turkey. Argentina was forced to reverse its current account quickly, from - 6 percent of GNP in 1980 to 3.6 percent of GNP during 1982-83. The

12

Susan M. Collins

improved external balance was achieved by cutting imports in half and through reduced investment and a domestic recession. Third, Argentina’s budget deficit more than doubled during 1981 -83, while Korea, Turkey, and Indonesia all undertook fiscal corrections. Reasons for the increase included the government’s unwise exchange rate guarantees, which deteriorated the budget as additional devaluations were required, and the military conflict in the Falklands. Dornbusch and de Pablo emphasize the interaction between the inflation explosion and the government budget deficit, which averaged 18 percent of GDP during 1981-83. Rising real interest rates and exchange rate depreciation increased the government’s debt service obligations and thereby the budget deficit, leading to additional money creation and fueling inflation. The process was exacerbated as institutional changes provided additional interest-bearing substitutes for money. As Alfonsin took office in 1984, large wage increases put further pressure on the budget and inflation accelerated. By the end of 1984, Argentina seemed to be headed toward hyperinflation. The Austral Plan, a heterodox shock treatment launched in June 1985, succeeded in bringing inflation down to 100-200 percent. The plan included a wage and price freeze, a promise not to create money to finance the deficit, and a rescheduling agreement with creditors. However, the budget deficit remained at 5 percent of GNP, while net private investment turned negative. Fiscal reform, especially broadening the tax base, remains an important issue on Argentina’s agenda. Brazil Brazil’s experience combines elements from Korea, Turkey, Mexico, and Argentina. Like Argentina, Brazil has a long history of inflation, debt, and macroeconomic crises and stabilization plans. Cardoso and Fishlow argue that external debt shifted from being part of the solution to attaining sustainable growth to being part of the problem. The years 1968 to 1973 were a period of strong growth. Import substitution coupled with some export promotion and financed by foreign borrowing seemed to be working well. When the first oil price shock hit, there was little political support for an orthodox policy response. Instead, Brazil elected to continue the expansion and embarked on a National Development Plan which stimulated public investment and stressed import substitution. It relied on external funds to avoid passing external price increases through to domestic consumers and to maintain domestic demand. Widespread indexation made the rising inflation tolerable. As the current account deteriorated, the government relied increasingly on direct market intervention and controls to restrict imports. While the real exchange rate was kept relatively constant, few incentives were created to encourage exports. Thus, when the second oil price shock hit, the Brazilian economy was in a precarious position.

13

Introduction

In some respects, Brazil’s experience in 1974-78 paralleled the Korean experience. Korea also decided to borrow through the oil shock, without altering its plans to invest in heavy and chemical industries. As difficulties emerged, government intervention and import restrictions increased. However, there were also some important differences. Korea passed through the external price changes. Many more of Korea’s investments were in exportable industries, which would later generate foreign exchange to help repay debts. Also, Korean domestic savings rose during periods of strong growth, helping to finance the investments. Finally, Korea began to reduce government direct intervention and the size of subsidies to special interests before the crisis became serious. Brazil attempted to implement an orthodox program, including fiscal reform, in March 1979. However, the approach was labeled “recessionist.” A heterodox plan resembling the Argentine approach to inflation reduction through preannouncing the exchange rate devaluation was adopted instead, The program maintained real growth, inflation rose, the real exchange rate appreciated, and a massive current account deficit led to rapid debt accumulation. The plan was abandoned at the end of 1980 under pressure from foreign creditors. Thus, like Korea, Brazil suffered a debt crisis in 1980. Restrictive macroeconomic policies in 198 1 resulted in a recession along with a trade surplus, restoring Brazil’s access to international capital markets. However, Brazil’s policy lacked attention to medium-term structural adjustment, and an opportunity was lost to put the economy back on track. The public deficit rose, the real exchange rate appreciated, and exports declined, forcing Brazil to go to the IMF and reschedule its debts. Under a series of IMF programs during 1983-84, Brazil managed to overfulfill external balance targets and to revive growth of exports and output, but inflation persisted. Cardoso and Fishlow believe that the policies pursued to generate trade surpluses to service the debt also fueled inflation. They emphasize the switch from external debt finance of the budget to internal debt finance which pushed up domestic interest rates, feeding the budget deficit and lowering investment. At the same time, agressive devaluation fueled inflation in the highly indexed economy. Brazil’s inertial inflation differs strikingly from inflation in Korea and the Philippines. At the same time, the government was unable to raise tax revenues, a problem shared by Argentina, Mexico, and the Philippines. Fiscal policy played a central role in Brazil’s transition from a “successful debtor” in 1984 to a “problem debtor” by 1986. Internal problems, especially inflation, were the dominant concern, not external factors. Brazil also launched a heterodox plan to stop inflation in 1986. The Cruzado Plan did stop inflation initially, but the gains were short lived. As November elections approached, fiscal policy became expansionary and monetary growth accelerated. When the price freeze was no longer

14

Susan M. Collins

capable of containing the repressed inflation, the entire program fell apart and inflation reached 800 percent by mid-1987. Cardoso and Fishlow again emphasize the need for budget corrections and for revived investment expenditures, if Brazil is to achieve growth with financial stability.

Common Themes There are many similarities between countries which did relatively well and those which are still struggling to resume growth with low inflation. Even Korea, Indonesia, and Turkey (the “high growers”), which grew at moderate to high rates and avoided rescheduling in the 1980s, have been forced to readjust economic policies in the face of external shocks and poor economic performance. Strikingly, external shocks were not the primary cause of economic difficulties in either the high growers or the debt reschedulers. Nor do the differences in the magnitudes of those shocks differentiate between the two groups. In all of the countries studied, problems emerged because domestic policies were ill suited to cope with external conditions. In Korea’s 1979-80 crisis, external (and internal) shocks were compounded by heavy imports, overvaluation, and increased fiscal deficits associated with the Big Push of the 1970s. In Turkey and Mexico, a foreign exchange windfall, not negative external shocks, set the stage for unsustainable fiscal expansion, real appreciation, and debt accumulation. In contrast, foreign exchange inflows in Indonesia led to a fiscal expansion, but not to a surge in foreign borrowing. Six Lessons There are six important differences between those countries which rescheduled in the 1980s and those which did not. First is the importance of the breathing space which some countries had at the initial stages of the adjustment. None of the three high growers undertook a structural adjustment at the same time that it converted its external deficit to a surplus. In all three cases, financing was available to maintain investments and to stimulate growth while the current account deficit was reduced. Indonesia received generous official assistance in the mid- 1960s. Turkey received substantial foreign inflows after its 1977 crisis. Korea continued to borrow heavily in international capital markets during 1979-81. These countries all ran into difficulties before the magnitude of the crisis became evident and before lending to developing countries contracted in 1982. Second, the successful countries maintained high rates of capital formation. They avoided cuts in investment and encouraged investment in export industries. When world demand conditions were favorable, they were in a prime position to expand their exports rapidly. Of course, high rates of investment are not enough to ensure high rates of growth, especially if there

15

Introduction

are severe microeconomic distortions. The Philippines during 1979- 83 provided a stark example of very high incremental capital output ratios when capital accumulation was concentrated in industries with low or negative value added at world prices. Third, the successful countries maintained stable and competitive real exchange rates. Indonesia’s aggressive exchange rate policy kept nonoil exports competitive. Korea maintained a relatively stable and competitive real exchange rate. Devaluations in Brazil, Indonesia, Korea, and Turkey were successful in expanding exports, although sometimes with a lag. Fourth, the importance of fiscal discipline and a strong tax base stands out from the country experiences. In Korea, budget deficits above 4 percent of GNP were considered very large and lasted for no more than two consecutive years before being reduced to 1-2 percent of GNP. Revenues were maintained and increased through the introduction of a value-added tax and other taxes. Korea’s public sector also contributed to a high saving rate. In contrast, Argentina has not had a budget deficit below 4 percent of GNP since 1970. The deficit reached 18 percent of GNP during 1981-83. In both Argentina and Brazil, the budget deficit played a central role in debt accumulation and rising inflation. How to broaden and deepen the tax base has plagued these two countries as well as Bolivia, Mexico, and the Philippines. The timing of policy response to external shocks is a fifth important factor. Indonesia and Korea initiated adjustments because of undesirable domestic performance even before a crisis emerged. Arguably, Brazil could have avoided much of the 1981-83 recession if it had undertaken policy adjustments in 1979. Finally, both Korea and Indonesia have enjoyed trend increases in domestic (especially private) savings. The rising saving rate enabled these countries to finance high rates of investment with declining reliance on foreign borrowing. It is not clear, however, how much of the saving behavior is attributable to government policies and how much to social and political factors. Tradeoffs in Economic Policy A final theme which runs throughout the country case studies is how to allocate domestic resources among consumption (public and private), investment, and net resource transfers abroad. In the years preceding a debt crisis, the typical pattern has been an inflow of resources from abroad allocated to a combination of higher domestic consumption and higher investment. In Korea, the resources went primarily to investment. In Mexico and the Philippines in the early 1980s, a large part of the resources went to increased consumption. Difficult policy decisions arise when the net resource transfer must go in the other direction, to repay debts. If consumption is to be squeezed, it also

16

Susan M. Collins

matters how the transfer is achieved. Alternatives include inflation, as in Argentina, Bolivia, Brazil, and perhaps Turkey in the late 1970s, income taxes, and cuts in government expenditures. The alternatives can have widely different distributional consequences. However, for political as well as economic reasons, there are limits to the amount consumption levels can be squeezed in already poor countries. This perspective highlights differences in the adjustments across countries. Initially, Korea and Indonesia were able to maintain investment because the net resource inflow continued. Over time, real growth stimulated domestic savings, reducing private consumption as a share of GNP. The “lower” consumption substituted for net inflows from abroad by freeing resources for investment. In Argentina, Bolivia, Brazil, Mexico, and the Philippines, the reversal of net resource inflows led to cuts in investment and private consumption to contain imports. Growth slowed and there was no rise in private savings to ease the transfer burden. Thus, there are strong arguments for reducing debt repayments in order to stimulate investment and growth. The tradeoffs involved in allocation of domestic resources are at the heart of the debates over schemes for debt reduction. The lesson which emerges strongly from the NBER country studies is that, while capital inflows are no panacea and need not lead to investment and growth, it is unrealistic to expect countries to revive growth without some “breathing space” while they initiate structural adjustments.

I

Indonesian Economic Policies and Their Relation to External Debt Management Wing Thye Woo Anwar Nasution

This Page Intentionally Left Blank

1

Introduction and Summary

1.1 Introduction Our aim is first to explain why Indonesia did not experience an external debt crisis in the 1982-84 period as did most of the Latin American countries, and then on the basis of the analysis make recommendations to deal with Indonesia’s present debt situation. We hope that the lessons drawn will be useful as well in the design of adjustment policies for other countries experiencing debt-servicing difficulties. While a great deal of this monograph is about the conduct of economic policies and their consequences, our study goes beyond the normal economic analysis by giving detailed attention to the Indonesian historical and political setting within which decisions about policies are made. Specifically, the questions we are interested in are: What were the economic policies adopted, and what have been the results? What are the different strategies for economic development considered by the Indonesian government? What is the political and economic basis for each strategy and what is its distributional implication? What are the reasons a particular strategy was chosen and then subsequently superceded? Finally, even though the actual combination of strategies warded off a debt crisis in 1982-84, how likely is an external debt crisis in the near future, taking into account the present economic policies and the political dynamics in Indonesia? In the course of answering these questions, we conclude that the major reason for the absence of a debt crisis was the satisfactory management of the exchange rate. This task was made easier because there were neither burgeoning budget deficits nor extended periods of loss of control over monetary growth. The absence of protracted exchange rate overvaluation from 1979 onward was fundamental in maintaining a strong nonoil tradable sector. The nonoil tradable sector was able to earn enough foreign exchange 19

20

Wing Thye Woo and Anwar Nasution

to service Indonesian debts when the external shock of high interest rates increased debt-service payments and the recession in industrialized countries lowered the price of oil. We also want to emphasize that the absence of extended exchange rate overvaluation kept the external debt down by not encouraging capital flight. We ascribe this use of the exchange rate to protect the tradable sector as much to the existence of an influential political constituency consisting of (comparatively) neoclassical economists, Javanese peasantry, and Outer Island residents, as to balance-of-payments considerations. This monograph is organized as follows: chapters 2 and 3 explore the historical and political dimensions of economic policymaking, and chapter 4 verifies the veracity of our political economy approach by analyzing the distributional consequences of the fiscal system. Chapter 5 reviews the conduct of monetary policy and finds that the choice of the monetary control mechanism in the 1970s was influenced by political economy considerations. Chapter 6 shows how exchange rate management since November 1978 has been sensitive to both the economic viability of the agricultural export sector and the balance-of-payments position. The difficulties that political economy considerations imposed on external debt management are illustrated by the Pertamina crisis discussed in chapter 7. Chapter 8 identifies three factors-concessional loans, prudent debt management, and export orientation-responsible for the non-crisis outcome in 1982-84, and then discusses the role for policy in ensuring a non-crisis outcome in the medium run. In the remainder of this introductory chapter, we give a summary of each of the succeeding chapters.

1.2 Political and Economic Instability, 1950-65 An understanding of the economic conditions prior to the establishment of the New Order government in 1966 is important because the economic policies of the 1950-65 period left a very strong imprint on the institutional memory of the new government, especially insofar as it resides in the person of President Soeharto. During this period, the Indonesian government was preoccupied with domestic political and military problems, with the restoration of sovereignty on West Irian, and with political recognition in world forums. Little attention and resources were devoted to economic development. An increasingly difficult budget situation made inflation a major problem. Taxes on trade were the major source of government revenue, but the overvalued multiple exchange rate system was reducing the profitability of the tradable sector, causing it to shrink. The twin rebellions on the islands of Sumatra and Sulawesi in 1958 constituted a simultaneous supply and demand shock to the budget. The rebellions forced big increases in military expenditure, at the same time reducing the government revenue base since

21

IndonesiaiChapter 1

both of these islands were important sources of export tax revenue. The monetization of the budget deficits raised the average 1958-61 inflation rate to 25 percent from the 1950-57 average of 17 percent. The budgetary pressures grew steadily worse, resulting in a period of even higher inflation in 1962-65. Between 1962 and 1964 both money supply and the cost-of-living index roughly doubled every year, and by the end of 1965 they were doubling every few weeks. Economic growth slowed to 0.8 percent per year in this turbulent period. The evolution of the export/GDP ratio tells the story of economic decline very well; it fell from 8.7 percent (1951-57) to 6.8 percent (1958-61), and then to 1.1 percent (1962-65). The internal political struggle culminated with the abortive coup by military personnel sympathetic to the Indonesian Communist Party (Purtai Kommunis Indonesia or PKI) on the night of 30 September 1965. The political instability aggravated the economic instability. The increasing economic difficulties speeded up the transfer of authority from President Soekarno to the anti-communist General Soeharto in the following year.

1.3 The Political Economy Factors in Policymaking The economic chaos of the 1958-65 period left such a deep impression that the new Soeharto government has had a “balanced” budget rule since 1968. In reality, this rule amounts to refusing to finance the deficit through money creation and limiting the deficit to the availability of foreign loans, which are officially described as foreign “revenue.”’ Another consequence of the chaos was the recognition that the exchange rate is an extremely potent policy instrument that can effect large-scale, economy-wide resource reallocation and income redistribution. President Soeharto has been in power since October 1965 and has not faced any serious challenges to his rule since the 1974 riots. The Indonesian political system can be described as a corporatist-authoritarian state. We were able to identify three political concerns that have significantly influenced his choice of economic policies. The first is to avoid conditions favorable to the revival of the PKI. Since the PKI was primarily a Javanese peasant-based movement, the policy implication is that conditions in the rural area must be improved. The second concern is to display equitable treatment of the Outer Islands, given their long history of secessionist movements. Since the economy of the Outer Islands depends heavily on tree crop exports, this further strengthens the case for promoting agricultural development. It must be added that Soeharto, because of his peasant origin, has consistently shown a strong personal commitment to eliminating rural poverty. Soeharto’s third concern is one common to all politicians: the maintenance of his power base. Political patronage has often taken the form of trade restrictions to benefit specific groups. It would not be correct, however, to attribute all trade restrictions to rent-seeking motivation. Ideology also plays an important

22

Wing Thye Woo and Anwar Nasution

role. Soeharto, like most members of the 1945 generation who fought in the bitter war for independence from the Netherlands, is influenced by an economic nationalism which is congruous with Indonesian political nationalism. Dutch economic policies were seen as designed to impose a plantation economy on Indonesia to serve the need of Dutch manufacturing industries for raw materials. In reaction, the generation of 1945 regards industrialization as synonymous with economic development. The policy translations of economic nationalism are high trade barriers to induce the development of a manufacturing sector and foreign investment laws more strict than those of neighboring countries. The primary reason Indonesia sometimes pursues a contradictory mix of liberal and protectionist policies is the above political, ideological, pecuniary, and personal elements working themselves through two groups of contending presidential economic advisors, popularly referred to as the technocrats and the technicians. The technocrats are neoclassical economists who subscribe to the comparative advantage principle. This has meant a favorable treatment of the agricultural sector because it supplies about 80 percent of nonoil exports. Exchange rate devaluations, rather than the removal of trade restrictions on imported inputs, are used to promote exports because the technocrats control the ministries that oversee macroeconomic policies but not the Ministry of Trade and the Ministry of Industry which have authority over trade restrictions. Their policies find favor with the president because they address his political concerns for raising rural income and for maintaining equitable treatment of the main islands. The second group of economic advisors, the technicians, is an amorphous collection of technicians-turned-managers, military advisors, and economists with structuralist inclinations. The technicians are united by their common belief in the general validity of the infant industry argument and by their common rejection of foreign capital ownership. The technicians’ push for import-substitution industrialization has won them the support of the army, the most powerful constituency in the country. Thanks to the dual function (dwifungsi) doctrine which legitimizes military participation in economic development, the expansion of state enterprises translates directly into more managerial positions for senior military personnel. It must be noted that since most of the import-competing industries are set up in urban Java, the higher prices of manufactured goods represent an implicit tax on residents of the rural sector and the Outer Islands. In looking at the political setting within which policies are chosen, we have identified an important political coalition of technocrats, Outer Islanders, and rural residents which favors a policy package emphasizing the maintenance of a competitive exchange rate. Since a debt crisis occurs when a government runs out of foreign reserves-either to service its guaranteed external debts or to permit private residents to convert their domestic currency to service private external debts-such a policy package reduces

23

IndonesidChapter 1

the probability of a debt crisis by keeping the (foreign exchange earning) nonoil export sector healthy.

1.4 The Fiscal System The most notable feature of the 1970s is the central government’s increasing reliance on oil as its chief source of revenue. Oil revenue as a share of total federal revenue rose from 26 percent in 1969/70 to peak at 71 percent in 1981/82. Nonoil revenue normalized by GDP fell from the 1969-71 average of 8 percent to the 1980-82 average of 6 percent. The fiscal danger of such a narrow tax base was brought home dramatically in 1982 when the global recession caused oil prices to collapse. Oil revenue (in 1980 rupiahs) fell from Rp 7.8 billion in 1981182 to Rp 6.9 billion in 1982/83, causing real total revenue to fall for the first time since the Soekamo years. It was clear that greater internal resource mobilization was necessary. A completely revised personal and corporate income tax code came into force in January 1984, a value-added tax in April 1985, and a consolidated property tax in 1986. The result was that real (1980 prices) nonoil revenue rose from Rp 3.6 billion in fiscal 1983 to Rp 4.7 billion in fiscal 1986. It is important to note that the oil revenue contributed to “undertaxation” in a subtle way which led to greater external debt accumulation. The two OPEC price increases encouraged undertaxation by giving Indonesia access to external credit on very favorable terms. The examination of the fiscal system supports our claim that the technocrats favor an economic strategy which leads to resource transfers to the Javanese rural areas and to the tree crop industries in the Outer Islands. This favorable tax treatment of the agricultural sector improves the rural-urban terms of trade and hence encourages the production of tradables, the presence of which determines a country’s ability to service its debts. In examining government expenditure, we surmised from fragmentary evidence that government spending was more likely to display a rural rather than an urban bias. Because of better data, stronger evidence could be garnered to support the hypothesis that budget allocations were very sensitive to inter-island equity. There is in fact evidence that inter-island equity takes precedence over rural-urban equity. This is consistent with our conjecture that concern for rural development stems more from a desire to eradicate poverty than to narrow the rural-urban gap. The analysis in chapter 4 sets the stage for our discussion in which we determine that political economy factors have been important in determining the debt outcome. To the extent that people are consistent in their actions, the fact that the technocrats implemented, with Soeharto’s approval, a fiscal policy favoring the tradable sector means that they would also advocate a

24

Wing Thye Woo and Anwar Nasution

similarly-oriented exchange rate policy. Such an exchange rate policy, as we will see, yields side effects that are salutary to external debt servicing.

1.5 Monetary Policy and Financial Structure While the balanced budget policy of the Soeharto government effectively ended the creation of money to finance budget deficits, it could not prevent monetary policy from moving in tandem with fiscal policy during the 1970s. This is because the main instrument for monetary control prior to April 1974 was the extension of central bank credits to the banking system, state enterprises, and private companies. Since these credits were extended for a contracted time period, the government was not in a position to engineer quick increases or reductions of the money stock. This meant that with the maintenance of a fixed dollar-rupiah exchange rate, the conversion of oil revenue from dollars to rupiahs in order to finance expanded government expenditure automatically increased the money supply. Thus, when the price of oil quadrupled at the end of 1973, encouraging the government to increase its spending, the monetary authorities lost control of the money supply. Reserve money grew 57 percent in 1974 and the inflation rate for that year was 41 percent. The central bank responded to this monetary anarchy by setting lending ceilings on the banking system in April 1974. It is understandable to use lending ceilings as a short-run, stopgap measure to control monetary growth, but Indonesia continued to rely upon them for monetary control until June 1983, despite their well-known deleterious effects. The reason is that credit ceilings gave the government an additional instrument to consolidate its political base. Over time, Bank Indonesia was instructed to introduce detailed ceilings by type of credit for each bank through an extensive selective credit system featuring subsidized interest rates to achieve other goals. For example, banks were assigned a civic function insofar as they restricted certain credit only to pribumis (indigenous people) and made establishing credit for them a priority in order to enhance pribumi participation in economic activities. The highest priority items were financing rural participation in the government’s rice intensification program and financing the scheme to stabilize the price of rice. Since most priority credits were handled by state banks, the policy of ceilings with selective credit became one of the major tools protecting the state banks from competition with private banks. Since state banks also had a wider network of branch offices than the private banks, the result of choosing this mechanism of monetary control was the preservation of the dominant position of the state banks. In the absence of pressures to innovate in order to be more efficient, the Indonesian financial system remained underdeveloped. One of the policy responses to the external shocks of the 1980s was an overhaul of the financial structure which was announced in June 1983. The financial refom package included partial deregulation of interest rates,

25

IndonesidChapter 1

elimination of credit ceilings, and reduction in the scope of Bank Indonesia's subsidized credits to state-owned banks. To increase the number of central bank instruments for open market operations, Bank Indonesia began reissuing Debt Certificates (SBIs) in February 1984 and introduced Money Market Instruments (SBPUs) in January 1985. It is clear that the financial instruments SBI and SBPU still did not provide sufficient control over monetary aggregates. This is evident in the way the money supply had to be contracted in response to a speculative run on the rupiah in June 1987. The minister of planning, in addition to asking Bank Indonesia to sell Rp 800 billion of open market instruments, ordered the state enterprises to withdraw Rp 1.3 trillion from state banks to be put into central bank securities. This action reveals that the market for both SBI and SBPU was still too shallow. It may be difficult to increase their role if the financial markets remain underdeveloped. Financial deepening is an important priority, but not only because of the need to enhance the effectiveness of the monetary instruments. Financial deepening would also better mobilize (and maybe increase) domestic saving, reduce dependence on external credit, and improve the overall allocation of capital within the economy. A boost to financial deepening would be the privatization of some of the state-owned enterprises, and an easing of external debt service would be achieved if foreigners were allowed to buy into these enterprises. 1.6

Exchange Rate Policy

Indonesian exchange rate policy is characterized by three distinct phases during the 1966-87 period. In the first phase, from October 1966 to July 1971, there was a steady dismantling of the multitiered exchange rate system into a unified exchange rate. This phase revealed a readiness to have medium-sized devaluations at short intervals in order to restore the competitiveness eroded by higher inflation in Indonesia. In the second phase, from August 1971 to October 1978, there was a fixed exchange rate. The reason for this remarkable stability is straightforward: the balance of payments was very strong throughout the period. The rapid development of the oil sector together with the 1973 OPEC price increase caused Indonesian oil exports to grow as follows (in U.S.$ billion): 0.4 in 1969, 0.9 in 1972, 5.2 in 1974, and then 7.4 in 1978. The result was a swelling of the nongold reserves, measured in the number of weeks of imports they could support: 8.1 in 1969, 19.1 in 1972, 20.2 in 1974, and 20.4 in 1978. The macroeconomic conditions also did not warrant any additional stimulus which a devaluation would bring. The sustained high income growth rates of this period-7.9 percent per year-were achieved with substantial overheating of the economy. The average 1973-78 inflation rate was 22 percent compared with 8 percent in 1970-72.

26

Wing Thye Woo and Anwar Nasution

The third phase involved three large devaluations in November 1978, March 1983, and September 1986, separated by moderately long periods of gradual exchange rate depreciation. There are two mutually compatible explanations for the 50 percent devaluation in 1978. The first explanation views the November 1978 devaluation primarily as an anticipatory action to the inevitable dropoff in oil export earnings due to resource depletion. The second explanation emphasizes the economic difficulties and political tensions associated with the reallocation of resources being forced upon the economy by the overvalued exchange rate. The overvaluation of the rupiah was the result of maintaining the exchange rate at 415 rupiahs to the dollar despite the large domestic inflations from 1974 to 1977. This meant that Indonesian producers of tradables were experiencing a profit squeeze. The prices of their output were fixed by international competition, but the prices of their domestic inputs were being driven up by the double-digit inflation. The result was reports of increasing unemployment in the tradables industries, particularly in the labor-intensive agricultural export sector. Indonesia was suffering from the “Dutch disease”.* In this second explanation, the 1978 devaluation was as much due to political concern about worsening conditions in the countryside as to economic concern about the desirability of the resulting composition of economic activities. After all, the level of reserves in 1978 could sustain the existing amount of imports much longer than was possible at any time during the 1969-76 period. It must be stated that the deleterious effects of the Dutch disease on the nonoil export sector are not obvious. Nonoil, non-liquified natural gas (LNG) exports, whether measured in physical units or in dollars or in the units of imports for which they can be exchanged, show steady growth throughout 1972-78. The production disincentive faced by the nonoil export industries is clearly seen only when the local purchasing power of their exports is measured. Even though nonoil exports were bringing in increasing amounts of foreign goods, the steady real appreciation of the exchange rate meant that the nonoil export industries were not being paid a greater number of baskets containing the mix of goods typically consumed by Indonesians. In terms of foreign purchasing power, these industries increased their revenues by 32 percentage points between 1973 and 1978, but their revenues were unchanged if measured in terms of local purchasing power. More direct evidence of the production disincentive experienced by the nonoil export sector is the movement of prices of tradables relative to prices of nontradables, PT/PN. Our three proxies of PT/PN show an average decline of 29 percent between 1973 and 1978. The 50 percent nominal exchange rate devaluation caused FT/PN to increase by 29 percent, almost restoring the profitability of the tradable sector. The speed and size of the response of nonoil, nonLNG exports were impressive. Export volume went up by 36 percent in one year, raising dollar earnings by 52 percent. The 32 percent growth in foreign purchasing power in 1979 translated into a domestic purchasing power increase of 78 percent.

27

IndonesidChapter 1

We interpret the quick and large response to the devaluation as proof of the Dutch disease, the mounting severity of which since 1974 caused an increase in excess capacity in the traditional export industries. Small producers of tree crops were spending more and more of their time in nontradable activities. Meanwhile, the real prices of their agricultural products sank as a constant nominal exchange rate was maintained in the face of big domestic price increases. The March 1983 and September 1986 devaluations were undertaken to boost nonoil exports in the face of big declines in oil export earnings. Manufacturing exports grew very rapidly after the March 1983 devaluation. They jumped (in U.S.$ million) from 850 in fiscal 1982 to 1,480 in fiscal 1983, and then to 2,166 in fiscal 1984. The strong export responses to the 1978 and 1983 devaluations indicate that export-oriented industrialization is a real possibility as long as favorable relative prices are maintained through appropriate exchange rate and trade policies. While it is clear that negative external demand shock had a role in worsening the balance of payments from 1983 to 1986, we want to point out that there were also internal developments during this period that caused substantial movements in relative prices which were unfavorable for the tradable sector. Specifically, we are refemng to the widespread use of quantitative restrictions (QRs) in the 1980s. Of the 5,229 items imported in 1985, 1,484 required import licenses and 296 were under quotas. The import licenses were usually given to only two or three traders or to the few firms producing the competing goods domestically and thus conferred a monopoly position to their holders. The range of activities that were protected by import licenses accounted for 32 percent of total domestic value added, excluding construction and services. If the petroleum sector, which requires no protection, is also excluded, then the coverage is 53 percent of total domestic value added. The implication of this microeconomic distortion for exchange rate management is profound. The intrusion of this distortion since late 1982 and its quick metastasis across the tradable sector renders invalid trying to draw conclusions about production incentive by examining the movements of the macroeconomic proxies for FT/PN. Output prices of tradables are set by international competition, while output prices of nontradables (which are generally very labor-intensive) are set by the domestic cost structure, the level of which is primarily determined by domestic wages on the supply side and domestic macro conditions on the demand side. Hence, the introduction of a quota on an imported input to the tradable sector will reduce the profitability of the tradable sector without any change in PT/PN. The point we want to emphasize is that although the Morgan Guaranty competitiveness index in predevaluation 1986 shows almost the same value as in postdevaluation 1979 (110 versus I l l ) , it does not indicate that the August 1986 exchange rate was not overvalued. In order to have the 1986 nonoil export supply schedule in the same position within the familiar

28

Wing Thye Woo and Anwar Nasution

Marshallian price-quantity space as in 1979, a devaluation was warranted, particularly in light of the shrunken gap between output and input prices. It is of course an empirical question how much the additional nonoil export earnings would have been in the absence of QRs, especially in comparison to the fall in oil export earnings. The current account deficit would still have widened in 1986, but not to the same extent.

1.7 External Debt Management The Soeharto government is no stranger to external debt management, having inherited an external public debt of $2 billion. It cut its teeth on the economic stabilization and rehabilitation program of 1966, within which rescheduling the Soekarno debts and arranging for new capital inflows to support the balance of payments were key components. Given the desperate situation of Indonesia, the Western countries set up the Inter-Governmental Group on Indonesia (IGGI) to ensure a long-term coordinated plan of official assistance. IGGI was generous both in the amount and the terms of assistance. In the 1967-70 period, it gave an average of $477 million a year, with a repayment period of 25 years, including 7 years of grace, and an interest rate of 3 percent. The mix of generous external assistance and corrective economic policies undertaken by the Soeharto government imparted a new dynamism to the Indonesian economy. The annual average growth rate from 1968 to 1972 was 8.2 percent compared to the average rate of 1.2 percent in the preceding five years. By 1974 the international credit markets had rescinded whatever credit restrictions they had imposed on borrowing by the Indonesian government in the aftermath of Soekamo’s economic Armageddon. The reasons for this change were threefold: one, the avalanche of oil revenue increased the creditworthiness of the Indonesian government; two, there was a boom in commodity prices in the early 1970s; and, three, lending opportunities decreased in OECD countries whose medium-term economic prospects were rather bleak after the 1973 OPEC price increase. On consumption-smoothing grounds, readmission into the external credit market resulted in a net gain to Indonesian national welfare. This welfare gain was not without its price: Indonesia was now exposed to two new risks. The first risk is systemic in nature and threatens every country with external debts. The second is the possibility of imprudent borrowing by Indonesia. This danger was realized in February 1975 when the state oil company, Pertamina, could not roll over a $400 million short-term loan and defaulted. It could be cogently argued that the Pertamina crisis was a blessing in disguise. The government, by denying all state-owned enterprises direct access to the external credit market after the Pertamina embarrassment, did not have as large a publicly-guaranteed external debt as it otherwise would have had at the beginning of 1982.

29

IndonesidChapter 1

While the Pertamina crisis was a deep one, it was still manageable. Real interest rates were low and Indonesian exports were in high demand. The collapse of oil prices since 1982 and the subsequent collapse of commodity prices have made external debt management more difficult. 1986 was a particularly bad year for Indonesia. The price of oil dropped precipitously from $28 per barrel in January 1986 to $10 in August 1986. The yen, in which more than a third of Indonesian public external debt is denominated, appreciated 21 percent against the dollar. The result was that the public debt-service ratio shot up to 29 percent at the end of 1986. The Indonesian government has shown itself to be prepared to make significant policy changes to ward off an external debt crisis. The rupiah was devalued by 38 percent in March 1983 and by 45 percent in September 1986. In October 1986 and January 1987, the input costs to the export sector were lowered by abolishing a substantial number of import restrictions. Nominal state expenditure for fiscal 1986 was reduced 8 percent from the previous year. In fact, real government expenditure has been steadily lowered since fiscal 1983. Real expenditure has declined more than the drop in real revenue, reducing the amount of real borrowing (in 1980 prices) from Rp 2.8 billion in 1983 to Rp 2.1 billion in 1986. Mobilization of domestic resources was also undertaken to slash foreign borrowing, with the income tax system completely overhauled in December 1983 and a value-added tax introduced in April 1985, followed by a more comprehensive land tax in June 1986. These tax increases, however, have not been able to make up for the fall in oil royalties. Nominal revenue in 1986 was Rp 1.4 billion lower than in 1985.

1.8

Conclusions and Prospects

As pointed out earlier, a debt crisis occurs not only when the government does not have the reserves to service the loans it has guaranteed, but also when it does not have the reserves to enable private domestic residents to convert their service payments on nonguaranteed debts from domestic currency to foreign ~ u r r e n c yTo . ~ take the second cause of a debt crisis into consideration, we define the total external debt service to be the sum of external short-term debt and the debt service on all external long-term debt, publicly-guaranteed and private nonguaranteed. We include short-term debt in our definition because we are interested in the financial resilience of a country to sudden protracted credit squeezes in international credit markets which make short-term borrowing extremely expensive, if not occasionally impossible. After all, the 1973-74 credit crunch did precipitate the 1975 Pertamina debt crisis in Indonesia. Since the reserve position of a country is crucial for avoiding debt crises, it is not appropriate to assess a country’s ability to pay by looking at the total external debt service with respect to its income. A more appropriate

30

Wing Thye Woo and Anwar Nasution

indicator is the debt-service ratio, DSR-the total external debt service normalized by the level of exports-because the official reserve position is determined primarily by the ability of the export sector to earn foreign exchange. By comparing the DSRs of Indonesia, Mexico, and Brazil, we identified three factors that explain why Indonesia did not experience a debt crisis in 1982-84 as did Mexico and Brazil. These factors are concessional loans, high export orientation, and prudent management of the maturity structure. Concessional loans. A high proportion of Indonesia’s external debt was borrowed at fixed concessionary rates from IGGI. This “IGGI effect” explains why the effective interest rate on Indonesian long-run debt averaged 16 percent against the 20 percent paid by Mexico and Brazil. Another result was that only about one-third of Indonesian debt was denominated in dollars compared to 90 percent of Mexican and Brazilian debt. This meant that the large appreciation of the dollar from 1979 to 1982 did not raise the effective interest rate for Indonesia as much as it did for Mexico and Brazil. High export orientation. The availability of significant amounts of other tradables prevented Indonesia’s debt-servicing capacity from collapsing as did Mexico’s when the price of oil dropped in early 1982. Appropriate exchange rate policies by Indonesia, exemplified by the 1978 devaluation, ensured a diversified export bundle as well as a high export orientation. The average 1980-82 export/GNP ratio was 27 percent for Indonesia, but only 14 percent for Mexico and 9.5 percent for Brazil. Indonesia’s political concern to keep the agricultural sector vibrant no doubt helped to maintain the observed export orientation. Prudent management of the maturity structure. This was evident in that only 14 percent of Indonesia’s debt in 1981 was short-term compared to the 19 percent for Brazil and 32 percent for Mexico. The shock of the Pertamina crisis caused official borrowing in Indonesia to take place very cautiously with regard to exposure in the short-term credit market. We can also call this third factor the Pertamina legacy. To quantify the relative importance of the three factors in explaining the absence of an Indonesian debt crisis, we calculated what the total DSRs in the 1980-82 period would have been if Indonesian debt was paying the same effective interest rates as Mexico, as well as bearing the same maturity structure, and if the Indonesian exporVGNP ratio was identical to Mexico’s. The average values of the six possible decompositions of the effect of each factor and the range of values achieved are reported in table 1.1. The decomposition identifies the export orientation of Indonesia as the most decisive factor in why Indonesia’s total debt-service/export ratio was so low compared with Mexico’s. Export orientation explained 31 of the 54 percentage point difference, accounting for 57 percent of the gap. The Pertamina legacy was of moderate importance. It contributed 18 percentage points, accounting for almost a third of the gap. Concessional interest rates

31

IndonesidChapter 1 Average Values and Range of Values Achieved

Table 1.1

Average Value Hypothetical 1980-82 average total debt-service ratio when Indonesia had all 3 Mexican features Contribution of (in percentage points):

IGGI Maturity management Export orientation Actual 1980-82 average total debt-service ratio

Range of Values for Each Factor

84.4 5.8 11.1 30.8

3.7-8.4 12.0-23.8 23.6-31.8

30.1

and currency composition of debt played only a minor role in reducing the DSRs, explaining less than 6 percentage points. Our finding that the IGGI effect contributed so little toward the reduction of the 1980-82 debt-service/export ratio is surprising because many of the informed observers we talked to cited foreign concessionary loans as the primary reason for the absence of an Indonesian debt crisis. The $1 billion saved annually in reduced debt service during 1980-824 is a large sum of money, but this amount would have been easily swamped by a Mexico-style loss of reserves if the Indonesian government had tried to prop up an overvalued exchange rate and was then forced to finance capital flight. Similarly, if exports were 12 percent below actual value because of an overvalued exchange rate, as suggested by the 1965-68 experience, the loss in foreign reserves would have also greatly exceeded this $1 billion saving. Our conclusion is that the Indonesian exchange rate policy was the most important reason Indonesia was able to meet its debt commitments in the 1982-84 period. The conduct of this exchange rate policy was greatly facilitated by the existence of a political lobby that promoted exchange rate protection and by the memory of the economy-wide negative effects of exchange rate overvaluation. While the decline of oil prices in early 1982 did make debt management more difficult-the total DSR rose from 26 percent in 1981 to 39 percent in 1982-it was still far from a crisis situation. The subsequent collapse of agricultural commodity prices and the rapid descent of the price of oil in 1986 produced a more ominous situation. The fall in export earnings from $21 billion in 1982 to $15 billion in 1986 caused the total DSR to soar to 68 percent. The 68 percent appreciation of the yen against the dollar between 1985:lQ and 1987:1Q caused debt-service payments to jump. While our analysis would place great emphasis upon an aggressive competitive real exchange rate policy to reduce the probability of a debt crisis through its effects on exports and capital flight, we strongly feel that there are a number of other policies which must be implemented

32

Wing Thye Woo and Anwar Nasution

immediately if a debt crisis is to be avoided in the medium run. We recommend the following policies: 1. Cut budget deficits by controlling spending and increasing taxes. This

would keep fiscal policy consistent with exchange rate policy. 2 . Maintain an anti-inflationary posture in monetary policy. 3. Amend the balanced budget rule to allow internal financing of government deficits. The deepening of the domestic market in government securities would make open-market operations by the central bank easier. The addition of this monetary tool to SBI and SBPU would tend to enhance monetary control, and thus macroeconomic stabilization efforts, tremendously. 4. Accelerate the development of the domestic financial system. Financial deepening could be boosted by the privatization of many of the state-owned enterprises. The debt situation would be improved if the government were to allow foreigners to purchase shares in the former state enterprises. 5 . Liberalize controls on foreign investments in the manufacturing and agricultural sectors, especially in industries that produce primarily for export markets. In short, state and private enterprises should issue equities instead of bonds to foreign investors when financing their capital expenditure. 6. Eliminate the wide array of monopoly import licenses. The growth of manufactured exports, spurred by access to cheaper inputs, would not only increase foreign exchange earnings but would also diversify the export bundle, hence reducing the sensitivity of the DSR to the prices of a few key commodity exports. 7 . Expand the tree crop sector. In addition to earning more foreign exchange, the strategy of accelerating the growth of agricultural export industries would also promote a more equitable, rural-urban-as well as inter-island-growth pattern, and ease population pressure on the urban areas.

2

Political and Economic Instability, 1950- 65

2.1 Introduction Indonesia achieved formal recognition of its political independence in December 1949 after a bitter, four-year struggle against the Dutch. President Soekamo was the dominant political figure in the 1950-65 period. The years

33

IndonesidChapter 2

from 1950 to 1959 are usually referred to as the period of liberal democracy when President Soekarno shared power with elected representatives. He was the only durable presence, however; the extreme fragmentation of political opinions brought about twelve cabinet changes in ten years. The corollary of this constant reshuffling of ministerial portfolios is that government policies toward economic development were in a constant state of flux. Needless to say, the continuous jockeying for power by various coalitions diminished public respect and support for parliamentary democracy. In June 1959 President Soekarno, with the support of the army, centralized political power in his hands by dismissing the elected representatives. The parliamentary seats were reassigned to representatives of regional interests and of “functional” groups. Examples of the latter are youth and women’s groups, and religious, labor, and peasant organizations. This form of “guided democracy” lasted until October 1965 when the armed forces took over after suppressing a communist coup. During the 1960-65 period Indonesia continually faced fiscal and trade problems, in addition to unemployment problems resulting from a high rate of population growth. The root of the fiscal problem was ever-increasing budget deficits. Since tax revenues could not rise as fast as expenditures and governments were unable to curtail spending, the deficits had to be financed by money creation. Government revenue was heavily dependent on the taxing of international trade. These trade taxes reduced profitability in the trade sector, cut real wages, and increased (disguised) unemployment, at least in the short run. High inflation and high domestic costs of production made exports uncompetitive in international markets. Occasional bans of some export commodities to stabilize their prices in the domestic market and the trade breaks with Singapore and Malaysia since 1963 made Indonesia an unreliable source of traded goods. All of these factors made Indonesia unable to translate its export potential into actual earnings.

2.2 The Course and Causes of Inflation The episodes of economic development in Indonesia during 1950-65 can be divided into three phases, 1950-57, 1958-61, and 1962-65. Despite the political instability and destruction during the Pacific War and the Revolution, the economy performed quite well during the first phase. Measured at 1955 constant market prices, the average annual growth rate was a respectable 5.4 percent per annum, and the average inflation rate was 16.5 percent per year (table 2.1). The trade sector was badly hit by a decline in the terms of trade after the Korean War boom. Despite that, the average current account deficit during this period was less than 1 percent of the national income. Table 2.1 shows that 1951 was the only year the government ever had a budget surplus. Since then, the budget has always been in deficit. The average amount of this deficit during 1951-57 was 1.5 percent of the

Table 2.1

Selected Economic Indicators, 1951-69

1951

1952

1953

1954

3.1 31.2 6.0 -27.6

8.8

6.5 13.4 5.6 - 10.1 - 19.3

6.9 48.5 6.4 3.1 - 17.7

5.8 10.0 35.0 9.1 .3

2.6 9.5 - 1.5 -2.0 36.3

17.4 15.7 I .7 .9 5.0 7.3

14.6 17.9 -3.3 -3.5 12.9 12.7

15.3 17.7 - 2.3 - 1.4 9.9 10.8

9.7 11.0 - 1.3 - .4 5.9 8.2

11.2 12.8 -1.6 .9 5.7 8.5

3.8

11.4

11.4

11.5

500.8 11,402.0 495.8

266.7 9,736.0 460.4

192.5 7,571.0 495.0

14.5

14.4

12.6

Growth rates National income Money supply Inflation Exports' Imports'

6.3 16.9

64.1 61.4 98.4

As percentage of national income Government revenue Government expenditure Government deficitb Current account Imports Exports Exchange rate (Rp/US$)' Prices of export goodsd Rubber (USelkg) Tin (US$/metric ton) Coffee (US$/kg) Income velocity of money

1955

1956

1957

1958

1959

7.1 41.2 54.9 3.0 -6.6

-4.1 55.3 18.0 - 17.2 - 32.3

2.4 18.8 13.2 17.7 - 11.4

12.2 13.3 -1.0 - 1.2 6.5 7.0

11.7 14.6 -2.9 - .5 5.2 6.3

12.7 19.2 -6.5 .8 3.4 5.0

11.5

11.5

11.5

200.0 7,619.0 678.5

326.0 7,713.0 537.4

268.6 7,923.0 595.3

13.1

10.9

11.8

1960

1961

1962

2.0 37.1 20.0 -9.7 19.9

5.1 41.4 95.2 -6.3 37.8

2.4 100.9 155.9 - 15.8 - 18.7

13.9 20.1 -6.2 .9 2.5 4.9

13.7 15.5 -1.8

5.5 9.1 -3.6

6.1 9.7

13.2 18.8 -5.6 -4.4 7.6 1.5

11.5

45.3

45.3

45.3

45.3

239.6 1,429.0 503.2

211.9 7,109.0 404.6

275.4 7,701.0 354.8

288.2 7,652.0 345.0

218.2 8,380.0 329.0

201.7 8,320.0 303.0

10.9

7.6

6.9

9.4

8.1

13.1

-.I

- .7

2.2 2.3

Annual Averages 1%3

1965

1964

1966

1967

1968

1%9

1951-57

1958-61

1%2-65

1951-65

Growth rates

National income Money supply Inflation Exports' Imports'

-2.4 93.8 128.8 5.2 - 19.4

As percentage of national income 5.1 Government revenue Government expendihm 10.3 Government deficitb -5.2 Current account - .3 Imports .7 Exports 1.0 Exchange rate (Rp/US$)' -

Prices of export goodsd Rubber (USGlkg) Tin (US$/metric ton) Coffee (USe/kg)

.o

3.8 156.3 135.3 3.8 30.4

.O 282.5 594.5 -2.3 2.2

2.3 763.5 635.4 -4.1 - 24.2

2.3 131.8 112.2 8.0 35.1

121.3 84.8 13.1 3.2

7.1 61.1 17.4 14.1 19.7

5.5 24.4 24.4 5.3 14.3

1.4 38.2 36.6 - 3.9 3.5

4.0 9.6 -5.6 -.l .4

3.9 10.7 -6.8 - .2 - .6 .6

4.1 9.3 -5.2 -6.1 -33.4 40.0

7.1 10.3 -3.2 -4.1 - 13.1 12.5

7.5 9.3 - 1.8 -3.2 -11.9 12.4

9.4 12.8 - 3.4 - 3.9 -11.5 11.5

13.2 14.7

13.4 18.4

- 1.5 - .7 7.3 8.7

5.1 6.8

1.1

10.9 14.4 -3.5 - .7 4.9 6.2

235.0

235.0

235.0

300.1

326.0

10.4

36.9

70.1

33.4

.5

-

11.1

-5.0 - .9

1

158.4 253.6 -2.3 - 1.4 4.6 9.9 -5.3 - .3 .7

3.2 63.8 88.8 .8 7.2

191.8 8,586.0 299.0

177.9 11,436.0 361.1

184.0 12,977.0 356.3

164.0 11,492.0 336.0

140. I 10,608.0 294.3

132.7 10,019.0 301.0

168.9 10,451.O 302.4

284.9 8.484.7 537.9

248.4 7,710.5 358.4

188.9 10,329.8 329.9

249.6 8,770.3 434.6

16.1

15.2

14.6

26.8

23.0

24.1

18.3

12.6

8.0

14.8

11.9

Income velocity of money

Sources: BPS, Statisticul Pocketbook of Indonesia. 1960-71 editions. Bank Indonesia, Report of the Governor, various issues covering financial years 1950151 through 1960/65, IMF and World Bank databases. Note: Dash indicates that data were not available. 'In foreign currency (US $). bMinus sign indicates a deficit. Wfficial exchange rates at Jakarta. 1980 constant US. dollars.

36

Wing Thye Woo and Anwar Nasution

national income. The Bank Indonesia Act of 1953 restrained the use of deficit financing to a level equal at most to five times the amount of the nation’s gold and foreign exchange reserves. Inflation during this phase was largely due to positive external shocks. The rise in exports because of the Korean War boom in 1950-51 increased domestic income as well as the monetary base. As shown in table 2.1, the rates of growth of the money supply and inflation rates fluctuated sharply during 1950-57, reflecting the efforts of various governments to check the inflation rate. The exceptional rate of growth of the money supply in 1954 was related to the rising campaign expenditures of various political parties for the general election in 1955, the first to be held after achieving independence. The incumbent government expanded bank credit and used the trade licensing system to, in essence, buy political supporters. Inflation during the second phase, 1958-61, was primarily related to a combination of a supply shock and a sharp increase in government expenditures. These simultaneous adverse events were the result of, first, the twin rebellions in 1958 on the wealthy, export-earning islands of Sumatra and Sulawesi and, second, the nationalization of foreign-owned firms and expulsion of expatriate managers and technicians in 1957-58. The combination of inflation, widespread trade licensing, and unrealistic multiple exchange rates reduced profitability in the traded sector, thus creating disincentives for the transfer of productive resources in that direction. As shown in table 2.la, the margin between official and market exchange rates became wider over time. The fall in production and exports eroded the tax base and reduced government tax revenue, both in real and nominal terms. At the same time, the rebellions created pressure for an expansion of military expenditures to restore political unity. The size of the budget deficit increased dramatically, and expenditure allocation for investment purposes was reduced substantially. The combination of supply shocks and pressure for higher budget expenditures accelerated during 1962-65. The political conflict with Singapore and Malaysia in 1963 led to a break of trade which further reduced exports. This was because Singapore and Penang were important entrepdt ports for Indonesian exports. As shown in table 2.1, the ratio of exports to GDP declined from 8.7 percent (1951-57) to 6.8 percent (1958-61), and then to 1.1 percent during the third phase (1962-65). The 1962-65 period was marked by an intensified contest for power among the president, the army, and the PKI. The contest was reflected in the making of policy, the exercise of patronage, and control over financial resources. As the struggle for budget allocation became more of a political dispute than a technical one, the political and administrative control of the Ministry of Finance declined. Since 1959 many ministries maintained extra budgets which were not reported in the budget document. The president and

37

IndonesidChapter 2

Table 2.la

Exchange Rates (rupiahRT.S. dollar)

Year

Market Rates

Year

Market Rates

1946 1947 1948 1949 1950 1951 1952 1953 1954 1955 1956 1957

2.7 2.7 2.7 3.8 3.8 3.8 11.4-1 1.7 11.4-1 1.7 11.4 11.4 11.4- 13.6= 22.7" 28.4-42.6b 56.8-78.1' 30.3' 37.9-56.gb 75.8-104.2' 36.P 45.0-67Sb 90.0-135.0'

1961

45.0' 45 .Oa- 72.Ob 200.W 45.0-173.0' 45.0-270.0b 900.0- 1 ,170.OC NA

1958

1959

1960

45.0'

45.0-72.0b 200.w

1962

1963 1964 1965

1966

1967

1%8

I969

NA

235.06 235.V 235.0' 235.0' 235.F 235.0' 235.06 235.0' 153.7' 326.0' 326.0' 300.1' 326.e 326.05 326.6

Source: International Financial Statistics, various issues. aF'rincipal export rates. bPrincipai import rates. 'Other import rates. 'The end-of-period national currency value of the SDR. 'Refers to end-of-period market exchange rates for countries quoting rates in units of national currency per

U .S . dollar.

'Refers to period averages of market exchange rates for countries quoting rates in units of national currency per U.S. dollar. NA: not available.

his cabinet ministries controlled the so-called Revolution Fund which was financed separately from the reported government revenues. The use of this fund was not reported to anybody but the president. During the later part of the term of President Soekamo, the government did not even bother to submit its draft budget proposal to get approval from the parliament. The supply of money and the inflation rate accelerated during 1962-64. In this period, both the money supply and the cost-of-living index roughly doubled every year, and by the end of 1965 they were doubling every few weeks. In May 1957 the government enacted Emergency Law No. 14 which freed Bank Indonesia from its obligation to maintain the 20 percent cover on gold and foreign exchange reserves. In 1965 Bank Indonesia merged with three of the four existing state commercial banks and the sole state-owned savings bank to form Bank Negara Indonesia. With these changes, the

38

Wing Thye Woo and Anwar Nasution

government budget deficits were automatically financed by credit from the central bank. The government made some effort to finance its budget deficits by selling bonds to the general public in the early 1950s. The Jakarta stock exchange, which was closed during the Pacific War, reopened on 4 June 1952. However, because of the public’s unfamiliarity with the stock exchange and because of their preference for other forms of assets, the government was not able to raise any significant amount of funds through the stock exchange. The erosion of real value by high inflation helped make financial assets, including government bonds, unattractive.

2.3 The Government Budget None of the coalition governments during the period of liberal democracy (1950-59) had an adequate mandate to take bold measures, nor survived long enough to implement an economic stabilization program. The key element in the contest for power within the coalition was the budget allocations. For example, one of the prime ministers, Wilopo, failed to prune back the personnel budget in 1952 because of strong political opposition from all over, including from his own party. The series of secessions between 1957 and 1965 led to greater military expenditures. In addition, President Soekamo, who had the dominant political position between 1959 and 1965 and who was not interested in economic issues, also demanded higher expenditures for his pet projects. Table 2.1 shows that the ability of the government to raise revenue declined from an average of 13 percent of GDP in 1951-61 to 5 percent in 1962-66. However, the average ratio of annual government expenditures to GDP rose from 14.7 percent (1951-57) to 18.4 percent (1958-61). This ratio dropped to 9.9 percent in 1962-66, in line with the decline in the government’s ability to tax.The ratio of budget deficit to GDP has risen from 1.5 percent in the first phase to around 5 percent in the second and third phases. The size of the government budget from 1950 to 1969 is given in table 2.2. Column (5) of the table, which gives the budget deficit as a percentage of gross expenditures, shows an exceptionally low percentage in 1956 (7.7 percent). This was due to extensive deregulations introduced by Professor Sumitro Djojohadikusumo, minister of economic affairs in the short-lived 1955 cabinet of Burhanudin Harahap. To combat inflation the government had liberalized imports and then released extra foreign exchange reserve to finance them. The increase in imports and exports raised revenue from the tax on foreign trade which, in turn, moderated the budget deficit. Table 2.3 depicts the major components of government expenditures and their rates of growth. The table shows that roughly half of government expenditures were for personnel and other current expenditures. Some of the

39

IndonesidChapter 2

lbble 2.2

Year (1) 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1%2 1963 1964 1965 1966 1967 1968 1969 1%9/70

Government Budget 1950-69

(in

billions of rupinhs)

Gross Revenue

Gross Expenditure

Total Budget Deficits

Deficits as 96 of (3)

(2)

(3)

(4)

(5)

7.00 11.80 12.30 13.60 12.50 14.20 19.20 20.60 25.30 30.50 49.82 62.20 75.00 161.00 282.10 960.77 13.14 60.2 149.8 45.9 251.6

8.70 10.60 15.00 15.70 15.40 16.30 20.80 25.60 35.50 44.30 58.34 88.50 122.1 329.80 681.30 2,526.32 29.43 87.55 185.3 58.6 342.7

- 1.7 1.2 -2.7 -2.1 -2.9 -2.1 - 1.6 -5.0 - 10.2 - 13.8 -8.5 -26.3 -47.1 - 168.8 -399.2 - 1,565.6 - 16.3 -27.3 -35.5 12.7 -91.1

-

- 19.5 11.3 - 18.0

- 13.4 - 18.8 - 12.9 -7.7

- 19.5 -28.7 -31.2 - 14.6 -29.7 -38.6 -51.2 -58.6 -62.0 -55.3 -31.2 - 19.2 -21.7 -26.6

Source: Bank Indonesia, Annual Reporr, various issues; World Bank sources. 1966 on: data in new rupiahs (1 new = 1,ooO old). 1969: Jan-Mar. 1%9/70 fiscal year that begins 1 April 1969.

current expenditures were subsidies for public utilities and state-owned enterprises. The next biggest expenditure item was for defense, which accounted for between 20 to (nearly) 47 percent of the budget throughout 1950-65. The share of defense outlays as a portion of total expenditures markedly increased during 1958-62 and again in 1965 to meet the security crises in these periods. Increases in the shares of current and defense expenditures reduced the shares of outlays for economic and social services, as well as for investment. Since 1963 the share of investment expenditure rose significantly, mainly to finance pet projects of the president such as the Senayan Sport Complex, the National Monument, the massive Istiqlal mosque, and the giant Ganefo complex. As shown in the lower section of table 2.3, the annual average growth rate of government expenditures increased from 15.8 percent in 1950-57 to 36.3 percent in 1958-61, and then jumped to 275 percent during 1962-66. During the first phase, the rate of growth of government expenditures was roughly equal to the average annual inflation rate at 15.8 percent. Since 1958, however, the rate of growth of government expenditure has consistently been higher than the inflation rate. The structure of government revenue, as shown in table 2.4, was heavily dependent upon taxing foreign trade (import and export taxes). As a result, tax revenue was quite sensitive to foreign trade activities and to government

40

Wing Thye Woo and Anwar Nasution

Table 2.3

Year

Major Components of Government Expenditures, 1951-66 (in millions of rupiahs)

Defense

Economic Service

Social Service ~

1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966”

3,269 3,032 3,892 3,327 3,937 4,075 6,051 11,085 14,071 22,431 41,607 57,438 97,900 195,200 1,009,100 2,260

(30.8) (20.2) (24.9) (22.0) (24.1) (20.4) (23.6) (31.4) (31.7) (38.5) (46.5) (46.6) (29.7) (28.7) (40.0) (9.3)

313 1,758 1,579 1,844 1,483 984 1,082 1,404 1,927 3,036 6,218 5,615 53,600 40,700 193,800

1,034 1,435 1,245 1,436 1,564 2,120 2,334 2,884 3,124 5,083 6,945 6,816 23,000 41,400 108,Mx) 18,400

Other Current Expenditure ~~

(9.7) (9.6) (8.0) (9.5) (9.6) (10.6) (9.1) (8.2) (7.0) (8.7) (7.8) (5.5) (7.0) (6.1) (4.3) (75.5)

Investment

~~~~

Total

~~

5,343 7,726 8,005 7,401 8,302 11,642 15,078 18,646 23,279 23,962 27,678 48,880 47,000 144,900 354,000

(50.3) (51.4) (51.1) (48.9) (50.9) (58.2) (58.9) (52.8) (52.5) (41.1) (30.0) (39.7) (14.3) (21.3) (14.0)

-

-

(6.3) 666 (7.1) 1,074 (6.0) 938 (7.5) 1,133 (6.3) 1,030 (5.9) 1,180 (4.2) 1,065 (3.7) 1,294 (4.4) 1,949 (6.6) 3,824 (7.9) 7,051 (3.7) 4,510 108,300 (32.8) 259,100 (38.0) 860,200 (34.1) 3,700 (15.2)

~

10,625 15.025 15,659 15,141 16.316 20,001 25,610 35,313 44.350 58,336 89,499 123,259 329,800 681.300 2,525.700 24,360

Averagc Annual Rates ot Growth

Defense Economic Cervicr Social \ervice Other current expenditure Investment Total expenditure Inflation rate

1951-57

193-6 I

I962 66

I95 I -66

10.8 23.0 14.5 18.9 81 15.8 24.4

55.4 64.2 34.0 14 I 76.0 36.3 36.6

150.5 225.6h 620.8 93.P 435.2 274.9 330.0

54.6 22.5’ 92.0 34.9’ 77.7 67.5 123.0

~

Source Biro Pusat Statistik

N o w . Dash indicatec that data were not available. Figures in parentheses are expenditures as a percentage of the total. “In new rupiahs beginning 1966 ( I new rupiah equals 1,000 old rupiahs). ”Avcrage growth ratea for 1962-65 ‘Average growth ratch for 1951-65.

policies toward the trade sector. Tax collection from the foreign trade sector was easy to administrate and faced practically no political opposition since the taxes were paid by foreigners (Dutch trading firms until 1958) or by the Chinese who dominated foreign trade activities. Revenue from the external sector was also raised by instituting different exchange rates for exports and imports. The result of multiple exchange rates was an upsurge in smuggling exports and imports. This illegal trade reduced the government’s tax revenue. At the same time, the high inflation rate and the inefficiency of domestic tax administration eroded revenue from personal and business income taxes. The share of nontax revenue increased steadily from 1959 on as there was an increasing contribution to the treasury of profit from the newly nationalized state enterprises.

41

IndonesidChapter 2

Table 2.4

Sources of Government Revenue 195169 (in percentages) ~~

~

Year

Personal Income Tax

Business Income Tax

on Income"

Tax on Consumption

Import Tax

Export Tax

Nontax Revenue

1951 1952 1953 1954 1955 1956 1957 I958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969

6.6 8.6 9.0 12.4 12.1 9.5 9.9 10.4 8.6 6.6 5.3 8.4 8.0 10.1 4.7 4.9 3.7 5.1 3.6

6.4 9.3 11.1 15.7 17.5 10.0 10.3 9.9 10.9 10.3 13.5 15.2 12.4 14.5 9.2 4.6 12.7 18.9 19.1

.3 .3 .4 .5 .7 .3 .3 .4 .3 .4 .4 .2 .6 .5 15.2 4.1 3.4 3.6 4.6

17.1 20.5 22.5 28.8 25.1 19.4 26.2 25.9 20.2 21.7 19.8 30.7 26.1 27.2 27.8 37.1 17.2 15.6 15.1

53.6 35.9 32.6 25.7 28.6 43.6 25.7 23.2 28.3 18.9 10.8 6.5 8.1 12.0 5.6 28.1 19.9 23.4 22.0

12.7 18.8 10.3 9.9 10.6 4.0 19.2 20.3 20.9 13.3 5.4 6.6 1.1 .3

3.3 6.7 13.2 6.9 5.4 13.2 8.8 9.8 10.6 27.4 29.4 30.3 43.4 34.7 37.5 6.2 1.5 2.6 .9

Other Tax

.o 15.0 12.5 11.7 7.4

~~

Development Revenue

.o .o

.o .o .o

.o .o .o .o 1.3 15.6 2.1 .3 .8

.o .o 29.1 19.2 27.2

Source: 1951-65: Bank Indonesia, Annual Reporr. 1951-52 through 1960-65 issues. 1966-69: Bank Indonesia, Indonesiun Financial Statistics. monthly bulletin, July 1969. "Includes land tax.

2.4

The Monetary Purges

The government of the newly independent Republic of Indonesia in January 1950 inherited two types of money in circulation. Each of the local governments of the Republic of Indonesia during the war for independence issued its own currency as legal tender in any temtory under its control. The Republican government issued currency known as Uang Republik Indonesia (URI) and the Dutch military authority issued currency notes popularly referred to as the Nica or "Red" money. At the end of 1949 the amount of URI of all types that was in circulation was estimated at about Rp 6 billion, and the amount of Dutch currency in circulation amounted to over Rp 3.3 billion. To bring monetary order and restore stability of the rupiah, the government decided upon the first monetary purge in April 1950. The Nica money and any Java Bank notes higher than Rp 2.50 were literally to be cut into two pieces. The right-hand side was to be exchanged for 3 percent government bonds and the left-hand side was to be exchanged for new Java Bank notes. Currency denominated at Rp 2.50 or less was excluded. Half of time and demand deposits over Rp 400 were to be exchanged for government bonds to be redeemed in installments within a forty-year period. The URI money with a denomination of Rp 50 or more could be exchanged for the

42

Wing Thye Woo and Anwar Nasution

new Java Bank notes at rates that varied from one region to another. URI money of smaller denomination could be deposited with the Bank Negara Indonesia. Monetary stability brought about by the monetary purge did not remain for long due to the way the government financed its budget deficit and to transmission of international inflation into the domestic economy during the Korean War boom. Lack of confidence in government domestic economic policy after the monetary purge raised the income velocity of money which, in turn, aggravated inflationary pressures. The lack of confidence was also due to an unequal distribution of the burdens of the monetary purge and inflation. The purge had sharply reduced the cash position of the business sector as shown by a 1.2 percent reduction in demand deposits in 1951 (table 2.5). To restore normal business operations, bank credit had to be expanded to the business sector. By September 1950, six months after the purge, the total money supply exceeded the amount in circulation before the purge. Inflationary pressures again rose to an alarming level in 1959. This was due to rapid growth of the money supply to finance the budget deficit, a Money Supply and Its Composition, 195(c67 (in billions of rupiah)

Table 2.5

Year

Currency (C)

Demand Deposit (DD)

Total Money (M)

1950 1951 1952 1953 1954 1955 1956 1957 1958 L959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969

2.58 3.33 4.35 5.22 7.47 8.65 9.37 14.09 19.87 26.38 34.08 48.54 102.28 175.46 447.00 1,811.00 14.36 34.10 74.68 115.70

1.73 1.71 2.26 2.27 3.64 3.59 4.02 4.82 9.49 8.50 13.76 19.10 33.62 87.90 278.00 761.00 7.85 17.37 39.21 67.74

4.31 5.03 6.60 7.49 11.12 12.23 13.39 18.91 29.37 34.88 47.84 67.64 135.90 263.36 725.00 2,572.00 22.21 51.47 113.89 183.44

Rate of Growth (96)

Percent of Total

Currency

Demand Deposit

59.9 66.1 65.9 69.7 67.2 70.7 70.0 74.5 67.7 75.6 71.2 71.8 75.3 67.0 61.6 70.4 64.7 66.2 65.6 63.1

40.1 33.9 34.1 30.3 32.8 29.3 30.0 25.5 32.3 24.4 28.8 24.7 33.0 38.4 29.6 35.3 33.8 34.4 36.9 38.2

Currency

Demand Deposit

Money Supply

Income Velocity ofMoney

47.4 29. I 30.6 20.0 43.1 15.8 8.3 50.4 41.0 32.8 29.2 42.4 110.7 71.5 154.8 305.1 692.9 137.5 119.0 54.9

10.9 -1.2 32.2 .4 60.4 -1.4 12.0 19.9 %.9 -10.4 61.9 38.8 76.0 161.5 216.3 173.7 931.5 121.3 125.7 72.8

30.2 16.7 31.2 13.5 48.5 10.0 9.5 41.2 55.3 18.8 37.2 41.4 100.9 93.8 175.3 254.8 763.5 131.7 121.3 61.1

14.5 14.4 12.6 13.1 10.9 11.8 10.9 7.6 6.9 9.4 8.1 13.1 16.1 15.2 14.6 26.8 23.0 24.1 18.3

-

Source: BPS, Srarisrical Pocketbook of Indonesia, various issues. Notes. Income velocity of money is defined as Y r / [ l / z ( [ M t - 11 money supply at time t .

New rupiah since 1966. Dash indicates that data were not available.

+ M I ) ] ,where Yr = GDP at time f and Mr

=

43

IndonesidChapter 2

reduction in exports as a result of the nationalization of foreign firms in 1957, as well as the closing of non-Indonesian Chinese rural stores. The black market value of the rupiah responded to the surge in inflation by dropping sharply from Rp 100 to Rp 150 per U.S. dollar between April and July 1959. On 25 August 1959 the government undertook a monetary purge for the second time. Bank notes with a value of Rp 500 or Rp 1000 had their face values reduced by a factor of ten. Ninety percent of all bank deposits of Rp 25,000 or more were frozen. The purge reduced the volume of currency in circulation from Rp 34 billion to Rp 21 billion overnight. The government also devalued the rupiah by almost 300 percent and increased taxes on trade. A 20 percent tax was put on all exports, and imports were classified into six categories with tariffs ranging up to 200 percent. The price stability brought by the second monetary purge and the tax increases again did not last long. Government expenditures on prestige projects financed by money creation started up again in 1961. To induce external aid as a way to contain the renewed inflation, in May 1963 the government implemented a stabilization program designed by the IMF. The program was a combination of budget austerity, relaxation of price controls, introduction of new foreign exchange regulations to encourage exports, and tight money policy. The effects of the stabilization program were dramatic but short-lived. The availability of foreign aid and loans increased imports and flattened out the general price index between June and August. In September 1963 Indonesia entered into a political conflict with Malaysia. The resulting increase in military expenditures widened the budget deficit so much that the stabilization program was undermined. Diversion of resources and transportation facilities from the civilian sector for military use reduced the productive capacity of the economy. In April 1964 the IMF stabilization program was formally abrogated. The government reverted to money-financing the budget deficits and reimposed extensive price controls. The money supply increased by 156 percent in 1964, 283 percent in 1965, and 764 percent in 1966. Under such circumstances, price controls were unable to curb the inflation rate. 2.5

The Foreign Sector

The Indonesian economy was heavily dependent on the export of a narrow range of primary products such as rubber, crude oil, tin, palm oil, and coffee. Rubber, for example, accounted for about 45 percent of the total value of Indonesia’s exports in 1955 and in 1960 (table 2.6). Export of petroleum amounted to 23 percent of the total value of exports in 1955 and 26 percent in 1960; copra accounted for nearly 4.5 percent of the total export proceeds in 1955 and 4 percent in 1960. The figures for tin were 6 percent in 1955 and in 1960.

Table 2.6

Export by Commodity as Rercentage of Total Exports (1951-69) ~~

~~

~

~

1951

1952

1953

1954

1955

I956

1957

1958

1959

Rubber Petroleum and petroleum products Tin and tin ore Copra and copra cakes Coffee Tea Tobacco Palm oil and palm kernels Other

50.6

44.9

33.0

30.5

45.4

40.1

36.0

34.6

45.0

14.3 6.3 10.3 1.7 2.8 1.8 2.9 9.4

20.5 9.2 6.0 1.9 2.5 2.5 3.3 9.3

24.5 9.9 7.8 3.7 2.9 3.0 4.1 11.2

26.2 7.1 6.7 4.6 4.6 3.8 3.5 13.1

22.8 6.3 4.5 1.7 3.3 2.9 2.8 10.3

25.5 7.2 5.1 3.4 3.4 3.3 3.5 8.7

33.3 5.6 4.4 3.0 3.1 3.5 3.1 8.0

37.4 5.0 2.8 2.4 3.3 4.0 3.7 6.9

30.7 3.9 3.6 2.0 3.2 2.6 2.6 6.4

Rubber Petroleum and petroleum products Tin and tin ore Copra and copra cakes Coffee Tea Tobacco Palm oil and palm kernels Other

7.8

7.2

5.5

5.8

5.8

5.0

3.6

3.7

69.1 0.5 6.1 0.2 0.4

76.5 0.5 4.2 0.2 0.3 0.1 1.5 9.5

78.7 0.4 3.4 0.3 0.2 0.1 1.4 10.0

84.6 0.4 3.4 0.3 0.3 0.2 1.4 3.7

79.5 0.4 3.2 0.2 0.2 0.1 1.3 9.4

80.8 0.4 3.0 0.4 0.3 0.1 1.3 8.8

86.8 0.2 2.4 0.3 0.2 0.1 0.9 5.6

86.8 0.2 1.5 0.2 0.2

0.1

1.3 14.6

Source: Bank Indonesia, Annual Report, various issues.

0.1

1.1 6.1

1960

1961

1962

1963

1964

I965

1966

1967

1968

1969

By Value 44.9 39.0

43.8

35.2

32.6

31.4

32.9

25.3

24.0

26.5

31.7 5.1 2.7 1.8 3.0 2.4 3.1 6.5

38.6 2.7 2.5 2.9 2.6 2.7 3.4 9.5

36.9 4.4 3.9 3.7 2.4 3.0 4.3 9.0

38.5 5.7 2.6 4.5 1.8 2.7 4.5 8.5

30.0 4.7 2.2 4.8 2.5 3.5 5.5 13.8

36.2 7.8 2.0 6.8 1.5 2.2 4.1 14.1

42.2 5.6 3.5 6.2 2.2 2.0 3.3 10.9

44.9 4.8 2.3 7.0 1.1 0.6 3.3 9.6

4.7

3.3

3.3

3.6

3.7

3.0

2.8

2.2

85.0 0.2 1.4 0.4 0.2 0.1 0.7 7.4

86.6 0.1 0.6 0.5 0.2 0.1 0.6 8.0

88.6 0.1 0.9 0.3 0.2 0.1 0.7 5.8

88.1 0.1 1.3 0.5 0.2 0.1 0.8 5.3

86.9 0.1 1.4 0.5 0.2 0.1 1.1 6.0

87.1 0.1 1.3 0.6 0.1 0.1 0.8 6.8

87.1 0.1 1.2 0.3 0.1 0.0 0.7 7.5

82.3 0.1 I .9 0.5 0.1 0.0 0.5 12.4

26.3 33.1 6.0 4.2 5.1 4.1 1.6 1.8 3.3 3.3 3.4 3.1 2.9 3.1 7.6 7.4 By Volume 4.9 3.6 3.6

84.7 0.2 1.0 0.3 0.2 0.1 0.7 8.0

86.1 0.2 1.0 0.3 0.2 0.1 0.7 7.2

86.2 0.1 1.3 0.4 0.2 0.1 0.7 7.5

45

IndonesidChapter 2

Table 2.7 shows an almost continuous deficit in the goods and services account during 1951-66. As has been pointed out earlier, since economic policies had been subordinated to political maneuvering, government expenditure could not be cut. The only adjustment measures that could be taken by the government were to reduce private consumption and private investment expenditures. To achieve the reduction in private spending, the government used a mixture of devaluation, tariff, trade restrictions, and export-inducement certificates. Devaluation gives a price incentive in domestic currency to factors of production in the traded sector. The price incentive is expected to encourage the movement of factors from the nontraded to the traded sector, thereby changing the structure of the economy to a more efficient one. The growth in output and the structural change in the economy, however, did not happen in Indonesia due to political instabilities and extensive dirigisme. Undoubtedly, quantitative import restrictions are more effective in restraining imports than a devaluation or a tariff, since their effects can be felt immediately. However, the resulting distortions decrease the efficiency of the economy and ultimately its growth potential. The first system of export-inducement certificates was introduced on 11 March 1950. In this system, exporters were required to sell their export proceeds to the Foreign Exchange Fund, LAAPLN (Lembaga Alat-Alat Pemabayaran Luar Negeri). In addition to the countervalue in rupiah at the official exchange rate (Rp 3.80/U.S.$), they received a certificate to buy foreign exchange equal to 50 percent of the rupiah value of the foreign exchange they turned in. The export certificates could be sold in the free market at a pegged price which was equal to 200 percent of the official rate. The buyers who wanted to use foreign exchange were required to turn in certificates equal to the full countervalue of the foreign exchange. In this system, therefore, exporters received 200 percent of the official exchange rate, importers paid 300 percent, and the government collected the difference as tax revenue. However, the constant need for additional government revenue made sustained export promotion efforts impossible. On 4 February 1952 the government imposed “additional export duties” on a wide range of export products to supplement the existing 8 percent general rate. The so-called strong products, such as rubber and copra, were subject to an additional 25 percent export tax, while the “weak products,” such as tin, palm oil, petroleum, and coffee, were subject to an additional 15 percent export tax. And finally, in February 1953 the export certificate system was abolished. The Burhanudin Harahap government introduced a major economic policy reform on 1 September 1955. To encourage exports, the authorities introduced a new export certificate, the BPE (Bukti Pendorong Ekspor) system. Export premium in the new system was given in foreign exchange, unlike in the old one which gave it in local currency. The percentage amount

Table 2.7

Indonesia’s Balance of Payments, 1950-69 (in millions of U.S. dollars)

Goods and services Merchandise Exports Imports Nonmonetary gold movement Foreign travel Transportation Insurance Investment income Government n.i.e. Miscellaneous Oil Companies Donations Private Official Total current transaction Private, exclusive banks Long-term loans Oil companies investment Other long-term loan movements Short-term loans Long-term improvements Short-term improvements Government and banks Long-term loans Short-term loans Liabilities to IMF & IBRD Other short-term loan movements Long-term improvements Short-term improvements Foreign exchange & gold holdings Foreign exchange holdings Monetary gold holdings Total movements of capital, foreign exchange & monetary gold Transirory items

1950

1951

1952

1953

1954

1955

1956

1957

1958

1959

174.0 250.4 629.1 -378.7 0.0 - 8.7 -21.5 - 10.0 -24.1 - 26.5 14.4 0.0 39.1 N.A. N.A. 213.2

164.2 401.6 1254.6 -853.0 0.0 -7.1 -5.8 -7.7 -42.3 -25.6 -43.4 - 105.5 0.4 N.A. N.A. 164.7

-265.4 -90.9 890.5 -981.4 0.0 -9.3 -11.1 - 12.1 -38.0 - 18.6 -32.3 -53.1 1.3 N.A. N.A. -258.2

- 135.8

-40.6 61.1 769.8 -708.7 0.3 -6.8 -7.7 -9.7 -70.3 -7.5 0.0 0.0 2.7 0.6 2.1 -44.1

102.2 251.0 877.4 -626.4 0.1 -7.1 -13.3 -12.4 - 107.6 -8.5 0.0 0.0 1.3 0.4 0.9 103.5

- 157.6

-42.7 23.2 425.6 -402.4 0.0 -5.2 -4.7 -4.6 -42.6 - 8.8 0.0 0.0 6.0 0.2 5.8 - 36.7

- 96.9

-48.6 845.7 -894.3 0.1 -9.7 -11.7 - 12.7 - 64.0 -11.0 0.0 0.0 1.4 0.6 0.8 - 156.2

83.6 961.2 - 877.6 0.0 -9.9 -20.2 -5.9 - 102.0 -42.5 0.0 0.0 280.4 0.3 280.1 183.4

25.2 234.8 816.9 -582. I 0.0 - 10.0 -53.7 -3.2 -71.5 -15.1 -50.1 0.0 16.7 0.2 16.5 41.8

-4.7 0.0 0.0 0.0 0.0

8.2 0.0 0.0 -0.1 0.0 0.0

7.9 0.0 0.0 0.0 0.0 0.0

-0.4 15.8 -16.2 -0.8 0.0 0.0

0.4 25.6 -25.2 0.7 0.0 0.0

0.1 2.3 -2.2 0.8 0.0 0.0

0.8 1.5 -0.7 0.0 0.2 0.6

5.7 7.8 - 2.0 0.0 0.2 1.2

1.4 0.0 1.4 0.0 -0.3

- 39.4

-78.4 0.0 0.0 0.0 0.0 0.0 0.0 - 181.1 -44.2 -295.7 37.5

-2.4 0.0 14.9 0.0 0.0

- 15.9

0.0 0.0 97.6 8.9 - 103.5

-2.5 0.0 0.0 22.9 - 1.7 0.0 0.0 12.1 2.7 35.1 1.7

- 3.4

0.0

-11.0 0.0 27.5 36.9 1.2 0.0 0.0 60.8 37.1 153.4 2.7

120.9 0.0 -9.1 8.7 0.0 -21.3 0.0 - 124.2 4.8 - 18.9 -22.8

0.0

7.6 161.4 0.0 0.0 0.0 0.0 116.0 0.0 0.0 194.7 18.4

0.0 0.0 0.0

0.0 0.0 0.0

76.6 -70.6 115.9 48.7

66.1 853.5 -787.4 1.7 -7.9 -13.1 - 12.9 -54.3 - 16.1 -44.4 -54.9 3.3 N.A. N.A. - 132.5 1.6 0.0 0.0 -0.1 0.0 0.0

-1.2 26.1 0.0 0.0 0.0

0.0 0.0 89.6 5.6 115.5 16.9

0.0 -22.6 55.3 44.1 0.0

0.0 0.0 0.0

0.0

0.0

0.0

0.0 11.2 - 17.9 0.0 0.0 - 170.9 2.2 - 171.8 -11.6

0.0

Goods and services Merchandise Exports Imports Nonmonetary gold movement Foreign travel Transportation Insurance Investment income

Government n i e . Miscellaneous Oil Companies Donations Private Official Total current transaction Privare, exclusive banks Long-term loans Oil companies investment Other long-term loan movements Short-term loans Long-term improvements Short-term improvements Governmenr and banks Long-term loans Short-term loans Liabilities to IMF & IBRD Other short-term loan movements Long-term improvements Short-term improvements Foreign exchange & gold holdings Foreign exchange holdings Monetary gold holdings Total movements of capital, foreign exchange & monetary gold Transitory items

1960

1961

1962

I963

1964

1965

1966

1967

1968

1969

-83.5 132.2 881.2 -749.0 0.0 -7.6 -50.7 -3.2 -67.0 -10.3 -76.9 0.0 24.5 0.8 23.8 -58.9

-523.5 -289.9 766.1 - 1056.0 0.0 -7.5 -65.6 - 3.4 -86.3 - 15.5 -55.2 0.0 60.0 0.3 59.7 -463.5

- 248.4

-227.8 54.4 616.4 -562.0 0.0 -5.6 -39.7 -3.2 -97.5 -50.8 -85.4 0.0 26.4 0.4 25.9 -201.5

-229.8 41.8 631.5 -589.8 0.0 -6.5 -44.1 - 1.2 -92.6 -52.7 74.4

- 248.0

- 132.0

- 282.0

39.0 634.0 - 595.0 0.0 8.0 -11.0 -60.0 -95.0 -52.0 -61.0 0.0 24.7 0.0 24.7 - 222.8

110.0 714.0 -604.0 0.0 - 18.0 -7.0 -63.0 -47.0 -35.0 -72.0 0.0 15.0 0.0 15.0 - 117.0

-35.0 770.0

-251.00 41.0 872.0 -831.0 0.0 - 14.0 - 3.0 -92.0 - 78.0 -23.0 - 82.0 0.0 26.0 0.0 26.0 -225.0

-361.00 2.0 995.0 -993.0 0.0 - 10.0 3.0 - 107.0 107.0 - 18.0 -118.0 0.0 47.0 0.0 47.0 -314.0

18.3 18.7 -0.4 0.0 0.1 1.8

- 11.7 - 12.5

25.1 25.0 0.1

18.0 18.0 0.0 0.0 0.0 0.0

- 16.0 - 16.0

- 10.0

0.0 0.0 0.0

9.4 9.7 0.4 0.0 0.0 0.0

0.0 0.0 50.0 0.0

2.0 28.0 66.0 0.0

3.0 -3.0 6.0 12.0 35.0 0.0

31.0 31.0 0.0 8.0 12.0 0.0

119.5 0.0 - 19.0 38.5 0.0 - 22.2 0.0 -42.6 - 25.4 68.8 -9.8

349.6 0.0 34.0 -42.0 0.0 -2.7

45.7 0.0 21.3 43.5 0.0 3.1 0.0 130.5 -0.7 255.9 -44.1

85.3 0.0 20.0 25.3 0.0 0.7 0.0 89. I 8.6 239.2 -37.7

78.8 0.0

5.8 0.0 0.0 222.7 0.0 -0.8

51.0 30.0

0.0 12.5

0.0 5.0 0.0 126.0 -9.0

187.0 4.0 0.0 0.0 0.0 0.0 0.0 -6.0 0.0 284.0 -30.0

179.0 12.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 229.0 -4.0

290.0 8.0 0.0 0.0 0.0 0.0 0.0 35.0 0.0 279.0 35.0

0.7

0.0 0.0 0.6

0.0 119.6 15.0 462.4 1.1

-25.4 710.8 - 736.2

0.0 -3.0 -51.8 -2.3 - 100.8 -23.8 -41.2 0.0 36.5 0.2 36.3 -211.8 12.5 11.5 0.9

~

Source; Bank Indonesia, Annual Report. various issues. Nore: N.A. = data not available. n.i.e. = not included elsewhere.

-

0.0 24.0 0.4 23.6 - 205.7

0.0 0.0 0.0

0.0 50.1 0.0 0.2 0.0 5.5 33.2 192.8 12.9 ~

-

0.0

0.0 0.0 0.0

0.0 258.3 -35.5 ~~~~

- 805.0 0.0 -4.0 - 9.0 -86.0 -63.0 -23.0 -62.0 0.0 28.0 0.0 28.0 -254.0

- 12.0

~~

-

-

48

Wing Thye Woo and Anwar Nasution

of export premium was divided into eight groups ranging from 2 to 20 percent. The BPE certificate could be sold in the free market within three months. In early September 1956 the free market value of the BPE certificate was 260 percent of the official rate. On 20 June 1957 the Monetary Board replaced the BPE certificate with the BE (Bukti Ekspor) certificate. The BE certificates, valued in rupiah at the official exchange rate (Rp 11.40), could be sold in the free market within two months. Purchase of the BE certificates was restricted to holders of import permits and licenses and of transfer permits. A 20 percent levy known as PBE (Pembuyuran Bukti Ekspor) was collected by the government from exports of goods and services. The market price of the new BE certificate immediately soared to 220 percent of the official rate. At the end of 1957 it rose to 250 percent and then to 332 percent in March 1958. In April 1958, however, the government ended the market determination of the BE exchange rate by fixing its value. The BE system was officially ended in August 1959. The vacillating treatment of the export sector described above characterizes very weil the inability of the government to conduct consistent economic policies within any reasonable time horizon. The on-again and off-again use of export incentives continued until the end of the Soekarno regime: the SIVA (Surat Zjin Vulutu Asing) export certificate in March 1962, the Additional Export-Inducement certificate and Special Export certificate in May 1963, and the SPP (Surat Pendorang Produksi) export certificate in April 1964. None of these export certificate schemes lasted long because the tension between taxing exports in order to reduce the budget deficit and promoting exports in order to attenuate the balance-of-payments deficit was never resolved.

2.6 The End of the Soekarno Regime The internal political struggle culminated with the abortive coup by the communists on the night of 30 September 1965. The political instability aggravated the economic instability. A combination of drastic devaluation and a third monetary purge on December 13 had no effect in controlling inflation. The increasing economic difficulties due to those measures speeded up the transfer of authority from President Soekarno to General Soeharto in the following year.

Appendix Table A2.1

Private consumption Government consumption Gross capital formation Exports Imports

1960

1961

1962

I963

311.4

335.8 (7.8) 42.0 (-6.9) 44.1 (43.6) 56.7 (9 0 ) 66.0 (34.7)

359.2 (7.0) 33.8 ( - 19.5) 40. I (-9.1) 51.8 ( - 8.6) 64.7 ( - 2.0)

345.0 (-4.0) 34.0 (.6) 30.6 (-23.7) 48.7 (-6.0) 47.5 (-26.6)

347.7 (.8) 40.0 (17.6) 34.8 (13.7) 54.5 (11.9) 51.7 (8.8)

356.0 (2.4) 29.0 ( - 27.5) 36.2 (4.0) 56.2 (3.1) 47.5 (-8.1)

390.2

412.6

420.2

410.8

425.3

390.2

470.1

1,335.1

3.208.8

7l.237.5

45.1 30.7 52.0 49.0

GDP at constant prices GDP at current prices Source:

Expenditure on GNP at Constant 1960 Prices (in billions of rupiahs)

Biro Pusat Statistik.

Note; Figures in parentheses are growth rates

1964

I965

I966

1967

1968

1969

350.8 1.5) 40.3 (39.0) 40.7 (12.4) 55.6

396.3 (3.8) 37.2 (3.9) 46.3 (39.5) 61.3 (10.5) 62.3 (6.9)

442.0

45.5 (-4.2)

381.8 (8.8) 35.8 (-11.2) 33.2 (~ 18.4) 55.5 ( - .2) 58.3 (28.1)

429.9

441.9

448.0

478.8

531.0

23.823.4

419.5

945.2

2.102.9

2,830.7

(-

( - 1.1)

(11.5)

42.0 (12.9) 52.0 (12.3) 70.0 (14.2) 75.0 (20.4)

Table A2.2

Rrcentage Distribution of National Income by Sector of Origin, 1953-69.

Sector

1953

1954

1955

Agriculture

57.0

58.0

56.0

55.0

36.2 7.9 4.5 4.2 3.0 2.3

36.9 7.9 3.8 4.9 2.9 2.2

36.2 8.0 3.8 4.6 2.9 1.4

36.6 7.1 3.5 4.5 2.8 1.4

1.1

.6

.9

.9

43.0

42.0

44.0

12.0 23.1

11.8 22.5

12.2 23.2

7.9

7.7

8.6

Peasant food crops Peasant export crops Estate crops Livestock Forestry Fishery Minus: export duties and statistical tax Nonagriculture Industryb Services Trade & financial Other Government

1956

1957

1958

1959

52.6

56.0

35.5 6.7 3.4 4.2 2.6 1.3

40.4 3.6 3.2

1966

55.4

58.7

38.0 5.7 2.6 2.8 2.3 4.0

35.8 9.1 3.6 3.5 .9 5.8

-

42.1 12.2 26.2 15.6 10.6 3.8

1961

55.2

53.9

47.9

59.0

57.9

3.0 1.6

35.3 7.7 3.1 4.7 2.2 2.0

34.4 7.2 3.3 4.8 2.4 1.9

27.3 7.2 2.9 4.7 2.5 3.4

42.8 5.4 2.6 3.6 1.4 3.1

37.7 6.8 2.3 3.7 1.2 6.1

.9

.8

-

-

-

-

-

45.0

47.4

44.0

44.8

46.1

52.1

41.0

12.5 23.4

14.7 23.1

11.0 19.3

9.1

10.2

13.7

16.5 23.2 11.8 11.4 5.0

14.4 27.2 15.3 11.9 4.5

15.4 31.0 18.7 12.4 5.6

11.1 27.3 16.1 11.3 2.5

5.0

Source: 1953-58: Biro Perancang Negara (1960, 101) 1959-66: BPS, SfafisticalPocketbook of Indonesia,various issues Nofe: Dash indicates that data were not available.

"Based on current prices. bIncludes mining, manufacturing, construction, and utilities.

1962

1965

1960

1%3

1964

1967

1968

1969

53.3

53.9

51.0

49.3

33.8 5.9 2.6 2.6 .8 7.7

35.5 5.4 2.3 3.9 .7 6.2

34.6 6.3 2.2 2.5 1.7 3.6

30.3 7.3 2.5 3.3 2.2 3.7

-

-

-

-

-

44.6

41.3

46.7

46.1

49.0

50.7

17.2 24.6 14.5 10.2 2.7

11.9 25.8 13.2 12.6 3.6

11.1 30.4 19.5 10.9 5.1

12.1 29.2 18.0 11.1 4.8

15.2 28.2 17.5 10.7 5.6

17.2 28.5 18.3 10.2 5.0

51

IndonesidChapter 2 Expenditure Composition of National Income (GDP), 1960-69 (in percentages)

lsble A2.3

1960 1961 1962 1963 1964 1965 1964 1961 1968 1969

private Consumption

Government Consumption

Investment

Export

Import

19.8 82.0 88.4 85.0 80.4 81.1 12.3 83.2 84.2 81.2

11.6 11.8 6.2 7.1 7.0 5.6 6.6 6.6 6.8 7.0

1.9 10.2 5.6 8.2 11.9 6.1 3.4 1.2 8.5 11.2

13.3 9.6 5.2 9.1 12.1 5.3 9.6 1.9 10.8 11.6

12.6 13.6 5.4 9.3 12.9 5.1 16.6 15.1 15.5 15.0

Source: Biro Pusat Statist&.

110' 96E' 829' CZO' LEI'E 909.02

IPI' 8L9' 9PO' 810' 612'

Em'

95Z' OL9'Z LLE'I I 09P'Z 098'91

WE

L8E' 8E9' ZZO' LL8'Z IZL'81

rn'

EPO'

LEI'

558. PPO' 610' 812' IEP. 9SZ' LL8'Z L69'Zl 260'2 OS6'51

091' 828' 8PO' 510' 8SZ. 6EE' ZEZ' E59'Z 811.01 098'1 LL9'PI

PPE'5I

KO' IPZ' LIL' 820' 86L'Z 655'51

PZO' E69'Z OIE'91

'

EW'

8PI. ILL' 6pO' €10'

SPZ' SIP' ZEZ'

EOI'E 8LZ'll PE9.Z

801'

EOE 858' 1EO' 8E9'Z 028'21 S91' 98L'

Em' EIO'

99Z' LSE' 812' 8E9'Z 1E1.6 596'1 619'PI

6EO' WZ'

PIS' KO' Z9P'Z 818'11

991' ZS8' PPO' 910' L9Z' 9PE' LOZ' 868'1 LIE'6 I L6' I ZEP'PI

OZO' ELI'

006' 9EO' IPOO'Z

E18'OI 691 ' 81L' LPOO' PIO' 882'

oop' 882' IlI'Z 695'6 OZL'Z 190'51

ZZO' 051' L68' PEO' E9L'I 19Z'OI 091' 029' LEO'

ZZO' 60E' 90E' POZ. 9Ll.Z ES6'8 518.1 t.9o'Pl

600' PPE' 696' 9EO' 6LE'l EZS'8 9PI' 8SP' LPO' €10' 662' 982' L91' 262'2 5ES.L 8E9' I ELL.ZI

0 ZW' 898'

IEO' EIO'I 660'8

IZI' 8ZP' 9mO'

ZIO' 612' 9LZ' P61'

EOE. I EEI'L 86E'I 896'1 I

Agriculture (million kgs) Paddy Maize Cassava Sweet Potatoes Peanuts Soya beans Rubber Coffee Tea

sugar Palm oil Mining (billion kgs) Crude petroleum Natural gas Tin in ore Coal Bauxite Manganese ore

1961

I962

1963

I964

1965

1966

1967

1968

1969

15.900 2.283 11.190 2.464 ,252 .426 .231 .019 ,034 ,643 ,146

17.111 3.243 11.386 3.680 ,261 ,397 ,218 ,013 ,047 .592 .142

15.276 2.358 11.679 3.070 .235 .350 ,218 ,019 ,039 .664 ,148

16.192 3.769 12.262 3.958 ,261 ,392 .234 .008 ,046 .655 .161

17.072 2.365 12.643 2.651 ,244 ,410 .228 ,020 ,044 .775 .I57

17.960 3.717 11.233 2.476 ,264 ,417 ,217 ,013 .612 ,174

17.398 2.369 10.747 2.144 ,241 ,416 ,205 ,019 ,034 ,666 ,174

20.032 3.102 11.268 2.282 ,273 ,389 ,215 ,014 ,042 ,603

20.464 2.271 10.845 1.904 ,257 ,416 .223 .016 ,047 .628 ,189

21.287 3.314 ,019 ,549 ,420 ,013

22.747 3.491 .Ol7 ,471 ,461 ,050

22.231 3.610 .013 ,650 ,506 ,001

26.851 3.524 .016

27.955 3.156 .015 ,281 ,688 .052

26.778 3. I62 ,013 ,320 ,701 .001

25.311 2.776 ,013 ,207 ,912 0

34.946 3.287 ,017 ,176 ,879 0

.446 ,648 ,005

,040

Sources: BPS, Statistical Yearbook of Indonesia, various issues. Bank Indonesia, Yearly Report, various issues

,188

43.086 2.604 ,017 ,191 ,765 0

54

Wing Thye Woo and Anwar Nasution

Table A2.5

Population of Indonesia, 1930-76 Population (millions)

Java-Madura Rural Urban Total Outer Islands Rural Urban Total Indonesia Rural Urban Total

Annual Growth Rate ( 0 )

1930

1961

1971

1976

1930-61

1961-71

1971-76

-

53.2 9.8 63.0

62.4 13.7 76. I

67.9 15.1 83.0

-

1.6 3.4 I .9

1.7 2.0 1.9

29.5 4.6 34.0

36.0 7. I 43.1

40.2 8.6 48.8

-

2.0 4.4 2.4

2.2 3.9 2.5

82.7 14.4 97.0

98.4 20.8 119.0

108.1 23.7 131.8

-

1.8 3.7 2.1

1.9 2.6 2.0

-

41.7

-

18.9 -

60.6

Note: Dash indicates that data were not available.

I .4

-

1.9

-

I .5

Table A2.6

Distribution and Size of Work Force in Indonesia, 19-76

(in percentages)

Indonesia

Java-Madura

Outer Islands

Sector

1930

1961

1971

1976

1930

1%1

1971

1976

1930

1961

1971

1976

Agricultural Manufacturing Trade Services Total Total number (millions) Annual growth (percent)

68.8 10.6 7.7 12.9 100.0 20.9

73.6 7.8 8.9 9.7 100.0 34.8

66.2 9.9 13.5 10.4 100.0 44.1

62.0 10.1 17.3 10.6 100.0 51

64.3 11.6 8.5 15.6 100.0 14.4

69.4 9.1 10.2 11.2 100.0 22.7

61.4 11.4 15.9 11.3 100.0 28.8

58.7 11.3 18.9 11.1 100.0 33.2

79.5 8.3 5.7 6.5 100.0 6.2

81.6 5.2 6.6 6.6 100.0 12. I

75.4 6.9 9.0 8.7 100.0 15.3

68.4 7.9 14.1 9.5 100.0 17.8

Source: David 0. Dapice (1980)

1.7

2.4

3.0

I .4

2.4

2.9

2.2

2.4

3.1

56

3

Wing Thye Woo and Anwar Nasution

The Political Economy Factors in Policymaking

3.1 Introduction The economic chaos of 1958-65 left such a deep impression on the institutional memory of the new Soeharto government that it has followed a “balanced” budget rule since 1968. In reality, this rule amounts to refusing to finance the deficit through money creation and limiting the deficit to the availability of foreign loans, which are officially described as foreign “revenue”. Another consequence of the chaos was the recognition that the exchange rate is an extremely potent policy instrument which can effect large-scale, economy-wide resource reallocation and income redistribution. The aim of this chapter is to show how the institutional memory has interacted with political economy factors to shape Indonesian economic management since 1965. In short, this chapter attempts to illuminate how the dead hand of history sets the constraints within which the invisible hand of economics operates. An emphasis on political factors should certainly be an important part of any discussion about external debt management. In the final analysis, it is these political considerations which determine whether a country will continue austerity measures in order to service its debts or will choose to repudiate its debts. In the case of Indonesia, an understanding of the political alignment and economic interests of the different constituencies is necessary in order to make sense out of the seemingly contradictory policies of the 1980s. For example, in March 1983 when it was clear that oil prices would be unlikely to soon recover, the Indonesian currency was devalued by 38 percent against the dollar in order to promote nonoil exports. The devaluation was followed by several fundamental reforms with the stated goal of liberalizing the economy in order to improve resource allocation which would then lead to higher long-term growth. The financial sector was thoroughly deregulated; the traditional instruments of monetary control-interest rate ceilings and credit rationing-were abandoned for more market-oriented financial instruments. The tax system was completely restructured: the tax code simplified, the tax base broadened, and marginal tax rates slashed. Economic liberalization was nevertheless not the overall thrust of Indonesian economic policies. At the time that the liberalizing measures were introduced, an avalanche of import restrictions was imposed. The new import restrictions generally took the form of monopoly licensing. The right to import was granted either to an individual, to a small group of firms, or to import-competing firms. The stated aim was the same as that for the liberalizing reforms-to achieve higher long-term growth. The means in this

’

57

IndonesidChapter 3

case were industrialization and the development of sectoral linkages. Furthermore, since many of the monopoly licenses were for important industrial materials which were inputs to many export industries, these new trade policies ran counter to the purpose of the March 1983 devaluation. Given that all of the policies adopted in 1983 were drastic departures from past norms, one may be puzzled by the absence of a clear direction in the new economic strategy. We attribute this curious mix of liberalizing steps and protectionist decrees to two factors: one, the existence of two groups of economic advisors who represent very different interests and ideological perspectives; and, two, the delicate role of the president in Indonesian politics as the supreme arbiter of the distribution of economic benefik2

3.2 The Dead Hand of History Indonesia is an authoritarian state which practices implicit corporati~m.~ President Soeharto has been in power since October 1965 and has not faced any serious challenges to his rule since the Malari riots of January 1974. Parliament is dominated by the government party, GOLKAR, which is an umbrella organization of civil service unions, trade associations, and youth, veteran, and women’s groups. GOLKAR was created by Soeharto and hence it does not formulate the national agenda to be implemented by the president. The opposition parties are in such disarray that they actually receive annual government subsidies in order to maintain their operations. The plight of the opposition parties is as much due to government intimidation as to their record for ineffectiveness and squabbling during the early days of the Republic. Although Soeharto is a former army general and the army is the backbone of GOLKAR, Indonesia is not a military state. The military does not have “important centers of power independent of the central authority” and the president does not engage in “a continuous process of bargaining with other officers” (Sundhaussen 1978, 78). The army supports Soeharto but it, like GOLKAR, does not set the national agenda. We pay particular attention to Soeharto’s preferences because of the unusual degree of decision making concentrated in, and the broad executive powers granted to, this one individual. Since the authoritarian nature of the state renders the medium-run accountability of the president’s actions to be very low, and there are substantial differences in interests and ideology within the implicit coalition, Soeharto must be considered an independent force rather than merely the compromise byproduct of competition among the elite groups. One way of describing the economic policymaking process is the following: The different lobbies and advisory groups propose policy initiatives, and the president adopts those which are either compatible with his innate preferences or vital to maintaining his position as the overarching

58

Wing Thye Woo and Anwar Nasution

patron. It is, therefore, only natural that we begin the task of explaining the choice of economic strategy in Indonesia by studying the events and forces which shaped Soeharto’s views on economic management. Indonesia was in a state of total economic chaos when Soeharto assumed formal executive power in March 1966. The economy had grown unsteadily at an annual average rate of 1.8 percent between 1960 and 1965. Since the annual population growth rate was 2.5 percent, per capita income declined over this period. The inflation rate was also accelerating. It was 128 percent in 1963, 135 percent in 1964, and 595 percent in 1965. In addition, Indonesia had defaulted on its external debts in 1965. The economic stagnation was inevitable given the many microeconomic distortions and the huge macroeconomic imbalances. An exchange rate that was overvalued for a prolonged period of time had caused the decline of the export industries, the most productive sector of the economy. The import-competing industries, kept alive by high tariffs, were notoriously inefficient. Extensive price controls which discouraged production, a complex tax system with high marginal rates, and a corrupt administrative structure also helped to cripple economic growth. The high inflation rate was the result of the monetization of the swingeing budget deficits and the government-ordered extension of central bank credit to state enterprises and to favored members of the private sector. Central government expenditure was 103 percent larger than its revenue in 1963, 140 percent in 1964, and 163 percent in 1965. The rapid expansion of money (Ml) in circulation reflected this fiscal mismanagement. The money stock increased 94 percent in 1963, 156 percent in 1964, and 283 percent in 1965. The inflation rates in 1963 and 1965 were actually higher than the money growth rates, 129 percent and 595 percent, respectively, producing a situation where money was rapidly losing its function as a medium of e~change.~ On 3 October 1966 the new Indonesian government announced a stabilization and rehabilitation program5 The central plank of the stabilization program was an unequivocal commitment to end the printing of money to finance government budget deficits. The government pledged that it would adhere to a “balanced” budget policy from 1967 on. In official Indonesian usage this meant that government expenditure was limited to the sum of domestic revenue and external loans. Since the Indonesian government was shut off from the external private credit market in 1966, such a budget rule was effective in limiting government expenditure because of the inelastic supply of foreign concessionary loans. Furthermore, central bank credit to state and private enterprises was severely curtailed and placed under strict supervision. The trend toward a demonetized economy was immediately reversedprices in 1967 rose only 112 percent when the currency stock went up by 132

59

IndonesidChapter 3

percent. After two years of balanced budgets and continued credit restraint, the restoration of confidence in the rupiah was complete. In 1969 prices rose only 17 percent in the face of a 61 percent increase in the currency stock. Inflation reached the extraordinarily low rate of 4 percent in 1971 despite a nearly 30 percent rise in the amount of currency in circulation. Soeharto obviously understood his first lesson in macroeconomic management because the balanced budget principle has never been compromised during his administration. To Soeharto, a prudent fiscal policy is the prerequisite for preventing runaway inflation, and fiscal prudence is understood to mean a “balanced” budget. The rehabilitation part of the October 1966 program was to allow market forces a greater (but by no means unrestricted) role in resource allocation. To reverse the trend toward subsistence fanning in the agricultural sector and, in particular, to improve the balance of payments, the government devalued the rupiah from 10 rupiahs/dollar to 100 rupiahs/dollar.6 Price controls on many commodities were removed to encourage their production. The government subsequently showed no reluctance to allow the rupiah to depreciate according to market conditions. The rupiah stood at 326 to the dollar at the end of 1968. Real nonoil exports (measured in foreign purchasing power) increased by 12 percent between 1965 and 1968.7 This is an underestimate because there was a large upsurge in smuggling to evade unauthorized taxes levied by corrupt officials who attempted to scoop some of the gains of the devaluation for themselves (see Penny and Thalib 1967). This export growth is still impressive, however, because this was a period of slow growth in the industrialized economies.* The surge in agricultural exports came from increased production by the small holders rather than by the estates. The production of farm nonfood crops increased by 19 percent compared to a 7 percent increase in estate output (World Bank 1975, table 2.2). The rehabilitation program really took effect in 1968 when real GDP increased by an unprecedented 11 percent. There was an annual average growth rate of over 8 percent from 1969 until the 1973 OPEC price increase. The distributional aspects of the devaluation are clear. Since the prices of commodities are set in dollars, a devaluation of the rupiah translated directly into increased income for the small agricultural producers. It would belittle Soeharto’s ability if we were to believe that he was not aware of either the political economy of the 1966 devaluation or its effectiveness as a policy instrument. Since then, devaluation has been a frequently used instrument; the rupiah was devalued in 1978 even when there was no balanceof-payments crisis ! Competitive exchange rate management alongside the balanced budget are the two constants in Soeharto’s economic policymaking, and they directly result from the memory of the economic chaos of the Soekamo years and of the 1966 economic program.

60

Wing Thye Woo and Anwar Nasution

3.3 Agrarian Radicalism, Separatism, and the Peasant Background of Soeharto It would be helpful at this point to use the issue of exchange rate management to lay out the political environment within which decisions are made about the goals of economic policies and the choice of instruments with which to implement them. The considerations that render Soeharto immune to a commitment to a strong (read “overvalued”) currency are the same ones that dictate budgetary choices during times when total spending has to be restrained to match lower revenue growth. Soeharto’s attitude to the exchange rate after observing the distributional impact of the 1966 devaluation is rooted in two political concerns and a strong personal commitment. The first political concern is to avoid conditions favorable to the resuscitation of the PKI. In 1965 Indonesia had “the strongest communist party outside the communist bloc, with a membership of over three million and affiliated mass organizations of farmers, workers, women, and students that claimed over 20 million followers” (Dake 1973, 1-2). The fact that its members were largely landless peasants in Central and East Java indicates that any prolonged impoverishment of the rural heartland could lead to the resurgence of the PKI. Soeharto is personally opposed to the communist ideology, and he has had two bloody encounters with the PKI. The first run-in was the internecine Madiun Affair in 1948 when the PKI tried to hijack the leadership of the independence movement. The second showdown was the 1965-66 aftermath of the abortive communist coup of September 30.9 The official casualty figure for the latter event is half a million, though unofficial reports of the period put it at one million. The political lesson is clear: the spectre of communism can be exorcised only by improvements in the lives of the rural population. Soeharto’s second political concern is the diversity of the ethnic groups which live in the 13,000 islands which make up Indonesia-“There are over three hundred different ethnic groups in Indonesia, each with its own cultural identity, and more than two hundred and fifty distinct languages are spoken in the archipelago. Religious beliefs, too, are varied” (Geertz 1963, 24). This diversity resulted in numerous secession attempts in the 1950s. Soeharto was personally involved in squashing a rebellion in Sulawesi in 1950. The sense of alienation in the Outer Islands has not been helped by the fact that the inner circles of the government have been dominated by the Javanese since independence in 1949. It has, hence, been necessary to assuage feelings of discrimination among the non-Javanese by proffering to them tangible economic benefits. The government has had to do more than raise living standards as it did in rural Java, it has also had to make it clear that inter-island (regional) equity is a primary goal of the Soeharto government. This has meant that spending for regional development has

61

IndonesidChapter 3

been a high priority, reinforcing Soeharto’s readiness to devalue the currency. Most of the import-substitution industries, whose products are practically nontradables because of cost inefficiency, are located in Java, and a devaluation always turns the regional terms of trade in favor of the agricultural-commodity-exporting Outer Islands. On the personal level, Soeharto has never downplayed his peasant origin.” From the earliest days of his presidency, he has consistently emphasized the need to improve the living standard in rural areas. Even if one were to dismiss Soeharto’s avowed commitment to alleviating rural poverty as political opportunism, the critic must concede that all this symbolism testifies to the importance which Soeharto places on rural development.” It may have been more than “mere” symbolism that the agricultural sector was singled out for attention in the first development plan, Repelita 1. This agricultural emphasis has been continued in subsequent development plans. It is the combination of the three factors-oncem about the traditional peasant base of PKI, experience with secessionist movements, and a personal commitment to rural development-that explains a great deal of the observed allocation of government expenditure in particular, and the conduct of economic management in general. The first and third concerns imply the need to improve the absolute level of the standard of living in rural areas, while the second concern implies the need to improve (or bring to par) the relative standard of living in the Outer Islands. Together these three factors focus attention on development of the agricultural sector, that is, rice in Java and agricultural commodities in the Outer Islands. l2 The general INPRES programs (development funds funneled directly at the president’s discretion through decrees, Znstruksi Presiden, to the local level), fertilizer subsidies, irrigation projects, and maintenance of a competitive exchange rate are the most explicit manifestations of the high priority placed on developing the agricultural sector.

3.4 Economic Nationalism, Political Patronage, Priburni-ism, and the Military Background of Soeharto Attention to its agricultural sector does not mean a strong overall agricultural bias in Indonesian economic policies. Rather, the bias is toward the industrial sector. There is widespread popular sentiment for, significant intellectual support of, and powerful special interests clamoring for the rapid development of a large and diversified industrial base. The president himself is sympathetic to this view because of his experience in the war for independence. As we explained in chapter 1, Dutch economic policies were seen as designed to impose a plantation economy on Indonesia to serve the needs of Dutch manufacturing industries for raw materials and to drain

62

Wing Thye Woo and Anwar Nasution

Indonesia of its wealth through profit repatriation. In reaction, the generation of 1945 regards industrialization as the key to economic prosperity because technological advancements were, supposedly, less likely to occur in the agricultural sector. They see the repatriation of profits to Holland as an additional reason for the absence of investment in manufacturing. So it is understandable that economic nationalism in postcolonial Indonesia took the form of state support for industrialization programs and intolerance for foreign ownership of capital (except in extractive industries in the Outer Islands where the capital requirements were immense). The policy translations of economic nationalism are high trade barriers to induce the development of a manufacturing sector and foreign investment laws (which still are) stricter than those of neighboring countries. It is Soeharto's economic nationalism which explains the vacillating attitude toward laissez-faire. The simultaneous introduction in 1983 of liberalizing measures, such as financial deregulation, together with interventionist measures, such as additional nontariff barriers, illustrates this ambiguity. Setting up inefficient industries behind trade barriers may reflect more than the fact that the army is the bastion of economic nationalism or the belief that industrial development is the long-run solution to rural poverty. An institutional legacy from the past may have been equally important. During the war for independence, the various army units were selfsupporting by circumstances, as there was no functioning central government to make budget allocations. In the Soekarno years, army generals were expected to continue supporting their troops by raising outside revenue to supplement their budget allocations. Joint business ventures involving senior army generals and the private sector were common. This practice has been expanded under the Soeharto regime, and the use of army personnel in business management has been justified by the doctrine of dwifungsi.'3 The imposition of trade barriers to give a monopoly position to manufacturing enterprises with army connections thus serves the dual purpose of catering to the army's economic nationalism and providing additional funds to the armed forces.14 The second purpose is an important reason Indonesian manufacturing industries are oriented toward internal rather than external markets. Competition in external markets may make production more efficient, but it also makes funding for the army more uncertain. This also helps to explain why quotas rather than tariffs are the favored form of import restriction.15 The granting of quota rights is a faster way of transferring funds to selected groups like the army, bypassing the laborious budgetary process. The result is that the budget understates the revenue and expenditure of the public sector. The widespread use of monopoly import licensing since 1982 serves in most cases the same transfer-of-funds function as quotas. Since the license holders are usually family members of government officials and political supporters, it is clear that the trade barriers were erected more for

63

IndonesidChapter 3

rent-seeking purposes than for infant industry reasons. l6 What is being revealed is the political patronage side and not the economic nationalist side of the Indonesian state. It has been argued that this privatization of public funds was actually a form of privatization of public works rather than the fruits of political patronage. According to Bustanil Arifin, the minister of cooperatives, the profits of the monopoly import licensing system are “being used to develop small industries, build mosques and churches and run public health program^."'^ The point is that given the considerable overlap of personnel in the public and private sectors (a practice sanctioned by dwifungsi), the line between these two sectors is blurred in Indonesia. Since private firms are expected to contribute for nonbudgeted public projects, it is perhaps not surprising that when the biggest cement company, Indocement, ran into financial difficulties in July 1985, the government reciprocated by spending over U.S. $325 million to bail it out. The ways in which Soeharto’s background has influenced his style of economic management have presented occasional problems for the budget. Perhaps, being a military man, it is natural that Soeharto shows impatience at the bureaucratic implementation of government programs. What is surprising is that instead of trying to streamline the bureaucracy, he regards its inefficiency as endemic. Two events reveal this attitude very clearly. Through INPRES since 1969, a sizable amount of development spending is directly channelled to the village and county levels, bypassing the usual disbursement mechanism. More recently, in 1985, the entire Indonesian customs service, well known for its corruption and delay, was put on indefinite paid leave. A private Swiss firm (SociCtC GCnCrale de Surveillance S.A. of Geneva) was hired in its place to clear the goods at their foreign ports of departure. In the eighteen years of his rule, Soeharto had plenty of time to revamp the customs service but he chose not to. And when he did respond to its inefficiency, the action reflected his pessimism about bureaucratic reform. While Soeharto highly valued the advice of his capable economic technocrats, he saw their penchant for detailed financial assessment and accountability as a drag on the pace of development. Soeharto’s appointments of, first, General Ibnu Sutowo, and then Technology Minister Habibie, as the czars of national industrialization, arise from his perception that the problems of a desperately poor country like Indonesia are obvious and what is needed are quick, decisive actions initiated and enforced by a capable and dedicated individual. Quick actions necessarily dictate that the individual be unconstrained by the usual bureaucratic checks which, after all, are meant only for men of lesser talent and dedication. Soeharto’s practice of allowing the “dynamiser” a free rein follows from his own experiences as the Central Java regional commander in the 1950s. Again, it was a time when military commanders had to generate substantial

64

Wing Thye Woo and Anwar Nasution

revenue to supplement the budget allocation in order to maintain the army units, the force that was preventing the disintegration of the new country into regional republics. Soeharto met this challenge with a series of very successful business ventures. In addition to the usual transportation and trading activities, Soeharto’s military command was the only one with a Development Contribution Fund financed by levies on the local business community. Soeharto’s style of granting broad discretionary powers to his industrialization czars resulted in the Pertamina crisis in February 1975. General Ibnu Sutowo, who in 1966 became head of the state oil company, Pertamina, expanded its nonoil investments tremendously. To finance this proliferation of activities, Pertamina borrowed heavily in the international credit market. A significant portion of Pertamina’s external debt was in short-term borrowing because the technocrats resented the evolution of Pertamina into an independent development agency and had, with the help of major aid donors, pressured Soeharto to restrict Pertamina’s borrowing in the medium-term credit market. It was this borrowing in the short-term market which precipitated the Pertamina crisis. In 1973 and 1974 Pertamina found itself paying higher and higher interest payments on the debt it was continually rolling over. (The interest rate rose because the central banks in the industrialized countries were stomping on their monetary brakes in order to dampen aggregate demand to offset the supply-side inflation caused by the OPEC-1 price increases.) The increased debt-service burden proved too much for Pertamina. In February 1975 it defaulted on a short-term loan, and Bank Indonesia had to take over Pertamina’s debt. Bank Indonesia’s foreign exchange reserves fell from $1.6 billion in January 1975 to $0.5 billion in September 1975, while at the same time its foreign liabilities rose from (nearly) zero to $0.8 billion, a reflection of the tremendous external borrowing required to meet Pertamina’s obligations during those seven months.” The big diversion of resources led to the scaling back of development expenditure in 1976/77 and 1977/78. It would be wrong to attribute the fiscal problems arising from the Pertamina affair entirely to Soeharto’s predilection for getting things done quickly. Special interests and ideology were also important in shaping events. The huge revenue generated by rapid development of the petroleum sector was very important to the Soeharto government in the early days of its administration because it obviated the unpopular steps of raising taxes to pay for routine government expenditures. l9 Since the Ministry of Finance had no knowledge of the amount of oil taxes and royalties that Pertamina had collected on the Ministry’s behalf, Pertamina retained part of the revenue for extrabudgetary activities. These activities helped to consolidate Soeharto’s power base by channelling resources to key constituents to meet what Karl Jackson (1978b) has called their “nonservice goals.” It is highly likely that

65

IndonesiaChapter 3

the armed forces, which had their official budget capped in order to convince external aid donors of the commitment by the Indonesian government to development, were a big recipient of extrabudgetary allowances from Pertamina. Given these extra functions, it was no surprise therefore that Sutowo adopted a management style which “ensured that few people apart from himself even had a rough overall picture of the finances of operations of the company” (McCawley 1978, 5). Nothing illustrated this attempt at obfuscation better than the fact that Pertamina had six uncoordinated accounting departments.*’ Ideology also plays an important role in explaining the Pertamina affair because there was a sizable group of intellectuals who concluded that the establishment of more quasi-state conglomerates a la Pertamina was essential for political stability. An underlying resentment among many indigenous Indonesians (pribumis) was that the Chinese Indonesians (peranakan and totok) wielded economic power disproportionate to their 3 percent share of the population. It was widely felt that this state of affairs originated from the victimization of the pribumis by Dutch colonial policies.21 This resentment against the Chinese led to occasional mob destruction of Chinese property. These intellectuals were pessimistic about the ability of laissez-faire to change this inequality drastically and quickly enough to be politically acceptable. They proposed that Chinese domination of business be reduced by launching large, state-sponsored enterprises, each headed by one of the small number of talented pribumi entrepreneurs in existence. By regarding these Indonesian zaibatsu as holding their capital in trust for the pribumis, the share of indigenous ownership of capital would be considerably increased very rapidly.22 This method of defusing racial tension received enthusiastic support from the economic nationalists within the army, the key constituency in Soeharto’s regime, making it easier for Soeharto to follow his preference for quick action.

3.5 The Embodiment of the Different Concerns at the Policymaking Level The primary reason Indonesia sometimes pursues a contradictory mix of liberalizing and protectionist policies is because of these ideological, pecuniary, and personal elements working themselves through two groups of contending presidential economic adivsors, popularly referred to as the technocrats and the technician^.^^ The technocrats are mostly economists of neoclassical persuasion who work at the Ministry of Finance and the National Planning Body (BAPPENAS). Their acceptance of the comparative advantage principle leads them to emphasize the development of nonoil export industries, particularly agricultural commodities and labor-intensive manufactured goods. This has meant a favorable treatment of the agricultural sector because it supplied 82 percent of nonoil exports in 1970 and 75

66

Wing Thye Woo and Aowar Nasution

percent in 1980. Exchange rate devaluations, rather than the removal of trade restrictions on imported inputs, are used to promote exports because the technocrats control the ministries that oversee macroeconomic policies but not the Ministry of Trade and the Ministry of Industry which have authority over trade restrictions. The technocrats do not have much of a domestic constituency outside of the universities. Consistently strong sources of support for their marketoriented policies are the foreign aid donors, the World Bank, the International Monetary Fund, and IGGI.24This foreign backing helped in the early years to establish the role and power of the technocrats because of Indonesia’s dependency at that time on foreign concessionary This aspect was very well illustrated by the strategy the technocrats adopted to rein in Pertamina when it was intruding upon the economic policymaking turf.26In March 1972 the minister of finance chose to enter into another standby agreement with the IMF, even though in the agreement document the IMF had concluded that: [the] proposed fiscal and credit policies, the recent depreciation in the effective exchange rate of the rupiah, the expected substantial increase in net receipts from crude oil exports should make possible the achievement [of the Indonesian goals which were] to achieve a higher rate of economic growth in conditions of relative price stability, and to increase net international reserves. (emphasis added) The result of this extreme aversion of the Ministry of Finance to the possibility of a balance-of-payments crisis led to the IMF setting a ceiling on medium-term external borrowing of $145 million for 1972/73 as part of the standby agreement. A decree was issued in October 1972 requiring all state bodies to seek approval from the Ministry of Finance before contracting medium-term foreign loans. When Pertamina ignored this decree, the United States (the biggest donor) suspended its aid. Pertamina then started borrowing in the short-term market to finance its long-term investments. As noted earlier, this switch led to the 1975 Pertamina debt crisis, after which the firm was taken over by army personnel sympathetic to the technocrats. Despite the big decline in the importance of foreign aid after the oil boom, the technocrats have maintained their influence on economic policies because of their proven competence. They designed and implemented the 1966 stabilization program, restructured Pertamina’s debts, and introduced a new professionalism into economic management. This explains why the president remains their patron. The second group of economic advisors, the technicians, is an amorphous collection of technicians-turned-managers, military advisors, and economists with structuralist inclination^.^^ The technicians are united by their common belief in the general validity of the infant industry argument and by their common rejection of foreign capital ownership. They see state enterprises

67

IndonesidChapter 4

like Pertamina (until its downfall) as vehicles to achieve these two objectives.’’ This position allies the technicians with members of the intelligentsia who see state enterprises as the way to counterbalance Chinese domination of the corporate sector. The technicians’ control of the Ministry of Trade has allowed them to encourage domestic production of manufactured goods, including airplanes. Given the strong sense of economic nationalism in Indonesia and the widespread belief that only industrialization (regardless of whether it is import-competing or export-oriented) holds the key to a higher standard of living, the technicians enjoy popular support among the Indonesian elite. Furthermore, their import-substitution industrialization has won them the support of the army, the most powerful constituency in the country. Thanks to the dwifungsi doctrine, the expansion of state enterprises translates directly into more managerial positions for senior military personnel. It must be noted that since most of the import-competing industries were set up in urban Java, the higher prices of manufactured goods represented an implicit tax on residents of the rural sector and the Outer Islands. In looking at the political setting within which policies are chosen, we have identified an important political coalition of technocrats, Outer Islanders, and rural residents which favors a political package emphasizing the maintenance of a competitive exchange rate. The fact that there exists a strong institutional memory about the economically destructive effects of an overvalued exchange rate means that the government is naturally disposed to the arguments for a competitive exchange rate. Since a debt crisis occurs when a government runs out of foreign reserves, this emphasis on avoiding prolonged exchange rate overvaluation reduces the probability of a debt crisis by keeping the (foreign exchange earning) nonoil export sector healthy and capital flight low. We shall show in subsequent chapters how these political and economic factors have influenced the setting of economic policies and, hence, the performance of the economy.

4

The Fiscal System

4.1 Introduction

The two arguments we are developing in this monograph are that appropriate exchange rate management was fundamental to why a debt crisis did not appear during 1982-84 and that the exchange rate policy was heavily influenced by political considerations. The aim of this chapter is to test the

68

Wing Thye Woo and Anwar Nasution

validity of the second argument by examining the fiscal system to see if its functioning is in accord with the political concerns to prevent impoverishment of the rural sector and to show equitable treatment of the main islands. We picked the fiscal system because it is largely controlled by the technocrats, and we have claimed in chapter 3 that the technocrats favor an economic strategy which has as its side effects (if not as its aims) the alleviation of rural poverty and the reduction of regional differences. We will discuss the control and performance of the overall fiscal balance in chapter 7 where the subject of external debt management is explored in detail. A comparative analysis of external debt due to cumulated budget deficits is done in chapter 8. It is more natural to examine the accumulation of external debt due to fiscal deficits together with the management issues involved.

4.2 The Revenue Structure, 1969-83 In assessing the tax system before the December 1983 tax reform, it is important to keep in mind that the collection of taxes did not usually correspond closely to the tax laws. The shortage of competent personnel made enforcement of the highly complicated Indonesian tax code impossible. The result was “that the tax revenue targets published in the budgets determined the amounts which administrators felt obliged to collect” (Booth and McCawley 1981b, 136). Since the amount of taxes actually paid was nearly always a negotiated outcome, annual changes in income taxes bore little relation to the marginal tax rates. In short, the vertical and horizontal equity aspects of the income tax system cannot be accurately gleaned from the pre-1984 tax code. Furthermore, the tax burden owed to the state was consistently understated by the total amount of revenue collected. This is because of widespread petty corruption and occasional unauthorized levies on business transactions by administrative and military personnel. The data in table 4.1 show the revenue structure of the central government, and they differ from the official classifications in three ways. Our total revenue figures from fiscal 1969 to fiscal 1971 are greater than the official figures by the amount of IPEDA because IPEDA was not included in central government revenue figures prior to 1972. (IPEDA stands for Iuran Pembangunan Daerah which means Contribution to Regional Development.) IPEDA is revenue which belongs to the provincial authorities and is collected on the provinces’ behalf by the central government. The “tax on nonoil income” category differs from the official definition in that it covers only personal income, corporate income, and withholding taxes; the “other taxes” subcategory under the official definition has been added to the “tax on nonoil domestic consumption” category in the table.’ The third difference is that we have constructed a “tax from oil sector” category by combining official subcategories-the “tax revenue from oil corporations”

Table 4.1

Revenue Structure, FY1969 to FY1983 (in billions of Rupiahs) 69/70

Total revenue Tax on nonoil income Tax on oil and gas Tax on nonoil domestic consumption Tax on international trade Tax on property, IPEDA Nontax receipt Share of total revenue (%) Tax on nonoil income Tax on oil and gas Tax on nonoil domestic consumption Tax on international trade Tax on property, IPEDA Nontax receipt Share of nonoil revenue (%) Income tax Domestic consumption tax International trade Property tax Nontax receipt Indicative ratios (%) Total revenue/GDP Oil and gas tax/ GDP Nonoil income tax/GDP Property tax/GDP Nonoil income tax/ domestic consumption tax Nonoil export tax/ total revenue Memo items Tax on nonoil exports Nominal GDP GDP deflator (1980= 100)

70171

71/72

72/73

73/74

74/75

75/76

76/77

77/78

78/79

79/80

80181

81/82

6.696.8 10,227.0 12,212.6 12,418.3 14,432.7 736.5 1,045.3 1,279.3 1,605.2 1,784.3 4,259.6 7,019.6 8,627.8 8,170.4 9,520.2

251.6 42.9 65.8

354.7 52.8 99.2

440.0 67.4 140.7

590.6 84.6 230.5

%7.7 135.4 382.2

1,753.7 217.8 957.2

2,241.9 287.2 1,248.0

2,906.0 359.8 1,635.3

3,534.4 475.8 1,948.7

4,266.1 581.2 2,308.7

50.8 81.0 8.0 3. I

61.8 117.8 10.0 13.1

72.5 119.9 12.0 27.5

92.1 133.7 15.1 34.6

133.3 247.5 19.5 49.8

184.3 299.8 28.0 66.6

253.6 308.1 34.6 110.4

328.9 421.3 42.2 118.5

432.1 481.7 52.5 143.6

534.7 587.0 63.1 191.4

599.0 843.0 71.4 187.3

811.1 948.1 87.2 315.7

986.7 887.9

17.1 26.2

14.9 28.0

15.3 32.0

14.3 39.0

14.0 39.5

12.4 54.6

12.8 55.7

12.4 56.3

13.5 55.1

13.6 54.1

11.0 63.6

20.2 32.2 3.2 1.2

17.4 33.2 2.8 3.7

16.5 27.3 2.7 6.3

15.6 22.6 2.6 5.9

13.8 25.6 2.0

11.3 13.7

11.3 14.5

12.2 13.6

12.5 13.8

1.5

1.5

1.5

1.5

5.1

10.5 17.1 1.6 3.8

4.9

4.1

4.1

23. I 27.3 43.6 4.3 1.7

20.7 24.2 46.1 3.9 5.1

22.5 24.2 40.1 4.0 9.2

23.5 25.6 37.1 4.2 9.6

23.1 22.8 42.3 3.3 8.5

27.3 23.1 37.6 3.5 8.4

28.9 25.5 31.0 3.5 11.1

28.3 25.9 33.2 3.3 9.3

9.3 2.4 1.6 .3

11.0 3.1 1.6 .3

12.0 3.8 1.8 .3

12.9 1.9 .3

14.3 5.7 2.0 .3

16.4 8.9 2.0 .3

17.7 9.9 2.3 .3

84.4

85.4

93.0

91.9

101.6

118.2

2.8

7.0

6.4

5.5

7.1

28.1 32.7 3,672.0 4,564.0 16.3 18.5

68.6 6,753.4 24.6

7.0 25.0 2,718.0 3,238.0 13.9 15.4

Nore: New GDP series used from 1978 onward.

5.1

82/83

83/84

336.4

1,266.5 835.4 105.2 435.6

1,560.3 916.0 132.4 519.5

10.2 68.6

10.5 70.6

12.9 65.8

12.4 66.0

4.5

8.9 12.6 1.1 2.8

7.9 9.3 .9 3.1

8.1 7.3 .8 2.8

10.2 6.7 .8 3.5

10.8 6.3 .9 3.6

30.0 27.2 30.4 3.3 9.1

29.7 27.3 30.0 3.2 9.8

30.2 24.6 34.6 2.9 7.7

32.6 25.3 29.6 2.7 9.8

35.7 27.5 24.8 2.6 9.4

37.8 29.8 19.7 2.5 10.3

36.3 31.8 18.6 2.7 10.6

18.8 10.6 2.3 .3

18.6 10.2 2.5 .3

17.8 9.6 2.4 .3

19.5 12.4 2.1 .2

20.9 14.4 2.1 .2

20.9 14.8 2.2 .2

19.8 13.0 2.6 .2

19.6 12.9 2.4 .2

113.2

109.4

110.1

108.7

123.0

128.9

129.7

126.7

114.4

4.0

2.7

2.1

2.3

3.9

5.8

3.0

1.1

.7

.7

70.3 10,708.0 36.2

61.6 12,642.5 40.7

61.7 15,466.7 46.6

94.5

104.0 128.4 389.1 82.5 305.0 80.2 166.2 19,033.0 24,002.5 34,344.7 48,913.5 58,421.3 62,646.5 73,697.6 58.4 77.4 52.7 100.0 111.2 119.6 136.3

70

Wing Thye Woo and Anwar Nasution

item from the direct tax category and the “other oil revenue” from the taxes on domestic consumption category. The second subcategory is zero after 1977/78 because the government stopped taxing the domestic use of oil and started subsidizing its use instead, changing what was formerly a revenue item to an expenditure item. Taxes on domestic consumption consist primarily of sales tax (which after 1984/85 includes value-added taxes) and excises. Prior to 1986 the property tax consisted only of IPEDA, the land tax. Nontax receipts are mainly the profits of state-owned enterprises. The most notable feature of table 4.1 is the central government’s increasing reliance on oil as its chief source of revenue. Oil revenue as a share of total federal revenue rose from 26 percent in 1969/70 to 55 percent in 1974/75, and peaked at 71 percent in 1981/82. The fiscal danger of such a narrow tax base was brought home dramatically in 1982 when the global recession caused oil prices to collapse. Oil revenue (in 1980 rupiahs) fell from Rp 7.8 billion in 1981/82 to Rp 6.9 billion in 1982/83, causing real total revenue to fall for the first time since the Soekarno years. The continued real revenue decline in the succeeding years as oil prices began their free fall from $36/barrel to $18/barrel wreaked havoc with the financing of expenditure, especially of development projects. It is, hence, only natural that since December 1983 the government has passed several tax reforms to broaden the tax base. The need for action is well illustrated by the fact that nonoil revenue normalized by GDP has fallen from the 1969-71 average of 8 percent to the 1980-82 average of 6 percent. The fact that Indonesia was adhering to its balanced budget rule and the inflow of oil revenue had been enormous does not justify the decision to allow the tax base to shrink. The government could have broadened the tax base and still adhered to its balanced budget rule. All it had to do was reduce its foreign borrowing. Foreign borrowing obviated the introduction of unpopular measures to make tax collection broader and more effective. Since the ease of external borrowing was no doubt helped by the existence of substantial oil reserves, we can attribute the absence of base broadening largely to the two oil booms. It appears that the “tax-negotiating’’ form of tax collection in the 1969-83 period did not undermine the spirit of progressiveness which the pre-1984 Indonesian tax code aspired to achieve. The fact that the nonoil-income-tax/GDP ratio exhibits a slow, rising trend indicates that the actual marginal rates are mildly progressive. Indirect evidence suggests quite strongly that progressivity increased in the 1969-83 period. This statement is based on the common belief that direct income taxes are progressive and indirect consumption taxes are regressive. The Indonesian ratio of direct income tax to indirect consumption tax has been increasing over time. Income taxes were only 84 percent of consumption taxes in 1969/70, but averaged over 120 percent in 1980-83.

71

IndonesidChapter 4

Indirect taxes increased more slowly than direct taxes as real per capita income doubled in the same period. It may be true that the rich did not pay the high taxes required by the tax code, but in light of the two preceding indicators of vertical equity, it cannot be said that this was done at the expense of the poor. We are unable to analyze the incidence of taxes along rural-urban and regional divisions because the revenue data is not disaggregated enough to permit such an examination. There are two items in table 4.1, however, that permit us to make a partial assessment of our hypothesis that an agricultural bias existed in Indonesian fiscal policy. The first item is the ratio of IPEDA land tax to nonoil income tax which has shown a secular decline since 1969/70; IPEDA is paid almost entirely by landowners in rural areas, while income tax is paid by urban residents. This ratio, together with the increasing direct tdindirect tax ratio, suggest that the tax burden in the rural sector has not increased at the same pace as in the urban sector. The second item suggestive of agricultural sector bias is the large decline in revenue from nonoil export taxes (see memo item in the table) and from import taxes. As pointed out earlier, nonoil exports are largely agricultural products like rubber, palm oil, timber, coffee, tea, and spices. In April 1976 the export duties on most agricultural exports were reduced from 10 to 5 percent, and subsequently to zero. While there is no doubt that the large decline in export tax revenue is a big transfer to the agricultural sector, the almost equally dramatic phasing out of import taxes did not result in the same degree of transfer.2This is because the import tariff in some cases was replaced by nontariff barriers such as quotas and monopoly import licenses. It is reasonable to believe that the government would not have allowed part of this import tax revenue to be transferred as economic rents to certain segments of the industrial elite if the treasury were not awash with oil revenue. This is the second instance of tax base erosion permitted by the oil boom, the first being the postponement of tax reforms because of easy external credit.

4.3 The 1984 Tax Changes and Their Aftermath With the onset of the worldwide recession in 1982, there was an across-the-board fall in Indonesian exports. The severity of this external shock caused the real GDP growth rate to fall from 6.8 percent in 1981 to 0.1 percent in 1982. The oil sector was particularly badly hit, with oil export earnings plummeting from U.S. $18 billion in 1981 to U.S. $15 billion in 1982. Because oil taxes accounted for 70 percent of domestic revenue in 1981, the collapse in oil exports exerted severe financial pressures on government spending. The Indonesian government reacted swiftly against the financial crisis and the low level of economic activity: the rupiah was

72

Wing Thye Woo and Anwar Nasution

devalued by 38 percent against the dollar in March 1983 to boost nonoil exports, and some forty-seven capital-intensive projects were postponed indefinitely. It was clear that greater internal resource mobilization was necessary to make up for the immediate revenue shortfall and to broaden the tax base to prevent the reoccurrence of major financial crises arising from overreliance on one revenue source. There were a number of hopeful signs that a lot more revenue could be extracted from the nonpetroleum sector if better enforcement were undertaken. The most hopeful sign was that only 60 percent of the taxpayers who had filed returns in 1979/80 did so again the following year. In December 1983 the Indonesian government announced a drastic revision of the personal and corporate income taxes which would take effect on 1 January 1984. The complicated and steeply progressive income tax structure was simplified to three rates-15 percent, 25 percent, and 35 percent-which applied to both personal and corporate taxpayers. To make enforcement easier, the cutoff point beyond which people had to pay income tax was doubled to render only 10 to 15 percent of the population eligible for income taxation. Greater attention was put on withholding as the way to collect personal income taxes. Corporations were required to withhold 15 percent of interest, rents, royalties, and dividends to domestic residents and 20 percent of these payments to foreigners. The time-consuming practice of collecting corporate taxes by negotiating individually with the firm concerned was replaced with complete self-assessments by the firms themselves. These self-assessments were subject to selective audit by the government to prevent abuses. The commitment to efficiency in the reform measures was emphasized by laying down specific time limits for the government to refund excess taxes and to respond to appeals against its rulings. Later on in 1984 the Indonesian government sought to increase the tax rolls by announcing that taxpayers who registered by June 1985 were eligible for a pardon of past unpaid taxes. The broadening of the tax base was impressive-there were 995,000 registered income taxpayers at the end of 1985 compared to the 550,000 registered in March 1984. The tax amnesty program resulted in adding Rp 52 billion to 1985186 revenue (IMF 1986a). It needs to be noted, however, that while the procedural reforms and simpler tax code reduced the administrative burden and the incentive to cheat, the biggest bugbear of the Indonesia tax system still remains: shortage of trained personnel. Until this outstanding personnel problem is resolved, nothing can be done about the fact that in 1985 only 50 percent of registered corporate taxpayers and 70 percent of registered personal-income taxpayers actually filed returns. In April 1985 the complicated sales tax with seven different rates was replaced with a value-added tax (VAT) of 10 percent. Like the response to

73

IndonesidChapter 4

the earlier income tax reforms, there was a surge in the number of registered VAT taxpayers40,OOO in December 1985, up from 25,000 in March 1985. The disappointing aspect,, as with the income tax case, is that only 41 percent of the registered VAT payers are filing the required monthly returns. In January 1986 a new property tax law that consolidated IPEDA with six other property taxes was introduced and the stamp duty laws were revised. The revenue-raising ability of the tax reform has been impressive (see table 4.2). Real nonoil income tax rose from 1.3 trillion rupiahs in fiscal 1983 to 1.7 trillion in fiscal 1986. The success of the VAT was even more impressive: it boosted the revenue from domestic consumption by 94 percent in the first year of its introduction. A further rise of 30 percent is expected

Table 4.2

Revenue Structure, FY1982 to FY1987 (in billions of rupiahs)

7ixal revenue Tax on nonoil income Tax on oil and gas Tax on nonoil domestic consumption Tax on international trade Tax on property, IPEDA Nontax receipt Share of total revenue (%) Tax on nonoil income Tax on oil and gas Tax on nonoil domestic consumption Tax on international trade Tax on property, IPEDA Nontax receipt Share of nonoil revenue (%) Income tax Domestic consumption tax International trade Property tax Nontax receipt Indicative ratios (%) Total revenueiGDP Oil and gas taxiCDP Nonoil income tax/CDP Property tax/GDP Nonoil income taxi domestic consumption tax Nonoil export taxitotal revenue Memo items Tax on nonoil exports Nominal GDP GDP deflator (1980= 100)

82/83

83/84

84/85

85/86

86/87

12,418.3 1,605.2 8,170.4

14,432.7 1,784.3 9,520.2

15,905.5 2,121.0 10,429.9

19,252.8 2,313.0 11,144.4

17,832.5 2,880.5 9,738.2

17,236.1 3,315.9 6,938.6

1,266.5 835.4 105.2 435.6

1,560.3 916.0 132.4 519.5

1,648.2 861.9 157.2 687.3

3,478.6 657.8 167.5 1,491.5

3,317.1 658.8 284.0 953.9

4,481.4 732.6 274.0 1,049.3

12.9 65.8

12.4 66.0

13.3 65.6

12.0 57.9

16.2 54.6

19.2 40.3

10.2 6.7 .8 3.5

10.8 6.3 .9 3.6

10.4 5.4 4.3

18.1 3.4 .9 7.7

18.6 3.7 I .6 5.3

26.0 4.3 I .6 6.1

37.8 29.8 19.7 2.5 10.3

36.3 31.8 18.6 2.7 10.6

38.7 30.1 15.7 2.9 12.6

28.5 42.9 8.1 2.1 18.4

35.6 41.0 8.1 3.5 11.8

32.2 43.5 7. I 2.7 10.2

19.8 13.0 2.6 .2

19.6 12.9 2.4 .2

18.2 11.9 2.4 .2

20.0 11.6 2.4 .2

17.6 9.6 2.8 .3

16.0 6.4 3.1 .3

126.7 .7

114.4 .7

128.7 .6

66.5 .3

86.8 .4

74.0 .4

82.5 62,646.5 119.6

104.0 13,697.6 136.3

91.0 87,535.5 152.6

50.5 96,132.4 165.8

78.8 101,491.2 170.9

70.9 107.672.0 176.0

1.o

87/88

Note: GDP deflator assumed to rise 3.1 percent in 1986 and 3 percent in 1987. Real GNP assumed to rise 2.4 percent in 1986 and 3 percent in 1987.

74

Wing Thye Woo and Anwar Nasution

for fiscal 1987, increasing (real) consumption taxes by almost two and half times over their fiscal 1984 value. The reduction of the maximum marginal rate from 50 percent to 35 percent did not decrease the progressiveness of the income tax system. Nonoil income taxes as a proportion of GDP rose from 2.4 percent in 1983/84 to 2.8 percent in 1986/87, and were expected to reach 3.1 percent in 1987/88. Progressiveness in income taxes was enhanced because the income tax reform greatly increased the number of people paying taxes. Furthermore, with the doubling of the threshold for tax eligibility, the majority of these new taxpayers are people who have incomes substantially above the average. It seems however that the overall effect of the whole tax reform package may not be a progressive one. This is because the regressive taxes on nonoil domestic consumption (sales tax, VAT, excises, and stamp duties) were raised much more than progressive taxes on income. The ratio of income taxes to consumption taxes fell by 50 percent in fiscal 1985, the year that VAT was introduced.

4.4 The Structure of Central Government Expenditure Table 4.3 shows the allocation of state expenditure according to function. Expenditure is divided into two categories: routine and development. Routine expenditure represents what is necessary to maintain the level of existing government services, while development spending represents capital deepening which expands the productive capacity of the economy. In the case of Indonesia, the official designation of development expenditure does not in many instances correspond to its economic definition. The biggest misnomer is generally believed to be the payment of salary supplements from the development budget to government employees for development-related activities. There is great incentive for government workers to initiate many minor development projects because the average salary in the public sector is rather low and the criteria for supplement awards is quite broad. Salary supplements are paid to civil servants for engaging in “development” activities such as serving on the steering committee of a new project, doing exercises in project evaluation, and travelling to inspect construction projects. The result is that many members of the bureaucracy receive regular supplements which amount to significant portions of their salaries. The worst feature of this scheme is that it encourages neglect of operations and maintenance activities in favor of starting new projects. Other items of routine expenditure that are included in the development budget are fertilizer subsidies and military expenditure. The former is clearly an input to the current production process, and the latter contributes to capacity creation only in the broadest sense that viable economic growth is

Table 4.3

Expenditure Structure of the Central Government Budget FY1969 to FY1987 (in billions of rupiahs) 69/70

70171

71/72

Total expenditure

342.7

467.8

Total routine expenditure Personnel Debt service External debt service Internal debt service Subsidies to regions Food subsidy Oil subsidy Other routine expenditures

216.5 103.8 14.4 12.7 1.7 44.1

288.2 131.4 25.6 23.6 2.0 56.2

.o .o

.o .o

54.2

75.0

72.4

100.4

265.1

179.4

325.7

Total development expenditure Regional development Fertilizer subsidy Agriculture & imgation, excluding fertilizer subsidy Industry & mining Electric power Transportation, tourism & communications Manpower & transmigration Education & culture Health & social welfare Housing & water supply General public services Government capital participation Other items Unknown allocation of project aid

126.2 12.6

179.6 43.5

207.9 49.6

298.2 55.7

.o

.o

.o

.o

450.9 70.1 33.0

961.8 136.0 227.2

1,397.7 173.0 134.0

25.0 5.8 4.1

32.1 1.8 7.1

46.6 8.1 14.1

39.6 4.7 16.2

45.0 5.3 21.6

74.8 71 .O 79.0

25.3 .1 9.1 4.5 1.2 11.8

17.7 1.o 8.9 3.5 2.6 14.6

42.4 .7 10.9 4.6 2.4 11.9

44.0 .3 16.2 7.3 4.4 16.0

57.0

124.0

.o

5.0

29.9 14.4 5.3 78.0

.o

Memo item GDP deflator (1980= 100) (continued)

72/73

73/74

74/75

75/76

76/77

77/78

78/79

557.0

736.3

1,164.2

1,977.9

2,730.3

3,684.3

4,305.7

5,299.3

349.1 163.3 46.6 37.2 8.4 66.8

438.1 200.4 53.4 44.1 83.9

713.3 268.9 70.7 62.6 11.1 108.6

.o .o

.o .o

1,332.6 593.9 78.5 71.7 6.8 284.5 50.0

1,629.8 636.6 189.5 165.1 24.4 313.0 39.0

2,148.9 893.2 228.3 220.9 7.4 478.4

.o .o

1,016.1 420. I 73.7 67.3 6.4 201.9 141.0

.o

.o 451.7

65. I 483.9

2,743.7 1,001.6 534.5 525.7 8.8 522.3 43.5 197.0 444.8

2,054.5 190.0 107.0

2,156.8 251.0 32.0

2,555.6 275.0 83.0

123.0 124.0 128.0

249.0 195.0 218.0

348.0 139.0 223.0

367.0 205.0 272.0

47.0 25.0 7.0 49.0

312.0 12.0 114.0 38.0 13.0 72.0

429.0 27.0 136.0 48.0 30.0 114.0

355.0 61.0 211.0 71.0 90.0 123.0

413.0 95.0 251.0 79.0 56.0 225.0

5.3

.o

.o

.o

1

.9

1.2

7.0 1.6

24.7 1.9

.O .O

98.0 19.0

115.0 40.0

225.0 87.0

190.0 63.0

162.0 73.0

25.8

44.6

8.0

67.2

91.3

.o

.o

.o

.o

.o

13.9

15.4

16.3

18.5

24.6

36.2

40.7

46.6

52.7

58.4

Table 4.3

(continued) 79/80

SOB1

81/82

82/83

83/84

84/85

85/86

86/87

87/88

Total expenditure

8,076.0

11,716.1

13,917.6

14,355.9

18.31 1.0

19,380.8

22,824.6

21,42 I .6

23,583.2

Total routine expenditure Rrsonnel Debt service External debt service Internal debt service Subsidies to regions Food subsidy Oil subsidy Other routine expenditures

4,061.8 1,419.9 684.1 647.6 36.5 669.9 124.9 534.9 628.1

5,800.0 2,023.3 784.8 754.0 30.8 976.1 281.6 1,022.0 712.2

6,977.6 2,277.7 931.0 915.0 16.0 1,209.4 224.0 1,316.0 1,019.5

6,996.3 2,418.1 1,224.5 1,204.7 19.8 1,315.4 1.o 962.0 1,075.3

8.41 1.8 2,757.0 2.102.7 2,072.9 29.8 1,546.9

11,951.5 4,018.3 3,323.1 3,303.1 20.0 2,489.0 374.2 1,746.9

13,125.6 4,212.6 4,223.2 4.183.2 40.0 2,639.7 .O 142.4 1.907.7

15,826.6 4,316.9 6,805.4 6,765.4 40.0 2,649. I

928. I 1,077.I

9.428.9 3.046.8 2,776.5 2.737.2 39.3 1,883.3 0.0 506.7 1,215.6

Total development expenditure Regional development Fertilizer subsidy Agriculture & irrigation, excluding fertilizer subsidy Indusuy & mining Electric power Transportation, tourism & communications Manpower & transmigration Education & culture Health & social welfare Housing & water supply General public services Government capital participation Other items Unknown allocation of project aid

4,014.2 336.0 85.0

5,916.1 482.0 283.0

6,940.0 616.0 371.0

7,359.6 711.0 420.0

9,899.2 749.0 324.0

9,951.9 791.0 732.0

10.873.1 850.0 477.1

8,296.0 939.0 672.0

7,756.6 873. I 203.5

423.0 356.0 376.0

646.0 491.0 431.0

583.0 827.0 530.0

511.0 913.0 758.0

589.0 2,153.0 660.0

967.0 839.0 911.0

660.9 1,189.0 1,447.0

434.0 737.0 788.0

977.2 349.9 1,008.9

466.0 162.0 362.0 142.0 117.0 473.0

780.0 326.0 575.0 218.0 191.0 700.0

807.0 417.0 726.0 286.0 166.0 800.0

876.0 436.0 703.0 259.0 151.0 785.0

1,527.0 456.0 1,032.0 279.0 221.0 899.0

1,428.0 422.0 1.231.0 320.0 224.0 927.0

1,484.0 665.0 1,413.0 398.0 335.0 977.0

1.063.0 394.0 1,146.0 312.0 333.0 722.0

1,288.I 156.6 1,021.5 207.7 412.0 569.4

466.0 250.0

389.0 404.0

389.0 422.0

281.0 556.0

234.0 776.0

292.0 868.0

221.0 758.0

202.0 554.0

191.1 497.0

.o

.o

.o

.o

.o

.a

.o

.o

.o

77.4

100.0

111.2

119.6

136.3

152.6

165.8

170.9

176.0

.o

.o

.o .o 2,055.2

Memo item GDP deflator (1980= 100)

77

IndonesidChapter 4

not possible without deterrence of foreign aggression. In the absence of more information, it is not possible for us to eliminate the discrepancy between the official and economic definitions of development expenditure. The allocation of official development expenditure by program is given in table 4.4. The general INPRES programs are funds channelled to local authorities to finance public works projects chosen at their own discretion. The INPRES village program was started at the very beginning of the Soeharto era to alleviate rural unemployment and to rebuild the rural infrastructure that had been allowed to deteriorate under Soekamo. With the growth of the oil sector in the early 1970s, the INPRES district and province programs were started partly to handle projects which affected more than one village and, partly, because the village administrators were not able to absorb more funds. This decentralized decision making in the public works projects of the general INPRES programs is very much in line with what we have earlier identified to be one of Soeharto’s traits-impatience with the bureaucracy. While efficiency was clearly an important concern, what may have been equally important was the political symbolism of commitment to rural development. With the increased inflow of oil revenue, sectoral INPRES programs in primary education, health, reforestation, market and road construction were started and funding to existing INPRES programs was increased. In real terms (1980 prices), the cost of INPRES programs rose from Rp 276 billion in fiscal 1973 to Rp 714 billion in fiscal 1980. The government also expanded its industrialization program: annual government capital participation rose from Rp 166 billion to Rp 477 billion in the same period, a 16 percent annual rate of increase. Given the easing of the budget constraint, it was only natural that subsequent spending was much broader in coverage.

4.5 Preferences as Revealed by the Expenditure Pattern In the discussion on political considerations in chapter 3, we identified a goal of the Soeharto government to be the improvement of the livelihood of the Javanese peasants. This goal is based on the fear of the reemergence of the PKI in its traditional rural stronghold. It must be emphasized that the logical policy translation of this primary policy concern is to reduce rural poverty and not the degree of rural-urban inequality. The policy emphasis is on the absolute standard of living rather than on the relative standard of living. Given the history of separatist movements and the fact that the Javanese dominated key government positions, we also identified regional equity to be another of Soeharto’s primary political concerns. The policy translation in this case is the attenuation of differences in the standard of living across islands. Finally, we argued that the technocrats, because of their belief in comparative advantage, would strive to maintain and improve the economic

Table 4.4

Distribution of Development Expenditure by Programs FY1972 to FY 1987 (in billions of rupiahs) Actual

1. Departments 2. General INPRES programs Subsidies to provinces Subsidies to kabupatens Subsidies to villages 3. Sectoral INPRES programs primary schools Health Markets Replanting/ afforestation Roads 4. IPEDA 5. Irian Jaya and East Timor Subtotal of transfers to lower levels of govemment (2-5) 6. Fertilizer subsidy 7. Government capital participation (PMP) 8. Others Total (1-8) 9. Project aid Total (1-9)

72/73

73/74

74/75

75/76

76/77

150.0 39.3 20.8 12.8 5.7 .O

.O

167.3 48.7 20.8 19.2 5.7 19.2 17.2

.o

.o

221.6 101.3 47.4 42.5 11.4 25.0 19.7 5.3

384.9 129.0 54.0 59.1 15.9 65.1 49.9 15.2

590.9 143.7 61.5 62.4 19.8 94.1 57.3 20.8

.o .o

16.0

77/78

78/79

79/80

80/81

81/82

82/83

83/84

84/85

2,533.2 336.8 166.7 119.4 50.7 377.2 249.8 50.4 2.5 48.6 25.9 87.2 6.4

2,724.6 448.1 215.0 162.6 70.5 584.5 374.5 78.8 6.0 70.4 54.8 94.5 6.8

3,260.9 535.3 253.0 193.9 88.4 444.2 267.4 80.3 4.5 49.6 42.4 105.2 5.7

3,219.5 538.8 253.1 194.1 91.6 771.2 549.3 87.3 10.6 59.4 64.6 132.4 5.2

3,474.4 540.4 253.0 194.6 92.8 824.4 572.0 64.6 25.5 61.2 101.1 157.2 4.2

4,466.5 2,087.7 574.5 599.7 287.3 280.0 188.6 220.8 98.6 98.9 753.7 715.5 526.1 417.2 110.6 114.5 11.5 4.4 42.5 42.3 70.1 130.0 167.5 255.6 6.9 7.2

807.6 283.6

1,133.9 371.4

1,090.4 420.1

1,447.6 324.2

1,526.2 731.6

1,502.6 477.1

252.8 476.5 480.9 336.6 591.7 128.5 326.7 448.7 291.0 385.5 565.3 75.1 1,568.3 2,697.9 4,486.4 5,276.1 5,434.7 6,031.7 987.3 1,316.3 1,429.7 1,663.9 1,924.9 3,867.5 2,555.6 4,014.2 5,916.1 6,940.0 7,359.6 9,899.2

336.1 474.9 6,543.2 3,408.7 9,951.9

851.0 181.6 86.8 70.9 23.9 176.0 111.8 26.9 1.3 36.0

.o

.o

.o

.o .o .o

.o

.o

.o

15.2 3.3

19.5 3.3

28.0 4.0

34.6 5.5

42.2 5.0

52.5 9.0

63.1 10.4

1,480.3 218.8 100.7 87. I 31.0 252.0 155.8 30.0 12.4 40.8 13.0 71.4 6.6

57.8 .O

85.7 33.0

158.3 227.2

234.2

285.0 107.3

366.2 31.8

431.1 82.6

548.8 125.0

22.5 5.6 235.9 62.3 298.2

40.8 10.0 336.3 114.1 450.9

91.1 67.7 765.9 195.9 961.8

217.9 108.7 64.0 79.8 926.3 1,280.9 471.4 773.6 1,397.7 2,054.5

166.9 109.8 1,419.2 737.6 2,156.8

.o .o

.o .o

134.5

.o

744.5 167.7 75.4 69.1 23.2 137.0 85.0 26.3 1.2 24.5

Budget 85/86

86/87

87/88 752.2 604.9 280.0 226.0 98.9 326.3 100.8 76.3 3.0 16.2 130.0 246.6 5.0

1,578.0 671.5

1,182.8 203.5

412.3 207.4 511.2 243.7 7,369.7 4,788.3 3,503.4 3,507.7 10,873.1 8,296.0

83.4 109.0 2,330.9 5,425.7 7,756.6

79

IndonesidChapter 4

incentives to produce Indonesia’s traditional exports, primary commodities. Since the production of primary commodities is confined to rural Java and the Outer Islands, the thrust of the technocrats’ economic program addresses two important political objectives of the government. It is impossible to conclude from tables 4.3 and 4.4 whether a rural bias exists because only a few of the items in them can be easily classified either as pro-rural or pro-urban. The obvious pro-rural items are “fertilizer subsidy” and “agriculture and imgation” in table 4.3 and most of the INPRES programs in table 4.4. The obvious pro-urban items are “food subsidy” and “housing and water supply” in table 4.3. Examination of these items shows that the pro-rural items tended to be financed first after the 1973 and 1979 OPEC price increases, and that they also tended to suffer smaller cuts when future revenue prospects turned gloomy as in 1977/78 and 1983/84. The 1986/87 fertilizer subsidy allocation provides a striking example of rural income maintenance. In the 1986/87 recession, while total nominal spending by the government fell by 7 percent in response to lower domestic revenue, fertilizer subsidies actually rose by 20 percent in an attempt to check the fall in rural Javanese income. What is really noteworthy about this is that the peasants were encouraged to grow more rice at a time when BULOG, the state rice agency which guarantees the floor price, was on the verge of bankruptcy because of the runaway costs of storing the excess rice from the bumper harvests of previous years! Another indication that the commitment to rural development is genuine is that the first two programs started right after the 1973 oil price increase were targeted toward the rural sector. Fertilizer subsidies benefited the agricultural sector directly, and the first sectoral INPRES program, by focusing on primary schools, benefited the rural sector disproportionately. The fact that food subsidies, which benefited urban residents disproportionately, were started in 1974/75 after the 1973 oil price increases does not overturn our hypotheses of rural bias in government policies. The history of food subsidies clearly shows the lack of a systematic urban bias. Food subsidies were considered dispensable. During 1977/78 when Pertamina needed a cash infusion of Rp 86.4 billion to meet its debt obligations, food subsidies were eliminated that year. With the weakening of oil prices in 1982, food subsidies were drastically reduced in 1982/83 and completely ended in 1983/84. Table 4.5 ranks the provinces by their nonmining regional gross domestic product (RGDP) and details the amount of central government expenditure under each program by province. Despite several sizable deviations, the central feature is that government outlay systematically varied inversely with the income of the province. The average total central government spending for the poorest one-third of the regions is 26 percent of RGDP, for the middle one-third, 21 percent, and for the richest one-third, 17 p e r ~ e n t . ~

80

Wing Thye Woo and Anwar Nasution

Table 4.5

Per Capita Budgetary and Central Government Direct Development Expenditures 1980181 1980 Nonmining RGDP Per Capita (in thousands Rp)

INPRES Grant Rp/cap

South East Sulawesi West Nusa Tenggara East Nusa Tenggara D.I. Jogyakarta Central Java West Java Lampung Jambi East Java Central Sulawesi Bengkulu West Sumatra Bali Aceh South Sulawesi West Kalimantan South Kalimantan North Sulawesi Irian Jaya North Sumatra Maluku South Sumatra Riau Central Kalimantan DKI Jakarta East Kalimantan

87 97 97 I19 127 131 146 I46 I47 I47 148 153 I53 I57 I58 I68 181 I96 I97 205 223 228 250 270 448 740

Indonesia

167

Province

Central Subsidy Rpicap

Central Development Expenditure Rp/cap

Total Spending & Nonmining RGDP

13,523 5,692 5,704 4,697 2,830 3,015 4,629 9,497 3,022 10,896 14,185 5,653 6,889 6,849 4,968 8,339 7,804 7,920 1 1.365 5,034 8,801 6,507 7,292 13,075 2,042 10,431

10,177 6,536 11,406 9,084 6,305 6,353 7,038 7,541 5,479 8.973 8,106 7,563 8,220 8,069 7.045 8,267 10,266 13,609 27,191 8.530 8,998 5,276 10,187 11,359 6,690 10,431

36.618 10,342 9,805 10,801 4,967 8,176 8,293 26,864 4,902 20,110 34,319 17,400 12,288 19,467 9,754 12,754 22,712 16,364 28,016 10.908 16,275 18,131 23,557 2 1,794 94.03 I 25,354

72 23 21 22 II 13

17 23 19 34 12 IS 13 16 17 23 6

4,465

7,111

9,961”

13

13

30 9 27 38 20 18 22 14

Source: World Bank (1984, 133) a Excluding DKI Jakarta. The figure including Jakarta is 13,661. The high figure for Jakarta reflects the substantial level of spending on the apparatus of the Central Government rather than on the development of Jakarta itself.

Examining the big outliers to the practice of awarding more aid to the poor provinces yields a very interesting finding. Four Javanese provincesJogjakarta, Central Java, West Java, and East Java-rank fourth, fifth, sixth, and ninth in terms of poverty, yet only Jogjakarta received aid higher than the average level of 13 percent. In Jogjakarta’s case, its higher aid level may have less to do with its poverty than with the fact that its sultan played an important role in Soeharto’s rise to power (he was Soeharto’s first vice president). East Java’s RGDP is indistinguishable from those of Lampung, Jambi, Central Sulawesi, and Bengkulu, but its aid level is only 9 percent while the others receive 13 percnt, 30 percent, 27 percent, and 28 percent, respectively.

81

IndonesidChapter 4

We interpret this Outer Island bias as a deliberate attempt to lessen the seemingly big differences in the absolute amount given to each island. As it is, three Javanese provinces (Central Java, West Java, and East Java) already account for Rp 119 billion of the total amount of Rp 313 billion spent, i.e., 38 percent. If Jakarta and Jogjakarta are included, then Java is receiving 61 percent of total budgetary transfer and direct development expenditures while contributing only 47 percent to total national income. It was, apparently, necessary to tolerate inequities toward these three Javanese provinces in order to have some semblance of regional (inter-island) equity. The political message of table 4.5 is clear: the top echelon of the Soeharto government may be dominated by Javanese, but the government is committed to improving the standard of living in the Outer Islands. This political message is the legacy of the many secessionist movements in the 1950s and early 1960s. Table 4.6 focuses on a number of welfare measures in order to provide an alternate way of determining the thrust of Indonesian economic policy, especially its fiscal policy. We have limited the welfare measures to those which particularly apply to the poorest segment. This is because the provision of services to meet the basic needs of the poor is heavily dependent on government expenditure. Part A of the table divides the population along rural-urban lines. In 1971 only 58 percent of rural children aged seven to twelve attended school compared with 73 percent among urban children. In 1980 the figures were 81 percent and 90 percent, respectively. In the same period the ratio of rural to urban infant deaths declined from 5.5 to 4.6. The surprising finding here is that the rural poor may actually eat better than the urban poor, 1,47 1 calories per day in the rural sector as against 1,433 in the urban sector. Together, these three basic-needs indicators paint a picture of improvement in the social services being provided in the countryside and imply that the Indonesian government does not neglect the rural population. Part B of table 4.6 provides a number of regional welfare measures to serve as a consistency check on the conclusions drawn from table 4.5. The most notable difference is that the provision of health care in the Outer Islands appears to be more pervasive. The minimum average number of health centers in the Outer Islands is at least one and a half times more than in Java. The basic health conditions appear to be at least as good in the Outer Islands as in Java; the infant mortality rate and life expectancy are almost indistinguishable across the main islands, except for the Eastern Islands. The primary school enrollment also saw uniform improvements-a 33 percent improvement in every region. It is clear that the expenditure of the INPRES primary school program was quite evenly spread among the islands. Perhaps the same could be said about the regional division of government programs in general because the poverty rate fell by approximately 17 percentage points in both Java and the Outer Islands.

82

Wing Thye Woo and Anwar Nasution ~

Table 4.6

Indicators of Distribution of Government Expenditures on Basic Needs Part A: Rural-Urban Equity

Ratio (ruraYurban) of number of infant deaths School enrollment ratio for 7- 12 year olds

1971 5.5

I976 4.6

51 73

81

Rural 1,471

Urban 1,434

Rural Urban Daily calorie intake among the poorest 40 percent of sector in 1976

90

Part B: Inter-Island Equity

Java

Sumatra

32

24

36

71

48

140 105

138 104

139 93

139 106

149 108

Indonesia Health centers per million people 1980 Infant mortality rate 1971 1980

Kalimantan

Life expectancy 1969 1978 (years increase)

46.7 52.9 (6.2)

47.1 53.1 (6.0)

47.0 55.4 (8.4)

46.8 53.8

Primary school enrollment ratio 1971 1980

60 84

59

64

85

84

Java

Outer Islands

Indonesia Poverty rate 1970 1980 Per capita consumption index, Indonesia = I00 1970 1980

4.6

40

57

65 47

43 28

100 100

88 97

130

Sulawesi

Eastern Islands 50

NA NA

(7.0)

45.2 52.3 (7.1)

42.7 41.8 (5.1)

60 71

62 82

63 84

105

A Summing Up

The examination of the fiscal system supports our claim that the technocrats favor an economic strategyy which leads to resource transfers to the Javanese hinterland and to the tree crop industries in the Outer Islands. The secular decline in trade taxes and low taxation of land relative to income reflect Soeharto’s political concern with communism and secession, as well as the technocrats’ neoclassical inclination toward the comparative advantage doctrine. This favorable tax treatment of the agricultural sector improves the rural-urban terms of trade and hence encourages the production of tradables, the presence of which determines a country’s ability to service its debts.

83

IndonesidChapter 5

In examining government expenditure, we surmised from fragmentary evidence that government spending was more likely to display a rural rather than an urban bias. In the absence of more detailed data, disproportionate weight was given to the budget allocations for fertilizer subsidies, irrigation projects, rural school programs, the INPRES village programs, and food subsidies. Because of better data, stronger evidence could be garnered to support the hypothesis that budget allocations were more sensitive to inter-island equity. There is in fact evidence that inter-island equity takes precedence over rural-urban equity. This is consistent with our conjecture that the concern for rural development stems more from a desire to eradicate poverty than to narrow the rural-urban gap. The analysis of this chapter sets the stage for our forthcoming discussion on the importance of political factors in determining the debt outcome. To the extent that people are consistent in their actions, the fact that the technocrats support, and Soeharto approves of, a fiscal policy which favors the tradable sector means that they would also advocate a similarly-oriented exchange rate policy. We will show in chapter 6 that exchange rate management has been tempered by political considerations, and will quantify in chapter 8 that this exchange rate policy resulted in Indonesia avoiding a debt crisis during 1982-84.

5

Monetary Policy and Financial Structure

5.1 Introduction

The purpose of this chapter is to analyze the conduct of monetary policy and the development of the financial sector since 1966. Along with other economic measures, financial policies have been actively used by the government to pursue its macroeconomic objectives. During the period of prosperity in the 1970s, mainly due to the two oil booms in that decade, there was no incentive for the government to reform the underdeveloped tax and banking systems which were inherited from the Dutch colonial administration. Major reforms to the financial system in order to mobilize domestic saving were initiated only after the bust of the second oil boom. In contrast to the 1966-67 reforms which accomplished a total turnaround of the economy in a relatively short period of time, recent reforms cannot produce quick results.

84

Wing Thye Woo and Anwar Nasution

5.2 Performance of Monetary Policy in the 1970s As we mentioned earlier, the balanced budget policy of the Soeharto government effectively ended the creation of money to finance budget deficits. This did not mean, however, that monetary creation in the 1970s was truly independent of the state of the government budget. This is because the money stock was also affected by changes in the foreign asset position of the central bank, Bank Indonesia. This balance-of-payments linkage to the monetary supply was what made the stance of monetary policy move in tandem with that of fiscal policy. As will be explained, when fiscal policy was expansionary in the wake of the two oil booms, it induced monetary policy to be expansionary too. The result was a big inflation spurt in the 1970s that had nothing to do with money-financing of government budget deficits, but rather with the lack of instruments to end the balanceof-payments linkage between government spending and the money supply. Prior to the introduction of ceilings on lending by the banking system in April 1974, the main instrument for monetary control was the extension of central bank credits to the banking system, state enterprises, and private companies. Since the central bank credits were extended for a contracted time period, the government was not in a position to engineer quick increases or reductions of the money stock. This reserve method of monetary control was shown to be grossly inadequate for stabilizing the economy when a large fiscal stimulus occurred, financed by increases in oil revenue. The reason for the synchronization of fiscal and monetary policy during the 1970s lies in the balance of payments. With the rapid development of the oil sector since 1970, government revenue from oil accelerated (see table 5.1). It climbed from Rp 99 billion in 1970 to Rp 141 billion in 1971, and then to Rp 231 billion in 1972 oil revenue was actually denominated in U.S. dollars). Since the primitive nature of the Indonesian financial system ruled out the possibility of open-market operations, the maintenance of a fixed dollar-rupiah exchange rate meant that the conversion of oil revenue from dollars to rupiahs in order to finance the expanded government expenditure automatically increased the money supply. This is clearly seen in 1972 when oil revenue increased by 90 billion rupiahs over the previous year. The conversion of this oil revenue (231 billion) led to a 122 billion rupiah increase in the reserve money base because of the monetary authorities’ inability to quickly sterilize the monetary consequences of foreign-exchange market transactions to peg the value of the exchange rate. The 1972 growth rate of reserve money was 46 percent compared to the 29 percent of the previous two years. The price of oil then quadrupled at the end of 1973, encouraging the government to increase its spending. In fact, the government augmented its expenditure well beyond the increase in oil revenue. This was possible because the creditworthiness of Indonesia soared along with the price of oil.

85

IndonesidChapter 5 Monetary Consequences of Changes in Official Foreign Asset Position

Table 5.1

1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

Oil and Gas Tax”

Budget Deficit

Foreign Assets of Central Bank as Proportion of Total Assets

65.8 99.2 140.7 230.5 382.2 957.2 1,248.0 1.635.3 1,948.7 2.308.7 4,259.6 7,019.6 8,627.8 8,170.4 9,520.2 10,429.9 11,144.4 9.738.2

91.1 113. I 117.0 145.7 196.5 224.2 488.4 778.3 771.3 1,033.2 1,379.2 1,489.1 1,705.0 1,937.6 3,878.3 3,475.3 3,571.8 3,589.1

18.3% 21.3 17.7 38.9 41.7 49.0 12.0b 22.7 31.2 32.3 39.3 46.1 39.6 28.4 34.5 39.0 39.8 32. I

Change in= Reserve Money

M1

60.0 47.0 60.0 122.0 153.0 308.0 282.0 248.0 340.0 165.0 593.0 897.0 545.0 187.0 1,031.0 563.0 1,020.0 1.449.0

67.0 67.0 69.0 155.0 197.0 271.0 332.0 327.0 405.0 482.0 828.0 1,695.0 1,463.0 646.0 456.0 1,005.0 1,543.0 1,507.0

Rate of Growth of Reserve Money 60.0% 29.4 29.0 45.7 39.3 56.8 33.2 21.9 24.6 9.6 31.5 36.2 16.1 4.8 25.1 11.0

17.9 21.6

M1

Inflation

57.8 36.6 27.6 48.6 41.6 40.4 35.2 25.7 25.3 24.0 33.3 51.1 29.2 10.0 6.4 13.3 18.0 14.9

17.4% 12.3 4.4 6.4 31.0 40.6 19.1 19.8 11.0 8.1 20.6 18.5 12.2 9.5 11.8 10.5 4.7 5.8

”In billions of rupiahs. ?he low ratio of foreign assets to total assets in 1975 is because of the bailing out of Pertamina by Bank Indonesia.

External credit, euphemistically referred to as “external revenue,” was pretty much available on demand. This is evident from the large jumps in the budget deficit; it rose from Rp 224 billion in 1974 to Rp 488 billion in 1975, and then to Rp 778 billion in 1976. The constantly increasing amount of oil revenue caused the monetary authorities to lose control of the money supply. Reserve money grew 57 percent in 1974, and the inflation rate for that year was 41 percent. The central bank responded to this monetary anarchy by setting lending ceilings on the banking system. The government tried to control the inflationary effects of this massive foreign wealth transfer by increasing imports in order to reduce the inflow of foreign reserves. Imports by the public sector were increased by restructuring budgetary expenditures toward those which were import-intensive. The private sector was encouraged to import more through a general tariff reduction and the removal of the ban list. In addition, the government built up stockpiles of imported commodities such as foodstuffs and basic materials. We want to emphasize that the expansion of government expenditure was not in any way a logical consequence of the balanced budget practice. The government need not have increased expenditures at all; it could have decreased its external “revenue” (external borrowing) in step with the

86

Wing Thye Woo and Anwar Nasution

increase in oil revenue. With access to external credit markets, the Indonesian balanced budget practice was neither a restraint on government spending nor a check on the growth of the money stock. The key point we want to make from the 1973-75 experience is not that the increase in government spending was undesirable, but that the instruments available for controlling monetary aggregates were grossly inadequate. Loan ceilings were a highly inefficient way to solve the problem. The need for better monetary instruments was again demonstrated in 1979 and 1980 after the doubling of oil prices. The great inflow of oil revenue via the balance of payments caused the monetary authorities to briefly lose control of the money supply. Reserve money grew 32 percent in 1979 and 36 percent in 1980, pushing inflation up to 20 percent in these years.

5.3 Financial Structure The structure of the organized financial system is shown in table 5.2.' The dominance of the banking sector in the system is evident from the data. At the end of 1985 the banking sector (Bank Indonesia and commercial banks) held more than 90 percent of the gross assets of the organized financial system. Table 5.2

Structure and Growth of the Organized Financial Sector, 197MQ

Gross Assets (in billions of rupiahs)

Number in 1982

1986

Bank Indonesia 1 Deposit money banks 113 National foreign exchange banksb 15 11 Foreign banks Other commercial banksC 50 Development banks 28 Nonbank financial institutionsd 13 Savings banks' 3 Insurance companies 83 Leasing companies 34 Other credit institutions' All institutions 6,106

Annual Growth of Assets (%)

Shares in Assets (5%)

1978

1982

1985

1978-82

1983-85

1978

1985

1 113

5,368 5,277

13,707 15,952

23,285 33,758

26.4 31.8

19.3 28.3

48 48

55

15 11 59 28

4,115

26,469 2,245 2,342 2,702

32.6 23.4 29.5 33.1

27.6 24.1 48.1 26.5

37 5 2

43 4 4

401

12,724 1,172 720 1,336

4

4

14 3

195 37 159 26 57 1 1,077

805 452 528 114 86 31,615

2,073 1,290 694 693 30 61,879

42.6 86.5 35.0

37.1 41.8 9.5 82.5

2

4 2 I

-

73 20 6,120

505 256

44.0

14.6 30.0

Source: Bank Indonesia, Indonesian Financial Sfatisfics.various issues. Note: Dash indicates data were not available.

"Annual compound rates. bFive state banks and ten national private banks. 'National private banks doing only domestic currency business. dNine investment finance, three development finance, and two other finance companies. 'One state savings banks and two private savings banks. 'Village banks, rural paddy banks, and government-owned pawnshops.

-

1

-

38

1

-

_

_

25.1

100

100

87

IndonesidChapter 5

About 70 percent of the assets and credit of the banking system are owned and provided by the seven state-owned banks, including Bank Indonesia itself. Until March 1984 Bank Indonesia provided credits directly to economic units, particularly to state-owned enterprises and to quasigovernment institutions. The rest of the bank credit is divided evenly between the foreign and joint venture banks and about seventy domestic private banks. The latter include twenty-seven Regional Development Banks (RDB) owned by provincial governments. Each province has an RDB restricted to operating within its own jurisdiction. In reality, RDBs operate like commercial banks. There are now twelve nonbank financial institutions (NBFI) operating in Jakarta, all established between 1972 and 1974. Nine of them are investment finance companies, and three are development finance corporations. The first type of NBFI acts as an intermediary and underwriter of financial paper and finances medium- and long-term investments. The second type, development finance corporations, concentrates only on medium- and long-term financing and equity participation. The state commercial banks and the central bank are the majority shareholders of nearly all NBFI. The minority owners are domestic and foreign private companies. An NBFI is required to have at least three foreign partners, each from a different country, with at least one partner an investment bank. Bank Indonesia is a major shareholder in two of the investment finance companies and in all three of the development finance corporations. The investment finance companies have one big advantage over the commercial banks: they are subject to only two regulations, a debtlequity ratio of 15 and a ceiling on foreign borrowing. The absence of other regulations has permitted the investment finance companies to adapt more rapidly and effectively than the banks to changes in the economic environment. In fact, these companies were used by the state banks and Bank Indonesia to extend credit and to invest in sectors and activities which banks could not service under the old credit regulations. The Jakarta stock market started operations on 4 June 1952, but it was closed in 1958 due to political and economic instabilities. It was reopened on 10 August 1977. Trading has been virtually inactive. As of September 1986 there were twenty-four equity stocks and three bonds listed in the infant Jakarta stock exchange. Sixteen of the listed companies issuing equity shares were foreign companies and eight were domestic privately-owned firms. The companies issuing bonds were public enterprises. All of the new share issues took place during 1981-84. The main motivation of foreign companies to go public was to comply with the “Indonesianization” process which is required after operating for a certain period of time in the country. Through their overseas networks,these companies had access to international markets. They really did not need to raise money in the small, fragmented, and high-cost Indonesian money market. In other words, raising capital was not

88

Wing Thye Woo and Anwar Nasution

the main reason for foreign companies to issue equity shares in the Jakarta stock exchange. Going public for a foreign company was like “paying an entrance fee” to the Indonesian market. At the same time, these companies derived benefits from the tax concessions offered for going public under the 1983 tax system. The tax benefits actually exceeded the value of the sale of shares! This tax concession has since been rescinded. On the supply side, there are many reasons why the supply of marketable securities is so limited in Indonesia. The government has never floated bonds in the domestic market because its budget deficits since 1968 have always been financed by foreign aid and loans. Only four of the 220 state-owned enterprises have floated bonds in the Jakarta stock exchange, and none of them has issued equity share. The capital needs of these public enterprises have been financed by direct government investment, foreign loans, and subsidized bank credit. For domestic private companies, debt financing was less expensive relative to the costs of raising and servicing equity because credit from the state banks carrying subsidized interest rates was plentiful and could be easily rolled over. During the period of high international inflation rates and low nominal interest rates in the 1970s, it was not hard to tap foreign financial centers such as Singapore and Hong Kong. Working capital could be obtained by issuing promissory notes to the NBFIs (merchant banks). In a country where the data base on taxpayers is poor, tax administration is inefficient, and the legal and accounting systems are underdeveloped, tax evasion is an important source of internal financing. For all of these reasons, the opportunity costs of going public (disclosure, regulation, tax liability, dividend payout, dilution of ownership and control) for corporations is too high. On the demand side, the general public is still unfamiliar with the function of modem financial institutions, including the stock exchange, and the benefits to be derived from them. This, and a history of financial instability and repression, have made investment in stocks unattractive. An example of financial repression in the Jakarta stock exchange is the excessive controls by PT Danareksa (the National Investment Trust), a public company, in stabilizing the stock prices. In order to avoid large capital losses, PT Danareksa stabilizes the share prices within a narrow band (4 percent daily maximum variation) to make a share similar to a fixed price (fixed coupon) marketable asset like a time deposit or bond.

5.4 Financial Repression of the 1970s The financial system in Indonesia was repressed for almost a decade after April 1974, up until the financial reforms of June 1983. To control the increase in money supply resulting from monetization of government oil revenue denominated in dollars, Bank Indonesia set ceilings on bank credits and other domestic assets. The original purpose of the credit ceilings was to

89

IndonesidChapter 5

directly control the expansionary impact of the oil boom in the 1970s since a reserve management approach was shown to be insufficient. Then the government realized that this rationing method of monetary control could also be used to address several important items on its political agenda. Over time, Bank Indonesia was instructed to introduce detailed ceilings by type of credit for each bank through an extensive selective credit system featuring subsidized interest rates. For example, banks were assigned a civic function insofar as they restricted certain credit only to pribumis and made establishing credit for them a priority in order to enhance pribumi participation in economic activities. With the allocation of credit in mind, the Bank of Indonesia set up complicated rediscount financing, rediscount rates, and state banks’ loan rates. For Bimas (a government rice-intensification program) and foodrelated activities, Bank Indonesia provided 100 percent of the funds loaned by state banks. For this rediscount financing the state banks act as agents of the government to finance its program. For other sectors, rediscount percentages range from 20 to 80 percent of the state banks’ loans. The higher the priority of the sector, the higher the percentage of financing. (Discount facilities for nonpriority sectors were subsequently made available in January 1978.) To encourage state banks to extend credit to the priority sectors, Bank Indonesia ensured them adequate profits by charging low rediscount rates ranging from one-fourth to one-half of their loan rates. High rediscount percentages and low rediscount rates increased the rate of return on state banks’ own funds. Because the discount facilities were primarily available to state banks, these facilities were another form of subsidy to their operations. The true subsidy was higher, taking into account the government’s share in the burden of bad debts of state banks. The bad-debt ratio of state banks was quite high. For example, of the Bimas loans made by Bank Rakyat Indonesia (BRI) in Java (85 percent of the total for Indonesia) during 1970-74, around 7 percent remained unpaid two years after the loans were made. Loans were supposed to be due at most seven months from disbursement. The proportion of unpaid debts rose to 13 percent for the 1974-75 crop and 22 percent for the 1975 dry season crop. Approximately Rp 7 billion (U.S. $17 million) was written off or rescheduled during 1974-77 because of crop failures. Risk-bearing in the Bimas loans was shared by Bank Indonesia and Bank Rakyat Indonesia (25 percent each) and the rest (50 percent) by the government. According to Bank Bumi Daya’s 1976 annual report, about 28 percent of its loans were rescheduled and 8 percent had to be written off in that year. The Bank Bumi Daya is the biggest state bank, and most of its unrepaid loans were investment credits. Since most priority credits were handled by state banks, the policy of ceilings with selective credit became one of the major tools protecting state

90

Wing Thye Woo and Anwar Nasution

banks from competition with private banks. In addition, state banks and RDBs are depositories for government institutions and state enterprises. Since they also have a wider network of branch offices than private banks, state banks and RDBs have great advantages in tapping domestic savings. These institutional features discouraged competition, preserved the status quo, and guaranteed the dominant positions of the state banks. Discrimination in access to Bank Indonesia’s liquidity credit by bank ownership has encouraged the banking system to hold excess reserves, both in rupiah and in foreign currencies. With their loans financed primarily through liquidity credit facilities, most of the funds generated by state banks contributed to excess reserves. With the secured loan refinancing, there was little need for state banks to mobilize domestic saving. On the other hand, savings were not attracted to these banks where deposit rates were set by authorities at levels which usually lagged behind inflation. Many RDBs also had excess reserves. The excess reserves of state banks and RDBs, channelled through the interbank market in Jakarta, have been an important source of funds for private banks and NBFIs. In theory such credit policies add to the distortion of resource allocation and preserve fragmentation in the financial market (see McKinnon 1973, Gablis 1977, Nasution 1983). An extensive ceiling is similar to credit rationing, i.e., a ceiling on what customers can borrow, regardless of their willingness to pay higher interest rates. It is hard to judge whether Indonesian authorities succeeded in allocating credit according to their original design. For one thing, there are no detailed data on how credit was allocated according to various government objectives or a scale of priorities. Second, there were too many simultaneous objectives the government had wanted to achieve with the selective credit policy: to redress racial, sectoral, firm size, and technological imbalances; to equalize distribution of income, increase employment, stabilize prices of basic commodities; and so forth. If not accompanied by other policies, these objectives were not likely to be achieved since their number is much greater than the number of instruments available to use in pursuing them. For example, without any talent, skills, or experience, a person cannot be turned into an entrepreneur overnight by credit provision alone. Also, real resource allocation might not be similar to the credit allocation. In order for selective credit policies to be effective, the degree of banks’ credit fungibility would have to be zero. This means that long- and short-term credits from the banking system are exclusively utilized by business firms to finance fixed assets and working capital, respectively. Fungibility in this context is defined as the ability of business firms to borrow credits for a particular purpose but to effectively use it for another. Another type of fungibility is when business firms use low-cost funds from the banking system to finance (bank-) approved uses, thereby releasing their own funds for purposes that otherwise could not be pursued without the

91

IndonesidChapter 5

availability of bank credit. For example, BRI provides low-interest credits for the purchase of hand tractors, and a farmer who receives such credit uses it for that purpose and uses his own funds to buy a color TV set. Although easily evaded by big borrowers such as Pertamina and the multinationals, credit ceilings are thought to have contributed to slowing the rate of growth in the Indonesian money supply (Amdt 1979). There is one extremely interesting puzzle in the credit ceiling experience, and most of the answers to it are rather damning to the dominant presence of the government in the financial sector. The puzzle is that state banks, unlike the private banks, seldom used up all of their prescribed ceilings and had to lend out their excess reserves in the interbank market. One can speculate, first, that the demand was actually high but bad bank practices by the state banks made them unable or reluctant to reach out to small customers because it involved cumbersome operations and low profit per customer. Second, it could be the inability of the officers of the state banks to select projects which were acceptable both economically and politically. Third, the officers may have demanded too high side payments from prospective borrowers (to divide the implicit rents from a negative, zero, or low real interest rate). It has been suggested that the graft could have been as high as 15 percent of the volume of the loan granted (Gray 1979). This would have resulted in the real cost of interest rates from state banks becoming too high either for prospective borrowers or to be competitive with interest rates at private banks. If the interest rates at state and private banks were about equal, borrowers would have preferred to borrow from the latter, especially from branches of foreign banks or NBFIs, to avoid long delays and harassment from state banks. All of these factors could have reduced the rate of growth of credit expansion to less than the permissible ceiling without necessarily satisfying demand.

5.5 Financial Sector Reform in the 1980s A series of negative external shocks began in 1982. The world recession and the first wave of weakness in the world oil markets was worse than had been expected. Economic recovery in the OECD countries since 1983 does not appear to have helped economic growth in Indonesia significantly. The price of oil dropped from $35 per barrel in 1982 to $25 in 1985, and then to $12 in 1986 (figure 5.1). This dismal picture was repeated for the prices of Indonesia’s nonoil exports. The worst is still to come. Nonoil export prices are expected to continue falling, bottoming out only in 1988 when they would be 25 percent of the 1982-84 level (figure 5.2). The decline in the terms of trade has been aggravated by the sharp currency realignment in 1985-87. Most of Indonesian exports are priced in terms of the U.S. dollar, but a large proportion of its imports and foreign debts are denominated in the appreciating currencies. The dollar depreciation worsened the current

92

Wing Thye Woo and Anwar Nasution

account deficit, added to the burden of foreign debts, and cut Indonesia’s real budgetary expenditures. The decline in the terms of trade plus the big reduction in government real nondebt budgetary expenditures produced a sharp decline in economic growth. The annual average real GDP growth since 1982 has declined to one-third to one-half of the level in the 1970s. In short, the end of the oil boom in 1982 made financing the investment-saving and foreign exchange gaps much harder. In response to the external shocks, the government adopted several economic programs, among which was an overhaul of the financial structure. The financial reforms since 1 June 1983 fall into two major categories. The first category comprises partial deregulation of interest rates, elimination of credit ceilings, and reduction in the scope of Bank Indonesia’s subsidized credits to state-owned banks. The second category consists of piecemeal measures. Direct Bank Indonesia intervention in the day-to-day operation of state banks was reduced significantly, and its direct lending to quasigovernment bodies and state-owned enterprises was replaced by state bank lending financed by Bank Indonesia’s liquidity credits. At present there is no treasury debt that can be used for open market operations because the government financed its budget deficits solely through foreign aid and foreign loans. To increase the number of central bank instruments for open market operations, Bank Indonesia began reissuing

93

IndonesidChapter 5

160 140

120 100

80 60 40 44 48 52 56 60 64 68 72 76 80 84 88

Year 1979-1981 = 100 Fig. 5.2 Primary product prices. Index of nonoil commodities (in constant dollars).

Debt Certificates (SBI) in February 1984 and introduced the Money Market Instruments (SBPU) on 28 January 1985.* The SBI had first been issued on 12 March 1970. However, issue ceased during the oil boom era in the 1970s as the authorities encouraged banks to invest their excess reserves in foreign assets in order to sterilize the inflow of oil money. At the beginning the interest rates on SBIs were set by Bank Indonesia at levels higher than the rates paid on excess reserves held at the central bank, but over time the sales of SBIs shifted to an auction system. Since 1984 the frequency of SBI auctions has increased, and rediscounting can be done either at Bank Indonesia or at Ficorinvest, an investment finance type of NBFI largely owned by Bank Indonesia. Bank Indonesia guarantees to rediscount the SBIs at their original auction prices, irrespective of their remaining maturities. Despite promotional efforts by Bank Indonesia, SBIs are not yet well accepted by the financial sector. The SBIs can be resold to the nonbank public. So far, however, the secondary market for SBIs is nonexistent. To maximize their profit from holding SBIs, financial institutions usually purchase a large volume of long-maturity SBIs and hold them for a short period before rediscounting.

94

Wing Thye Woo and Anwar Nasution

SBPUs are contingent liabilities of the bank of NBFI that first endorsed them. There are three types of SBPUs, namely: (1) promissory notes by eligible banks and NBFIs; (2) promissory notes issued by customers of eligible banks and NBFIs when borrowing from them; and (3) bills of exchange issued by third parties and endorsed by eligible banks and NBFIs. Initially the maturity of SBPUs was set between one to three months. However, on 7 August 1985 the upper limit was raised to six months. In reality, 98 percent of the SBPUs are in the form of promissory notes with maturity dates ranging form one to fourteen days. Only institutions which have signed repurchase contracts with Ficorinvest are eligible to endorse an SBPU. The repurchase contract prescribes the upper limit of each institution in rediscounting SBPUs. There are no data available on the total volume of SBPUs in circulation. However, the growth in the market for SBPUs must have been rapid since there is a high rate of growth of Bank Indonesia’s claims on banks arising from the use of the central bank’s rediscount facility on these instruments. As it now stands, the SBPU system is only beneficial for large financial institutions, especially the foreign exchange banks. Its eligibility standards discriminate against small private banks whose customers are mainly small and medium-sized firms. It is also discriminatory against unit banks which operate outside the capital city, since Ficorinvest has no branch office outside Jakarta.

5.6 Evaluating the 1983 Financial Sector Reform Our first observation is that the financial instruments SBI and SBPU still do not provide sufficient control over monetary aggregates. This is clearly seen in the way that the money supply had to be contracted in response to a speculative run on the rupiah in the first half of 1987. Capital flight began in earnest in the second quarter of 1987 when a higher than expected current account deficit was reported. (This overreaction was mostly due to nervous bankers and investors who had suffered losses in the two closely spaced devaluations of May 1983 and September 1986.) In June 1987 the minister of planning, Dr. Sumarlin, after concluding that open market operations would not be able to raise interest rates quickly enough, ordered state enterprises to withdraw Rp 1.3 trillion from state banks to be placed in central bank se~urities.~ This action, together with the sale of Rp 800 billion of open market instruments to banks during the month, sharply reduced bank liquidity and forced the banking system to sell its foreign assets in order to meet rupiah funding requirements. Other domestic corporations began repatriating capital to meet current operating needs. This severe credit squeeze succeeded in convincing private agents that the government was prepared to adjust other policies in order to ensure the viability of the existing exchange rate, and the speculation against the rupiah came to an end.

95

IndonesiaChapter5

Our second observation is that the response of long-term financial intermediation to liberalization of the financial sector is a slow one, and its initial reaction may even be perverse. While the response by depositors has been very favorable, the response by financial institutions has not been so. The volatility of sources and costs of funds forced the banking industry to adjust its lending terms in order to minimize interest rate risks. This resulted in an increasing share of short-term fixed rate assets and a large proportion of credits canying adjustable rates. Since interest rate volatility also increases borrowing risks, the banks have become more cautious in extending credit. The third observation we want to make is that the deregulation should have been preceded by institutional reforms in the state enterprise system. The bureaucratization of the state-owned banks in the protected climate of past credit policy raised their overhead cost and subsequently their cost of intermediation. High arrears in all sectors and all credit schemes is a reflection of high credit risks which resulted from the inadequate selection and supervision of customers. These deficiencies were largely due to the fact that all of their lending risks were passed on to the government. After the reforms, the opportunity cost of funds increased while the cost of intermediation and credit risks remained high. The high cost of intermediation was also caused by the state banks being prohibited from reducing personnel. These are additional reasons why the lending rates of state banks rose after deregulation. Our fourth observation is that the initial balance sheet conditions of the financial institutions should not have been overlooked. The present newly competitive market requires capital bases stronger than those of many of the financial institutions. Due to the past credit policy of subsidized interest rates, a large proportion of the assets of state-owned banks were extended at negative real interest rates. The weakness of the capital base of the private national banks is clearly shown by several recent bank failures. The final observation we want to make is that deregulation of the financial sector should have been accompanied by greater supervision of the banking system. The banking system is not insulated well enough in practice from the fortunes of individual clients. This is because the interconnections of family ownership closely link many of the private national banks to the performance of sister companies. Indeed, some of these banks have been established mainly to secure funds for the nonbank business ventures of their owners. 5.7

Concluding Remarks

The emphasis on the development of two open market instruments, SBI and SBPU, is a long overdue step toward better control over monetary growth. An important obstacle to improved macroeconomic management seems to be that open market operations have been mainly geared to keep interest rates constant. Monetary policy was particularly unresponsive to

96

Wing Thye Woo and Anwar Nasution

external interest rate developments. One reason for the capital outflow prior to July 1987 was that Bank Indonesia had kept the SBI rate flat for a relatively long period, causing domestic interest rates to diverge from international interest rates. In a financially open economy like that of Indonesia, it is essential to recognize that external shocks will frequently make tradeoffs among interest rate stability, domestic income stability, and exchange rate stability inevitable. It is clear from the manner in which the monetary contraction of June 1987 had to be implemented that the market for both SBI and SBPU was still too shallow. It may be difficult to increase their role if the financial markets remain underdeveloped. Financial deepening is an important priority, but not only because of the need to enhance the effectiveness of the monetary instruments. Financial deepening would also better mobilize (and maybe increase) domestic savings, reduce dependence on external credit, and improve the overall allocation of capital within the economy. One of the first steps that could be undertaken to boost development of the financial sector would be to privatize some of the state enterprises. It would certainly ease Indonesia’s external debt burden if a minority portion of these state enterprises were sold to foreigners. The possible increase in efficiency of these enterprises would be an added bonus.

6

Exchange Rate Policy

6.1 Introduction In chapter 3 we identified an important political constituency (technocrats, Javanese peasants, and Outer Island residents) which is opposed to the maintenance of an overvalued exchange rate. We will show in this chapter that this constituency has been successful in influencing exchange rate policy, with the result that there is an asymmetry in policy response to changes in the balance of payments. It makes good economic sense to devalue the real exchange rate when a balance-of-payments deficit occurs, but due to the existence of this constituency it makes good political sense not to allow the real exchange rate to revalue when a surplus occurs. The fact that the institutional memory was impressed by the potency of the exchange rate in effecting economy-wide resource reallocation and income redistribution during the 1966 economic rehabilitation program helps to strengthen the economic argument for a devaluation whenever the balance-of-payments situation demands it. This exchange rate policy, as we will argue in chapter 7, played a crucial role in helping Indonesia to avoid a debt crisis in

97

IndonesidChapter 6

1982-84 because it maintained a large and healthy tradable sector and discouraged capital flight. Given the crucial role of the exchange rate in external debt management, we focus on exchange rate management in some detail. The statistical profile of the rupiah-dollar exchange rate in figure 6.1 is characterized by three distinct phases. The first phase, from October 1966 to July 1971, saw a steady dismantling of the multitiered exchange rate system into a unified exchange rate, revealing a readiness to have medium-sized devaluations at short intervals. In the second phase, from August 1971 to October 1978, there was a fixed exchange rate. The third phase occurred from November 1978 to March 1987 and was a time of large devaluations separated by moderately long periods of gradual exchange rate depreciation. Since these three phases reflect policy responses to changes in the external environment and to developments internal to Indonesia, we will use them as a convenient way to organize our discussion of Indonesian exchange rate management. The discussion of phase 3 will revolve around the November 1978, March 1983, and September 1986 devaluations.

6.2 Phase 1, Pre-August 1971 When the New Order government of General Soeharto took power in October 1965, it inherited a system of multiple exchange rates.' This system not only allowed corruption because of the discretionary element inherent in

60 50 40 30 20 10 0 -1[

68 69 70 71 72 73 74 75 76 78 79 80 81 82 83 84 85 86 Fig. 6.1 Rupiah-dollar exchange rate changes (%), 1968-86

98

Wing Thye Woo and Anwar Nasution

its administration, it also made smuggling an extremely profitable business. Since the array of exchange rates was set to promote domestic industrialization and extract revenue for the government, the exchange rate system discriminated against the traditional agricultural commodity exports. The fact that the whole exchange rate structure had become increasingly overvalued because of chronic high domestic inflation meant an acceleration in the pace of resource shift away from export activities. Many rural residents abandoned the cultivation of tree crops for subsistence farming. The Soeharto regime made the maintenance of a competitive exchange rate and the simplification of the exchange rate system a key element in its economic rehabilitation program. A unified exchange rate was achieved in April 1970 when the government set its major import exchange rate (Bonus Ekspor, BE) to be the same as that of the then free market exchange rate (Devisa Pelangkap, DP) of 378 rupiahs to the dollar.2 The government in this period displayed no reluctance to change the exchange rate whenever it seemed that a balance-of-payments problem was appearing. Exchange rate realignments were quite frequent, with medium-sized devaluations undertaken at short intervals in order to preserve the competitiveness of Indonesian goods in the face of high domestic inflation rates. A good example is the August 1971 devaluation which brought the exchange rate to 415 rupiahs to the dollar. This devaluation was clearly implemented in response to the worsening of the current account deficit, which widened from 3.4 percent of GDP in 1970 to 4.0 percent in 1971 (see table 6.1). The deterioration in the balance of payments was partly the result of a slowdown in world economic growth and partly the result of the real appreciation of the rupiah caused by the relatively higher inflation rate in Indonesia during 1969 and 1970.

6.3 Phase 2, August 1971 to October 1978 The exchange rate remained at 415 rupiahs/dollar for a record seven years. The reason for this remarkable stability was straightforward: the balance of payments was very strong throughout the period. The current account deficit during phase 2 stayed below the 1971 figure of 4 percent of GDP (see table 6.2). The largest current account deficit occurred in 1975, 3.6 percent of Table 6.1

Background to the August 1971 Devaluation (in percentages)

Current account balance to GDP Industrial countries’ GDP growth rate Industrial countries’ inflation rate Indonesia’s inflation rate

I969

1970

1971

1972

-4.0 4.9 4.1

- 3.4

-4.0 3.5 5.2 4.4

-3.0 5.2 4.6 6.4

17.4

Source: International Financial Sfatistics, 1986 yearbmk.

2.7 5.6 12.3

99

IndonesidChapter 6

Table 6.2

Economic Conditions During Rriod of Fixed Exchange Rate, August 1971to November 1978 ~~

Current balance as %of GDP Current account receipts as % of GDP Nongold reserveslimports (in weeks) GDP growth rate (8) Inflation rate (%)

1970

1971

1972

1973

1974

1975

1976

1917

1978

-3.4

-4.0

-3.0

-2.9

2.3

-3.6

-2.4

-0.1

2.7

13.6

14.8

17.2

20.7

29.1

23.1

23.6

23.9

22.0

8.1 6.5 12.3

8.7 7.0 4.4

19.1 9.4 6.4

15.3 11.3 31.0

20.2 7.6 40.6

6.4 5.0 19.1

13.7 6.9 19.8

20.9 8.9 11.0

20.4 7.7

8.1

GDP, when the rosy economic prospects induced by the oil boom caused both private and public spending to soar. The proportion of national income from current account receipts leaped from the 1970-72 average of 15 percent to an average of 24 percent during 1973-78. The fear of a balance-of-payments crisis was never further away from the authorities’ minds; there was little fear of not being able to accommodate any short-run speculative flight. The level of nongold reserves, measured as the number of weeks the nongold reserves could sustain existing import levels, was consistently higher than the 4.8 weeks of the 1967-69 period and the 12 weeks of the 1970-72 period^.^ The impressive balance-of-payments performance is largely due to the rapid development of the petroleum and LNG sectors and the fourfold oil price increase at the end of 1973. Besides the absence of a balance-of-payments reason for changing the exchange rate, the macroeconomic conditions of this period also did not warrant any additional stimulus which a devaluation would surely bring. The sustained high income growth rates of this period were without precedent in the Soekarno years. This high average income growth rate of 7.9 percent was achieved with substantial overheating of the economy-the average inflation rate was 22 percent compared with the 8 percent of 1970-72.

6.4 The November 1978 Devaluation For most observers the devaluation of the rupiah on 15 November 1978 was a surprise. While it was generally agreed that the real exchange rate had appreciated significantly since 1971 and that there would be a need in the future, when oil reserves were closer to depletion, to devalue in order to boost nonoil exports, there was little expectation of an immediate deval~ation.~ There were no signs that the balance of payments was deteriorating-the current account deficit was 2.7 percent of GDP in 1978 and 0.1 percent in 1977, compared with 2.4 percent in 1976 and 3.6 percent in 1975. There were, in fact, numerous speculations in the Indonesian press during April and May 1978 that a revaluation of the rupiah might be

100

Wing Thye Woo and Anwar Nasution

necessary given the plunge of the dollar vis-6-vis the other major currencie~.~ Many explanations have been offered for the timing of the exchange rate realignment but they all tend to be combinations, with different emphasis, of two main interpretations. In the first interpretation the November 1978 devaluation is seen as an anticipatory action to the inevitable dropoff in oil export earnings due to resource depletion.6 Arguments in support of the anticipatory devaluation interpretation are: (1) that it is better to act before a balance-of-payments crisis actually develops because this would prevent the financial chaos attendant upon a speculative outflow of domestic capital; and (2) since exports and imports react to relative price changes with substantial lags, a devaluation during a balance-of-payments crisis would have to be larger than is really necessary in order to have any immediate beneficial effects. The second interpretation emphasizes the economic difficulties and political tensions associated with the reallocation of resources being forced upon the economy by the overvalued exchange rate. The overvaluation of the rupiah was the result of maintaining the exchange rate at 415 rupiahs/ dollar despite the large domestic inflations from 1974 to 1977. This meant that Indonesian producers of tradables were experiencing a profit squeezethe prices of their output were fixed by international competition, but the prices of their domestic inputs were being driven up by the double-digit inflation. The result was reports of increasing unemployment in the tradables industries, particularly in the labor-intensive agricultural export sector. The stagnation of the manufacturing export industries was worrying because, being of a labor-intensive nature, they were looked upon as the means to soak up the natural increases in the labor force. Indonesia was suffering from the Dutch disease. The growth of the extremely capital-intensive oil industry caused the real exchange rate to appreciate, hence decimating the labor-intensive tradables industries. Since the oil industry constituted an enclave export sector with minimal linkages to the rest of the economy, the steady movement of resources into the service (nontradables) industries was a threat to the long-run growth rate of the economy. The movement of resources out of the rural sector was hastened by another development. Protection was increasingly granted to importcompeting industries in order to offset the profit squeeze caused by the overvalued rupiah, and this protectionist policy deteriorated the rural-urban terms of trade. Politically, the distress in the rural sector was undermining the efforts of the Soeharto regime to prevent the resurgence of the PKI, and it was also raising inter-island tensions because the Outer Islands depended heavily on the export of agricultural products. Given these economic and political costs of the Dutch disease, it was therefore not surprising that the government devalued, even in the absence of a balance-of-payments crisis.7 Max Corden (1982) has aptly labelled the use of the exchange rate to protect

101

IndonesidChapter 6

the tradable sector for reasons unrelated to balance-of-payments considerations as “exchange rate protection.” The anticipatory devaluation interpretation and the Dutch disease interpretation could both be right. They do not contradict each other. It must be mentioned, however, that anticipatory devaluations are extremely rare events, not only by the past experiences of other LDCs but also by Indonesia’s own history of devaluations. Devaluations, including the Indonesian ones prior to 1978 and those of 1983 and 1986, usually occur either in the midst of a balance-of-payments problem or when one is imminent. And as shown in table 6.2, the level of foreign reserves in 1978 would have been able to sustain the existing amount of imports longer than at any time during the 1970-76 period. Further, the anticipatory devaluation explanation can be judged plausible only if it explains why the technocrats reacted quite differently in 1978 toward potential balance-of-payments problems than at any time before or after 1978. It must be admitted, however, that the alleged deleterious effects of the Dutch disease on the nonoil export sector are not obvious. Nonoil, nonLNG exports, whether measured in physical units or in dollars or in the units of imports for which they can be exchanged, show steady growth throughout the Dutch disease period of 1972-78 (see the first three columns in table 6.3). The 1975 dip in export earnings is due to a recession in the industrial hrfomance of Nonoll NonLNG Exports (1974 = 100)

Table 6.3

Physical Volume NA NA 73.9 83.4 96.3 100.0 99.6 111.9 121.0 118.0 160.0 144.5

1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

U.S. Dollars

Foreign Purchasing Power

Domestic Purchasing Power

52.2 58.5 59.4 60.2 91.1 100.0 74.2 103.8 133.0 123.0 162.4 155.9

51.6 60.1 66.9 73.7 103.2 100.0 69.5 80.8 100.9 103.6 184.6 170.8

28.6 33.6 36.0

40.0 73.2 100.0 82.6 115.2 159.7 166.4 253.7 276.4

Memo item: Average annual growth rate of nonoil exports in 1973-78 period. Exports in US. dollars

Growth rate (70)

Indonesia

Malaysia

Thailand

South Korea

Hong Kong

Singapore

15.5

32.3

19.9

30.6

19.0

20.0

Nore: Physical volume from deflating rupiah value series by nonoil export price index. Foreign purchasing

power from deflating U.S. dollar value series by export unit value of industrial countries. Local purchasing power from deflating rupiah value series by Indonesian CPI. The CPI contains prices of imported consumption goods in its construction. NA

=

not available

102

Wing Thye Woo and Anwar Nasution

countries rather than to a fall in domestic production. The fact that the value indices of columns 1, 2, and 3 went up in 1979 and 1980 only shows that a devaluation is effective in increasing supply rather than that there was stagnation in the nonoil export sector. The production disincentive faced by the nonoil export industries is clearly seen only when one measures the amount of local purchasing power which their exports are able to command (see column 4 of table 6.3). Even though the nonoil exports were bringing in increasing amounts of foreign goods, the steady real appreciation of the exchange rate meant that the nonoil export industries were not being paid a greater number of baskets containing the mix of goods typically consumed by Indonesians. The first three measures show that total export earnings in 1976 and 1977 were at unprecedented heights, whereas the fourth measure puts the 1976 and 1977 earnings below that of 1973. In terms of foreign purchasing power (column 3), the nonoil export industries increased their revenues by 32 percent between 1973 and 1978, but their revenues were unchanged if measured in local purchasing power. Another indicator that the Indonesian tradables sector was suffering from the Dutch disease is its poor growth performance compared to Malaysia and Thailand, which exported similar products (see memo item in table 6.3). The respective annual growth rates of nonoil exports over the 1973-78 period for Indonesia, Malaysia, and Thailand were 16 percent, 32 percent, and 20 percent, respectively. The most telling comparison is with Malaysia, which had an oil boom like Indonesia and which also kept its currency fixed (almost) to the dollar. The big difference was that the average annual inflation rate for Malaysia was 7.5 percent as against Indonesia’s 21.6 percent. Table 6.4 gives the prices of the five largest agricultural exports, measured in different ways. Other than the price of coffee, the other agricultural prices move more or less in tandem. Again, because of real exchange rate appreciation, the foreign purchasing power measures (item a) of palm oil, rubber, log, and plywood prices gave a less bleak picture of the 1974-78 period than the local purchasing power measure (item b). The average fall in prices was 12 percent by the first measure, but 29 percent by the second measure. Local purchasing power is the relevant measure for assessing the degree of profit squeeze on the smallholders. To smooth out individual price deviations, as in the case of coffee prices, in table 6.4, aggregate price indices were constructed to study overall movements in the ratio of prices of tradables to prices of nontradables, PT/PN (see table 6.5). Two proxies for the ratio were obtained by (1) normalizing the output price indices of several sectors by the CPI and (2) normalizing the sectoral prices by the housing component of the CPI.8 Since housing cost is a more direct proxy for nontradables, we would expect the second ratio to move more than the first.

103

IndonesidChapter 6

Table 6.4

Commodity Price Indices (1974= 100) Coffee

Palm Oil

Rubber

Logs

Plywood

56.7 74.4 65.7 74.2 91.5 100.0 106.7 208.9 337.1 228.1 249.4 221.8

28.0 42.9 43.4 36.9 47.8 100.0 94.6 71.3 102.0 106.3 118.7 105.8

68.0 57.1 43.6 41.3 83.3 100.0 75.3 102.0 110.3 131.0 170.3 194.5

47.7 52.9 53.1 49.8 83.5 100.0 82.8 112.8 110.1 112.3 197.0 236.7

55.3 67.5 53.7 63.4 124.3 100.0 79.8 96.7 105.8 124.1 171.9 179.3

124.1 99.4 71.9 62.2 103.7 100.0 67.6 91.9 91.9 96.8 109.1 109.7

87.1 92.2 87.5 74.8 103.9 100.0 74.4 101.7 91.8 83.0 126.1 133.5

100.9 117.5 88.6 95.4 154.7 100.0 71.6 87.1 88.1 91.7 110.0 101.1

(a) In U.S. dollars

1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

(b) In terms of real exports from industrial countries 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

103.5 129.4 108.4 111.6 113.9 100.0 95.8 188.2 280.9 168.6 159.7 125.1

51.1 74.7 71.6 55.5 59.5 100.0 84.9 64.2 85.0 78.6 76.0 59.7

(c) In terms of baskets of domestic:ally consumed goods (real rupiah prices as given 1969 102.3 50.5 122.7 1970 133.0 76.7 102.1 1971 122.0 80.6 81.0 1972 136.7 67.9 76.1 1973 129.0 67.4 117.5 1974 100.0 100.0 100.0 1975 89.8 79.6 63.4 1976 146.6 50.0 71.6 1977 213.1 64.5 69.7 1978 134.4 62.6 77.2 86.4 123.9 1979 181.5 1980 137.1 65.4 120.2

by CPI) 86.2 94.7 98.5 91.7 117.7 100.0 69.7 79.2 69.6 66.2 143.3 146.2

99.7 120.6 99.7 116.9 175.2 100.0 67.1 67.8 66.9 73.1 125.0 110.8

In our opinion, the most reliable indicators of PT/PN are the normalized wholesale price indices of imports and of nonoil exports. This is because: 1. Beginning in 1974, some segments of the manufacturing sector started receiving quantitative restrictions on imports to protect domestic industries hurt by the real appreciation and to promote import substitution. Goods sheltered by quantitative restrictions are effectively nontradables from the analytical viewpoint. This is because imports cannot enter, regardless of the spread between domestic and international

104

Wing Thye Woo and Anwar Nasution

prices. With a given quota, prices of the domestically produced substitutes are insulated from international price movements and move only in response to changes in domestic demand and cost conditions.' 2. Rice is a very large portion of domestic agricultural output, and its price has been deliberately set to increase slightly more than CPI movements in order to promote the goal of self-sufficiency in rice. Prices of a number of other food crops such as corn, soybean, and sugar are also protected from external competition. This means that tree crops are the main tradable component of the agricultural sector. Part (a) of table 6.5 shows that, except for agriculture, wholesale prices normalized on the CPI show a downward trend from 1973 to 1978, indicating pressure on producers of tradables to shift to nontradables. We suspect that it is for the reasons given above that the production incentive in the manufacturing sector fell less than in the general tradables sectors. Judging from the normalized export and import indices, the incentive to produce tradable as against the general basket of consumption goods fell 22 to 27 percent between 1973 and 1978." When prices were normalized by the nontradable price variable, housing cost, all series showed a downward trend (see part b of table 6.5). Again, the differences in the decline of the series, PT/PN, reflect the fact that the food component of agriculture is essentially a nontradable because of the self-sufficiency goals, and that the manufacturing sector has been receiving increasing quota protection over the period. In this direct measure of FT/PN, the production disincentive increased by about 26 percent in the 1973-78 interval. Table 6.5

Indicators of Tradable-NontradablePrice Ratio, PT/PN (1974= 100) 1980

Wholesale F'rices

1971

1972

1973

1974

1975

1976

1977

197V

1979

(a) Relative to Jakarta CPI Imports Exports, nonpetroleum Agriculture Manufacturing

105.9 90.2 90.2 103.9

111.1 88.9

107.0 107.0 102.8 114.1

100.0 100.0 100.0 100.0

91.6 69.7 98.3 89.9

81.8 72.0 102.8 88.1

76.7 83.0 113.2 88.1

78.3 84.3 116.9 92.2

87.0 110.2 120.8 94.9

84.8 112.9 127.7 97.7

(h) Relative to housing component in Jakarta CPI Imports 75.0 82.2 91.6 Exports, nonpetroleum 63.9 65.8 91.6 Agriculture 63.9 74.0 88.0 Manufacturing 73.6 79.5 97.6

100.0 100.0 100.0 100.0

87.2 66.4 93.6 85.6

74.5 65.6 93.6 80.3

66.3 71.7 97.8 76.1

65.3 70.4 97.5 16.9

73.7 93.3 102.4 80.4

70.5 93.8 106.2 81.2

(c) Morgan Guaranty's competitiveness measure

100.0

87.3

74.8

74.2

79.6

111.3

101.1

114.1

100.0 107.4

127.1

120.3

-

Note: Part (a) and (b) are calculated from table 3 in Wan (1986), but some of our calculated numbers differ from the

calculations in his table 4. The series in this table and those in table 6.7 are not comparable because the definition of wholesale price index changed over the two periods. The competitiveness measure is from inverting the Morgan Guaranty real exchange rate. "January to October 1978

105

IndonesidChapter 6

6.5 Effects of the 1978 Devaluation The 50 percent devaluation caused a much bigger jump in the normalized nonoil export prices than in the normalized import prices; on average, a 24 percentage point jump in the former versus 8 percentage points in the latter. The minor improvements in the normalized manufacturing and agricultural prices may reflect the significant use of quantitative restrictions (QRs) which have pushed these goods closer to the nontradables category. The 24 percentage point improvement in relative prices for the tradable sector may not be an exaggeration because the rise in the Morgan Guaranty measure of competitiveness was even more substantial-32 percentage points. l 2 The speed and size of the response of nonoil, nonLNG exports were extremely impres~ive.'~ The devaluation happened at the end of 1978, and the expansion of nonoil exports in 1979 was considerable according to all four of the earning measures used in table 6.3. Export volume went up by 36 percent in one year, raising dollar earnings by 52 percent. Measured in units of exports from industrial countries, the value growth from the 1978 level was 32 percent in 1979 and 27 percent in 1980. This growth in foreign purchasing power translated into domestic purchasing power increases of 78 percent in 1979 and 65 percent in 1980. The nonoil export response caused the nongold reserves of the central bank to swell from twenty weeks of imports at the end of 1978 to twenty-six weeks in the third quarter of 1979, just before the OPEC-2 October price increase further boosted official reserves. What makes this export achievement particularly notable is that 1979 was the beginning of the slide into the deep recession of 1982. The real GDP of the industrial countries grew only 3 percent in 1979 and 0.6 percent in 1980, compared to a 1976-78 average of 4.4 percent. While the big response by nonoil exports may not be surprising given the large devaluation, what may not have been expected at the outset was the speed of the response. This is because the bulk of Indonesian nonoil exports is agricultural raw materials and minerals, with manufactured exports averaging only 6.7 percent of nonoil exports in the 1972-78 period. l4 Since the supply of both of these primary commodities is typically assumed to be inelastic in the short run, either because of their long gestation periods (tree crops) or their heavy capital-intensive nature (minerals), the 36 percent expansion in the physical volume of nonoil exports in 1979 was bewildering. The quick response is proof of the Dutch disease. The mounting severity of the Dutch disease since 1974 caused an increase in excess capacity in the traditional export industries. Small producers of tree crops were spending more and more of their time in nontradable activities as the real prices of their agricultural products sank with the maintenance of a constant nominal exchange rate in the face of big domestic price increases. Another source of the excess capacity in the agricultural raw materials sector was that, given the low real prices, producers were not fully exploiting the now matured

106

Wing Thye Woo and Anwar Nasution

trees planted in the early 1970s in the wake of the stabilization and rehabilitation of the economy. Hence, production was easily increased when PT/PN was improved by the 50 percent devaluation. The above point is very well brought out by table 6.6 which shows the value of nonoil exports before and after the November 1978 devaluation. Total nonoil exports, measured in dollars, went up by 54 percent in fiscal 1979.l5 The rates of increase for the seven biggest nonoil export items were all at the two-digit level: timber, 91 percent; rubber, 42 percent; palm oil, 16 percent; coffee, 41 percent; animals and animal products, 19 percent; and manufactured goods, 82 percent. The sizable expansion of the tree crop Table 6.6

Nonoil, NonLNG Exports of Indonesia

Product

1977178

I978179

I979180

(a) In millions of U.S.dollars Timber Rubber Palm oil Coffee Tea Tobacco Pepper Palm kernel Copra cake Tapioca Other foodstuffs Animals and animal products Tin Copper Other minerals Manufactured products Total value of nonoil exports

943.0 608.0 202.0 626.0 120.0 59.0 62.0 5.0 33.0 13.0 48.0 179.0 253.0 74.0 36.0 245.0 3,506.0

I , 130.0 774.0 221.0 508.0 98.0 58.0 66.0 2.0 34.0 28.0 65.0 214.0 324.0 64.0 49.0 361.0 3,996.0

2,166.0 I.101 .o 257.0 715.0 91.0 60.0 46.0 12.0 65.0 59.0 79.0 255.0 388.0 95.0 126.0 656.0 6,171.0

-

(b) Rate of change from preceding year ( I ) Timber Rubber Palm oil Coffee Tea Tobacco Pepper Palm kernel Copra cake Tapioca Other foodstuffs Animals and animal products Tin Copper Other minerals Manufactured products Total value of nonoil products Memo item: Manufactured exports as proportion of total nonoil exports

19.8 27.3 9.4 - 18.8 - 18.3 -1.7 6.5 -60.0 3.0 115.4 35.4 19.6 28.1 - 13.5 36.1 47.3 14.0

7.0

9.0

91.7 42.2 16.3 40.7 -7.1 3.4 -30.3 500.0 91.2 110.7 21.5

19.2 19.8 48.4 157.1 81.7 54.4

10.6

107

IndonesidChapter 6

exports-rubber, palm oil, and coffee-and tin exports testified eloquently to the presence of excess production capacity in these industries prior to 1979. The large reaction of manufactured exports was particularly gratifying to the technocrats who had advocated the devaluation because they viewed the labor-intensive manufactured industries as a crucial sector for Indonesian industrialization and the creation of employment. l 6 6.6 The March 1983 Devaluation

Management of the exchange rate after the 1978 devaluation was much more flexible; the rupiah glided gently downward against a basket of currencies to compensate for the higher inflation rate in Indonesia. The OPEC-2 shock in November 1979, however, unleashed external and internal forces which led ultimately to the 38 percent rupiah devaluation in March 1983. Specifically, the doubling of oil prices provoked the industrial countries to tighten their monetary policies to ward off the cost-push inflation, and the result was three years of negligible growth with its nadir in 1982 when the world experienced its deepest recession since the Great Depression. At the same time, the great inflow of oil revenue caused the Indonesian government to augment its investment spending as dictated by the balanced budget rule. Because of the primitive state of domestic financial markets, ruling out the use of open market operations by the central bank, the conversion of the dollar-denominated oil revenue by the government into rupiah expenditure led to an explosion of the money supply. As in the aftermath of the OPEC-1 shock, Bank Indonesia temporarily lost control of the money supply. Reserve money grew by 28 percent in 1979 and 40 percent in 1980. The result was that the expected one-time price level increase due to the November 1978 devaluation was given new momentum; the inflation rate was 18.5 percent in 1979 and 12.2 percent in 1980. The inflation rate would have been higher if stagnation in the industrial countries had not exerted a moderating effect on prices via lower import prices. The internal shock of high inflation, which appreciated the real exchange rate, and the external shock OF low OECD growth caused the balance of payments to deteriorate. The slowdown of the OECD economies shifted the demand for Indonesian exports downward, and the real exchange rate appreciation decreased the supply of nonoil exports. The import price index normalized by housing cost went from 74 to 66 in the 1979-82 period, while the Morgan Guaranty competitiveness index declined from 111 to 80; a fall of 10 and 20 percent, respectively (see table 6.7). The consequence of these internal and external shocks was that both the volume and real domestic value of nonoil exports in 1982 were half of their 1979 level. The current account deficit was a record 6 percent of GDP, with reserves falling to ten weeks of imports.

108 Table 6.7

Wing Thye Woo and Anwar Nasution Background to Devaluations of 1983 and 1986

(a) General economic conditions Income growth in industrial nations ( W ) Real price of oil in foreign purchasing power (1980= 100) Indonesian inflation rate ( W ) Indonesian growth rate ( Q )

1979

1980

1981

1982

1983

1984

1985

3.4

1.3

1.4

67.7 20.6 6.3

100.0 18.5 9.9

119.2 12.2 7.9

27.5 1.9 29.3

27.9 4.0 25.9

13.5

65.2 99.0 111.3 160.0 184.6 162.4

1986

-0.4

2.7

4.7

3.0

2.4

124.1 9.5 2.2

111.6 11.8 4.2

109.1 10.5 6.6

104.1 4.7 1.1

59.1 5.9 2.4

25.2 -0.7 19.6

20.8 -5.9 9.7

23.1 -7.8 11.8

24.3 -2.5 17.9

21.4 -2.1 20.5

21.4 -4.3 23.0

7.9

8.2

10.6

12.8

14.7

20.1

29.3

62.8 101.0

63.0 91.1 89.0 90.2 111.8 118.1

56.9

63.2 105.9 98.4 100.9 148.1 142.6

59.8 102.9 91.8

58.2 98.4 92.7 129.2 188.5 177.0

57.7 98.9 109.7 142.7 222.4 168.2

(b) Balance-of-payments situation Merchandise exports to GDP (%) Current account balance to GDP ( 8 ) Reserves to imports ratio (weeks) External long-term public debt service 10 exports (%I (c) Nonoil export sector (1974= 100) Import price deflated by housing cost Export price deflated by housing COSI Competitiveness B la Morgan Guaranty In physical volume In local purchasing power In foreign purchasing power

101.1 144.5 170.8 155.9

80.5 79.5 80.3 93.5 107.3

115.5

175.0 169.3

Note: IWPN proxies for 1983 are for post-March devaluation and for 1986 are pre-September devaluation. PT/PN proxies here are not comparable to those in table 6.5 because of changes in the definition of price indices. This is why the 1979 and 1980 figures in this table are different from table 6.5.

The balance-of-payments picture worsened in the first quarter of 1983. Imports continued to grow with no sign of export recovery, and capital outflow started accelerating. The category of errors and omissions, into which the official balance-of-payments account put all private portfolio capital flows, soared from -$0.6 million in 1979 to over -$2.0 billion annually during 1980-82. The weak export earning together with this avalanche of capital outflow caused total nongold reserves to fall to 5.3 weeks of imports by the end of the first quarter of 1983. The grim balance-of-payments picture at that point was definitely the reason for the 38 percent devaluation. An economic stimulus coming at the time when the economy was growing at 2.2 percent was an added incentive to devalue. Because Mexico was unable to meet its debt service in August 1982 and two other borrowers, Argentina and Brazil, were on the brink of a debt crisis, it was prudent for Indonesia to take some preventive measures, especially since the price of oil was moving downward. It seems likely that the external debt did not play more than a cautionary role in the government’s decision to devalue in 1983. Although the external public debt-service ratio rose from 8 percent in 1980 to 11 percent in 1982, it was still well below the 1981 (pre-crisis) Mexican debt-service ratio of 28 percent. l7

109

IndonesidChapter 6

The March 1983 devaluation restored PT/PN back to the level set by the devaluation of November 1978. The response of the nonoil export sector was impressive, as in the previous devaluation-exports expanded 26 percent in physical volume and 58 percent in local purchasing power. The reason the 1983 nonoil export levels (in real terms) were significantly lower than the 1979 levels, even though the value of the real exchange rate was the same in both instances, is that foreign demand was much lower in 1983 than in 1979. Real GDP of industrial countries grew 2.7 percent in 1983, compared to 3.4 percent in 1979. Nevertheless, the increase in nonoil exports was enough to shrink the current account deficit to 2.5 percent of GDP in 1984. Manufacturing exports grew especially rapidly, jumping from $850 million in fiscal 1982 to $1,480 million in fiscal 1983, and then to $2,166 million in fiscal 1984 (see table 6.8). It is noteworthy that a greater variety of manufactured goods were being exported because of the favorable PT/PN. Manufactured goods in the “others” category shot up by 300 percent in real terms in just two years. This reaction of the manufacturing sector strongly indicates that export-oriented industrialization is a real possibility as long as favorable relative prices are maintained through appropriate exchange rate and trade policies. It must be emphasized that the government supported the 1983 exchange rate devaluation with conservative macroeconomic policies. Both monetary and fiscal policies were tightened. The latter was done by massive postponement of capital-intensive (hence, import-intensive) projects and by increasing tax revenue through streamlining a cumbersome tax system. The government budget deficit went from Rp 13 billion in 1982 to Rp 10 billion in 1983, and then to Rp 0.5 billion in 1984.” The conservative macroeconomic policies succeeded in keeping inflation to just a shade over 10 percent in 1983 and 1984, hence preventing a fast reversal of the real depreciation of the exchange rate. The government may also have sought to improve the trade balance directly by rapidly expanding the list of import items subject to quotas. While imposition of quotas may be due largely to the efforts of infant industries advocates and rent-seekers, the timing of the flood of quantitative Table 6.8

The Response of the Manufacturing Sector to the March 1983 Devaluation (in millions of dollars)

Plywood Textiles and clothing Electrical appliances (e.g., T.V.. semiconductor, transistor) Others (e.g., fertilizer, cement, iron and steel, floor coverings) Total manufactured exports Source: World Bank (1985, 20; 1986, 91; 1987a, 16).

1982183

1983/84

1984/85

320 I80 1 LO 240 850

580

697 519 I35 815 2,166

360 130 410 1,480

110

Wing Thye Woo and Anwar Nasution

restrictions seems to suggest a role for balance-of-payments considerations as of November 1982. The increased protectionism may help the balance of payments in the short run, but the quota form of protectionism (for reasons given in section 6.8 below) is counterproductive in the medium and long run. Quantitative restrictions could very well have contributed substantially to the need for another devaluation in September 1986.

6.7 The September 1986 Devaluation The world economy showed no signs of returning to sustained economic growth after the deep 1982 recession. There was a spurt of activity in 1984 (see table 6.7) which was normal after such a deep recession and was helped along by the large U.S. budget deficits. But U.S. budget deficits could not keep on widening indefinitely in order to provide the same stimulus. In 1985 the growth rate in the industrial countries slowed down to 3.0 percent. After the middle of 1986 it was clear that immediate growth prospects were lower than anticipated. The IMF revised its growth rate projections for industrial countries downward, from 3.0 to 2.7 percent for 1986, and from 3.2 to 3.1 percent for 1987 (IMF 1986b, 1986~).The actual 1986 growth rate turned out to be even lower than the midyear forecast; it was 2.4 percent.19 For Indonesia, the slow global economic recovery translated directly into uncharacteristically low oil and commodity prices. The average oil price in fiscal 1985 was $25 per barrel, and it fell to $13 per barrel in fiscal 1986.*’ The nonoil terms of trade also turned harshly against Indonesia with the result that even though nonoil exports increased by 10 percent in physical volume, their value decreased by 5 percent in terms of foreign purchasing power. Despite the adoption of stringent macroeconomic policies and the steady floating down of the exchange rate from 970 rupiahs/dollar in 1983:lQ to 1,131 rupiahs/dollar in 1986:2Q, the current account deficit doubled to 4.3 percent of GDP from 2.1 percent. Added to the balance-of-payments problems of 1986 was a quickening of the rise in the external public (medium- and long-term) debt-service ratio since 1984. The 1986 debt-service ratio stood at 29 percent, the same level as the Mexican debt-service ratio in 1981. The primary factors behind this drastic rise were the export collapse, which decreased the denominator, and the “uncontrollable” increased debt payments, which increased the numerator. The increased debt-service payments were termed ‘‘uncontrollable” because since less than 30 percent of Indonesia’s external debt is denominated in U.S. dollars, the drastic drop of the dollar against the other currencies accounted for more than 70 percent of the $1.1 billion increase in annual debt service over the 1984-86 period (World Bank 1987a). Given that Japan is Indonesia’ biggest creditor (36 percent of debt), as well as its biggest trade partner, the 29 percent dollar depreciation against the yen in 1986 put Indonesia in the uncomfortable position of receiving dollars for its

111

IndonesidChapter 6

chief export, oil, and paying yen for its imports and for a third of its debt service. With the worsening of the trade balance, the growing shadow of a debt crisis, and the slowing down of economic activities, the September 1986 devaluation of 45 percent against the dollar was the single most effective step Indonesia could have taken to simultaneously improve its capacity to earn foreign exchange and stimulate its economy.

6.8 Relative Prices and the 1986 Devaluation While it is clear that negative external demand shock had a role in worsening the balance of payments, we want to point out that there were also internal developments over this period which caused substantial movements in relative prices unfavorable to the tradable sector. Specifically, the widespread use of QRs in the 1980s depressed the outputhnput price ratio faced by the tradable sector and caused the supply of nonoil exports to fall. Our point is that the introduction of QRs on the imported inputs of the tradables sector transfers part of the profits previously received by the producers of tradables to the holders of the input quota. QRs on inputs and real exchange rate appreciation are alike in that they both cause a profit squeeze in the tradables sector.21 The use of QRs has a long history. (QRs take the form of either import licenses or quotas.) The first use of QRs in the Soeharto regime was in 1970, and by the end of 1971 twenty-four items were under this form of protection. The use of nontariff barriers increased significantly throughout the 1970s, partly in response to the Dutch disease squeeze on the profit margins of the import-substituting industries.22 In November 1982, when the balanceof-payments situation looked precarious, the use of QRs accelerated.23 By 1985 QRs were undoubtedly the dominant form of protection in Indonesia. The pervasiveness of QRs is very well shown in table 6.9. Of the 5,229 items imported in 1985, 1,484 required import licenses and 296 were under quotas. The import licenses were usually given to only two or three traders or to the few firms producing the competing goods domestically. The method of license issuance effectively conferred monopoly status on the license recipient. Quotas spanned the whole spectrum from zero to a discretionary quantity decided by a bureaucrat at the time the import application was submitted. License restrictions covered 30 percent of total import value. The range of activities protected by import licenses accounted for 32 percent of total domestic value added (excluding construction and services).24 If the petroleum sector, which requires no protection, is also excluded, then the coverage is 53 percent of total domestic value added. The types of goods under QRs is very diverse, ranging from basic inputs to consumption goods. Basic inputs under monopoly import licenses include cold-rolled steel sheets, key chemicals for making plastics, and tin plate.25It

112

Wing Thye Woo and Anwar Nasution

Table 6.9

Import Licensing in 1985 Agriculture

Number of CCCN items Total Under license Under quota Import value" Total Under license Share of value addedb In sector Under license

Minerals

1,024 122

139 2

64 127 170

40.6 21.4

Manufacturing

Total

1

4,066 1.360 23 1

5,229 1.484 296

1,451 I

8,082 2,539

8,987 2.710

42.4 .I

17.0 10.3

100.0 31.8

Source: Central Bureau of Statistics, Indonesia, and World Bank staff estimates. Nore: CCCN = Customs Cooperation Council Nomenclature.

"In millions of U.S. dollars. Based on CBS data for 1985, which vary from balance of payments estimates for 1985186 "In percentages. Based on the 1980 input-output table. Excludes construction and services. Due to differences in classification, these numbers may vary from national account estimates.

is clear that protectionism is not always extended for infant industry reasons; for example, there is no domestic producer of cold-rolled steel. In the case of plastic inputs (where there is one domestic producer), the monopoly importer imposed an administrative fee for each raw material which amounted to about 18 percent of its value, resulting in a 30 to 40 percent rise in costs to the end users.26 The implication of this microeconomic distortion for exchange rate management is profound. The intrusion of this distortion since late 1982 and its quick metastasis across the tradable sector renders invalid drawing conclusions from the movements of the macroecnomic proxies for PT/PN, as we did for table 6.5. This is because the introduction of quotas on inputs to the tradables sector reduces the sector's production incentive for a given PT/PN. Output prices of tradables are set by international competition, while output prices of nontradables (which are generally very labor-intensive) are set by the domestic cost structure whose level is determined by, primarily, domestic wages on the supply side and domestic macro conditions on the demand side. Hence, the introduction of a quota on an imported input to the tradables sector will reduce the profitability of the tradable sector without any change in the proxies for PT/PN. The fact that in table 6.7 the two proxies for PT/PN in predevaluation 1986 are at least as favorable as in postdevaluation 1979 does not imply that the production incentive (measured in terms of local purchasing power) has not worsened if we abstract from demand conditions. The economic effects of a QR can be modelled by the addition of another cost, henceforth called rent, to the production of the good. The limit of the rent is determined chiefly by the shape of the demand curve for the output

113

IndonesidChapter 6

and by the cost of ~muggling.~’The existence of this rent imposes a potential check on the ability of devaluation to restore international competitiveness. This is because the effectiveness of a devaluation in boosting production of tradables depends crucially on its ability to raise the domestic output price without a corresponding rise in the domestic cost of nontraded domestic inputs. As a first approximation, a devaluation works by increasing the real profits of the tradable sector by cutting the real wage (which is more easily achieved if austere macro policies are undertaken simultaneously). We can think of the QR-introduced rent as a payment for a nontradable input service. Since there is no competitive determination if this rent, its level is at the discretion of the monopoly import license holder. How the license holder reacts to a devaluation determines the effectiveness of the devaluation in boosting production of tradables. If the license holder keeps the rent constant, either in nominal terms or in local purchasing power, then production of tradables will increase as long as the costs of other nontraded inputs fall in terms of local purchasing power. However, if the license holder increases his rent so that the loss of other nontraded inputs is entirely transferred to him, then the production level will remain unchanged. As a practical matter, it may be reasonable to assume that the license holder is usually not able to scoop all of the “released payments,” and hence devaluation would in most cases increase the output of tradables. This means that a devaluation in the presence of QRs will have to be larger than one undertaken in their absence in order to achieve the same output response. We can say that the August 1986 real exchange rate was overvalued in the sense that the introduction of QRs caused a drop in the supply of nonoil exportables which a devaluation would be able to offset. The point we want to emphasize is that although the Morgan Guaranty competitiveness index in predevaluation 1986 shows almost the same value as in postdevaluation 1979 (110 versus l l l ) , it does not indicate that the August 1986 exchange rate was not overvalued. In order to have the 1986 nonoil export supply schedule in the same position within the familiar Marshallian price-quantity space as in 1979, a devaluation was clearly warranted in light of the shrunken gap between output and input prices.28 It is of course an empirical question how much the additional nonoil export earnings would have been in the absence of QRs, especially in comparison to the fall in oil export earnings. The current account deficit would still have widened in 1986, but it may not have doubled as it did.29

114

7

Wing Thye Woo and Anwar Nasution

External Debt Management

7.1 Introduction In the preceding chapters we identified historical and political economy factors and showed how they have influenced economic policymaking in Indonesia. The distributional impact of fiscal policy (chapter 4) was the outcome of the interaction between the comparative-advantage-based economic program of the technocrats and the political concerns of President Soeharto. The pattern of allocation of subsidized credit (chapter 5) and the granting of monopoly import licenses (chapter 6) showed clearly the need to cater to both economic nationalist aspirations and the rent-seeking activities of key political allies. In this chapter, we will see how these economic policies (especially fiscal policy) have affected the accumulation of external debt and how the resulting Pertamina crisis set the tone for external debt management after mid- 1975.

7.2 Trial by Fire The Soeharto government is no stranger to external debt management, having inherited an external public debt of $2 billion. It cut its teeth on the economic stabilization and rehabilitation program of 1966, within which rescheduling the Soekarno debts and arranging for new capital inflows to support the balance of payments were key components. The situation was grim. Indonesia had defaulted on its 1965 debt service because the swingeing current account deficits of 1961-65 had completely drained the foreign exchange reserves and turbulent domestic conditions had caused the country to be shut out from the private external credit market. Faced with debt payments (including arrears) of $530 million for 1966, which was 70 percent of GDP and 132 percent of exports, Indonesia requested a meeting with its debtors to organize her debt.' The result of the debt rescheduling meetings that year in Tokyo (September) and in Paris (December) was that the major Western countries gave Indonesia the following: 100 percent relief from principal and interest payments on credits of more than 180 days, related to contracts effective prior to July 1, 1966 [and which were due in 1966 and 19671. The new schedule of payments [was] to start January 1, 1971, after a 4-year grace period, and the rescheduled or refinanced amount [was] to be repaid over a period of eight years on an ascending scale starting at 5 percent in 1971 and reaching 20 percent in 1978 . . . The Paris meeting also reaffirmed that, in respect of the interest rate on the rescheduled payments, interest during the respective grace periods (moratorium interest) should not exceed 4 percent per annum; that

115

IndonesidChapter 7

this interest should not be payable during the grace periods, and when paid, should not be compounded. (World Bank 1968, 54-55) Given the desperate situation of Indonesia, the Western countries established the IGGI to draw up a long-term plan of official assistance and to coordinate the form of aid to maximize its effectiveness. To ensure maximum institutional flexibility, IGGI was not given formal status and, operationally, The IGGI is a series of meetings between Indonesia and its donor countries and organizations; it is not based on an international agreement, nor do its conclusions and recommendations represent such agreements. (Posthumus 1971, 7) The terms of IGGI lending were kept as soft as possible: repayment period of twenty-five years, including seven years of grace, and an interest rate of 3 percent. IGGI was also generous in the amount of official assistance it granted (in millions of dollars): 1967, 167.3; 1968, 361.2; 1969/70, 507.7; 1970/71, 609.7; 1971/72, 633.7; and 1972/73, 670.0 requested (Posthumus 1972). The mix of generous external assistance and corrective economic policies implemented by the Soeharto government imparted a new dynamism to the Indonesian economy and boosted it to a higher growth path. The annual average growth rate from 1968 to 1972 was 8.2 percent compared to the average rate of 1.2 percent in the preceding five years. 7.3 Debt Management in the 1970s The inflow of foreign resources to finance government spending was substantial in the early days of the Soeharto regime. They financed 20 to 28 percent of total government expenditure from 1967 to 1972/73. Since 1968, all of foreign borrowing has been officially designated in the budget under the development expenditure category. Whether Indonesia’s dependence on external funds has decreased or increased in the 1970s depends on the measure of dependence. If the focus is strictly on the actual financing of government expenditure, then the role of foreign borrowing has diminished. The ratio of foreign resources to expenditure declined secularly from 27 percent in 1969/70 to 17 percent in 1979/80 (see table 7.1). The growth of the petroleum and LNG sector since 1971 brought in tax revenues at a rate faster than the rise in government expenditure, hence reducing the role of foreign financing. On the other hand, if the issue of foreign borrowing rests on fears of an excessive debt-service burden in the future, then an appropriate measure of external financial dependence is the amount of annual government borrowing in the international markets normalized by the GDP. The movement of this

Table 7.1

Total expenditure Total revenue Foreign borrowing As percentage of expenditure As percentage of GDP Memo item GDP deflator (1980= 100)

Total expenditure Total revenue Foreign borrowing As percentage of expenditure As percentage of GDP Memo item GDP deflator (1980= loo)

Trends in Aggregate Expenditure and Revenue, FY1969 to FY1979 (in billions ofrupiahs) 69/70

70171

71172

72173

73/74

74175

75176

76/77

77/78

78179

342.7 251.6 91.1

467.8 354.7 113.1

557.0 440.0 117.0

736.3 590.6 145.7

1,164.2 967.7 196.5

1,977.9 1,753.7 224.2

2,730.3 2,241.9 488.4

3,684.3 2,906.0 778.3

4,305.7 3.534.4 771.3

5,299.3 4,266.1 1,033.2

26.6

24.2

21.0

19.8

16.9

11.3

17.9

21.1

17.9

19.5

3.4

3.5

3.2

3.2

2.9

2. I

3.9

5.0

4.1

4.3

13.9

15.4

16.3

18.5

24.6

36.2

40.7

46.6

52.7

58.4

79/80

80181

81182

82/83

83/84

84/85

85/86

86/87

87/88

8,076.0 6,696.8 1.379.2

11,716.1 10,227.0 1,489.1

13,917.6 12,212.6 1,705.0

14,355.9 12,418.3 1,937.6

18,311.0 14,432.7 3,878.3

19,380.8 15.905.5 3,475.3

22,824.6 19,252.8 3,571.8

21,421.6 17,832.5 3,589.1

23,583.2 17,236.1 6,347.1

17.1

12.7

12.3

13.5

21.2

17.9

15.6

16.8

26.9

4.0

3.0

2.9

3. I

5.3

4.0

3.7

3.5

5 .9

77.4

100.0

111.2

119.6

136.3

152.6

165.8

170.9

176.0

117

IndonesidChapter 7

measure in the face of the rapid economic growth which Indonesia experienced indicates the extent to which the government had mobilized internal resources to replace foreign credit in development financing. The ratio of foreign borrowing to GDP showed a slight increase, from 2.9 percent in 1970-74 to 4.5 percent in 1975-79. This resource inflow measured in real terms-foreign borrowing normalized on the GDP deflator indexed on 1 9 8 C w a s Rp 657 billion in 1969-70 and Rp 1,468 billion in 1977178, a doubling of the annual inflow in eight years. The continued heavy borrowing was partly the result of the government’s unwillingness to increase taxes by broadening its tax base and partly the result of the greater availability of foreign credit at favorable interest rates. By 1974 international credit markets had rescinded whatever credit restrictions they had imposed on borrowing by the Indonesian government in the aftermath of Soekamo’s economic Armageddon. Three reasons were responsible for this change: one, the avalanche of oil revenue increased the creditworthiness of the Indonesian government; two, there was a boom in commodity prices in the early 1970s; and, three, lending opportunities decreased in OECD countries whose medium-term economic prospects were rather bleak after the 1973 OPEC price increase. The eagerness of the international financial community to lend to Indonesia is best captured by the casualness with which it committed $10.5 billion in loans (as of February 1975) to the state oil company, Pertamina, most of it in the two years before 1975. These loans were extended to Pertamina without access to any detailed official financial statements of the oil company! The improvement in the creditworthiness of Indonesia can be clearly seen by comparing the years 1970 and 1980 in table 7.2. Of the $12.5 billion increase in publicly-guaranteed debt in this period, 41 percent of it came from private creditors. Concessionary lending from official creditors was no longer the only important source of external funds, and it was only natural after 1973 that official loans declined. The IGGI loans were no longer as eagerly sought after the OPEC-I oil price increase because the advice (“policy recommendations”) which was dispensed with the loans was resented in many circles as a foreign intrusion into the domestic policymaking process. Furthermore, the newfound petroleum wealth of the Indonesian economy reduced the intensity of the charitable feelings which made concessionary loans possible in the first place. The ending of private credit rationing against Indonesia was amply justified by political and economic developments. Since 1970 the Soeharto regime had been firmly in power with neither strong domestic opposition nor outstanding regional tension being a credible threat. On the economic front, the export sector showed an amazing ability to earn foreign exchange and the economy was increasingly showing signs of growing beyond 7 percent annually. Despite the sixfold increase in debt between 1970 and 1980, the

Table 7.2

External Public and Publicly-Guaranteed Debt" of Indonesia, 1970-80

Debt outstanding and undisbursed Debt outstanding and disbursed (DOD) Private creditors Total debt service (TDS) Private creditors Principal ratios ( 8 ) DODIXGSb DODIGNP TDSIXGS TDSIGNP Proportion of DOD which: Is concessionary Bears variable i-rates Is from private creditors Memo items Proportion of debt service paid to private creditors Central Bank net foreign assets

1970

1973

1974

1975

1976

2,947.2 2,443.2 282.6 82.2 42.4

6,693.6 5,248.8 1,218.9 207.5 114.8

9.060.9 6,358.2 1,739.4 291.6 168.4

1 1.741.2

7,994.0 2,990.1 523.5 388.9

14,572.5 10,001.6 4,089. I 760.6 573.6

205.5 27.1 6.9 .9

158.8 33.5 6.3 1.3

85.2 25.8 3.9 1.2

113.8 27.4 7.5 I .8

78.0 11.6

75.3 4.5 23.2

71.3 6.8 27.4

51.6 55.6

55.3 800.0

57.8 1,494.0

.o

1977

1978

1979

1980

16,134.6 4,583.1 1,261.7 985.3

19,037.3 13.149.7 4,761.1 2.062. I 1.632.9

21,199.8 13,277.8 4,767.8 2,099.6 1,535.7

24,451.9 14,971.3 5,464.9 1,758.5 1,127.6

114.0 27.7 8.7 2. I

106.7 26.4 11.5 2.9

116.3 26.6 18.2 4.2

85.5 27. I 13.5 4.3

67.4 21.6 7.9 2.5

60.8 19.4 37.4

53.4 20.7 40.9

52.2 18.7 39.3

53.2 15.0 36.2

51.5 14.5 35.9

50.2 16.8 36.5

74.3 -691.6

75.4 - 159.0

78. I 1,108.4

79.2 1,312.0

73.1 3,282.3

64.I 5,868.4

1 1,658.3

Note: Central Bank assets position calculated from Infernational Financial Srutisrics data prior to February 1987, using series I 1 and 16c. Bank Indonesia changed its valuation procedures in February 1987. The new procedures caused net asset position to improve quite substantially.

"Debt reported is end of period, in millions of U.S. dollars. bXGS

=

export of goods and services.

119

IndonesidChapter 7

debvexport ratio fell from 206 percent to 67 percent, and the debt/GNP ratio fell from 27 percent to 22 percent. The eagerness of the banks to extend credit appears to have been matched by the eagerness of the Indonesian public sector to borrow. It was quite normal for Indonesia to contract additional loan commitments even when it was unable to spend these funds quickly enough. The ratio of funds drawn to the total committed was 61 percent in 1980 compared to 83 percent in 1970. It must be mentioned at this point that there is no paradox in why in 1970-80 the debt-service/export and debt-service/GNP ratios went up as the debvexport and debVGNP ratios moved down. The rise in the former two ratios was a good sign in this case because it meant that Indonesia had regained access to the private capital market. The increase in the DSRs was inevitable for two reasons. First, Indonesia was now borrowing an increasing proportion of its loans at market interest rates rather than at concessionary IGGI rates. Only 50 percent of the outstanding long-term Indonesian debt in 1980 was concessionary compared with 78 percent in 1970. Second, private credit generally has a shorter maturity than official credit. In 1970 only 11 percent of total debt was owed to private creditors compared with 37 percent in 1980. So it is not that the Indonesian DSRs in 1980 were high, but that these ratios were unusually low in 1970. The low 1970 DSRs were the result of Indonesia being shut off from the private credit market and of IGGI having to forward emergency loans at 3 percent to help the Soeharto government weather its inherited financial crisis. The point is that the higher DSRs in 1980 were the result of an economic situation superior to that in 1970. On consumption-smoothing grounds, readmission into the external credit market resulted in a net gain to Indonesian national welfare. This welfare gain was not without its price, however, as Indonesia was now exposed to two new risks. The first risk is systemic in nature and threatens every country with external debts. An example of such a systemic risk is the simultaneous collapse of a debtor’s export earnings and a large increase in the real rate of interest. The second new risk is the possibility of imprudent borrowing by Indonesia. This danger was realized in February 1975 when Pertamina could not roll over a $400 million short-term loan and defaulted. The government bailed Pertamina out by guaranteeing repayment of all of its debts. The enormity of the Pertamina rescue operation is clearly indicated by the jump in the principal ratios in table 7.2. The debvexport ratio jumped from 85 percent in 1974 to 114 percent in 1975, and it returned to the 85 percent level only in 1979. By taking over Pertamina’s debts, the proportion of private credit to total credit jumped by 10 percentage points-a magnitude unprecedented and not repeated. It is clear that most of Pertamina’s debt bore variable interest rates because the percentage of variable interest rate loans almost trebled from 1974 to 1975 (see table 7.1).

120

Wing Thye Woo and Anwar Nasution

7.4 The Pertamina Debt Crisis Pertamina, on the eve of the crisis, was more than an oil company.* It was an extremely diversified conglomerate with the distinction of being the largest corporation in Asia outside of Japan. Table 7.3 gives a partial listing of the range of businesses engaged in by Pertamina. As discussed earlier, one of Soeharto’s key operating styles is to appoint a “dynamiser” to solve problems of industrial development and to grant him broad discretion in achieving the goals. General Ibnu Sutowo, who became chairman of the National Oil Committee in 1966 and president-director of Pertamina in 1968, lived up to the “dynamiser” image. Oil production rose from less than half a million barrels a day in 1966 to 1.4 million barrels a day in 1973. Furthermore, it was widely accepted that General Sutowo had driven a hard bargain with the oil companies, receiving terms which were more favorable than those received by the Saudis (Hunter 1967). His innovative “production-sharing’’ scheme avoided many of the monitoring difficulties of the “posted price” approach adopted by the Gulf states.

Table 7.3

F’ertamina Subsidiaries and Joint Ventures

Company and Status Whollv owned subsidiaries in Indonesia PT Electronika Nusantara (Elnusa)

Functions

Services for marine, land, and offshore operations

PT Palembang Rice Estate

Large-scale rice project in South Sumatra

PT Patra Jasa

Providing facilities to oil and service contractors (offices, housing, and land transport)

PT Pelita Air Service Pertamina Gulf Industrial Processing

Air services Packaging of fertilizer and other chemical products Operating non-vessel tankers

PT Pertamina Tongkang Wholly owned subsidiaries outside lndonesia Ocean Petrol Ltd. (Hong Kong)

Joinr ventures in lndonesia PT Arun Natural Gas Liquefaction Co. (Pertamina 55%; Mobil 30%; Jilco 15%) PT Badak Natural Gas Liquefaction Co. (Pertamina 5 5 6 ; Huffco 30%; Jilco 15%) PT Brown and Root Indonesia (Pertamina 20%; Brown & Root USA 80%)

PT Chicago Bridge and Iron lndonesia (Pertamina 51%; Chicago Bridge and Iron Co.. USA 49%)

Operating and managing ocean-going tankers Processing and sale of LNG produced in Aceh Processing and sale of LNG produced in East Kalimantan Manufacture of components and appurtenances for offshore constructions; concrete coating of steel pipes: design and construction of processing plants and engineering works for oil and gas; procurement and storage of materials To furnish metal plate structure, process facilities equipment, and construction services throughout Indonesia for government agencies of Indonesia and companies operating in Indonesia

121

IndonesidChapter 7

Table 7.3

(continued)

Company and Status

PT Dresser Magcobar (Pertamina 10%; Dresser Magcobar. USA 90%) FT Indonesia Chemical Co. (Pertamina 60%; PT Sempurna 10%; Teijin Ltd and Toyo Menka 30%)

PT Krakatau Steel (infrastructure; on behalf of the GOI, $6 million)

FT Kuda Laut Batam Island (Pertamina 50%; Intera-

Functions Mud for drilling To produce annually 100,000 tons of peravele and 120,000 tons of dimethyl telethalate (DMT) in South Sumatra To rehabilitate and operate the abandoned Soviet steel mill project at Cilegon Supply frozen and dry foodstuffs

gencies Hong Kong 50%)

PT Nippon Steel Construction (Nisconi) (Pertamina 10%;Nippon Steel Japan 90%)

PT Patra Vickers Batam (Pertamina 50%; Vickers Ruwolt Australia 50%) PT Permiko Engineering and Construction (Pertamina 10%;Nippon Kokan KK and Mitsubishi 90%)

FT F'ertafenikki (Pertamina 30%; Japan Gasoline 60%; Far East Trading Co. 10%) Pexa Oil Co. (Pertamina 25%; Pexa Oil Co. 75%)

I T Burna Bina Indonesia (Pertamina 51%; Bechtel

To provide support for oil and gas industry including: fabrication, assembling, and construction of steel structures; coating of gas and oil pipes; supply storage and servicing Heavy engineering facilities to service the oil, mineral processing extraction, and other manufacturing industries. Fabrication, coating, assembly, installation. and construction of pipelines and steel structures for oil and gas exploration drilling; production and distribution; supply of services including design, inspection, testing, repairing, and surveying for gas and oil; storage and lease of goods and equipment related to these Consulting engineering Oil exploration onshore South and East Kalimantan Engineering consulting

Inc. 49%)

PT Sankyu International (Pertamina 10%;Sankyu Tokyo 90%) PT Toyo Kanetsu (Pertamina 51%; Toyo Kanetsu 35%: Nissho lwai 14%) Joint ventures outside Indonesia Far East Oil Trading Co. Ltd. (Japan) (Pertamina 50%; various Japanese companies 50%) Indonesian Enterprises Ltd., USA (Pertamina 50%; vanous companies 50%) Japan-Indonesia LNG Import Co. (Jilco) (Pertamina 15% through Far East Oil Trading Co.; five Japanese companies 5 I %; Tokyo Electric and Tokyo Gas 4%; Industrial Bank of Japan 6%; Nisano Iwai 15%; other trading companies 9%) Japan Indonesian Oil Kabushiki (Pertamina 50%; Toyota Motor Sales Co., The Tokyo Electric Power Co., The Kansai Electric Power Co., The Chuba Electric Power Co.. Maruzen Oil Co.. Daikyo Oil Co., ldemitsu Kosan Co., total 50%) Perta Oil Co.. USA (Pertamina 50%; United States International Investment Corp. 50%) Tugu Insurance Co. Ltd., Hong Kong (Pertamina 40%; private investors 64%)

Fabrication, assembling, installation, and construction of pipelines and steel structures Engineering consulting

Marketing of crude oil in Japan Promotion of tourism in the USA

To supply the low sulphur crude oil produced in Indonesia to Japan, and other associated matters

Transport and marketing of Indonesian crude oil Insurance

Source: Bulletin of Indonesian Economic Studiey (July 1974)

122

Wing Thye Woo and Anwar Nasution

Sutowo’s ability to get things done, albeit many times by paying a high p r e m i ~ m resulted ,~ in him being asked by the president to take over lagging projects. The biggest of such projects was the completion of the Krakatau Steel Mill for which Pertamina had assumed responsibility in 1970 and which had been abandoned uncompleted by the Soviet Union in 1966. Partly by allowing itself to be volunteered to take over the management of more and more state projects, and partly because of Sutowo’s desire to play a pioneering role in Indonesian economic development, Pertamina developed into an autonomous development agency independent of the control of the technocrats at BAPPENAS. Pertamina improved harbors, developed residential and commercial estates, and built roads and hospitals. Although these expanded activities of Pertamina intruded upon the economic policymaking turf of the technocrats, there was not much domestic pressure which they could bring to reverse this state of affairs. There were good reasons why the president chose to maintain the status quo with Pertamina. An obvious one was that criticisms of a success story were hard to make stick, especially when Pertamina’s largess had won it many supporters in the president’s circle. An equally important reason could be that General Ibnu Sutowo had come to epitomize the kind of economic nationalism favored by a large segment of the military and the intellectuals. As explained in chapter 3, one thorny problem of Indonesian economic development has been the disproportionate economic power of the Indonesian Chinese community. General Sutowo’s much heralded success as a big-time businessman was a source of considerable ethnic pride. It was widely held in some intellectual circles that the only effective containment of Chinese economic strength was to allow the small number of capable priburni entrepreneurs to each head a gigantic state enterprise modelled after the Japanese zaibatsu. It was believed that only such gigantic enterprises could reap the economies of scale and sustain the short-term losses of an infant industry. What was really alarming to the technocrats who had gone through the trials of rescheduling Soekamo’s debts was that this new independent development agency was now borrowing heavily in the international credit market to finance its nonoil activities. In March 1972 the minister of finance chose to enter into another standby agreement with the IMF, even after the IMF had concluded that a balance-of-payments crisis was unlikely. The standby agreement set a ceiling of $14 million on medium-term external borrowing for 1972/73. Given this ceiling, a decree was issued requiring all state bodies to seek approval from the Ministry of Finance before contracting foreign borrowing. Pertamina ignored this decree, borrowing $350 million in short-and medium-term debt in 1972 without informing the Ministry of Finance. When this transgression came to the attention of the United States, the biggest aid donor, American economic aid was suspended. And in February 1973 U.S.

123

IndonesidChapter 7

Vice President Spiro Agnew raised the issue of Pertamina’s borrowing with President Soeharto. The president’s reaction was very much in line with his belief that a talented problem-solver like Sutowo ought to be given free rein. Soeharto “delivered a vigorous defense of Ibnu Sutowo, saying he had personally charged him with important national projects and trusted the oil chief to find his own finance” (McDonald 1980, 155). Nevertheless, when U.S. aid resumed, Pertamina stopped borrowing in the medium-term market and started financing its long-term projects with short-term loans. It was this borrowing in the short-term market which precipitated the Pertamina crisis. Short rates rose dramatically after the OPEC-1 price increase because central banks in the industrialized countries began tightening their monetary policies to dampen aggregate demand to offset the price pressures from the supply side. At the end of 1974, the discount rate was 7.8 percent in the United States and 9 percent in Japan, compared with their respective 1972 discount rates of 4.5 percent and 4.3 percent. It was also at this time that the international banking community was shocked into greater cautiousness by the failures of the Franklin National Bank in the United States and the Herstatt Bank in Germany due to foreign exchange speculation. The result was that the banks took a harder look at Pertamina’s borrowing. They were alarmed that they had extended so much credit (it turned out to be $10.5 billion) to Pertamina without any one of them having seen a full statement of Pertamina’s finances. While it was true that the future of the oil industry looked very promising in 1975, bankers were troubled because significant proportion of the borrowing was for nonoil projects, the debt was not guaranteed by the Indonesian government, and Pertamina was showing increasing signs of mismanagement. In the face of these reservations, the banks refused to automatically roll over existing debts unless they were given more information about Pertamina’s financial position. The upshot was that Pertamina defaulted on 20 February 1975 when it could not meet a $400 million payment to the Republic National Bank of Dallas. The Indonesian government announced that it would assume responsibility for Pertamina’s debt, of which $1.5 billion was in short-term loans. The rescue operation obligated the Indonesian government to undertake its first major borrowing in the external credit market since it had been readmitted into the private credit market. The Pertamina crisis was solved by a combination of repayment, rolling over part of the existing debt into longer term instruments, and cancellation of contracts. Even with access to the international financial market, Indonesian resources were still stretched thin. The net foreign reserve position of Bank Indonesia fell from $1.5 billion at the end of 1974 to a deficit of $0.7 billion at the end of 1975. An indication of the magnitude of money involved in the rescue operation can be obtained from table 7.4 in which the long-term publicly-guaranteed debt is divided between the government sector and the public enterprise

Table 7.4

Distribution of External Sovereign Debt’ by Sectors, FYI973 to FYI986 73/74

Government sector Net drawings Adjustment Outstanding debt

75/76

76/77

77/78

571.0

1,918.0

1,657.0

1,345.0

3,979.0 4,550.0

6,468.0

8,125.0

9,470.0

562.0

74/75

Public enterprises Net drawings Adjustments Outstanding debt

1,665.5 2,220.7

Unattributable drawings or adjustments during 1973-77

-118.3

Total public sector Net drawings Adjustment plus unattributed sums Outstanding debt Memo item Governmentborrowingforbudget Government borrowing from balance of payments Ratio of public enterprise to total public sectordebt (%)

352.6

555.2

-3.6

914.6

1,126.2 -3.6 5,526.2 6,767.1 - 118.3

-109.3

78/79

1,425.0 1.263.0 571.0 -975.0 11,466.0 11,754.0

-327.1

-10.0

2,162.1

1,835.0

.O 1,825.0

128.0

726.1

81/82

1,381.0 -62.0 13,073.0

2,271.0 -787.0 14,557.0

49.0 18.0 1,802.0

683.0 -5.0 2,480.0

82/83

83/84

3,226.0 3,883.0 5.0 -471.0 17,788.0 21,103.0 1,017.0 -122.0 3,375.0

84/85

85/86

86/87

4,710.0 1,678.0 3,134.0 -2,038.0 4,487.0 3,603.0 21,179.0 27,344.0 34,081.0 -408.0 77.0 3,175.0

1,430.0 2,954.0 4,243.0 3,960.0 1,017.9 1,415.0 1,190.0 1,808.7 1,707.7 -117.0 -417.0 -83.5 128.0 726.1 571.0 -992.0 -44.0 -792.0 8,495.9 10,415.1 12,031.1 13,291.0 13,489.0 14,875.0 17,037.0 21,163.0 24,609.0

4,302.0 -1,961.0 24,354.0

-83.5

-73.0 - 17.0 1,735.0

SOB1

77.0 54.0 3,506.0

2,111.4

50.7

79/80

-325.0 41.0 2,891.0

-383.0 18.0 2,526.0

1,353.0 2,751.0 4,528.0 3,621.0 30,235.0 36,607.0

491.3

559.0

1,184.6

1,888.7

1,863.6

2,105.1

2,204.1

2,382.5

2,685.0

2,877.9

3,947.9

3,314.0

3,189.3

2,544.4

643.0

660.0

1,995.0

1,823.0

2,106.0

2,208.0

2,690.0

2,684.0

3,521.0

5,011.0

5,793.0

3,519.0

3,432.0

5,296.0

30.1

32.8

24.9

20.8

15.3

13.7

12.9

12.1

14.6

15.9

14.2

13.0

9.6

6.9

Note: Figures for 1978/79 onward are from an IMF document. For the earlier years, net drawing of public sector is from the World Bank’s World Debt Tables and net drawing of government sector is from the balance of payments. The difference between the two numbers is attributed to net drawing by public enterprises. The difference between the cumulated Rows and the stocks in the World Debt Tables is reported in the “Unattributable drawings” item. ”Debt reported is end of period, in U.S. dollars.

125

IndonesidChapter 7

sector. The debt flow of 1975/76 was extraordinary compared with the preceding two years-the $1.9 billion increase in the long-term debt of the government sector was more than three times that in 1973/74 and 1974/75. Four reasons have been advanced for this large increase. The first two reasons are the conversion of part of the $1.5 billion short-term obligations into long-term debt and the amortization of Pertamina’s long-term debt. The third reason is that by 1975 the government had been convinced by the 1973 oil price increase that the future income stream of the economy had been markedly increased. It decided, therefore, to embark on an expanded program of development spending to be financed by external borrowing. The fourth reason is that development expenditure in 1975 overshot its targeted level by $313 million, requiring the government to undertake additional borrowing. Since detailed data on the Pertamina rescue operation have not been released, any estimates of how much of the $2.0 billion worth of public borrowing, as shown in the balance-of-payments account, is Pertaminarelated is necessarily speculative. It could be argued that the Pertamina affair accounted for only 40 percent of the borrowing because $1.2 billion was required to finance the budget. On the other hand, one has to take into account that Pertamina had collected $8 19 million in oil taxes on behalf of the Ministry of Finance, which it had kept for its own use. If this oil revenue had been forwarded to the government, only $365.6 million would have been needed for budgetary reasons. This meant that more than 80 percent of the large external borrowing in 1975/76 was caused by the mismanagement of ~ e r t a m i n a . ~ For the next several years, Pertamina continued to be a drain on the budget and obliged the government to undertake additional external borrowing. In 1976/77 the government budget showed a debt-service transfer of $75 million to Pertamina; in 1977/78 the debt-service transfer was $208 million; and in 1979/80 the government extended a subsidy of $77 million to Pertamina.

7.5

Economic and Political Effects of the Pertamina Crisis

Since the 4.9 percent real GDP growth rate in 1975 was the lowest in the thirteen-year period from 1968 to 1981, one may be tempted to attribute the drop in aggregate demand to the across-the-board cancellation of Pertamina’s numerous investments. We do not think so, however. Instead, we believe that the large plunge of the 1975 real GDP growth rate from the rates of 9 percent in 1972, 11 percent in 1973, and 8 percent in 1974, was largely the result of the 1975 global recession induced by high oil prices. Decomposition of the sources of growth in 1975 revealed that it was external, not internal, factors which were responsible for the low growth rate: real domestic absorption increased by 17 percent, while real exports fell by 18 percent and real imports rose 8 percent.

126

Wing Thye Woo and Anwar Nasution

The level of real government expenditure did grow more slowly in 1976/77 and 1977/78 because the technocrats deemed it prudent to lower the growth of external debt and, hence, diverted domestic revenue to keep Pertamina afloat. If one is looking for a silver lining in the Pertamina crisis, it could be argued that the fiscal conservatism of this period was a desirable outcome because inflation has been above 18 percent since 1973 and economic growth has picked up again.5 The Pertamina debacle was a major setback to the military advisors who favored the nationalist zaibatsu approach to economic management. The technocrats at the BAPPENAS and Ministry of Finance were once again unchallenged in the sphere of economic policymaking. The technocrats were asked to oversee the reorganization of the Pertamina empire and were granted real control over all external borrowing (short- and long-term) by state enterprises. There is no doubt that the technocrats immediately used this new authority not only to sharply curtail foreign borrowing by the state enterprises but also to reduce their outstanding debt. At the time of the Pertamina crisis, not counting short-term debt, the long-term publicly-guaranteed external debt of state enterprises stood at $2.2 billion, which was 33 percent of the total outstanding public debt. By the end of March 1979, the debt of the state enterprises was reduced to $1.7 billion, which was only 13 percent of the total. The days of Pertamina-style borrowing by public enterprises were clearly over. It could be cogently argued that the Pertamina crisis was a blessing in disguise. By reminding the Indonesian government of the importance of being prudent in external borrowing, it could be the reason why Indonesia did not experience a debt crisis in 1982. The timing of the Pertamina crisis could not have been better. The real price of oil was at an all-time high, so the international credit markets were willing to roll over the now-guaranteed debts; and the real interest rate was low, so the debt burden was not increased disproportionately. If Pertamina had not been prevented from further borrowing in the medium-term credit market and had, therefore, over-reached itself in the short-term credit market, it could have accumulated by 1982 a foreign debt at least as big as the $20 billion debt of PEMEX, the Mexican state oil company. After all, in February 1975 Pertamina already had $10.5 billion in loans (including undisbursed), and the 1979 OPEC-2 price increase would have further expanded its ability to borrow. The failure of an unchecked Pertamina in 1982 would have had cataclysmic effects on the economy. Not only would the record high real interest rates of 1982 have made the debt service painful, but the lower oil prices would have forced the implementation of more draconian austerity policies. By denying all state-owned enterprises direct access to the external credit market after the Pertamina embarrassment, Indonesia did not have as large a debt as it otherwise would

127

IndonesidChapter 7

have had when it entered 1982. With the benefit of hindsight, the Pertamina debacle could certainly be viewed as a vaccination against excessive borrowing. It is important to note that the Pertamina crisis did not reduce the role of the army in the management of state and private enterprises. The doctrine of dwijiingsi remained intact. For example, even though the technocrats were asked to reorganize Pertamina, it was an army man, General Piet Haryono, not a civilian administrator, who was appointed to replace General Ibnu Sutowo. And it was another army man who succeeded General Haryono in 1981. The point is that it is vital for the army to control key sectors of the economy, partly because of the political need to channel resources to retain the support of the military and partly because of the president’s belief that a ‘‘dynamiser” should have discretionary funds available for off-budget development projects. The legacy of the Pertamina crisis is that these extrabudgetary allocations are now unlikely to lead to an external debt crisis.

7.6 External Debt in the 1980s The Pertamina crisis also resulted in major changes in the way in which the external debt is managed. As a commitment to prudent debt management, the government now eschews short-term loans in its borrowing. Furthermore, all external borrowing by the government, government agencies, and state-owned enterprises must be approved, negotiated, and administered by the Ministry of Finance and Bank Indonesia. By law, private Indonesian enterprises are required to report all their external borrowing and to hand over the administration of these debts to Bank Indonesia. In practice, the government has control only over official borrowing. This state of affairs is due in small part to underreporting by the private sector and in large part to the Indonesian government’s pledge of maintaining an open capital account. There is close collaboration between the Ministry of Finance and Bank Indonesia in managing the external debt.6 In addition to monthly and detailed quarterly reports on changes in the external debt, Bank Indonesia also submits a weekly report of its external-debt-related activities to the Ministry of Finance. Before any payment of principal and interest can be made, Bank Indonesia must request approval from the Ministry of Finance which would then recalculate the debt service using its own records. The slow computerization of external debt information unduly increased the difficulties of debt management in the early 1980s when Latin American countries were slipping into debt crises. Manual handling and careful recalculations meant that as late as 1984, “late payments [were] frequent, resulting in late payment penalties and even defaults” (Haryono 1985, 229). The manual handling of the debt records meant that it was almost impossible to generate debt-service scenarios under different refinancing schemes to

128

Wing Thye Woo and Anwar Nasution

change the maturity structure and under different assumptions about foreign interest rate and exchange rate developments. While the Pertamina crisis was a deep one, it was still manageable. Real interest rates were low and Indonesian exports were in high demand. The collapse of oil prices in early 1982 and the subsequent collapse of commodity prices have produced a situation which is more ominous. In 1985 public long-term debt-service ratio went above 20 percent for the first time, rising from a ‘‘comfortable’’ 15 percent in 1984 (see table 7.5). The significant rise in the DSR occurred despite a 38 percent devaluation of the rupiah in March 1983, and the large-scale postponement of public investment projects in fiscal 1983 which saved $10 billion in foreign exchange (World Bank 1987b, 24-25). External debt management in the 1980s has also been made more difficult by capital flight. The (net) errors and omissions item (which contains all private portfolio flows) in the balance-of-payments accounts shows a cumulative deficit of $6.3 billion for 1980-85 compared to one of $1.4 billion for 1969-79. Without the loss of reserves from the 1980-85 capital flight, the DSR in 1985 may have been 15 percent instead of 20 percent. Two events occurred in 1986 which worsened the debt situation dramatically. The price of oil dropped precipitously from $28 per barrel in January to $10 in August (World Bank 1987b, 24-25). The yen, in which more than a third of Indonesian public external debt is denominated, appreciated 21 percent against the dollar. The result was that the public debt-service ratio shot up to 29 percent at the end of 1986. (If external credits for the expansion of LNG production are included, the ratio stood at 32 percent.) The Indonesian government has shown itself to be prepared to make significant policy changes to ward off an external debt crisis. The rupiah was devalued by 31 percent in September 1986, and, in October 1986 and January 1987, the input costs to the export sector were lowered with the abolition of a substantial number of import restrictions. Nominal state expenditure for fiscal 1986 was cut 6 percent from the previous year (see table 4.3). In fact, real government expenditure has been steadily reduced since fiscal 1983. Real expenditure has declined more than the drop in real revenue, reducing the amount of real borrowing (in 1980 prices) from Rp 2.8 billion in 1983 to Rp 2.1 billion in 1986. Mobilization of domestic resources was also undertaken to slash foreign borrowing-a value-added tax was introduced in April 1985, followed by a more comprehensive land tax in June 1986. These tax increases, however, have not been able to make up for the fall in oil royalties. Nominal revenue in 1986 was Rp 1.4 billion lower than in 1985.

Table 7.5

External Debt' of Indonesia, 1980-86

(a) External public and publicly-guaranteed long-term debt Debt outstanding and undisbursed Debt outstanding and disbursed (DOD) Private creditors Total debt service (TDS) Private creditors Principal ratios(%) DOD/XGSb WDIGNP TDSi XGS TDSIGNP Proportion of DOD which: Is concessionary Bears variable i-rates Is from private creditors Proportion of public debt service paid to private creditors (b) External private nonguaranteed long-term debt Debt outstanding and disbursed Total debt service

1980

1981

1982

1983

1984

1985

1986

24,451.9 14,911.3 5,464.9 1.758.5 1,127.6

27.21 1 .0 15,870.3 5,812.5 2.047.2 1,336.4

32,216.0 18,515.0 7,403.4 2,246.6 1,374.3

35.492.0 21.689.5 9.649.6 2.550.8 1,526.8

36.922.8 22.861.9 10,048.6 3,251.0 2.069.9

41.872.6 26,624.6 11.667.9 4,015.1 2,631.7

48,712.0 32,119.0 14.556.0 4.401.0 2,549.0

67.4 21.6 7.9 2.5

63.7 19.2 8.2 2.5

87. I 21.2 10.6 2.6

108.8 28.0 12.8 3.4

103.2 26.5 14.7 3.8

133.0 33.4 20.1 5.0

212.8 48.2 29.3 6.6

50.2 16.8 36.5 64.1

48.5 17.8 36.6 65.3

42.7 20.0 40.0 61.2

37.3 22.7 44.5 59.9

34.9 23.1 43.9 63.6

35.1 21.7 43.6 65.5

NA NA 45.3 57.9

3,142.0

3,579.0 1.173.0

3,200.0 1,260.0

3.400.0 1 , I 16.0

3,800.0 960.0

3,810.0 1,036.0

3,800.0 800.0

1,051.0

(c) External short-term debt

2,775.0

3,274.0

4.787.0

4,639.0

5,384.0

5.280.0

5.000.0

Total external debt

20,888.3

22,123.3

26,502.0

29,128.5

32,045.9

35.714.6

40.919.0

22,208.0

24,926.3

21,262.1

19,932.6

22,163.8

20,014.5

15,094.0

Memo XGS

Source: The World Bank's Wor/d Debt Tables and Country Reports. Figures do not include LNG expansion credits.

"Debt reported is end of period, in millions of US. dollars. bXGS = export of goods and services.

130

8

Wing Thye Woo and Anwar Nasution

Conclusions and Prospects

8.1 Introduction In this chapter, we attempt to explain why Indonesia did not experience a debt crisis in 1982-84. We do so by comparing the Indonesian debt situation with that of Mexico and Brazil, two countries which have debt-servicing difficulties. In section 8.2 we examine the most common explanation for external crises which puts the blame on excessive budget deficits that force the government to borrow from abroad. Section 8.3 identifies the factors which have prevented an Indonesian debt crisis in 1982-84, and section 8.4 estimates the relative importance of each factor by using Mexico as the reference. In section 8.5 we discuss the prospects of Indonesia avoiding a future debt crisis and the role for policy in ensuring such an outcome.

8.2 A Comparative View of Fiscal Imbalances When a debt crisis occurs because the government is unable to service the external debts that it has guaranteed, it is a truism to claim that government budget deficits, i.e., fiscal imbalances, are the root of all external sovereign debt crises. By definition, an external public debt can be incurred only when a government borrows from abroad to finance part (or all) of its expenditure. In order for fiscal imbalances to have more than a tautological role in precipitating an external debt crisis, it is necessary to have a criterion which would enable one to assess whether the amount of foreign borrowing being undertaken is excessive in an ex ante sense. The statistic usually cited in support of this fiscal imbalance view is the ratio of official long-term debt to GNP, DGNP.' The official long-term debt is taken to represent the cumulated amount of fiscal deficits financed by external borrowing, and the normalization by GNP is to indicate the extent to which the country has been made to live beyond its income by the budget deficits. The fiscal imbalance explanation of external debt crises points out that DGNP rose very rapidly for those countries which experienced a debt c r i s k 2 From a value of 9.1 percent in 1970, Mexico's DGNP increased 18.7 percent in 1981, the eve of the debt crisis. In the same time period, DGNP went up from 8.7 to 19.5 percent for Argentina, from 8.2 to 17 percent for Brazil, from 8.1 to 20 percent for the Philippines, and from 6.6 to 17.2 percent for Venezuela (see top half of table 8.1). It is true that these governments increased their budget deficits significantly during this period, but it is not true that they have been cumulatively more profligate than the countries in the bottom half of table 8. I , which did not fall into debt crises.

131

IndonesidChapter 8 Public and Publicly-Guaranteed Debt as Fkrcentage of GNP, DGNP, 1970-85

Table 8.1 1970

1975

1978

1980

1981

1982

1983

1984

1985

19.5 17.0 18.7 20.0 17.2

29.1 18.7 33.3* 22.7 18.3

42.6' 30.3' 50.1 31.0' 19.8

36.6 34.7 43.4 36.0 36.7*

58.7 35.5 43.4 41.7 34.3

17.7 28.3 23.1 14.6

20.5 29.7 31.4 17.2

28.0 29.8 37.5 17.9

26.5 30.3 38.6 18.6

33.4 35.0 47.8 26.4

Countries with serious debt problems in 1982-85 Argentina Brazil Mexico Philippines Venezuela

8.7 8.2 9.1 8.1 6.6

8.2 11.6 13.2 8.8 4.6

16.0 14.8 25.3 17.8 17.4

19.2 16.6 18.8 18.7 18.4

Countries with no serious debt problems in 1982-85 Indonesia Korea Malaysia Thailand

25.2 21.2 9.7 5.0

25.5 27.8 14.2 4.2

26.6 22.5 16.9 8.0

20.0 26.9 16.3 12.5

Note; Figures for 1978 are from 1985-86 edition of the World Bank's World Debr Table, and the rest are from the 1986-87 edition. *Year in which debt crisis began.

In fact, Korea and Malaysia had the highest DGNPs in 1981. There are just not enough differences in the 1981 DGNPs of Mexico, Brazil, and Indonesia to explain why Indonesia alone avoided a debt crisis in the following two years. The huge jumps in the DGNPs of Argentina, Brazil, Mexico, the Philippines, and Venezuela occurred only after they were unable to service their debts, forcing them to devalue their currencies. It must be mentioned that DGNP is a flawed indicator of public profligacy. First, the stock of long-term official debt can understate as well as overstate the amount of borrowing for budgetary reasons. This is because the government can borrow short-term to finance budget deficits and long-term to finance foreign market interventions. Second, DGNP is a measure of profligacy only in the sense of living beyond income and not in the sense of being unable to service the acquired external debt. An indicator of the latter would normalize the external debt by the level of exports, the foreign exchange earning capacity of the country. To us, the most interesting fact from table 8.1 is how much the DGNP of each country soared in the year in which its debt crisis happened. Since the big DGNP movement was the result of a devaluation, this suggests that the pre-crisis DGNPs may have been understated and hence provided misleading impressions to policymakers and bankers. This implies that inappropriate exchange rate management may have been a very important factor in precipitating a country's debt-servicing problem. The currency overvaluation not only understated the amount by which the country had lived beyond its means but, perhaps more importantly, debilitated the export sector, the foreign exchange generator for the economy.

132

8.3

Wing Thye Woo and Anwar Nasution

Explaining the Absence of a 1982-84 Debt Crisis

As we pointed out in chapter 1, a debt crisis does not occur only when the government does not have the reserves to service the loans it has guaranteed. A debt crisis also occurs when the government does not have the reserves to enable private domestic residents to convert their service payments on nonguaranteed debts from domestic to foreign currency. When a private borrower cannot come up with the service payments in domestic currency for his private nonguaranteed external debt, we do not consider it a national debt crisis because the government did not cause the default except in the broadest sense of not creating more favorable macroeconomic conditions, if it were able to do so. For the case of a debt crisis caused by a shortage of foreign reserves to allow conversion of private debt-service payments, we define the total external debt service to be the sum of external short-term debt and the debt service on all external long-term debt, publicly-guaranteed and private nonguaranteed. We include short-term debt in our definition because we are interested in the financial resilience of a country to sudden protracted credit squeezes in international credit markets which make short-term borrowing extremely expensive, if not occasionally impossible. After all, the 1973-74 credit crunch did precipitate the 1975 Pertamina debt crisis, and the 1980-8 1 financial squeeze precipitated the PEMEX crisis of Mexico. Since the reserve position of the country is crucial for avoiding debt crises, it is not appropriate to assess the country’s ability to pay by looking at the total external debt service with respect to its income. A more appropriate indicator is the debt-service ratio-the debt service normalized by the level of exports-because the official reserve position is determined primarily by the ability of the export sector to earn foreign exchange. This point is well illustrated by parts (a) and (b) in table 8.2. Even though the 1980-82 debt-serviceiGNP ratios for Brazil and Indonesia are quite close, Brazil had an average total DSR of over 100 percent compared to the Indonesian average of 30 percent. And Brazil experienced a debt crisis after 1982 and Indonesia did not. One reason why the Mexican and Brazilian DSRs are so much larger than that of Indonesia is because Indonesia is a much more export-oriented economy. The average 1980-82 export/GNP ratio was 27 percent for Indonesia, but only 14 percent for Mexico and 9.5 percent for Brazil (see memo item in the table). Even if we ignore short-term debt by assuming (unrealistically) that it could always be rolled over, we see in part (c) of table 8.2 that the long-term DSRs for Mexico and Brazil were still very high compared to Indonesia. The 1980-82 average was 40 percent for Mexico, 62 percent for Brazil, and 14 percent for Indonesia. Items (b) and (c) in table 8.2 together explain why Mexico defaulted before Brazil even though their total DSRs were almost the same in 1981,

133

IndonesidChapter 8

Table 8.2

Debt Characteristics of Mexico, Brazil, and Indonesia, 1978-86 (in percentages) ~~

1978

1980

1981

1982

1983

1984

1985

1986

(a) All short- and long-term debt service as ratio of GNP Mexico 12.0 14.1 15.5 Brazil 7.5 11.0 11.6 Indonesia 9.3 7.5 7.3

24.8 12.8 9.2

17.4 13.0 10.7

13.9 11.0 11.1

11.8 10.2 13.0

NA NA

(b) All short- and long-term debt service as ratio of exports Mexico 105.8 103.6 117.1 Brazil 106.5 114.5 113.6 Indonesia 40.8 25.1 26.1

138.9 146.0 39.0

80.8 104.5 41.7

69.0 72. I 43.3

66.5 72.6 51.6

(c) Public and private long-term debt service as ratio of exports Mexico 62.4 38.0 35.0 Brazil 57.6 56.4 56.8 Indonesia 25.0 12.7 12.9

44.6 71.7 16.5

45.4 46.2 18.4

49.2 34.1 19.0

48.2 34.9 25.2

(d) Proportion of debt which is short term Mexico 14.0 Brazil 13.2 Indonesia 9.9

30.5 19.3 18.1

11.1 14.9 15.6

6.8 11.6 16.8

5.8 10.8 14.8

12.2

(e) Proportion of publicly-guaranteed long-term debt which has variable rate Mexico 59.5 71.5 75.4 76.7 82.7 Brazil 56.8 61.0 67.1 69.3 70.1 Indonesia 15.0 16.2 17.8 20.0 22.8

83.6 73.1 23.7

so. 1 71.5 21.7

NA NA NA NA NA

28.3 19.3 13.3

32.1 19.2 14.4

15.3

NA NA 67.6

NA NA 34.5

NA NA

(f) Effective interest rate for all long-term debt, calculated by (debt service/debt) Mexico Brazil Indonesia

23.4 18.0 17.5

22.8 23.3 15.5

20.1 23.7 16.6

20.8 23.0 16.1

15.9 13.9 14.6

18.0 11.7 15.8

16.1 11.2 16.6

11.3 7.1 22.8

13.7 9.6 29.7

13.2 10.2 27.9

17.9 8.7 23.6

21.5 12.4 25.8

20.1 15.3 25.7

17.8 14.0 25. I

22.6

9.3

13.4

16.1

13.7

10.2

11.8

9. I

7.0

14.5

Memo item Export/GNP ratio Mexico Brazil Indonesia One-year LIBOR for dollar deposits NA

=

NA NA

not available

117 percent versus 114 percent. This is because 70 percent of Mexican total debt service in 1981 consisted of rolling over short-term loans, as against Brazil’s 50 percent, and this was the period when the one-year London interbank offer rate for dollar deposits (LIBOR) reached and remained at historic highs. LIBOR was over 10 percent from 1979 to 1984, with a peak of 16 percent in 1981 (see item d and the memo item in the table). When it was clear in 1980 that short rates would remain high, long rates rose too. Since almost 70 percent of Brazilian publicly-guaranteed debt was on variable rates, the effective interest rate on Brazilian long-term debt rose from 18 percent in 1978 to over 23 percent during the 1980-83 period (see items c and f in table 8.2). The additional interest payments, together with the collapse of its exports due to the deep global recession in 1982, brought about the Brazilian debt crisis in 1983.

Wing Thye Woo and Anwar Nasution

134

Many authors have cited capital flight as a major cause for the external debt crisis in some of the Latin American c ~ u n t r i e s .Part ~ 1 of table 8.3 reports two sets of capital flight estimates. These estimates ought to be treated with caution; different studies have come up with significantly different figure^.^ Sometimes the sign is not even certain; for example, estimates for capital flight from Brazil ranged from -$0.2 billion to $3.9 billion. The point we want to make here is that imprudent maturity structure management may have contributed more to the Mexican debt-servicing difficulties than capital flight per se. To see this, we allow capital flight to have maximum impact on the actual DSR by assuming that Mexico financed the capital flight entirely by short-term debt. Part 2 of table 8.3 shows that Mexico’s DSR in 1982 would have dropped from the actual 138 percent to 64 percent if financing had been done with long-term loans instead. Without capital flight, the DSR would have been 45 percent. In short, the major reason why Mexico’s DSR was so high was because the way in which the government financed the capital flight added 74 percentage points. Capital flight per se added only 19 percentage points. Our conclusion of imprudence in the management of maturity structure can be shown in another way. To see that much of the Mexican short-term debt was from borrowing by the government rather than from commercial credits to finance imports, we make use of the fact that the Indonesian Table 8.3

The Role of Capital Flight in Precipitating Debt Crises

Part I : Cumulated capital flight amount up to 1982 (in billions of dollars) Khan-Haque Estimate

Morgan Guaranty Estimate

29.4 - .2 NA

36.0 3.0 6.0

Mexico Brazil Indonesia

Part 2: Constructing total debt service/export ratios in 1982 Actual Ratio

Mexico Brazil Indonesia

138.9 146.0 39.0

Counterfactual Ratios If capital flight had been financed by longterm instead of shortterm loans 64.2 143.2 20.1

No capital Right 44.6 133.2 16.5

Note: Counterfactual ratios were calculated by assuming that capital flight had maximum impact on the actual debt-service ratios. This means that capital Rights were assumed to have been financed entirely by short-term borrowing. The maximum amount of capital flight was the actual short-term debt of the country. For Mexico and Indonesia, capital flights were assumed to equal actual short-term debts in 1982; and for Brazil the Morgan Guaranty estimate was used. NA

=

data not available

135

IndonesidChapter 8

government, since the Pertamina crisis, has avoided short-term external borrowing as much as possible. Assuming that the Indonesian ratio of short-term debt to imports reflects normal trade financing, we can attribute 77 percent of Mexican short-term debt in 1981 and 1982 to government borrowing. The 1981 and 1982 figures for Brazil are 68 percent and 57 percent, respectively. There is a tradeoff in external debt management between generally lower interest payments and predictability of debt-service payments. Short-term liabilities pay lower interest rates most of the time, but it is risky to rely on a strategy which rolls over large amounts of short-term debt every period. An unforeseen credit crunch would force the country to increase borrowing in order to cover its interest payments. If the credit squeeze persisted for more than three years and was accompanied by a prolonged fall in the country’s exports, this extra borrowing would be difficult to sustain because the situation smacks increasingly of a Ponzi game. Capital flight can be an important mechanism in bringing about a debt crisis, but we cannot view capital flight as an exogenous shock in the Latin American debt crises. Enders and Mattione (1984) and Dornbusch (1987) have emphasized that the large capital flight in Mexico and Argentina was the result of highly overvalued exchange rates. Even with overvalued exchange rates, our discussion based on table 8.3 concludes that capital flight would not have hurt Mexico’s debt-service capacity if it had been financed by long-term, rather than short-term, external borrowing. Pulling all of the observations about tables 8.2 and 8.3 together, we attribute the absence of an Indonesian debt crisis in 1982-84 to three factors: 1. A high proportion of Indonesia’s external debt was borrowed at fixed concessionary rates from IGGI. This ZGGZ efSect explains why the effective interest rate on Indonesian long-run debt averaged 16 percent against the 20 percent paid by Mexico and Brazil (part f of table 8.2). Another result was that only about one-third of Indonesian debt was denominated in dollars compared to 90 percent of Mexican and Brazilian debt. This meant that the large appreciation of the dollar from 1979 to 1982 did not raise the effective interest rate paid by Indonesia as much as it did for that paid by Mexico and Brazil. The high degree of export orientation in Indonesia prevented its debt-servicing capacity from collapsing as did Mexico’s when the price of oil dropped in early 1982. Appropriate exchange rate policies by Indonesia, exemplified by the 1978 devaluation, ensured a diversified export bundle as well as a high export orientation. Indonesia’s political concern to keep the agricultural sector vibrant no doubt helped to maintain the observed export orientation. The shock of the 1975 Pertamina crisis caused official borrowing in Indonesia to take place very cautiously with regard to exposure in the

136

Wing Thye Woo and Anwar Nasution

short-term credit market. The resulting Indonesian prudence is the major reason the maturity structure of Indonesian debt was so drastically different from that of Mexico. We could also refer to the this factor as the Pertamina legacy.

8.4 The Relative Contribution of Each Factor To get a sense of the relative contribution of concessional IGGI loans, prudence in debt management, and export orientation in explaining the absence of an Indonesian debt crisis, we decompose the average 1980-82 DSR. This is done by comparing the Indonesian debt situation with that of Mexico. We chose Mexico over Brazil because the former is an oil-exporting country like Indonesia. On the eve of the debt crisis, oil was the chief foreign exchange earner and the biggest source of government revenue in both Indonesia and Mexico. We calculate what the DSRs would have been if Indonesia: (i) paid the same effective interest rates as Mexico; (ii) had managed its debt such that its maturity structure was the same as Mexico’s; and (iii) had the same export/GNP ratio as Mexico. In the construction of these counterfactual DSRs, we assume that the total debt of Indonesia remained unchanged in these alternative scenarios. Item (i) in part A of table 8.4 reports the DSRs normalized by the actual export level after the IGGI effect and the prudence factor were removed. Item (ii) normalizes the different debt services by the level of exports that would have come about if Indonesia had the same export/GNP ratio as Mexico. The last entry in item (ii) reports that if the Indonesian debt and economy assumed all three Mexican features, the resultant DSRs in 1980-82 would be two to three times larger than the actual, making a debt crisis highly probable. On average, Indonesia’s DSR would be 54 percentage points higher if it had all three Mexican features. Part B of table 8.4 reports the range of values assumed by each factor in six decompositions of their effect, along with the average contribution of each factor. We use average contribution in our discussion because theory gives us no guidance as to which decomposition is most natural. The results show that the export orientation of Indonesia is the most decisive factor in why Indonesia’s total DSR is so low compared with Mexico’s. Export orientation explains 3 1 of the 54 percentage point difference, accounting for 57 percent of the gap. The prudence factor was of moderate importance, contributing 18 percentage points and thus accounting for almost one-third of the gap. Concessional interest rates and the currency composition of debt played only a minor role in reducing the DSR, explaining less than 6 percentage points. Our finding that the IGGI effect contributed so little toward the reduction of the 1980-82 debt-serviceiexport ratio is surprising because many of the

137

IndonesidChapter 8 ~

Table 8.4

Relative Importance of IGGI, Maturity Structure, and Export Orientation Part A: Construction of counterfactual ratios for decomposition 1980 1981 1982

-

-

(i) Total debt-service normalized by actual exports Actual maturity structure 26.1 At actual interest rates 25.1 28.8 At Mexican interest rates 31.1 Mexican maturity structure 39.5 At actual interest rat37.1 41.7 At Mexican interest rates 42.0 (ii) Total debt-service normalized by counterfactual exports Actual maturity structure 54.9 At actual interest rates 54.6 60.8 At Mexican interest rates 67.6 Mexican maturity structure 83.3 At actual interest rates 80.5 87.9 At Mexican interest rates 91.3

-

Avg (80-82)

39.0 43.7

30.1 34.6

52.0 56.1

42.9 46.6

51.5 57.7

53.7 62.0

68.7 74.0

77.5 84.4

Part B: Relative importance as determined by average of the six possible decompositions Hypothetical 1980-82 average ratio when Indonesia has all three Mexican features is 84.4 percent Actual 1980-82 average ratio is 30.1 percent.

IGGI concessional loans Maturity structure due to Pertamina legacy and absence of capital flight Export orientation

Average Value

Percent of Gap Between Actual and Counterfactual Ratios Accounted For

Range of Values Assumed in the Six Decompositions

5.8

11%

3.7-8.4

17.7 30.8

32 57

12.0-23.8 23.6-37.8

informed observers we talked to cited foreign concessionary loans as the primary reason for the absence of an Indonesian debt crisis. Our point is that while the $1 billion saved annually in reduced debt service during 1980-82 is a large sum of money,5 this amount would have been easily swamped by a Mexico-style loss of reserves if the Indonesian government had tried to prop up an overvalued exchange rate and was then forced to finance capital flight. Similarly, if exports were 12 percent below actual value because of an overvalued exchange rate, as suggested by the 1965-68 experience, the loss in foreign reserves would have also greatly exceeded this $ 1 billion saving. Our conclusion is that the Indonesian exchange rate policy was the most important reason Indonesia was able to meet its debt commitments in the 1982-84 period. The conduct of this exchange rate policy was greatly facilitated by the existence of a political lobby which promotes exchange rate protection and by the memory of the economy-wide negative effects of exchange rate overvaluation. The fact that neither the budget deficits nor the money growth rates went out of their historical range for extended periods also helped to make exchange rate management easier.

138

8.5

Wing Thye Woo and Anwar Nasution

Prospects

Since over 30 percent of Indonesia’s public debt is denominated in yen, debt-service payments have jumped in the face of the 68 percent appreciation of the yen against the dollar in the 1985:lQ to 1987:lQ period. Debt management has become more difficult since 1984. Agricultural commodity prices and oil prices have continued to decline, often very rapidly. The price of oil plunged from $28/barrel to $lO/barrel between January and August 1986. The fall in export earnings from $21 billion in 1982 to $15 billion in 1986 has caused the total debt-servicelexport ratio to soar to 68 percent. The situation is ominous. While our analysis would place the greatest of emphasis upon an aggressive competitive real exchange policy to reduce the probability of a debt crisis through its effects on exports and capital flight, we strongly feel that there are a number of other policies which must be implemented immediately if a debt crisis is to be avoided in the medium run. The supplementary policies which we recommend can be divided into two groups: (1) those which affect the debt service directly, and (2) those which affect export earnings. The policy measures which would ameliorate the debt service burden directly, through the reduction of foreign borrowing, are: 1. Cut budget deficits by controlling spending and increasing taxes. The fiscal policy posture must be kept consistent with that of the exchange rate. The tax reforms since January 1984 have raised domestic revenue considerably, but their implementation has not been wholly satisfactory. While the number of registered taxpayers has increased from 550,000 before the tax reform to 995,000 at the end of 1985, only 50 percent of the companies and 70 percent of registered taxpayers actually filed tax returns in 1985 (World Bank 1986, 13). The elimination of this slack in tax collection should not cost too much given that the offenders are already known. The task now is to fully enforce existing tax laws. 2. Maintain an anti-inflationary stance in monetary policy. This would help to keep trade account deficits down by reducing absorption relative to income. Interest rates should be kept internationally competitive to discourage capital flows. Both of these outcomes would make exchange rate management easier. 3. Amend the “balanced” budget rule to allow internal financing of government deficits. This rule was introduced to prevent the reoccurrence of the kind of inflation in 1961-65 that resulted from the monetization of the budget deficits. It may seem imprudent to remove this institutionalized practice but the fact is that the inflation of the final Soekarno years was the result of the breakdown of the political system which made austerity policies impossible. And if political conditions were to really degenerate, no government would deem itself bound to this practice

139

IndonesidChapter 8

anyway. Furthermore, at present, while this rule prevents monetization of budgetary expenditure, it does not prevent the monetization of nonbudgetary expenditure, e.g., central bank credits to BULOG. Since bold austerity policies during bad times have never been avoided by the Soeharto regime, it makes little sense not to amend the balanced budget practice in order to reduce reliance on external funds. In addition, the development of a domestic market in government securities would make open market operations by the central bank possible. The existence of this monetary tool would enhance monetary control, and thus macroeconomic stabilization efforts, tremendously. 4. Accelerate the development of the domestic financial system. Besides further deregulation of the financial sector, financial deepening could be boosted by the privatization of many of the state-owned enterprises. The balance-of-payments position would be improved if the government were to allow foreigners to purchase shares in the former state enterprises. A developed financial market would lower intermediation costs, allow better monetary control, and, possibly, encourage savings. 5. Liberalize the controls on foreign investments in the manufacturing and agricultural sector, especially in industries that produce primarily for the export markets. This step will increase capital formation without the need of incurring external debt and will also increase foreign exchange earnings. In short, state and private enterprises should issue equities instead of bonds to foreign investors when financing their capital expenditure. It is important that protection not be used as a means of inducing foreign investments because, in all likelihood, the resulting enterprise would be inefficient and cause a net loss of foreign exchange for Indonesia. Our proposal for liberalizing foreign investment laws may be a hard one to accept given the prevailing economic nationalism in Indonesia, but it should be seriously considered if the debt situation takes another turn for the worse. The second group of supplementary external debt management policies are those which focus on the denominator of the DSR. Our analyses suggest that the viability and expansion of the Indonesian export sector depends crucially on : 1. The elimination of the wide array of monopoly import licenses. The present efforts to replace import licenses with tariffs is an improvement but it is still a second-best solution. It is important that tariffs not take the place of the import licenses removed from the imported basic inputs. The growth of manufactured exports, spurred by access to cheaper inputs, will not only increase foreign exchange earnings but will also diversify the export bundle, hence reducing the sensitivity of the DSR to the prices of a few key commodity exports. Furthermore, any Indonesian manufactured export industry that is internationally competitive will be one which uses Indonesia’s abundant semiskilled labor force intensively in its

140

Wing Thye Woo and Anwar Nasution

production process. The favorable employment effects alone should be justification enough to eliminate these monopoly import licenses. 2. The expansion of the tree crop sector. Indonesia has cheaper labor than Malaysia, and with additional investments in transportation, Indonesia could potentially outcompete Malaysia in the production of rubber and palm oil. In addition to earning more foreign exchange, the strategy of accelerating the growth of agricultural export industries will also promote a more equitable rural-urban, as well as inter-island, growth pattern and ease population pressure on the urban areas.

A final cautionary word on external debt management from the political perspective is pertinent. The Pertamina crisis has led to close supervision by the Ministry of Finance of external borrowing by all state enterprises, making it unlikely that a debt crisis would ever again emerge from the external adventurism of an economic fiefdom. The danger now may be the absorption of private external debts in order to save large domestic firms when they get into financial problems, as in the Indocement case. As we described in chapter 3, in July 1985 Indocement, the biggest cement company in Indonesia, began to experience cash flow problems because the recession-induced collapse of the construction industry led to a cement glut. The response of the Indonesian government was to inject U.S. $325 million in cash to acquire a 35 percent share of the company, and to form a consortium of four state banks to “convert into a rupiah liability a U.S. $120 million syndicated loan that Indocement took out in 1981.’16 If a few more such rescues are allowed, then the habit may well be impossible to break without the government having to put to the test an important source of its political power-the cohesiveness of the bureaucratic and military elite. Given the widespread participation in large private business ventures by government officials and their family members, the selective use of financial rescue will threaten the political unity of the group. If such political pressure were able to completely eradicate the already blurred line between public and large private enterprises, then the vulnerability of Indonesia to a debt crisis would be greatly increased. External debt management would become impossible because no one would know what the size of the sovereign debt really was, and the size of this debt could increase very quickly given the openness of the private capital account.

Notes Acknowledgments We received a lot of help in the writing of this monograph. We are beholden to the many government officials, policy analysts, diplomatic observers, bankers, journal-

141

IndonesidNotes

ists, and academic researchers who graciously agreed to discuss the current debt situation with us. In order to keep this note brief, we have not cited each by name. Our apologies, as well as our thanks, to our unnamed benefactors. We are very grateful to the late Professor Dr. Jusuf Panglaykim Pangestu for arranging visiting privileges for Wing T. Woo at the Center for Strategic and International Studies, and to Mr. Abdulgani for doing the same at Bank Duta. Bruce Glassburner helped us greatly by giving generous access to his collection of materials on Indonesia. We also want to thank Willem Bake, Suhadi Mangkusuwondo, Masaaki Horiguchi, Richard Mattione, Stephen Mink, Gordon Nelson, Robert Rucker, Syahrir Sabirin, Pande Silalahi, Djisman Simandjuntak, Sutopo, Ast, Peter Timmer, Ralph Beak, Richard Porter, David Dapice, and Khalid Wajid for data and references which permitted us to do a more complete analysis. John Roemer, Thomas Mayer, Kevin Hoover, Tun-Jen Cheng, William Liddle, Anne Booth, John Legge, Edward Buffie, Dani Rodrik, Donald Snodgrass, Mark Sundberg, Heinz Arndt, Eric Ramstetter, and Seiji Naya read portions of an earlier draft and gave us extremely helpful comments. Evie Anwar, Ratna Astiti, Sirajudin Ahmad, Alian, Wendy Eudey, Petrina Ho, Carl Galopin, Elmer Bartley, and Jonathan Kendall provided extremely effective research support which enabled us to meet the NBER deadlines. The warm hospitality extended to Wing T. Woo during his stay in Jakarta by Donald and Oralia Foster-Gross, Mari Elka Pangestu, Sjahrir, Richard Monteverde, Ibrahim Hasan, Mr. and Mrs. Van Rennes, Mr. and Mrs. Rucker, and Ayuna Nasution made his research experience a very enjoyable one. We are deeply indebted to Max Corden, David Dapice, Peter McCawley, Sarath Rajapatirana, and Jeffrey Sachs for the many perceptive comments and advice we received while writing this monograph. Wing Thye Woo is responsible for writing chapters 1, 3, 4, 6, 7, and 8. These chapters are partially based on three earlier papers: “The Dead Hand of History in Indonesian Economic Policymaking” and “The External Debt Situation in Indonesia: Performance and Prospects” (no, 43 and no. 48, respectively, in the Applied Macroeconomics Working Paper Series issued by the University of California, Davis); and “The Economic Policy-Making Equation in Indonesia” (Pacific Rim Studies Program Working Paper no. 6, University of California, Davis). Anwar Nasution is responsible for writing chapters 2 and 5. These chapters draw on some materials from “Instruments of Monetary Policy in Indonesia after the 1983 Banking Deregulation” (a paper prepared for the Conference on Financial Research in Indonesia, August 1986) and from Financial Institutions and Policies in Indonesia, Institute of Southeast Asian Studies, Singapore, 1983.

Chapter 1 1. The “balanced” budget rule ceased to be a binding constraint on expenditure after the 1973 OPEC-1 price increase. The Indonesian government (perhaps until very recently) was pretty much able to get whatever amount of credit it wanted. This budget rule does not prevent the financing of nonbudgetary expenditure by money creation, e.g., central bank credits to state agencies. 2. The term “Dutch disease” originated with the observation that the 1973-78 growth of the Dutch natural gas industry was accompanied by a steady decline in the traditional export sector.

142

Wing Thye Woo and Anwar Nasution

3. In the case where a private borrower could not come up with the service payments in domestic currency for his private nonguaranteed external debt, we do not consider it a national debt crisis because the government did not cause the default except in the broadest sense of not creating more favorable macroeconomic conditions, if it were able to do so. 4. This is calculated assuming that Indonesia would pay the same effective interest rate as Mexico.

Chapter 3 1. Sachs (1985) argues that it is political considerations which explain why Latin American countries, but not the East Asian countries, pursued exchange rate and trade policies which rendered their economies vulnerable to an external debt crisis. 2. Other viewpoints on the political economy aspects of Indonesian policymaking are: Jackson (1978a, 1978b), Palmer (1978), Mortimer (1974), Glassburner (1978), and Liddle (1987). 3. This characterization-of state-society relations is developed in Wing Thye WOO (1988). His article also reviews and tests the competing models of the Indonesian state. 4. Since 1961 there has been a secular decline in the real money stock. See table 11.8 in Amdt (1971). 5. We postpone discussion of the reorganization of the external debt to chapter 7. 6. This figure is not strictly accurate because multiple official exchange rates for different types of transactions still existed at this time. Data is from Arndt (1966). 7. This was calculated by deflating the dollar earnings of nonoil exports by the dollar export unit value of the industrial countries. The absence of a reliable representative exchange rate for 1965 makes the calculation in local purchasing power impossible. 8. The annual real GDP growth rate of the industrialized countries averaged 5.5 percent in the three years prior to 1965 and averaged 4.9 percent in the 1965-68 period. 9. This coup is commonly referred to as the GESTAPU coup, with GESTAPU standing for Gerakan September Tiga Puloh-the September 30 Movement. 10. For example, he maintains a working farm for recreational purposes. 11. The great care which Soeharto takes to preserve his country-boy image is seen by the POP affair. POP, a popular culture magazine, claimed in 1974 that Soeharto was actually a descendent of the Yogjakarta royal family. Instead of ignoring this gossip which would only add to the legitimacy of his rule among the superstitious and conservative peasants, Soeharto publicly rebuked the close confidant who owned the magazine. 12. See Timmer (1975) for an excellent discussion of the political economy of Indonesia’s rice policy. 13. Dwifungsi is a transliteration of “dual function” and refers to the army’s dual role in defense and economic development. Harold Crouch (1978) cites official statements in 1969 and 1970 that the armed forces budget covered only 30 to 50 percent of expenses. 14. Access to these funds may be the reason the defense budget has not jumped dramatically in the face of continued military operations in East Timor.

143

IndonesidNotes

15. Another important reason for the use of quotas over tariffs is that tariffs would violate the GA’IT agreement. 16. See The Asian Wall Street Journal Weekfy, 1 December 1986. 17. INSIGHT, 9 February 1987, p. 33. 18. Bank Indonesia, embarrassed by the size of the rescue operations, delayed the release of information about its financial status for over a year. 19. Soekarno, Soeharto’s predecessor, had resorted to the inflation tax to pay for government operations. 20. Glassburner (1976) cited this as an instance of managerial incompetence, so he may disagree with our interpretation here. 21. Siregar (n.d., 5) noted that “it was generally believed that while the Dutch colonial rule had a stifling effect on autochthonous Indonesian entrepreneurs, it had fostered the entrepreneurial activity of the Dutch and other foreigners, particularly the Chinese.” 22. In the preface to their book, Panglaykim and Palmer (1969) advanced the proposition that state corporations could be used to provide a counterweight to Chinese domination of the private trade sector. (Zaibatsus were large holding companies that dominated the Japanese economy until the end of World War 11.) 23. Liddle (1987) labels the two groups economists and engineers, respectively. 24. An account of IGGI’s role in the 1966-70 period is given in Posthumus (1971). 25. Because of their influential foreign constituency, their lack of ideological aversion toward foreign investment, and the fact that several of them received doctorates from the University of California, the technocrats were called the “Berkeley Mafia” by their detractors. 26. Two earlier attempts to check Pertamina’s influence had failed. When the minister of mining attempted to assert his control over Pertamina in 1969, Soeharto transferred the company out of the minister’s jurisdiction. In 1971 Soeharto ignored the recommendation by a presidential commission on corruption that Pertamina be placed under the supervision of the Ministry of Finance. 27. The term “technician” is used because of the increasingly prominent role of Minister of Technology Habibie, who is an outstanding aerospace engineer. 28. This belief is evident in the following editorial in a magazine identified with the military advisors to the president: One [path of economic development] leads via the BAPPENAS technocrats to a free-flight and laissez fake pattern of development in the Western and American fashion. Another [path] takes the form of cooperation with Japan on the basis of one’s own strength without loans from the IGGI, the IMF and the World Bank with Pertarnina as guarantee. (Ekpres, 18 January 1974, quoted in Fur Eastern Economic Review, 4 February 1974)

Chapter 4 1. The Indonesian government made this change beginning with 1984185 but did not revise the figures of earlier years, so our revision merely makes the official categories consistent across time. 2. Real import tax revenue, measured in 1980 prices, fell by Rp 62 billion in the 1969-83 period.

144

Wing Thye Woo and Anwar Nasution

3. The rise in oil tax revenues in 1983/84 is illusory because the 1983 devaluation increases the rupiah value of this revenue, which is paid in U.S. dollars. 4. Nonmining RGDP is used instead of RGDP because while the presence of oil and LNG industries increases a region’s production tremendously, their impact on the region’s income is small. This is because almost the entire revenue is taken by the central government and, being extremely capital-intensive industries, their local wage bills are small. 5 . Jakarta is excluded here because its high figure “reflects the substantial level of spending on the apparatus of the Central Government rather than on the development of Jakarta” (World Bank 1984, 133).

Chapter 5 1. More detailed descriptions of the Indonesian financial markets from 1970 to 1979 are found in Nasution (1983) and Suwidjana (1984). 2. A fuller discussion of the new monetary instruments is in Nasution (1986). 3. This liquidity squeeze has been immortalized in the Indonesian press as the Sumarlin shock treatment, Gebrakan Sumarlin.

Chapter 6 1. See Corden and Mackie (1962) and Kanesa-Thasan (1966) for analyses of the Indonesian exchange rate system prior to the Soeharto years. 2. Glassburner (1970) and Pitt (1971) describe the pricing of foreign exchange during the transition to the unified exchange rate. 3. The low figure of 6.4 weeks for 1975 shown in table 6.2 was not because of a sharp drop in exports or a big increase in imports; it was because of the transitional financing of Pertamina’s debts while negotiations were underway to convert the company’s short-term debts to long-term debts. 4. On the eve of the devaluation, Arndt (1978, 10) wrote:

The overall picture for Indonesia’s balance of payments in the current fiscal year is thus still reasonably reassuring. On a longer view, it is difficult to resist the traditional balance of payments pessimism of economists everywhere, official and unofficial alike.

5. See stories in Warta Berita, April 6; Sinar Harapan, April 13; and Merdeka, May 5 and May 24. 6. An exponent of this view is Kincaid (1984, 88), who wrote that the 1978 “exchange rate adjustment” was prompted by “the balance of payments constraint.” 7. Notable supporters of the Dutch disease interpretation of the 1978 devaluation are Malcolm Gillis, Dwight Perkins, Michael Roemer, and Donald Snodgrass (1987, 539) who wrote: [In 19781 the country had ample foreign exchange reserves to defend the vale of the rupiah, and the business community viewed devaluation as unthinkable. Nevertheless, the Indonesian rupiah was devalued by 50 percent in November 1978, not to protect a precarious foreign exchange reserve situation but primarily to encourage labor-intensive manufactured and agricultural exports.

145

IndonesidNotes

8. These normalizations were suggested by Peter Warr (1986). 9. This statement is true as long as international prices do not rise above the domestic prices of the quota-protected goods. Tariff-protected goods are still considered tradables because their domestic prices are higher than international prices only by the amount of the tariff and they move in tandem with international prices. 10. The OPEC-1 price increase occurred in October 1973. 1I . Pangestu (1986) constructed several price indices for tradables and nontradables by decomposing the gross domestic product, the CPI, and the WPI. Her results are qualitatively the same as those reported in table 6.5. 12. The Morgan Guaranty measure of competitiveness is the inverse of the real exchange rate reported in World Financial Markets published by Morgan Guaranty. The unit of measurement of competitiveness here is rupiahs per basket of foreign currencies. 13. Kincaid (1984) and Pangestu (1986) provide formal econometric explorations of the devaluation. 14. We have. as Gillis (1984) did in his table 4, identified the “miscellaneous” category in the World Bank list of nonoil exports as manufactured goods. 15. These disaggregated nonoil export data are available only in fiscal year intervals. Fiscal year 1979 goes from April 1979 to March 1980 and is henceforth represented by 1979/80. 16. McCawley (1980) discusses the need for other economic policy changes in order for the beneficial structural effects of the devaluation to be viable. 17. Debt here refers to medium- and long-term external public debt. 18. International Financial Statistics, June 1987. 19. International Financial Statistics, June 1987. The IMF, at the beginning of 1987, further lowered its prediction of the 1987 growth rate to 2.3 percent (IMF 1987). 20. Even after taking into account the fall in import prices, the real price of oil measured in terms of foreign purchasing power fell a whopping 40 percent in one year. 21. In this section, we confine our attention to the consequences of quotas on the supply of exportables within a static setting. For an excellent discussion of the impact of protection on growth, see Corden (1972). 22. Pitt (1971) and World Bank (1981) are two comprehensive studies of effective rates of protection in 1970 and 1975, respectively. 23. As pointed out in section 6.4, goods with QR protection are effectively sheltered from international price trends and are hence regarded as nontradables from an analytical viewpoint. 24. This is not the total level of protection as this figure does not take into account activities protected by quotas and tariffs. 25. See Wall Street Journal, 24 November 1986 for a longer list of vital inputs with a small number of approved traders. 26. Figures are from The Asian Wall Street Journal Weekly, 1 December 1986. 27. This assumes that the rent-seeker is rational enough not to demand too high a price and obliterate the industry. 28. We are abstracting from natural growth considerations here to make this point within a static context. 29. Current account expressed as a percentage of GDP.

146

Wing Thye Woo and Anwar Nasution

Chapter 7 1. The conversion to rupiah was done using the average BE Market rate of October to December, 99.5 rupiahs/dollar. The average curb market rate was 130 rupiahsldollar. Data are from the World Bank (1968), Basic Data section and pp. 39, 50, and 54. 2. Lipsky (1978), Glassburner (1976), and McCawley (1978) are excellent sources about and analyses of the Pertamina crisis. 3. According to Glassburner (1976), the power installation at the Krakatau Steel Mill cost three times more than the same installation in Taiwan. 4. These figures are only guesses because the same item from three official sources-balance of payments, the government debt record, and the government budget-sometimes shows gaping discrepancies. See also the note in table 7.4 on how the 1973/74 through 1977/78 data were constructed. 5. The rate of change in the CPI was 31 percent in 1973, 41 percent in 1974, and 19 percent in 1975; and real GDP growth was 7 percent in 1976, 9 percent in 1977 and 8 percent in 1978. 6. This paragraph draws upon the article by Haryono (1985). Chapter 8 1. A textbook exposition of this fiscal cause of external debt crisis is Rivera-Batiz and Rivera-Batiz (1985). 2. For example, see table 17.5 of Rivera-Batiz and Rivera-Batiz (1985, 557). 3. For example, Dombusch (1987), Lever and Huhne (1986), and Cline (1984). 4. Cuddington (1986), Dooley (1986), Khan and Haque (1987), and Morgan Guaranty (1986). 5. This is calculated assuming that Indonesia would pay the same effective interest rate as Mexico. 6. Quote is from the Far Eastern Economic Review, 25 July 1985. The Asian Wall Street Journal Weekly, 12 May 1986, put the cash injection at U.S. $360 million.

References Amdt, H. W. 1966. Survey of current developments. Bulletin of Indonesian Economic Studies October. -. 1971. Banking in hyperinflation and stabilization. In The economy of Indonesia: Selected readings, ed. by B. Glassburner. Ithaca, N.Y.: Cornell University Press. -. 1978. Survey of recent developments. Bulletin of Indonesian Economic Studies November. -. 1979. Monetary policy instruments in Indonesia. Bulletin of Indonesian Economic Studies November. Bank Indonesia. Annual Report, various issues. Jakarta.

147

-.

IndonesidReferences

Indonesian Financial Statistics, various issues. Jakarta. Report of the Governor, various issues covering financial years 1950/51 through 1960/65. Jakarta. -. Yearly Report, various issues. Jakarta. Biro Perancang Negara. 1960. Report on the execution of the five-year development plan, 195640: Covering the years 1956, 1957, and 1958. Jakarta, August. Biro Pusat Statistik. Statistical Pocketbook of Indonesia, various issues. -. Statistical Yearbook of Indonesia, various issues. Booth, Anne, and Peter McCawley, ed. 1981a. The Indonesian economy during the Soeharto era. Oxford: Oxford University Press. . 1981b. Fiscal policy. In Booth and McCawley (1981a). Cline, William. 1984. International debt: Systemic risk and policy response. Washington, D.C.: Institute for International Economics. Corden, Max. 1972. Protection and growth. In International economics and development: Essays in honor of Raul Prebisch, ed. Luis Marco. New York: Academic Press. . 1982. Exchange rate protection. In The international monetary system under flexible exchange rates, ed. Richard Cooper, Peter Kenen, George de Macedo, and Jacques van Ypersele. Cambridge, Mass.: Ballinger. Corden, Max, and J. A. C. Mackie. 1962. The development of the Indonesian exchange rate system. Malayan Economic Review April. Crouch, Harold. 1978. The army and politics in Indonesia. Ithaca, N.Y.: Cornell University Press. Cuddington, John. 1986. Capital flight: Estimates, issues and explanations. Princeton Studies in International Finance 58 (December). Dake, Antonie. 1973. In the spirit of the Red Banteng: Indonesian communists between Moscow and Peking, 1959-1965. The Netherlands: Mouton. Dapice, David 0. 1980. An overview of the Indonesian economy. In The Indonesian economy, ed. Gustav F . Papanek. New York: Praeger. Dooley, Michael. 1986. Capital flight: A response to differences in financial risks. Mimeo, July. Dornbusch, Rudiger. 1987. The world debt problem: Anatomy and solutions. Study paper for the Twentieth Century Fund, New York. Enders, Thomas, and Richard Mattione. 1984. Latin America: The crisis of debt and growth. Washington, D.C.: Brookings Institution. Gablis, V. 1977. Financial intermediation and economic growth in LDC: A theoretical approach. Journal of Development Studies January. Geertz, Hildred. 1963. Indonesia culture and community. In Indonesia, ed. Ruth McVey. New Haven, CT: Yale University Press. Gillis, Malcolm. 1984. Episodes in Indonesian economic growth. In World economic growth, ed. Arnold Harberger Institute of Contemporary Studies. Gillis, Malcolm, Dwight Perkins, Michael Roemer, and Donald Snodgrass, 1987. Economics of development. 2d.ed. New York: W. W. Norton. Glassburner, Bruce. 1970. Pricing of foreign exchange in Indonesia, 1966- 1967. Economic Development and Cultural Change January. -, ed. 1971. The economy of Indonesia: Selected readings. Ithaca, N.Y.: Cornell University Press.

-.

148

Wing Thye Woo and Anwar Nasution

. 1976. In the wake of General Ibnu: Crisis in the Indonesian oil industry. Asian Survey December. . 1978. Political economy and the Soeharto regime. Bulletin of Indonesian Economic Studies November. Gray, Clive. 1979. Civil service compensation in Indonesia. Bulletin of Indonesian Economic Studies March. Haryono, Subaliono. 1985. Indonesia. In External debt management, ed. Hassanali Mehran. Washington, D.C.: International Monetary Fund. Hunter, Alex. 1967. Indonesian oil: A new generation of explorers. Bulktin of Indonesian Economic Studies October. International Financial Statistics. various issues. International Monetary Fund. International Monetary Fund. 1986a. Indonesia Country Report. -. 1986b. World Economic Outlook April. -. 1986c. World Economic Outlook October. -. 1987. World Economic Outlook, April. Jackson, Karl. 1978a. The prospects for bureaucratic polity in Indonesia. In Jackson and Pye (1978). . 1978b. Bureaucratic polity: A theoretical framework for the analysis of power and communications in Indonesia. In Jackson and Pye (1978). Jackson, Karl, and Lucian Pye, ed. 1978. Political power and communications in Indonesia. Berkeley, Calif.: University of California Press. Kanesa-Thasan, S. 1966. Multiple exchange rates: The Indonesian experience. International Monetary Fund Staff Papers, July. Khan, Mohsin, and Nadeem Haque. 1987. Capital flight from developing countries. Finance and Development 24, no. 1 (March) 2-5. Kincaid, G . Russell 1984. A test of the efficacy of exchange rate adjustments in Indonesia. International Monetary Fund Staff Papers, March. Lever, Harold, and Christopher Huhne. 1986. Debt and danger: The worldfinancial crisis. Boston, Mass.: Atlantic Monthly Press. Liddle, William. 1987. Indonesia in 1986: Contending with scarcity. Asian Survey February. Lipsky, Seth. 1978. The billion dollar bubble. In The billion dollar bubble, ed. Seth Lipsky. Hong Kong: Dow Jones. McCawley, Peter. 1978. Some consequences of the Pertamina crisis in Indonesia. Journal of Southeast Asian Studies March. -. 1980. The devaluation and structural change in Indonesia. Southeast Asian Afsairs 1980. Singapore: Institute of Southeast Asian Studies. McDonald, Hamish. 1980. Suharto’s Indonesia. London: Fontana. McKinnon, Ronald. 1973. Money and capital in economic development. Washington, D.C.: Brookings Institution. McVey, Ruth, ed. 1963. Indonesia. New Haven, CT: Yale University Press. Morgan Guaranty. 1986. LDC capital flight. World Financial Markets March. Mortimer, Rex, ed. 1974. Showcase state: The illusion of Indonesia’s “accelerated modernization”. London: Angus and Robertson. Nasution, Anwar. 1983. Financial institutions and policies in Indonesia. Singapore: Institute of Southeast Asian Studies. -. 1986. Instruments of monetary policy in Indonesia after the 1983 banking deregulation. Paper presented to the Conference on Financial Research in Indonesia. August. Jakarta, Indonesia.

149

IndonesidReferences

Newmann, John. 1974. Inflation in Indonesia: A case study of causes, stabilization policies and implementation (1966- 1970). Ph.D. diss., Tufts University. Palmer, Ingrid. 1978. The Indonesian economy since 1965: A case study of political economy. London: Frank Cass. Pangestu, Mari Elka. 1986. The effects of an oil boom on a small oil exporting country: The case of Indonesia. Ph.D. diss. University of California, Davis, September. Panglaykim, Jusuf, and Ingrid Palmer. 1969. State trading corporations. Rotterdam. Papanek, Gustav, F., ed. 1980. The Indonesian economy. New York: Praeger. Penny, D. H., and Dahlan Thalib. 1967. Survey of current developments. Bulletin of Indonesian Economic Studies February. Pitt, Mark. 1971. Alternative trade strategies and employment in Indonesia. In Trade and employment in developing countries, volume I , ed. Anne Krueger, Hal Lary, Terry Monson, and Narongchai Akransee. Chicago: University of Chicago Press. Posthumus, G. A. 1971. The Inter-Governmental Group on Indonesia. Rotterdam: Rotterdam University Press. . 1972. The Inter-Governmental Group on Indonesia. Bulletin of Indonesian Economic Studies July. Rivera-Batiz, Francisco, and Luis Rivera-Batiz. 1985. International finance and open economy macroeconomics. New York: Macmillan. Sachs, Jeffrey. 1985. External debt and macroeconomic performance in Latin America and East Asia. Brookings Papers on Economic Activity 2. Siregar, Arifin. n.d. Indonesian entrepreneurs. Mimeo. Sundhaussen, Ulf. 1978. The military: Structure, procedure and effects on Indonesian society. In Political power and communications in Indonesia, ed. K. Jackson and L. Pye. Berkeley, Calif.: University of California Press. Suwidjana, Njoman. 1984. Jakarta dollar market: A case offinancial development in ASEAN. Occasional paper no. 76. Singapore: Institute of Southeast Asian Studies. Timmer. Peter. 1975. The political economy of rice in Asia: Indonesia. Food Research Institute Studies 14(3). Warr, Peter. 1986. Indonesia’s other Dutch disease: Economic effects of the petroleum boom. Working Papers in Trade and Development no. 86/2. Research School of Pacific Studies, Australian National University, April. Woo, Wing Thye. 1988. The economic policy-making equation in Indonesia. Pacific Rim Studies Program Working Paper no. 6. University of California, Davis. World Bank. 1968. World Bank country report on Indonesia. . 1975. World Bank country report on Indonesia. -. 198I . Indonesia: Selected issues of industrial development and trade strategy, annex 2: The foreign trade regime. -. 1984. World Bank country report on Indonesia. . 1985. World Bank country report on Indonesia. -. 1986. World Bank country report on Indonesia. -. 1987a. World Bank country report on Indonesia. . 1987b. World development report 1987.

This Page Intentionally Left Blank

I1

External Debt and Macroeconomic Performance in South Korea Susan M. Collins Won-Am Park

This Page Intentionally Left Blank

1

Introduction

In 198 1 South Korea was the world’s fourth largest debtor country and in the midst of an economic crisis. It had accumulated $17.6 billion of debt within three years, raising its debt stock to $32.4 billion and its debt/GDP ratio to 49 percent. Output had declined by 4.8 percent in 1980, compared to average growth rates in excess of 9 percent during 1970-79. Inflation had doubled from 14.4 percent in 1978 to 28.7 percent in 1980.’ Korea’s adjustment to the 1979-82 debt crisis has been remarkable. Some of the key elements in the adjustment are shown in table 1 . I . By 1986 it had substantially reduced the debt burden. Inflation had fallen to just 3 percent, while the government budget deficit had been cut in half. Exports grew by 26.6 percent, fueling a 12.5 percent increase in output and a current account surplus (4.9 percent of GNP). At the same time, real wages, per capita income, and consumption all increased, and the country maintained historically high levels of fixed capital formation. In stark contrast, the World Bank’s World Development Report (1986, 54) describes the plight of seventeen of the middle-income debtor countries as follows: The bulk of the adjustment has been undertaken through lower demand, which has meant, in practice, reducing imports and investment. The volume of imports for the heavily indebted middle income countries in 1985 was 32 percent below its 1981 level. The ratio of investment to GDP fell from 25 percent in 1981 to 18 percent in 1985. GDP has stagnated since 1980, and per capita incomes have declined substantially. The The views expressed in this study represent those of the authors and not necessarily those of their respective institutions. The authors would like to thank Vittorio Corbo, Anne Krueger, and an anonymous reviewer for comments and suggestions, and Joo-yup Ahn, long-ha Yoon, Jee-won Park. and Ana Revenga for research assistance. Susan Collins gratefully acknowledges financial support from an Olin Fellowship.

153

154

Susan M. Collins and Won-Am Park

Table 1.1

Korea’s Economic Reeoverv

A: Average Growth Rates

1970-79

1980-82

1983-85

1986

9.5 14.9

2.3 18.8

B: External Indicators (selected years)

1978

1980

1982

1985

1986

Debt (billions U.S. $) DebVGNP Current accounVGNP

14.8 28.5 - 3.0

27.2 45.0 -9.6

37.1 53.5 -4.5

46.8 56.3 - 1.8

44.5 46.8 4.9

GNP growth Consumer prices

8.5 2.7

12.5 2.3

reduction in demand has pushed the collective trade balance of these countries into a large surplus, which has brought their current account into rough balance. Yet the main indicators of debt at the end of 1985 were close to their previous peaks. Despite their adjustment efforts, these countries seem to be as far as they ever were from reconciling growth and creditworthiness. Thus, it is not surprising that Korea’s experience has been labeled “a case of successful adjustment” (Aghevli and Marquez-Ruarte 1985) raising a number of important and provocative questions. What were the secrets of Korea’s performance? How important were economic structure, policy choices, social and political factors, and external developments? Are there lessons to be learned which could help other debtor countries to achieve a more favorable balance between growth and external adjustment? This study analyzes Korea’s macroeconomic performance, policy, and prospects so as to provide answers to these questions. Particular emphasis is given to the role of external debt in contributing to the crises as well as to the recoveries. Korea’s position in the limelight is not new. The remarkable transformation from a war-devastated economic “basket case” heavily dependent on foreign aid in the 1950s to a newly industrialized “economic miracle” with impressive growth rates during the 1970s has been well documented.2 Clearly, this historical development is linked to Korea’s ability to adjust to the recent crisis so rapidly. We pay close attention to the implications of Korea’s structural development in putting together the pieces to explain the 1980s performance. Korea has also received international attention due to labor unrest and to opposition to the slow progress toward a democratic political process. While an in-depth analysis of the interactions between politics and economic performance is beyond the scope of this study, we recognize that political and social factors are integral forces in the process of economic development. Thus we have attempted to integrate some of these factors into our discussion in places where we found them to be especially relevant. For example, the sociopolitical environment seems to have had an influence over

155

KoredChapter 1

wage determination, over the extent to which announced economic plans have been viewed as credible, and over the enforceability of economic policy. The purpose of this chapter is to provide an overview of the key elements of Korea’s experience so as to set the stage for the remaining chapters. It is composed of two sections. The first section briefly reviews Korea’s macroeconomic performance, identifying the central issues to be analyzed. The second section outlines the rest of the chapters in the study.

1.1 Overview of Macroeconomic Experience Korea’s macroeconomic history can be divided roughly into five periods. The early period, from 1945 to 1953, was one of continued disruption. First came the division into North and South Korea at the 38th parallel after World War 11. The South was left with rich agricultural lands and light manufacturing industries, but almost no heavy industry or power facilities. Attempts to begin economic recovery were interrupted by the devastation of the Korean War which is estimated to have killed over one million people and destroyed over one-third of South Korea’s physical capital. Another development during this period, with lasting implications for Korean development, was a major land reform. During 1947-49, farmland previously owned by Japanese landlords was either redistributed or sold, dramatically decreasing the concentration of land ownership. This is perhaps the most important factor in explaining the relatively egalitarian distribution of income in Korea. The second period (1953-60) was one of slow recovery financed by massive foreign aid, primarily from the United States. Foreign aid inflows averaged nearly U.S. $300 million per annum during 1955-59, reaching 16 percent of GNP in 1957. Inflation rates jumped to 60 percent immediately following the war, while output growth remained moderate. Under the complex system of trade restrictions erected by the Syngman Rhee dictatorship, exports grew by only 1.3 percent per year. In contrast, the third period, from 1960-73, saw a dramatic economic turnaround fueled by rapid rates of export growth. Exports grew by 40-50 percent per year during 1960-73, while output grew by more than 10 percent during 1965-73. The economic transition coincided with a change in political regime and economic policy. Syngman Rhee was forced to resign in 1960 after a student uprising. The new government, led by Chang Myon, collapsed in May 1961 following a military coup led by General Park Chung Hee, who remained president of Korea until a second coup in 1979. Under General Park, Korea switched from an import-substitution strategy to an active export-promotion strategy. The first of a series of five-year plans, initiated in 1962, identified investment and economic growth as the

156

Susan M. Collins and Won-Am Park

number one priorities. Other hallmarks of the strategy were extensive government intervention in domestic and international capital markets, the development of close links between government and industry, import liberalization, and the more active use of exchange rates to maintain competitiveness. Foreign aid inflows fell dramatically during the period. During 1960-64, they averaged $210 million per year, over ten times the average annual accumulation of external debt. This inflow dropped to $1 10 million per year during 1965-69, just one-third of the average annual debt accumulation, and to only $28 million per year during 1970-74, or 0.03 percent of the debt accumulation. Foreign aid to Korea had essentially ended by 1975. Gross fixed investment was raised from 15 percent of GNP in 1965 to 26 percent in 1969. To finance the investment, declining foreign aid flows were replaced by increased reliance on external borrowing and by increased domestic saving. Firms (especially exporters) were given strong incentives to borrow abroad. A system of loan guarantees substantially reduced the risks. Furthermore, the real cost of borrowing abroad (in won) turned negative. External debt jumped to 27 percent of GNP by 1969. Difficulties emerged during 1970-72. As growth slowed, domestic savings dropped, increasing the current account deficit and reducing Korea’s debt service ability. A devaluation to stimulate exports exacerbated repayment difficulties for externally indebted firms. The government bailed them out and continued to pursue its investment strategy. Further depreciation was combined with some monetary and fiscal restraint. Taking advantage of strong world demand, exports grew by 90 percent in 1973. Output growth rose to a record 16 percent, stimulating a spurt in domestic saving and pulling Korea out of the first period of debt difficulties. The fourth period (1973-78) included a second period of rapid debt accumulation, economic difficulty, and recovery. It also coincided with a major shift in economic strategy-a renewed industrialization, coupled with increased government intervention. The “Big Push” was a massive investment program in heavy and chemical industries initiated in 1973 both because policymakers feared that Korea’s comparative advantage was shifting away from light industry and because they wished to strengthen Korea’s defense capabilities. The program coincided with a resurgence in inflation, a slowdown in export growth, a rise in the incremental capital output ratio, and a deterioration in the distribution of income. Import restrictions and credit rationing increased, and the exchange rate was fixed (1975-79) and allowed to appreciate. Although widely viewed as a policy mistake, some of the investments (e.g., autos) have begun to pay off. Economic growth slowed during 1974-75 in the aftermath of the oil and commodity price rise. Again, there was a drop in domestic savings, increasing the borrowing necessary to finance the investment program.

157

KoreaiChapter 1

Korea elected to “borrow its way” through the crisis so as to fulfill planned investment and to relax monetary and fiscal policies. As world demand recovered during 1976-78, high growth rates resumed, raising domestic savings and improving the debt position. In 1979 Korea underwent another shift in economic strategy. Motivated by concern over rising inflation rates and economic distortions from the Big Push, the new stabilization plan included monetary and fiscal restraint plus the gradual reduction of price controls, import restrictions, and financial market interventions. However, 1979-82 were years of crisis for Korea. The 1979 assassination of President Park together with a disastrous agricultural harvest and the second oil shock all contributed to a severe economic and political crisis in 1980. The military assumed effective control of the country in May 1980 under General Chun Doo Hwan. Chun formally assumed power in September, promulgated a new constitution in November, and became president in March 1981 when his Democratic Justice Party (DJP) won a majority of seats in the new National Assembly. Chun was succeeded by Roh Tae Woo after the DJP won the December 1987 presidential election. The poor performance in 1979-82 is documented in table 1.1. Output stagnated, actually declining ( - 4.8 percent) during 1980. As domestic savings plunged, the current account deficit mushroomed, financed by massive external borrowing. Korea accumulated over $22 billion of debt during 1979-82, raising its debt stock to 53.5 percent of GNP. Unlike the earlier episodes, the 1979-82 period was an economic crisis, comparable to the crises experienced in many other large debtor countries after the second oil price rise. During 1980-81 the exchange rate was devalued, while the position of monetary and fiscal policies alternated. Korea continued to borrow heavily to maintain investment. By 1982 growth was still low by Korean standards (5-6 percent) and exports stagnated, but inflation and the current account deficits had fallen significantly. The government initiated a more expansionary policy to stimulate growth. As world demand recovered and the terms of trade improved during 1983-84, Korea again underwent a remarkable economic recovery. Growth rates spurted. Savings rose reducing the current account deficit. Authorities responded to the 1985 slowdown in export growth as world demand stagnated with a 6 percent real depreciation and a further 15 percent real depreciation in 1986. By 1986 the economy was booming, inflationary difficulties had been resolved, and there was a substantial trade surplus. In contrast to many of the other large Third World debtor countries currently negotiating rescheduling arrangements with their creditors, Korea not only met all debt service obligations, but began to repay the principal, reducing its debt stock by nearly 5 percent.

158

Susan M. Collins and Won-Am Park

1.2 Overview of the Chapters Four questions emerge from our summary: 1. What caused Korea’s debt crises? 2. How was Korea able to achieve rapid, successful recoveries? 3. What role has external borrowing played in Korean development? 4. What are the lessons for other debtor countries?

Answering these questions involves examining and synthesizing a number of interrelated factors. We focus on the individual factors in the body of our study, bringing them together to address the four questions above in the final chapter. The study is composed of three parts. The first part, chapters 2-5, provides a detailed discussion of Korea’s macroeconomic experience and the role of external debt. The historical background given in chapter 2 is a review of the experience prior to 1962 which set the stage for the impressive economic developments during the 1960s. Chapter 3 gives an overview of external debt, presenting a variety of debt statistics. It highlights the fact that external debt has gone primarily to finance current account deficits and not capital flight. The point is important because it focuses attention on the domestic counterpart to current account imbalance-an excess of domestic investment over savings. It also discusses the process of borrowing in Korea. Chapter 4 examines the three periods of rapid debt accumulation in detail. It reviews the economic and political developments during each cycle of debt accumulation, difficulty, and recovery, discussing the roles of policy and external shocks. Chapter 5 provides a further analysis of the current account deficits which triggered the heavy external borrowing during 1974-77 and 1979-82. Using both accounting decompositions and simulations from a macroeconomic model for Korea, it examines the importance of external shocks in the current account deteriorations. The chronological analysis in part 1 identifies a number of key factors in the experience to be examined individually in the second part of the study (chap. 6- 12). After a brief introduction in chapter 6, we examine economic growth in chapter 7. Chapter 8 analyzes saving behavior and the role of investment and Korea’s five-year plans. Exchange rate, trade, and industrial policy are studied in chapter 9, while in chapter 10 we discuss the important linkages between wages, productivity, and international competitiveness. Chapter 1 I examines monetary and fiscal policy, and income distribution is discussed in chapter 12. Finally, we provide a synthesis of these pieces in part 3 (chap. 13) and discuss the lessons from Korea’s experience.

159

2

KoredChapter 2

Historical Background: Economic Development Prior to 1962

Korea, once known as the "Hermit Kingdom," has been among the fastest growing economies, attracting the attention of both developing and developed countries. The outward-looking strategy, adopted since the establishment of the first five-year development plan in 1962, has led the economy to record impressive growth rates. The economic performance has been cited as the most successful export-led growth among the developing countries. From 1962 to 1984, the Korean economy grew at an average annual rate of 8.4 percent. Within that time, Korea was transformed from a largely agricultural, subsistence economy to a newly industrialized one, despite initially low levels of domestic savings and a lack of natural resources. This chapter briefly describes the sociocultural development, the political history, and the economic factors that brought about the rapid growth in the 1960s.' We divide the period prior to the initiation of the first five-yew economic plan into three subperiods: (1) the colonial period, ending in 1945; (2) the post-liberation and Korean War period from 1945 to 1953; and (3) the construction period after the war from 1954 to 1962. The phases of economic development cannot be completely matched with these subperiods. For example, severe economic dislocations prior to the end of the Korean War could be regarded as starting conditions for later development. Even the later reconstruction process could be viewed as a continuation of the economic rehabilitation program prior to the war, which aimed to overcome the barriers to self-sustaining economic growth with foreign aid. However, political and social changes have a significant impact on the economy, so it is useful to divide the period according to political chronology. Unfortunately, most data in the early period are scattered, and even those available seem to be inconsistent. However, all efforts have been made to gather available information.

2.1 The Colonial Period, 1910-45 Traditionally, China was a suzerain to Korea and, not surprisingly, Korea's trade was dominated by China. But as Japan began to modernize after the Meiji Restoration, it became more powerful and began to exert more influence over Korea. In 1876 Korea could not resist the military pressures from Japan to open trade and signed a trade treaty in Kang-Wha. Other countries such as the United States, Great Britain, China, Germany,

160

Susan M. Collins and Won-Am h r k

Russia, Italy, and France followed, coming to Korea and signing similar treaties on trade and commerce. Korea became a battleground for many foreign powers during the late nineteenth century, but Japan’s decisive victories in the Sino-Japanese War (1884-85) and Russo-Japanese War (1904-5) enabled Japan to ward off other foreign powers. Japan’s dominance over Korea intensified and eventually, in 1910, Japan officially annexed Korea. During the colonial period, 1910-45, the colonial government introduced many new social and economic institutions and built agricultural and social infrastructure. Also, a large number of Koreans gained experience in factories and received education at modem schools. As a result, Korea was able to post high rates of growth in both agricultural and industrial sectors by the standards of that time. In addition, colonial forces improved the industrial structure by increasing the share of manufacturing and mining sectors from a mere 4.1 percent in 1910-20 to 26.4 percent in 1938-40. But this structural change had been catalyzed by Japan’s desire to utilize Korea as its logistical base. In a real sense, the period can be labelled as the industrialization of Korea for Japan. The Korean economy was distorted into becoming an adjunct to Japan, controlled by the totalitarian rule of the colonial government. The pattern of trade was typified by the export of primary goods such as rice and mineral ores to Japan and the import of finished manufactured goods from Japan. Also, the industrialization process in colonial times prevented indigenous development of an entrepreneurial class and hindered the emergence of technical and skilled manpower among the Koreans. The ownership of physical capital was also heavily concentrated among the Japanese.

2.2 Post-Liberation Development and the Korean War, 1945-53 Japan’s direct influence over Korea ended with the Second World War. The liberation on 15 August 1945 divided Korea into two regions, and the U.S. Army Military Government in Korea (USAMGIK) was established to control the political and social turmoil and support the transition of South Korea (henceforth, Korea) into an independent republic. However, severe dislocations were associated both with the departure of the Japanese and with the partition of the peninsula. The resources of the two regions were almost completely complementary, with most of the mineral ores and hydraulic power resources concentrated in the North, and the agricultural resources predominant in the South. The South was deprived of its traditional sources of intermediate and capital goods. USAMGIK implemented measures to revitalize the economy which suffered from a drastic decline in domestic production, severe food shortages, rapid population increase owing to an influx of refugees,

161

KoredChapter 2

primarily from North Korea, and rampant inflation. Some of the measures, particularly redistribution of Japanese-owned land and modern education programs, laid the foundation for further economic development. USAMGIK also initiated the first foreign trade regulations to eliminate smuggling and attempted to control inflation by rationing essential goods and directly suppressing their prices. These economic rehabilitation programs were implemented with massive foreign aid as discussed below. According to the preset political schedule, the first Republic of Korea was established on 15 August 1948, replacing USAMGIK. Its immediate tasks were to reorganize the governmental administration and reduce social unrest and economic instability following the governmental change. For the most part it adopted USAMGIK’s economic policies, in particular by continuing the land reform, managing the grain supply, and organizing trade and foreign exchange policies. There were times, however, when the U.S. and Korean governments had conflicting opinions about the appropriate forms of foreign assistance, exchange rate management, and monetary and fiscal stance. The abrupt outbreak of the Korean War in 1950 halted the rehabilitation of the economy. The war destroyed production facilities and took the lives of many hundreds of thousands of military and civilian people. The three-year war ended in a stalemate in 1953, with the same hostile forces facing each other. The armistice signed between the Soviet and UN forces left Korea in total devastation and extreme poverty, bringing it back to the starting point of economic reconstruction. We turn next to documenting economic growth, inflation, the role of foreign aid, and the major institutional reforms during this period. 2.2.1

Economic Stagnation

Production fell far below the pre-World War I1 level as both the liberation and sudden partition disconnected the economic links between North and South Korea as well as between Korea and Japan. Average rice production for 1945-64 was 9.3 percent below that for 1940-44. In 1948 manufacturing output was about 15 percent of the 1939 level. As shown in table 2.1, the economy started to recover slowly during 1947-49 as a result of increases in industrial production and tungsten exports until production fell sharply in the first two years of the Korean War. As the war gradually stalemated during 1952-53, production recovered to well above the prewar level. Still, net commodity product for 1953 was 27 percent less than that for 1940 in the southern region, as shown in table 2.2. Table 2.3 also shows that the volume of commercial and governmentfinanced trade flows during the 1946-53 period was almost trivial by preliberation standards. Adjusted for the sharp rise in the wholesale price index, the annual trade volume prior to 1953 never exceeded 7 percent of the 1940 level.

162

Susan M. Collins and Won-Am Park

Table 2.1

Average index

Production Indexes, 1946-53 1946

1947

1948

1949

1950

1951

1952

1953

100

115.1

148.8

208.8

131.8

106.2

183.6

239.6

Source: K. S. Kim and Roemer (1979, 29).

Table 2.2

Net Commodity Product, 1940 and 1953 (in billions of constant 1953 won)

Agriculture Forestry Fishery Mining Manufacturing Total commodity product (net)

26.44 2.62 3.38 0.84 4.44 37.72

Percentage Change

1953

I940 (70.2) (6.9) (9.0) (2.2) (11.7) (100.0)

21.20 1.16 0.43 0.53 4.28 27.60

(76.8) (4.2) (1.6) (1.9)

(15.5) (100.0)

- 19.8

-55.7 -87.3 -36.9 -2.7 -26.8

~~~~~~

Source Kim and Roemer (1979, 35)

Table 2.3

Merchandise Exports and Imports for South Korea, 194@-53 (in millions of won) Current Prices

I940 1946 1947 I948 1949 1950 1951 1952 1953

1947 Constant Prices

Exports”

Imports”

Seoul WPI (1947= 100)

0.48 0.05 1.11 7.20 11.27 32.57b 45.91 194.96 398.72

0.77 0.16 2.09 8.86 14.74 5.2Ib 121.83 704.42 2,237.01

55.0 100.0 162.9 222.8 348.0‘ 2,194.Id 4,570.8 5,951.0

0.4

Exports

Imports

118.75 0.09 1.11 4.42 5.06 9.36b 2.09 4.27 6.70

192.50 0.29 2.09 5.44 6.62 1 SOb

5.55 15.41 37.59

Source: BOK, Economic Review 4 (1949):51, and BOK, Annual Economic Review, 1955. See also Kim and Roemer (1979, 31). Nore: Trade volumes for 1940 represent one-half of all Korean exports and imports. The trade data for 1946-53 include only private and government trade, and thus exclude financed imports, transactions with North Korea, and smuggling. ‘Exports and imports were originally valued in won (yen in 1940) on the basis of f.0.b. export or c.i.f. import prices until March 1951 and thereafter on the basis of domestic market prices (tariffs, domestic taxes, and trade margins were subtracted from domestic prices to estimate the price of imports). bImports and exports made through Inchon and Seoul customs offices during March-June 1950 were not included because of missing data due to the war. ‘Average index for June 1950. dAverage index for April-December in F’usan

163

KoredChapter 2

2.2.2 Foreign Aid and Foreign Exchange Regimes Dislocation, poverty, and extreme shortages of many commodities made the survival of South Korea as an independent political entity after the liberation infeasible without massive foreign aid. The United States provided most of the foreign aid mainly in the form of food, clothing, medicines, fuel, and fertilizers to meet emergency needs and to raise rice production toward self-sufficiency. A small amount was spent on reconstruction works. Table 2.4 lists the foreign aid agencies and their contributions. The Government Appropriations for Relief in Occupied Areas (GARIOA) aid program started early in the USAMGIK period, and it provided nearly half of the total aid during 1945-53. The GARIOA program was followed by the Economic Cooperation Administration (ECA) program, which emphasized capital development and restrictive economic policies. Although there were often donor-recipient conflicts over the primary role for foreign aid, pressing demands from the U.S. for economic (especially price) stabilization led to a significant reduction in the money supply before the outbreak of the Korean War. The war reshuffled the entire recovery program, with more emphasis on relief than development. The Civil Relief in Korea (CRIK) program, mainly financed by the U.S., provided almost all assistance during 1952-53, followed by the United Nations Korean Reconstruction Agency (UNKRA) which was multinationally based. These programs financed nearly 70 percent of total imports during 1951-53 as commercial export earnings for imports were almost negligible. Since the export performance was so poor, imports were constrained with trade regulations. Indeed, USAMGIK had initiated strong foreign trade regulations that restricted exports and imports only to licensed traders. Exports were linked to imports by a barter trading system, and imports faced a 10 percent across-the-board tariff. Also, quantitative restrictions were pervasive, and the overvalued exchange system was maintained. Economic Assistance to Korea, 194553 (in millions of U.S. dollars)

Table 2.4

GARIOA ~~~~

~

1945 1946 I947 1948 1949 1950 1951 I952 I953

Total

4.9 49.9 175.4 179.6 92.7

502.5

ECA

CRIK

~~

23.8 49.3 32.0 3.8 0.2 109.1

Source: BOK, Economic Statistics Yearbook. various years.

9.4 74.4 155.2 158.8 397.8

Total

UNKRA ~~

~~

4.9 49.9 175.4 179.6 116.5 58.7

0. I 2.0 29.6 31.7

106.5

161.0 188.4 1,040.9

164

Susan M. Collins and Won-Am Park

The overvaluation of Korean currency had been the focus of conflicts between the U . S . and Korea. Currency overvaluation discouraged exports and, as exports could not match import demand, excess demand for foreign exchange produced a black market in which rates were substantially higher than the official ones. Because foreign exchange earnings from sales to U . S . and UN forces were substantial, the Korean government intended to overvalue its currency to gain more foreign exchange for a given amount of won. The same intent was apparent in the giving of a large won “advance” to the UN Command, the repayment of which in U.S. dollars would occur in due time. Thus, the Korean government was more concerned with obtaining foreign exchange through won sales to the U.S. and the UN than with promoting exports, because of the belief that devaluation would be ineffective in improving the balance. It was thought that the barter system, with some export incentives, tariffs, and strict quantity restriction on imports could effectively reduce the trade deficit. These trade and payment regimes prevailed for the rest of the 1950s without major changes. 2.2.3

War-related Inflation

Very high inflation rates first appeared as strict price controls were removed after the liberation. The price index jumped twenty-five-fold between July and August of 1945, although the issue of notes stayed moderate during the period. Rapid inflation continued in 1946 with a 366 percent increase in prices. However, inflation rates declined steadily during 1947-49, as economic stabilization measures were implemented. Monetary expansion during the Korean War gave a second momentum for hyperinflation. Prices shot up drastically after the outbreak of the war and surged again in the next year with an expansion in government military expenditure. In 1951 the annual inflation rate reached 530 percent. However, as table 2.5 shows, inflation had been reduced to 30 percent by 1953. During 1945-53 inflation rates were high compared with those during the colonial period, reflecting the expansionary effects of banks’ lending to the government for economic recovery and war financing. Also important was the role of aid goods and foreign monies such as U.S. dollars and U.S. military payment certificates (MPCs) that were circulated in the unorganized financial markets in exchange for goods. The hyperinflationary process invoked a flight from money into goods, which in turn brought about a flow of commodities and foreign monies through the unorganized financial markets. While a detailed analysis of this early inflationary episode is beyond the scope of this chapter, it does seem that the provision of massive aid immediately after the war helped to arrest the hyperinflationary bias. 2.2.4

Educational Expansion and Land Reform

Two institutional changes, educational expansion and land reform, seem to have been critical for the subsequent success of export-led development. Both changes were motivated by egalitarianism.

KoreafChapter 2

165

lsble 2.5

Money Supply and Inllation, 194653 Bank of Chosun Notes Issued (in million won”)

Seoul WPI (1947= 100)

Inflation over Previous Period (9%)

3.14 4.70 4.98 8.50 8.76

0.60 0.69 17.0b 1 1.2b ll.gb

15 2,364 - 34 5

December 1944 July 1945 August 15, 1945 September 8, 1945 December 3 I, 1945

1946 I947 1948 1949 1950 1951 1952 1953

Money Supply Index (1947= 100)

Seoul WPI (I947= 100)

Annual Inflation (%)

49.I 100.0 166.2 264.7 458.3 1.588.4 3.148.2 6,762.6

55.0 100.0 162.9 222.8 348.0‘ 2.194.Id 4.570.8 5,951.0

366 82 63 37 56 530 108 30

Source: BOK, Annual Economic Review. 1947. 1955, and Bloomfield and Jenson (1951).

‘The amount of bank notes issued in old won denomination is converted into the presently used won denomination. bMonthly average. ‘Average index for June 1950. dAverage index for April-December in F’usan.

The high aspirations for educational attainment typical of Koreans stems from their cultural traditions. The Japanese had established facilities for primary education, however, these were used to “Japanize” the Koreans. All instruction was in Japanese prior to 1945. Furthermore, the opportunities for higher education were extremely limited. The liberation opened the way to the higher education that was a prerequisite for social advancement. An extensive public education system (with instruction in Korean) was initiated during the USAMGIK period. The initial focus on elementary levels was followed by a rapid expansion of the school population in higher education. Another social program that modernized Korea on a egalitarian basis was land reform. The measure was first implemented by USAMGIK in 1947 by distributing farmland previously owned by the Japanese to the tenants. The new Korean government continued the process in 1949, so that most of the land had either been redistributed or sold privately by landlords who had anticipated the redistribution. As a result, the number of farmers owning all the land they farmed increased dramatically, and the number of tenants fell from 48.9 percent of the total in 1945 to just 5-7 percent by 1952. The reform effectively reduced the concentration of land ownership, and it led to further-reaching social consequences: it ushered in the decline of the

166

Susan M. Collins and Won-Am Park

landlord class which had formed the backbone of the traditional society for centuries. In sum, the two institutional changes put Korea in a far better position to achieve growth with relative equity than many other developing countries. Equity and income distribution are discussed further in chapter 12.

2.3 The Postwar Reconstruction Period, 1954-62 When the Korean War ended with the armistice in 1953, all efforts were made to reconstruct the infrastructure and industrial facilities of Korea. The government endeavored to curb inflationary pressures, viewing inflation as a stumbling block to the effective implementation of monetary, fiscal, and trade policies. However, continuing hostility between North and South Korea necessitated big military spending that otherwise would have been used for reconstruction and stabilization. Thus, reconstruction, stabilization, and strong defense were the main tasks of the government after the war. No doubt the government was also concerned about economic growth, but it remained of secondary importance. 2.3.1 Slow Economic Growth with Import Substitution With the aid of the reconstruction and stabilization programs of the aid agencies, the economy grew at 4.1 percent during the period from 1953-55 to 1960-62, and the industrial sector grew over 10 percent per year (table 2.6), with its share of GNP increasing from 13.6 to 18 percent. Table 2.6

Economic Growth, 1953-62 (in percentages) Industry

Year

Primary

Total

1954 1955 1956 1957 1958 1959 1960 1961 1962

6.7 2.7 -5.8 9.4 6. I -0.9 -0.5 11.8 -5.0

20.0 17.1 13.3 11.8 8.1 11.3 6.7 4.6 14.0

2.0

10.8

45.2 39.9

13.6 18.2

Average 1953-55 to 1960-62 Share of GNP 1953-55 1960-62

(Manufacturing)

Services 2.5 5.7 4.0 5.8 3.5 7.5 2.8

GNP

(Commodity Export) 36.0)

( - 20.3)

8.9

5.5 5.4 0.4 1.7 5.2 3.9 1.9 4.8 3. I

(13.2)

4.3

4. I

(7.6)

(10.5) (13.8)

40.0 41.0

100.0 100.0

- 1.1

Source: BOK, National Income in Korea, 1975. Nofe: Percentages based on constant I970 market prices

(-

( - 27.9)

(42.1)

(1.5) (11.3) (54.1) (22.7) (30.0)

167

KoredChapter 2

Industry grew at a much faster rate than agriculture because industry began from such a low base after the devastation of the war. Still, the major source of industrial growth in the 1950s was the government’s policy of import substitution. A study by Frank, Kim, and Westphal (1975) showed that the direct contribution of import substitution to the growth of the manufacturing sector was far greater than that of export expansion and that the shares of exports and imports in GDP were appreciably below Chenery’s estimates of the structural “norm” of countries at similar stages of development and capital flows (Chenery, Robinson, and Syrquin 1986). 2.3.2 Aid, Trade, and Payments Foreign aid and the trade and payments regimes were highly interdependent. As the value of commodity imports in current U.S. dollars was more than eleven times the value of exports on average from 1953 to 1962, foreign aid financed commodity imports. The major sources of foreign exchange were aid and the local currency expenditures of the UN forces. When aid imports were sold in the domestic market, the sales receipts were deposited in “counterpart funds.” The Korean government could then use those funds only for purposes mutually agreed upon with the American government. Therefore, the impact of aid on Korean economic policies, either through the commodity imports or the counterpart funds, was tremendous because the inflow of commodity imports was crucial to economic reconstruction and price stabilization and because the government deficit was largely financed by counterpart funds. In this situation conflicts were bound to arise between the aid donor and recipient about the objectives of aid flows. The frictions centered on whether the primary objective was aid to finance economic reconstruction or to attain self-sufficiency. Table 2.7 quantifies the significance of foreign assistance. Unfortunately, the foreign aid figures do not coincide according to sources and coverage of foreign aid. In addition to data problems, the overvaluation of the won leads to an underestimation of aid and imports as a fraction of GNP. Aid, however, seems to constitute over 70 percent of imports during 1953-60. The trade and payments regime continued to consist of the exchange controls and import licensing that had begun in 1949, however, the impact of multiple exchange rates, tariffs, and quantitative controls intensified. Some partial steps toward export promotion policies were undertaken in 1954 and 1955. The government introduced an export-import linkage system which stipulated that some essential materials could be imported only with the foreign exchange earned through exports. Special exchange rates were applied to specific commodity exports whereby the exchange rates were set at 10 to 50 percent above the legal rate to compensate for the loss of profit in export production. In addition, the exporters of particular commodities who were losing international competitiveness due to high domestic production

168

Susan M. Collins and Won-Am Park

Table 2.7

U.S. bilateral CRlK UNKRA Total Total imports Aid as a percentage of Imports Imports as a percentage of GNP (current prices)

Foreign Aid, 195340 (in millions of U.S. dollars) 1953

1954

1955

1956

1957

1958

1959

1960

12.8 158.8 29.6

108.4 50.2 21.3

205.8 8.7 22.2

271.0 0.3 22.4

368.8

313.6

219.7

245.2

-

-

-

-

_

_

-

-

_

_

-

14.1

1.7

2.5

-

-

0.2

-

201.2

179.9

236.7

293.7

382.9

321.3

222.2

245.4

345.4

243.3

341.4

386.1

442.1

378.2

303.8

343.5

58.3

73.9

69.3

76.1

86.6

84.9

73.1

71.4

12.9

7.3

9.8

13.1

12.0

10.7

10.1

12.6

Source: BOK, Economic Srarisrrcs Yearbook. 1960. 1974, and IMF. lnrernarional financial Sraristics. May 1976. Nore: Imports as a percentage of GNP are calculated including imports of both goods and services. Dashes indicate that data were not available.

costs were authorized to import high profit commodities such as sugar, newspaper, and wool yam. Finally, some direct subsidies were given to exporters of commodities that were experiencing international price declines. However, these export promotion policies were relatively ineffective because they were insufficient to compensate for the growing overvaluation of the won. Industrial products remained uncompetitive, and there seems to have been little increase in incentives to produce manufactured goods for export. Exports were mainly primary commodities such as mineral ores and agricultural and fishery products. The trade balance fluctuated with changes in foreign aid and grain harvests. Still, it is important to note that most manufacturing sectors had recovered beyond the prewar level by 1956 and the structure of exports had begun to shift toward light manufacturing. The import structure changed more visibly. Although the import volume fluctuated widely according to the foreign aid amount, the portion of imports such as food, crude materials, and chemicals rose, while the share of mineral fuels and manufactured goods declined. This indicates clearly the direction of import substitution in the 1950s. Table 2.7 shows that foreign aid peaked in 1957. The decline in aid flows (which had been preannounced) brought about a number of policy changes. In particular, the reduced aid set the stage for shifts toward export promotion and toward reliance on foreign borrowing and domestic savings to finance investment and growth. These issues are discussed in detail in subsequent chapters.

2.3.3 Chronic Inflation The inflation stemming from the high rate of monetary expansion maintained its fast pace throughout the postwar period. During 1953-62

169

KoredChapter 2

annual inflation averaged around 17-20 percent (table 2.8). However, inflation rates varied from period to period, mainly a result of fluctuating food prices. In the face of the chronically high inflation, pressure for a financial stabilization program came from U.S. aid officials in the process of negotiating with the Korean government about the level of aid and its objectives. Beginning in the second half of 1957, the restrictive program set quarterly ceilings for monetary expansion and enforced strict budgetary reduction. Consequently, inflation declined to very low rates of 2-4 percent, although decreases in food prices resulting from bumper crops in 1957 and 1958 also helped. The declining inflation during the implementation of the program was accompanied by an economic slowdown (see table 2.6), similar to the stabilization experiences of other developing countries. The austerity measures together with the declining aid flows also explain the decline in import volumes (see table 2.7).

2.4 The Korean Economy in 1960: An Overview We end this chapter by summarizing Korea’s economic position at the beginning of the 1960s, on the eve of the industrial transformation and impressive growth. The trade regime was characterized by import substitution, including the familiar complex set of multiple exchange rates, import licensing, and overvaluation. The economy was heavily reliant on foreign aid, with exports accounting for only 3 percent of output, while imports amounted to over 10 percent. While inflation had been brought under control, output growth rates were stagnant. Furthermore, manufacturing

Table 2.8

Inflation Rates, 195342 Wholesale Price of Foods

Seoul CPI

Period

GNP Deflator

WPI

1954 1955 1956 1957 1958 1959

31.6 65.3 30.6 20.4 -0.5 2.6 9.5 15. I 13.9

28.0 81.9 31.4 16.3 -6.5 2.6 10.7 13.2 9.4

- 19.8

12.8

36.0 69.6 22.5 23.1 -3.1 4.3 8.3 8.0 6.5

36.0 3.8 19.6

37.4 2.0 18.7

36.8 -4.9 16.7

36.6 3.1 17.9

1960

1961 1962 1953-57 1957-60 1953-62

Source: BOK, Economic Sratistics Yearbook, various issues

136.9 60.4 15.0 - 18.0 - 12.4 19.6 18.4

170

Susan M. Collins and Won-Am Park

accounted for just 11 percent of GNP, with over 45 percent of output concentrated in the primary sector. The next chapters examine Korea’s transformation into a rapidly growing economic miracle.

3

An Overview of Korea’s External Debt

In 1962 Korea’s external debt stood at $58 million, only 2.5 percent of GNP. By 1982 Korea had accumulated over $37 billion in external debt, or 52.7 percent of GNP, ranking it fourth in the list of debtor countries. After an impressive turnaround, Korea began to reduce its external debt in 1986. This chapter presents and discusses a number of debt statistics. The objective is to identify the key trends to be examined in detail in subsequent chapters. In addition to the tables in the text, we refer extensively to debt tables from the Data Appendix.

3.1 Korea’s External Debt Table 3.1 traces the accumulation of Korean debt from 1961 to 1986. As shown, the debt stock rose steadily until 1985, with an average annual increase of $2 billion. Over 75 percent of the increase between 1962 and 1982 occurred during three periods of rapid debt accumulation. From 1966 to 1969 the debt rose from $0.35 to $1.8 billion, or from 9.6 percent to 27.2 percent of GNP. A second jump followed the 1973 increase in oil prices. The debt stock increased by $4.2 billion from 1973 to 1975, pushing the debt/GNP ratio to 40.6 percent. As debt accumulation slowed during the recovery period from 1975 to 1978, the debt/GNP ratio fell back to 28.6 percent. The third period of rapid accumulation began in 1979. Over the next three years, the debt stock rose from $14.9 to $37.3 billion, while the debt/GNP ratio jumped to 52.7 percent. During the subsequent recovery (1982-83, the debtXNP ratio averaged 53.8 percent before falling below 47 percent in 1986. The three periods of accumulation will be examined in chapters 4 and 5. We return to the recent debt decumulation in chapter 13. Table A2.1 in the Data Appendix provides a more detailed breakdown of external debt by maturity and borrower. It identifies two other aspects of Korea’s external debt history. It shows that there have been considerable shifts in the term structure of the debt. There have also been shifts in the composition of the debt between the public sector and the bank and nonbank private sector.

171

KoredChapter 3

Table 3.1 Debt Total foreign debt Foreign direct investment ForeigndebUGNP Foreign debt plus direct investmenUGNP Debt service ratio' Debt

Korea's External Debt, 1961-86 (in millions of U.S. dollars) 1961

1962

1963

1964

1965

1966

1967

1968

1969

83

89

157

177

206

392

645

1,199

1.800

3.9

3.8

3 5.8

6 6.2

16 6.9

21 10.7

34 15.1

49 22.9

56 21.2

3.9 0.6

3.9 0.8

5.9 1.0

6.4 2.6

7.4 5.0

11.3 3.2

15.9 5.4

23.9 5.4

28.0 8.6

-

1970

1

1971

1972

1973

1974

1975

1976

1977

I978

Total foreign debt 2,245 2,922 3,589 4,257 5.933 8,443 10,520 12,649 14,823 Foreign direct investment 81 117 175 329 486 549 650 741 830 Foreign debVGNP 28.7 31.2 34.0 31.5 32.0 40.5 36.7 33.8 28.5 Foreign debt plus direct investmenUGNP 29.7 32.4 35.6 34.0 34.6 43.1 38.9 35.8 30.1 Debt service ratio' 18.5 21.0 18.7 14.8 14.4 14.4 12.1 11.1 13.9 Debt

1979

1980

1981

1982

1983

1984

1985

1986

Total foreign debt 20,287 27,170 32.433 37,083 40,378 43,053 46,762 44,510 Foreign direct investment 866 873 975 1,044 1,112 1,222 1.456 1,891 Foreign debUGNP 32.5 45.0 49.0 53.5 53.1 52.3 56.3 46.8 Foreign debt plus direct investmenUGNP 33.9 46.5 50.4 55.0 54.6 53.7 58.0 48.8 Debt service ratio' 16.3 18.5 20.1 20.6 18.8 20.4 21.7 22.7 'Includes interest on short-term debt.

In the table, debt is divided into three maturities. Long term refers to over three-year maturity, while medium and short term refer to one-to-three year and under one year, respectively. In 1962 medium-term trade credits amounted to over 80 percent of Korea's total debt. The next decade saw a consistent decline in the share of medium-term debt to 3.8 percent by 1971. Although the share of medium-term debt remained more or less constant during 1971-85, there have been significant changes in the relative importance of short- and long-term debt. Short-term debt increased dramatically during each period of rapid accumulation and fell, after a lag, during the subsequent recovery. The share of short-term debt jumped from 16.4 percent in 1973 to 28.5 percent in 1975. In 1978 it declined to 21.2 percent. By 1980 the share had soared to 34.5 percent. Since then, it has been substantially reduced to less than 21 percent in 1986. There have also been shifts in the distribution of the debt between the public sector, the nonbank private sector, and financial institutions. In

Susan M. Collins and Won-Am Park

172

interpreting these numbers, it is important to bear in mind that the typical distinctions between public, private, and bank debt are somewhat misleading in Korea. First, external debt is overseen by the Ministry of Finance (MOF), and all borrowing requires prior approval. In effect, all loans are “publicly guaranteed” in the sense that they are ultimately backed by the Bank of Korea. Second, bank debt comprises loans to the banking system. These are treated as below-the-line, or accommodating flows, in Korean balanceof-payments statistics. The funds are then lent out to the private sector. During 1961-67 all long-term Korean debt was either public or private borrowing, except for some usage of IMF facilities beginning in 1965. From 1966 to 1971 private debt substantially exceeded public debt. During 1971-78 the two remained of comparable magnitudes. Since 1978 the importance of public debt has risen substantially relative to private debt. However, private and bank debts together have consistently exceeded public borrowing. Public and private borrowing continued to constitute 90 percent of the total until 1978. These items had fallen to 70 percent of the total by 1982 and to only 52 percent by 1985. From 1978 to 1982 the declining importance of public and private debt is attributable to the growth in bank loans, foreign bank “A” accounts,’ and IMF facilities. After 1982 bonds were increased dramatically.

3.2 Korean Corporations with Foreign Branches Borrowing by Korean enterprises with branches abroad is not included in external debt statistics. However, these figures are monitored by the MOF and are subject to the regulations of the Foreign Exchange Control Act. As shown in table 3.2, this borrowing doubled between 1979 and 1982, reaching $5.4 billion. Since 1984 this borrowing by foreign branches has exceeded the total of private long-term loans.

Table 3.2

Foreign Financing of Korean Companies with Branches Abroad, 1979-86 (in millions of U.S. dollars)

Total Trade companies Construction companies Other Memo Foreign finance as a percentage of long term pnvate loans Source:

Ministry of Finance

1979

1980

1981

1982

1983

1984

1985

1986

2,447 1,404 964 79

3.712 1.863 1.758 91

4,463 1,711 2,649 103

5,377 1.791 3,456 130

5.710 1.885 3.672 150

5.976 2,167 3,631

I78

6.076 2,175 3,560 341

5,619 2.534 2,722 363

044

060

0 69

0 85

0 93

I01

I06

104

173

KoredChapter 3

3.3 Indicators of Debt Burden In absolute terms, Korea is one of the most heavily indebted countries. Among the twenty largest debtor countries in 1983, Korea was fourth in terms of the level of its gross debt. However, the gross volume of debt can be a misleading indicator of the real burden of external borrowing. In 1983 Korea was eleventh in terms of its debt/GNP ratio and fifteenth in terms of the ratio of debt service to exports.2 Table 3.3 presents a number of measures of the burden of Korea’s debt. The debt/GNP ratio is given in the first column. The second and third columns show the ratios of total debt service to GNP and to exports. The final column shows the ratio of foreign exchange reserves to external debt. From the table it can be seen that the ratio of debt to GNP jumped in 1975 and again in 1980 after an intermediate period of decline. The ratio continued to rise after 1980. The table also shows that, despite rapid debt accumulation over the period from 1970 to 1982, service payments fell as a proportion of foreign exchange earnings. Although debt rose from 34 to 53 percent of GNP, rapid growth of exports has meant that the share of export revenues needed to service the debt has risen much more slowly. Also, receipts from invisibles have grown very quickly since the mid-1970s. By 1982 service payments accounted for 21 percent of export receipts, but for less than 16 percent of total current revenues. Thus, the rapid growth of foreign exchange earnings plays a critical role in Korea’s experience with

Table 3.3

Debt Burden Indicators, 1966-86 (in percentages)

Year

DebVGNP

ServicelGNP

ServicelExports

ReservesfDebl

1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

10.7 15.1 22.9 27.2 28.7 31.2 34.0 31.5 32.0 40.5 36.7 33.8 28.5 32.5 45.0 49.0 53.5 53.1 52.3 56.3 46.8

0.4 0.8 0.9 I .5 3.3 3.6 3.9 4.5 3.8 4.1 4.0 4.2 4.6 5.1 6.9 8.2 8.2 7.6 8.3 8.6 10.0

3.2 5.4 8.6 18.5 21.0 18.7 14.8 14.4 14.4 12.1 11.1 13.9 16.3 16.0 18.5 20.1 20.6 18.8 20.4 21.7 22.7

60.6 54.3 32.6

30.7 26.I 18.4

20.6 25.7 17.8 18.3 28.1

34.0 33.2 27.8

20 21.2 18.7 16.8 17.7 16.5 17.9

174

Susan M. Collins and Won-Am Park

external debt. Table 3.3 also shows that Korea accumulated foreign exchange reserves relative to debt during the recovery from the first and second crises, but did not replenish the reserve stock in the first few years after the most recent crisis. How does Korea compare to other debtor countries? Table 3.4 provides comparison debvexport and debt/GNP figures for three groups of countries. In 1980 Korea’s debt to export ratio was high relative to both the group of fifteen heavily indebted countries and the group which has not experienced debt servicing difficulties. The debt to output ratio was considerably higher than even the average for countries that did experience servicing difficulties. By 1986 Korea’s relative position had improved considerably. Its debt to export ratio had declined by 12 percent. In contrast, the ratio had risen by 44 percent for countries without difficulties and 164 percent for those with difficulties. Korea’s debt to output ratio had fallen below the ratio for the countries having difficulties, but remained significantly higher than the ratio for the countries without difficulties. It is useful to consider all of these indicators because none is an ideal measure of the debt burden. The debt to GNP measure relates the total amount owed abroad to total domestic output, but does not indicate a country’s ability to transfer domestic into foreign resources so as to pay external debts. The debt to export ratio does focus on access to foreign exchange earnings. However, this indicator is also problematic because countries would differ in their foreign exchange requirements even if they had no external debts. In particular, Korea relies heavily on imported intermediates and raw materials for domestic production. The import requirements for exportables and for investment were 35 percent and 42 percent, respectively, in 1980. Over 1980-83 imports averaged 39.8 percent of GNP. In contrast, imports averaged 18.7 percent of output for the ten principal Baker

Table 3.4

Cross-country Comparisons of Debt Burden 1980

1982

1984

I986

A. Long- and Short-Tern External Debt Relative to Exports of Goods and Services

Korea 135 heavily indebted countries Countries with recent debt servicing problems Countries without recent debt servicing problems

120.3 109.3 151.2 79.1

130.8 178.7 241.5 92.8

127.9 178.9 247.2 96.3

288.1 302.4 114.0

45.1 30.8 33.6 20.5

53.5 41.7 45.5 24.9

52.2 46.8 51.1 27.3

46.8 48.4 54.8 32.5

106.1

B. Long- and Short-Term External Debt Relative to GDP Korea 15 heavily indebted countries Countries with recent debt servicing problems Countries without recent debt servicing problems

~~

Source. IMF, World Economrc Outlook. April 1987, for all countries except Korea (uses new SNA)

175

KoreaChapter 3

countries. The figure was 22-25 percent for Chile, Ecuador, Nigeria, Peru, the Philippines, and Venezuela and just 9-12 percent for Argentina, Brazil, and Mexico.

3.4 Usage of External Debt The balance of payments accounts imply that increases in the stock of gross external debt must be equal to the current account deficit plus acquisitions of official foreign exchange reserves plus capital inflows. A useful way to write the identity is given below.

A Gross External Debt

=

Current Account Deficit

+

A Official Reserves

+

Short-Term Long-Term Private - and Direct Capital Capital outflows Inflows

The equation points out that current account deficits, reserve accumulation, and short-term capital flows (capital flight if it leaves the country) must be financed either by long-term capital movements and direct foreign investment, or by accumulation of external debt. Exactly how foreign borrowing has been used has an important bearing on the ease with which a country can repay its debts. There are two key issues. The first is that debt which financed private capital outflows can be extremely difficult to repay because it does not increase domestic resources. Instead, a few private citizens hold assets abroad-the counterpart to the country’s external debt. To repay its liabilities, the government must mobilize and transfer domestic resources to the rest of the world. This is typically accomplished through subsidy cuts and tax increases to improve the government budget, and through real exchange rate depreciation and real wage reductions to increase competitiveness and to shift resources into the production of tradable goods. The transfer may well lead to a deterioration in the standard of living and/or in the distribution of income. The second issue concerns the sources of the current account deficit. National income accounts imply that a current account deficit is the foreign savings counterpart to the difference between domestic savings and investment. Countries can run large current account deficits because of high investments which will pay off through increased future productive capacity. They can also run large deficits with low investment when domestic savings are small, perhaps because of government budget problems or because of a spurt in imports of consumer goods. To the extent that external borrowing goes to finance a current account deficit which reflects strong investments (particularly in the traded goods sectors), a country should have little difficulty in repaying its obligations. Although there may be problems of liquidity in the short to medium term, resources should eventually become available to transfer abroad. However,

176

Susan M. Collins and Won-Am Park

countries that borrow abroad to substitute for low private and/or government savings are likely to face the same difficulties of repayment as those that borrowed to finance capital flight. Korea’s use of debt distinguishes it from many of the other large debtor countries. For example, Dornbusch (1985b) points out that external debt accumulated in Argentina during 1978-82 went primarily to finance capital flight. In Brazil the debt financed current account deficits, but these reflected government dissavings and not high investments. The public sector had not adequately adjusted to the severe external shocks. Table 3.5 breaks down the use of external debt for Korea. The seven time periods include the three periods of rapid debt accumulation and the subsequent recoveries, as well as the 1986 developments. During each of the three accumulation periods the current account deficit accounts for the bulk of the increase. The current account deficit, together with reserve accumulation and errors and omissions from the balance of payments, accounted for at least 78 percent of the debt in all six period^.^ In 1983-85 much of the large discrepancy was due to an increase in exports on credit. Finally, about half of the huge 1986 current account surplus went to reducing external debt. Debt accumulation has not gone to finance capital flight. Except for the increased reserves (and more recently the increases in other assets), the primary usage of external borrowing has been to finance the imbalance between investment and domestic savings, In fact, Korea has maintained consistently high and rising investment rates, and domestic saving rates rose substantially during 1965-85. Unlike for either Argentina or Brazil, investment has played a central role in Korea’s debt accumulation. Chapter 8 on savings and investment behavior explores these issues in more detail.

3.5 The Cost of Foreign Borrowing Why was the private sector willing to borrow so much, so rapidly? Table 3.6 gives a variety of interest rates, in addition to inflation and exchange rate Table 3.5

Use of External Debt, 1966-86 (in billions of U.S. dollars)

1966-69 Debt Current account deficit Foreign exchange accumulation Errors and omissions ( - ) Direct foreign investment ( - ) Discrepancy

1970-73

1974-75

1976-78

1979-82

1983-85

1986

1,594

245

4,186

6,380

22,260

9.679

-2,252

1,285

2,150

3,910

1,387

16,768

3,866

-4,617

41 I

484

507

3.396

2.047

796

207

I

- 57

94

585

2.406

2,716

544

-40 - 63

-281 161

-232 - 93

-308 1,320

-430 1.469

-522 2,893

- 477 -2.091

177

KoredChapter 3

Table 3.6

The Cost of Foreign Capital, 1966-86 (annual average percentages)

Item

1966-70

I . Domestic bank lending ratea Curb market interest rate 2. Foreign interest rate' 3. Exchange rate depreciation' 4. Domestic inflation rate (GDP deflator) 5. Interest rate differential between home and foreign markets [ ( I ) - (2) - (3)l 6. Real private cost of borrowing abroad [(2) + (3) - (4)]

1971-75

1976-80

24.4 54.2 7.2 3.1 15.4

17.4 40.1 7.9 9.3 18.8

18.0 41.4 9.5 4.7 20.9

14.1

0.2

3.8

-5.1

-1.6

-6.7

1981-83 13.8 30.5 13.0 8.5 8.5

-7.7 13.0

1984-85 10.0 24.4 9.5 5.9 4.0

-5.4 11.4

1986 10.0 23.2 6.7 1.3 1.4

2.0 6.6

Source: BOK, Monthly Bulletin, various issues, and IMF, International Financial Sratistics. various issues.

*Discounts on bills of Deposit Money Banks. 'Ninety-day Euro-dollar rate.

'Period average.

depreciation. The figures show that there were strong reasons to borrow abroad instead of domestically, even for those firms that had access to credit from the official banking system. The incentives were even stronger for those firms that could only borrow domestically through the unorganized (curb) markets. (Korean financial markets are discussed in chapter 11.) The interest differential between domestic and foreign borrowing was a full 14 percent during 1966-70, the first period of rapid debt accumulation. It fell to about 1 percent during 1971-80, however, the real cost of foreign borrowing remained negative as domestic inflation outstripped depreciation of the won. During 1981-85 the incentives reversed dramatically. The interest differential between domestic and foreign borrowing had turned negative. The slowdown in domestic inflation, combined with a substantial depreciation, made the real cost to foreign borrowing jump from -6.7 percent in 1976-80 to 13.0 percent in 1981-83 and then to 11.4 percent in 1984-85. During 1986 the exchange rate began to appreciate, while domestic inflation and foreign interest rates continued to fall. Despite little change in domestic nominal rates, borrowing abroad once again became relatively less expensive than borrowing from domestic banks.

3.6 Foreign Aid and Concessional Lending We have seen that virtually all of Korean imports and gross investment was financed by foreign aid during the 1950s. However, the importance of foreign aid declined precipitously during the 1960s, becoming a negligible source of funding by the mid-1970s. Table 3.7 summarizes aid flows to Korea during 1948-83. (A more detailed breakdown of foreign aid is given in the Data Appendix.) As shown, the United States is by far the largest donor. The flows rose to a high of 16

178

Table 3.7

Susan M. Collins and Won-Am Park

Average Annual Aid Received, 1948-83 (in millions of U.S. dollars) Period

Total

U.S.

1948-55

150.9 299.7 185.8

84.4 290.3 185.8 91.2

1956-60 1961-65 1966-71 1972-77

1978-83

91.2 2.0 0.2

2.0 0.2

Source: BOK, Economics Statistics Yearbook. 1972, 1984.

percent of GNP in 1957, averaged 8-9 percent during 1959-62, 2 percent during 1966-68, 1 percent during 1969-71, and have been negligible since 1972. The majority of the flows were nonproject assistance used to finance imports of raw materials and capital goods. The magnitude of these flows implied that the United States had a critical influence over Korean investment decisions during this p e r i ~ d . ~ After the overthrow of Syngman Rhee in 1960, Korean policies increasingly encouraged foreign borrowing from private sources. By 1967 foreign loans and foreign direct investment each played a more important role than foreign aid. However, these figures underestimate the effective amount of aid because some of Korea’s public loans during the 1960s and 1970s were on concessional terms and the grant element of these loans has not been included. As Krueger (1982, 154) points out, almost all public borrowing between 1966 and 1969 came from either the United States or Japan, so that one sensible correction focuses on public debt from these sources. In fact, PL 480 and development loans came almost exclusively from the United States during 1961-75. The magnitude of these loans increased as other U.S. aid tapered off, implying that the total remained approximately constant in nominal terms through 1972. In 1965 the Japanese Settlement (treated as a reparations settlement and not as aid by Korean authorities) called for $500 million in grants and public loans to be disbursed over the next decade. An alternative approach to estimating the grant element in Korea’s foreign loans is to examine the terms of the borrowing. Table 3.8 shows interest rates and terms of repayment for (committed) foreign capital during 1959-74. There is a marked difference between public and commercial loans. Taking the weighted average figures, public loans enjoyed a 3 percent reduction in interest rates, a 4.7 year increase in the grace period, and a repayment period almost 16 years longer than that for commercial loans. Strictly comparable data is not available for the more recent period. To provide an updated indication of the amount of Korea’s concessional lending, table 3.9 computes the average interest rates for public, commer-

179

KoredChapter 3

Table 3.8

Foreign Loans by Interest Rates and Terms of Repayment, 1959-74 (commitment basis, weiahted averages)

~

Commercial

Public

7.1 2.5 10.1 4,166.5

4.1 7.2 26.0 2,764.4

~~

Interest rate (7%) Grace period (years) Repayment period (years) Total (millions of U.S. $) Source: Krueger (1982, 156-57, table 45)

Average Interest Rates on Long-term Loans, 1972-86

Table 3.9 Year

Public

Commercial

Bank

LIBOR"

1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

6.0 4.7 3.3 4.5 5.3 5.7 6.6 6.5 7. I 6.8 7.8 6.2 6.6 5.8 6.9

7. I 8.5 9.4 9.3 8.4 9.2 10.0 10.0 12.0 12.7 12.4 10.0 10.8 9.5 10.1

8.5 11.1 23.4 16.2 12.0 15.7 17.8 27.9 16.9 19.8 15.2 11.4 11.3 8.8 8.4

5.4 9.4 10.9 7.0 5.6 6.0 8.9 12.1 14.2 16.9 13.3 9.7 10.9 8.4 6.9

Source: EPB, Major Statistics: MOF, Fiscal and Banking Statisrics; and IMF, International Financial Statistics. various issues. Note: Average interest rates are computed as total interest payments divided by debt outstanding at the end of the preceding year.

'LIBOR = London interbank offer rate for dollar deposits.

cial, and bank long-term loans from 1972-86.5 The figures show that interest rates on bank loans are higher and more variable than rates on other private sector loans and that public loans have enjoyed the lowest rates. This component includes the public long-term debt originally lent at concessionary terms. Also, the majority of the fixed interest loans are public loans, while commercial and bank borrowing is typically at variable interest rates (see table 3.10). During 1980-84 approximately 70 percent of public debt was at fixed interest rates as compared to only 7 percent of bank debt and virtually no private debt.

3.7 The Process of Borrowing and Repayment Guarantees Finally, we turn to another aspect of Korea's debt which distinguishes it from borrowing in many other debtor countries. As we mentioned earlier, the

180

Susan M. Collins and Won-Am Park

Table 3.10

Fixed rates Public Private Bank Variable rates Public Private Bank Source:

Fixed versus Variable Interest Rates as a Percentage of Total Debt, 1980-84 1980

1981

1982

1983

1984

34.0 30.5 0.0 3.5 66.0 16.2 18.0 31.8

34.1 31.2 0. I 2.8 65.9 12.8 14.8 38.3

32.8 30. I 0. I 2.6 61.2 12. I 12.5 42.6

31.6 28.8 0.3 2.5 68.4 11.9 14.6 41.9

31.2 27.8 0.6 2.8 68.8 11.6 11.9 45.3

MOF, White Rook. 1986, p. 36

MOF monitors and oversees all borrowing activities in Korea. In practice, all loans require prior approval and can be considered “government guaranteed .’’ Debt monitoring has had important implications because it has meant that Korean authorities have kept up to date about the volume of external borrowing. In fact, as discussed further in part 2 (see esp. ch. 8), borrowing has figured prominently in the five-year plans as a means of financing investment, and a large portion of the debt accumulation was anticipated. The very strict borrowing process has also played an integral part in Korea’s industrial and development policy by directing the allocation of foreign funds to particular industries and to particular firms, focusing on successful exporters. In describing the loan application process it is useful to distinguish between three types of loans. First, there are loans which are directly controlled by MOF. These include public sector loans, financial credits to special banks (Korea Exchange Bank [KEB], Korea Development Bank [KDB], and Korea Export Import Bank [KEXIM] and import credits of less than three years original maturity. The second category is borrowing by financial institutions. These loans are subject to the foreign exchange regulations administered by the Bank of Korea (BOK). The third category requires application to the Economic Planning Board (EPB) for appraisal. These include loans to nonfinancial private borrowers and import credits of more than three-year maturity. The application includes a report on the firm’s creditworthiness and on the desired usage of the funds. Projects are selected depending on whether they are judged to be consistent with developmental goals specified in the current five-year plan-expanding targeted industrial sectors and improving the balance of payments. The EPB is responsible for choosing among competing projects, typically favoring solicited ones. One example of government influence resulted from the shifts in loan allocation that came with the priority shift of the Big Push toward heavy and chemical industry. Table 3. I 1 shows that the share of foreign loans going to

181

KoredChapter 3

Table 3. I I

Foreign Loans by Destination, 1966-82 (shares of total)

Debtination Agriculture, forestry. and fisheries Mining Manufacturing Heavy and chemical Light Social overhead Services Other Total (million U.S. $)

1966-70 11.4 I .0 39.8 (22.7) (17.1) 39.5 6.5 1.8

1,693.2

1971-75

1976- 80

13.0

6.7 0.1 39.4 (30.8)

-

38.8 (26.3) (12.5) 29.8 13.4 5.0 4,523.2

(8.6)

38.8 14.5

0.3 I1 1,810.5

1981-82

9.2 0.2 15.2 (12.8) (2.4) 55.5 14.4 5.3 5.734. I

Soirrce: EPB, Economic lridirarors of Korco. 1983 Nore; Dash indicates that data were not available.

this sector rose from 23 percent in the late 1960s to 31 percent during the Big Push and fell to just 13 percent once the Big Push had ended in the 1980s. Allocation of domestic loans has also been an important issue in Korean development. As discussed in chapter 11, rankings of priority industries and loan ceilings have been used to ration credit. Virtually all foreign loans require repayment guarantees. Originally (1963-66), the KDB issued foreign loan guarantees (in won) to the BOK which issued a guarantee of convertibility directly to the foreign lender. The guarantees had to be authorized by the National Assembly, often involving special bargains and inducements from the EPB and MOF.6 Since 1966 commercial and specialized banks have been allowed to issue repayment guarantees for private foreign loans without prior authorization from the National Assembly. They guarantee the loan (in won) to the BOK, which assures convertibility. (Guarantees issued by the Foreign Exchange Bank are secondary acceptances on the guarantees of other banks.) Loans which are judged to be difficult for commercial banks to guarantee (for example, large loans to public enterprise) receive government guarantees through the KDB. All commercial banks in Korea were government owned until the early 1980s. Since 1982, five large commercial banks have been transferred to private ownership. In practice, the banks usually acted as passive partners in issuing repayment guarantees. They did not actively examine the loan applications once they had been arranged between the borrower and lender and approved by the EPB or MOF. Thus, when faced with the problem of whether to bail out the banks during difficult periods in which firms, and therefore banks, were unable to pay, the government found it hard to hold the banks responsible.

3.8 Summary Foreign capital inflows have played a central role in Korean economic development. We have already emphasized the importance of foreign aid

182

Susan M. Collins and Won-Am Park

during the initial recovery stages. This chapter has documented the rapid accumulation of external debts beginning in the early 1960s. The chapter has made four main points. First, most of the debt accumulation took place during 1966-69, 1974-75, and 1979-82. Second, growth of the nominal debt stock overstates the burden of the debt because of the very rapid growth rates of GNP and exports. As we shall see in chapter 7, a substantial portion of Korean growth is attributable to investments financed by foreign borrowing. Third, Korea’s debt has been used primarily to finance current account deficits. For this reason, subsequent chapters will focus on the behavior of domestic savings and investment, recognizing that the excess of domestic savings over investment is the counterpart to a current account imbalance. Finally, Korea, unlike many other developing countries, has carefully monitored foreign borrowing. Up-to-date and accurate statistics are maintained. In fact, the allocation of foreign (and domestic) credit has played a central role in Korea’s growth strategy, facilitating the rapid growth of exports. Comprehensive and current information has also enabled policymakers to react relatively quickly to external and internal economic developments.

4

Three Cycles of Debt Accumulation, 1960- 86

This chapter examines Korea’s macroeconomic performance and experience with external debt from 1960 to 1986. As pointed out in chapter 3, most of Korea’s debt was accumulated during one of three periods: 1966-69, 1974-75, or 1979-81. Each period can be characterized as a cycle in which an initial phase of economic difficulty and growth slowdown was followed by a subsequent recovery with resumed growth. As we shall see, only the economic downturn during the third cycle was severe enough to be classified as a crisis by international standards. However, all three declines in performance were viewed with concern by Korean policymakers. Each of the three cycles also involved important shifts in economic policy as domestic authorities responded to external developments and to changes in domestic macroeconomic performance. While it is convenient to discuss each cycle separately, it is also important to identify the broad trends which developed throughout Korea’s recent history. In particular, when we pick up the story, Korea has a war-devastated economy, heavily dependent on foreign aid. By 1986 it has successfully

183

KoredChapter 4

weathered the international debt crisis. In sharp contrast to most other developing country debtors in which policy has remained focused on macroeconomic stabilization (balance of payments and/or prices), the focus of Korean policy returned to the issues of long-term growth and structural development. The major external “problem” was a large current account surplus-a problem which placed Korean policy debates much closer to those of Japan than to those of other debtor countries. 4.1 Economic Growth and External Borrowing, 1960-73

Korea’s first cycle of debt accumulation, crisis, and recovery coincides with a number of changes in the Korean economy. First, shifts in economic policies following the 1961 military coup have generally been identified as the beginning of Korea’s export-oriented growth, with rapid expansions of both exports and GNP. Second, the period follows shortly after the decline in grants and military aid from the United States and the subsequent push for substitute funding by the Korean government. Third, the growth rate of the Korean capital stock accelerates markedly after 1966 following relatively slow growth during the period 1953-66. On the one hand, the growth rates of exports and GNP responded very favorably, jumping from annual averages of 8 and 3 percent, respectively, during 1953-66 to 37 and 10 percent during 1966-70. At the same time, inflation rates remained stable but quite high (15- 16 percent). The period is characterized by rapidly increasing employment, increases in both manufacturing wages and farm incomes, and rising wage-rental ratios. On the other hand, investment exceeded domestic saving, despite the rise in saving following the 1965 financial reforms. Korea ran large current account deficits during the period from 1965 to 1969 and financed the deficits by external borrowing. As a share of GNP, debt rose from 6.9 percent in 1965 to 27.2 percent in 1969. Severe problems had emerged by 1970. The contributing factors included a sharp drop in private saving rates, an overvalued exchange rate, and rising unit labor costs. By 1973, however, the economy was booming. We begin with a background review of developments during 1960-65 Section 4.1.2 gives an analysis of the debt accumulation period from 1966 to 1969 leading up to the crisis. Section 4.1.3 provides an examination of the components of the subsequent recovery. In section 4.1.4 we assess the extent to which any underlying structural weaknesses had been addressed, and examine the relative roles of policy, luck, and economic structure in the 1973 performance. The discussion refers to the economic indicators given in table 4.1. 4.1.1 Background, 1960-65 The period 1960-65 was a time of major transitions. At the outset, two critical features of the Korean economy were its trade policy of “import

184

Table 4. I

Susan M. Collins and Won-Am Park

Major Economic Indicators, 1964-73 1964-65

GNP growth rate Export growth rate Inflation (CPI) Current account (9% GNP) Fixed investment (9% GNP) Domestic savings (9% GNP) M2 growth rate Budget deficit (% GNP) Growth rates: Nominal wages Real wages Labor productivity Valued added KPC index’ Terms of trade Real effective exchange rate Won/$

1966-67

1968-69

1970

1971

1972

1973

7.7 42.1 18.1 0.3 15.0 14.2 33.8

9.7 35.4 11.0 -3.7 21.1 17.0 61.7

12.3 39.5 15.5 -8.4 26.5 20.8 66.7

9.7 34.2 15.9 -7.7 24.7 18.6 27.4 I .6

9.1 27.8 13.5 - 8.9 22.5 16.2 20.8 2.3

5.3 52.1 11.7 -3.5 20.4 18.3 33.8 4.6

14.0 98.6 2.3 - 2.3 23.2 24.1 36.4 I .6

20.3 1.6

19.9 8. I

30.6 16.9

26.9 9.3

16.2 2.4

13.9 2.0

18.0 14.3

2.9 13.2 84.6 116.7 263.0

3.9 10.9 97.1 104.3 269.0

13.3 23.2 101.0 98.0 282.0

22.3 12.1 100.0 100.0 310.6

13.9 9.6 99.2 105.6 347.2

5.0 8.8 98.7 114.1 392.9

5.0 8.8 93.7 132.5 398.3

Source: EPB, Major Statistics of Korean Economy, review issues, and BOK, Economic Sratisrics Yearbook. Note: National income data prior to 1970 are based on 1975 constant prices, old SNA. 1970-73 data are based on new SNA.

‘From Korea productivity center, output per production worker.

substitution of nondurable consumer and intermediate goods behind the protective wall of tariffs and quotas” (Hong 1979, 245) and its overvalued exchange rate. Growth rates were low, however, in contrast to the high inflation in the early 1950s-a financial stabilization program (including quarterly ceilings for the growth of monetary aggregates) combined with restrictive fiscal policy helped to stabilize prices during 1957-61. Political developments set the stage for a significant policy shift. The student uprising in April 1960 force the resignation of President Syngman Rhee. The new government, led by Chang Myon, collapsed following a military coup in May 1961 led by General Park Chung Hee. General Park was elected president of a civilian government in 1964. The new government embarked on an active, comprehensive policy of export promotion to encourage growth. Although the policies have also involved some import substitution, and although some measure were undertaken in 1961 (notably the unification of a complex system of multiple exchange rates), we identify 1962 as the beginning of the “export-orientation’’ phase of Korean development. The cornerstone of the new approach to economic management has been a series of five-year development plans. As we shall see, the plans have involved shifting combinations of liberalization (particularly in the trade regime), government intervention (most obviously through financial markets), and concern over macroeconomic stability. The mainstay has been a desire to maintain high rates of growth. This has been achieved through

185

KoredChapter 4

increasingly high rates of capital formation in export industries. Except perhaps in the most recent period, this has put stable, credible incentives for exporters as a top priority. The first five-year plan (1962-66) targeted fixed capital formation to grow at an average rate of 14.6 percent. However, domestic sources of financing were limited: domestic bank savings were small, and domestic commercial banks were not accustomed to or equipped for long-term loans, unless ordered to undertake them by the government. Hong (1979, 142, 257) estimates that short-term credit for exports and long-term credit for export promotion amounted to only 3 percent and 1-2 percent of total bank loans, respectively. (He uses medium Industry Bank Loans and foreign currency loans to estimate total long-term loans for export promotion.) The major source of domestic long-term funding, the Korea Reconstruction Bank, had access to only limited funds through the government. Furthermore, the slowdown of aid inflows after massive foreign aid during 1957-61 signaled a critical need for alternative financing. The government had begun a concerted effort to encourage foreign loans and investments in 1960. The Foreign Capital Inducement and Promotion law, the first of a series of new laws and regulations, focused on foreign loans, foreign direct or joint investments, and capital and technology inducements. It granted a number of special incentives, including special income tax provisions for interest earnings arising from foreign loans. Foreign investment businesses were allowed exemptions on income and corporate taxes and on tariffs on their imports of capital equipment (Hong 1979, 141). In 1962 the government instituted the system of guarantees to foreign lenders and investors. As described in chapter 3, each private loan or project was examined individually. Those which were authorized also received a guarantee of repayment from the KDB and BOK, together with a guarantee of repatriation of funds. Two problems emerged in 1963: a resurgence of inflation and a deterioration in the balance of payments. A number of factors contributed. Macroeconomic policies had been very expansionary during the military government of 1960-6 1-large fiscal deficits were financed through borrowing from the BOK. There were two poor agricultural harvests-rice in fall 1962 and barley in spring 1963. U.S. aid flows declined substantially. Multiple exchange rates were reintroduced during 1963, and import controls were tightened. However, it is important to note that incentives to exporters were kept relatively constant during this period (Frank, Kim, and Westphal 1975). A joint U.S.-Korea stabilization agreement during 1963-64 reduced the fiscal deficit, introduced credit ceilings, and controlled lending to the private sector. It is also notable that Korea began its industrialization with a period of wage restraint. Real wages fell by over 10 percent between 1962 and 1964. Available evidence suggests that labor productivity increased strongly during the same period.'

186

Susan M. Collins and Won-Am Park

A series of reforms were instituted following the 1964 election. Under U.S. pressure, the exchange rate was devalued and import controls were reduced. Beginning &I964 the exchange rate took on a more prominent role in Korean economic management. Measures were undertaken to increase both public and private savings. Partly in response to these contractionary measures, in 1964 there was an improved current account, a sharp decline in imports, and reduced industrial growth. That year also seems to have marked the beginning of a more active role for unofficial financial markets. In 1965 the government undertook a major interest rate reform. Some authors have cited this as the reason for the dramatic increase in domestic (private) savings in the late 1960s.’ However, our analysis of savings in the more recent period finds interest rates to be of little importance (see ch. 8). This finding is consistent with Giovannini’s (1983) conclusion that interest rate elasticities of savings are small in developing countries and with work by van Wijnbergen (1983b). At the same time, diplomatic and commercial relations with Japan were normalized, generating a renewed inflow of funds which partially substituted for the decline in foreign aid from the United States. From 1966 on, for the first time commercial banks were allowed to issue foreign loan guarantees, and a series of strong incentives were put in place for exporters to invest and to borrow abroad. To summarize, three critical developments had occurred by 1964-65. First, the shift to export promotion as the means to economic growth elevated capital formation to top priority. Second, changes in government policy and external environment had set the stage for heavy reliance on external debt as a source of finance. Finally, the five-year plans identified an important role for government intervention in the allocation of resources, setting the stage for government control over (organized) financial markets and, therefore, the allocation of domestic and foreign finance. This was in marked contrast to the period prior to 1961 in which the United States played the major role in allocating foreign capital inflows. During 1964-65 growth of output and exports had resumed, the current account deficit had fallen to a manageable 0.3 percent of GNP, and the 1964 devaluation together with real wage declines had resulted in a competitive labor force. 4.1.2 Rapid Growth, 1966-69 The years 1966-69 were a period of high growth and stable inflation. However, increasing external imbalance and the rapid accumulation of external debt presented potential difficulties for the macroeconomy. As was shown in tables 3.1 and 3.3, external debt jumped from $392 million in 1962 (10.7 percent of GNP) to $1800 million at the end of 1969 (27.2 percent of GNP). Many factors facilitated these massive inflows. On the foreign lenders’ side, risk was substantially reduced because of the loan guarantee

187

KoredChapter 4

system. In addition, many borrowers received guarantees from their own domestic governments. Domestic borrowers were given strong incentives. In practice, applications for loans to fund investment in priority sectors were encouraged and usually approved. As discussed in chapter 3, the interest cost of domestic bank loans exceeded the average cost of borrowing abroad by 14.1 percent during 1966-70, and loans from the curb market were considerably more expensive. The real private cost to borrowing abroad was - 5.1 percent. At a time when domestic bank loans were strictly rationed, the Foreign Capital Inducement law in 1966 introduced a more flexible process for foreign loan approval. Total loan guarantees grew at an average annual rate of 5 percent during this period as compared to average growth rates of 30 percent for bank credit to the public and private sectors. The foreign capital inflows sustained high investment. Nearly 40 percent of total foreign loans during 1966-70 were allocated to manufacturing, with another 40 percent to social overhead investments, 11 percent to agriculture, and 6.5 percent to services. Gross fixed investment jumped from less than 15 percent of GNP in 1965 to 20 percent in 1966, and then to 26 percent in 1969. As we have already seen, 83.8 percent of the increase in external debt can be accounted for by the current account deficit. Reserve accumulation amount to 20.2 percent of the increase. Three other developments occurred during the period. In 1967 there was a liberalization of the trade regime as the government switched from a positive to a negative list for restricting imports. Second, the Law for Fostering Capital Markets in 1968 was the first in a series of measure to encourage public borrowing. It is also important to stress the developing role of financial policies. The years 1965-70 were a period of rapid growth of commercial and specialized banks. Interest rate subsidies on foreign loans also increased markedly after 1966. Hong (1979, 260-61) estimates that tariff exemptions were much less important than interest rate subsidies on loans as an incentive for investment. 4.1.3 Economic Downturn and Recovery, 1969-73 By 1969-70 Korea was faced with four major difficulties. The first was the preciptious rise in the burden of external debt. Despite the exemplary export performance, the debt service ratio (long term) escalated from 7.8 percent in 1969 to 18.2 percent in 1970. A second difficulty was that domestic savings dropped by 3 percent of GNP between 1969 and 1970. One reason cited for the decline is the reduction in real interest rates as a result of increasing overvaluation (Y. C. Park 198%). However, an alternative explanation begins by pointing out that the real question may be not why saving rates fell in 1970, but why they

188

Susan M. Collins and Won-Am Park

were so high in 1969. In 1970 saving rates returned to their 1968 level and remained roughly constant for three years. A sensible answer to the latter question is based on the dramatic jump in real growth rates during 1968-69.3 Domestic residents may well have perceived these rates as temporary so that one would expect little adjustment of consumption. In fact, this rationale also helps to explain the 1974 “drop” in savings to 19.9 percent of GNP. Savings had jumped from 16.5 percent of GNP in 1972, with a 5.3 percent growth rate, to 22.8 percent in 1973, with a 14.0 percent growth rate. The third factor was consistently high investment relative to domestic savings. Although fixed investment declined slightly as a share of GNP during 1970 and 1971, inventory accumulation jumped sharply in 1969, remaining high through 1971. Much of the 1969 increase in inventories was from the accumulation of agricultural products arising from high grain imports and from a large rice harvest. The increases in 1971-72 were primarily in manufactured goods, presumably in response to the increasing overvaluation and expected depreciation. The fourth problem arose from wage and exchange rate developments. During 1966-70 nominal wages rose by over 160 percent, implying a 65 percent increase in domestic real wages. However, the nominal exchange rate (won/$) depreciated by less than 15 percent. The result was a deterioration in international competitiveness. The extent of the loss depends on which measure of labor productivity is used. Using the KPC measure, productivity rose by 101.1 percent during the period, implying a 14.4 percent rise in unit labor costs measured in dollars. However, using the value-added index, productivity grew much more slowly, implying a 50.8 percent increase in dollar unit labor costs. A series of adjustments were undertaken beginning in 1970. In accordance with an IMF standby arrangement, medium-term loans were strictly limited, slowing the growth of external debt. Monetary expansion was also tightened. By 1971 a slowdown in economic activity was evident. Real growth rates declined as did the growth of imports, particularly capital goods imports, resulting in a dampening of capital formation. Authorities were reticent to pursue expansionary monetary or fiscal policies for fear of worsening the current account. In June 1971 the exchange rate was devalued in hopes of expanding the economy, without deteriorating the external balance, by stimulating exports. After an initial 13 percent devaluation relative to the dollar, the won was gradually devalued until June 1972 when the exchange rate was fixed at 400 won/$. There were also adjustments of the dollar vis-a-vis other major currencies during 1972-73.4 In real terms the won depreciated by 11.9 percent during 1970-72 and by an additional 15.6 percent during 1973. Nominal wage growth slowed. Consequently, although unit labor costs continued to rise when measured in won, when measured in dollars they fell

189

KoredChapter 4

by 19 percent from 1970 to 1973 using the KPC index, or by 5 percent using the value-added index. In fact, both monetary and fiscal policies were loosening during 1971-72. Two developments contributed to this policy shift. Agricultural production (in particular, food grains) was low throughout 1970-73, with the yield of 1971 crops especially disappointing. As a result, there were large deficits in the government's Grain Management Fund financed by domestic credit expansion. Second, there was a financial crisis in 1972.' Because of devaluation and export difficulties, many firms with foreign debts were forced close to bankruptcy. To avoid jeopardizing Korea's standing in international credit markets, the government elected to bail out these firms from their difficulties. Outstanding guarantees on foreign loans fell in 1972, and few new ones were issued. The government instituted measures to restrict the expansion of the unofficial financial market. A presidential decree, announced on 3 August 1972, is especially notable because it reversed almost all of the financial liberalizations that had been instituted since 1965. The decree replaced all existing agreements between firms and unofficial lenders with new ones more favorable to borrowers. For example, many short-term, high interest loans were replaced by longer term, low interest rate ones. The measure mitigated the difficulties of many debt-ridden firms and effectively shifted adjustment to the financial crisis to the crub market. The unofficial market almost disappeared in the aftermath of the crisis and was not revived until after the 1973 jump in oil prices. Overall, 1970-78 was a period of slowed growth of the banking system. Emphasis was placed on the partially regulated nonbank financial institutions, especially investment and finance corporations, which were given incentives and encouraged to grow. In 1972 inflation accelerated and real growth slowed even further, despite improved export performance and the more expansionary macroeconomic policies. The primary factors seem to have been, on the demand side, a drop in private consumption, and on the supply side, poor performance in services and manufacturing as well as agriculture. On the brighter side, the substantial improvement in the current account position is primarily attributable to export growth and not to a contraction of imports. To further encourage investment the government took a more active role. On 12 October 1972 explicit priority sectors were introduced for the inducement of foreign investments.'j Nineteen seventy-three was an extremely favorable year for the Korean economy. Exports and GNP boomed. The debt situation improved. The current account deficit relative to GNP fell even further, as domestic saving rates soared. There was some decline in inflation, and the growth in real wages resumed, exceeding the rise in labor productivity.

190

Susan M. Collins and Won-Am Park

Why was 1973 such a good year? Three factors were the very strong world economy, the lagged impact of real depreciation and expansionary macroeconomic policies, and the favorable private savings outturn. However, an important point is that Korea avoided more substantial macroeconomic stabilization measures because of its history of well-placed investments which enabled it to resume the high growth rates of the 1960s as soon as favorable external conditions returned. 4.1.4

The Korean Economy in 1973

We end this section by asking whether the 1973 boom signified a complete recovery from the problems which emerged during the early 1970% or whether underlying weaknesses remained. There is considerable evidence (high and growing investment with high rates of return, rising labor productivity, a competitive real exchange rate) that it would be difficult to dispute the very favorable prospects for rapid continued growth. Certainly, this was one important strength. However, some aspects of Korea’s structure left the economy particularly sensitive to unfavorable external developments. Investment rates targeted in the economic growth plans exceeded realistic forecasts of domestic savings. The high investment and shifting economic structure implied increasing dependence on imports of raw materials and capital goods. Furthermore, GNP growth was closely linked to the growth in world demand for Korean exports. Difficulties emerged when savings fell relative to investment. Given the high fixed investment, the problem was overly variable saving and inventory behavior. The larger current account deficit required additional external borrowing, increasing the burden of debt. The problem could then be exacerbated by external factors, namely higher interest rates or a world recession which slowed the growth of exports. It could also be exacerbated by internal factors such as a rise in the (planned) capital formation component of investment. With variations, these are exactly the elements of both the second and the third crises. From this perspective, it is sensible to ask whether Korea would have been better off overall by choosing somewhat smaller investment targets. Potential advantages would have been a reduction in the sensitivity to unexpected internal and external developments. With a smaller trend current account deficit and less accumulation of external debt, the economy might have been able to weather a jump in inventories or a drop in savings. However, this view is misleading. As we argue in chapter 7, foreign borrowing contributed significantly to the growth of output. A ballpark estimate is that the economy would have grown only half as quickly during 1961-71 without the external finance, and only two-thirds as quickly during 1972-76. Frank, Kim, and Westphal (1975) reach similar conclusions in their estimates of the costs of lower investment. It is economically sensible for an economy with very profitable investment opportunities to supplement

191

KoredChapter 4

domestic savings with external funds. During 1962-72 Korea very successfully encouraged industries for export-oriented growth.

4.2 The Second Period of Rapid Debt Accumulation, 1974-78 We turn next to the second period of difficulty, 1974-75, and subsequent recovery, 1976-78. Just as in the first episode, this period coincides with a major shift in economic policy and a significant increase in fixed capital formation. At the beginning of the 1970s, Korean policymakers saw a decline in competitiveness which they felt necessitated further structural shifts in order to maintain future growth prospects. They felt that the rising real wages and capital intensity in manufacturing undermined Korea’s ability to compete in light manufacturing and signaled a shift in its comparative advantage toward higher skill-intensive and technology-intensive products. The U.S. decision to reduce the number of troops stationed in Korea reinforced the desire of policymakers to invest more heavily in defense. As a consequence, a massive investment program was initiated in 1973 to develop heavy and chemical (HC) industries. The program remained in effect through 1979. A primary difference between the second period of rapid debt accumulation and the first is that, in addition to internal factors, the economy was forced to adjust to unfavorable external developments-the jump in oil prices followed by the slowdown in world activity. The major facts to be explained are as follows. During 1974-75 there was a drop in real growth rates, a jump in inflation, and a substantial increase in external borrowing. During 1976-78, however, Korea was able to resume its high growth rates and to improve its debt position. In addition, there was some reduction in inflation at first (1976-77), but a resurgence in 1978. Section 4.2.1 examines the period of poor performance, assessing the relative importance of internal and external factors. Section 4.2.2 turns to the recovery period and to a discussion of the strengths and weaknesses of the economy in 1978, the threshold to the third and most serious crisis. Throughout the discussion, we refer to the economic indicators in table 4.2. 4.2.1 The Problem Years, 1974-75 Table 4.2 shows that economic performance deteriorated in 1974. By Latin American standards, the outturn, with its real growth rate in excess of 8 percent, can hardly be called a crisis. But Korean policymakers were quite concerned about the developments. The growth rate dropped by nearly 40 percent. Inflation surged to 24.3 percent. Even more striking was the unprecedented increase in the current account deficit, which jumped from 2.3 percent to 10.8 percent of GNP within one year. External debt grew by 37 percent from $4.3 to $5.9 billion. However, the debt GNP ratio rose only marginally from 3 1.5 percent to 32

192

Table 4.2

Susan M. Collins and Won-Am Park Major Economic Indicators, 1973-78

GNP growth rate Export growth rate Inflation (CPI) Current account (% GNP) Fixed investment (% GNP) Domesitc savingsiCNP M2 growth rate Budget deficiUGNP Growth rates: Nominal wages Real wages Labor productivity Value added KPC index’ Terms of trade Redl effective exchange rate Won/$

1973

1974

1975

1976

1977

1978

14. I 98.6 3.1 - 2.3 23.2 22.8 36.6 1.6

8.5 38.3 24.3 - 10.8 25.6 19.9 24.0 4.0

6.8 13.9 25.3 -9.1 25.3 19.1 28.2 4.6

13.4 51.8 15.3 -1.1 24.4 23.9 33.5 2.9

10.7 30.2 10. I 27.3 27.5 39.7 2.6

11.0 26.5 14.4 -2.1 31.3 28.5 35.0 2.5

18.0 14.3

35.3 8.8

27.0 1.4

34.7 16.8

33.8 21.5

34.3 17.4

5.0 8.8 136.2 117.1 398.3

2.4 11.4 110.9 101.1

2.2 11.6 100.0 100.0 484.0

2.4 7.5 114.1 93.6 484.0

10.3 10.5 122.0 94.6 484.0

12.6 11.9 127.9 97.8 484.0

404.5

0.0

Source: Economic Planning Board and Bank of Korea. Note: Based on new SNA method

‘From Korea Productivity Center, output per production worker.

percent. The debt service ratio fell slightly to 14.4 percent, substantially below its 197 1 level of 2 1 percent. More worrisome developments were the rise in the share of short-term debt to nearly 21 percent. It is noteworthy that all of this rise was in loans to the banking sector. Unlike short-term loans to the private sector, which fell between 1973 and 1974, these “accommodating” capital inflows can be considered unplanned. Total long-term loans grew more slowly during 1974 than they had during 1973, again except for a jump in long-term loans to the banking sector. Referring to the decompositions given in chapter 7, poor performance in construction and manufacturing accounts for a 4 percent decline in GNP growth, with most of the rest due to slower growth of other services. On the demand side, most of the slowdown is attributable to exports. The jump in inflation is not surprising. Domestic credit expansion averaged 35 percent during 1972-73 compared to only 24 percent during 1970-71. There was also a large shock from external price increases (Korea imports oil as well as primary commodities). Unit import prices rose by 55 percent between 1973 and 1974. In addition, nominal wages rose by 35 percent (a real wage gain of 8.8 percent), while labor productivity increases amounted to less than 12 percent (less than 3 percent using the value-added index). The rapid nominal wage growth has been attributed to tight labor markets in the mid 1970s, as the Big Push created an excess demand for many types of skilled labor.

193

KoredChapter 4

The counterpart to the current account deficit was increased fixed and inventory investment combined with a drop in savings (relative to output). The rise in fixed capital formation was to be expected given the shift in development strategy. As an indication of the magnitudes of the shifts during the early seventies, it is interesting to compare the sectoral allocation of loans. Although the share of total foreign loans which went to manufacturing fell slightly from 39.8 percent during 1966-70 to 38.8 percent during 1971-75, the percentage of these going to HC industries rose from 57 to 68 percent. Most of this increase is accounted for by changes in allocation during 1973-75. It is also likely that investment in HC industries during 1971-75 was concentrated in 1974, because investors anticipated a devaluation in the wake of the first oil shock that did not occur until December 1974. It is interesting that fixed capital formation grew more quickly during the third five-year plan (1972-76) than had been targeted: 13.2 vs. 7.6 percent. Part of the explanation for this may be the increase in investment expenditures on residential construction between 1973 and 1974 which accounted for nearly half of the increased fixed capital formation, the remainder being attributed primarily to an increase in investment expenditures on transport equipment. A second factor was the decline in domestic savings. Y. C . Park (1985c, 304) writes that “mostly as a reflection of the short-run difficulty in the adjustment of consumption to a lower real income, and of a high rate of inflation, domestic savings as a fraction of GNP plunged by four percentage points to 19% in 1975 from about 23% in 1973.” Other authors also argue that the large unexpected drop in savings was a major cause of the crisis. However, as discussed above, the high saving rate in 1973 was more out of line than the lower one in 1974. The 1974 rate exceeded the average rate of 16 percent during the less inflationary period, 1968-72, and remained approximately constant through 1975. The main reason for the jump in 1973 seems to have been the unexpectedly rapid real growth. From a planner’s perspective, the more surprising outturn must have been the unprecedented jump in inventory investment. Eighty percent of the 1974 increase came from accumulation of manufactures (including capital goods) and raw materials. The large increases can be partially explained by the combination of an imminent expected depreciation and an unanticipated reduction in export growth. (Y.C. Park 1985c, 304). It is also useful to identify the components of the current account deterioration. A little over 20 percent of the increased deficit came from a worsening in the invisibles balance, primarily due to increased payments for transport and investment income. Eighty percent came from the trade balance. There was the expected surge in imports. However, only 26 percent of the jump is accounted for by oil payments. Another 26 percent was from imports of capital goods, and the remainder was due to raw materials

194

Susan M. Collins and Won-Am Park

imports. Payments for imports rose not only because of the price hike, but also because of a rise in the volume of imports. At the same time, the growth of export receipts slowed relative to 1972-73, returning to the average 1966-72 performance. While the world recession caused a reduction in the total volume of exports, the unit value of exports jumped by 27 percent between 1973 and 1974, dampening the deterioration in Korea’s terms of trade. To summarize the 1974 experience, Korean export growth was slowed by a combination of the oil and commodity price rise and the ensuing world recession. Slower export growth, in conjunction with the Big Push toward HC industries resulted in a jump in investment (fixed capital formation and especially inventories of imported capital goods and intermediates). At the same time, saving rates fell from their high level in the boom year of 1973. The result was an enormous current account deficit. The jump in inflation rates came both from higher oil prices and from rapid nominal wage growth. We return to a discussion of the relative importance of internal and external developments in chapter 5. The year 1974 was the beginning of the Big Push toward promotion of heavy industries. The decision was made to continue this effort, borrowing to finance the required imports instead of contracting the economy to adjust to external shocks. BOK secured loans for the banking sector. Taxes were raised to conserve oil consumption. Unlike the response in many other developing countries, domestic oil prices were increased. The predeposit requirement on imports was also raised. At the same time, incentives for exporters came from lowered interest rates and expanded access to export credits. In December the won was devalued from 400 to 484 won/$, a rate which prevailed until January 1980. The devaluation resulted in a 7.2 percent real depreciation of the won relative to its average 1972-73 level. However, unit labor costs in dollars rose by about 4 percent during 1972-74 because of large nominal wage gains. Finally, the National Investment Fund (NIF) was created in 1974. Its purpose was to generate additional domestic savings and to channel them to targeted sectors and projects consistent with the development plan. More specifically, it was to mobilize employee pension funds. In encouraging banks to make preferential loans, the policy marked the beginning of additional government intervention in the financial sector through credit allocation. As a share of bank credit, preferential loans were to grow from 40 percent in 1971 to 55 percent in 1976-77, and then to 70 percent in 1978. It is interesting that the interest rate incentives to borrow abroad actually declined during 1971-75 relative to 1966-70 because of higher foreign rates, the depreciation, and a decline in domestic bank loan rates (see table 3.6). However, access to loans from domestic banks remained severely

195

KoredChapter 4

limited. The real cost of borrowing abroad remained negative, - 1.6 percent. Overall, the situation deteriorated during 1975. The outcome was slightly better in terms of inflation and the current account deficit, but both remained extremely high. There was some furhter slowdown in real growth. The situation was much worse in terms of external debt. Korea borrowed an additional $2.5 billion, escalating the debt-GNP ratio to 40 percent. Although the debt service ratio remained at 14.4 percent, the share of short-term debt to the total jumped from 20.9 percent to 28.5 percent. In marked contrast to 1974, 64 percent of the rise in short-term debt went to the private sector, with only 46 percent going to “accommodating” bank loans. Similarly, most of the rise in long-term debt went to the public or the private sectors. The counterpart to the current account improvement was a decline in inventory accumulation. This portion of investment remained high, although the accumulation was concentrated primarily in agricultural, not manufacturing, products. Fixed capital formation rose somewhat, and there was a slight further decline in the saving ratio. The trade balance improved somewhat, primarily because of the small increase in the value of imports. In particular there was a substantial decline in the imports of manufactures, offsetting further increases in the prices of capital goods and oil. It is not surprising that inflation remained relatively high as the impact of the December 1974 devaluation filtered into domestic prices. However, nominal wage growth slowed somewhat to 27 percent, with the increase in labor productivity growth remaining constant. With no additional external shocks and with a sustained moderation in wage growth relative to productivity, inflation rates would have been expected to drop further during 1976-77. One sign pointing in this direction was the declining growth of wholesale prices-26.5 percent in 1975 compared to 42.1 percent in 1974. The high inflation in 1974-75 was in large part a one-shot reaction to the oil price shock and devaluation. This perspective, combined with labor market developments, makes the rapid decline in inflation during 1976-78 less surprising. Fiscal policy continued to be expansionary, financed primarily by external borrowing. Thus, during 1975 there was no significant change in domestic saving. Furthermore, the 1974 depreciation did not succeed in reviving exports, primarily because of stagnant world demand, rising unit labor costs, and the resulting decline in competitiveness. In summary, three major problems characterized 1974-75. The first was a slowdown in growth of exports and GNP. The second was an unsustainable current account deficit and the implied rapid accumulation of external debt. Current account deficits during these two years accounted for 93 percent of the increased external debt. This problem was exacerbated by a worrisome

196

Susan M. Collins and Won-Am Park

shift to short-term borrowing. Finally, policymakers were concerned about the high rates of inflation. 4.2.2

Recovery, 1976-78

Table 4.2 shows the rapid recovery which began in 1976. Growth rates of GNP and exports surged to 14 percent and 51 percent, respectively, while inflation continued to decline. Most striking is the drop in the current account deficit from 9.1 percent in 1975 to 1.1 percent within one year, and to 0.0 percent in 1977. This section examines how these dramatic improvements came about. It concludes with a discussion of the state of the Korean economy in 1978, the year before the severe 1979-80 crisis. The current account improvement during 1976-77 is attributable to a rise in domestic savings as a share of income and to a decline in inventory investment. On the other side, very rapid export growth, fueled by the 1974 devaluation and the recovery in world demand, contributed to an export boom during 1976. Korea was also beginning to enjoy growing receipts from construction activity in the Middle East. Thus, we can identify four factors which explain how Korea’s current account deficits recovered so quickly. One factor is the strong recovery in world demand which stimulated demand for Korean exports. A second is the increased fixed capital formation which expanded potential export production. Exports of chemicals plus machinery and transport equipment grew from 14 percent to 24 percent of total exports between 1973 and 1978.7 Third, the large increases in savings, attributable primarily to rapid income growth, enabled Korea to finance the bulk of its investment domestically by 1976. Finally, by 1978 the negative impact of higher oil prices had been dampened considerably by the inflows from construction in the Middle East. Oil payments had averaged $0.3 billion per year during 1972-73, while construction revenues had averaged $0.014 billion. During 1974-78, oil payments and construction revenues totaled $5.8 billion and $3.9 billion, respectively, so that 90 percent of the additional oil payments were offset by additional foreign exchange inflows from construction. It is important to stress that substantial capital inflows continued during this recovery period. External debt increased by approximately $2 billion in each of the three years. The real cost of foreign borrowing remained negative during 1975-78. There was relatively easy access to foreign credit, including import financing and prepayment of exports. Domestic bank credits, however, were subject to increasing restrictions. Inflation fell from 29.5 percent in 1975 to 15.7 percent in 1977. As argued above, much of the 1974-75 jump in inflation should be interpreted as a one-time adjustment to the terms of trade shock and to devaluation. Given an economy without backward-looking wage indexation, and given that import

197

KoredChapter 4

prices remained stable during 1976-78, reduced inflation is not surprising. The two issues which do warrant explanation are, first, that inflation did not decline by more and, second, that it was reignited during 1978. Two factors help to explain why inflation rates did not fall below 15 percent: rapid wage inflation and rapid monetary expansion. Nominal wages increased by 142 percent between 1975 and 1978, while consumer prices and labor productivity rose by only 45 percent and 33 percent, respectively. As will be argued in chapter 10, the wage growth was fueled by an increasingly tight domestic labor market. In particular, the combination of the accelerating demand for labor from the Big Push and the reduced supply of skilled labor for foreign construction projects pushed up wages in some sectors, filtering across to wages elsewhere in the economy. The wage growth together with a fixed nominal exchange rate implied a deteriorating competitiveness of Korean workers relative to the country’s major competitors-Singapore, Hong Kong, and Taiwan. It is noteworthy, however, that existing data points to a deteriorating distribution of income during the late 1970s, following two decades of continued improvement. The monetary expansion arose both from domestic credit expansion and from the foreign sector. The continued capital inflows and growing net foreign asset position has been mentioned above. In addition, large deficits in the Grain Management Fund were financed through money creation. In an effort to promote self-sufficiency, the price at which the government purchased rice grew 30 percent more rapidly than the price at which the rice was sold during 1975-78. The government became increasingly concerned about domestic inflation. During the late 1970s, a variety of price controls, ceilings, and guidelines proliferated. Prices in monopolistic and oligopolistic industries were controlled by the government, which authorized all increases. As the industrial concentration grew, these controls accounted for an increasingly large share of the CPI. In addition, the prices of many essential products were monitored by the government. Nam claims that government pricing policies led to many problems during the late 1970s (1984). The “stop-go” approach to allowing price increases created supply shortages, declining product quality, reduced investments, and distorted resource allocation during a time of substantial structural readjustment. Black markets for some essential consumer goods emerged. There is a general consensus that 1975-78 was a period of increasing misallocation of resources and increasing industrial concentration. Seventyseven percent of all investment in equipment in the manufacturing sector went to HC industries, although these industries accounted for only 55 percent of total production. The chaeboE-large-scale industrial conglomerates-became a significant share of the business sector in the mid-1970s. Although they participate in all sectors of the economy, they have been the most prominent in heavy and

198

Susan M. Collins and Won-Am Park

chemical manufacturing. Statistics are difficult to obtain, however Jones and Sakong (1980, 304) provide estimates for 1975 which suggest that the forty-six largest chaebol produced 37 percent of value added in manufacturing and 13 percent in GNP, and that business concentration was increasing rapidly. Westphal (1984) states that by 1980-81, the list of officially recognized chaebol had 26 large groups, which together controlled 465 firms. Eight of these, along with two public conglomerates, appear on Fortune’s 1980 list of the 500 largest industrial corporations outside of the United States. One, the Hyundai Group, was the largest nonpetroleum corporation resident in the less developed countries. Financial and trade policies also became more restrictive during this period (see ch. 9 and 11). Financial market restrictions increased and credit rationing was tightened, with preference given to HC industries and to large firms. Extremely high corporate debt-equity ratios contributed to the fragility of the banking sector-in the manufacturing sector, the debt-equity ratio rose from an already high 3.16 in 1974 to 3.77 in 1979, and then to 4.88 in 1980. (It would fall to 3.86 by 1982, following a massive bailout and the growth of Korean stock markets.) By 1978 the economic situation looked somewhat less promising. Growth rates declined further. The current account deficit reemerged. This time, the increase was attributable to increased fixed capital formation. Domestic savings continued to rise as a share of income. There was also a jump in inflation. On the positive side, the debt of GNP ratio declined, with a reduction in the share of short-term debt. Thus, a number of structural weaknesses faced the Korean economy at the beginning of 1979. The major ones were the recurrent imbalance between investment and domestic savings, growing fragility of financial markets, and increased government intervention in trade, the financial sector, and pricing. Furthermore, the Big Push to HC industries contributed to a misallocation of domestic resources and to excess capacity in these sectors.

4.3 The Third Period of Crisis and Recovery, 1979-86 The final period of major debt accumulation, crisis, and recovery is perhaps the most interesting. It was certainly the most severe, including one year (1980) in which output declined by nearly 5 percent. By 1983, however, high growth had resumed, combined with substantial improvements in inflation and external balance. Korea’s impressive performance stands in marked contrast to the majority of heavily indebted countries, which continue to struggle in the aftermath of multiple painful external shocks since 1979. The rapid and sustained turnaround in Korea’s economic performance has been widely cited as a model of successful adjustment and held up as an example of the favorable

199

KoredChapter 4

outcomes from the correct application of macroeconomic stabilization policies. For Korea in this period, as in the two earlier episodes and just as for many other countries, internal developments combined with external ones to create the economic crisis. By 1979 Korea was again in the midst of a shift in the government's fundamental economic strategy. Performance during 1974-78 had convinced policymakers to step back from the Big Push, with its reliance on widespread government intervention, and to refocus from viewing industrial policy as a tool to promote rapid economic growth to having a growing concern about price stability as a necessary precondition to continued growth. The policy shift was confounded by increasing social unrest, the assassination of President Park, and agricultural disasters during 1978-80. On net, complicated interactions between internal and external factors make it extremely difficult to identify the relative importance of particular elements in explaining outcomes. We return to these issues in chapter 5 . The discussion is divided into four remaining sections. Section 4.3.1 discusses the policy shift embodied in the 1979 Comprehensive Stabilization Plan (CSP). Section 4.3.2 examines the 1979-80 crisis period. Sections 4.3.3 and 4.3.4 analyze the early recovery period from 1981 to 1983 and the strong performance period, 1983-86. Throughout the discussion, we refer to the economic indicators in table 4.3.

Table 4.3

Major Economic Indicators, 1978-86

GNP growth rate Export growth rate Inflation (CPI) Current account (% GNP) Fixed investment (% GNP) Domestic savings (% GNP) M2 growth rate GNP) I Budget deficit ( Growth rates Nominal wages Real wages Labor productivity Value added KPC index' Terms of trade Real effective exchange rate Won/$

1978

1979

1980

1981

1982

1983

I984

1985

1986P

11.0

26.5 14.4 -2.1 31.3 28.5 35.0 2.5

7.0 18.4 18.3 -6.8 33.2 28.1 24.6 1.4

-4.8 16.3 28.7 -8.8 32.3 23.5 26.9 3.2

6.6 21.4 21.3 -7.0 28.7 23.5 25.0 4.7

5.4 2.8 7.2 -3.8 30.5 24.0 27.0 4.4

11.9 11.9 3.4 -2.1 31.3 27.9 15.2 1.6

8.5 19.6 2.3 - 1.7 31.3 30.3 7.7 1.4

5.4 3.6 2.5

12.5 14.6 2.3 4.9 31.3 34.8 18.6

34.3 17.4

28.6 8.7

22.7 -4.7

20.1 -2.6

14.7 6.9

12.2 10.4

8.1 5.7

9.9 7.3

9.1 6.7

12.6 11.9 117.8 109.0 484.0

16.0 15.9 115.3 97.2 484.0

-3.9 10.6 100.0 100.0 607.4

11.1 18.1 97.9 103.6 681.0

- 1.8

4.2 13.6 103.1 110.6 775.8

12.0 10.5 105.3 114.4 806.0

-0.8 7.1 105.9 121.2 870.0

7.6 13.6 114.7 139.2 881.5

Source: Economic Planning Board and Bank of Korea. Note: Based on new SNA method.

'From Korea Productivity Center, output per production worker. PPreliminary.

7.8 102.2 103.2 831.1

-1.1

30.8 30.7 15.6 1.0

1.8

200

Susan M. Collins and Won-Am Park

4.3.1

Policy Refocus, 1977-79

As government concern over persistently high inflation grew, policymakers began to reassess the approach embodied in the Big Push. A series of measures were introduced. During 1977 these included restraints on monetary and fiscal expansion to contain aggregate demand. The government also attempted to eliminate shortages through improvements in the distribution system (in particular, for agricultural products), increases in a number of controlled prices, and acceleration of import liberalization. Additional measures were undertaken during 1978. On the rnonetary/fiscal side, short-term trade credits were discouraged in an effort to reduce the contribution of the foreign sector to monetary expansion. Ceilings were placed on credit to the private sector. Interest rates on bank loans and deposits were increased as part of a nationwide savings campaign. In addition, it was hoped that the August 1978 Comprehensive Measure to Curb Speculative Real Estate Investment would shift savings from real assets to the banking sector. Government spending was reduced, in part through deferment of construction projects. On the trade side, the import liberalization ratio was raised and tariff rates on some imported raw materials were adjusted so as to absorb increasing prices. Limitations were imposed on the exports of some items with domestic shortages. The CSP was announced in April 1979. This plan has been described as a “landmark” (Nam 1984) because it was the first of its kind to put control of inflation as the number one priority. In the past, the government had been primarily concerned about investment for growth and had consistently been willing to use external and/or internal credit to finance real expansion, despite any unfavorable implications for price stability. Furthermore, the CSP stated that pervasive government intervention to direct economic development was appropriate in the early stages, but argued that it was also appropriate to rely increasingly on market forces at later stages. As such, the government accepted part of the blame for existing economic difficulties. The new approach, which combined proposals from BOK, the Korea Development Institute (KDI), and the Economic and Scientific Council, was strongly supported by a newly appointed deputy prime minister, Shin Hyon Whak. The CSP had four major components. The first was a more restrictive monetary policy, including improvements in the preferential loan system, and increased interest rates. Second, fiscal policy was to be contracted through a 5-percentage-point cut in spending and additional deferments of large public investment projects. Third, the policy stepped back from the focus on HC industries by calling for a reallocation of investment toward other manufacturing and nonmanfacturing sectors. Finally, the government redoubled its efforts to prevent real estate speculation and to increase the supply and stabilize the prices of essential commodities.

201

KoredChapter 4

4.3.2

The Crisis, 1979-80

Macroeconomic performance deteriorated during 1979. Output and export growth rates continued their decline. Inflation rates remained high. The current account deficit jumped to 2.2 percent of GNP, while external debt rose by $5.5 billion to 32.5 percent of GNP. It was a year of increasing domestic unrest. Partially in response to worsening income distribution, there were a number of demonstrations. The situation culminated in the widespread political uncertainties following the death of President Park in October. Macroeconomic policies were relatively contractionary during 1979. Money growth was kept within the CSP’s targets, and government expenditures fell relative to GNP, leading to a reduction in the fiscal deficit. The counterpart to the larger current account deficit was a jump in fixed and inventory investment. Savings remained high. As was the story during 1974-75, unanticipated slowdown of export and output growth helps to explain the inventory jump and subsequent external imbalance. A large trade deficit accounts for most of the current account deterioration. Higher import prices led to a substantial rise in the value of imports, while export receipts stagnated. Increasing real appreciation and labor costs help to explain the poor export performance. Between 1978 and 1979, the real exchange rate appreciated by 9 percent, while unit labor costs rose by 1 1 percent. Nominal wages, real wages, and labor productivity grew by 29 percent, 9 percent, and 16 percent, respectively, marking an end to the 1976-78 period of real wage gains in excess of productivity and the beginning of a period of restrained nominal wage gains. Cumulatively, unit labor costs more than doubled during 1975-79, while the exchange rate remained fixed. We look next at the declining growth rates. A simple accounting decomposition on the demand side (see ch. 7) shows that, although there was a massive (7 percent) reduction in the contribution of exports to growth between 1978 and 1979, this decline was offset by the extremely slow growth of imports. The net contribution of trade to growth remained roughly constant between 1978 and 1979. On the other hand, the drop in the growth of fixed investment was only partially offset by inventory accumulation. Total investment contributed a full 3 percent to the reduction in growth between 1978 and 1979. However, this simple approach underestimates the total effects from external developments because it ignores resulting changes in endogenous variables. Our counterfactual examples using the KDI Quarterly Macroeconomic model of the Korean economy (see ch. 5) imply that with no deterioration in external conditions (i.e., with unchanged oil prices, foreign prices, foreign growth rates, and interest rates), Korean growth would have been considerably stronger (9 percent in 1979) while the current account deficit would have been 17 percent ($0.7 billion) smaller.

202

Susan M. Collins and Won-Am Park

Nineteen eighty was a crisis year for the Korean economy. Real output declined by 4.8 percent. Inflation reached over 25 percent. The current account deficit rose to 8.7 percent of GNP. External debt jumped from 32.9 percent of GNP at the end of 1979 to 44.7 percent by the end of 1980. Again, there were both internal and external reasons for the 1980 outturn. There were two major internal developments. First, the death of President Park created a climate of political uncertainty and social unrest which is difficult to quantify. The second arose from the agricultural sector. After poor grain harvests in both 1978 and 1979, the rice crop failed in 1980. Grain imports increased substantially during this period. The sector’s contribution to total GNP growth was -3.4 percent in 1980. In contrast, agriculture’s annual contribution to growth had ranged from 0.8 to 2.3 percent during 1971-77. External factors included the terms of trade deterioration following the second oil shock (there was a 17 percent decline between 1978 and 1981), the slowdown in world economic activity, and the increased cost of servicing the external debt due to the rise in interest rates. Referring again to simulations from the KDI Quarterly model, our results suggest that if external conditions had not deteriorated, real growth would have been positive (5 percent) and the current account deficit would have been only half as large (as improvement of $2.7 billion). Three factors contributed to the inflation: devaluation, the oil price jump, and the gradual decontrol of prices. The model simulations suggest that inflation would have been about 9 percentage points lower in the absence of the unfavorable external developments. A stabilization package was initiated in January 1980, supported by a two-year IMF standby arrangement. The exchange rate was devalued by 17 percent, and at the same time, a more flexible exchange rate regime was introduced in which the won-dollar exchange rate was to be determined based on external conditions and on the value of a basket of currencies. During 1980, the (trade weighted) nominal exchange rate depreciated by 18.9 percent in nominal terms and 9.7 percent in real terms. Domestic interest rates, bank loans, and deposits were increased 5-6 percent and the higher oil prices were passed through to domestic consumer^.^ The plan also called for a tightening of monetary and fiscal policy, in the hopes of counteracting the inflationary impact of devaluaton. However, conditions deteriorated during the year. Employment and output stagnated, student demonstrations and labor unrest increased in the spring, and firms were having severe difficulties meeting their debt obligations as a result of the devaluation and the economic recession. The high debt-equity ratios contributed to the precarious financial situation. In response the government relaxed monetary and fiscal policy in a series of measures in June, September, and November. In June, interest rates were raised 1-2 percent and domestic credit was expanded, particularly to small and medium-sized businesses and to low income housing construction.

203

KoredChapter 4

Government expenditures on social services were increased, and the target money growth rates were raised slightly. The September and November measures reduced selected taxes and the interest rates on loans, and expanded credit for residential construction. 4.3.3 Early Recovery, 1981- 83 Korea had weathered the two previous crises by borrowing extensively and smoothing the adjustment instead of contracting the economy. However, policymakers were skeptical about the feasibility of this option. Their debt stock was already very large and prospects for a quick recovery of world demand for Korean exports looked dim. Instead, macroeconomic stabilization especially a reduction in inflation rates, remained the top priority. The fifth five-year plan, formulated in 1981, launched a major new stabilization effort. It gave first priority to reducing inflation. In response to dissatisfaction with the role of government intervention in the unfavorable economic performance, second priority was given to economic liberalization. As will be discussed further in chapters 9 and 11, the trade regime has since been liberalized substantially, while liberalization of domestic financial markets has proceeded more slowly. The program included a wide variety of measures. Tax reforms reduced individual income taxes, extended the value-added tax, and restructured corporate taxes, eliminating many special advantages. Price controls were eliminated. The number of restricted imports was reduced as part of the trade liberalization. Again, the actual restrictiveness of macroeconomic policies varied as a number of additional measures were undertaken during the year. In April, policy was loosened as additional credit was given to exporters and to small and medium-sized firms. In June the government tried to further stimulate construction. Interest rates were reduced by 3 percent, lagging behind the declines in inflation. The government also began to rely more heavily on incomes policy in an attempt to keep wages down. There were some improvements in the state of the economy during 1981. In particular, there was a one-year turnaround in the growth rate-the economy grew strongly at 6.6 percent. A sectoral decomposition shows that agriculture grew very strongly (contributing over 3 percent to the GNP growth rate as compared to - 3 percent in 1980), with some recovery in manufacturing. Inflation fell from 28.7 to 21.3 percent within the year. However, the current account deficit remained at nearly 7 percent of GNP and external debt had risen to 48.4 percent of GNP, with a womsome 26.1 percent of the debt being short term. Inflation remained high by historical standards. Furthermore, gross fixed investment had fallen from 31.8 percent of GNP during 1978-80 to 28.9 percent of GNP during 1981. A new policy package to revive the economy was introduced in January 1982. The interest differential on preferential loans was eliminated. The money supply was increased to stimulate investment. A financial scandal in

204

Susan M. Collins and Won-Am Park

May 1982 resulted in further credit expansion in order to bail out firms in trouble." The growth rate of M1 jumped to over 45 percent. At the same time, there was little change in the fiscal position and the real effective exchange rate appreciated by nearly 4 percent. Furthermore, world demand stagnated. Economic performance in 1982 was mixed. The growth of exports fell from 20.1 percent in 1981 to only 1 percent in 1982. As a consequence, there was a moderation of output growth. This time, neither agriculture nor manufacturing grew strongly. Instead, construction and other services were the sources of growth. External debt rose an additional 4 percent of GNP to 52.7 percent. However, there were substantial improvements in the current account and in inflation. The current account deficit declined from 6.9 to 3.7 percent of GNP. We return to the discussion of current account improvement with growth in chapter 7. Even more striking is that inflation fell from 21.3 to 7.2 percent. Three factors contributed to the large drop. The first was the sustained slowdown in nominal wage growth. Real wages had declined in both 1980 and 1982. The second was a small terms of trade improvement. The third was a real currency appreciation. Although the won depreciated against the dollar, the nominal effective exchange rate remained constant and the real effective exchange rate appreciated. By 1983 the Korean economy was performing strongly. Real growth was nearly 12 percent, while inflation had fallen below 4 percent and the current account deficit had been reduced to just 2 percent of GNP. Where did the 1983 boom come from? The simple accounting decomposition (ch. 7) shows that Korean exports, investment, and private consumption all grew strongly. The expansion was not attributable to increased government spending. The sectoral decomposition shows that expansion of manufacturing contributed nearly 4 percentage points, as compared to just 1.3 percentage points in 1982. There had been some improvement in external conditions. World growth had resumed-industrial countries grew by 2.6 percent in 1983 as compared to - 0.2 percent in 1982 and an average of 1.4 percent per year during 1980-81. Increased international competitiveness enabled Korea to take advantage of the stronger world demand. Further nominal exchange rate adjustment had led to over 10 percent additional real depreciation since 1980. Domestic wage growth had also slowed. Despite a slowdown in labor productivity during the early 1980s, unit labor costs measured in dollars declined by 16.6 percent over 1979-83. During the same period, (dollar) unit labor costs remained constant for Hong Kong and rose by 28 percent for Taiwan. l 1 Other internal factors had also improved. Many of the controls and restrictions introduced during the 1970s had been relaxed. Agricultural output had revived. In addition, the social and political climate had eased considerably.

205

KoredChapter 4

By 1983 Korea had dealt with the major economic difficulties from 1979-80. Furthermore, macroeconomic stabilization had been achieved without compromising high rates of capital formation. Investment had remained strong throughout 1980-82 even though domestic savings did not begin to recover until 1983. A critical point here is that Korea was able to continue to borrow from abroad during its crisis period and these funds were used to maintain investment. It is very unlikely that Korea would have had this option if the crisis years had been 1982-83. Korea was lucky to run into difficulty before most of the other debtor countries. 4.3.4

Successful Adjustment, 1983-86

As a result of the very favorable 1983 economic performance, Korean policy shifted away from a focus on short-run macroeconomic stabilization (prices and the balance of payments), turning again to issues of long-run structural development. The point is important in contrasting Korea’s experience with that of other developing country debtors. For most of them, 1983 was the beginning of the crisis. For Korea, the major adjustments had already been accomplished. The government launched a revision of the fifth five-year plan, to be in effect from 1984 to 1986. In the revised plan it was explained that the economy had already achieved the major goals of price stability and renewed export and output growth, as set forth in the original plan. The revision, ‘‘rather than being oriented to quantitative targets, emphasizes institutional reforms and structural improvements ...to make a major shift in the style of economic management toward relying more on competition and market mechanism and to solve the problems of imbalance” (Government of Korea 1983, iii, 3). The revised plan very clearly shows the policy shift to structural adjustment and long-term growth. For example, it states that Korea’s “remarkable [ 1980-831 performance has laid the foundation for another economic takeoff” and that Korea was “forging ahead towards joining the ranks of advanced industrial countries” (3). Against this backdrop, both monetary and fiscal policy were tightened significantly in conjunction with a new IMF program, in effect from July 1983 through March 1985. The fiscal deficit was reduced from 4.3 percent of GP in 1982 to 1.6 percent in 1983. M I growth was slowed to 17.0 percent during 1983 and 0.5 percent during 1984. The nominal exchange rate was managed so as to depreciate the won by 5.7 percent in real terms from 1982 to 1984. Economic performance remained strong in 1984. Growth exceeded 8 percent. The current account improved further as domestic savings rose. Inflation fell below 3 percent. In 1985 the real growth rate slowed to 5 percent. This development is partially attributable to a slowdown in world economic activity. The dollar value of Korean exports grew by just 4 percent and exports contributed only 1 percent to GNP growth, compared to 4 percent in 1984 and 6 percent in

206

Susan M. Collins and Won-Am Park

1983. However, inflation rates remained low and the current account continued to improve. Korea’s debt position also improved. Short-term debt, as a share of total debt, declined from 26 percent in 1981 to 19 percent in 1985, and the ratio of debt service to exports dropped from 57 percent in 1982 to 49 percent in 1985. The government initiated further depreciation of the won in order to bolster Korea’s competitiveness. In real terms, the won depreciated by 6 percent during 1985 and by an additional 15 percent in 1986. Nineteen eighty-six was a banner year for the Korean economy. Real growth reached 12.5 percent, inflation remained at just 2.3 percent and the current account registered a $4.6 billion surplus (nearly 5 percent of GNP).’* In stark contrast to most of the other debtor countries which experienced further deterioration in their debt indicators, l 3 Korea’s debt to GNP ratio fell from 56.3 to 46.8 percent as its debt stock was reduced by $2.25 billion. Strong growth in the industrial countries, lower interest rates, a dramatic terms of trade improvement (primarily from the drop in oil prices), and the substantial real depreciation all contributed to the impressive performance. Korea’s adjustment has been extremely successful on the macroeconomic stabilization front. The balance of payments, inflation, growth, and the debt burden have all improved dramatically since, 1979-81. In the following chapters, we turn from a chronological analysis to an examination of individual pieces of the performance. These pieces are synthesized and our main conclusions are summarized in the final chapter.

5

Internal versus External Shocks

AS described in chapter 4, Korea experienced large current account deficits, slowdowns in growth, and rapid accumulation of external debt during 1974-77 and again during 1979-83. In both periods, the poor performance coincided with internal as well as external developments. This chapter evaluates the relative importance of internal versus external factors in explaining the current account imbalances during each of these periods. Our analysis draws from two approaches. The first begins with the current account identity and decomposes the change in the current account from a base year into price, income, interest rate, and other effects. This approach does not take into account shifts in behavior of domestic residents (importers, monetary authorities, etc.). Our second decomposition, based on the KDI Quarterly Macroeconomic model, incorporates a more fully specified set of behavioral relationships. The basic characteristics of the

207

KoredChapter 5

model are summarized at the end of this chapter. The model is described in more detail in W. A. Park (1986).

5.1 Current Account Performance, 1974-77 During 1972-73, the average current account deficit was just 2.9 percent of GNP. In 1974 and 1975 this figure jumped to 10.8 percent and 9.1 percent, respectively (see table 4.2). As discussed in chapter 4, internal as well as external developments seem to have contributed to the deterioration. On the external side, there was the dramatic rise in oil and commodity prices. On the internal side, Korea was beginning the Big Push to develop HC industries. The massive investments called for increased imports of intermediates and capital goods. Thus, an interesting question is how much of the larger current account deficit can be attributed to the external terms of trade shock. In a very provocative analysis of this question, Y. C. Park (198%) argues that the terms of trade deterioration was not the most important factor. He finds that increased nonoil imports were almost twice as important. Because many of these imports are attributable to the Big Push, he concludes that the internal policy shift significantly outweighed external factors in explaining the poor current account outcome. To support this view, Park decomposes the current account deterioration in each year, 1974 to 1977, using 1972-73 as a base. His components are world interest rates, import and export price changes, import and export volume changes, and a (domestic) aggregate demand component. Import price and volume are further decomposed into oil, capital goods, and other imports. His results are reproduced in table 5.1. He finds that the deterioration associated with the terms of trade loss . . . amounted to an increase of 5 percentage points in the current account/potential GNP ratio . . . however, the sum of the expansion of capital goods in relation to fixed investment, and other imports, excluding oil, relative to GNP was the main element producing imbalance in the current account. This jump, equivalent to a deterioration of 10 percentage points, was larger than the actual increase in the deficit ratio. A similar development took place during 1975. (302) However, Park’s results overestimate the contribution of increased nonoil import volume effects and underestimate the contribution of external price developments. The difficulty arises from the price-volume decomposition of imports. In particular, Park’s analysis requires indices for unit value and volume for capital goods and “other” imports on a balance of payments basis. Proxies for these series are unreliable and potentially misleading. To make the point, we present an alternative decomposition. We follow Park’s basic procedure, but divide imports into oil, nonoil goods, nonfactor services.2 This decomposition enables us to use a more reliable data series,

’

Susan M. Collins and Won-Am Park

208

Table 5.1

Current AccountIPotential GNP Ratio, 1974-77 (base period 1972-73) ~~~

Item

1974

I . Current account imbalances potential output (actual change)

2. Terms of trade effect Import price Capital goods Oil Other Export price 3. Interest rate effect

~~

~~

1975

I976

1977

7.258

5.351

- 1.782

-3.478

4.893 -0.462 - 1.816 3.064

4.284 - 1.573 - 1.621 3.302 -3.255 5.858

1.183 -6.772 -2.605 2.954 -7.121 7.956

-0.254 - I I .495 - 3.048 2.608 - 11.055 11.242 0.040

-

1.710

5.355 0.167

0.458

0.023

4. Accumulated debt effect

- 0.142

0.134

0.210

0.170

5 Import replacement Capital goods Noncapital goods Oil conservation efforts

9.528 3.385 6.358 -0.215

7.621 2.150 6.312 0. I59

10.727 2.742 7.546 0.439

14.769 1.907 12.364 0.498

6. Export promotion Construction services

-7.883 -0.031

-7.791 -0.098

-

15.932

- 1.632

-21.667 -3.954 4.067 2.463 1.604

7. Aggregate demand adjustment Fixed investment Domestic output

1.183 0.680 0.503

1.123 0.736 0.387

2.532 1.292 1.241

8. Total effect (2 to 7)

7.747

5.829

- 1.256

-2.874

-0.526

-0.602

9. Interaction effects and adding-up errors [ ( l ) through (8)]

- 0.489

-0.478

Source: Y . C. Park (198%. table 11.8).

Nore: The dccompositjon factors were calculated by using an average of cument ycar and base period weights. A negative sign indicates a balance of payments improvement.

the BOK’s unit value index for commodity imports. We also use the Saudi Arabian petroleum price index reported by the IMF. The results are shown in table 5.2. Because we have used revised National Accounts data, we obtain a somewhat different base deficit to potential GNP (row 1). However, our primary interest is in the share of the additional deficit attributable to terms of trade changes. Using our decomposition, it is clear that the terms of trade deterioration is the predominant factor explaining the 1974-75 current account imbalance. The rise in oil and commodity prices accounts for 90 percent of the imbalance in 1974 and over 100 percent of the imbalance in 1975. The impact of the import volume changes is quite small in 1974 (6 percent contribution) and in fact contributes to an improvement in the 1975 current account equivalent to 14 percent of the imbalance. To further investigate the impact of the oil price rise and other external shocks, we turn next to the implications of the KDI Quarterly Macroeconomic model. Unlike the simple decompositions reported above, the model allows us to incorporate endogenous changes in behavior, e.g., changes in domestic prices, output, and investment as a result of exogenous shocks.

KoredChapter 5

209

Table 5.2

Current Account/Potential Nonagricultural G N P Ratio, 1974-77 (base period 1972-73)

Item

1974

1975

1976

1977

1. Current account deficitlpotential

nonagricultural GNP 2. Terms of trade effect

lmport Price Oil Nonoil goods Nonfactor services Export price

3. Import replacement Oil Nonoil goods Nonfactor services

10.04

7.41

-2.29

-3.71

7.01 9.17 5.46 3.58 0.13 - 2.16

10.05 8.54 5.67 2.61 0.26 1.52

4.77 -0.54 4.88 -5.27 -0.15 5.31

1.85 -6.03 4.37 - 10.02 -0.38 7.89

0.64 -0.41 0.81 0.23

- 1.07

2.69 -0.48 2.00 1.18

3.08 -0.53 1.94 1.67

-2.04

-4.07

- 13.96

- 18.27

4.42 4.25 0.17

9.16 8.80 0.36

-0.66 - 1.56 1.15

4. Export promotion of goods and

nonfactor services 5 . Aggregate demand adjustment

Fixed investment Domestic output 6. Interest rate effect 7. Accumulated debt effect

2.79 2.76 0.03

1.57 1.74 -0.17

I .52

-0.45

-0.93

0.06

8. Exports of factor services

0.53

0.62

9. Net transfers

0.41

0.58

10. Total effect (2 to 9)

9.92

7.29

11. lnteraction effects and adding-up errors

0.12

0.12

-0.71 0.36 -0.33 0.38 -2.40 0.11

1.oo 0.28 - 2.02

1.15 -3.78 0.07

Source: Authors’ calculations. See text. Nore: The decomposition factors were calculated by using an average of current year and base period

weights. A negative sign indicates a balance of payments improvement.

The model allows us to simulate the behavior of the current account under alternative assumptions about the paths of exogenous variables. We then compare these counterfactual paths with the actual performance. Of course, our simulations cannot tell us what policymakers would have done in the absence of external shocks, or how the U.S. and Japanese economies might have performed differently. The exercises cannot fully disentangle the role of policy adjustment and “luck,” however, they do provide measures of the effect of key external variables. We begin with the following counterfactual exercise (exercise A). Taking the paths of other exogenous variables as given, how would Korean economic performance have been different from the actual experience if oil prices had remained fixed at their 1972-73 average level? The results are reported in tables 5.3 and 5.4. In table 5.3 we examine the current account imbalance, providing a similar decomposition to that in table 5.2. For each year, the table gives the estimated outcome and, in parenthesis, the difference between the actual and the counterfactual estimate. Additional

Susan M. Collins and Won-Am Park

210

Table 5.3

Current Account/Potential Nonagricultural GNP with Fixed Oil Prices

Item

1974

1975

1976

1977

6.04 (-4.00)

3.33 (-4.08)

-6.05 (-3.76)

-6.56 (-2.84)

3.19 (-3.82) 2.74 ( - 6.43) -0.64 (-6.10) 3.57 (-0.00) -0.19 (-0.33) 0.44 (2.60)

6.25 (-3.80) 2.21 (-6.33) -0.69 (-6.35) 3.12 (0.51) -0.22 (-0.48) 4.04 (2.52)

6.08 (-4.09) -7.19 (-6.66) -1.07 (-5.95) - 5.44 (-0.17) -0.69 (-0.54) 7.87 (2.57)

1-10( - 1.73) -0.22 (0.18) - 1.36 (-2.17) 0.49 (0.25)

-4.00 (-2.93) -0.37 (0.30) -5.28 (-3.71) 1.64 (0.49)

1.29 (-1.40) -0.31 (0.18) -0.41 (-2.40) 2.00 (0.82)

-3.45 ( - 1.41)

--7.84 (-3.77)

1 . Current account deficitlpotential

nonagricultural GNP 2. Terms of trade effect Import price Oil Nonoil goods Nonfactor services Export price 3. Import replacement Oil Nonoil goods Nonfactor services

-

4. Export promotion of goods and nonfactor services 5. Aggregate demand adjustment Fixed investment Domestic output

6. Interest rate effect

5.74 (2.94) 5.28 (2.52) 0.46 (0.43)

7.90 (6.33) 7.08 (5.34) 0.82 (0.99)

- 18.24 (-4.28)

10.25 (5.84) 9.15 (4.90) 1.10 (0.94) 0.45 (1.16)

2.65 (1.13)

0.22 (0.67)

-0.60 (0.37)

0.45 (0.39)

8. Exports of factor services

0.53 (0.00)

0.62 (0.00)

-0.31 (0.04)

9. Net transfers

0.41 (0.01)

0.59 (0.01)

0.43 (0.05)

7. Accumulated debt effect

10. Total effect (2 to 9)

-0.53 (-0.89)

-2.44 (-4.29) - 13.10 ( - 7.07) - 1.29 (-5.66) - 10.76 (-0.74)

(-0.67) 10.66 (2.77)

- 1.06

3.07 (-0.01) -0.30 (0.23) 0.44 ( - I .50) 2.93 ( I .26) -20.43 (-2.16) 12.70 (3.54) 11.72 (2.92) 0.99 (0.62) 2.84 ( I .85) -1.61 (-1.89) -1.84(0.18) 1.20 (0.06)

7.42 ( - 2.50)

4.19 (-3.10)

-5.97 (-3.57)

-6.51 (-2.73)

- 1.38 ( - 1.50)

-0.86 (-0.98)

-0.08 (-0.18)

-0.05 (-0.11)

11. Interaction, adding-up, and

simulation errors

Nore: Using the KDI Quarterly Macroeconomic model, the counterfactual fixes oil prices at their (average) 1972-73 level. The decomposition factors were calculated by using an average of current year and base period weights. A negative sign indicates a balance of payments improvement. Numbers in parentheses indicate the difference between the actual and counterfactual values.

Table 5.4

Macroeconomic Performance: Fixed Oil Prices versus Actual

Item Real GNP growth Real fixed investment (IF) growth WPI inflation CPI inflation Exports Imports Oil Nonoil Trade balance Exports of nonfactor services Imports of nonfactors services Imports of factor services Current balance

1974

1975

I976

1977

5.4 9.4 - 13.6 - 8.6

7.1 11.8 -7.0 -5.0

1.8 0.7 0.8 1.3

-2.0 -5.9 2.4 2.0

-0. I -0.8 -0.9 0.1 0.7 -0.1 0.0 0.0 0.6

0.3 -0.6 -1.0 0.4 0.9 -0.0 0. I 0.1 0.7

0.7 -0.5 - 1.3 0.8 1.2 - 0.0 0.2 0. I 0.9

0.7 -0.7 - 1.5 0.8 I .4 -0.0 0.3 0.2 0.9

Nore: These figures are deviations between the baseline path and a counterfactual in which oil prices are fixed

at their (average) 1972-73 level. The top panel gives percentage deviations, while the bottom panel gives billions of U.S. dollars. The KDI Quarterly Macroeconomic model is used for simulations.

211

KoredChapter 5

comparisons of the actual and the counterfactual performance are given in table 5.4. The top panel of that table shows the difference in GNP growth, investment growth, and inflation for each year. The bottom panel shows the absolute difference (in billions of U.S. dollars) for various components of the current account. The tables imply that the 1974 current account deficit would have been only 60 percent as large as it actually was if oil prices had been fixed. The improvement amounts to $0.6 billion, or about 4 percent of potential nonagricultural GNP. In contrast, the simple accounting decomposition in table 5.2 estimates that the oil shock increased the current account deficit by 5.5 percent of potential GNP. In fact, the accounting decomposition suggests a larger role for the oil price rise for every year during 1974-77 than is suggested from the model simulations. The model estimates an impact of 4.1, 3.8, and 2.8 percent of potential GNP during 1975, 1976, and 1977, respectively, as compared to 5.7, 4.9, and 4.4 percent from the accounting decompositions. In the model the improvement from stronger terms of trade is partially offset by endogenous changes in growth, inflation, investment, and other domestic variables. The key factors explaining the results from the model are as follows. Lower oil prices would have led to lower domestic inflation and lower prices of domestic exports. They would also have led to faster domestic growth, with especially strong effects on domestic investment (table 5.4). These factors have conflicting effects on the external balance. A decline in export prices decreases the dollar value of exports (holding export volumes fixed). This channel worsens the current account relative to the base by 2.6 percent of potential GNP. The lower oil prices would also have led to an increase in export volume, tending to improve the current account (table 5.3, row 4). The aggregate demand expansion is estimated to contribute an additional 3 percent of potential GNP to the current account deficit. This is partially offset by substitution effects from the decline in domestic prices on nonoil import demand (table 5.3, row 3). The results from a second counterfactual exercise (exercise B) are reported in table 5.5. This exercise provides a rough measure of the overall impact from external shocks during 1974-77. Here the world interest rate is assumed to be fixed at its 1973 level. Oil prices, a weighted index of real GNP of Korea’s major trading partners, and a weighted index of foreign prices are all assumed to increase at their three-year average rate of increase during 1970-73. We compare the results in tables 5.4 and 5.5 so as to discuss the additional impact of external factors other than oil prices. The tables show that growth rates in 1974 would have been substantially higher under B than with just the fixed oil prices of A. However, there is little difference in the inflation rates in the two cases. Furthermore, the current account improves by only an additional 17 percent. (Recall that the fixed oil price in A led to a 40 percent

212

Table 5.5

Susan M. Collins and Won-Am Park Macroeconomic Performance: Fixed External Conditions versus Actual

Item Real GNP growth Real fixed investment (IF) growth WPI inflation CPI inflation Exports Imports 011

Nonoil Trade balance Exports of nonfactor services Imports of nonfactor services Imports of factor services Current balance

I974

1975

1976

1977

8.9 14.7 - 13.2 - 8.2

7.3 11.7 -4.4 -2.4

2. I 0.2 3.9 4. I

1.2 -1 6

0.3 -0.4 -0.7 0.3 0.7 -0.0 0.0 -0.0 0.7

I .0 0. I -0.7 0.8 0.9 0. I 0. I 0. I 0.8

2.5 0.7 -0.7 I .4 I .7 0.2 0.2 0.3 1.4

3.9 I .4 -0.7 2.I 2.5 0.5 0.5 0.5 2.0

3.4

3.3

oil prices, foreign GNP. and foreign prices are assumed to increase at the three-year average rate prior to the oil shock. The world interest rate is assumed fixed at the preshock level. The KDI Quarterly Macroeconomic model is used for simulations. Data in the top panel and percentages; in the bottom panel, billions of U.S. dollars. Nore: These figures are deviations between the baseline and a counterfactual in which

current account improvement relative to the actual outcome.) By 1975 growth rates are nearly the same under the two scenarios, with B implying a somewhat higher inflation and a 14 percent current account improvement relative to A. While the additional external factors had a relatively small impact on the current account during 1974-75, the simulations suggest that there would have been strong benefits by 1976 from an external environment in which there was continued growth by Korea’s trading partners. The simulations estimate substantial additional current account improvement in B compared to A during 1976-77

5.2 Current Account Performance, 1979-83 By 1977 the current account had improved substantially (see table 4.2). However, there was a renewed deterioration during the 1979-81 economic crisis. Again, our discussion in chapter 4 identified both internal and external factors which contributed to this outcome. External factors included the second oil price shock as well as increased world interest rates and a slowdown in world growth. Internal factors included the death of President Park, the associated social and political turmoil, and the disastrous agricultural harvests. Again we begin with simple accounting decompositions of the current account. We present both the decompositions from Y. C. Park (198%) and our revised version. The revision decomposes imports into oil, nonoil commodities, and nonfactor services instead of oil, capital goods, and other imports. As before, our decomposition enables us to use more reliable import value deflators to separate changes in value from changes in volume.

213

KoredChapter 5

In contrast to the 1974-75 episode, Park concludes that external price developments were the most important cause of the current account deterioration after the second oil shock. As shown in table 5.6, he finds that the increased prices of oil and capital goods can more than explain the external deficits. The terms of trade effects worsen during 1981 before improving somewhat in 1982-83. In fact, Park finds that if oil prices, interest rates, and construction service exports had remained at their 1978 levels, Korea would have run a substantial current account deficit during 1979, a small deficit (less than 1 percent of potential output) during 1980, and surpluses during 1981-83. Our results (table 5.7) also point to the critical role of terms of trade changes, particularly the oil price rise, in explaining the current account deterioration. However, our decomposition implies a smaller role for external price changes and a larger role, especially after 1981, for “import replacement” or the growing import volumes. Both decompositions identify poor export performance as the primary reason for the external imbalance in 1979. Both decompositions attribute the

Table 5.6

Current Aceount/Potential GNP Ratio, 1979-83 (base period 1977-78)

Item I . Current account imbalances potential output (actual change)

2. Terms of trade effect Import price Capital goods Oil Other Export price

1979

1980

1981

4.838

6.099

4.398

1.675

0.607

- 1.581

6.186 6.490 -0.645 3.892 3.242 -0.303

8.014 8.666 -0.331 4.472 4.524 -0.651

4.860 3.719 -0.139 3.973 -0.114 1.141

4.015 3.142 0.216 3.274 -0.348 0.873

- 2.846 - 1.507 0.389

- 1.728 1.265

1982

1983

0.586

1.308

1.746

1.534

0.945

4. Accumulated debt effect

- 0.324

0.082

0.457

0.653

0.886

5. Import replacement Capital goods Noncapital goods Oil conservation efforts

2. I43 1.003 1.666 -0.526

-1.192 0.644 - 1.095 -0.741

-0.724 -0.246 0.310 -0.788

-0.842 0.061 0.271 -1.174

2.753 1.212

2.216 0.986

0.123 1.674

3. Interest rate effect

6. Export promotion Construction services 7. Aggregate demand adjustment Fixed investment Domestic output

5.106 0.789

- 1.213 0.228 - 1.442

-0.641 - 0.278

-0.284 -0.079 4.309 1.464 -5.269 -1.154 -4.114

-7.538 -2.281 -5.257

-7.012 -1.778 -5.234

-4.771 -0.981 -3.790

8. Total effect (2 to 7)

4.715

5.977

4.240

1.527

0.455

9. Interaction effects and adding-up errors [ ( l ) through (8)]

0.123

0.122

0.158

0.148

0. I52

Source: Y. C. Park (1985~.table 11.11) Nore: The decomposition factors were calculated by using an average of current year and base period

weights. A negative sign indicates a balance of payments improvement.

214

Susan M. Collins and Won-Am Park

Table 5.7

Current Accouut/Po&entialNonagricultural GNP Ratio, 1979-83

Item

1979

1980

1981

1982

1983

6.42

7.72

5.79

2.48

0.86

2. Terms of trade effect Import price Oil Nonoil goods Nonfactor services Export price

-0.55 -2.32 0. I3 -2.12 -0.32 1 .I7

5.71 5.45 4.24 0.98 0.23 0.26

6.48 6.41 5.03 0.92 0.47 0.07

4.36 2.92 5.01 -2.45 0.36 I .44

3.41 1.25 3.88 -2.51 -0.12 2.16

3. Import replacement Oil Nonoil goods Nonfactor services

0.49 -0.20 0.20 0.49

1.85 0.43 0.32 1.10

6.01 0.11 4.01 1.89

3.38 -0.91 2.91 1.38

2.18 -1.40 2.16 1.42

1. Current account deficivpotential GNP

4. Expert promotion of goods and nonfactor services 5. Aggregate demand adjustment Fixed investment Domestic output

4.32

0.98

-1.20

6.38 -4.69 - 1.69

-10.16 -7.97 -2.18

-8.06 -5.95 -2.11

-4.83 -3.47 -1.37

-0.71

-1.79

-3.18

5.46 0.98 1.30 -0.32

-

1.31

0.49

0.47

- 1.30

0.09

I .42

2.22

8. Exports of factor services

0.46

1.21

I .oo

0.85

1.34

9. Net transfers

0.16

0.22

0.21

0.26

0. I6

6. Interest rate effect 7. Accumulated debt effect

2.71

10. Total effect (2 to 9)

6.18

7.48

5.56

2.20

0.59

11. Interaction effects and adding up errors

0.25

0.24

0.23

0.28

0.27

Source: Authors’ calculations. See text Nore: The decomposition factors were calculated by using an average of current year and base period weights. A negative sign indicates a balance of payments improvement.

outcome in 1980 to the terms of trade, poor export performance, and higher interest rates, partially offset by the drop in domestic aggregate demand. Our decomposition suggests that the terms of trade shock accounts for 74 percent of the added external imbalance, compared to over 100 percent in Park’s decomposition. Similarly, the nonoil commodity price deflator (which we feel to be more reliable) does not show the same strong increase as the nonoil, non-capital goods deflator during 1981-83. Consequently, our decomposition suggests that growing import volumes contributed nearly as much to 1981-83 current account deficits as did high import prices. It is interesting to compare the 1974-77 experience (table 5.2) with the 1979-83 experience (table 5.7). The major difference is that the current account deficit was more persistent in the latter period. The current account imbalance was reversed two years after the onset of large deficits in 1974, but only four to five years after the onset in 1979. Two factors help to explain this difference. First the terms of trade shock was initially larger but less persistent in the first episode. Second, strong

215

KoreaXhapter 5

export performance during 1974-77 contributed to current account improvement. In contrast, export volumes contributed to the external deficits during 1979-82. Reasons for the poor export performance include the deterioration in Korea’s external competitiveness and weak world demand. However, these two factors were partially offset by the slow Korean growth (aggregate demand effect) and by the deceleration in the growth of fixed investment as the economy pulled back from the Big Push during the early 1980s. Again, we use the KDI Quarterly Macroeconomic model to further examine the impact of the oil price increases. Tables 5.8 and 5.9 present the results from a counterfactual simulation holding oil prices fixed at their 1978 level (exercise A). As before, the figures in parentheses show the difference between the actual outcome and simulated value. Table 5.8 shows the importance of the oil price path very clearly. Holding oil prices fixed would have resulted in a 14 percent improvement in the current account as a percentage of potential GNP in 1979, a 46 percent improvement in 1980, a 68 percent improvement in 1981, and small current account surpluses in 1982-83. As before, the model suggests a smaller role for the price of oil when behavioral relationships are taken into account. The simulations imply that, as a percentage of potential GNP, oil price changes accounted for a 3.6 percent current account deficit during 1980, and 4.0, 2.6, and 1.6 percent deficits during 1981, 1982, and 1983, respectively. The comparable figures from the accounting decompositions were 4.2, 5.0, 5.0, and 3.9 percent. The reasons for this difference are precisely the same during 1980-83 as they were during 1974-76. Without the rise in oil prices, Korean growth would have been faster and investment would have been higher, tending to raise nonoil imports. Finally, we use the model to simulate the economic performance assuming an unchanged overall external environment (exercise B). Interest rates are fixed, while the average growth rates of oil prices, foreign GNP, and foreign prices are assumed equal to the average growth rates during 1976-78. Comparing the results from B (table 5.10) with those from A (table 5.9), and those from the 1974-77 experiments, it is clear that nonoil external factors were more important during 1979-82 than 1974-77. Over the 1979-81 period, B implied an improvement in the current account deficit of 34 percent more than A. The major reason for the improvement comes from the considerably stronger export Performance that can occur when world demand continues to grow. Stronger exports also contribute to more rapid domestic growth rates.

5.3 The KDI Quarterly Macroeconomic Model We would like to summarize here the key features of the KDI Quarterly Macroeconomic model developed by Won-Am Park. Additional information

Table 5.8

Current AccounUPotential Nonagricultural GNP with Fixed Oil Prices 1979

I . Current account deficitlpotential nonagricultural CNP

5.50 (-0.93)

2. Terms of trade effect Import price Oil Nonoil goods Nonfactor services Export price

- I .50 ( - 0.95)

3. import Replacement Oil Nonoil goods Nonfactor services

0.37 (-0.12) -0.15 (0.05) -0.07 (-0.27) 0.58 (0.09)

-3.84

1.52) 1.55) -1.98 (0.14) -0.43 (-0.11) 2.34 (0.57) (-

- 1.43 ( -

1980

1981

1982

1983

1.85 (-3.94)

-0.08 (-2.55)

-0.72

2.42 (-3.29) 0.08 (-5.37) - 1.35 (-5.59) 1.71 (0.73) -0.28 (-0.51) 2.34 (2.08)

2.21 (-4.27) 0.28 (-6.13) -1.45(-6.47) 1.94 (1.02) -0.21 (-0.68) 1.93 (1.86)

-0.49 (-4.85 -4.15 (-7.08) -1.46(-6.47) -2.32 (0.13) -0.37 (-0.73) 3.66 (2.23)

-0.15 (-4.46) -5.74 (-6.99) -1.46 (-5.34) -3.26 (-0.75) -1.02 (-0.90) 4.69 (2.53)

0.16(-1.69)

3.73 ( - 2.08) -0.25 (-0.36) 1.28 (-2.73) 2.70 (0.82)

3.27 (-0.11) -0.81 (0.10) 1.34 ( - 1.56) 2.73 (1.36)

4. I4

(-

3.57)

0.10 (-0.33) - 1.52 ( -

1.84) 1.58 (0.48)

4. Export promotion of goods and nonfactor services

5.13 (-0.33)

1.95 (-2.37)

5. Aggregate demand adjustment Fixed investment Domestic output

1.42 (0.44) 1.60 (0.30) -0.18 (0.14)

-3.00 (3.38) -2.31 (2.38) -0.69 (1.00)

-3.37 (-4.68) -3.28 (6.88) -3.07 (4.90) -0.21 (1.98)

-2.71 (-3.69)

(-

1.58)

3.99 (1.81)

- 1.06 (0.34) 1.71 (-0.45) 3.34 (1.92) - 3. I 2 ( - 1.92)

-2 17 (5.89) - 1.86 (4.08) -0.30 (1.81)

- 1.88 (2.95) - 1.50 (1.97) -0.38 (0.98)

1.32 (0.84)

0.87 (0.39)

-0.09 (0.61)

-0.78 (1.00)

-1.93 (1.25)

7. Accumulated debt effect

-0.57 (0.73)

0.90 (0.81)

0.86 (-0.56)

0.67 ( - 1.55)

0.33 ( - 2.38)

8. Exports of factor services

0.45 (-0.01)

1.16 (-0.05)

0.96 ( - 0.04)

0.93 (0.08)

1.50 (0.16)

9. Net transfers

0.16 (-0.00)

0.20 ( - 0.02)

0.20 (-0.01)

0.28 (0.02)

0.22 (0.06)

6.77 (0.59)

4.65 (-2.83)

1.22 (-4.34)

- 1.27 ( - 1.52)

-0.51 (-0.75)

6. Interest rate effect

10. Total effect (2 to 9) I I . Interaction, adding-up, and simulation errors

0.63 (0.40)

-1.00(-3.20) 0.92 (0.64)

-

1.95 (-2.54) 1.23 (0.96)

Nore: Using the KDI Quarterly Macroeconomic model, the counterfactual fixes oil prices at their (average) 1977-78 level. The decomposition

factos were calculated by using an average of current year and base period weights. A negative sign indicates a balance of payments improvement. Numbers in parentheses indicate the difference between the actual and counterfactual values.

217

KoredChapter 5

Table 5.9

Macroeconomic firformanee: Fixed Oil Prices versus Actual

Real GNP growth Real fixed investment (IF) growth WPI inflation CPI inflation Exports Imports Oil Nonoil Trade balance Exports of nonfactor services Imports of nonfactor services Imports of factor services Current balance

I979

1980

1981

1982

1.2 1.3 - 2.7 - 2.0

6.4 7.3 - 13.0 - 9.0

8.1 10.4 -6.3 -4.2

-0.7 I .o 1.7

0.0 -2.6 -3.0 0.4 2.6 -0.2 0.1 0.1 2.2

I .5 - 1.8 -3.6 1.8 3.3 -0.0 0.5 0.2 2.6

-0.1 -0.7 -0.7 0.0 0.6 -0.1

0.0 0.0 0.5

1.1

I-.

r

- 1.1

-3.5 2.3 2.8 0. I 1 .o

0.2 I .7

Nofe: These figures are deviations from the baseline path and a counterfactual in which oil prices are fixed at their (average) 1977-78 level. The top panel gives percentage deviations, while the bottom panel gives billions of U.S. dollars. The KDI Quarterly Macroeconomic model is used for simulations.

%ble 5.10

Counterfactual Analysis on External Shock 1979

I980

1981

1982

Real GNP growth Real fixed investment (IF) growth WPI inflation CPI Inflation

2.4 2.3 -2.5 - 1.8

10.1 11.3 - 13.7 -9.3

9.1 11.9 -5.1 -2.8

12.3 10.8 0.8 2.5

Exports Imports Oil Nonoil Trade balance Exports of nonfactor services Imports of nonfactor services Imports of factor services Current balance

0.2 -0.3 - 0.6 0.3 0.5 - 0.0 0.0 - 0.2 0.7

1.3

4.0 0.6 - 2.9 3.5 3.4 0.4 0.7 -0.7 3.7

11.1 4.7 -2.3 7.0 6.4 I .6 1.7 -0.0 6.3

- 1.1

-2.6 1.4 2.4 0.0 0.2 -0.5 2.7

Nore: These figures are deviations from the baseline and a counterfactual in which oil prices, foreign GNP, and foreign prices are assumed to increase at the three-year average rate prior to the oil shock. The world interest rate is assumed fixed at the preshock level. The KDI Quarterly Macroeconomic model is used for simulations. Data in the top panel are percentages; in the bottom panel, billions of U.S. dollars.

about the model, including the actual equations and the estimation results, are available on request. A major focus of the model is to interrelate real and financial sectors of the Korean economy. Thus, the model incorporates credit availability to firms for investment, includes money as a determinant of consumption, and emphasizes links between the monetary sector and the balance of payments. The model has been estimated using quarterly data over 1972:I to 1985:IV. Seasonal dummies were included in the regressions and, where appropriate,

218

Susan M. Collins and Won-Am Park

the estimation was corrected for serial correlation. The model consists of six blocks of equations: GNP, government sector, labor market, wages and prices, balance of payments, and financial sector. Real gross national expenditure is composed of private consumption expenditure, private fixed investment, inventory investment, government expenditure, exports and imports of commodities and nonfactor services, and net factor income from abroad. Real GNP is divided into two components: production from agriculture, forestry, and fisheries, and other production. The supply and demand for money are determined in the financial block, where interest rates in the unorganized money market adjust to equilibrate the market. The overall balance of payments and the government budget deficit are both linked to the money supply. Prices are subject to both demand-push and cost-pull factors. Wholesale prices are determined by firm’s production costs. The unit value index for exports in dollar terms is assumed to be influenced by world demand for Korean exports as well as by export production costs (wages and intermediate input costs). Import unit values are determined by the prices of capital goods imports and raw materials, including oil. The wage equation is an expectations-augmented Phillips curve. Finally, the unemployment rate is determined by the gap between potential and actual output, a variant of Okun’s law.

6

Introduction to Part Two

Korea’s macroeconomic performance, with its three cycles of debt accumulation and recovery, presents a number of puzzles which will be examined in the remaining chapters. Thus, in summarizing the experience (particularly during 1979-85) described in the first part of our study, we will introduce part 2 (ch. 7-12). The first puzzle is how Korea has managed to consistently maintain such high growth rates. Certainly its rapid growth rates for output and exports have helped to hold in check the burden of external debt. A related issue is how Korea was able to achieve a substantial improvement in the current account while output was growing strongly. In practice, most debtor countries have improved their current accounts through a domestic recession which cuts imports. Improvement with growth is a much more palatable option. Another puzzle arises from the large fluctuations in domestic savings. How was Korea able to increase saving rates so dramatically from the

219

KoredChapter 7

mid-1960s to the mid-I970s? Why have there been periodic drops in saving rates, and what has determined the rates of investment? These questions are important because each of the three debt crises was precipitated by a large drop in domestic savings relative to investment and because most of Korea’s external debt has been used to finance the savingshvestment gap. A third puzzle is why Korea was able to combine a real depreciation with improved competitiveness and an increase in real wages. Policymakers elsewhere often resist devaluation precisely because they expect it to reduce real wages and the standard of living. We shall argue that one of the most interesting and important aspects of Korean development has been the interrelationship between exchange rates, wages, and labor productivity. In addition to these three puzzles about macroeconomic performance, our study will consider the distribution of income to Korea. And finally, we will look at the role of domestic policy. How did exchange rate, trade, and industrial policies influence growth and external balance? How did monetary and fiscal policies contribute to economic performance and to the accumulation of external debt? One view is that “Korea’s experience following the second wave of oil price increases is an excellent example of how orthodox stabilization policies, effectively implemented, can help a country adjust to external shocks” (Aghevli and Marquez-Ruarte 1985, 1). An alternative view is that “domestic stabilization measures were at best a way of muddling through, and contributed little to improving the current account during 1982-83 (Y. C. Park 1985c, 308). Part 2 is composed of six remaining chapters. In chapter 7 we examine the sources of growth. Chapter 8 is an analysis of the rapid rise in Korean savings and looks at the role of investment and the series of five-year plans. In chapter 9 we discuss exchange rate, trade, and industrial policies. The important link between wages, productivity, and international competitiveness are explored in chapter 10. In chapter 11 we examine monetary and fiscal policies, and then discuss income distribution in chapter 12. In part 3 (ch. 13) we will provide a synthesis and discuss the lessons from Korea’s experience.

7

Korea’s Rapid Growth

One of the most notable features of Korea’s experience has been its consistently high rates of growth. Growth rates for Korea and a number of other countries are given in table 7.1. The sample includes developed as well as developing countries, Asian as well as Latin American countries, and

220

Susan M. Collins and Won-Am Park Economic Growth Rates (GDP)

Table 7.1 Country

1963-72

1972-81

1981-84

Korea

9.84

7.89

7.36

Argentina Brazil Mexico

3.90 8.51 7 90

1.24 6.51 6.59

0.38 0.71 -0.85

Indonesia Philippines Turkey

4.88 4.70 6.30

7.63 6.09 4.89

3.17 -0.75 3.88

Singapore Hang Kong

9 84 -

8.05 8.68,

7.20 5.69

Japan U.S. France Germany U.K.

9.95 3.93 5.41 4.40 2.71

4.30 2.70 2.68 2.27 I .39

3.73 2.31 1.13 0.89 1.39

Source: IMF. International Financial Statistics. for all countries except Hang Kong. Data for Hang Kong is from the Korean Economic Planning Board. "1973-81

debtor as well as nondebtor countries. Three facts stand out from these figures. First, all countries exhibit some slowdown in growth between 1972-81 and 1981-84. Second, Asian countries (with the exception of the Philippines) have maintained relatively high growth rates. The Asian developing countries have grown more quickly on average than developing countries in Latin America, while Japan has maintained rapid growth relative to other OECD countries. Finally, even though Japan and a number of developing countries grew rapidly during the 1960s, only Singapore and Korea maintained growth rates in excess of 7 percent through 1984. How Korea achieved its rapid growth is one of the keys to understanding Korean adjustment. The rapid growth, particularly of exports, has enabled Korea to continue to service its large debt with less disruption than has occurred elsewhere. Large increases in labor productivity have contributed to the country's ability to depreciate while raising real wages. In addition, one of the most striking aspects of Korea's recovery from the 1979-80 crisis has been the ability to improve the trade balance, not from a recession-induced cut in imports, but with growth. In this chapter we examine Korea's growth rates in more detail. We begin with an analysis of the role of external debt in contributing to growth through financing investment. We then turn to decomposition of the sources of growth by sectoral origin and by type of expenditure. In the next section we discuss the results from an accounting decomposition of the role of factor inputs and productivity gains. The recent current account improvement with growth is discussed in a final section.

221

KoredChapter 7

7.1 Growth Effects of External Debt We begin with a simple exercise to estimate the role of foreign borrowing in Korean growth. Our decomposition has two steps. In the first, the growth of GNP is equal to the incremental capital output ratio (ICOR) multiplied by the increase in the capital stock (investment). The ICOR gives the additional capital required to generate an additional unit of output. It changes over time as the stock of capital and other inputs changes and as a result of positive or negative shocks to the output supply function. For example, technological improvements would tend to lower the ICOR, while a jump in oil prices would tend to raise it. In the second step, total investment is financed by a combination of domestic and foreign savings. With no external borrowing, investment would be constrained by domestic savings resulting in a smaller output expansion. Thus, it is possible to use the ICOR, investment rates, and the share of investment which was financed from abroad to decompose GNP growth into two parts-growth exclusive of foreign borrowing and growth attributable to external debt. The major shortcoming of this approach is that it assumes that the ICOR would have been the same if there had been no external borrowing and if investment had been smaller. For this reason, the results should be interpreted as indicative only. Table 7.2 provides the relevant data for five periods between 1962 and 1982. It shows that the ratio of investment and domestic savings to GNP both rose steadily until 1982. Foreign savings increased during 1962-71 and has since fallen relative to GNP and to national savings. The ICOR has risen continually since 1962, with a large jump in 1982. (Using the 1982 figure one is likely to overestimate the appropriate figure for 1983-85.) The marginal product of capital would be expected to fall as the capital stock increased. However, this tendency would be dampened by increases in other inputs (in particular, employment and human capital) and Table 7.2

Growth Effect of External Debt

GNP growth rate, % (A) Investment ratio National saving ratio Fbreign saving ratio Marginal total capital coefficient' GNP growth rate without foreign savings, % (B) Growth effect of external debt, % (A - B)

1962-66

1967-71

1972-76

1977-81

1982

1962-82

7.9 16.3 8.0 8.6

9.7 25.4 15.1 10.0

10.2 29.0 20.4 6.7

5.7 31.0 25.5 5.6

5.3 27.0 22.4 4.5

8.2

2.3

3.1

3.4

4.8

6.2

3.4

3.8

4.9

6.9

4.1

4.1

4.9

4.1

4.8

2.3

1.5

1.2

3.3

Note: GNP calculated at 1975 constant prices.

"Total capital includes domestic gross fixed formation, increases in inventories, and statistical discrepancy.

222

Susan M. Collins and Won-Am Park

by positive shifts (improved know-how, economies of scale). The data imply that the combination of increased capital and unfavorable developments (oil shocks and perhaps resource misallocations) have had the larger impact. The last rows of the table decompose GNP growth into a domestic and an external part. Approximately half of the growth during 1962-71 is attributable to external borrowing. If investment had been financed only by domestic savings, GNP would have grown by 3.8 percent in 1962-66 and 4.9 percent in 1967-71. In comparison, the actual growth rates were 7.9 and 9.7 percent, respectively. The contribution of debt to growth declined somewhat in 1972-76 and declined significantly during 1977-82. Without borrowing, growth would have been 6.9 percent (compared to the actual 10.2 percent) during 1972-76 and 4.1 percent (compared to 5.5 percent) during 1977-82. Foreign-financed investment has clearly played an important part in Korea’s remarkable growth performance. Overall, about 3.3 percent a year, or 40 percent of total actual growth, is attributable to external borrowing. The importance of borrowing diminished over time, but still accounted for over 20 percent of growth during 1977-81.

7.2 Decompositions of Growth In this section we consider some simple growth accounting. Contributions to growth by industrial sector are examined in table 7.3, and in table 7.4 we decompose GNP growth by type of expenditure. In each case, the contribution of an individual component to total GNP growth is computed as that component’s growth rate multiplied by its share in total output. The growth rates are given for five time periods, beginning with the 1971-73 recovery from the first period of rapid debt accumulation and including the second and third debt accumulation periods and the subsequent recoveries. The tables show the high growth rates of GNP during the early 1970s, the slowdown following the first oil shock, the strong rebound during 1976-78,

Table 7.3

Contributionsto Growth by Sectoral Origin (in percentages)

GNP Agriculture, forestry, and fishing Mining and quarrying Manufacturing Construction Electricity, gas, and water Transportation Other

1971-73

1974-75

1976-78

1979-82

1983-85

9.46 1.15 0.10 3.44 0.39 0.17 0.78 3.42

7.65 1.42 0.18 2.94 0.72 0.19 0.49 I .71

11.71 0.33 0.11 4.98 1.08 0.30 1.10 3.40

3.57 0.42 0.01 I .57 0.33 0.18 0.58 0.48

8.58 0.62 0.12 3.17 0.88 0.59 0.66 2.53

Source: EPB. Mujor Stufisficsof Korean Economy, 1987

223 lsble 7.4

KoredChapter 7

Contributions to Growth by Expenditure

GNP Private consumption Government consumption Fixed investment Exports Imports OtheP

1971-73

1974-75

1976-78

1979-82

1983-85

9.46 5.83 0.60 1.87 5.39 -4.54 0.30

7.65 6.76 1.57 2.47 1.89 -2.87 -2.11

11.74 4.24 0.88 6.89 6.77 -7.97 1.37

3.57 2.64 0.38

8.58 3.88 0.42 3.43 3.54 -2.61 -0.08

0.44 2.59 - 1.31 -1.15

Source: EPB,Major Statistics of Korean Economy, 1987.

Includes inventories, net factor payments from abroad, and any rounding errors

the real stagnation during 1979-82, and the recent recovery. A few general observations are useful. First, on average, private consumption has grown more slowly then income, declining from 77 percent of GNP in 1966 to 61 percent in 1985. The notable exception to this trend came during the 1979-82 crisis. Second, government consumption has declined, both as a share of GNP and in terms of its contribution to growth. It is interesting that government consumption is often countercyclical. For example, the growth rate of government consumption dropped during the 1973 real growth spurt and increased sharply as GNP growth slowed during 1974. (Fiscal policy will be examined in more detail in ch. 11.) Gross fixed investment has risen steadily as a share of GNP, from 18 percent in 1966 to 30-33 percent during 1978-85. The dramatic increase in capital accumulation during the late 1970s was reversed during the 1979-82 crisis. Since 1982 investment growth rates have returned to their early 1970s average. Stocks have been quite variable. On average, inventories accumulated from 1971-75 and have decumulated since 1976, with especially rapid declines during 1979-82. Table 7.4 also shows the well-documented rise in exports. From only 7.3 percent of GNP in 1966, they had grown to 22.4 percent by 1973 and have remained at 38-39 percent since 1982. Imports also rose rapidly, from 16 percent of GNP in 1966 to 43 percent by 1980, declining to 38 percent by 1985. There are significant differences in the sources of GNP growth across the five time periods. During 1971-73, the rapid growth was attributable almost equally to private consumption and to exports, with fixed investment considerably less important. Net exports contributed almost 1 percentage point to growth, on average. Table 7.3 shows that manufacturing and services contributed 3.4 percent each, while agriculture contributed 1.1 percent.

224

Susan M. Collins and Won-Am Park

The slowdown in growth in the second period occurred as export growth dropped sharply. This was only partially offset by a decline in import growth and increases in private and government consumption. Exports and fixed investment were the two major sources of growth during the 1976-78 boom, contributing 6.8 and 6.9 percent, respectively. However, there was also a large jump in imports, so that the contribution of net exports to growth remained negative. On the industrial side, there was a rebound in both manufacturing and other services. The sources of stagnant real growth during the last crisis period differ considerably from the 1974-75 slowdown. As a result of very poor harvests, agriculture contributed - 3.6 percent to growth in 1980. However, this decline is offset by an unusually large contribution during 1981. (See table 7.3.) Unlike the earlier period in which the contribution of total investment actually rose (from 1.9 percent in 1971-73 to 2.5 percent in 1974-73, investment declined precipitously from 6.9 percent during 1976-78 to 0.4 percent in 1979-82. Export growth also fell, but to 2.6 percent as compared to 1.9 percent during 1974-75. Furthermore, import growth slowed even more dramatically so that the contribution of net exports to GNP growth turned positive. There was also a very large decline in private consumption growth. Again, the sources of the 1983-85 rebound are quite different from the sources of earlier recoveries. In particular, growth rates for fixed investment, exports, and private consumption increase only moderately, with the contribution of private consumption to growth exceeding the contributions of the other two components. Government consumption does not play a role in the improved performance. A key factor is the surprisingly small increase in import growth-the contribution of net exports remains positive. We summarize our discussion by making four points. The first is that the well-documented role of exports as a source of growth is especially relevant in the early period. Second, investment has played a critical role, often rivaling exports in making the largest contribution to growth. Gross fixed investment contributes to capacity and is a critical determinant of future growth potential. The figures reported in table 7.4 strongly suggest an investment-led growth for Korea in which current investment stimulates future net exports, both through a rise in export production capacity and through a reduction in required manufactured imports. The high investment during the 1976-78 period helps to explain the reversal in the contribution of net exports to growth, from a negative one during 1974-78 to a positive one during 1979-85. Third, there has been a recent shift in the sources of growth. Moderate import expansion has permitted strong GNP performance despite the slowdown in growth rates of exports, investment, and private consumption. Finally, government consumption has played a small role throughout in explaining the performance.

KoredChapter 7

225

7.3

The Role of Imports: Further Analysis

The majority of imports are capital goods and raw materials. Final consumption goods constitute less than 5 percent of imports, Increases in investment and exports lead to direct increases in imports. This linkage implies that the figures in table 7.4 overestimate the contributions of these components to growth because they do not correct for the associated import growth. Korean input-output tables from the Bank of Korea provide “import requirement coefficients” by sector for selected years. Table 7.5 shows the actual coefficients for consumption, investment, and exports for four years: 1970, 1975, 1980, and 1983. The numbers imply that a one-unit rise in consumption would have increased imports by 0.13 units in 1970 and by 0.23 units in 1980. The import coefficients for investment jump from 0.39 in 1970 to 0.48 in 1975, declining to 0.38 by 1983.’ The coefficients for exports jumped from 0.26 in 1970 to 0.36 in 1975, and have remained in that range through 1983. Using these data, the annual import coefficients for each sector can be approximated by interpolation. The next step is to decompose imports. There are seven components of demand ( Y ) : private and government consumption, C and G; fixed investment, IF; inventory investment, Inv; exports, X ; imports, M ;and nonfactor income, N F . Imports are attributed to consumption, investment, exports, and a residual, E. The import coefficient for consumption, g , is assumed to be equally relevant for private and government consumption. The coefficient for investment, yc, is assumed to apply only to fixed investment, and not to changes in inventories.

+

Y =C G +IF

(1)

(2) (3)

+I

~ +vX - M

+NF

M=y,(C+G)+yiIF+y,X+~

+

+

Y = ( 1 - yc)C+ (1 - yc)G (1 -?;)IF + I ~ v ( I - y,)X-

E

+NF

Table 7.6 uses the adjusted expenditure components to decompose the sources of GNP growth for the five subperiods. The same general trends emerge here as in table 7.4. However, the revised figures show that the simple decomposition overstates the contribution to Korean output coming Import Requirement Coefficients for Korea (per unit final output)

Table 7.5

Year

Consumption

Investment

Exports

1970 1975 19x0 1983

0.13 0.18 0.23 0.22

0.39 0.48 0.42 0.38

0.26 0.36 0.35 0.36

Source: BOK, input-output tables

226

Susan M. Collins and Won-Am Park

Table 7.6

Contributions to Growth by Expenditure (revised for import dependence) 1971-73

GNP Private consumption Government consumption Fixed investment Exports Imports (residudl) OtheP

1974-75 ~ _ _ _ _ _ _

~

9 46 4 93 0 51 I 05 3 73 - 1 07 0 30

7 65 5 50 I29 I31 I21 0 50 - 2 II

1976-78

1979-82

1983-85

3.57 2.05 0.29 0.2 I 1.19 0.97 - 1.15

8.58 3.03 0.33 2.18 2.26 0.86 -0.08

~

11.74 3.35

0.70 3.40 3.82 -0.46 1.37

Source: EPB, Major Statistics of Korean Economy, 1987.

"Includes inventories, net factor payments from abroad, and any rounding errors

from growth of exports and fixed investment, particularly during the Big Push with its heavily import-dependent investments.

7.4 Supply-side Sources of Korean Growth This section reviews the economic sources of Korean growth during 1963-72 and 1972-82. The discussion is based on work by Kim and Park (1985), who have used the Denison growth accounting framework to analyze Korea.* The data in all of the tables reported below come from their study. Estimates of the sources of growth for the whole Korean economy are given in table 7.7. Beginning with the experience over the entire period, we point out three characteristics. First, as discussed above, Korea maintained very high rates of growth. Second, about two-thirds of its performance is attributable to increased factor inputs, with two-thirds of that increase arising from labor inputs and only one-third from additions to the capital stock. We shall see that the absolute and relative importance of factor inputs are distinguishing characteristics of Korea's experience. Finally, the most important contributions to Korean growth have been a result of increased employment, additions to nonresidential structures and equipment, and economies of scale. While these three characteristics provide a general picture of the sources of Korean growth, they do not capture the shifts in these sources over time or highlight those aspects which differentiate Korea's experience from that of other countries. After a brief discussion of the differences between the sources of growth during 1963-72 and 1972-82, we will compare Korean growth sources with those in other countries. As shown, the average growth rates of national income fell slightly from 8.22 percent during 1963-72 to 7.05 percent during 1972-82. In the earlier period, approximately equal parts of the growth were attributable to increased factor inputs and to increased output per unit input.3 In the later period, however, actor inputs account for nearly 80 percent of the growth, with a corresponding decline in the contribution of productivity increases.

227

Table 7.7

KoreaChapter 7 Sources of Growth of the Korean Economy _

~

Source National income (growth rate) Total factor input Labor Employment Average hours Age-sex composition Education Efficiency offset Unallocated Capital Nonresidential structure and equipment Inventories International assets Dwellings Land Output per unit of input Improved resource allocations Contraction of agricultural inputs Contraction of nonagricultural self-employment Economies of scale Measured in U.S. prices Income elasticities Irregular factors Effect of weather on fanning Intensity of demand Advances in knowledge and n.e.c.

1963-72 8.22 (100.00) 4.19 (51.00) 3.05 (37.10) 2.36 (28.70) 0.32 (3.90) -0.01 (-0.10) 0.31 (3.80) 0.01 (0.10) 0.06 (0.70) 1.14 (13.90) 1.09 (13.30) 0.20 (2.40) -0.25 (-3.00) 0.10 (1.20) 0.00 (0.00) 4.03 (49.00) 0.63 (7.70) 0.53 (6.40)

0.10 (1.20) 1.52 (18.50) 0.87 (10.60) 0.65 (7.90) -0.01 (-0.10) 0.03 (0.40) -0.04 (-0.50) 1.89 (23.00)

_

_

1972-82 7.05 (100.00) 5.58 (79.10) 3.48 (49.40) 2.03 (28.80) 0.45 (6.40) 0.15 (2.10) 0.44 (6.20) 0.05 (0.70) 0.36 (5.10) 2.10 (29.80) 2.59 (36.70) 0.31 (4.40) -0.89 ( - 12.60) 0.09 (1.30) 0.00 (0.00) 1.47 (20.90) 0.68 (9.60) 0.64 (9.10) 0.03

(0.40) (20.70) 0.85 (12.10) 0.61 (8.70) -0.97 ( - 13.80) 0.01 (0.10) -0.98 ( - 13.90) 0.30 (4.30) 1.46

~

_

_

_

1963-82 7.61 (100.00) 4.89 (64.30) 3.31 (43.50) 2.18 (28.60) 0.40 (5.30) 0.06 (0.80) 0.39 (5.10) 0.03 (0.40) 0.25 (3.30) 1.58 (20.80) 1.80 (23.70) 0.26 (3.40) -0.58 (-7.60) 0.10 (1.30) 0.00 (0.00) 2.72 (35.70) 0.67 (8.70) 0.60 (7.90)

0.06 (0.80) 1.49 (19.60) 0.86 (11.30) 0.63 (8.30) -0.52 (-6.80) 0.01 (0.10) -0.53 (-7.00) 1.09 (14.30)

Source: Kim and Park (1985) table 4.6 Note; Growth rates and contributions to growth are in percentages. The share of total growth is given in parentheses. n.e.c. = not elsewhere classified.

The contribution of factor inputs to growth rose in absolute as well as relative terms. In the first period, the total contribution was 4.19 percent as compared to 5.58 percent in the second period. From 1963 to 1972, labor accounts for 73 percent of the increase due to factor inputs, with 88 percent of the labor increase explained by greater employment and a rise in average work hours. From 1972 to 1982, the importance of labor declined somewhat to 62 percent of the total factor input, 71 percent of which is explained by increased employment and average hours. In both periods, equal contributions to growth come from education and from increased work hours. The importance of capital nearly doubles from the first to the second period. This increase is due almost exclusively to additions of nonresidential structure and equipment, reflecting the rapid investment buildup which was initiated in the 1970s. The total contribution of capital would have been even larger if not for the high negative contribution from international assetsKorea’s net investment income from abroad has been negative since the mid- 1960s.

228

Susan M. Collins and Won-Am Park

During 1972-82, the contribution of output per unit input fell to barely one-third of its level during the previous period. This development is not explained by either resource allocation or economies of scale, which retained approximately the same contribution to growth during both periods. The large decline came from two “irregular factors”-the impact of poor weather on farming and unfavorable developments in what Kim and Park (1985) label the intensity of demand. There was also some decline in the residual, as discussed further below. The above discussion highlights the importance of increased employment in explaining Korean growth. We find that it is also useful to examine the sources of growth per person employed. Table 7.8 shows that the average annual increase in labor productivity fell from 4.59 percent during 1963-72 to 3.88 percent during 1972-82. Again, this aggregate figure masks a dramatic change between periods. In the earlier period, only 13 percent of growth per person employed is attributable to capital (human and/or physical), shifts in sectoral allocation of workers, or changes in land, while 87 percent is attributable to increased productivity. After 1972, factor inputs account for 62 percent of growth per worker, with the increases in human and physical capital partially substituting for the substantial decline in productivity growth. Again, a large share of the reduction in productivity growth shows up in the residual. It is possible to suggest some explanations for this. Kim and Park (1985, 174-75) mention four alternatives. The first is changes in the efficiency of resource allocation arising from shifts in Korea’s industrialization strategy: “the initial positive momentum from opening-up to semi-free trade was largely spent by the early seventies. . . . Korea again emphasized

Table 7.8

Sources of Growth of the Korean Economy per Person Employed

Source National income per person employed Irregular factors Factor inputs Labor Changes in hours Increased education More capital Less land OutpuVinput Less labor misallocation Economies of scale Advances in knowledge

1963-72

4.59 -0.01 0.56 0.69 0.38 0.31 0.32 -0.45 4.04 0.63 1.52 1.89’

Source: Kim and Park (1985, table 8.2).

Note: Percentages in parentheses.

(100.00) (-0.20) ( 12.20) (15.00) (8.30) (6.80) (7.00) ( - 9.80) (88.00) ( I 3.70) (33.10) (41.20)

1972-82

3.88 -0.97 2.41 1.45 1.01

0.44 1.42 -0.46 2.44 0.68 1.46 0.30

(100.00) (-25.00) (62.10) (37.40) (26.00) (11.30) (36.60) (-11.90 (62.90) (17.50) (37.60) (7.70)

I963-82

4.23 -0.50 1.51 1.13 0.74 0.39 0.80 -0.42 3.22 0.66 1.48 1.08

(100.00) (-11.80) (35.70) (26.70) (17.50) (9.20) (18.90) (-9.90) (76.10) (15.60) (35.00) (25.50)

229

KoredChapter 7

the import substitution of intermediate and capital goods during the seventies, in connection with the government promoted construction of heavy and chemical industries” (173). A second source of a decline in residual contributions to growth is increases in hours spent in required military and civil defense training. Since some of the implied absences from work are not counted in labor force statistics, the contribution of labor input to growth may have been overestimated, with the residual contribution underestimated (174). The third possibility arises from increased industrial concentration and, presumably, associated reductions in productive efficiency. The fourth possibility comes from the jump in oil prices after 1973.

7.5 International Comparisons of the Sources of Growth Table 7.9 compares the sources of growth for Korea with the sources for Japan, the United States, Canada, and eight European countries over various time periods during 1950-73.4 As shown, Japan and Korea had the highest growth rates, followed by West Germany. In absolute terms, Korea and the faster growing European countries saw comparable gains from increased productivity. However, in terms of the composition of the sources of growth, Korea has less in common with the other rapid growth countries than with the slower growing United States and Canada, where increased factor inputs also accounted for two-thirds of growth. The rapid growth rates in Korea and Japan have frequently been compared. Although the aggregate growth levels are comparable, the sources are quite different. Korea had significantly higher growth of labor inputs (arising from increased employment and work hours) and fixed capital formation, but lower inventories5 and substantial outflows of international capital which decreased the overall contribution of capital. In terms of output per unit input, Japan recorded larger contributions to growth across the board, achieving larger gains from productivity increases than any of the other countries. Thus, the Korea-Japan comparison is somewhat misleading, because it fails to highlight the overwhelming importance of increased factor inputs in explaining the rapid Korean growth. In summary, the critical factor in Korea’s high and sustained growth has been the increasingly large contribution of factor inputs. Korea was able to maintain high growth rates in the 1970s, despite negative shocks and declines in the contribution of productivity, by raising the contributions from both labor and capital inputs. 7.6

Current Account Improvement with Growth

One of the striking aspects of Korea’s adjustment to the 1979-80 crisis was that a turnaround from negative to positive growth was combined with a

Table 7.9

International Comparisons of Sources of Growth Percentage of Growth Growth Rate (standardized)

Factor Inputs

Korea Japan

8.13

us.

3.79 4.95 3.03 3.63 4.70 6.27 5.60 4.07 3.43 2.38

4.89 3.95 2.13 3.02 1.17 1.55 1.24 2.78 1.66 1.91 1.04 1.11

Country

Canada Belgium Denmark France West Germany Italy Netherlands Noway U.K.

8.81

Total

Labor

Capital

0.60

0.41 0.21 0.37 0.37 0.25 0.16 0.10 0.22 0.17 0.21 0.04 0.25

0.19 0.24 0.19 0.23

0.45

0.56 0.61 0.39 0.43 0.26 0.44 0.30 0.47 0.30 0.47

0.14

0.26 0.17 0.22 0.13 0.26 0.26 0.21

Percentage of Growth Output per Unit Input 3.24 4.86 1.66 1.96 1.86 2.08 3.46 3.49 3.94 2.16 2.39 1.27

Total 0.40 0.55

0.44 0.40 0.61 0.57 0.74 0.56 0.70 0.53 0.70 0.53

Source: Kim and Park (1985, 67-69, table 4.7) Note: Data reported are for the period 1950-62 for all countries except the following: Korea, 1963-82; Japan, 1953-71; U.S., 1948-73; and Canada, 1950-67.

Resource Allocations

Economies of Scale

Other

0.08 0.11 0.08 0.13 0.17 0.19 0.20 0.16 0.25 0.15 0.27 0.05

0.18 0.22 0.08 0.13 0.17 0.18 0.21 0.26 0.22 0.19 0.17 0.15

0.13 0.22 0.28 0.13 0.17 0.21 0.32 0.14 0.23 0.18 0.26 0.33

KoredChapter 7

231

substantial improvement in the current account. From a theoretical standpoint, there is nothing particularly surprising about this achievement. Strong export performance will stimulate output while reducing a trade deficit. However, an observer aware of only recent country experiences would be left with the impression that, at least in the short run of one to two years, countries improving their trade balance are most likely to do so by a domestic contraction which reduces imports.6 The natural question becomes why was Korea able to avoid the typical scenario and are there lessons to be learned for others that would like to follow suit. We focus here on the experience through 1985. The 1986 current account surplus, which raises a number of other issues, is discussed in detail by Dornbusch and Park (1987). Table 7.10 gives an overview of the Korean balance of payments. As shown, the current account worsened markedly from 2.1 percent of GNP in 1978 to an average of 7.5 percent of GNP during 1979-81. The deterioration is accounted for primarily by a jump in imports and in payments for invisibles (primarily transport and investment income). Beginning in 1982, the current account deficit was steadily reduced, reaching 1 percent of GNP by 1985. In 1982 the improvement came in almost equal parts from the trade balance and from the balance on invisibles. There was essentially no growth in export receipts measured in U.S. dollars, so that the gains were due to a 3 percent decline in the value of imports. This decline is only partially explained by the 5 percent reduction in import prices. The unusual aspect is that imports rose by only 0.2 percent in physical terms even though real GNP grew by 5.4 percent. In 1983 the current account improvement was due primarily to strong export performance. In 1984 and 1985 the improvement continued, despite renewed deterioration of the invisibles account, because of a dramatic reduction in the trade deficit. Real output has grown strongly in each of these years, including the very rapid rates achieved during 1983 and 1984. Current Account and GNP, 1977-86 (in billions of U.S. dollars and percentages)

Table 7.10

Year 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

Current Account 0.012

- 1.085 -4.151 -5.321 -4.646 -2.650 - 1.606 - 1.373 -0.887 4.617

WGNP (70) 0.0 -2.1 -6.8 -8.8 -7.0 -3.8 -2.1 -1.7 -1.1 4.9

Trade Balance -0,477 - 1.781

-4.396 -4.384 -3.628 -2.594 - 1.764 - 1.036 -0.019 4.206

Exports

Imports

Invisibles Account

10.046 12.711 14.704 17.214 20.671 20.879 23.204 26.335 26.442 33.913

10.523 14.491 19.100 21.598 24.299 23.474 24.967 27.371 26.460 29.707

0.266 0.224 -0.195 - 1.386 - 1.518 -0.554 -0.435 -0.878 - 1.446 -0.628

Transfer Account 0.223 0.472 0.439 0.449 0.501 0.499 0.592 0.541 0.578 1.039

GNP Growth

(a)

10.7 11.0 7.0 -4.8 6.6 5.4 11.9 8.4 5.4 12.5

232

Susan M. Collins and Won-Am Park

The real puzzle then is that imports did not rise more strongly with the recovery in growth. To examine import behavior more carefully, in table 7.11 we decompose imports by major commodity group, while in table 7.12 we show the annual growth rates for import values, import volumes, and unit import values for 1977-85. As expected, the tables show the large rise in import prices in 1979-81 and the increase in imports of crude materials and mineral fuels. The Korean oil bill rose from 17.5 percent of total imports during 1976-78 to 27.3 percent during 1980-82, as the value of petroleum imports increased by 228.4 percent between the two periods. It is important to keep in mind that the OPEC revenue gains had favorable feedback effects for Korea. In particular, the value of overseas construction contracts to the Middle East were $14.2 billion larger during 1980-82 than during 1977-79, as compared with a $10.2 billion dollar rise in the value of imported petroleum and petroleum products between the same three-year periods.’ However, there were also very large increases in imported foods and machinery and transport equipment, especially during 1981. Imports declined in 1982. There was a substantial reduction in world prices of Korean imports. Furthermore, import volumes grew very slowly. The major reason was the large decline in food imports as the agricultural sector registered a banner harvest. Although normal import growth resumed in 1983, import price developments continued to be extremely favorable. The question raised at the beginning of this section was how Korea managed to improve its trade balance while maintaining rapid real growth rates. Although export growth resumed by 1983, the initial improvements depended on the behavior of imports. The analysis above points to four factors. The first is the favorable price developments during 1982-85. The second is the role of weather conditions in creating an unusual sectoral composition of growth post-1979. A bad harvest leads to increased food imports. When domestic agriculture recovers, output and the trade balance are improved simultaneously. The fourth factor involves policy more directly. Korea has restricted consumer imports and thereby limits those imports which tend to have the highest income elasticities. Most imports in Korea are tied to industrial requirements for imported materials. Are there lessons to be learned from this episode, in terms of policy advice to other countries? It is very difficult to argue that policy choices were a determining factor in Korea’s ability to use growth to improve the current account. Other countries which followed the same policies, but had different shocks (e.g., external prices and harvests) should not expect the same outcome. Furthermore, there is no reason to expect that Korea, faced with a large trade deficit in the future, will be able to repeat this performance, unless the deficit is again linked to a temporary decline in agriculture and to a temporary deterioration in the terms of trade.

Table 7.11

Year

Food, Etc.

Imports by Major Commodity Group, 1976-84 Crude Materials

Mineral Fuels

Machinery Petroleum

Chemicals

Manufactures

& Transportation

Ships

Other

Total

1,658 2,065 2,312 3,416 6.164 6,918 6,740 6,195 6,414

866 1,005 1,298 2,009 1,836 2,109 2,084 2,281 2,762

1,479 1,929 2,782 3,440 3,122 3.562 3,394 3,942 4,881

2,387 2,908 4,947 6,125 4,977 6.000 6.009 7,556 9.797

397 193 402 316 472 873 1,119 1,798 2,704

10 13 II 71 93 141 93 93 102

8,774 10,811 14.972 20,339 22,292 26.131 24,251 26,192 30,631

0.19 0.19 0.15 0.17 0.28 0.26 0.28 0.24

0.10 0.09 0.09 0.10 0.08 0.08 0.09 0.09 0.09

0. I7 0.18 0.19 0.17 0.14 0.14 0.14 0.15 0.16

0.27 0.27 0.33 0.30 0.22 0.23 0.25 0.29 0.32

0.05 0.02 0.03 0.02 0.02 0.03 0.05 0.07 0.09

0.00 0.00 0.00 0.00 0.00 0.01 0.00 0.00 0.00

1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00

Panel A (in rnillirms of U S . dollars) 1976 1977 1978 1979 1980 1981 1982 1983 1984

720 834 1,087 1,655 1,993 2,926 1,708 1,886 1,864

1,565 1,941 2,395 3,260 3,634 3,630 3,370 3,480 3,951

1,747 2,179 2,453 3,779 6,638 7,765 7,593 6,958 7,274

Panel B (in percentage of total imports) 1976 1977 1978 1979 1980 1981 1982 1983 1984

0.08 0.08 0.07 0.08 0.09 0.11 0.07 0.07 0.06

0.18 0.18 0.16 0.16 0.16 0.14 0.14 0.13 0.13

0.20 0.20 0. I6 0.19 0.30 0.30 0.31 0.27 0.24

0.21

234

Susan M. Collins and Won-Am Park

Table 7.12

Percentage Changes in Components of Import Growth, 1977-85

Year

Import Value

Import Volume

Unit Value

Real GNP

1977 1978 1979 1980 1981 I982 1983 1984 1985

23.2 38.5 35.8 9.6 17.2 -7.2 8.0 16.9 1.6

20.5 31.2

2.2 5.6 22.2 20.3 5.4 -7.4 -4.7 1.3 -4.2

10.7 11.0 7.0 -4.8 6.6 5.4 11.9 8.5 5.4

8

11.1

-8.9 11.2 2.2 13.3 15.4 6.2

Savings and Investment

During each of Korea’s periods of rapid debt accumulation, virtually all of the additional foreign borrowing was used to finance current account deficits. Since domestic investment must be financed through some combination of domestic and foreign savings, foreign savings-r the deficit in the current account-is exactly equal to the imbalance between domestic savings and investment. In this chapter we examine the behavior of the current account from the savings-investment perspective. The decomposition is especially interesting for Korea because its experience differs markedly from that of many other debtor countries. A frequently observed pattern is for the current account deficit to increase as government savings decline and then for a current account improvement to be attained, at least in the short run, through cuts in (public and private) investment and in government expenditure, thus raising government savings. Relatively little of the adjustment tends to be achieved through private sector savings. Korean experience contrasts with the “stylized” scenario with respect to the roles of investment, public savings, and private savings. First, fiscal deficits have played at most a minor role in current account deterioration. Instead, increases in fixed investment, associated with new economic development strategies, have outpaced rising private savings. This leaves the door open for a jump in required foreign financing to cover either unexpected surges in inventory accumulation or unexpected drops in private savings. The series of five-year economic and social plans have played a critical role through their impact on investment. Second, the reduction of the current account deficit during the recovery is achieved without a substantial decline

235

KoredChapter 8

in investment. The adjustment comes from increased domestic savings, the lion's share of which is generated by the household sector. This pattern is illustrated by figure 8.1. The plot shows the behavior of gross fixed investment and of domestic savings less inventory accumulation, each as a share of GNP. Accounting identities imply that the difference between these two variables is equal to the current account. When fixed investment is larger than the excess of domestic savings over inventory accumulation, the current account is in deficit, while small investment relative to the adjusted domestic savings is the counterpart of current account surplus. In broad terms, fixed investment has behaved like a step function with rapid increases during 1965-68 and jumps in 1974 and 1979. There has been considerably more variation in the adjusted savings variable. The large current account deficits during 1970-71, 1974-75, and 1980-81 follow rises in fixed investment, but are precipitated primarily by reduced savings and/or increased inventories. Similarly, the current account improvements are explained by increased domestic savings relative to inventory accumulation, with only a small role for reduced fixed investment. The remaining sections of the chapter examine savings and investment behavior in more detail. We turn first to domestic savings in section 8.1, and then to investment and the role of the five-year plans in section 8.2. The data used in this chapter are based on the old System of National Accounts (SNA) decomposition. This allowed a long enough time series for the empirical

IF/GNP

.35

-

(Sav-InvVGNP -.-.-.

.30-

' l o - 66

'

L8

I

7b

I

i!2

I

l'4

'

76

I

i!8

I

i0

I

i2

I

84

Year IF = Gross Fixed Investment Inv = Accumulation of Stocks Sav= Domestic Savings

Fig. 8.1 Current account imbalance: ratios of investment and savings to GNP

236

Susan M. Collins and Won-Am Park

estimates. Furthermore, the new SNA data required to decompose domestic savings were only available for a few years at the time of this analysis. The data and the methods used for disaggregation are described in section 8.3. 8.1 Domestic Savings

In table 8.1 we show the behavior of domestic savings, foreign savings, and investment from 1965 to 1984. The top panel gives the variables in levels, while the bottom takes each variable relative to GNP. Evident from the table is the rise in savings from less than 15 percent of GNP during the mid-1960s to nearly 30 percent by the end of the 1970s. Also clear is that this impressive growth has been interrupted by drops of as much as 6 percent from one year to the next. Three sources of domestic savings are also identified: general government, public and private corporations, and other (including households and unincorporated businesses). Unfortunately, it is not possible to break the components down more finely. The data show there has been a significant shift from government and corporations to households as a source of savings. In 1965 government savings constituted nearly half of the total, with household and other savings the smallest component, accounting for less than 18 percent of the total. By 1984 household savings had grown to nearly 45 percent of the total. The share of corporate savings was 27 percent, representing a drop of 10 percent. The contribution of government savings had fallen even more, also accounting for about 28 percent of the total by 1984. Figure 8.2 shows government, corporate, and household savings relative to GNP during 1965-84. The three series have behaved quite differently. With the exception of the permanent increase in 1972, corporate savings have remained relatively flat, declining only slightly during economic downturns. One explanation for the increase in 1972 is the impact of the financial crisis and the August Emergency Decree, which effectively caused a transfer from lenders in the unofficial money market (UMM) to borrowers. The measure thereby succeeded in significantly reducing the activities of the UMM during the next year, nearly eliminating an important source of corporate finance. The transfer and the tighter access to funds would both be expected to increase corporate savings. Cole and Park (1983, 167) argue that any effects of the decree were short-lived, however, there does seem to have been a sustained effect on corporate savings. Corporate savings averaged 0.053 percent of GNP from 1965 to 1971 and, excluding the jump to 0.09 percent during 1972, averaged 0.075 percent during 1973-85. Government savings have fluctuated more than corporate savings, but have been considerably less variable than household savings. The graph in figure 8.2 documents a small rise from 1967 to 1970, a decline during 1971-75, and a gradual return to government saving rates on the order of 6 percent of

Domestic Savings by Source, 1965-84

Table 8.1

Domestic Savings Year

Total

Government

Corporate

Other

41.90 53.50 68.10 93.80 121.40 132.50 157.10 357.00 442.90 619.10 719.90 1,058.70 1,469.60 1,779.70 2,288.10 2,943.80 3,201 .SO 3,719.70 4,404.40 4,998.40

19.50 67.30 49.50 83.90 198.00 152.80 195.10 357.00 627.10 743.90 909.30 1,540.60 2,643.90 3,788. LO 4,519.80 3,265.10 4,143.70 5,004.90 6,075.40 8,155.60

Current Account Deficit

Gross Investment

Statistical Discrepancy

Total

Fixed

Stocks

10.00 13.50 22.10 -6.00 - 13.60 12.80 2.70 24.40 -43.10 -28.30 79.70 -74.80 -55.40 - 92.90 136.40 0.90 -82.30 66.30 30.00 54.70

120.90 223.90 280.70 427.70 621.30 693.20 848.10 923.00 I ,38 I . 20 2,374.80 3,030.10 3,556.90 5,026.50 7,554.90 I I , 139.40 11.630.20 13.343.00 13,979.80 16,225.40 19,447.60

119.00 209.80 274.60 413.60 555.80 627.10 726.40 830.80 1,257.70 I ,898.80 2,573.40 3,343.30 4,830.00 7,463.60 10,239.70 11,873.90 13,208.10 15,675.60 18,604.80 20,175.50

1.90 14.10 6.10 14.10 65.50 66.10 121.70 92.20 123.50 476.00 456.70 213.60 196.50 91.30 899.70 -243.70 134.90 - 1,695.80 -2,379.40 -727.90

Panel A (in billions of won) 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984

113.30 182.40 206.70 311.90 476.80 487.00 550.70 753.90 1,301.10 1,582.80 2,037.10 3.480.00 5,087.80 7,122.50 8,993.80 8,405.00 10,260.60 11,960.00 14,974.90 18.298.30

(continued)

51.90 61.60 89.00 134.20 157.40 201.60 198.50 163.40 231.00 219.80 407.90 880.70 974.40 1,554.70 2,185.90 2,196.10 2,915.10 3,235.00 4,495.00 5,144.40

- 2.40 28.10 51.90 121.80 158.20 193.50 294.70 144.60 123.20 820.40 913.20 151.80 -6.00 525.30 2,009.10 3,224.30 3,164.70 1,945.00 1,220.50 1,094.50

Table 8.1

(continued) Domestic Savings

Year

Total

Government

Panel B (in percentage of GNP) I965 14.06 6.44 I966 17.59 5.94 1967 16.13 6.95 1968 18.87 8.12 1969 22.12 7.30 1970 17.68 7.32 1971 16.32 5.88 1972 18. I5 3.93 1973 24.19 4.29 1974 21.10 2.93 1975 20.18 4.04 1976 25.07 6.34 1977 28.09 5.38 1978 29.40 6.42 1979 28.78 7.00 1980 22.92 5.99 1981 22.74 6.46 1982 23.58 6.38 1983 25.39 7.62 1984 27.55 7.75

Corporate

Other

Current Account Deficit

5.20 5.16 5.32 5.67 5.63 4.81 4.65 8.59 8.23 8.25 7.13 7.63 8.11 7.35 7.32 8.03 7.10 7.33 7.41 7.53

2.42 6.49 3.86 5.08 9.19 5.55 5.78 8.59 11.66 9.91 9.01 11.10 14.59 15.64 14.46 8.90 9.18 9.87 10.30 12.28

-0.30 2.71 4.05 7.37 1.34 7.03 8.73 3.48 2.29 10.93 9.05 1.09 -0.03 2.17 6.43 8.79 7.01 3.83 2.07 1.65

Gross Investment

Statistical Discrepancy

Total

Fixed

Stocks

1.24 1.30 1.72 -0.36 - 0.63 0.46 0.08 0.59 -0.80 -0.38 0.79 -0.54 -0.31 -0.38 0.44 0.00 -0.18 0.13 0.05 0.08

15.01 21.59 21.91 25.88 28.83 25.17 25.13 22.22 25.68 31.65 30.02 25.62 27.75 31.19 35.65 31.71 29.57 27.56 27.51 29.28

14.77 20.23 21.43 25.02 25.79 22.77 21.52 20.00 23.38 25.31 25.50 24.09 26.66 30.81 32.17 32.38 29.27 30.90 31.54 30.38

0.24 1.36 0.48 0.85 3.04 2.40 3.61 2.22 2.30 6.34 4.53 1.54 I .08 0.38 2.88 -0.66 0.30 -3.34 -4.03 -1.10

Source: EPB, Major Statistics of Korean Economy, 1986, and BOK, Flow of Funds Statistics, 1984. Note: Statistical Discrepancy is the difference between the depreciation allowance from the two sources (see discussion in sec. 8.3).

239

KoredChapter 8

17.5

Household Savings ---Government Savings -

I

Corporate Savings

SAVl NGS/GNP

Fig. 8.2 Components of savings/GNP

GNP. It also provides some evidence for an inverse relationship between government and household savings, particularly during the mid- 1970s. We turn next to an analysis of household savings, the most variable and, since 1972, the largest component of domestic savings. There are two facts to be explained. First, how did Korea managed to triple household saving rates from 5 percent during 1966-68 to 15 percent during 1977-79? And second, why have there been such large fluctuations in the household saving rate? Household savings are computed as the difference between domestic savings and the sum of general government and corporate savings. As already mentioned, they also include the savings of unincorporated businesses. Therefore, part of this component is a capital consumption allowance. Table 8.2 divides household savings into depreciation (HSD) and other (HSO). Not surprisingly, the depreciation has been quite stable as a share of GNP, ranging from 2.0 to 2.7 percent. Movements in this component clearly cannot explain the large fluctuations in the total. Figure 8.3 plots the ratio of household savings excluding depreciation to disposable personal income (HSOIYD). It shows that savings have risen relative to income in a ratchet fashion, which is suggestive of a permanent income model of consumption. In such a model, temporary and unexpected rises in income will have a relatively small affect on consumption, leading to short-run jumps in savings. The model seems particularly relevant for the Korean economy with its periodic growth spurts and slowdowns. Another factor which may help to explain the large fluctuations in savings is the real interest rate. Some authors have argued that increases in interest

240

Susan M. Collins and Won-Am Park

Table 8.2

Household Savings ~

Household Savings Year

Depreciation

Other

Disposable Income

(SND/YD)"

(SND/Y)b

1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980

18.00 24.40 32.40 35.60 45.60 54.80 66.60 89.20 131.00 199.00 241 .SO 317.60 416.10 570.50 678.00 799.20 959.40 1,260.90

1S O 42.90 17.10 48.30 152.40 98.00 128.50 267.80 496.10 544.90 667.50 1,223.00 2,227.80 3,217.60 3,84 I . 80 2,465.90 3,184.30 3,744.00

853.70 1,022.40 1,279.30 I,681.10 2,058.70 2,138.40 2,676.30 3,322.30 4,210.00 5,867.30 7,855.90 10,385.20 10,385.20 13,500.70 23,260.90 27,294.20 33.815.60 37,863.60

0.00 0.04 0.01 0.03 0.08 0.05 0.05

0.00 0.04 0.01 0.03 0.07 0.04 0.04 0.06 0.09 0.07 0.07 0.09 0.12 0.13 0.12 0.07 0.07 0.07

1981

1982

0.08

0.12 0.10 0.09 0.12 0.22 0.25 0.17 0.09 0.10 0.10

"Nondepreciation sdvings/disposable income. bNondepreciation savingsiGNP.

0.25-

0.2 I88 -

SAVINGS/INCOME

0.18750.1563 0.125-

-

0.0938

0.0625 -

0.0313-

Year Fig. 8.3 Ratio of household savings to disposable income Note: Household saving and disposable income both exclude depreciation allowance.

241

KoredChapter 8

rates relative to inflation generate additional savings and that financial reform in 1965 was the key to understanding the jump in Korean saving rates during the mid-1960s.' Giovannini (1983) and others have found little sensitivity of saving to interest rates in empirical studies of a broad sample of developing countries. Thus, it is interesting to examine the relationship between saving and interest rates for the more recent period, 1965-82. Korean bank deposits have been adjusted a number of times during this interval, but adjustments have typically not kept pace with inflation, leading to substantial variation in real interest rates. We assume a simple structure to explain saving behavior. Consumption is assumed to depend positively on income and negatively on the real return to saving. The consumption function, given by equation ( I ) , allows for different marginal propensities to consume out of permanent and transitory incomes. Equation (2) state that household savings are identically equal to disposable income less consumption. Disposable income is taken net of depreciation, so that the structure determines only the determinants of nondepreciation savings. C=

(1)

+ (YIYP i

O L ~

HSO

(2)

=

+

O L ~ Y ~ a3RR '

+E

YD - C

+

YT = Y - T - HSD; Y is personal income; YD is where YD = YP personal disposable income, less depreciation; YP and YT are (perceived) permanent and transitory income; C is household consumption; RR is the real interest rate; and HSO and HSD are the depreciation and nondepreciation savings. Temporary and permanent income were established as follows. Disposable income in period t was estimated as a linear function of disposable income from periods t - 1 and t - 2. Permanent income was measured as the fitted values from this regression while residuals were taken as a measure of transitory income. Using this procedure, transitory income ranged from 2 percent to nearly 10 percent of total disposable income. As expected, transitory income is very large and negative during 1970, 1980, and 1982. It is large and positive during 1974, 1978, and 1981. We use the nominal interest rate on one-year time deposits, deflated by the CPI. Other specifications, for example the use of curb market interest rates, did not significantly alter the estimation results. The estimation results from equation (1) are given below in equation (3), with t-statistics in parentheses. They very strongly support a permanenttemporary income model for Korean consumption behavior.

(3)

+

+ 0.88 YP 0.55 YT - 16.50 RR (0.44) (77.62) (4.26) (-0.95)

C = 91.20

Sample: 1965-82; adjusted R2 = 0.98; Durbin Watson = 1.78.

242

Susan M. Collins and Won-Am Park

They show a marginal propensity to consume out of permanent income of 0.88. The marginal propensity to consume out of temporary income is significantly smaller, 0.55. The estimates for the constant term and the influence of the real interest rate, however, are measured imprecisely and are not significantly different from zero. The regression does not provide support for the view that changes in real interest rates have been a critical determinant of saving behavior. Instead, it emphasizes the importance of the high average growth of disposable income as an explanation for the impressive rise in household savings, and the large swings in real growth rates as an explanation for the cyclic fluctuations in saving rates. However, the very dramatic rise in Korean household savings is unusual and has played an important part in Korea’s successful adjustment. Korean saving behavior clearly warrants additional analysis. 8.2

8.2.1

Investment Trends in Fixed Capital Formation

We begin with a discussion of the general trends in gross investment, focusing on the behavior of fixed capital formation. As already mentioned, and as illustrated in Figure 8.1, investment as a share of income follows cycles which coincide with the five-year development plans. In the first or second year of each plan, there has been a sudden rise in fixed capital formation. The increases continue through the third or fourth year of the plan and taper off somewhat in the last one to two years. The main thrust of each five-year plan is summarized in table 8.3. In table 8.4 we summarize the shares of gross domestic capital formation by industry and by type of capital good during 1972-84. This period includes the third, fourth, and original fifth five-year plans. The major developments during each plan period do in fact coincide with the stated plan objectives. During the third plan, 1972-76, and particularly during 1972-74, there was an increase in allocation to agriculture. Most of this increase came from declines in allocation to social overhead capital-transportation, storage, and communication. However, this sector retained the largest share of total investment, with manufacturing a close second. During the fourth plan there was a decline in the share of investment in agriculture. Allocation to manufacturing and social overhead also fell, with services growing substantially (especially wholesale and retail trade and public administration). The decline in manufacturing occurred during the 1980-81 retrenchment from the Big Push. During 1977-81, over 77 percent of investment in equipment in manufacturing went to the HC industries. It is also interesting that equipment, as a share of investment, jumped from 31 to

243

KoredChapter 8

Table 8.3

Main Focus of Five-Year Plans

First plan (1962-66)

Emphasis on basic industries for import substitution and expansion of social overhead capital

Second plan (1967-71)

Export-oriented industrialization to promote labor-intensive light manufacturing

Third plan (1972-76)

Development of rural sector-balanced economic growth and stability Deepening industrial structure through promotion of HC industries

Fourth plan (1977-81)

Initially-continued push toward HC industries From 1978/7%shift toward industrial restructuring and price stabilization

Fifth plan Original (1982-86)

Priority to economic stabilization, given expectation of unfavorable domestic and foreign conditions

Revised (1984- 86)

Table 8.4

Shift away from government intervention, including trade and financial liberalization

Composition of Investment During Five-Year Plans (as shares of total, period averages) 1972-76

1977-81

1982

1983

1984

10.6 21.7 32.5 35.2

6.9 18.1 40.6 31.1

6. I 13.6 44.5 35.8

6.1 13.2 46.3 34.1

46.2 32.1

31.2 53.8 15.0

39.6 54.4 6.0

38.9 65.1 -4.0

36.8 69.8 -6.6

38.0 62.6 -0.6

Sectors

Agriculture, forestry, and fishery Manufacturing Services Social overhead capital Type of good Transport, machinery, and other equipment Construction Change in stocks

5.8 15.5

Source: Ministry of Finance, Economic Sfofistics Yearbook, various issues.

40 percent as stock accumulations declined. Construction’s share remained relatively constant. The table shows that the shift toward investment in services continued during the fifth plan.3 However, the increase came at the expense of manufacturing, not agriculture. The counterpart has been a shift from stock accumulation to construction, with equipment retaining 37-38 percent of total investment.

244

Susan M. Collins and Won-Am Park

8.2.2 Five-Year Plans Design This section uses the revised fifth five-year economic and social plan to illustrate how the plans are put t ~ g e t h e r .There ~ are four basic steps described below. Relevant figures are given in table 8.5. The first step is to target a real growth rate. Here, labor force projections implied a required 3 percent per year increase in job openings to maintain employment. Labor productivity was projected to grow at 4.5 percent per year. Thus, a 7.5 percent annual increase in real GNP was targeted for 1984- 86. The second step was to identify the fixed capital formation required to generate the target growth rate. Given the estimated incremental capital output ratio of 4.72, targeted fixed investment in real terms could be calculated. Together with assumptions about inventory behavior, projections about gross investment were formulated. The third step was to make projections about inflation and use them to translate the real targets for output and investment into nominal series. The final step began with the realization that gross investment must be financed through a combination of domestic and foreign savings. The Table 8.5

Revised Fifth Five-Year Plan Fifth Plan

1980 Constant prices GNP growth (%) (marginal capital coefficient) (%) Required fixed capital formation Gross investment Currenr price, GNP ( Y ) Investment (I) Domestic savings (S) Marginal propensity to save (MPS) WY (Yo) SIY (Yo) of which: Household (9%) Corporate (9%) Government (%)

1982”

1983b

1984

1985

1986

5.6

9.3

7.5

7.5

7.5

(5.83)

(3.88)

(4.73)

(4.72)

(4.75)

12,984.5 12,480.6

15,136.4 14,217.3

16,196.0 15,894.6

17,378.3 17,378.3

18,768.5 18,975.5

51,786.6 13,979.8 11.594.0

58,297.7 16,107.2 14,252.2

63,277.2 18,137.9 16.877.1

69.383.5 20,206.9 19,504.9

76,079.0 22.436.6 22,280.6

(0.279) 27.0 22.4 6.6 9.6 6.2

(0.409) 27.6 24.4

(0.525) 28.7 26.7

7.1 10.2 7. I

11.0

Source: Government of Korea (1983).

“Actual. hProjected Note: Figures in billions of won unless otherwise indicated

8.2 7.5

(0.430) 29.1 28. I

(0.415) 29.5 29.3

8.9 11.4 7.9

9.3 11.7 8.3

245

KoredChapter 8

technique for projecting domestic savings is to predict the ratios of household, government, and corporate savings to GNP. As shown, each was expected to rise over time in connection with a variety of measures designed to encourage savings. For example, household savings were predicted to rise in response to expanded financial instruments and banking services. Together with projected nominal GNP, the ratios were used to predict total nominal domestic savings. Foreign savings were then given by the difference between investment and domestic savings. Perhaps the key implication of the way that the five-year plans were formulated is that foreign savings is determined as a residual. The five-year plans also contain detailed projections for current account behavior. Documentation of earlier plans contained projections about the path of external debt, including projected debt service payments, and the allocation of debt between short- and long-term borrowing. Unfortunately, these figures are not readily available for the original or the revised fifth five-year plans. Three questions emerge. First, have the plans successfully achieved their investment targets, and what are the implications for the determinants of investment in Korea? Second, how successful have planners been in forecasting savings? And finally, we look at the other side of the equation to examine the implications for current account behavior and debt accumulation in Korea. Investment Targets

Table 8.6 shows planned and actual investment as a share of GNP, and real growth rates during the fourth and the revised fifth five-year plans. During the fourth plan, investment consistently exceeded target as a share of GNP. It is interesting that this was true both during the beginning of the plan period, Table 8.6

Planned versus Actual Rates of Investment and Growth Real Growth (70)

InvestmenVGNP Total

Fourth plan 1977 1978 1979 1980 1981 Fifth plan 1984 1985

Fixed

Plan

Actual

Plan

Actual

Plan

Actual

10.0 9.0 9.0 9.0 9.0

12.7 9.7 6.5 -5.2

28.0 31.0 36.0 32.0 30.0

-

6.6

27.0 26.3 25.9 25.9 26.0

-

26.7 30.8 32.8 32.3 28.7

7.5 7.5

8.4 5. I

28.7 29. I

29

31.5 31.1

30.5 31.5

~

~

Source: Government of Korea (1976): Government of Korea (1983): and EPB. Major Economic Stofistics. Nure: Dashes indicate that data were not available.

246

Susan M. Collins and Won-Am Park

when real growth rates were higher than projected, and during the second half of the program, when the 1979-80 crisis gave rise to an unanticipated decline in economic activity. What is an appropriate model for the determination of fixed capital formation? The alternative suggested by the preceding discussion is that government policies and incentives essentially set a minimum investment level as part of the five-year plans and they ensure adequate (domestic or foreign) financing for any approved investment project, soliciting enough to ensure that the minimum level is met or exceeded. In such a framework, firms on the periphery are totally at the mercy of market conditions in obtaining financing, however, variations in their position may have little impact on the aggregate investment figures. Fixed capital formation as a share of GNP has not been very cyclical during 1970-85. For example, the investment ratio jumped between the boom year 1973 and the downturn in 1974-75.5 Some authors have focused on credit access as the key to investment determination.6 These models suggest that private credit availability and curb market loan rates be included in regressions to explain investment. In regressions on quarterly data, curb market rates have negative coefficients, but do not tend to be significant. On balance, it is difficult to assess the quantitative importance of the standard neoclassical variables as determinants of investment. However, the role of the government in explicitly allocating credit across industries and to particular firms is clear.

Planned versus Actual Behavior of the Current Account and External Debt When actual economic performance diverges from the five-year plans, the differences have tended to be higher investment than projected, with smaller domestic savings, implying a deterioration of the current account. How have the authorities tended to react to this situation? There are at least two alternatives from which they could choose. They could simply make up the shortfall in financing by increasing external borrowing (presumably through bank loans), and accept the resulting increase in external debt. Alternatively, they could hold firmly onto their projected path of foreign borrowing and finance the current account deficit through a reduction in foreign exchange reserves or, if possible, through foreign direct investment. Table 8.7 shows planned and actual figures for the key variables for 1977 and 1978. In 1977 the plan predicted the trade balance quite accurately, however, a much stronger service account than anticipated (from overseas construction) meant that the current account was about $650 million larger than expected. This favorable outturn was offset by an additional $635 million accumulation of foreign exchange, not by a reduction in external borrowing. In fact, Korea borrowed almost $600 million more than projected. In 1978 the trade balance was much worse than projected. Exports did better, but imports, particularly machinery and transport equipment, rose

247

Table 8.7

KoredChapter 8 External Balance: Planned versus Actual, 1977-78 (in millions of U.S. dollars) 1977

Plan

1978

Actual

Plan

Actual ~~

634 71 1

Current account deficit Reserve accumulation Emrs and omissions

-

Increased foreign debt

1,542

- 12.3 1,346 31.7 2,129

237 611

1,667

~

1,085.2 63 I 312.0 2,174

Source: Government of Korea (1976).

considerably. Korean authorities did not offset the development through reserve depletion to dampen the effect of external debt. Instead, they accumulated foreign exchange reserves, approximately in line with the plan targets. Again, external borrowing exceeded the projection, with an especially large jump in private long-term borrowing. During 1979-8 1 unexpected domestic and foreign developments drastically altered the environment and the economic performance. Arguably, circumstances had changed by so much that the targets and the projections from the fourth plan were no longer relevant, and that it is not meaningful to compare these targets with actual outcomes. It is notable, however, that investment remained high and stable during this period, financed by extensive foreign borrowing. The clear pattern through the fourth five-year plan is one which places investment as the number one priority, financing it with external borrowing whenever necessary, in spite of potential consequences for domestic price stability and the burden of the debt. Since the 1979-80 crisis, the government has stated that economic stabilization has been named the top priority and that concern over debt accumulation would preclude continued treatment of foreign borrowing as a residual. We conclude this section by asking whether there is any evidence of such a shift in policy. Unfortunately, the revised fifth plan does not make the projected debt accumulation explicit. Table 8.8 focuses on the current account and reserves. Errors and omissions, which became large during the early 1980s, are also reported. Again, in 1984 the current account deficit was larger than anticipated, as were errors and omissions. Authorities did not finance some of the deficit with reserves, but accumulated one and a half times the target amount. Presumably, the increase in external debt also exceeded the projection. There is no evidence here of a shift from an approach to macroeconomic management in which external borrowing is residual. The outcomes in 1985 and 1986 are more ambiguous. The 1985 current account deficit was larger than expected, but this time foreign exchange accumulation slowed down, mitigating the implied rise in borrowing. This

248

Susan M. Collins and Won-Am Park

Table 8.8

Reserve Accumulation: Planned versus Actual, 1984-86 (in millions of U.S. dollars) 1985

1984

Plan Current account deficit Reserve accumulation Errors and omissions Foreign reserves

1

.ooo 490

600 7,400

1986

Actual

Plan

Actual

Plan

Actual

1,373 740 894 7,650

300 400 600 7,800

887 99 880 7,749

-400 700 300 8,500

4,617 207 543 7,955

Source: Government of Korea (1983). and EPB, Major Statistics of the Korean Economy, various issues

episode provides some support for a policy shift such that variables other than external debt could adjust to unexpected developments. However, the evidence is not particularly strong when considered cumulatively. The total reserve accumulation during 1984-85 was very close to the cumulated projection, and in that sense there was no adjustment in reserve accumulation. Finally, in 1986 there was a massive external surplus. The plan had predicted a small surplus of $400 million. The actual surplus was more than ten times that figure, enabling Korea to reduce its external debt stock. Although reserve accumulation was smaller than projected, the episode provides little information about which external variables would be treated as residuals if domestic savings were too low to cover investment.

8.3 Disaggregation of Domestic Savings Data In order to accurately examine the determinants of Korean saving behavior, it was important to disaggregate savings. Given the available data, the finest decomposition possible was into three categories: general government, public and private corporations, and households. Unfortunately, the household category includes households, nonprofit institutions, and unincorporated businesses. It is not possible to separately identify these elements. In this section we describe the data and method used to compute the domestic savings figures that were given in tables 8.1 and 8.2. EPB and BOK report domestic savings for government, corporations, and households. These data include net transfers from abroad, however, they exclude allowances for capital consumption, which are reported separately. Flow of funds tables were used to assign depreciation allowances between the three sectors. BOK National Income Accounts statistics also provide a breakdown, however, those figures assume that no depreciation is attributable to the household sector. This is unrealistic, given that this sector includes unincorporated businesses.

249

KoredChapter 9

Two problems arise. First, the total depreciation from the FOF data is consistently smaller than the total given by BOK or EPB in the National Income Accounts. The discrepancy ranged from less than 1 percent of total gross investment to as high as 10 percent in a few years. The average was 3 percent of total gross investment. The discrepancy was assigned to corporate depreciation, which is therefore measured as the residual. The second problem is that of the FOF disaggregation is currently available only through 1982. For 1983 and 1984 the decomposition of depreciation was estimated based on the average shares of each sector in the total during 1976- 82. Korean GDP is computed from the expenditure side. Therefore, the residual appears in the expenditure side of the accounts and is not included in the savings estimates. This explains why the column for statistical discrepancy appears in the tables. Finally, the method for computing National Income Accounts data has recently been revised. The data used in this chapter are based on consistent data, using the old method, through 1984. These figures are unfortunately not comparable with figures based on data using the new method. In particular, the two methods give very different figures for fixed investment as a share of GNP during 1980-84. However, the trends in the two series are similar. Disposable income data were computed from the BOK National Income Accounts. The data subtract direct taxes and net transfers from the household sector to the government and to the rest of the world.

9

Exchange Rate, Trade, and Industrial Policy

The Korean economy has been one of the world’s most rapidly growing economies in recent decades. Since Korea launched its first five-year plan in 1962, it has grown at over 8 percent per year on average. The growth pace has slowed on occasion when the economy was faced with oil shocks and sluggish world demand, but, overall, exports have fueled growth even in periods when adverse situations abroad reduced foreign demand for export goods and raised domestic inflation. Simply on the basis of growth performance, the adoption of an outward-oriented growth strategy in place of import substitution could be considered an epochal change in trade and industrial policies. The period from May 1960 to 1965 is regarded as a time of transition during which Korean trade and industrial policies were reoriented toward

250

Susan M. Collins and Won-Am Park

active export promotion instead of the import substitution of the 1950s. The Chang Myon civilian government took office in May 1960, just after President Syngman Rhee was ousted by the student revolution. The new government opened the way for export promotion by implementing two devaluations. By that time, import substitution in nondurable consumer goods had been completed and further import substitution in machinery, consumer durables, and intermediate products seemed inappropriate due to the small domestic market and the large capital requirements involved. Rather than pursuing the slow growth path with import substitution, policymakers chose export-driven high growth. The military government of General Park that replaced Chang Myon's civilian government reinforced the policy switch toward export orientation by further devaluing the won, extending the scope of export incentives, and reducing quantitative restrictions on imports. Although the government encountered many difficulties, such as accelerating inflation during the transition, it revealed its strong intention to pursue consistent export-promotion policies by implementing monetary and fiscal reforms. Nineteen sixty-four may be identified as the watershed year after which the government depended on a comprehensive, export-driven policy. As a result of the strong export promotion, the nominal value of commodity exports increased about 480 times from U.S. $55 million in 1962 to $26.4 billion in 1985. The annual growth rate of nominal exports averaged 31 percent, substantially higher than the 20 percent growth of nominal imports. Despite this remarkable export growth, the trade balance continued to show a deficit, reflecting the huge initial deficit which was about ten times as large as exports. In this chapter we investigate the role of exchange rate, trade, and industrial policies.' Section 9.1 describes the pattern of changes in trade and industrial structures since 1962. In section 9.2 we investigate the exchange rate policies that were crucial to export promotion. Further discussion of the link between exchange rates, wages, productivity, and competitiveness will be given in chapter 10. Export incentives and import restriction measures are analyzed in section 9.3, with a brief chronology of those measures. In section 9.4 we discuss the industrial policies that have been pursued up to the present in association with trade and growth.

9.1 Changes in the Trade and Industrial Structure Economic development since the early 1960s can perhaps best be described as export-oriented industrialization. Economic growth has been led by exports of manufacturing goods. As can be seen in table 9.1, the rapid contraction of the primary sector was roughly counterbalanced by the rapid expansion of the manufacturing sector, while the tertiary sector constituted a relatively stable portion of GNP. The share of agricultural, forestry, and

251

KoredChapter 9 Industrial Origin of GNP (composition and growth rate, in percentages)

Table 9.1

Agriculture, forestry, and fishing Mining Manufacturing Light Heavy Othersa GNP

1962

1965

1970

1975

1980

1984

43.3 2.0 9.1 6.8 2.3 45.6 100.0

42.9 (7.8) 2.0 (8.7) ll.O(l5.4) 7.4(11.6) 3.6 (25.3) 44.0 (6.8) 100.0 (8.2)

30.4 (1.0) 1.6 (6.0) 17.8(22.6) 10.4(18.9) 7.4 (28.9) 50.1(13.4) 100.0 (9.8)

24.2 (5.3) 2.0 (8.6) 21.6(17.8) 12.3(15.1) 9.3 (22.1) 52.3 (7.1) 100.0 (8.6)

14.4 (-6.1) 1.4 (0.1) 28.8 (11.3) 14.1 (8.4) 14.7 (14.5) 55.4 (7.5) 100.0 (5.7)

15.1 (3.5) 1.4 (4.4) 34.1 (13.4) 15.0 (10.0) 19.1 (16.5) 49.4 (5.3) 100.0 (7.6)

Now: Data prior to 1970 are based on 1975 constant market prices. Data beginning from 1971 are based on 1980 constant market prices. "Others include wholesale and retail trade, restaurants and hotels, social and personal services, transport and communications, and construction.

fishing products in GNP decreased from 43 percent in 1962 to 15 percent in 1984, while the share of manufacturing products increased from 9 percent in 1962 to 34 percent in 1984. The service industry also expanded, but its gain was small compared with changes in the primary and manufacturing sectors. The changing composition of GNP among industries shows that manufacturing grew much more quickly than the GNP growth rate, the primary sector grew more slowly, and other sectors grew at approximately the same rate as GNP. It is also interesting to focus on the development of the manufacturing sector in terms of production factor intensity. The light manufacturing industries that were generally labor intensive had higher export ratios and produced more than the heavy manufacturing industries did until 1980 (tables 9.1 and 9.2). At the same time, it should be noted that the heavy industries grew faster than the light industries after 1962, exceeding the light industrial output by 1980. The export ratio of the heavy industrial output rose constantly, equaling that of the light industrial output by 1983. As we have discussed in previous chapters, there were clearly some unfavorable aspects of the Big Push toward HC industries. We will return to these issues in section 9.4. The rapid expansion of capital-intensive industries in a labor-abundant economy raises issues of efficiency because the industrial mix may result from excessively protectionist policies. In fact, arguments both for and against protection on efficiency grounds were advanced during the debate over the optimal growth strategy: import substitution or export promotion. As we shall see, protectionism did play a role in the development of the HC industries. Korea's experience provides an interesting episode to be used in assessing favorable versus unfavorable aspects of extensive government intervention. In table 9.3 we describe changes in Korea's manufacturing

lsble 9.2

Value-Added, Export, and Import Dependency Ratios by Industry (in percentages) Value-Added Ratio

Agriculture, forestry, and fishety Mining Manufacturing Light Heavy Average for all industries

Export Ratio

Import Dependency Ratio

1970

1975

1980

1983

1970

1975

1980

1983

1970

1975

1980

1983

74.0 74.9 25.9 (24.1) (30.1)

75.8 69.4 22.8 (21.5) (24.7)

69.8 68.6 22.8 (23.4) (22.2)

69.6 65.3 23.9 (24.4) (23.5)

2.7 19.8 10.6 (11.9) (7.4)

5.8 12.4 18.2 (20.8) (14.5)

5.6 5.5 19.2 (20.8) (17.6)

5.0 3.2 21.2 (21.7) (20.7)

1.1 1.6 17.3 (12.3) (29.1)

2.4 4.0 22.0 (14.2) (33.1)

2.2 0.6 22.7 (15.0) (30.2)

1.4 0.9 22.2 (14.7) (28.2)

49.7

42.5

39.6

40.5

6.9

12.4

13.3

14.4

8.6

12.8

14.2

13.5

Source: BOK, Cornpilatory Report on 1980 and 1983 Input-Output Tables, 1983, 1985 Note: Value-added ratio = (value-added amountidomestic production) x 100. Export ratio dependency ratio = (amount of imported intermediate inputidomestic production) x 100.

=

(export amountidomestic production) x 100. Import

253

Table 9.3

KoredChapter 9 Value-Added Ratio in Principal Manufacturing (composition, in percentages)

Light industries Food. beverage, and tobacco Textile, wearing apparel, and leather Wocd and wood products, including furniture Printing and publishing Rubber and plastic products Miscellaneous" Heavy industries Paper and paper products Chemical and petroleum products Nonmetallic mineral products Basic metal Fabricated metal products, machinery, and equipment Manufacturing

1%2

1965

1970

1975

1980

1984

(74.2) 37.7 23.0

(67.0) 35.6 18.2

(58.2) 30.2 16.4

(56.9) 24.1 20.6

(49.0) 19.8 17.1

(43.9) 18.5 15.1

3.2

2.7 4.8 1.9 3.9 (33.0) 3.1 11.6 5.4 2.9

2.7 2.6 2.3 4.1 (41.8) 2.5 21.6 5.6 2.4

1.7

4.8 1.8 3.8 (25.8) 2.8 5.0 4.5 2.9

3.4 5.2 (43.1) I .9 16.2 5.1 4.6

1.3 1.8 4.1 4.9 (51.0) 1.9 17.5 5.1 6.5

1.1 1.5 4.2 3.6 (56.1) 1.8 15.1 4.8 7.0

10.6 100.0

10.0 100.0

9.8 100.0

15.4 100.0

20.0 100.0

27.4 100.0

2.0

Source: BOK, National Income Accounts, 1985. Note: Data prior to 1970 are based on 1975 constant market prices. Data beginning from 1971 are based on 1980 constant market prices.

'Includes miscellaneous products of petroleum and coal and professional and scientific measuring and controlling equipment.

structure. The most salient features were a decrease in the share of food, beverage, and tobacco industries among the light manufacturing and an increase in the share of fabricated metal products, machinery, and equipment among the heavy manufacturing. This pattern of change in industrial structure is also reflected in the changing pattern of exports and imports by commodity group. In 1962, exports of food, live animals, and crude materials such as iron, tungsten ores, raw silk, and agar-agar, were 75 percent of total commodity exports. In 1985 these commodities of SITC Group 0 and 2 were only 5 percent of total exports, whereas commodities of SITC group 6- 8-manufactured material goods (iron and steel products, textile fabrics, textile yams, and plywoods), machinery and transport equipment, and miscellaneous manufactures (mostly clothing and footwear)--constituted 89 percent of total exports (table 9.4). The changes in imports by commodity groups were not as distinctive as those in exports, but imports of some commodities such as mineral fuels, mainly comprised of petroleum, and machinery and transport equipment, increased quite rapidly as shown in table 9.5. The increase in mineral fuels can be largely explained by the quadrupling of oil prices in 1974 which led to about a threefold increase in the share of crude oil (table 9.6). The share increased further in 1980 as a result of the second oil shock and thereafter declined with the fall in oil prices to reach 18 percent in 1985.

254

Susan M. Collins and Won-Am Park

Table 9.4

Exports by Commodity Groups (composition, in percentages)

SITC Group

1962

1965

1970

1975

1980

1985

(0)Food and live animals

40.0 0.3 35.2

16.1 0.5 21.2

7.8 1.7 12.0

11.9

1.3 2.8

6.6 0.7 1.9

3.8 0.4 1.0

5.0 0.1 1.8 11.3 2.6 3.6 0.1 100.0

I.! 0.0 0.2 37.9 3.1 19.7 0.1 100.0

1.0 0.0 1.4 26.4 7.4 42.2 0.0 100.0

2.2 0.0 1.3 29.4 15.0 35.8 0.2 100.0

0.3 0.1 4.3 35.7 20.3 29.9 0.3 100.0

3.1 0.0 3.1 23.3 37.6 27.6 0.1 100.0

(1) Beverages and tobacco (2) Crude materials, inedible (except fuels) (3) Mineral fuels, lubricants, and related materials (4) Animal and vegetable oils and fats (5) Chemicals (6) Manufactured goods classified by material (7) Machinery and transport equipment (8) Miscellaneous goods (9) Not classifiable Total

Source: EPB, Major Statistics of Korean Economy, various issues

Table 9.5

Imports by Commodity Groups (composition, in percentages)

SITC Group

1962

I965

1970

1975

1980

1985

(0)Food and live animals (1) Beverages and tobacco (2) Crude materials, inedible (except fuels) (3) Mineral fuels, lubricants, and related materials (4) Animal and vegetable oils and fats (5) Chemicals (6) Manufactured goods classified by material (7)Machinery and transport equipment (8) Miscellaneous goods (9) Not classifiable Total

11.5 0.0 21.3

13.7 0.0 23.7

16.1 0.1 0.4

13.0 0.2 15.4

8.1 0.4 16.3

4.5 0.2 12.4

7.3 0.9 22.4 17.3 16.5 2.4 0.3 100.0

6.7 0.8 22.3 15.3 15.9 1.5 0.0 100.0

6.9 0.8 8.3 15.4 29.7 2.4 0.0 100.0

19.1 0.7 10.7 11.9 26.5 2.3 0.1 100.0

29.9 0.5 8.1 11.0 22.4 3.1 0.3

100.0

23.6 0.5 9.0 11.4 34.2 4.0 0.3 100.0

Source: EPB, Major Stutistics of Korean Economy. various issues

Table 9.6

Imports by Type of Goods (composition, in percentages)

Capital goods Raw materials for export Raw materials for domestic use and others Crude oil Total

1962

1965

1970

1975

1980

1985

16.5

12.9 2.2 78.6 6.2 100.0

29.7 19.5 44.1 6.7 100.0

26.2 20.0 36.3 17.5 100.0

23.0 17.0 34.7 25.3 100.0

35.6 21.9 24.5 17.9 100.0

-

76.7 6.7 100.0

Source: EPB, Major Statistics of Korean Economy. 1982. 1986.

255

KoredChapter 9

The share of machinery and transport equipment among total imports in table 9.5 is roughly the same as the share of capital goods in table 9.6, since machinery and transport equipment are imported to be used largely as capital goods. The share of capital goods did not increase steadily but was quite variable, even if imports of crude oil are excluded from total imports. A large part of the variability of capital goods can be explained by heavy industrialization in the 1970s and, more importantly, by imports of raw materials for exports or domestic use, which in turn are sensitive to fluctuating raw material prices. The input-output structures that were summarized in table 9.2 suggest that the variability of the capital goods share should not be attributed to changes in the input-output structure. The import dependency ratio increased sharply between 1970 and 1975 and, as a mirror image, the value-added ratio decreased by a substantial margin during the same period. Afterward, however, the two ratios remained relatively stable, although import dependence was higher in the late 1970s than in the 1980s.

9.2 Exchange Rate Policies After export promotion policies were adopted in the early 1960s, the exchange rate emerged as a major economic policy variable with significant influence on the volume of exports and imports. Before the government shifted from import substitution to actively promoting exports, the exchange rate remained overvalued. Two arguments for the overvaluation during the 1950s were to avoid inflation acceleration and to earn more foreign currency in exchange for the sale of won currency to UN forces. As Korea’s major exports were primary goods such as tungsten ore and agar-agar, policymakers overlooked the export incentives from devaluation, depending instead on tight quantitative restrictions on imports. Exporters were granted import rights and could obtain foreign exchange premiums on the domestic market. Furthermore, they were provided with sizable export subsidies. This exchange rate policy was altered in the early 1960s as foreign aid was reduced and the government dedicated itself to the goal of export-driven growth. In 1961 the official exchange rate was devalued 104 percent from 62.5 won to 127.5 won per dollar. This drastic devaluation contributed to absorbing the import premiums caused by quantitative controls and unifying the multiple exchange rates for commodities. However, the expansionary monetary and fiscal policies of the military government caused accelerating inflation. This lessened the effect of currency devaluation, so that another devaluation was soon needed to depreciate the real exchange rate. The second large devaluation from 130 won to 256 won per dollar was carried out in 1964, but it was accompanied by fiscal and monetary reforms to reduce the inflationary pressure of devaluation.

Susan M. Collins and Won-Am Park

256

In March 1965 the government implemented a floating, unified exchange rate to maintain real exchange rate stability. However, the regime was interrupted by government intervention near the end of 1965, after which the won was maintained at approximately 271 won per dollar. The real exchange rate appreciated by 14 percent during this period. As shown in table 9.7, nominal depreciations maintained a constant real exchange rate during 1968-71, These adjustments were accelerated during 1971-73, leading to a 32 percent real depreciation. During the Big Push (1973-79), the focus of Korean exchange rate policy shifted from maintaining competitiveness. Inflation had jumped to 24 percent during 1974, following the rise in oil prices. Although the won was devalued by 21.3 percent vis-8-vis the U.S. dollar in December 1974, it was then pegged at 484 won/$ until January 1980, even though inflation averaged

Table 9.7

Year

Exchange Rates (in won per U.S dollar) and Terms of Trade

Nominal Exchange Ratea

1963 130.00 255.00 1964 271.00 1965 270.00 1966 1967 268.00 276.65 1968 288. I6 1969 310.56 1970 1971 347. I5 392.89 1972 398.32 1973 404.47 1974 1975 484.00 484.00 1976 1977 484.00 484.00 1978 484.00 1979 07.43 1980 68 1.03 1981 731.08 1982 775.75 1983 805.98 1984 1985 870.02 881.45 1986 Coefficient of variationC

Index of Exchange Rate

Nominal Effective Exchange Rateb

Purchasing Power Parityb

Real Effective Exchange Rateb

21.40 41.98 44.61 44.45 44.12 45.54 47.44 51.13 57. I5 64.68 65.58 66.59 79.68 79.68 79.68 79.68 79.68 100.00 112.12 120.36 127.71 132.69 143.23 145.I 1 0.44

18.20 34.78 36.96 36.19 35.32 36.58 37.88 41.09 46.78 56.04 60.40 59.46 70.57 70.37 74.15 83.18 80.38 100.00 117.54 118.66 126.62 129.85 138.47 164.37 0.52

95.42 139.33 138.11 130.96 123.83 119.83 119.81 122.64 127.16 129.78 141.76 126.16 125.90 118.15 113.34 104.40 96.31 100.00 98.87 104.67 111.51 116.91 125.39 122.93 0.12

81.16 115.43 114.40 106.63 99.12 96.25 95.66 98.56 104.09 112.44 130.58 112.65 111.50 104.35 105.47 108.98 97.15 100.00 103.65 103.19 110.56 114.41 121.22 139.24 0.10

Terms

of Trade 111.00 111.90 114.40 127.70 132.20 137.70 132.60 133.80 132.70 132.10 125.40 102.10 92.10 105.10 112.40 117.80 115.30 100.00 97.90 102.20 103.10 105.30 105.90 114.7

“Yearly average as weights each year’s trade volume of seven major trading partners, i.e., U S . , West Germany, Netherlands, Japan, U.K., Canada, France. Purchasing power parity = (index of exchange rate) x (relative price). Real effective exchange rate = (nominal effective exchange rate) x (relative price). ‘Standard deviationhear. It covers the period 1963-85

257

KoredChapter 9

16.6 percent during 1975-79. Consequently, Korea’s real exchange rate appreciated by 13 percent. The pegging of the won to the U.S. dollar ended in 1980 with Korea’s 20 percent devaluation in January and the subsequent adoption of a new exchange rate regime called a double basket system in February. Under the new regime, the won-dollar exchange rate was to be determined on the basis of movements of the exchange rates of major trading partners and other factors affecting Korea’s external position. With this currency basket system, the exchange rate has been managed more flexibly to maintain external competitiveness. As shown in table 9.7, significant further real depreciations occurred during 1985 and 1986. It is notable that the won continued to depreciate against the U.S. dollar during much of the dollar’s 1985-86 depreciation. Thus, two major switches of the exchange regime-from unified float to dollar peg in 1974, and from dollar peg to basket system in 198&involved one-shot devaluations of approximately 20 percent. Each was part of a policy package in response to current account difficulties following an oil price shock. However, in contrast to the devaluation in 1974 which ended in a five-year peg, there was a further substantial real depreciation during 1983-86. Korea has succeeded in maintaining external competitiveness throughout most of the period since 1962. Notice that the 1980 base year follows a 20 percent nominal devaluation and that the real exchange rate was considerably more depreciated than this base during the mid-l960s, most of the 1970s, and since 1983. Korea’s experience with real exchange rates contrasts sharply with that of its Latin American counterparts. Table 9.8 shows the variable real exchange Table 9.8

Real Exchange Rates in Latin America (idex 1980-82= 100)

Year

Argentina

Brazil

Chile

Mexico

Venezuela

1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

66 81

123 122 119 108 97 85 I03 I12 86 86 85 75

66 74 79 12 79 95 108 97 87

107 I06 93 94 98

94 97 96 93 89 91 109 109 116 86 93 94

64 74 101 I16 107 76 71 80 71 63

90 79 72

Source: Morgan Guaranty (1987); also from Dornbusch (1986). Nore: Higher values mean real appreciation.

104

114 82 78 92 90 68

258

Susan M. Collins and Won-Am Park

rates in Latin America. In particular, Argentina, Chile, and Mexico experienced extremely variable real exchange rates and, even more damaging, massive appreciation during 1978-81, the years prior to the current debt crisis. Korean’s real effective exchange rate also showed variations. We divided the period from 1960 to 1985 into five subperiods as shown in table 9.9, according to the behavior of the real exchange rate. The real appreciation during the 1964-69 and 1973-79 periods stemmed largely from the slow depreciation of the nominal exchange rate. For example, the nominal exchange rate depreciated by only 2.5 percent per year during 1964-69 and 3.3 percent per year on average during 1973-79, while it depreciated 40.7 percent annually during 1960-64 and 8.4 percent during 1969-73. The reason for this is that the government was concerned about the domestic inflation caused by an exchange rate depreciation. The authorities tried to avoid further devaluation whenever they thought that export incentives, other than currency devaluation and the favorable external conditions, would allow them to achieve the year’s export target. We return to this point below.

9.3 Trade Liberalization Policies Developing countries typically implement exchange rate policies with accompanying changes in trade measures other than an exchange rate adjustment. Sometimes commercial policies (export subsidies, tariffs, and quotas) have a greater impact on trade than exchange rate policies. The complicated picture of exchange rate with subsidy policies has been a focal point of analysis. Korea is not an exception in this respect. Its periodic devaluation has sometimes been accompanied by enlarged export subsidies or tightened quantitative restrictions on imports. But, from a long-term perspective, Table 9.9

Period 1960-64 1964-69 1969-73 1973-79 1979-85

Exchange Rates, Wages, and Productivity in Manufacturing (average annual percentage change)

Nominal Depreciation

Real Depreciation

Nominal Wage

Consumer Price Index’

Real Wageb

Labor Productivity‘

Unit Labor Cost(W)

Unit Labor Cost($Id

40.7 2.5 8.4 3.3 10.3

-3.7 8.1 -4.8 3.8

13.6 23.8 18.6 32.3 14.5

15.6 11.9 11.0 17.9 10.4

-1.7 10.6 6.9 12.2 3.7

7.5 16.9 9.9 11.5 11.2

5.8 5.9 7.9 18.6 3.0

-24.8 3.3 -0.5 14.9 -6.6

‘Using consumer price index for Seoul City for the 1960-64 period bNominal wageiconsumer price index. ‘Korea Productivity Center figures. dunit labor cost in won currencyinominal exchange rate.

259

KoredChapter 9

foreign trade has been liberalized throughout the period. In particular, the periods 1965-67 and 1978-79 exhibited rapid liberalization mainly due to the relaxed quantitative restrictions on imports. It is interesting, however, that real appreciation of the currency occurred during those periods. In this section we describe briefly the export and import liberalizations in terms of export subsidies, import restrictions, and tariffs. 9.3.1 Export Subsidies The concern here is how important the various export incentives were to the management of the official exchange rates. In table 9.10 we summarize net and gross export subsidies estimated in terms of won subsidies per U.S. dollar of export. The export incentives considered in the table are: (1) direct cash subsidies (abolished since 1964); (2) export dollar premium attained by linking export performance to imports (abolished since 1964); (3) direct tax reduction on income earned from exporting (abolished since 1973); (4) export credit at preferential interest rates (abolished since 1982); (5) indirect domestic tax exemptions on intermediate inputs used for export production and export sales; and (6) tariff exemptions on imports of raw materials for export production (drawback system). The first four incentives (col. 2-5 in table 9.10) represent subsidies which directly decrease the costs of exporting firms. As shown, there was a dramatic reduction in these direct subsidies over 1960-64, primarily because of the elimination of special exchange rates to exporters. Since 1965 the subsidies have come primarily through preferential interest rates. The subsidies ranged from 6 to 7 percent during 1968-71 and 2 to 3 percent during 1972-81, and were eliminated during 1982-83 (see col. 10). Declines in direct subsidies have been partially offset by indirect tax exemptions and tariff exemptions. As shown in column 11, there is no persistent trend in the gross subsidy to exporters between 1961 and 1980. It rose to a high of 30 percent in 1971, declining to 17 percent during 1975-76 and returning to 21 percent by 1980. It is interesting that exchange rate policy has often offset reductions in export subsidies. This was particularly true during 1967-73. From 1967 to 1970 the real exchange rate was appreciating (see table 9.7), however, other incentives for exporters (tax benefits, interest rate preferences, and tariff exemptions) were all increased. During 1970-73 the decline in export subsidies coincided with a real depreciation. During 1961-62, net export subsidies declined greatly because of the decrease in export dollar premiums as the government devalued the domestic currency and unified the exchange rates. As inflation negated the effect of the two devaluations in 1961, the government reintroduced the full-scale export-import link system in 1963 under which nonaid imports were constrained to export earnings. The export dollar premiums again became the major content of export subsidy during 1963-64.

Table 9.10

Export Subsidies, 1958-85 (annual averages) Won Subsidies Per U . S . Dollar of Export

Official Exchange Rate(won/$) Year

(1)

1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1917 1978 1979 1980 1981 1982 1983 1984 1985

50.0 50.0 62.5 127.5 130.0 130.0 214.3 265.4 271.3 270.7 276.6 288.2 310.7 347.7 391.8 398.3 407.0 484.0 484.0 484.0 484.0 484.0 618.5 686.0 737.7 781.2 807.1 871.7

Direct Cash Subsidies (2) .0 .0

.o 7.5 10.3 4.1 2.9

.o

.o .0

.o .o .o .0

.o

Export Dollar Premium (3) 64.0 84.7 83.9 14.6

.o 39.8 39.7

.o .0

.o .o .o .0

.o

.o

.0 .0 .0

.0 .0

.o .o

.o .o .o

.0 .0

.0

.0 .0

.o .o

.o .0

.o

.0

.0 .0 .0

.o

.o

Direct Tax Reduction (4) .0 .0

.o .0 0.6 0.8 0.7 2.3 2.3 5.2 3.0 3.7 3.5 4.8 I .9 1.4

.o .0

.o .o .o .o .o .o .o .0

.o .o

Interest Rate Preference (5)

1.2 I .3 I .2 1 .o

.9 2.9 6.0 7.6 10.3 14.7 15.2 14.7 17.3 18.I 10.5 7.4 8.6 12.9 12.3 9.4 11.0 11.0 20.6 15.0 3.0

.o .0 .0

Net Subsidies (6= 2 + 3 + 4 + 5 ) 65.2 86.0 85.1 23.1 11.8 47.6 49.3 9.9 12.5 20.0 18.2 18.4 20.8 22.8 12.5 8.7 8.6 12.9 12.3 9.4 11.0 11.0 20.6 15.0 3.0

.o .o .0

Source: Westphal and Kim (1977) for 1962-75 data; Nam (1981) for 1976-78 data; and K. S. Kim (1986) Note: n.a. = not available.

Ratio to Exchange Rate(%)

Indirect Tax Exemptions (7)

Tariff Exemptions (8)

.o .o .o

.0 .0

.0 5. I 5.3 7.6 13.9 17.8 17.8 19.9 27.4 27.0 32.2 26.4 21.0 22.5 33.8 33.6 53.1 53.6 56.6 74.6 n.a. n.a. n.a. n.a. n.a.

.0 4.7 6.6 10. I 15.4 21.3 24.6 39.6 34.3 40.4 48.0 66.3 64.4 55. I 34.3 35.9 30.6 30.0 30.3 36.4 n.a. n.a. n.a. n.a. n.a.

.o

Gross Subsidies (9=6+7+8j

Net Subsidies (10=6/1j

Gross Subsidies (11=9/1)

65.2 86.0 85. I 23.1 21.6 59.5 67.0 39.2 51.6 62.4 77.7 80. I 88. I 103.0 105.2 94.2 86.3 81.0 81.8 93.1 94.6 97.9 131.6 n.a. n.a. n.a. n.a. n.a.

130.4 172.0 136.2 18.1 9. I 36.6 23.0 3.7 4.6 7.4 6.6 6.4 6.7 6.6 3.2 2.2 2.1 2.7 2.5 1.9 2.3 2.3 3.3 2.2 .4

130.4 172.0 136.2 18.1 16.6 48.8 31.3 14.8 19.0 23. I 28.1 27.8 28.4 29.6 26.9 23.7 21.2 16.7 16.9 19.2 19.5 20.2 21.3 n.a. n.a. n.a. n.a. n.a.

.o .0 .0

261

KoredChapter 9

Substantial changes occurred in 1965. The drastic devaluation in 1964 and the following transition to a system of unified floating exchange rates choked off export dollar premiums, but the interest rate reform of 1965 substantially widened the interest rate differential between export credits and nonpreferential bank loans. Thus, export credit at a preferential interest rate became a major subsidy from 1965 until 1982. The tariff exemptions on raw material imports for exports and indirect domestic tax exemptions on intermediate inputs for exports substituted for export dollar premiums, and so gross subsidies remained substantial, ranging from 17 to 30 percent during the 1970s. 9.3.2 Import Restrictions The most important aspect of import liberalization has been the loosening of quantitative restrictions. Government approval was required for all imports until 1955. As can be seen in table 9.11, the semiannual trade program in the first half of 1955 listed 207 import-permissible commodities. Among these, imports of 22 commodities were restricted to prior approval of the government and imports of the other commodities were automatically approved. The degree of import liberalization, measured as a ratio of the number of automatically approved items to the number of total permissible items, has not shown a clear pattern of increase or decrease. The number of automatically approved items increased rapidly until 1958 as foreign aid replenished the foreign exchange resource for imports. After 1958 no obvious trend in import liberalization could be found until 1965. The government started to follow a consistent trade policy of liberalization as of 1965. The number of automatically approved items increased from 1,447 in the first half of 1965 to 2,950 in the first half of 1967. In the second half of 1967 the government changed the semiannual trade program from a positive list system into a negative list system, a major step for import liberalization. Under the negative list system, only the prohibited or restricted items were listed, so imports of nonlisted items were considered to be automatically approved. Introducing the new system, the government adopted the commodity classification method of the UN’s Standard International Trade Classification (SITC). No specific method of commodity classification had been applied under the positive list system. The system of commodity classification again changed from SITC to the four-digit Customs Cooperation Council Nomenclature (CCCN) in the second half of 1977, followed by the eight-digit CCCN in the second half of 1981. The adjusted import liberalization ratios based on the four-digit CCCN are provided in the last column of table 9.12 for comparison. Based on this index, the rate of import liberalization jumped from about 12 percent in the first half of 1967 to 59 percent in the second half of that year with the introduction of the negative list system.

Susan M. Collins and Won-Am Park

262

Import Restrictions by Semiannual Trade Program, 1955-67 (in number of commodities)

Table 9.11

Period 1955:I

II 19563

U 1957:I I1 1958:I 11 19591

II 19603 U l%l:I

U 1%2:I U 1%3:1 I1 1964:I

II 1%5:1 I1 19663 I1 1967:I

Import Permissible (1) 207 298 558 1,145 1,678 1,916 2,243 2,155 2,296 1,812 1,836 1,878 1,581 1,132 1,314 1,498 1,489 1,033 1,124 4% 1,558 1,633 2.240 2.446 3,082

Restricted (2) 22 51 172 242 29 1 282 410 562 125 622 619 613 35 17 119 121 713 924

* *

111 138 136 139 132

Automatic Approval iAA) (3=1-2) 185 247 386 903 1,387 1,634 1,833 1,593 1,571 1,190 1,217 1,265 1,546 1,015

1,195 1.377 776 109 1,124* 496' 1,447 1,495 2,104 2,307 2,950

Prohibited (4)

** ** ** ** ** **

** ** 356 297 315 326 305 355 366 433 442 414 617 631 624 620 583 386 362

(5 = 1 +4)

Index of Number of AA items (1967:1= 100) (6)

207 298 558 1,145 1,678 1,916 2,243 2,155 2,652 2,109 2,151 2,204 1,886 1,487 1,680 1,931 1,931 1,447 1.741 1,127 2,182 2,253 2,823 2,832 3,444

6.3 8.4 13.1 30.6 47.0 55.4 62.1 54.0 51.4 40.3 41.3 42.9 52.4 34.4 40.5 46.7 26.3 3.7 38.1 16.8 49.1 50.7 71.3 78.2 100.0

Total

Source: Korean Ministry of Commerce and Industry, Semi-AnnualTrade Programs, for respective periods. Also, K. S. Kim (1986). *Not divided between automatic approval items and restricted items. **Not specified.

Following the substantial progress toward import liberalization in 1967, the pace for liberalization slowed down until 1978-80. The push toward liberalization resumed in the second half of 1981. By 1985 the liberalization ratio had reached 87.7 percent. The government planned to raise the ratio to 95.2 percent by 1988. The ratio of automatically approved items to total commodity items has limitations as the true measure of import liberalization, and it may overestimate the extent of liberalization. Nonetheless, it points to the general trend in Korea's import liberalization when detailed data are difficult to obtain.

263

KoredChapter 9 ~

lsble 9.12

Period (Original Program)' 1967:I 11 1%8:I

I1 1%9:I

U 19703 I1 1971:I 11 1972:I

I1 1973:I

I1 1974:I

I1 197% 11 19761 I1 1977:I

I1 19783 11

19793

Import Restrictions by Semiannual Trade Program, 1%7-86 (in number of commodities)

Prohibited

Restricted

42 118 116 71 71 75 74 73 73 73 73 73 73 73 73 73 71 66 66 64 63 (54) 54 50

1,114 402 386 479 508 514 530 526 524 518 570 571 569 556 570 574 592 602 584 579 580 (499) 4% 458 424 349 (335) 327 312 312 (2,282) 1,886 1,769 1,482 1,203 970

-

-

I1 19801 1980IV1981:I 1981:IV 1982:I 1982:11/1983:1 1983:IV1984I 1984:IV 1985:I 1985:11/1986:1

-

-

-

Automatic Approval (A) 156 792 810 756 728 723 708 713 715 721 669 668 670 683 669 665 649

644 662 669 669 (544) 547 589 673 748 (675) 683 698 693 (5,183) 5,579 5.79 1 6,078 6,712 6,945

Rate of Import Liberalization (%)

Total (B)

NB

Adjustedb

1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 1,312 (1,097) 1,097 1,097 1,097 1,097 (1,010) 1,010 1.010 1,010 (7,465) 7,465 7,560 7,560 7,915 7,915

11.9 60.4 61.7 57.6 55.5 55.1 54.0 54.3 54.5 55.0 51.0 50.9 51.1 52.1 51.0 50.7 49.5 49.1 50.5 51.0 51.0 (49.6) 49.9 53.7 61.3 68.2 (66.8) 67.6 69.1 68.6 (69.4) 74.7 76.6 80.4 84.5 87.7

11.6 58.8 60.0 56.0 54.0 53.6 52.5 52.8 53.0 53.5 49.6 49.5 49.7 50.7 49.6 49.3 48.2 47.8 49.1 49.6 49.6 49.9 53.7 61.3 68.2 69.1 70.6 70.1 75.5 77.4 81.2 85.4 88.6

Source: Korean Ministry of Commerce and Industry. Also, fmm K. S. Kim 1986. Note: The classification of import items was based on SITC basic codes through the first half of 1977, on the four-digit CCCN codes during 1977-81 (until the first half), and on the eight-digit CCCN codes thereafter. Figures in parentheses indicate the number of commodity items based on the new system of classification used beginning in the following period.

Wriginal import program based on positive list system i s reclassified to make it comparable with the trade program for the following periods which are based on a negative list system. %e rate of import liberalization is adjusted to make it comparable over time on the basis of the same system of Classification as the four-digit CCCN codes (1,097 items) used during 1977-79 (until the first half).

264

Susan M. Collins and Won-Am Park

9.3.3 Tariffs There have been eight major tariff reforms from 1952 to 1984, as shown in table 9.13, during which the simple average tariff rate initially rose and then gradually declined from the peak of 40 percent in 1962 to 22 percent in 1984. The coefficient of variation in tariff rates showed only minor change. Although the simple regular tariff rate decreased by 18 percent during 1962-84, the actual tariff rates (estimated by dividing actual collections of tariffs and equivalents by the won value of commodity imports) did not show any discernible decreasing trend during the same period (table 9.14). This implies that the imports of high tariff items increased. On the other hand, the actual tariff rates were substantially below the legal tariff rates. The difference between the average legal rate and the average actual rate can be explained by the large portion of raw material and machinery and equipment imports on which tariff charges were exempted or reduced. 9.3.4

Summary

To summarize, exporters have received substantial incentives throughout most of Korea’s industrialization. With the exception of the overvaluation during the Big Push, the exchange rate has been maintained at extremely competitive levels. Furthermore, exports consistently received special subsidies throughout the 1960s and 1970s. While the total gross subsidy has remained relatively constant, the major sources shifted from direct export premiums to access to loans at preferential rates and to indirect exemptions from taxes and tariffs. On the import side, the broad characterization highlights three periods. The years 1967-68 were an early period of substantial liberalization. The Big Push in 1973-79 was a period of retrenchment and increased restrictiveness. Since 1979, as part of an overall policy shift away from interventionism, import liberalization has been resumed. It is, of course, difficult to quantify the actual effect of changes in regulations, but most observers conclude that the trade liberalization has been substantial. For Structure of Regular Tariff Rates, 1952-84

Table 9.13 Year

Simple Average ( W )

Coefficient of Variation

1952 1957 1962 I968 1973 1977 1979 1984

25.4

0.70 0.70 0.77

30.3 40.0 39.1 31.5 29.7 24.8 21.9

0.71 0.70 0.61 0.69 0.61

Source: MOF, TurifSchedules of Korea, various years. Also, from K . S. Kim (1986)

265

KoredChapter 9

Table 9.14

Source:

Estimation of Actual Tariff Rates, 1958-85 (in millions of current won and percentages)

Year

Actual Collections of Tariffs and Equivalents

Imports"

Actual Tariff Rates (%)

1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

4,394 8,281 10,196 5,557 6,824 6,358 8,231 12,576 17,635 25,413 37,881 44,724 50,924 52,187 59,106 82,371 126,698 181,004 275,512 385 3 7 1 646,425 732,294 766.063 890,615 1,012,564 1,463,200 1,593,959 1,686,852

18.910 15.190 22,328 41,093 54,834 72,839 103,526 126,091 194,503 273,557 411,806 555,286 628,333 893,792 1,006,026 1,685,519 3,316,271 3,520,810 4,246,422 5,232,282 7,246,400 9,843,882 14,710,292 18,305,045 18,158,999 20,835,895 25,344,454 27,716,961

23.2 54.5 45.7 13.5 12.4 8.7 8.0 10.0 9.1 9.3 9.2 8. I 8. I 5.8 5.9 4.9 3.8 5.1 6.5 7.4 8.9 7.4 5.2 4.9 5.6 7.0 6.3 6. I

K. S. Kim (1986).

aMerchandise imports in U.S. dollar terms multiplied by the official exchange rate for respective years

example, Y. C. Park (1985a) argues that liberalization has proceeded much further in international trade than in financial markets. We return to this discussion below. 9.4

Industrial Policies

Industrial policies have interacted with export promotion strategies in promoting Korea's rapid growth. Each phase of the development can be distinguished by policy concern about specific industries and by active provision of tax and financial incentives. A major aim of these protective measures was to enhance labor productivity and thereby increase export competitiveness. In particular, the government promoted HC industries during the 1970s. However, this focus led to imbalances among industrial sectors and to high inflation, and contributed to the deterioration of the trade

266

Susan M. Collins and Won-Am Park

balance. This section turns to a discussion of Korea’s industrial policy and to the background which gave rise to policy changes.*

9.4.1 Export-Oriented Growth, 1962-72 With the initiation of the first five-year economic development plan in 1962, Korea embarked on a policy of outward-oriented growth. The new government took the initiative in redirecting the economy from the import substitution of the 1950s to export promotion based on the abundant labor supply. The first five-year plan set up as major objectives the development of basic infrastructure, the modernization of the industrial structure, the development of several key raw-material-supplying industries, and the growth of exports. Measures adopted to achieve these objectives included the sizable devaluation of the domestic currency, the liberalization of quantitative import controls, and monetary and fiscal reforms. Various export incentives were also introduced-export loans at preferential interest rates (abolished since 1982), tariff exemptions on raw material imports for exports, and the reduction of corporate taxes for exporters (abolished since 1973). These policies of the 1960s contrasted with those of the 1950s in that the government played an active role in export promotion. Major policy changes were adopted to improve the trade structure, and the industrial policies implemented during this period were closely interrelated with trade policies. Although import substitution was also accomplished in key raw-materialsupplying industries, such as fertilizers, petroleum refining, cement, and chemical fibers, by offering tax incentives, there was little intervention in specific export items or industries. Export incentives were provided for all export-oriented industries. Even import restrictions were not used to protect specific industries. They were generally prohibitive. Furthermore, importsubstituting efforts were focused only on very infant industries such as fertilizers and chemical fibers where foreign investors were motivated by the incentives of domestic tax exemptions. But few foreign investments were made in the 1960s. Foreign investors showed little interest in Korean markets until the early 1970s when the various incentives, including the opening of a free export zone, took effect.

9.4.2 Shift to Import Substitution, 1973-79 In the early 1970s the external situation turned unfavorable for Korea’s export-led growth. The commodity boom brought the gloomy prospect of importing raw materials at an increasing price. The breakdown of the Bretton Woods system aroused worldwide protectionist sentiments rather than stimulating free trade. In addition, Korea’s comparative advantage based on labor-intensive exports was threatened by the emergence of new competitors such as China. These unfavorable economic conditions abroad

267

KoredChapter 9

coincided with the announcement of the Nixon doctrine in 1971 in which a partial withdrawal of American forces from Korea was threatened. Faced with these adverse situations, the president of the Republic called for the development of HC industries in January 1973 to strengthen export-led growth and to develop national defense industries. The major items of new measures taken to develop the “strategic” industries were as follows. First, quantitative restrictions on imports were reinforced for heavy industries and even more so for the strategic industries. The (nonadjusted) import liberalization ratio decreased from 61.7 percent in 1968 to 50.5 percent in 1976 (see table 9.12). For the machinery industry, which includes most of the strategic industries such as industrial machinery, electronics, automobiles, shipbuilding, and metal product, it declined from 55.9 percent in 1968 to 35.4 percent in 1976. Second, the “strategic” sectors were subsidized excessively through preferential financing and tax incentives. Investment projects of the HC industries were financed by long-term loans from commercial banks and public finance institutions like the KDB. Furthermore, the National Investment Fund (NIF) was established to provide loans at preferential interest rates. Domestic tax incentives in the form of direct tax exemption, tax holidays, special depreciation, and temporary investment tax credit have been utilized effectively to direct investment resources into several key industrial sectors. These incentives were rearranged by the Tax Exemption and Reduction Control law in 1975 under the heading of “Special Tax Treatment for Key Industries.’’ Third, the government announced its General Guidelines for Foreign Direct Investment to reinforce entry restrictions. Joint ventures became more acceptable than wholly-owned foreign firms, except that the joint-venture firms could not compete with domestic firms in overseas markets or take on technology-intensive projects. The restrictions on foreign direct investment were based on a fear that the various encouragements in place until the early 1970s might result in foreign firms’ dominance of domestic industries, and on the difficulty in implementing efficient development strategies. Thus, the government became more stringent on export requirements and foreign ownership. These strong attitudes toward foreign direct investment made Korea rely less on foreign direct investment in the pursuit of strategic development and in the related financing of imported capital goods. Foreign capital inflows and economies of scale each played an important role in the rapid growth of HC industries. Motivated by the promise of high returns, foreign capital inflows supported high rates of investment in the capital-intensive industries. At the same time, production in these sectors was concentrated among a few firms, leading to monopolistic or oligopolistic

268

Susan M. Collins and Won-Am Park

market structures. On the one hand, the concentration was justified by economies of scale; on the other hand, it was used to justify increased government intervention. Heavy and chemical industries as a share of total manufacturing value added rose from 42 percent in 1970 to 51 percent in 1980. There was also a jump in the export ratio of light industries during 1970-75. This suggests that exports in the labor-intensive industries were promoted despite the shift of emphasis to import substitution. However, the rapid change in industrial structure produced distortions in resource allocation. The push for the capital-intensive projects of the HC industries and the provision of a large amount of low-cost funds through various sources resulted in a shortage of skilled labor and rapid money growth, which in turn raised wages and inflation. Large firms rapidly expanded their power in the domestic market. Moreover, a number of public enterprises began to manage several investment projects on social overhead capital, further complicating the market structure. Overall, the shifting of the industrial structure and the development of strategic sectors through a strong government’s protection and assistance did not come without cost. In particular, because the changes in policy direction of the 1970s originated from unfavorable external conditions in the early part of the decade, and because the 1970s were jolted by two oil shocks, it is very difficult to measure the true cost of the policy changes toward more intervention in resource allocation and the backward linkage of the capital-intensive industries with the labor-intensive industries. The question is whether the government could have avoided some of the undesirable outcomes of the 1970s and what the optimal extent of government intervention in the 1970s would have been. The general consensus at the time of the second oil shock seemed to be that the costs borne by the domestic economy during the heavy industrialization process were excessive.

9.4.3 Restructuring Industrial Growth, 1980 to the Present Toward the end of the 1970s, the government’s promotion of HC industries gave rise to internal and external imbalances in the economy and less efficiency in resource allocation. The overcapacity problem in those sectors appeared as massive investments in strategic sectors were countered by the declining worldwide demand for, and international competitiveness of, exports. To make things worse, the oil price hike, social and political turmoil, and a rice crop failure during 1979-80 had a stagflationary impact on the economy. In 1980 the GNP shrank 4.8 percent, wholesale prices rose 38.9 percent, the current account deficit widened to $5.3 billion, and foreign debt increased by $6.9 billion. The poor performance of the economy forced the government to reevaluate the industrial policies implemented during the 1970s. To remedy the overinvestment in HC industries and the distortions in resource allocation

269

KoredChapter 9

caused by strong legislative protection and support, the government intended to rely more on market mechanisms by liberalizing the economy both internally and externally and by reinforcing antimonopoly and fair trade. We have already mentioned that the government carried out substantial liberalization of imports, raising the import liberalization ratio from 68.6 percent in 1979 to 91.6 percent in 1986. Also effective were the tariff reforms to reduce the average nominal tariff rate for all commodities from 35.7 percent in 1978 to 18.1 percent by 1988. In the financial sector the number of government-controlled policy loans has been reduced since 1980. Furthermore, preferential interest rates applicable to policy loans were abolished in June 1982. Financial liberalization efforts, which have been stepped up since 1982, include turning over the government’s equity share of commercial banks to the private sector, minimizing government control over banking operations, reducing entry barriers to the financial sector, and adopting policies to encourage the development of a universal banking system. We return to financial policies in chapter 1 1. Domestic tax incentives given to strategic sectors were sharply reduced in the early 1980s. The number of strategic industries eligible for tax incentives declined from fourteen to six. The content of tax incentives also changed to indirect tax preferences through accelerated depreciation instead of direct exemptions of corporate taxes. Foreign direct investment policies were reexamined. To promote competition in the domestic market, many new industries were opened to foreign investors in the early 1980s. This liberalizing trend culminated in the establishment of the new Foreign Capital Inducement Act in December 1983 which introduced a negative list system. Foreign investments were allowed unless they were directed toward the listed industries. Furthermore, projects that satisfied certain requirements were automatically approved without procedural difficulties imposed by bureaucratic custom. These new policies, the correction of the distortions created by the strongly protectionist policies of the 1970s, and the favorable economic conditions in the 1980s all contributed to a strong recovery of economic growth. Both light and heavy industries expanded. In fact, heavy industrial output surpassed light industrial output in the 1980s. Exports of heavy industrial products accelerated causing a rapid increase in the ratio of heavy to light manufactured exports (see table 9.2). From this perspective, it is misleading to assert that the heavy industrialization in the 1970s played an insignificant or detrimental role in Korea’s industrial growth. We do not believe there is a simple answer to the question of whether the Big Push was a mistake. While it is possible to list both favorable and unfavorable consequences of the strategy to date, and while many observers feel that the unfavorable ones dominate, the jury is still out on the ultimate costs versus benefits of the resource shift toward heavy industry.

270

10

Susan M. Collins and Won-Am Park

Exchange Rates, Wages, and Productivity

In this chapter we continue our discussion of exchange rate policy in Korea. We focus on the linkage between exchange rates and competitiveness, and relate them to labor productivity and to the behavior of nominal wages. The issues are especially interesting because of one unusual aspect of Korea's adjustment: real wages rose by 43 percent during 1982-86 despite a 34 percent depreciation of the real exchange rate.' The puzzle emerges, not because negatively correlated real wages and exchange rates are theoretically implausible, but because there seem to be so few examples in practice. Many countries would like to devalue so as to improve competitiveness and external balance, but avoid doing so precisely because of a desire to maintain real incomes and consumption. The links between real exchange rates and real wages are important precisely because they embody the tradeoffs between competitiveness and the standard of living. It is widely recognized that a nominal depreciation which does not result in a real depreciation because of induced rises in domestic goods and factor prices is likely to have little effect on the trade balance. At the same time, domestic real incomes will decline if wages do not rise enough to offset the loss in purchasing power from higher traded goods prices. A reduced standard of living is often viewed as the price paid for an increase in competitiveness. In addition to these issues, capital flight problems and fiscal and monetary policy are also integral to exchange rate policy decisions (Diaz Alejandro 1981 and Dornbusch 1985b). Latin American countries, in particular, have suffered from large budget deficits financed by money creation, massive capital flight, high inflation rates, and overvalued currencies. Accelerating wages and prices which exacerbate the overvaluation are especially likely when a government adopts an accommodating macroeconomic policyexpectations of an accommodating policy will tend to result in slower adjustment of wages in response to unemployment (Dornbusch 1982). Korea has a very different background: sound fiscal policy, strict capital controls which rule out capital flight, and wages which are not indexed to past inflation. Furthermore, rapid productivity growth has mitigated the conflicts between competitiveness and real income. Active government policies in allocation of resources seem to have enhanced productivity growth. This chapter explores lessons from the Korean episode. The first section discusses the relationship between the various relative price measures. In particular, we focus on the real wage, an important internal relative price, and various external relative prices, including a number of measures of the

KoredChapter 10

271

real exchange rate. Section 10.2 documents the behavior of real wages and exchange rates. In Section 10.3 we examine the determinants of wages. The chapter concludes with a discussion of the lessons to be learned and the implications for policymakers in other countries. 10.1 Theoretical Background This section specifies the relationship between real wages and real exchange rates. We begin by defining key variables. e

=

wonlforeign currency (nominal effective exchange rate);

w is the nominal wage;

P i , i = X , M , N are indices for export, import and nontraded goods prices; p = Pg P& P; is the Korean CPI, where a + /3 + y = 1; p* is a foreign price index; PT = P;;/(a+P)Pg(a++P) is an index of traded goods prices; Oi, i = T,N are indices of labor productivity (output/worker) in the tradable and nontradable goods sectors.

There are four variables of interest. Equations (1) and (2) give two measures of the real exchange rate. The first is the typical measure relating domestic and foreign prices. The second is the ratio of traded to nontraded goods prices in Korea. An increase in either represents a real depreciation. (1)

R = ep*/p

(2)

/A.

= PTIPN

Equation (3) denotes the real wage while equation (4) denotes unit labor costs of tradable goods measured in foreign currency, another measure of competitiveness. (3) (4)

w =w / p

{=wl(e.O,)

To highlight the role of labor productivity in determining the behavior of these four variables, we assume a very simple price setting structure. Korea is assumed to take the price of imported goods as given. Export and nontraded goods prices are assumed to be determined by costs. Both types of goods are produced using labor and imported intermediates. (5)

(7)

PM=e.P&

272

Susan M. Collins and Won-Am Park

Substituting (5)-(7) into equations (1)-(3) allows us to rewrite the real exchange rates and the real wage in terms of labor productivity, nominal wages, the nominal exchange rate, and the world price of imports.

*

+y b

ey'eGb

(i) aa

R=p*

(p&)l-aa-yb

A comparison of equations (8) and (9) shows how the two real exchange rate measures might move in opposite directions. If labor productivity grows more quickly in the tradables sector than in the nontradables sector, R will depreciate but the domestic relative price of nontradables will rise, implying that p will appreciate. Note that b - a 8 is likely to exceed one. It must do so if tradable goods production uses more imported intermediates than nontradables. Equation (10) shows that the real wage increases when labor productivity rises in either sector, when nominal wages rise relative to the domestic price of importables, or when the nominal exchange rate appreciates. However, real depreciations result from nominal wage declines or nominal depreciations (eq. [8] and [9]), creating a tradeoff between competitiveness and the standard of living, Equations (8) and (10) also show that productivity growth can eliminate the sharp conflict between these two objectives by creating a cushion. Real wages may rise while the real exchange rate depreciates (and unit labor costs in foreign currency fall), and as long as productivity is growing strongly enough. The condition for this scenario is that nominal wage growth exceed the domestic inflation rate but not the sum of nominal depreciation and productivity growth. 10.2

The Korean Experience

In table 10.1 we present data on the behavior of wages, prices, productivity, and unit labor costs in manufacturing since 1960. The table presents two measures of productivity. One (col. A) comes from surveys conducted by the KPC, while the other (col. B) gives value added per employee (VA). Neither is an ideal measure, and it is difficult to classify one as consistently better. Unfortunately, the two tell different stories. While real wages grew on average by 8.5 percent per year during 1964-86, manufacturing productivity grew by 6.8 percent according to VA but by 12.4

Table 10.1

Wages, Productivity, and Unit Labor Cost in Manufacturing (in percentages)

Nominal Wages (A)

Real Wages

CPI

Year 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

Index

1.6 1.8 1.9 2.2 2.6 3.1 3.7 4.5 5.7 7.7 9.7 11.3 12.9 15.2 20.6 26.2 35.2 47.2 63.3 81.5 100.0

120.1 137.8 154.6 167.2 183.8 200.6

Change

(1980-100) Index

12.0 6.5 14.4 22.0 18.6 17.8 22.0 27.I 34.2 26.9 16.2 13.9 18.0 35.3 27.0 34.7 33.8 34.3 28.6 22.7 20.I 14.7 12.2 8.1 9.9 9.1

6.4b 6.9b 7.4b 8.9b 11.9

12.5 13.9 15.3 17.0 19.1 22.2 25.2 28.1 29.0 36.1

45.2 52.1 57.4 65.7 77.7 100.0 123.3 132.3 134.5 137.6 141.0 144.2

24.9 25.7 25.7 24.5 23.0 25.1 26.6 29.4 33.7 40.2 43.9 45.0 45.9 52.5 57.1 57.9 67.6 82.2 96.5 104.9 100.0 97.4 104.1 115.0 121.5 130.4 139.1

Change

Labor Productivity

KPC Index (BY

Change

11.1

3.2 -0.0

-4.9 -6.2 9.5 5.9 10.4 14.6 19.3 9.3 2.4 2.0 14.3 8.8 1.4

16.8 21.5 17.4 8.7 -4.7 -2.6 6.9 10.4 5.7 7.3 6.7

12.5 12.8 13.6 14.8 17.4 18.1

21.3 25.6 32.3 36.4 39.9 43.4 47.2 62.6 58.7 63.I 69.7 78.0 90.4 100.0

118.1 127.3 144.6 159.8 171.1 194.4

Source: BOK, Economic Statistics Yearbook, various issues. "Index made by Korea Productivity Center using output per production worker. bConsumer price index in Seoul.

Change

Value added per employee (C)

Unit Labor Cost (Won)

12.6 2.4 6.3 8.8 17.6 4.0 17.7 20.2 26.2 12.7 9.6 8.8 8.8 11.4 11.6

7.5 10.5 11.9 15.9 10.6 18.1 7.8 13.6

10.5 7.i 13.6

A/B

14.3 14.2 14.8 15.9 17.9 18.0 20.4 21.2 22.4 23.8 26.8 28.4 29.7 32.3 39.2 44.6 55.8 67.7 81.2

Change

104.0

16.0

90.1

-0.5 4.0 7.7 12.1 0.8 13.3 3.6 5.7 6.3 12.6 6.0 4.7 8.5 21.4 13.8 25.3 21.2 20.0 11.0

100.0 111.1

-3.9 11.1 - 1.8 4.2 12.0 -0.8 7.6

100.0

11.0

101.7 108.2 106.9 104.6 107.4 103.2

1.7 6.4 -1.2 -2.1 2.7 -3.9

30.3 31.9 31.7 34.5 34.2 37.8 43.9 53.6 61.1 64.1 67.3 68.9 70.5 72.2 79.6 89.6

109.1

113.7 127.4 126.4 135.9

5.3 -0.6 8.7 -0.8 10.4 16.I 22.3 13.9 5.0 5.0 2.4 2.2 2.4 10.3 12.6

A/C

7.2 8.3 9.9 10.7 13.2 15.2 17.5 18.5 18.5 20.1 22.6 29.9 37.1 48.8 59.2 70.7 78.3 100.0 108.I 126.3 136.0 131.2 145.5 147.7

Unit Labor Cost ( U S . dollars)

Change

15.9 19.2 8.4 23.4 14.6 15.6 5.6 0.2 8.5 12.4 32.1 24.3 31.5 21.3 19.3 10.9 27.7 8.1 16.9 7.7 -3.5 10.8

1.5

A/B

Change

69.2 74.3 7.4 50.8 -31.6 41.1 -19.2 45.7 11.2 47.6 4.2 49.2 3.3 1.9 50.1 52.3 4.4 49.2 -6.0 45.8 -6.9 42.3 7.6 58.7 19.1 56.0 -4.6 70.1 25.1 85.0 21.3 101.9 20.0 11.0 113.1 100.0 - 11.6 90.7 -9.3 -0.9 89.9 83.7 -6.9 78.8 -5.8 -4.9 75.0 71.1 -5.1

A/C

Change

33.4 23.6 -29.5 22.6 -4.3 24.0 6.5 29.7 23.8 33.3 12.0 10.9 36.9 -2.0 36.2 10.4 32.4 -4.1 31.1 10.9 34.5 30.I 44.9 46.6 3.9 61.3 31.5 74.3 21.3 88.7 19.3 10.9 98.3 I .7 100.0 96.4 -3.6 8.9 104.9 1.5 106.5 98.9 -7.1 2.7 101.6 0.2 101.8 -

274

Susan M. Collins and Won-Am Park

percent according to the KPC. While the KPC measure was widely used during the 1970s, it suggests implausibly rapid productivity growth during the crisis of the early 1980s. For this reason, and for reasons of comparability, the VA measure has become more widely used recently. Our discussion will refer to both series. A key point that emerges from table 10.1, together with the exchange rate data in table 9.7, is that Korea has experienced a number of years in which real wages grew while the real exchange rate depreciated. The combination occurred during both 1971-73 and 1982-85. However, it is important to point out that real wages have not increased continuously during the Korean industrialization. They declined both at the outset of Korea’s export-led growth and as Korea reestablished its competitive position after the 1975-79 real appreciation. Real wages fell by 10.5 percent during 1962-64 despite a 15.1 percent increase in labor productivity (KPC), and by 7.1 percent during 1980-81 despite 30.6 percent (KPC) or 6.7 percent (VA) growth in productivity. Both measures of productivity identify the 1973-79 Big Push as a period in which rapid real wage gains outstripped productivity growth. As discussed further below, the rapid nominal wage increases during this period have been attributed to competition for scarce skilled labor, in conjunction with the push toward heavy industry. At the same time, the expansion of overseas construction contracts exacerbated the shortages of some types of domestic labor, with the resultant wage increases spreading to workers elsewhere. We examined the real wage behavior in more detail over four time periods from 1964 to 1985 in table 10.2. From equation (10) the key factors are the nominal wage relative to the domestic price of imports and labor productivity. In the discussion below, we focus on the VA measure of productivity. The table shows that the 1969-73 slowdown in real wages is in part attributable to a slowdown in overall productivity, but that the more important factor is a decline in nominal wages relative to imports. This represents both a moderation of nominal wage gains and a deterioration in the terms of trade. Real wage growth accelerated during 1973-79. During this period, very rapid nominal wage gains offset continued terms of trade deterioration. The slowdown in 1979-85 again arises from reduced productivity growth combined with a substantial deceleration of nominal wage gains. Real wages fell at the beginning of the recent adjustment (1980-81), with all productivity gains going to increase competitiveness. This, plus exchange rate depreciation, improved Korea’s competitive position. The table very clearly shows that real wages have grown more quickly during real appreciations. However, there has been no clear relation between real wage growth and the terms of trade. Not surprisingly, rising domestic production costs during periods of rapid real wage growth have tended to increase the price of nontradables relative to imported goods. In addition,

Table 10.2

Determinants of the Real WageExchange Rate Linkage (average annual percentage change)

Period

Real Wage

Real Effective Exchange Rate

Relative Price of Imported to Nontraded"

10.6 6.9 12.2 3.7

-3.7 8.1 -4.8 3.8

- 10.0

1964-69 1969-73 1913-19 1979-85

2.3 - 5.9 1.4

Labor Productivity (Value Added)

Labor

Terms of Trade

Productivity (KPC)

Manufacturing

Nonmanufacturing

3.5 -1.4 - 1.4 -1.4

16.9 9.9 11.5 11.2

6.6 11.3 7.5 3.3

5.9 2.7 4.9 4.5

'(Dollar unit price of imports x nominal exchange rate)/nonmanufacturing deflator.

bNominal wages in manufacturin@(doIlarunit price of imports 'Based on calculation in K. S. Kim (1986).

X

nominal exchange rate)

Total

Wages in Terms of Importsb

Real Capital Cost

5.9 4.2 6.1 4.2

21.3 0.7 13.4 2.5

3.0 -3.3 - 0.4 -0.1

Trade Liberalization Ratio' 93.0 - 1.4

5.3 4.2

276

Susan M. Collins and Won-Am Park

there is some evidence of a positive correlation between real wage gains and labor productivity growth. From the table, we can also support the view that exchange rate policy has been used to offset slowdowns in productivity growth and to maintain Korea’s competitiveness in international markets. Authorities depreciated the real exchange rate during 1969-73 and 1979-85. Overall productivity (VA) had declined in both periods. In contrast, the fixed exchange rate and real depreciation of the 1970s coincided with rapid overall productivity gains, although productivity growth slowed in the manufacturing sector. We have already seen that the real exchange rate (R) appreciated in some periods but depreciated in others, despite the fact that productivity grew rapidly by international standards throughout. However, using the price of traded relative to nontraded goods (p) as a measure of the real exchange rate, Korea has experienced a continuous real appreciation, as shown in table 10.3. To compute these figures, we use manufacturing and nonmanufacturing as proxies for the traded and nontraded goods sectors respectively. One reason for the faster inflation in the nontraded goods sector has been relatively slower productivity growth in that sector. Differential inflation rates emerged in the mid- 1960s as productivity growth accelerated in the manufacturing sector. According to the VA measure, productivity growth in nontradables began to outpace productivity growth in manufacturing in :he 1980s. As shown, the inflation differential narrowed considerably during this period. It is interesting to compare unit labor costs in Korea with the costs of its main trading partners and with costs in other newly industrialized nations, which compete with Korea in third markets. Korean unit labor costs measured in U.S. dollars declined by 30.3 percent from 1979 to 1984. In contrast, the U.S. Department of Labor reports that dollar unit labor costs for U.S. industries rose by over 22 percent during the same period. Japanese unit labor costs declined by 3.7 percent measured in yen and 11.6 percent measured in dollars. The figures in table 10.4 compare the Korean and Taiwanese growth rates of unit labor costs measured in U.S. dollars. During the late 1970s, the rapid increases in Korean wages implied a substantial loss in competitiveness vis-a-vis Taiwan. During 1979-82, however, Korean labor costs grew by just 2 percent per year, compared with nearly 10 percent annual growth in Taiwan. The divergence persisted during 1982-86 as Korea’s major depreciation led to a decline in labor costs. Although the countries have had similar gains in productivity, exchange rate policy in Korea has significantly improved its position relative to that of Taiwan.

10.3 Wage Determination in Korea The above discussion highlights the magnitude of nominal wage adjustments as a factor in Korea’s ability to combine depreciation with real

Relative Price of Manufacturing Goods and Productivity (average annual percent change)

Table 10.3

Deflator

Period 1960-64 1964-69 1969-73 1973-79 1979-85

Labor Productivity"

Manufacturing (A)

Nonmanufacturing (B)

Relative Price WB)

22.0 8.0 9.3 17.2 8.0

22.2 13.3 15.3 23.8 10.2

-0.2 -4.6 -5.2 -5.4 -2.1

Manufacturing (A)

Nonmanufacturing (B)

Relative Productivity WB)

- 1.4

8.2 5.9 2.7 4.9 4.5

-8.9 0.6 8.4 2.5 -1.1

6.6 11.3 7.5 3.3

"Labor productivity is defined here as the value-added per worker. b(Dollar unit price of imports

X

nominal exchange rate)inonmanufacturing deflator.

Real Effective Exchange Rate

Relative Price of Imported to Nontradedb

-3.7 8. I -4.8 3.8

- 10.0

2.3 -5.9 1.4

278

Susan M. Collins and Won-Am Park

Table 10.4

Unit Labor Costs in Manufacturing (in U.S. dollars, average annual percentage change) Period

Korea

Taiwan

1976- 79 1979-82 1982-86

17.06 2.20 -0.76

13.98 9.74 4.95

Source: BOK, Economics Statistics Yearbook, various issues, and Statistical Yearbook of the Republic of China. various issues. Note; Unit labor costs are defined as the nominal wages relative to value-added productivities.

wage increases. Throughout most of its recent history, nominal wages have grown more quickly than prices, however real wage increases have frequently been bounded by productivity growth (see table 10.1). This section provides an overview of key aspects of Korean labor markets to shed some light on the determinants of nominal wage growth. We focus on characteristics evident during Korea’s industrialization and adjustment to the 1979-80 crisis. The demonstrations, strikes, and other labor activities since 1986 may signify some important changes in wage determination and in the relationship between workers, management, and the government. However, it is too early to assess these developments. The discussion, which draws heavily on work by Kim Sookon (1982) and Lindauer (1984), is based on data for wages and Compensation of private, nonagricultural workers in the formal sector and of public sector employees. Unfortunately, earnings data for the informal urban sector, consisting of small-scale and family businesses, are not available.2 The discussion begins with an outline of compensation, labor mobility, and the role of institutional factors in Korean labor markets. It then focuses on wage determination, considering the relevance of a competitive labor market model for Korea, the link between wages and prices, and the importance of government intervention. Employee compensation in Korea is quite ~ o m p l e x .The ~ total is composed of a basic wage, allowances, and a bonus. The basic wage includes a starting wage plus annual increments arising, for example, from seniority, merit, and cost-of-living increases. It is typically the largest part of total compensation, ranging from 50 to 60 percent for production workers, and sometimes reaching 80 percent of compensation for managers, professionals, and technician^.^ The importance of allowances varies widely by industry and occupation. Some allowances, such as overtime and annual leave, are stipulated by the Labor Standards law. Many others, including allowances for special skills, family, housing, and transportation, are not. Their coverage differs widely across firms and across workers within firms.5 Bonuses are not required by law, but remain extremely widespread. In one study, every firm had paid out bonuses. Civil servants and public enterprise

279

KoredChapter 10

employees also received bonuses. On average, bonuses amounted to 400 percent of base monthly earnings or about 15 percent of total compensation, however, again there was a wide variance, with large firms tending to pay out more. It is difficult to assess the extent to which bonuses are considered part of anticipated compensation. In general, they have fluctuated with market conditions, although some large firms have maintained the level of bonuses during downturns. Overall, bonuses seem to have been increasing as a share of compensation. One implication of the special structure of Compensation in Korea is that neither basic wage figures nor total compensation is an ideal measure of “required” unit labor costs, i.e., an indicator of competitiveness. The wage numbers underestimate costs since they exclude some required payments. However, total compensation may also be biased because of its endogeneity. An increase in bonuses during a profitable year would increase measured unit labor costs, erroneously indicating that Korea was becoming less competitive vis-5-vis other countries. A better measure would combine basic wage with those allowances which the firm was obligated to pay and with the minimum bonus from the implicit contract between employer and employees. Of course, such a measure is unavailable. A second issue frequently discussed is the extent to which labor markets in Korea are characterized by Japanese-style lifetime employment. On the one hand, surveys show that 56 percent of Korean workers would expect to continue full-time work at normal pay during a major downturn.6 On the other hand, there are no explicit guarantees, and there is substantial job turnover. Average monthly separation rates in manufacturing are above 5 percent in Korea, as compared to 4 percent in the U.S. and 2 percent in Japan. (S. Kim 1982, 27). Lindauer concludes that “lifetime or permanent employment systems such as those that exist in Japan are not a feature of any significant sector of the Korean economy” (1984, 61). A third issue is whether institutional factors, such as unions and/or government interventions, played a significant role from the 1960s through 1985. The union movement in Korea remained weak and subject to strict government regulations. Until 1981 these regulations included a ban on strikes and a requirement of prior government approval for any collective bargaining activities. The Worker Council law in 1980 called for all firms with thirty or more employees to hold council meetings, with management and labor equally represented, to discuss productivity and other issues.’ However, the right to negotiate wages was not stipulated. Only a few industries, notably textiles and some public enterprises (e.g., rail, telephone and telegraph, electric) had unions. At most, 20 percent of the industrial work force belonged to a union. A consensus view is that unions have had a negligible impact on wages or total compensation, but that they have helped to increase job security. Those labor disputes which did occur focused not on wages and work conditions, but on issues of worker rights in the work place (S. Kim 1982, 62). Using

280

Susan M. Collins and Won-Am Park

wage regressions we find the coefficient on unionization to be insignificantly different from zero, while separation rates tend to be substantially lower in unionized establishments. Although the government did not establish explicit wage guidelines, there were a variety of less formal ways in which it could exert pressure on wage determination. There is also some evidence of intervention in private sector wage determination. In 1977, concern over real wage increases in excess of productivity growth led the government to announce that for monopolistic firms with controlled prices there would be a ceiling on allowed price increases due to rising labor costs.’ At the same time, the government began to follow a policy which based wage increases on productivity increases. A reduction in the growth of public sector wages was announced in 1980. While the Ministry of Labor continues to take a stand against direct government intervention in wage negotiations, the Ministry of Finance has seemed to favor some intervention since 1981. In November 1981 the BOK directed all banks to enforce a Korea Bankers Association (KBA) resolution to stop new loans to firms which, despite financial difficulties, increased wages beyond labor productivity.’ This resolution was reiterated in 1982 as part of a nationwide mass media campaign to bring down inflation. If implicit or explicit government policies significantly influenced wage determination, one would expect public sector wages to act as a signal for appropriate wage growth in the private sector. lo Empirical evidence provides little support for the view that the government acted as a wage leader prior to 1980. There is no obvious correlation between public and private earnings. Public sector employees earned less than those working in the private sector, with the differential increasing with skill and educational levels. In response, public sector earnings rose much more rapidly than did private sector earnings from 1972 to 1976. During the push toward heavy industry in 1976-79, earnings grew more quickly in the private than in the public sector, outstripping productivity gains. Since 1979 both public and private earnings growth rates have declined substantially. The moderation of public sector wages began in 1981 as part of the effort to reduce the fiscal deficit. Since then, the government has taken a more active stand on incomes policy, as discussed above. It is difficult to determine the importance of these factors, relative to the importance of changing economic conditions (notably the drop in inflation and the relative scarcity of skilled labor), in the subsequent slowdown of private wage growth. Given all of the factors discussed above, what is an appropriate model for wage determination in Korea?” Most authors conclude that wages have been determined primarily by market forces since the early to mid- 1970s. Lindauer bases his conclusion on the following findings for the formal sector: that real wage trends have been similar across industries, that the structure of interindustry earnings has been stable with a recent narrowing of the dispersion, and that educational wage differentials track relative

281

KoredChapter 10

scarcities in skilled 1ab0r.l~He concludes that the major inefficiency in Korean labor markets arises from the large and persistent wage and employment differences by sex. The rapid real wage increases during the 1970s have no obvious institutional explanation, but can be readily understood from changes in the structures of labor supply and demand. Amsden (1986), who finds a much greater role for government intervention, also concludes that market forces were the key factors in wage determination. The evidence for a competitive model of wage determination for the 1980s is much less clear. As pointed out above, it is difficult to distinguish market pressures from direct and indirect government pressures. It is also difficult to assess the extent to which the government became more interventionist. On the one hand, collective bargaining regulations were relaxed. On the other hand, government attention to incomes policies clearly increased. This concern, together with the increased leverage of the banking system over private firms, expands the scope for intervention. We conclude this section by highlighting some features of the Korean labor market. First, there is relatively little inertia in the wage-setting process in Korea. Instead of a backward-looking or indexation scheme, wages seem to react quickly to changing market conditions. Second, the increased reliance of the private sector on the organized domestic financial sector during the early 1980s expanded the government’s ability to exert an influence on private sector wage determination. Finally, organized resistance to any pressure (actual or potential) on wages from the government was negligible through the mid-1980s. There has been a marked increase in worker activism since 1986, however, it is too early to assess the longer term implications.

10.4 Discussion This chapter has highlighted two factors in explaining Korea’s ability to combine a real depreciation with real wage growth. The key has been rapid increases in labor productivity which drive a wedge between the minimum wage increase for real wage gains and the maximum increase for competitiveness gains. As argued in chapter 7, the key to Korea’s growth has been its very rapid augmentation of both capital and labor. The second factor has been the determination of wages. Weak unions and worker organizations have had a negligible effect on wage adjustments. Instead, Korean wages seemed to adjust relatively quickly to changing market conditions throughout the 1970s. The lack of wage indexation has removed some of the inertia in wage adjustment frequently seen in Latin American countries. Additional flexibility is introduced by the system of compensation in which a substantial share of worker compensation is in the form of bonus payments, which can be reduced during downturns. In some respects, Korea workers with their growing real wages have fared well under

282

Susan M. Collins and Won-Am Park

this system. The costs, which come in terms of limited influence over worker rights and work conditions, are very difficult to quantify. A final point worth stressing is that’tradeoffs between real incomes and competitiveness are only avoided once the investment-productivity gain cycle gets going. Korea cut real wages to give an initial boost and to get the “engine” moving both in the early 1960s and during adjustments in 1980-81.

11

Fiscal and Monetary Policy

In this chapter we assess the role of fiscal and monetary policy in Korea’s experience with external debt. One important issue is the financing of fiscal deficits. Did the government borrow heavily from abroad or rapidly expand the domestic money supply in order to finance large budget deficits? Both factors figured prominently in the experience of many Latin American debtor countries, however, both turn out to play much smaller parts for Korea. Still they are of interest precisely because they highlight some of the aspects which distinguish Korea’s debt history from the history of many other countries which have had less successful recoveries. A second issue is the role of fiscal and monetary policies in achieving the phenomenal growth rates which have enabled Korea to service very large external debts. To summarize our conclusions at the outset, we argue that fiscal policies have been used countercyclically, but that they were not the predominant explanation for rapid growth. Monetary policies, on the other hand, have played a central role, although not through excessive inflation finance because the allocation of domestic credit has been a centerpiece in the government’s industrial policies which have successfully targeted high-growth export industries. 11.1 Brief History An overview of the development of Korea’s financial and fiscal sectors provides a useful base for examining the current systems.’ The key issues of the linkages between government finances, monetary policy, and external borrowing are not new, but emerged at the outset as Korea recovered first from World War I1 and then from the Korean War. The developments through the early 1970s can be divided into three stages. In the early stage, prior to 1945, Korea enjoyed a very highly developed financial system run by the Japanese to mobilize resources for the colonial expansion and later to help finance military spending. The system

283

KoredChapter 11

was modeled on Japan’s, with the very close relationship between business and government which characterizes their approach and which has had significant influence on the current Korean system. In fact, statistics suggest that the fiscal and banking systems were more developed in 1940 than they were in 1975. Government revenues were 21 percent of GNP in 1940 as compared to 16 percent in 1975, while the ratios of M2 to GNP were 44 percent in 1940 and only 34 percent in 1975. The second stage, between the collapse of the Japanese system in 1945 and the beginnings of an independent Korean system in the mid-l960s,was dominated by the role of foreign aid inflows and the interactions between the U.S. and Korean governments. The systems stood in market contrast to those that had collapsed, with “no money and capital markets in the accepted sense of the terms and no really adequate facilities for mobilizing such savings as are currently made and for channeling them into productive investments” (Mason et al. 1980, 301). Two critical problems were that the experienced money and fiscal managers had been Japanese and that hyperinflation had removed confidence in the organized banking system. The gap was partially filled by expansion of the unofficial money market. Unlike official institutions, the UMM could operate using U.S. dollars and U.S. military payment certificates. Such curb markets have continued to play an important financial role in Korea. In this second phase, foreign aid flows and counterpart funds were the major sources of funding for the government, giving rise to a seesaw for control between the Korean and U.S. governments. On the one hand, the U.S. wanted a more western system with an independent central bank and a revised tax system to provide resources for government spending. They wished to have aid flows be conditional on the fulfillment of specified criteria-the government deficit and the growth of bank credit. The Koreans, on the other hand, pushed for continued aid, allowing the government continued control over the allocation of credit to finance reconstruction, and maintaining close ties with the business sector. The decade from 1954 to 1964, which resembled a tug-of-war between the two approaches, can be divided into four periods. From 1954 to 1956 increasing government expenditures and bank loans were financed by significant aid inflows, while domestic (bank) savings declined relative to output. From 1957 to 1960 the situation was reversed, with aid, government spending, bank lending, and real growth all falling, and government revenues and bank deposits both rising relative to GNP. The initial years of the Park regime, 1961-62, were again expansionary, followed by another U.S.-imposed contraction in 1964-65. The 1961-62 period is especially interesting. The new government instituted a number of financial reforms, all of which significantly increased bank credit. It reorganized agricultural financing institutions, created the Small and Medium Industry Bank, and authorized the KDB to guarantee foreign loans and to borrow abroad. It also regained ownership of the

284

Susan M. Collins and Won-Am Park

commercial banks and brought the BOK under the control of the Ministry of Finance. In addition to the accelerated money growth (nearly 50 percent from June 1961 to June 1962), a rise in government spending from 18 to 24 percent of GNP pushed the budget deficit from 2 to 4 percent. An unfortunate consequence of the expansion was a revival of inflation. In response, the government attempted a currency reform in June 1962. A new currency was issued with limited conversion. However, the resulting lack of funds crippled all business activity so severely that the government began to relax the measures within one week, and within five weeks the measures had been totally eliminated. The ineffective reform was succeeded by a U.S.-Korean stabilization plan which cut public expenditure from 24.1 percent in 1962 to 11.5 percent of GNP by 1964. M2 fell from nearly 15 percent of GNP in 1962 to barely 9 percent in 1964, as lack of confidence in the official banking system contributed to a new growth spurt in the unofficial money market. By 1965 Korea had entered a third stage of financial/fiscal development in which external governments played a diminishing role in the decision making. The Park administration had formed a system with the financial sector firmly controlled by the government and with the government, through the allocation of (domestic and foreign) credit, firmly linked to business decision making. The major remaining issue was how to replace the declining aid inflows, and the government turned to the problem of mobilizing domestic savings and nonaid external funds to finance government spending and investment. The system of foreign loan guarantees, combined with special incentives to exporters, had already begun to generate foreign (nonaid) inflows. The 1965 financial reform, which raised interest rates on bank deposits, was undertaken in the hope of stimulating private savings and channeling it to official financial institutions. As we have seen, both elements proved extremely successful. By the beginning of the 1970s, private savings had risen from 6 to 7 percent to 18 percent of GNP and foreign debt had jumped from 7 percent to over 30 percent of GNP. The key pieces in the story continued to be investment, private savings, government savings, and foreign savings. Investment, private savings, and foreign savings have been discussed in previous chapters. Government savings (revenues and expenditures) and the role of fiscal policy in Korean macroeconomic performance are discussed in section 11.2 of this chapter. In section 1 1.3 we examine the financial system and the allocation of credit.

11.2 Fiscal Policy 11.2.1 Structure Korea’s public sector is quite complex. It is composed of a central government and five special public enterprise funds, which together make up

285

KoreafChapter 11

the consolidated public sector, In addition, there are local governments and a number of nonfinancial public enterprises. Because of data delays and revisions, and because of the difficulties of adequately accounting for intergovernmental transfers, the consolidated public sector excludes the local governments. We focus on the activities of the consolidated public sector. The central government consists of the general account, which accounts for most of total revenues and approximately 80 percent of expenditures, fifteen special accounts, and twenty-four special funds. The public enterprise funds include the Grain Management Fund, which purchases and sells grains and which became very important during the disastrous agricultural output during 1978-82. Although their expenditures have been very large in some years, these funds contribute relatively little to the consolidated budget because only the net surplus or deficit of the public enterprise funds enters the accounts. Table 11.1 shows the revenues, expenditures, and budget deficits for the central government and the consolidated public sector. The top panel of the table gives the figures in billions of won, while the bottom panel takes ratios of each variable to GNP. The net financial transactions column gives the deficit or surplus in the public enterprise funds, which is included with the central government deficit in the consolidated budget deficit. The last four columns of the table give the sources of deficit financing. The domestic bank financing is taken from the monetary, and not the fiscal, accounts and is identically equal to the change in banking sector credit to the government.’ 11.2.2

General Trends

Figure 11.1 plots public sector revenues and expenditures relative to GNP from 1970 to 1985. It shows large swings in both series between 1970 and 1975. Since then, revenues have been considerably more stable than expenditures, rising gradually until 1981 and then tapering off slightly. Thus, recent changes in the budget were due primarily to changes in spending. As we shall see, there have also been significant changes in the composition of expenditures. There has also been considerable variance in the sources of finance for the public sector deficit. The share financed by domestic banks has tended to increase as the size of the deficit has grown. In 1972, 1975, and 1981, approximately 45 percent of a deficit which was 4-5 percent of GNP was funded by domestic credit. Recoveries from each of the three debt crises have involved reductions in the total bank credit to the public sector. In the remainder of this section, we look at spending and revenues more closely. In addition to table 11.1, the discussion refers to the fiscal statistics in the Data Appendix. These tables include a decomposition of revenues by type, and functional and economic decompositions of public sector expenditure. For most of the 1970-86 period, the stance of Korean fiscal policy seems closely tied to the performance of the domestic economy. During the

Table 11.1

Consolidated Public Sector Budget Central Government

Year

Revenues

Expenditures

Deficit

Net Financial Transactions

Consolidated Public Sector

Domestic

Revenues

Expenditures

Deficit

Net Financing

487.60 565.70 654.00 757.70 1.1 17.70 1.692.50 2,511.40 3,184.90 4,385.20 5.769.80 7,280.80 9,246.70 10,074.30 11,595.50 13,039.60 I3,8 13.80

512.30 642.10 846.50 844.70 1,418.60 2,158.60 2,909.70 3,660.50 5,001.00 6,210.10 8.454.50 I 1,357.60 12,296.40 12,546.10 13,962.50 14,820.00

-24.70 -76.40 192.50 - 87 .OO - 300.90 -466.10 - 398.30 -475.60 -615.80 -440.30 1,173.70 -2.1 10.90 -2,222.10 - 950.60 -922.90 - 1,006.20

24.70 76.40 192.50 87 .OO 300.90 466.10 398.30 475.60 615.80 440.30 1 ,173.70 2.1 10.90 2,222.10 950.60 922.90 1,006.20

Bank

Nonbank

Foreign

-28.50

26.60 21.10 61.60 11.70 118.50 110.40 217.10 301.00 187.10 303.10 489.70 649.90 1,130.20 797.10 653.80 506.70

26.60 71.20 50.00 59.90 79.00 143.20 225.50 279.30 359.20 266.30 317.70 533.00 675.80 398.80 309.50 459.50

Panel A (in billions of won) 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

-

-

-

-

-

-

-

-

1,038.70 1,653.60 2,326.60 2,958.40 4,107.70 5,445.40 6,833.20 8.604.80 9,983.20 11,537.50 12,603.30 13,638.30

1,203.00 1,765.30 2,518.90 3,274.40 4,408.00 5.990.00 7.682.00 10,189.00 11,639.20 12,200.10 13,444.60 14,653.90

-

-

- 164.30

- 136.60

-111.70 - 192.30 -316.00 - 3oO.30 -544.60 - 848.80 - 1,584.20 - 1,656.00 -662.60 - 84 1.30 1,015.60

- 354.40

~

- 206.00 159.60 -315.50 104.30 324.90 - 526.70 - 566.10 -288.00 -81.60 9.40 -

~

~

~

- 15.90

80.90 15.40 103.40 212.50 -44.30 -104.70 69.50 -129.10 366.30 928.00 416.10 -245.30 -40.40 40.00

Panel B (in percentage GNP) 1970 1971 1972 1973 1974 13.84 1975 16.38 1976 16.76 1977 16.33 1978 16.96 1979 17.43 1980 18.63 1981 19.07 1982 19.68 1983 19.56 I984 18.98 1985 18.86 ~

-

-

-

-

16.03 17.49 18. I5 18.08 18.20 19.17 20.95 22.58 22.95 20.68 20.25 20.26

-2.19 -1.11 - 1.39 - 1.74 - 1.24 - 1.74 -2.31 -3.51 -3.26 - 1.12 - 1.27 - 1.40

- 1.82

-3.51 -1.48 -0.88 - 1.30 0.33 -0.89 -1.17 -1.12 -0.49 -0.12 0.01

17.82 16.76 15.74 14.09 14.90 16.77 18.09 17.58 18.10 18.46 19.85 20.49 19.86 19.66 19.64 19.10

Source: EPB, Korean Economic Indicarors, and MOF, Government Finance Statistics in Korea. Nore:

- = not available.

18.73 19.02 20.38 15.71 18.91 21.39 20.96 20.21 20.64 19.87 23.05 25.17 24.24 21.21 21.03 20.49

-0.90 -2.26 -4.63 -1.62 -4.01 -4.62 -2.87 -2.63 - 2.54 - 1.41 -3.20 -4.68 -4.38 - 1.61 - 1.39 - 1.39

0.90 2.26 4.63 1.62 4.01 4.62 2.87 2.63 2.54 I .41 3.20 4.68 4.38 1.61 1.39 1.39

115.38 -20.81 42.03 17.70 34.96 45.59 -11.12 -22.01 11.29 -29.32 31.21 43.96 18.73 -25.80 -4.38 3.98

-

107.69 27.62 32.00 13.45 39.38 23.69 54.51 63.29 30.38 68.84 41.72 30.79 50.86 83.85 70.84 50.36

107.69 93.19 25.97 68.85 26.25 30.72 56.62 58.73 58.33 60.48 27.07 25.25 30.41 41.95 33.54 45.67

288

Susan M. Collins and Won-Am Park

ohGNP Expenditure -

Revenue ----

27.5-

25.0 -

22.5 -

12.5 70

1

1

72

1

1

74

1

1

1

76

1

78

1

1

1

80

1

82

1

1

84

Year

Fig. 11.1 Public sector revenues and expenditures (% GNP)

1970-72 economic slowdown following a period of high growth in 1966-69, revenues declined and government expenditures increased. The deficit was financed primarily from abroad during 1970-71, but as aid flows declined, the financing shifted toward domestic sources. Expenditure was sharply contracted in the 1973 economic boom. At the same time, the government undertook a revision of the tax and tax collection systems. However, the fiscal deficit reemerged in 1974-75 during the aftermath of the first oil shock, despite a rise in revenues. Expenditures were increased back up to 20 percent of output. The period of economic recovery from 1976 to 1979 included the stabilization of government expenditures and rising revenues, particularly from the value-added tax (VAT) and from income taxes. There was some fiscal contraction during 1977 and 1979, as part of the Comprehensive Stabilization Plan (CSP) and because of government concern over rising inflation. The budget deficit fell from 5 percent of GNP in 1975 to just 1 percent in 1979. The episode in 1979-83 is of particular interest. Social unrest, the bad harvests, and poor economic performance led the government to increase spending rapidly from 19 percent of GNP in 1979 to 23 percent in 1981, pushing the deficit back to 5 percent of GNP. As shown in table 11.2, the rise coincided with a shift in the composition of spending toward social service^.^ Social services increased from 21.6 to 29.3 percent of total spending, with an almost comparable decline in the share of economic services. In 1980 the big increases came in expenditure for housing. In 1981 the additional expenditure was allocated to education, social security, and welfare (for old age disabilities and government employees). Expenditure growth slowed in 1982 as the size and definition of the public sector was r e d ~ c e d The . ~ removal of some activities from the

289 Table 11.2

KoredChapter 11 Composition of Expenditure Social Services

Period 1975-79 1980-85

Defense

Economic Services

Total

Education Only

29.4 27.9

29.1 22.1

21.6 29.3

13.8 16.3

public sector helps to explain the rapid drop in spending between 1981 and 1983. This shift toward social services, including housing and education, has been maintained during 1983-86. One interesting point is that expenditures as well as revenues remained stable as a share of output during 1983-86, holding the deficit at just 1-2 percent of GNP. In contrast to previous episodes, expenditures were not increased (relative to output) as GNP growth slowed from 11 percent in 1983 to 8.5 percent in 1984, and to just 5.4percent in 1985. Instead, as discussed in chapter 9, the government reacted to the slowdown by further depreciating the exchange rate in hopes of stimulating the export sector. 11.2.3 Fiscal Policy and the Business Cycle We have seen that swings in public sector expenditures have brought about swings in the public deficit. Has the government actively used fiscal policies to influence economic activity? In fact, there is considerable evidence that fiscal policy has been used as a countercyclic policy tool, at least through 1983. There are a variety of difficulties in computing an appropriate fiscal policy indicator for use in assessing the effect on policy. One simple indicator is the relationship between government expenditures and economic growth. This is shown in figure 11.2 in which we plotted the real economic growth rate and expenditures as a share of GNP from 1970 to 1985. The figure shows a clear inverse relationship between the two series. Total expenditures are an inadequate reflection of fiscal policy because they include automatic stabilizers, are sensitive to inflation rate and interest rate changes, and because they do not incorporate changes in tax policy. In a recent paper, Corbo and Nam (1976) have considered alternative measures of “fiscal impulse” which adjust for some of these factor^.^ One measure is calculated using the IMF definition, which takes the difference between the actual budget deficit and a measure of the “cyclically neutral” deficit as a measure of fiscal stimulus, as given in equation (1).

B, is the actual budget surplus, Y, and T a r e actual and potential income, and t and g are the ratios of revenue and expenditure to GNP in a base year (when actual and potential GNP were judged to be equal). The fiscal impulse

290

Susan M. Collins and Won-Am Park

EXP('oGNP)

26 I

Expenditure - GNP Growth

----

GNP GROWTH

0.1563

Fig. 11.2 Fiscal policy and economic growth

measure, then, is FZ,,the change in fiscal stimulus as a share of income (eq. 2).

(2)

FZ, = A(FISt/Yt)

Table 11.3 reproduced from the Corbo and Nam paper, shows the IMF fiscal measure together with real output growth, the actual budget deficit as a share of output each year, and the change from the previous year. Although the actual budget change and the IMF fiscal measure are nearly identical in some years (such as 1978 and 1982), it is clear from the table that cyclic factors were sometimes quite important. For example, the unadjusted measure overstates the expansionary stance of fiscal policy during 1971-72 and especially during 1980, when the adjusted indicator shows a contractionary fiscal policy instead of the strongly expansionary policy suggested by the unadjusted measure. Despite these differences, the adjusted fiscal indicator retains a strong inverse relationship to economic growth over most of the sample period.6 The only exception is the crisis in 1979-80 during which fiscal policy remained contractionary despite the slowdown in real growth. The policy reaction was delayed until 1981, when a strong fiscal expansion took place.

11.3 Monetary Policy and Financial Markets 11.3.1 Financial Markets

We begin with a brief overview of financial markets in Korea. Cole and Park (1983) give an in-depth analysis of the 1945-78 period, while Cole and

291

KoredChapter 11

Table 11.3

Fiscal Impulse IMF Fiscal Measure

Year

Real GNP

Actual Deficit (% of GNP)

1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984

8.8 5.7 14. I 7.7 6.9 14.1 12.7 9.7 6.5 -5.2 -6.2 5.6 9.5 7.5

2.3 4.6 1.6 4.0 4.6 2.9 2.6 2.5 1.4 3.2 4.6 4.3 1.6 1.4

Change

A

B

1.4 2.4 -3.0 2.4 0.6 - 1.8 -0.2 -0.1 -1.1 1.7 1.5 -0.3 -2.7 -0.2

0.7 1.6 -2.2 2.1 0.4 - 1.0 0.1 -0.1 - 1.6 - 0.4 1.7 -0.3 - 1.9

1.3 1.7 -2.2 2.1 0.2 - 1.2 -0.1 -0.0 -1.1 -0.1 1.6 -0.5 -2.1 0.I

0.2

Source: Corbo and Nam (1987b, table 9). Note: The public sector includes Ihe central government (general account, 12 special accounts, and 21 funds) and five public enterprise accounts (grain management, monopoly, railways, communications, and supply), together with two related funds (grain management and supply). Fiscal impulse measure A uses potential GNP obtained from a regression equation. while measure B uses potential GNP from peak-through interpolation.

Cho (1986) and Y. C. Park (1985) provide additional details about recent developments. Readers are referred to these sources for further discussion. In addition to the text tables, we refer to the monetary statistics in the Data Appendix. Korea’s financial system is composed of three segments. Official banking institutions include five commercial banks and six special banks. These institutions have been strictly regulated since they were developed in 1950. Until 1982 the government operated the special banks and, as majority stockholder, indirectly managed the five nationwide commercial banks. Interest rates on deposits and loans were specified by the government. These rates have typically been low relative to inflation rates and to rates available elsewhere, creating a persistent excess demand for credit from the banking system (see app. table A4.4). Government officials have therefore directly influenced the allocation of loans to industrial sectors. Almost no credit from the banking system has been supplied to consumers--consumer loans come almost exclusively from curb markets. The Ministry of Finance has been de facto responsible for making these decisions which are actually carried out by the Monetary Board.7 Liberalizations put into place since 1982 are discussed in section 11.3.5. The second segment, nonbank financial institutions (development institutions, savings institutions, life insurance companies, and investment corporations) have been subject to limited supervision. The third segment is

292

Susan M. Collins and Won-Am Park

the “unregulated” financial institutions, also called the curb market. While not subject to direct controls, government policies have sometimes had a significant impact on them as well. 11.3.2 Controlled Liberalization, 1966-72

The years 1966-69 were boom ones for Korea, with real growth rates averaging 11 percent. Cole and Park (1983) label this period “controlled liberalization” in financial markets .8 There were three major developments. The first was the very rapid growth of bank deposits (table 11.4). Deposits grew by just 19 percent per year during 1961-64 and by less than 3 percent during 1962-64. They grew by 81 percent in 1965 and by 58 percent during 1966-70, rising from 12 percent of GNP in 1966 to 29 percent in 1969. The main reason was the monetary reform of 1965, which had increased the Deposits, Loans, and Foreign Loan Guarantees in the Major Banks, 1%1-85

Table 11.4 A. All Banks’

Deposits Year

Amount

1961 1964 1967 1970 1973 1976 1979 1982 1985

26.4 44.3 208.4

Loans Growth

Amount

-

52.3 84.7 230.4 851.5 ,906.0 ,464.8 1 ,115.7 2 ,895.9 40,724.6

18.8 67.6 56.6 30.2 28.3 38.4 29.2 13.6

800.5

1,766.0 3,725.9 9,878.1 21.309.5 31,221.2

Guaranteesb Growth

Amount

Growth

-

1.6 48.2 196.0 749.3 1,381.3 4,711.6 10,109.0 18,658.3 20,489.0

211.1 59.6 56.4 22.6 50.5 29.0 22.7 3.2

17.4 40.0

54.6 30.8 32.8 35.5 30.8 17.8

-

B. Commerical Banks ~~

Guarantees

Loans

DepoSItS

YW

Amount

% of Total

Amount

% of Total

Amount

1961 1964 1967 1970 1973 1976 1979 1982 1985

19.3 28.2 137.0 505.4 1,179.2 2,495.4 6.042.8 13,080.3 18,157.0

73.1 63.7

12.8 23.1 105.6 441.8 987.5 2,411.3 5,634.9 12,172.2 19,800.4

24.4 27.3 45.8 51.9 51.8 54.0 50.7 48.9 48.6

1.4 9.9 50.3 319.3 585.3 2,165.3 6,217.6 14.322.0 16,382.4

66.0 63.3 66.8 67.0 61.2 61.4 58.2

% of Total

87.5 20.5

25.7 42.6 42.4 46.0 61.5 76.8 79.5

Nore: Data reported in billions of won and as average growth rate (96) over the preceding three-year period. Includes commercial banks, specialized banks, and the Korea Development Bank. bAcceptances of special banks omits those of the Foreign Exchange Bank which are secondary guarantees of acceptances of the other banks.

293

KoredChapter 11

interest rate ceiling on time deposits from 15 to 30 percent. As a result, real interest rates on time deposits averaged 18.9 percent during 1966-69 as compared to -4.6 percent during 1962-64. A second development was that the deposit growth was accompanied by equally rapid growth in loans from the banking system (see table 11.4). However, all industrial sectors did not have equal access to this credit. The government had begun to target specific export industries in conjunction with the first five-year plan. These were given preferential access to loans. Short-term export loans, which were 4 percent of total bank loans during 1966, jumped to 12 percent by 1971. During 1967-71, short- and longterm export loans accounted for 55 percent of the total increase in bank notes and 29 percent of the increase in the money supply (Hong 1979, 117-30). Furthermore, preferential interest rates were established for exporters, for purchases of imported intermediates, and for equipment purchases by export and other target industries. Table 11.5 compares interest rates for export loans and discounts during selected years. As shown, the rates were equal in 1961. But in 1965, exporters paid only 27 percent of the standard discount rate. It is also interesting to compare these interest rates to the costs of borrowing abroad or borrowing in curb markets. In chapter 3, we showed in table 3.6 that the average cost of borrowing in curb markets was 54 percent during 1966-70. With an interest differential of 12.1 percent, it was significantly more expensive to borrow at the domestic discount rate than to borrow aboard. However, the differential between borrowing domestically at the preferential export rate and borrowing abroad was - 5.6 percent. The third development was the massive inflow of foreign funds, guaranteed by the banking system. As discussed in chapter 3, commercial banks began to issue guarantees for loans which had been authorized by the government after 1966. However, these banks simply “facilitated” the

Table 11.5

Export Promotion: Interest Rates Year

Loans for Export

Discount on Bills

1961 1965 1972 1976 1979 1980 1981 1982 1984

13.9 6.5 6.0 8.0 19.0 15.0 15.0 10.0 10.0

13.9 24.0 15.5 17.8 18.5 19.5 16.5 10.0 10.0-11.5

Ratio 1 .oo 0.27 0.39 0.45 0.49 0.77 0.91

I .oo

-

Source: BOK, Money and Banking Stotisrics, 1984, pp. 384-87.

Note: End-of-year interest rates on discounts of deposit money banks. Discounts refer to rates for “superior enterprises,” 1976-81.

Susan M. Collins and Won-Am Park

294

loans. They did not evaluate them and, therefore, it was difficult to hold the banks responsible when firms ran into repayment difficulties. These foreign loan guarantees amounted to just 3 percent of total bank loans in 1961. By 1964 this figure had rise to 57 percent and by 1967, to 85 percent. The growth was especially rapid during 1966-70 (see table 11.4). Thus, real growth was financed by a rapidly expanding financial sector. During 1966-71 domestic banks and nonbanks financed 40.5 percent of corporate sector borrowing, while foreign loans financed 3 1 percent (table 11.6). As we have seen, difficulties emerged during 1970. Growth slowed, the current account deficit rose. Monetary growth was restrained in conjunction with an IMF standby arrangement. Korean firms began to have difficulties servicing their external debts. The difficulties were exacerbated by the devaluation undertaken in 1971 and 1972 to stimulate exports. These factors contributed to a financial crisis in 1972. The government responded by issuing a presidential decree on 3 August 1972. The purpose of the decree was to help out firms which were close to bankruptcy and to stimulate economic growth. In the process, the measure also eliminated many of the liberalizations which had been instituted since 1965. There were five major elements of the decree.' New terms were established for all loans from unofficial lenders to licensed businesses, specifying a three-year grace period, a five-year repayment period, and a 1.35 percent monthly interest rate. More favorable terms were established for some short-term, high interest rate bank loans. A credit guarantee fund was set up to help small and medium-sized industries as well as agricultural businesses and fisheries. The government supplied 50 billion won to an industrial rationalization fund for long-term, low interest rate loans. Finally, interest rates in banking institutions were reduced. The time deposit rate was

lhble 11.6

sourns ofFunds by Corporate Sector (i

sector

Borrowing tiurn monetary sources BanLs Nonbanks Securities Bonds Stock Capital paid in Corporate biUs Government loans Borrowings tiurn abmad Total Soure:

percentages)

1966-71

1972-76

1977-79

1980-83

1984

40.5 (31.9) (8.6) 14.2 (0.7) (12.0) (1.5)

43.3 (29.2) (14.1) 18.5 (1.9) (15.2) (1.4) 1.9 0.0 26.3 100.0

50.4 (32.0) (18.4) 22.4 (6.5) (14.5) (1.5) 2.4 0.1 13.0 100.0

33.3 (17.4) (15.9) 25.5 (10.3) (7.7) (7.6) 5.6 2.6 15.7 100.0

54.4 (19.4) (35.0) 27.7 (12.5) (12.1) (3.1) -0.7 0.7 22.4 100.0

-

0.7 30.9 100.0

BOK. Flow of F u d Accounts and Economic Statistics Yearbook.

Note: Data include nonmpomte enterprises and government enterprises since 1980.

-

295

KoredChapter 11

lowered from 17.4 to 12.6 percent, while the rate on general loans (one year or less) was lowered from 19 to 15.5 percent. One of the most important aspects of the decree was that it implied a significant transfer from lenders to the unofficial market to borrowers. The market seems to have almost disappeared for nearly a year following the decree, however, it reemerged after the rise in oil prices at the end of 1973. The decree also provided the government with an unusual opportunity to collect relatively accurate statistics about the unofficial market as of August 1972. Unfortunately, comparable figures for other years are not available. As Cole and Park (1983, 163-54) discuss, many of the discoveries were surprising. In particular, the total volume of all informal loans amounted to nearly 80 percent of the money supply in 1972. Loans were made to large as well as to small firms, and the industrial distribution of the loans was similar to the distribution of loans from the banking system. 11.3.3 Intervention During the Big Push, 1973-80 Financial market developments during 1973-80 contrast sharply with the growth and liberalization of the late 1960s. Commercial and special banks were heavily regulated, with low nominal interest rates, implying negative real rates throughout much of the period. Consequently, the growth of the banking sector slowed considerably. M2 did not grow relative to GNP. At the same time, the government was in the midst of a major industrial restructuring and was actively promoting the growth of HC industries. Furthermore, interest rate developments significantly increased the attractiveness of bank loans to all domestic borrowers. The result was a substantial increase in government intervention to allocate bank credit, combined with increased expansion of the nonbank financial institutions. As was shown in table 11.5, interest rates on export and other preferential loans continued to be subsidized. But discount rates had been reduced, while rates on loans to exporters had been raised. The subsidy on commercial bank loans to exporters narrowed from 76 percent in 1969 to 42 percent in 1974. As verified in table 11.7, the average cost of borrowing (in sixty-eight manufacturing industries) fell from 18 percent during 1970-1971 to 12 percent during 1973-74, before rising back to 17 percent by 1979. Even more striking is that the variance in borrowing costs across industries ranged from 56-83 percent during 1970-71, but 14-21 percent during 1973-79. However, these figures do not include loans from unofficial sources and they do not incorporate the fact that many firms who would have liked to borrow from the banking system were unable to do so. The figures merely point out that, for those firms with access, the range of interest rates on bank loans narrowed significantly after 1972. In chapter 3 table 3.6 showed that the interest differential between home and foreign markets fell from 12 percent during 1966-70 to 1 to 3 percent during 1972-80. Domestic credit had become much more attractive, relative

296

Susan M. Collins and Won-Am Park ~

Table 11.7

~~

~~~~

Cost of Borrowing in Manufacturing Industries (in percentages) Year

Average

Variance

1970 1971 1972 1973 1974 I975 1976 1977 I978 1979 1980 1981 1982 1983 1984

17.92 18.40 15.05 11.49 12.47 13.59 14.58 15.16 15.52 17.17 20.47 19.50 16.89 14.33 14.46

83.18 55.73 43. I4 14.38 17.56 15.60 16.13 18.96 18.96 21.44 20.99 13.20 8.33 8.05 5.91

Source: BOK, Financial Statement Andvsis. various issues, cited in Cole and Cho (1986, table 7).

Note: Data drawn from sixty-eight different industries and based on the four-digit code classification of the Korea Standard Industry Classification (KSIC).

to external borrowing, for those who had access to bank loans, even if the loans were not at subsidized rates. Table 11.8 shows how additional banking sector credit was allocated across manufacturing industries during 1973-85. The figures clearly show the shift toward HC industries associated with the Big Push. During 1973-74, 66.1 percent of incremental credit went to light industries. The allocation was almost reversed during 1975-79, when 59.1 percent of the incremental credit went to heavy industry. In 1975 heavy industry accounted for only 42 percent of value added in manufacturing. By 1979 its share had risen to 51 percent."

Table 11.8

Incremental Credit Allocation of the Banking Sector (in percentages) Year 1973-74 1975-79 1980 1981 1982 1983 1984 1985

Heavy Industry

Light Industry

33.9 59.1 59.8 52.5 68.4 58.3 56.3 63.4

66.1 40.9 40.2 47.5 31.6 41.7 43.7 36.6

Source: World Bank (1987, table 2.5).

Note: These figures are the share of net credit increase of deposit money banks and the Korea Development Bank. Light industry includes food and beverages, textiles and apparel, wood and furniture, paper and printing, nonmetallic mineral products, and other manufacturing. Heavy industry includes chemicals, petroleum and coal, basic metals and fabricated metal products and equipment.

297

KoredChapter 11

In 1977 the government switched from a positive list loan allocation system in which priority sectors were explicitly listed, to a negative list system; however, the excess demand for funds from the banking system continued to imply that loans were severely rationed. Table 11.6 showed that loans from the banking sector continued to account for approximately 30 percent of total corporate sector financing during 1972-79. However, the importance of foreign borrowing had fallen to just 13 percent during 1977-79. Nonbank financial institutions became considerably more important, accounting for 18 percent of total corporate finance in 1977-79 as compared to 8 percent during 1966-71. As discussed above, a number of measures were undertaken during the 1970s to encourage the growth of the nonbanks, in the hopes of channeling funds away from the unofficial money market. Deposit growth provides one measure of the increasing importance of this sector. While the growth of bank deposits slowed (see table 11.4 and A4.3), deposits in nonbank financial institutions increased from 16 percent of total bank and nonbank deposits in 1971 to 30 percent in 1980. The development is important because, as Cole and Cho (1986) discuss, the expansion of this partially regulated sector offset some of the effects of increased interventions in the banking sector. 11.3.4 Economic Crisis, 1979-81 Thus, 1973-79 was a period of considerable government intervention in financial markets. However, as discussed in chapter 5, concern over inflation and resource misallocations associated with the Big Push led to the CSP announced in April 1979. One of the plan’s hallmarks was that, for the first time, it expressed the view that current government intervention was excessive and that, at Korea’s present stage of development, it was appropriate to begin to liberalize both trade restrictions and financial markets. We have already noted that one component of the CSP was a fiscal contraction. On the monetary side, the plan called for more restrictive monetary policy, increased nominal (and real) interest rates, and an improvement in the preferential loan allocation scheme. In fact, M2 growth slowed from over 35 percent per year during 1976-78 to 25 percent in 1979. There was also a slight increase in interest rates, but accelerating inflation meant that the real interest rate fell to 0.2 percent on discounts and to - 9.3 percent on loans to exporters. By the end of 1979, Korea was in the midst of an economic crisis. The second oil shock, the agricultural disasters, and the death of President Park all contributed to the severe difficulties in 1980. Resuming positive growth and reducing inflation and the debt burden became the government’s top priorities. A stabilization package was initiated in January 1980, supported by a two-year IMF standby arrangement. The package included the familiar

298

Susan M. Collins and Won-Am Park

combination of devaluation, fiscal and monetary restraint, and higher interest rates. In addition, the higher oil prices were to be passed through to consumers. The government raised interest rates in June 1980. The discount rate was increased by 1-2 percent, while the rate on loans to exporters was increased by 6 percent, substantially narrowing the differential (see table 11.5). However, inflation rose from 18 to 29 percent. Futhermore, the deteriorating situation led to a relaxation of some other policies. As discussed above, government expenditures on social services were increased. In addition, the target money growth rates were increased slightly. M2 grew by 26.7 percent in 1980 compared to 25 percent in 1979. In 1981 there was some improvement in economic performance. Growth rates were positive, while inflation and the current account deficit began to decline. As we have seen, there was a further depreciation and a further fiscal expansion. However, monetary policy remained restrictive. M1 grew by just 4.6 percent compared to 18 percent during 1979-80. M2 growth remained relatively constant at 25 percent. The government also continued its financial liberalizations, this time reducing the discount rate (see table 11.5). Finally, price controls were eliminated on a number of key items. 11.3.5 Financial Market Liberalization, 1982-86 Two developments took place during 1982-86. First, the government continued to pursue a restrictive monetary policy, helping to reduce inflation to 2.3 percent. Second, additional steps have been taken toward financial liberalization. We conclude this chapter by discussing each development. A new policy package to revive the domestic economy was initiated in January 1982. The package included further liberalization (to be discussed below), and also called for a loosening of monetary policy. In fact, a financial scandal in the curb market in May 1982 forced two large corporations into bankruptcy. The incident triggered a contraction in loans available from the curb market, threatening many other firms with bankruptcy. In order to bail out these firms, there was a major credit expansion-MI grew by over 45 percent during 1982. (However, M2 growth increased only marginally.) Since then, monetary growth has been quite restrictive. M1 (M2) grew by just 17 percent (15 percent) in 1983 and 0.5 percent (8 percent) in 1985. It has remained low by historical standards during 1985-86. Banking sector growth has slowed markedly since 1979. However, nonbank financial institutions (NBFI) have continued to grow quite rapidly. The ratio of deposits in NBFI to total bank deposits (demand, time, and savings) rose from 36.1 percent in 1979 to 71.7 percent in 1984, and then to 94.8 percent in 1986. The rapid expansion of this sector suggests that the slowdown in bank growth overstates the extent to which financial developments have constrained real activity. Although the annual growth rate

299

KoredChapter 1 1

of M1 slowed from 23.9 percent during 1973-79 to 15.3 percent during 1979-86, M3 growth (which includes NBFI) slowed only slightly, from 27.9 percent during 1973-79 to 26.5 percent during 1979-86. We turn finally to financial liberalization. The fifth five-year plan (1982-86), formulated during 1981, emphasized trade and financial liberalization and a commitment to more neutral government policies. In contrast to recent experiences in Latin America, the financial market liberalization was to be undertaken gradually. Although a number of steps have been taken, an evaluation of this policy shift remains premature." During 1981-83, the government sold its shares in the large commercial banks. It also attempted to restrict ownership by single shareholders to 8 percent. However, as shown in table 11.9, ownership of many of these banks is concentrated among the chaebol. The government has authorized two new commercial banks. It has also relaxed the restrictions on chartering NBFI and on the activities of the branches of foreign banks in Korea. Interest rates were restructured in 1982, although the government continues to set ceilings for bank loans and deposits. As was shown in table 11.5, the subsidy to export loans was eliminated. In table 11.7, we also showed the decline in the variance of borrowing costs across industries, from 21 percent during 1979-80 to just 6 percent by 1984. The government also acted to redress the discrimination against small firms during the Big Push. Small firms have received slightly lower rates than large firms since 1982. In 1984 access to additional credit for the large conglomerates was restricted, increasing the availability of credit to small firms. In 1982 the government also abolished direct credit controls for deposit money banks, switching to a monetary policy based on specified reserve Table 11.9

Conglomerate 1 . Hyundai 2. Daewoo 3. Samsung 4. Lucky Goldstar 5. Hanjin 6. Taekwang 7. Ssangyong 8 . Daelim 9. Shindongah 10. Dong Ah 11. Hanil-Kukje Memo item: ownership by top 1 1

Source: Business Korea.

"Quasi-lead bank. bLead bank.

Conglomerate Ownership of Banks (in percentages)

Cho Heung

2.14 1.23 8.34 1.71

-

3.77 5.57 -

7.98

Korea First

9.35 23.82b 5.69 5.30"

-

7.24

Hanil

7.27 2.22 9.72 5.87" 8.4Sb

9.29b

39.79

54.58

56.54

-

4.48 15.97b

-

2.18

Commercial Bank

4.56 9.90

4.05

-

11.938 5.29

-

10.03 3.69b

-

Bank of Seoul

-

1.91 31.68

22.36

300

Susan M. Collins and Won-Am Park

requirements, rediscounts, and open market operations. However, the government does continue to have considerable influence over the allocation of bank credit. It has intervened heavily to restructure industries which built up overcapacity during the Big Push. As was shown in table 11.8, this implied a shift in credit allocation bank to heavy industry during 1985. Thus, Korea has made some steps toward financial market liberalization in equalizing borrowing costs across industries. Furthermore, as Cole and Cho (1986) point out, the expansion of the only partially regulated NBFI has contributed to a de facto liberalization of the overall financial system. However, authorities have proceeded cautiously, continuing to influence credit allocation. In this sense, the policy shifts may have been more a matter of degree than an “about face” in direction. This viewpoint is advanced by Y. C. Park (1985a). It is too early to evaluate the results of the liberalization, or to attempt to draw lessons from the experience. Korea may soon have some interesting lessons to teach about the economic consequences of a controlled financial liberalization.

12

Income Distribution

As we have studied in detail in previous chapters, Korea underwent a successful macroeconomic adjustment while maintaining high rates of growth. In many cases, rapidly expanding developing countries have been able to achieve remarkable increases in per capita incomes, but one of the costs has been the deterioration of an already skewed income distribution. Consequently, the gains have bypassed a large part of the population. This chapter examines distributive aspects of Korea’s experience from the 1960s to the 1980s. There have been a number of studies of income distribution in Korea. We will refer to them throughout the chapter. Those focusing on the first half of Korea’s rapid growth (through the early 1970s) include Adelman and Robinson (1978), Rao (1978), Renaud (1976) and Mason et al. (1980). The studies consistently found that income was equitably distributed in Korea relative to other developing countries, and that Korea’s economic growth did not require or result in a deterioration. In fact, the rapid economic growth fueled by expansion of labor-intensive export sectors was widely believed to have improved the distribution of income during this period. However, later studies caused considerable concern among policymakers because they seemed to show a noticeable deterioration of income distribution during the 1970s. See, for example, Choo (1977), Szall (1981), and Jung (1982).

301

KoredChapter 12

Ideally, our discussion would review measures of income inequality from the early 1960s to the present so as to examine changes during the economic development. Unfortunately, available studies do not provide consistent time series over the entire period. We have chosen to rely extensively on estimates from Choo (1977, 1978, 1985) because they provide consistent series over the longest time period (1965-82). The EPB uses Choo’s figures as official figures. They are also quoted in the World Bank’s World Development Reports. Data on income distribution typically suffer fram a number of shortcomings, and Korean data are no exception. While there are certainly problems with existing statistics, it is reassuring that there is a general consensus-most of the other studies mentioned portray similar trends. Figure 12.1 shows Choo’s estimates of Gini coefficients and of the decile distribution ratio (the ratio of the income share of the bottom 40 percent to that of the top 20 percent, henceforth denoted as DDR). We follow Choo in identifying four time periods which are evident from the graph: an initial stage, a second period of slight improvement (1965-70), a third period in which the distribution of income deteriorated (1970-76), and a final period in which the deterioration was reversed (1976-82). This chapter is divided into five remaining sections. The next four sections examine the four time periods. In each case, we discuss the probable factors which contributed to changes in distribution. The final section turns to a comparison of Korea with other countries. Using cross-country data, we examine the widely held view that Korea has maintained one of the most equitable income distributions among developing countries.

Year m GIN1

0DDR

Fig. 12.1 Measures of income inequality: Gini coefficients and decile distribution ratios (DDR)

302

Susan M. Collins and Won-Am Park

12.1 The Initial Stages A consensus view is that Korea began its rapid industrialization with a relatively equitable distribution of income. As one indication, in table 12.1 we compare Gini coefficients for selected countries.’ In addition to Korea and other developing countries at early stages of development, we have included figures for Japan and the United States. With the exception of Yugoslavia, the only socialist country represented, Korea has the lowest coefficient. It is substantially below the measure for other developing countries, and comparable to the measure for Japan. The rest of this section investigates some of the social and historical factors which are widely believed to explain the relative equality at the initial stages of Korea’s development. 12.1.1 Homogeneous People In many countries, a significant portion of the inequalities in the distribution of income are associated with cultural, religious, and social differences in the population. In contrast, Korean society is extremely homogeneous, without ethnic minorities, distinct military or bureaucratic classes, or divisive political loyalties. Income differences arising from these factors have been essentially nonexistent. 12.1.2 Japanese Colonization During the Japanese occupation, ownership of property and accumulation of wealth by Koreans was severely limited. About 90 percent of the nation’s industrial assets were managed by the Japanese. Most Koreans were poor, Table 12.1

Gini Coefficients at Initial Stages of Development Country

Year

Gini

Yugoslavia Korea

1963 1964 1963 1953 1963 1963 1960 1961 1962 1961 1953 1960 1963 1964 1961

0.33 0.34 0.35 0.40 0.42 0.45 0.49 0.50 0.52 0.53 0.55 0.56

Japan* India

us.*

Sri Lanka Panama Philippines Venezuela El Salvador

Taiwan

Brazil Mexico Colombia Peru Source: Ahluwalia and Chenery (1974, 42). *These figures are from Renaud (1976, 1 I).

0.56

0.57 0.59

303

KoredChapter 12

essentially relegated to second-class citizens. With the withdrawal of the Japanese after the Second World War, most of the assets which had been owned by the Japanese came into the hands of the Government. These assets were then distributed to the private sector over a period of more than a decade. 12.1.3 Land Reforms After World War 11, the economy of South Korea was critically tied to agriculture, so that equality in the distribution of income depended largely on the distribution of agricultural assets, especially land. Therefore, extensive land reforms after the war played a central role in flattening the distribution of wealth. These reforms are often viewed as the key factor in explaining Korea’s relatively equitable distribution. Reforms under the auspices of the U.S. military government, beginning in 1947, focused on the redistribution of government-owned and vested land. In 1949 the newly established Korean government undertook the redistribution of land owned by big farmers and absentee landlords.2 The substantial impact of these reforms can be seen in figure 12.2. The main beneficiaries were the tenants to whom the land was distributed. In 1947 only 17 percent of rural households fully owned the land they farmed. By 1960 this figure had risen to 74 percent. There are no available data to compare the size distribution of f m income before and after the two land reforrns. However, the change in the structure of land ownership strongly suggests an

EZZ Full Owner

Year hz9 Owner-Tenant

Fig. 12.2 Land ownership composition of rural households Source: Mason et al. (1980, 298).

Tenant

304

Susan M. Collins and Won-Am Park

improvement in the distributive equity within the agricultural sector as a result of the reforms. 12.1.4 Korean War Korea experienced another ‘‘wholesale redistribution” of material wealth and capital in the Korean War (1950-52). Over 40 percent of the manufacturing facilities and 20 percent of the net capital stock were destroyed. The presumption is that the damages were especially detrimental to the upper echelons, where the ownership of these assets was concentrated and, therefore, that one effect of the war was to flatten the distribution of nonagricultural assets. Also, as is often the case during war, the terms of trade turned in favor of the relatively poor agricultural sector and against the nonagricultural sector. This is also believed to have decreased intersectoral inequalities. 12.1.5 Illegally Accumulated Wealth Large amounts of wealth had been accumulated by a favored few during the 1950s. The beneficiaries included individuals who had profited during wartime business activities and corrupt officials of the Syngman Rhee regime who had profited from disposing public and vested property, from bribes and tax evasion, and so on. In the early 1960s, backed by the military, the new government took strong measures to confiscate illegally accumulated wealth. Even though the initial penalties of more than 20 billion won were reduced to 4 billion won, the total amount of confiscated wealth was significant. It represented 16.6 percent of total corporate savings in 1962 and 10.3 percent in 1963. A transfer of this magnitude also contributed, at least temporarily, to a reduction in disparity. However, the implied improvement in the distribution of wealth would have been more substantial if the transfer had come from the most wealthy owners and businessmen, many of whom were in the emerging manufacturing sectors. 12.1.6 Relatively Fluid Society

Two final factors, inherited from the colonial period, contributed to Korea’s relatively equitable distribution: the Japanese system of government based primarily on cooperation and merit and a modern education system. Koreans have a strong traditional drive for learning, and enrollments at the primary, middle, and high school levels have consistently been high. Furthermore, the Korean system earns high marks for maintaining relatively equal educational opportunity.

12.2 Period of Improvement, 1965-70 Countries often experience a deterioration in the distribution of income during the early stages of development. As was shown in figure 12.1, Korea

305

KoredChapter 12

experienced a slight improvement. The Gini coefficient fell from 0.344 in 1965 to 0.332 in 1970. In this section we examine some of the contributing factors. 12.2.1 Employment Creation Korea’s rapid growth, fueled by the promotion of labor-intensive exports, successfully provided jobs for a growing labor force, reducing unemployment and underemployment. Some key labor market statistics for 1963-70 are given in table 12.2. The figures show the rapid decrease in the unemployment rate from 8.2 percent in 1963 to just 4.5 percent in 1970. The reduction is attributable primarily to a dramatic decrease in unemployment in the nonagricultural sector from 16.4 percent in 1963 to 7.4 percent in 1970. The numbers look all the more impressive given that the reduction was achieved while the labor force was growing at an annual rate of 2.1-3.5 percent. Employment creation in the manufacturing and social overhead capital (SOC) sectors was explosive. Jobs in manufacturing grew at an average annual rate of 11.2 percent during 1963-70, while jobs in SOC grew at an average rate of 9.5 percent. Table 12.3, in which we summarize the contribution of commodity exports to sectoral employment, is even more revealing. In manufacturing, for example, exports accounted for just 3 percent of employment in 1960. This had increased to 25 percent by 1970. Job creation in the urban sector meant a significant improvement in the distribution of income among employees. This point is made clearly in table 12.4, in which we compare changes in Gini coefficients over time. The figures are disaggregated into three groups: rural workers, employees, and employers. The figures show that in 1965 employees were the group with the greatest within-group inequality (0.399). By 1970 the figures show considerable improvement-the Gini coefficient had fallen to 0.304, which was well below the estimate of overall inequality (0.332). Table 12.2

The Korean Labor Market, 1963-70 (in percentages) Unemployment Rate

Underemployment Rate*

Year

Labor Force Growth Rate

Farm

Nonfarm

Total

Farm

Nonfarm

Total

1963 1964 1965 1966 1961 1968 1969 1910

2.1 2.2 2.3 3.3 2.4 2.4 2.8 3.5

2.9 3.5 3.1 3.1 2.3 1.9 2.2 1.6

16.4 14.4 13.5 12.8 11.1 9.0 7.8 7.4

8.2 7.7 1.4 7.1 6.2 5.1 4.8 4.5

6.3 6.9 6.2 7.3 6.2 4.9 3.2 4.7

1.9 1.7 1.3 1.1 0.9 0.8 0.5 0.8

4.4 4.1 4.1 4.5 3.1 2.9 1.8 2.1

Source: EPB. Annual Report on the Economically Active, Population Survey, and Major Statistics of Korean Economy, for relevant years.

*EPD defines underemployment to be working less than eighteen hours a week

306

Susan M. Collins and Won-Am Park Contribution of Exports to Sectoral Employment (in thousands of persons)

Table 12.3

Primary

Manufacturing

SOC s n d Service

Whole Industry

Year

Total

Exports

Total

Exports

Total

Exports

Total

Exports

1960 1963 1966 1968 1970

4,680 4,864 4,956 4,907 5,027

56 96

471 610 833 1,176 1,284

15 39 130 199 319

1,871 2,158 2,634 3,972 3,434

5 11 36 53 88

7.028 7,662 8,423 9,155 9,745

76 146 270 371 583

104

118 175

Source: Hong (1980,84) Note: Total employment represents total number of employed persons, while number of workers related to exports is on a man-year basis.

Gini Coefficients by Sector, 1965-82

Table 12.4

Overall

Rural Workers

Employees

Employers

Year

Gini

%

Gini

%

Gini

%

Gini

%

1965 1970 1976 1982

0.344 0.332 0.381 0.357

-

0.285 0.295 0.327 0.306

-

0.399 0.304 0.355 0.309

-

-3.5 14.8 -6.3

-23.8 16.8 -13.0

0.384 0.353 0.449 0.447

-8.1 27.2 0.0

3.5 10.8 -6.4

-

Source: Derived from Choo (1985,12- 15).

The government also initiated efforts to expand job opportunities to those who were not so easily absorbed by job creation in urban areas. For example, the underemployed constituted 6-7 percent of the total agricultural labor force. The comparable figure for the nonagricultural sector was 1-2 percent. The opposite was true of measured unemployment, which was much higher in the nonagricultural sector. The government provided additional job opportunities, primarily in the agricultural sector during the off-peak season. The public works programs included land reclamation, land improvement, reforestation, multiplication of marine resources, and feeder road construction. By the late 1960s, the underemployment rate in the agricultural sector had fallen to 3-4 percent. In sum, the labor-intensive, export-led growth generated enough new jobs to absorb the growing labor force as well as the unemployed nonagricultural workers. The government made deliberate efforts to reduce underemployment in the agricultural sector. Employment opportunities continue to be a very important factor in explaining Korea’s relatively equitable distribution.

12.2.2 The Terms of Trade Existing evidence suggests that although differences between the relatively poor rural sector and the wealthier industrial sector have contributed to overall inequality, between-group inequality has traditionally been much less

307

KoredChapter 12

important that within-group inequality in Korea. The evidence also points to some narrowing of the differences during the 1960s. One development which is consistent with a reduction in the intersectoral inequity during this period is the trend in the terms of trade between the primary and manufacturing sectors. We take the wholesale price index (WPI) for agricultural and marine foods as a proxy for prices in the primary sector, and the WPI for other goods as a proxy for manufacturing prices. The figures are given in table 12.5. The table also provides the weights of each sector to show the changes in the economic structure. While the share of the primary sector fell by nearly 50 percent, from 34.5 to 17.8 percent, its prices increased more than fivefold from 1960 to 197 1. The other goods price index only tripled over the same period. With the exception of 1964-65, the terms of trade continually moved in favor of the primary sector. The government also contributed a little to improving the living standards of the rural population. Although its primary focus was the promotion of industrialization during the 1960s, it did undertake the following measures to help farmers. It helped to settle usurious debts incurred by farmers. It increased the availability and supported the prices of chemical fertilizers, insecticides, and water pumps. It also introduced high-yield rice and cash crops. These policies helped to increase productivity and to improve the living standards of the lower income classes in the agricultural sector.

12.3 Period of Deterioration, 1970-76 As discussed in the introduction, the distribution of income deteriorated substantially in the 1970s. The Gini coefficient increased by 15 percent from Trends in the Terms of Trade between the Primary and Manufacturing sectors

Table 12.5

Other Goods

Agricultural and Marine Foods Year

WPI

1960 1%1 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971

100.0 118.3 128.7 188.7 240.0 243.5 266.1 292.6 330.0 383.1 434.8 530.0

Weight (8) 34.5 33.1 31.8 30.4 29.0 27.6 25.7 23.8 21.8 19.9 17.9 17.8

Term? of Trade 2

WPI

Weight (%)

1

100.0 110.5 120.7 135.2 185.2 208.2 225.0 236.4 251.7 262.8 284.1 300.3

65.5 66.9 68.2 69.6 71.0 12.4 74.3 76.2 78.2 80.1 82.1 82.2

-

-

1.07 1.07 1.40 1.30 1.17 1.18 1.24 1.31 1.46 1.53 1.76

1.07 1.00 1.31 0.93 0.90 1.01 1.05 1.06 1.11 1.05 1.15

Source: Derived from BOK, Economic Sfatistics Yearbook, various years. Note: Terms of trade 1 was calculated using the WPI of 1960 as the base. Terms of trade 2 was calculated using the WPI of the preceding year as the base.

308

Susan M. Collins and Won-Am Park

0.332 in 1970 to 0.381 in 1976. The deteriorating trend was also clearly reflected in a public survey conducted by Hoon Yu (1979), which is summarized in table 12.6. Most of the respondents believed that there was a severe gap between the wealthy and the poor groups in the society. In this section, we examine some of the factors which are likely to have led to the increased inequality. 12.3.1 Growth of Big Business As a catalyst for economic growth, the Korean government thought it advantageous to promote the growth of business conglomerates so as to take advantage of economies of scale and to generate rapid growth and ample employment opportunities. The government provided large export firms with various incentives and preferential arrangements. It made disproportionately large amounts of domestic and foreign capital available at lower interest rates and negotiated abroad for foreign capital. It provided other financial incentives including tax reductions and/or exemptions. It also provided large firms with various kinds of technical and infrastructure support. While these incentives certainly helped to encourage exports, they also led to the concentration of economic activity among a few private conglomerates, the chaebol. Figure 12.3 illustrates the increased concentration during the 1970s. As a percentage of GDP, value added accounted for by the chaebol roughly doubled between 1973 and 1978. Table 12.7 shows that concentration was especially great in the manufacturing sector. Value added of businesses in manufacturing controlled by the largest five chaebols reached a staggering 18.4 percent of GDP by 1978, and by the largest forty-six, 43.0 percent of GDP. Even more striking is the fact that the figure grew at an average annual rate of 35.7 percent, more than twice the average annual GDP growth rate, which Sakong (1980) lists as 17.2 percent. D. M. Kim (1979, 288) estimates that inequality arising from the manufacturing sector accounted for 45.6 percent of the overall inequality in 1971, but increased to 63.8 percent by 1977. These developments had severe effects on small businesses. The shortage of capital reduced rates of productivity and decelerated the growth of value added. As a result, intersectoral inequality among employers deepened Table 12.6

Views on the Level of Income Inequality in Korea Response Extreme inequality Strong inequality Moderate inequality Little inequality No inequality

Source: Yu (1979. 45).

Percentage of Respondents 45.0 41.5 11.3 1.9 0.2

309

KoredChapter 12

I8 r

I

I

I6 -

I4 -

-

12.3

greatly. As was shown in table 12.4, the Gini coefficient for employers increased by 27.2 percent from 0.353 in 1970 to 0.449 in 1976. Choo's ( 1978) Theil decomposition analysis also revealed that inequality among employers was the largest component of overall income inequality in this period. 12.3.2 Rural-Urban Migration The rapid growth of income and the greater employment opportunities in urban areas resulted in a large continuous migration from rural to urban areas. Szalls (1981) points out that the number of workers in rural areas increased by only 0.5 percent annually during 1970-79, while the number in Value Added by Chaebols in the Manufacturing Sector

Table 12.7

Value Added Percentage of GDP

Annual Growtha

Chaebols

1913

1918

1973-78

Top 5 Top 10 Top 20 Top 46

8.8 13.9 21.8 31.8

18.4 23.4 33.2 43.0

35.7 30.0 21.5 24.4

Source: Sakong (1980, 6).

"Compounded annual rate of growth

310

Susan M. Collins and Won-Am Park

urban areas increased by 6.1 percent. However, the increased labor force in urban areas was primarily an increase of low-skilled workers with low wages. Figure 12.4 presents the results from Suh's (1980) study on poverty trends in Korea during 1965-76. The figure reinforces the finding that urbanites were the main beneficiaries of the rapid industrialization during the 1960s. The percentage of urban households classified as poor decreased from 54.9 percent in 1965 to 16.2 percent in 1970.4 However, the figure rose to 18.1 percent in 1976. More revealing is the fact that the distribution of the poor shifted dramatically toward the urban areas during the 1970s. In 1970 only 28.1 percent of the total poor households were in urban areas. This figure had increased to an astonishing 61 percent only five years later. It is also important to point out that income inequality deepened in the industrial sector. As shown in table 12.8, the distribution has been less equitable in the nonagricultural sector than in the agricultural sector, and increasingly so. The nonagricultural DDR was 80 percent of agricultural DDR in 1970, and had fallen to 66 percent by 1976. This sectoral difference implies that the exodus from rural to urban areas contributed to an increase in overall inequality. The influx into urban areas also caused serious difficulties and bottlenecks as the social infrastructure failed to keep pace with growing needs. Problems emerged in the areas of housing, health, education, transportation, cultural

70 60

50 U

g 40

z 2

30 20

10

0

1965

1970

1976

1965

' 7 of Poor Households

eZa Urban

1970 Distribution

1976

Rural

Fig. 12.4 Poor households: percentage of total households and urban-rural distribution Source: Suh (1980, 30).

311

KoredChapter 12 Income inequality by Sector’

Table 12.8

Agriculture

Nonagriculture

Year

Gini

DDR

Theil (%)

Gini

DDR

1965 1970 1976 1982

0.29 0.30 0.33 0.31

0.59 0.55 0.48 0.54

36.0 24.5 24.6 13.6

0.41 0.35 0.42 0.37

0.30

0.44 0.32 0.41

Theil (a) 55.9 63.1 69.4 84.9

Source: Choo (1985, 12-16). *The balance is the Theil share of between-sector inequality.

facilities, and recreation. Shortages in housing, for example, contributed to rampant speculation in real estate, with the relatively privileged classes capitalizing on the opportunities. Overemphasis on growth with inadequate attention to the development of infrastructure meant that available services were poorly distributed, with the upper income groups in urban areas receiving a disproportionate share. This also contributed to a deterioration in the pattern of distribution. 12.3.3 The Reduction of Unemployment

In the previous section, we noted that during the 1960s, rapidly declining unemployment and underemployment rates especially reduced intrasectoral inequality between employees. However, both rates seemed to reach minimum levels at the beginning of the 1970s. Employment hovered around ‘‘full employment” throughout the 1970s until the crisis of 1980. Thus, there is little evidence that the labor-intensive, export-led growth continued to generate improvements in the intrasectoral distribution of income during this period. 12.3.4 Government Control of Labor

In order to maintain competitiveness in world markets, the government exercised tight control over labor activities in key export industries. For example, it is widely believed that the leaders and the activities of labor organizations have been severely restricted. Strikes are forbidden in industries involving foreign capital. Labor has limited negotiating power in those industries under the heaviest control. Consequently, wages are kept artificially low in these industries compared to elsewhere in the economy. Table 12.9 presents average (1976) wages in manufacturing as a percentage of the wages in a number of sectors. Manufacturing, which accounts for over 90 percent of export production, has by far the lowest wages. For example, the average manufacturing wage was less than 40 percent of the average wage in industries related to electricity, gas, and water. Strict control of labor activities and suppression of wages in export-oriented manufacturing

312

Susan M. Collins and Won-Am h r k

Table 12.9

Wages in Manufacturing Industry as a fircentage of Wages in Other

Industries, 1976 Other Industries Electricity, gas, and water Finance, insurance, real estate, and business service Construction Social and personal service Wholesale and retail trade, and hotels and restaurants Mining Transportation, storage, and communication

Manufacturing Wages 38.7 42.9 44.8 52.5 68.7 78.2 80.4

Source: EPB, Major Statistics of Korean Economy, 1986, p. 277.

industries seems to have contributed to greater wage dispersion during the 1970s. 12.3.5 Inflation Rates

Korea experienced relatively high rates of inflation during the 1970s, especially after the first oil price shock. The annual inflation rate averaged 19.8 percent. This inflation especially hurt relatively poor, fixed wage earners because wages are not indexed in Korea, but benefited the relatively more wealthy property owners. Of particular note was the rapid rise in housing prices in the mid- 1970s when speculative real estate investment was at its peak. Many of the low wage urban workers were unable to find housing, even on a temporary basis. Inflation also seems to have widened the income gap between the middle and the lower working classes.

12.4 Renewed Improvement, 1976-82 The trend toward increasing income inequality seems to have reversed during the late 1970s and early 1980s. The Gini coefficient declined from 0.391 in 1976 to 0.357 in 1982, while the DDR rose from 0.371 to 0.431. This section discusses some of the structural and macroeconomic factors we believe contributed to this improvement. 12.4.1 The Wage Structure As we saw in table 12.4, the most important factor in explaining improvements in the overall distribution of income was an improvement in its distribution among employee households. The Gini coefficient for employee households declined by 13 percent from 1976 to 1982. The coefficient for rural households fell by 6.4 percent, while that for employer households hardly changed. Since wages and salaries constitute about 90 percent of employee income, convergence of wages/salaries among different groups of employees provides strong evidence for an improved distribution among employee

313

KoredChapter 12

households. Table 12.10, in which we disaggregate wages according to occupational category, provides strong evidence of such a convergencethere was a drastic shift in the wage structure during 1976-82. While wages in relatively high paying categories (managers, professionals, and technicians) increased substantially relative to other categories during 1971-76, this trend reversed during 1976-85. For example, the average wage of managers was almost five times that of production workers in 1976, but had fallen to 3.4 times that of production workers by 1984. Table 12.11, in which we disaggregate wages by level of education, shows a similar pattern. The wage gap between people with different levels of education had declined continuously since the mid- 1970s. Reasons for the declining wage differential come from both the demand and the supply side. After the great expansion of heavy industry and of large firms in manufacturing, construction, and finance during the Big Push of the 1970s, there was a slowdown in the growth of demand for managers and skilled workers. Furthermore, a nationwide increase in the level of education expanded the relative supply of skilled and highly educated workers. Table 12.10

Relative Wages by Occupation (production workers = 100)

Year

Professionals/ Technicians

Managers

Clerical

Sales

Service

1971 1975 1976 1980 1981 1982 1983 1984

280 266 29 1 243 230 24 1 24 1 235

428 458 474 395 367 345 343 336

243 215 222 162 163 158 155 153

140

107 104 103 100 I00 102 101 101

I23 I12 89 96 134 129 128

Source: Park and Castaneda (1987, 33).

Note: Wages include regular pay, overtime, and bonus payments.

Table 12.11

Year 1975 I977 I979 1980” 1981 1982 1983 1984

Relative Wages by Education (primary school = 100) College/ University

Junior College

High School

306 305 280 256 256 252 245 240

200 206 187 170 168 151

154 I48 135 127 I27 126 124 121

146

145

Source: Park and Castaneda (1987, 34).

aFrom 1980 on, middle school and below

=

100

Middle School

314

Susan M. Collins and Won-Am Park

12.4.2 Price Trends for Agricultural Goods Figure 12.5 shows that prices of agricultural goods rose more quickly than those of nonagricultural goods. This trend was accompanied by a government agricultural policy which maintained a relatively stable ratio of the prices paid by farmers relative to the prices received by farmers. Also, Choo (1985, 22-23) mentions that the ratio of off-farm to total income of small, land-owning farmers increased from 43.5 percent in 1976 to 62.3 percent in 1982. There has also been an increase in average farm household income as a percentage of average urban household income, which has helped to raise the relative living standards of poor rural households. 12.4.3 Inflation and Capital Gains Stabilization policies in the early 1980s concentrated attention on combating inflation. The impressive results can be seen in table 12.12. Since inflation tends to distribute income to property holders and profit earners, the arrest of inflation is likely to have had a positive influence on distribution. The government also took steps to control windfall capital gains. A fiscal reform introduced higher rates of taxation for capital gains and revenues from real estate than for other types of income. The measures seem to have discouraged the rampant, nonproductive, speculative investments in real estate which we have argued contributed to the deteriorating position of the poor during the mid-1970s.

I

uu -

600 -

-

500 -

-

- 400a

-

a) x

1983 Year rn Agriculture

Fig. 12.5 WPI by sector, 1973-85 (1973 = 100) Source: Bank of

Korea (1986, 221).

Manufacturing

1985

315

KoredChapter 12

Table 12.12

WPI CPI

WPI and Seoul CPI Trend, 1974-83

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

42.1 24.3

26.5 25.3

12.2 15.3

9.0 10.1

11.6 14.4

18.8 18.3

38.9 28.7

20.4 21.3

4.7 7.2

0.2 3.4

Source: EPB, Major Szarisrics of Korean Economy, 1986, pp. 206- 10.

12.4.4 Welfare Policies for the Urban Poor Prior to the mid-l970s, there had been little serious government effort toward improving welfare programs and implementing welfare laws for the poor. Policymakers directed their concern to economic growth and industrial development, hoping that the fruits of growth would spread to the lower income groups. The government was involved in small-scale social welfare programs that had been initiated by foreign agencies. Over time, the government became more actively involved in social welfare. It began to set and enforce guidelines and laws; for example, increasing minimum wages and promoting better working conditions and social welfare facilities, etc. It is difficult to ascertain the magnitude of the redistributive measures. However, table 12.13 does indicate that policymakers became increasingly involved in social and distributive aspects of growth. Total expenditure on social welfare programs rose from just 20 percent of total expenditures in the 1970s to over 30 percent in 1980.

12.5 International Comparisons We conclude this chapter with a discussion of income distribution in Korea vis-a-vis the distribution in other countries. There have been a number of other cross-country comparisons. Although the studies are typically conscious of the importance of different stages of development and different levels of income, one can draw only limited conclusions because of the unreliability and inconsistency of available data. The analysis presented here is certainly Table 12.13

Trends in Government Social Welfare Expenditures (in billions of won)

Category Education Health Social Security Housing Others Total

1974 154.7 13.1 61.8 16.2 11.2 257.0

1978 (12.9) (1.1)

(5.1) (1.3) (0.9) (21.4)

605.0 68.3 189.2 55.0 37.0 954.5

1982 (13.7)

(1.5) (4.3) (1.2) (0.4) (21.7)

Source: Suh and Yeon (1986, 5). Note: Expenditure as a percentage of total government expenditures is in parenthesis

1,980.5 140.6 991.5 383.4 77.9 3.573.9

(17.0) (1.2) (8.5) (3.3) (0.7) (30.7)

316

Susan M. Collins and Won-Am Park

no exception. Nonetheless, the findings are interesting and indicative. Our comparisons are based on data from the World Bank and the IMF. We begin by using Kuznets’ well-known “U-Shaped Hypothesis” as a framework to make comparisons across countries. The hypothesis states that as a country develops, the distribution of income tends to improve only after an initial phase of deterioration. One indicator of Korea’s relative position comes from a comparison of Korean data at various stages of development to the U-shaped curve estimated from a cross section of countries. We use the ratio of the income share of the poorest 20 percent of the population to the income share of the richest 10 percent (denoted as RATIO) as a measure of inequality. GNP per capita is used as a proxy for the level of economic development. The figures are listed in ascending order of GNP per capita in table 12.14. The patterns of income distribution are illustrated in figure 12.6. The plot does offer some support for the U-shaped hypothesis. In general, countries with mid-range GNP per capita tend to have lower RATIOs than countries with lower GNP per capita. There is no doubt, however, that the countries with the most equitable distribution are those with the highest income per capita. We explore the relationship more formally by regressing the measure of inequality on GNP per capita. The resulting estimates (using ordinary least squares) are given below, with t-statistics in parentheses. RATIO = 1.280 - 0.344 . log(GNP/Capita) (4.259)(-4.156)

+ 0.025

. log(GNP/Capita)2

(4.591)

-

R2 = 0.506; Standard Error of Estimate = 0.065; Durbin Watson = 2.212; number of observations = 47. All of the coefficients are significantly different from zero. The estimates do indeed imply a U-shaped relationship. This is evident from figure 12.6 which shows both the actual and the estimated RATIO for each country as a function of its GNP per capita. For all four years included, Korea’s actual RATIO is substantially greater than the estimate for Korea’s income level. A second comparison is to examine the relationship between rates of growth and the distribution of income. This is done in figure 12.7, which plots the RATIOs against each country’s average GNP growth rate over the five-year period ending in the year in which the RATIO was obtained. Although the figure indicates considerable diversity, even among countries with similar rates of growth, there seems to be a negative overall correlation between the two variables. The straight line in the figure represents the estimated RATIOs obtained from an OLS regression. Again, the four RATIOs for Korea are well above the estimated “norm.” This is particularly true of the figure for 1970. As mentioned at the outset, these findings are based on relatively simplistic analysis with faulty data. They should be interpreted with caution.

317

KoredChapter 12

Table 12.14

International Comparison of Income Distribution

Country

Year

GNP per Capita in 1980 (U.S. $)

Bangladesh Nepal Sri Lanka India Tanzania Sierra Leone Kenya Sudan Egypt Indonesia Philippines Korea1 Thailand Zambia Brazil Korea2 El Salvador Turkey Mauritius Peru Malaysia Costa Rica Korea3 Panama Korea4 Mexico Portugal Argentina Venezuela Ireland Hong Kong Trinidad Spain

1977 1977 I970 1976 1969 1969 I976 1968 I974 1976 1971 1965 1976 1976 1972 1970 1977 1973 1981 1972 1973 197I 1976 1970 1982 1977 1974 1970 1970 1973 1980 1976 1981 1980 1977 1982 1979 1979 1975 1981 I979 1976 1981 1978 1981 1981 1980 198I 1982 1978

119.0 169.1 201.5 217.1 253.0 322.0 336.0 355.0 370.0 424.0 555.3 572.3 604.3 710.0 728.0 834.6 975.0 1,095.7 I , 135.0 1.154.2 1,171.3 1,238.0 1.330.2 1,454.7 1,657.5 1,833.2 2,062.7 2,605.8 2,924.0 4,406.0 4,578.6 5,003.4 5,037.0 5,146.7 5,920.0 7,242.0 8,684.0 9,242.0 9,501 .O 10,138.0 10,144.0 10,283.0 10,536.0 11,228.0 11,445.0 11,457.0 12,838.0 13,364.0 13,452.0 14,916.0

Israel Italy New Zealand Japan U.K. France Finland Belgium Australia Nether lands West Germany Canada Denmark

us.

Sweden Noway Switzerland Source: World Bank

Average GNP Growth (for last 5 years)

W Income Owned by Bottom 20% ('4)

% Income Owned by Top 10% (B)

7.2 2.8 7.9 3.0 4.8 6.3 5.8 2.6 2.9 8.4 5.2 6.5 7.0 3.5 9.3 12.7 5.4 6.8 3.1 4.6 9.4 6.8 9.6 7. I 5.4 5 '5 7.2 4.3 4.9 5.5 8.9 4.1 2.0 2.6 3.1 1.2 4.7 0.8 4.0 3.5 1.9 3.4 0.2 2.0 2.6 - 1.3 3.2 1.1 3.0 1.4

6.2 4.6 7.5 7.0 5.8 5.6 2.6 4.0 5.8 6.6 5.2 5.8 5.6 3.4 2.0 7.3 5.5 3.5 4.0 1.9 3.5 3.3 5.7 2.0 6.9 2.9 5.2 4.4 3.0 7.2 5.4 4.2 6.9 6.0 6.2 5.1 8.7 7.0 5.3 6.3 7.9 5.4 8.3 1.9 5.3 5.4 5.3 7.4 6.0 6.6

32.0 46.5 28.2 33.6 35.6 32.8 45.8 34.6 33.2 34.0 38.5 25.8 34.1 46.3 50.6 25.4 29.5 40.7 46.7 42.9 39.8 39.5 27.5 44.2 28.1 40.6 33.4 35.2 35.7 25.1 31.3 31.8 24.5 22.6 28.1 28.7 22.4 23.4 30.5 21.7 21.5 30.5 21.5 24.0 23.8 22.3 23.3 28.1 22.8 23.7

A/B Ratio

.i9 .10 .27 .21 .16 .17 .06 .12 .17 .19 .14 .22 .16 .07 .04 .29 .19

.09 .09 .04 .09 .08 .21 .05 .25 .07 .16 .13 .08 .29 .17 .13 .28 .27 .22 .18 .39 .30 .17 .29 .37 .18 .39 .33 .22 .24 .23 .26 .26 .28

Susan M. Collins and Won-Am Wrk

318

o"'"''""'' 2

2.6 2.8 3.0 3.2 3.4 3.6 3.8 Logged GNP per Capita (1980US$) o Actual Ratio - Estimated Ratio 2.2

2.4

4.0

4.2

Fig. 12.6 Income ratio versus GNP per capita n A7 . "_

JAP

0 0

0.35-

oWG KOR2 C 0

0.054 2 "

0

" '2

1

4

1

I

6

1

I

8

I

I

I

10

I

12

Avg GNP Growth Rote over last 5 years - Estimated Ratio Actual Ratio

Fig. 12.7 Income ratio versus average GNP growth over last five years

However, it seems difficult to refute the view that Korea's distribution of income has remained far better than the distribution in other countries at similar levels of development. Although Korea did experience a period of deteriorating distribution during its development, the conclusion that Korea achieved rapid growth rates with relative equity by international standards seems fairly certain.

319

13

KoredChapter 13

Lessons From Korea’s Experience: A Synthesis

The purpose of this chapter is to synthesize our analysis of Korea’s experience. There are four central questions: 1. 2. 3. 4.

What caused Korea’s debt crisis? How was Korea able to achieve rapid, successful recoveries? What role has external borrowing played in the experience? Are there lessons for other debtor countries?

To answer these questions involves integrating a number of interrelated factors. In section 13.1 we summarize our conclusions about each of these pieces individually. We put the pieces together and examine the implications, answering questions 1-3, in section 13.2. The final sections discuss the lessons to be learned and the prospects for the Korean economy.

13.1 The Pieces 13.1.1 External Debt Foreign capital inflows have played a critical role throughout Korea’s recent development. The preceding discussion has already emphasized the importance of foreign aid in the decade following the Korean War and documented the rapid accumulation of external debt, concentrated during 1966-69, 1974-75, and 1979-82. Rapid growth of output and exports has meant that Korea’s actual debt burden grew much more slowly than the nominal debt stock. Although the debt (denominated in U.S. dollars) grew at an average rate of 34.6 percent in the eighteen years from 1964 to 1982, the debt to GNP ratio reached 53.5 percent, while the ratio of debt service to exports reached only 20.6 percent. Korea ranked only eleventh in terms of its debt/GDP ratio and fifteenth in terms of its debt service ratio.’ Korea’s growth performance is a key piece of the puzzle surrounding the quick adjustment to the debt crisis in 1979-82. External borrowing in Korea was used primarily to finance current account deficits. In particular, there has been little capital flight. This points to an analysis of domestic savings and investment as the key to explaining debt accumulation, because the current account deficit, or foreign savings, finances the portion of investment not financed domestically. It is also notable that Korean debt has been carefully monitored by the Ministry of Finance since the borrowing began in the early 1960s. Applications for loans must be approved, and the government has actively used the allocation of foreign (and domestic) credit as part of the industrial policy, providing growth incentives for particular industries and firms.

320

Susan M. Collins and Won-Am Park

Borrowing is a central component of economic planning in Korea. In many periods, the amount of borrowing required to finance desired investment was forecast quite accurately, however, unexpected external and internal developments during 1974-75 and 1979-81 meant that the forecast turned out to be a sizable underestimate. In any case, the Korean government has maintained excellent debt statistics throughout the period. It was not faced with the additional difficulty of faulty or incomplete information in responding to the 1979-80 crisis. 13.1.2 Economic Growth Korea’s phenomenal growth rates since 1965 have been well documented. Of particular significance is that Korea was able to avoid the dramatic slowdown which most of the other fast growers experienced after the first oil price shock. A detailed analysis of the economic sources of Korea’s growth identifies fixed capital accumulation as the central factor. Korean growth during the 1960s was attributable to a combination of increased factor accumulation, improved resource allocation, economies of scale, and technological improvement. Fixed capital accumulation accounts for the 1.1 percent average annual growth during 1963-72. In contrast, capital accumulation accounts for a growth rate of 2.6 percent during 1973-82. Korea offset reductions in factor productivity after the first oil shock with a substantial increase in investment. Increased labor has also played a key role. The average work week increased throughout the period to 54.8 hours, placing Korea at the top of the International Labor Organization’s (ILO) list. Furthermore, the work force is well-educated and disciplined. It is interesting to point out that the sources of Korean growth are quite different from the sources of Japanese growth during its 1953-71 period of rapid acceleration. Factor accumulation explains only 45 percent of the Japanese growth rates as compared to 60 percent of Korean growth rates. A decomposition from the demand side shows that government consumption played at best a minor role. Investment was consistently strong. However, since import requirements for investment ranged from 0.38 to 0.48, investment has been only a moderate source of demand for domestic output. Not surprisingly, exports emerged as the “engine of growth” in Korea during the 1975-85 period, as well as previously. The data also document that labor productivity has consistently grown faster in the manufacturing than in the nonmanufacturing sector. The domestic price of manufactured goods-a proxy for the tradable goods sector-rose relative to the price of other-nontraded-goods throughout the 1960-85 period. However, this real appreciation has represented technical progress and not a deterioration in external competitiveness or a reallocation of resources away from the production of tradables. One of the most enviable aspects of Korea’s recent recovery has been trade balance improvement combined with growth. In contrast, most debtor

321

KoredChapter 13

countries have achieved trade surpluses through recession-induced reductions in imports. In fact, the very low income elasticities of Korean imports during 1981-83 are unusual by Korean standards. They are explained in large part by disastrous harvests during 1978-80, necessitating a surge in food imports, followed by a very favorable harvest during 1981-82 which both raised domestic output and reduced imports. Exports did not begin to recover until 1983, and this turnaround is explained by a combination of increased world demand, a terms of trade improvement, and the lagged impact of a real depreciation and numerous targeted investments gradually coming on stream. 13.1.3 Investment and the Five-Year Plans

As we have discussed, Korea instituted a series of five-year economic plans, beginning in 1962. The first step in the formulation of these plans was to determine the investment required to achieve a desired rate of growth. Thus, investment for growth has been the number one priority, while external borrowing emerges at the other end as the residual-the gap between investment and available domestic financing. In the mid-1960s it was an important supplement to declining foreign aid. More recently it has been used to substitute for shortfalls in domestic (especially household) savings. The plans also identify particular sectors of the economy for growth, and the government has actively controlled the allocation of credit, thereby playing a key role in determining the industrial concentration of capital accumulation. Even the best plan will have little impact if it cannot be implemented. Korean policymakers have used exchange rate, tax, and credit allocation policies to establish the appropriate incentives to domestic firms. However, a large part of the success of the five-year plans is attributable to Korea’s centralized decision making combined with a very close link between government and business. Authorities maintain current data, including information about individual firms’ performance. Decisions are made quickly, and policies are pragmatic, often intervening directly at the firm level. One implication of this approach has been that, by selecting previously successful firms to undertake new projects, the government has helped to create a number of large conglomerates (chaebol) and a highly concentrated industrial structure. 13.1.4 Saving Behavior Korea’s saving rate has risen from 14 percent in 1965 to 28 percent in 1986, however, the remarkable secular increase has been interrupted periodically. These plunges have accelerated foreign borrowing so as to finance desired rates of investment, leading to a crisis. Two aspects are especially notable. First, savings declines are primarily attributable to drops in household savings and not to deteriorating

322

Susan M. Collins and Won-Am Park

government budgets. Second, current account improvement during the adjustment has not been brought about by cuts in investment to close the gap. Instead, the key has been the recovery of household savings, supplemented by increased government savings. Disaggregation shows that both the secular rise and plunges occurred in the household sector. The performance is explained quite well by a model in which the marginal propensity to consume out of permanent income is higher than out of transitory income. Although interest rates are estimated to affect savings positively, we do not find the estimates to be significantly different from zero. Thus, Korea’s strong growth, leading to upward revisions in permanent income, accounts for the secular rise, while growth slowdowns account for the 1970-71, 1975, and 1980-81 plunges, as households reduced savings to smooth consumption. 13.1.5 Exchange Rate Policy Overall, Korea has followed a consistent, credible exchange rate policy, maintaining a competitive, sometimes undervalued, real exchange rate with low variance. In adjusting to external imbalance during both 1974 and 1980, the policy packages included a substantial (20 percent) one-shot devaluation in addition to a change in the exchange rate regime. The nominal exchange rate was fixed to the U.S. dollar during 1975-79, during which time authorities permitted a 14 percent real appreciation. Since 1980 the exchange rate has been continually adjusted vis-h-vis a basket of currencies. The real exchange rate depreciated by 6 percent during 1980-82 and by a further 14 percent during 1982-86. It appreciated gradually during 1987. 13.1.6 Wages and Competitiveness Even more striking than Korea’s success in maintaining external competitiveness throughout most of the 1965-86 period is the fact that real depreciations were often (e.g., in 1973 and 1983-86) accompanied by real wage increases. Again, rapidly increasing labor productivity is the key to the puzzle, providing a buffer which can be divided between increased competitiveness and increased real income. During 1965-72 real wages grew at an average annual rate of 9.0 percent, while productivity grew by 14.4 percent. However, during 1973-79 real wages grew by 12.5 percent, outpacing the 11.1 percent productivity growth. shortages in skilled labor associated with the Big Push toward heavy industrialization led to rapid nominal wage gains. Unit labor costs, measured in dollars, grew 2.3 times more quickly for Korea than for Taiwan, a major competitor in third markets. It is important to point out that real wages declined both at the outset of Korea’s export-led growth and as Korea reestablished its competitive position after the real appreciation in 1975-79. During 1960-64 the average

323

KoredChapter 13

annual real wage decline was 1.96 percent, despite 7.46 percent productivity growth. Real wages fell at the beginning of the adjustment (1981-82), with all of the productivity gains going to reduce unit labor costs. This, plus exchange rate depreciations, has dramatically improved Korea’s competitive position since 1982. We note a few other characteristics of Korea’s labor market. Worker organizations are extremely weak. There is evidence that they have increased job security, but not that they have influenced wages. Bonuses average 15 percent of employee compensation, which enhances flexibility. Finally, the fact that wages are not indexed to past inflation rates has meant that inflation shows little inertia. 13.1.7 Trade Policy

Korea’s well-documented switch from a policy of import substitution to one of export promotion during 1960-64 involved a significant import liberalization. However, trade policies continued to play an important role, with tax preferences and interest rate subsidies becoming the primary mechanisms after 1965. Through the mid- 1970s, export incentives were maintained with little variability. Subsidies were used to compensate exporters during periods of real appreciation. Imports were substantially liberalized during the 1960s. Restrictions increased during the Big Push and have been gradually relaxed since 1980. Quantitative restrictions, domestic content, and other regulations have remained critical, so that tariff rates substantially underestimate the degree of protection. For example, the share of manufactured items subject to import restriction jumped from 34 percent in 1968 to 61 percent in 1978. These restrictions have been important in developing infant industries such as automobiles and steel, allowing Korea to become competitive enough to begin exporting these products. The restrictions help to explain why almost all Korean imports are raw materials, intermediate products, or capital goods, with consumer products amounting to less than 5 percent of Korean imports. Korea also stands out in not maintaining a structure of protection which penalizes agriculture. The political economy of that outcome seems to be linked to the relatively equitable income distribution due primarily to a major land reform undertaken during the 1950s. 13.1.8 Industrial Policy

Korea has been extremely successful in selecting growth industries and in managing their industrial transition. A large part of the success lies in having developed credible, comprehensive strategies in which investment projects formed the cornerstones of five-year macroeconomic plans. Korean businesses targeted for expansion have not been concerned about policy inconsistencies or government policy reversals. They have been given

324

Susan M. Collins and Won-Am Park

preferential access to domestic credit, to external funds, and to imported materials. The government has maintained its commitment, bailing out firms threatened with bankruptcy during downturns or financial panics. It has also created a few conglomerates which are enormous, even by world standards. In retrospect, some of the policies were mistakes, particularly during the Big Push in 1974-79. For example, government intervention led to substantial overcapacity in petrochemicals. However, the entire policy should by no means be written off as a mistake. Many of the investments in heavy industries are beginning to pay off, and exports of these products are growing rapidly. 13.1.9 Fiscal Policy Fiscal policy in Korea is perhaps most notable for the role it did not play in the accumulation of external debt. Government savings have been positive in every year since 1962. The budget deficit (which includes public investment as an outlay) has been kept under control, ranging from 1 to 4 percent of output. A tax reform and switch to VAT in the 1970s succeeded in raising revenues from 15 to 18 percent of GNP. Large deficits in 1975 and 1980-81 are attributable primarily to increased expenditures in the Grain Management Fund. Social expenditures, such as education and housing, have been low historically, but have risen over time. Since 1980 they have amounted to 30 percent. Indicators of fiscal stance show that fiscal policy has been countercyclical, used by the government in attempts to “fine tune” economic performance. Overall, fiscal deficits have not been financed through rapid money creation. The deficits themselves have been relatively small. Also, authorities have alternated between domestic and foreign credit. For example, after jumps in bank credit to the public sector during 1980-81, net credit was reduced during 1982-84. 13.1.10 Monetary Policy The banking system, including the Bank of Korea, has been monitored by the MOF since 1962 so that macroeconomic policymaking is extremely centralized. We highlight four aspects. The first is the key role for credit allocation in the industrial strategy, as discussed above. A second objective of monetary policies (especially interest rate adjustments) has been to increase household savings. It is very difficult to quantify how effective this tactic has been. Third, Korean financial markets have three levels. The official banking sector is highly controlled, although there has been some liberalization since 1982, including the privatization of five commercial banks. There is also a partially controlled nonbank financial sector and an unorganized curb market. The latter two have added flexibility to Korea’s financial system, providing credit (often at high interest rates) to those firms which were not given access to scarce bank credit. Since a financial scandal in 1982,

325

KoredChapter 13

however, the curb market has shrunk considerably. NBFI’s have been growing rapidly, accounting for half of all deposits of banks plus nonbanks in 1985, as compared to one-fifth in 1978. Korea’s financial system has been anything but a unified system in which credit is allocated by market forces. While it is certain that the outcomes under such a system would have been different, it is very difficult to assess whether they would have been “better” or “worse.” To us, the most sensible conclusion is that the Korean government successfully used an active and pervasive policy of intervening in financial markets to promote its growth objectives. There has been some movement toward financial liberalization of the banking sector. But unlike the trade liberalization, the changes seem to have been greater on paper than in practice. Credit allocation remains a cornerstone of Korean industrial policy.

13.1.1 1 Income Distribution Korea began industrialization with a relatively equitable distribution of income. The main reasons for this initial situation were the major land reform and the devastation from the Korean War which helped to level wealth holdings. During the rapid growth of the 1960s, Korea seems to have been able to, if not improve the income distribution, at least maintain the existing one. Unlike many other developing countries, Korea avoided a deterioration in distribution during the early stages of industrialization. One of the main factors seems to have been the rapidly expanding employment opportunities generated by the export-led growth. The distribution of income began to deteriorate during the 1970s. One factor was the explosive growth of big business during the Big Push and the resulting growth in inequality among employers. However, available evidence does suggest a renewed trend toward increasing equality since the late 1970s.

13.1.12 Two Themes Two unifying themes emerge from these eleven pieces in the puzzle of Korea’s successful performance. The first is the importance of rapid growth rates, rising labor productivity, and expanding human and physical capital resources. These factors gave Korea the leeway to borrow heavily, while keeping the burden of debt repayments manageable, and to avoid squeezing real incomes when increasing international competitiveness. The rapid productivity growth in tradable goods production has eased the problem of mobilizing and transferring domestic resources in order to pay external debts. The second theme is the use of active, interventionist government policy which is credible, consistent, and coherent. These policies placed investment as the number one priority and led the economy through a fundamental industrial restructuring.

326

Susan M. Collins and Won-Am Park

13.2 Implications: A Synthesis In this section, we synthesize the pieces discussed above in order to answer the questions posed at the outset. The first question, important in distinguishing Korea’s experience from that of many other debtor countries, is why did the debt crises occur. Since 1965 Korea has been vulnerable to external and internal shocks because of its determined investment policy which left no buffers between desired investment and domestic savings. External borrowing was treated as the buffer or residual. The country was hit by a number of external shocks, in particular oil price and interest rate changes, but the role of internal shocks must not be underestimated. During 1974-75, terms of trade deterioration accounts for only one-third of the current account deficit. And like 1970-72, this period seems better described as a slowdown than as an economic crisis. External factors were more important during 1979-80. However, the crisis would have been much less severe if these factors had not been exacerbated by the agricultural disaster, political turmoil, and previous policy mistakes. How was Korea able to recover so quickly from slowdowns and crises? We believe the central factor has been successfully distinguishing between permanent and temporary shocks and responding appropriately. The devastation of the Korean War was clearly a permanent shock. In designing and carrying through the impressive structural readjustment of the 1960s, policymakers learned how to put together an adjustment package that worked. They chose to embark on another structural readjustment during 1973-79 because of pessimistic forecasts for medium-term growth using the industrialization path of the 1960s. In contrast, Korea borrowed to smooth adjustment to the 1973 jump in oil prices because the shock was judged unlikely to alter the medium- to long-run prospects for heavy industry. However, policymakers have not been rigid. A third shift in focus came as doubts emerged about the efficacy of further heavy industrialization, and the economy found itself saddled with massive debts accumulated during 1979-80. This point is closely linked to the role of external debt in Korea’s adjustment. The debt has been used to supplement domestic savings in financing investment, enabling faster rates of growth. The debt has also been used to smooth over temporary shocks, without jeopardizing the ongoing structural adjustment plan. However, Korea has been admirable in not using external borrowing to avoid undertaking a structural readjustment. What is the adjustment package that has worked for Korea? The centerpiece has been a comprehensive investment plan, operationalized through competitive exchange rates, credit rationing, tax and other incentives for targeted industries, trade policies, and allocation of external credit. Initial declines in real wages have helped to boost competitiveness,

327

KoredChapter 13

but once the investment-growth cycle got going, productivity gains were split between raising wages and enhancing competitiveness. Traditional macroeconomic “stabilization” tools-monetary expansion and fiscal deficits-have been important in the passive sense that they have been kept in line. Fiscal deficits have remained small and authorities have been careful to limit domestic credit expansion to the public sector. However, these policies played at best a supporting role in pulling Korea out of slowdowns and crises. Both were quite variable with many reversals during 1980-8 1. By the time a definite monetary-fiscal expansion emerged in 1982, Korea was already well on the way to recovery. Good fortune has also helped Korea to recover. In particular, the first oil shock gave Korea an unexpected boost during 1976-78 through revenues from construction in the Middle East, and the recent recovery was fueled by terms of trade improvements beginning in 1981.

13.3 Lessons We begin by pointing out two lessons which, most certainly, cannot be learned from the Korean experience. The first is how to design short-run macroeconomic stabilization packages. There are no ‘‘quick fixes” in Korea’s recent history. The second is the benefits of liberalized trade regimes and domestic and international capital markets. Active intervention has been a mainstay of Korean policy. However, there are numerous examples of extensive intervention in other countries which has coincided with poor economic performance. Korea surely contains lessons about which types of intervention are likely to be effective. We can draw four lessons from Korea’s experience. First, the value of credibility, consistency, and coherence in economic policy. As in Korea, this may well necessitate coordinated trade, industrial, and credit policies in order to promote infant industries. It certainly includes maintaining a competitive real exchange rate, together with a sustainable fiscal policy, and moderate monetary growth. A second lesson is the value of long-term structural adjustment policy with investment as the top priority. When things have gone well in Korea, high rates of investment have stimulated growth, raising domestic savings and enabling Korea to finance external debts. When difficulties emerged, Korea consistently avoided cutting investment so that the economy was poised to resume growth when external and/or internal conditions improved. Of course, the difficulty with such an investment program is that it must be financed, and extensive borrowing can lead to repayment difficulties. The Korean experience highlights a third lesson-the value of external borrowing in enabling an investment policy to be carried through, as distinguished from external borrowing used to avoid structural adjustment.

328

Susan M. Collins and Won-Am Park

Finally, Korea’s ability to recover from downturns emphasizes the value of monitoring economic performance and maintaining accurate statistics for key variables.

13.4 Prospects The prospects for rapid growth to continue over the short- to medium-term are excellent. Our view is based both on Korea’s recent good fortune (especially the decline in oil prices and interest rates and the appreciation of the Japanese yen) and on Korea’s very competitive position as a result of real depreciations in 1985-86 and heavy investments over the past decade which are beginning to pay off. Nonetheless, there are challenges and uncertainties. For example, the favorable external developments from which Korea has benefited greatly in recent years may not continue. Tensions in the Middle East heighten uncertainties about oil price movements. Exchange rate movements, particularly the yen-dollar-won rates, pose new issues for Korean competitiveness. While some gradual real appreciation is unlikely to disrupt growth prospects, it may well be important to mitigate protectionism in the United States. Continued access, especially to U.S. markets, is critical to the continuation of Korea’s export-led growth. Current efforts to identify new markets for Korean products and to reduce dependence on the United States are especially timely given the uncertainties about U.S. trade policy. Many have expressed surprise that Korea decided to reduce external debt. There remain many high-return investments. There are also arguments for borrowing to take advantage of current favorable external conditions through investment and stockpiles. On the other hand, Korea has a substantial external debt, and reducing it will reduce the potential for future debt crises. Furthermore, careful and forward-looking decision making has been an asset in the past. Caution today may well pay off handsomely as external conditions become less favorable down the road. In addition to the uncertainties about external developments, there were changes in the internal political and economic situation following the ruling party’s announcement in June 1987 that it would initiate steps to promote a more democratic society. The new atmosphere generated a number of volatile domestic issues. Labor disputes stand out as the most important, with wage compensation and labor rights emerging as major issues in negotiations between workers and employers. It is impossible to predict how the balance between these groups will shift, how much wages will increase, or how severe work stoppages will be. The most recent forecasts for Korean macroeconomic performance remain very favorable. We conclude our discussion with a look at the outcome projected in the sixth five-year plan, initiated in 1987. The major targets are given in table 13.1. As shown, real GNP is projected to grow at an average annual rate of 7.3 percent during 1987-91, with WPI inflation just 2 percent. The target current account deficit is $5 billion.

Table 13.1

Selected Macroeconomic Targets Sixth Plan Targetsb 1987-91

Units”

1985

1986

1987

1988

1989

1990

1991

(annual average)

GNP (current) GNP (1980 prices) Growth rate

$ billion

$ billion %

83.7 86.6 5.4

94.0 97.0 12.0

108.9 104.8 8.0

124.1 112.6 7.5

139. I 120.5 7.0

156.0 128.9 7.0

175.0 137.9 7.0

13.2 7.3 7.3

Total investment to GNP Domestic savings to GNP Foreign savings to GNP

9% 8 %

31.1 28.6 3.1

30.2 32.8 -2.7

29.1 32.8 -3.1

30.5 33.0 -2.5

30.8 33.2 -2.4

31.0 33.3 -2.3

31.3 33.5 -2.2

30.7 33.2 -2.5

GNP deflator Wholesale price

Increase rate (%) Increase rate (%)

4.1 0.9

1.5 -2.2

3.5 2.0

3.5 2.0

3.5 2.0

3.5 2.0

3.5 2.0

3.5 2.0

Current account

$ billion

-0.9

4.5

5.0

5.0

5.0

5.0

5.0

5.0

Trade balance Export of commodities Import of commodities

$ billion $ billion $ billion

-0.0 26.4 26.4

4.3 33.6 29.3

5.0 38.0 33.0

5.3 41.9 36.6

5.5 45.8 40.3

5.5 49.9 44.4

5.5 54.4 48.9

9.4 10.3

Invisible trade balance

$ billion

- 1.5

- 1.0

- 1.0

- 1.2

- 1.2

Total external debt Total external assets Net external debt

$ billion

46.8 11.2 35.6

41.8 13.0 28.8

39.0 14.3 24.7

36.5 15.8 20.7

34.5 17.5 17.0

$ billion $ billion

-0.8 44.5 11.8 32.7

-1.2 32.9 19.4 13.5

Source: Government of Korea (1986)

%ased on new SNA. Fur a comparison of the old and the new SNA methods, see the Data Appendix. By and large, the new SNA, by including previously unrecorded or underrecorded economic activities, yields higher GNP growth estimates. bThese are preliminary estimates based on data available.

330

Susan M. Collins and Won-Am Park

Data Appendix Overall, Korean data are excellent. Therefore, the main purpose of this appendix is to provide additional tables as a supplement to those in the text. Most of the data are taken directly from published statistical sources. They do not require detailed explanations or qualifications. There are, however, two issues which do warrant attention. The first, and more important, is that a new System of National Accounts (SNA) was recently adopted by the Bank of Korea. National Accounts data based on this new system are only available for 1970 and after. The revisions have had a substantial effect on some figures. For example, average GNP growth during 1980-84 was 7.2 percent under the old SNA and 8.1 percent under the new SNA. Unless noted otherwise, calculations and discussions in the text, as well as data tables in both the text and the appendix, are based on the new SNA. Unfortunately, comparisons with studies which rely on data based on the old SNA should be undertaken with caution. There are six major changes in the new SNA. This summary is based on “A Note on National Accounts Methodology” in World Bank (1987, v. 1, p. 133). 1. Transactions of goods and services are classified by activity, those of income and expenditure are classified by institutional sector. 2. Government and private nonprofit institutions are classified independently. 3. Both gross intermediate output and final products are estimated. 4. Imputed financial services and import duties are introduced. 5. Intermediate inputs are subtracted from gross outputs to estimate current value added; and 6. The double deflation method is used extensively in the estimation of current value added. The second issue is that the Korean balance of payments presentation treats foreign borrowing by domestic banks as “below the line.” Therefore, it does not show up in the capital account figures, appearing instead as one component of net financial borrowing. In addition to the Korean presentation, this appendix gives an alternative capital account presentation which allocates Korean bank loans and credits between short- and long-term capital flows. It also separates out usage of Fund credit from net financial borrowing. The appendix tables are presented in groups as follows:

Basic Indicators A1 .1 Social Indicators Al.2 GNP (Expenditure)

331

KoredData Appendix

Al.3 Al.4 Al.5

GNP (Industrial Origin) Domestic Capital Formation Population and Employment

External Debt A2.1 External Debt by Borrower and Type A2.2 Usage of Debt A2.3 Long-Term Debt Service A2.4 Allocation of Private Debt A2.5 Allocation of Government Debt A2.6 Foreign Aid Balance of Payments A3.1 Korean Summary Presentation A3.2 Breakdown of Net Financial Borrowing A3.3 Korean Capital Account, Alternative Presentation A3.4 Exports by Type of Good A3.5 Imports by Type of Good A3.6 Invisible Trade Balance A3.7 Long-Term Capital Flows A3.8 Short-Term Capital Flows A3.9 Foreign Trade Indexes and Terms of Trade Money A4.1 A4.2 A4.3 A4.4 A4.5

and Prices Monetary Aggregates Financial Survey Deposits in Banks and Nonbank Financial Institutions Interest Rates Price Trends

Fiscal Indicators A5.1 Consolidated Government Revenue A5.2 Consolidated Government Expenditure and Net kndingFunctional Classification A5.3 Consolidated Government Expenditure and Net Lending-Economic Classification

Table Al.1

Social Indicators

Year

Per Capita GNP (current $)

Tax Burden Ratio to GNP

Hours of WorWWeek"

1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

100 I03 I05 125 142 I69 210 252 288 318 395 540 590 797 I.008 1.392 1,640 1,589 1,719 1,773 1,914 2,044 2,047 2,296

8.6% 7. I 8.6 10.7 12.0 13.9 14.6 14.6 14.7 12.7 12.5 14.3 15.5 16.9 17.0 17.3 17.9 18.4 18.8 19.5 20.0 19.2 19.0 18.8

50.3 56.0 57.0 57.4 53.8 57.6 56.3 52.3a 51.9 51.6 52.3 49.9 50.5 52.5 52.9 53.0 52.0 53.1 53.7 53.8 54.4 54.3 53.8 54.7

Union Membership Rate 20.0% 23.0 22.2 22.4 22.0 20.9 21.1 19.8 19.6 20.1 20.2 21.7 22.8 23.0 24.1 23.8 23.4 20.0 19.5 19.0 18.0 16.8 15.7

-

Primary School Entrance Rate

Housing Supply Rate

-

82.2% 81.8 80.7 80.7 79.6 78.7 78. I 78.2 77.8 78.2 77.4 77.3 74.4 77.4 77.2 76.8 76.5 71.2 73.7 73.1 71.8 71.6 69.7 69.7

-

95.1% 94.5 96.7 96.3 96.7 97.0 97.6 97.5 98.1 97.5 97.3 97.0 97.6 98. I 98.9 97.9 98.4 98.4 98.3 98.3 98.6 98.4

Average Nutrient Intake Per PersodDay (cal.)

Persons Served by Each Medical Personnel

-

2,981 2,742 2,609 2,059 1.986 1,911 1,856 1,773 1,681 1,639 1,702 1,879 1,801 1,732 1,675 1,612 1,552 1,490 1,441 1,386 1.336 1,282 1,230 1.166

-

2.105 2,150 2,065 1,904 2,059 2,054 1,992 1,926 2,134 1,833 2,098 2,052 2,042 1,991 2,012 1,901 1,936 -

Water Supply Service Rate

Rate of Roads Paved

18.4% 18.6 20.9 22.0 24.7 26.4 29.1 32.4 35.3 36.6 38.0 40.6 42.4 50.0 51.0 51.5 52.8 54.6 57.0 59.4 61.9 64.7 60.2

14.7% 16.3 17.7 16.5 17.6 18.1 20.4 23.7 28.3 32.3 34.6 37.1 44.0 45.5 51.3 60.4 60.7 67.4 55.3 60.5 67.0 68.8 73.7 77.1

Source: BOK, Nationul Accounts; EPB, Social Indicators in Korea and Major Sfatistics of Korean Economy: ILO, Yearbook of Labour Statistics, various issues.

"In manufacturing. bNew series.

-

Expenditure on GNP (in billions of won at 1980 constant prices)

Table A1.2

Year

GNP

Consumption Expenditure

Gross Fixed Capital Formation

Increase in Stocks

Exports"

Importsb

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

17,013.0 18,563.9 19,546.8 22,278.4 24,177.0 25,815.7 29,285.5 32,407.9 35,981.1 38,502.7 36,672.3 39,088.7 41.21 1.6 46,109.1 50,003.0 52,705.4 59,289.8

14,801.9 16,185.8 17,013.9 18,325.8 19,923.3 21,213.7 22,863.1 24,424.3 26,890.3 28,968.2 29.054.1 30.1 10.2 3 1,367.1 33,626.7 35,373.3 37,191.6 39,689.5

3,771.8 3.881.1 3,912.6 4,853.9 5,564.4 5,985.7 7,139.4 9,147.6 12,174.6 13,218.3 11,835.7 11,359.5 12,820.1 15,016.9 16,617.5 17,355.7 19,950.4

414.8 637.0 320.8 508.3 1,463.3 832.5 693.3 730.5 646.2 1,645.7 -46.8 1,176.4 - 266.9 -271.9 874.6 413.4 -311.1

2,006.4 2,430.2 3,304.8 5,056.5 5,017.5 5,972.4 8,448.6 10,413.3 11,719.7 11,586.6 12,765.4 14,684.5 15,637.9 18,054.0 19,854.7 20,279.5 25,682.3

4,002.8 4,799.5 4,854.9 6,545.5 7,680.8 7,837.3 9,721.4 11,708.2 14,896.0 16,617.7 15,729.3 16,432.1 16,763.3 18,593.4 20,464.7 20,124.1 23,859.0

Source: BOK, National Accounts, 1987.

Note: Data are based on the new SNA method

"Exports of goods and services. bIrnports of goods and services.

Statistical Discrepancy

Net Factor Income

- 106.8

127.7 21.7 -26.6 - 101.8 - 138.4 -332.1 -318.4 -425.6 -429.9 - 596.9 - 1,242.6 - 1,634.6 - 1,824.1 - 1.634.6 - 1,869.5 - 1,969.0 - 1,876.6

207.6 - 123.8 181.2 27.7 - 19.2 180.9 - 174.0 - 123.8 298.5 35.8 175.2 240.8 - 88.6 -382.9 -441.7 14.3

-

Table A1.3

Composition of GDP by Economic Activity (in percentages at current prices) Industries

Year

Total

Agriculture

Mining and Manufacturing

Manufacturing

Electricityb

Others

Government Services'

Import Duties

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

88.3 88.5 88.9 90.2 90.4 89.4 87.8 87.7 87.9 88.2 81.7 87.5 86.9 86.6 87.2 87.2 81.2

26.0 26.6 26.1 24.5 24.3 24.5 23.2 22.0 20.2 18.8 14.6 15.8 14.6 13.6 13.3 13.5 12.3

22.4 22.5 23.5 26.2 27.2 27.6 28.8 28.9 29.4 29.8 30.9 30.7 29.8 29.8 30.5 29.6 31.4

21.2 21.3 22.4 25.1 26.0 26.2 27.6 27.5 28.0 28.1 29.6 29.2 28.3 28.4 29.1 28.2 30.0

6.7 6.1 5.8 5.8 5.4 6.0 5.9 7.1 9.0 10.2 10.4 9.4 10.2 10.8 11.1 11.3 11.0

33.2 33.3 33.5 33.6 33.5 31.3 29.9 29.7 29.3 29.3 31.8 31.5 32.3 32.4 32.3 32.8 32.5

9.5 9.7 9.5 8.1 7.7 8.3 9.2 9.2 8.7 8.6 9.3 9.4 10.2 10.0 9.6 9.9 9.6

2.1 1.8 1.6 1.7 1.9 2.3 3.0 3.1 3.5 3.2 3.0 3.1 2.9 3.4 3.2 2.9 3.2

Source: BOK, Notional Accounts, 1987. Note: Data are based on the new SNA method.

%eludes forestry and fishing. bIncludes gas, water, and construction. 'Includes private, nonprofit services to household

Table A1.4

Composition of Gross Capital Formation by Sector (in billions of won at 1980 constant prices)

Gross

Industries

Year

Fixed lnvestment

Total

Agriculture

Mining

Manufacturing

Elechicity"

Others

Government Services

Increase in Stocks

Total

1970 1971 1912 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

3,771.8 3,881.1 3,912.6 4,853.9 5,564.4 5,985.7 7,139.4 9,147.6 12,174.6 13,218.3 11,835.7 11,359.5 12,820.1 15,016.9 16,617.5 17,355.7

2,995.9 3,100.6 3,220.2 4,219.5 4,976.6 5,191.8 6,227.1 7,969.9 10,676.6 11,502.4 10,175.3 9,691.6 10,895.0 12,773.6 14,098.5 14,705.6

669.7 608.5 650.9 769.9 1,094.8 738.7 827.8 1,031.5 1,186.2 1,202.1 955.0 863.6 863.2 1,085.4 1,197.6 1,120.8

13.6 23.8 22.3 28.0 36.8 33.2 54.6 72.8 89.5 93.2 80.8 93.8 105.8 107.8 121.1 132.6

554.5 708.9 537.5 950.7 1,063.3 1,408.6 1,815.1 2,595.0 3,460.4 3,683.2 2,650.7 2,454.0 2,694.3 2,754.6 3,675.5 4,494.7

219.9 326.2 379.7 365.0 334.2 444.6 513.6 912.9 1,137.4 1.479.1 1,433.4 1,573.8 1,690.3 1,935.5 1,844.9 1,855.4

1,538.2 1,433.2 1,629.8 2,096.9 2,447.5 2,566.7 3,016.0 3,357.7 4,803.1 5,044.8 5,055.4 4,706.4 5,541.4 6,890.3 7,259.4 7,102.1

775.9 780.5 692.4 634.2 537.3 793.9 912.3 1,177.7 1,498.0 1,715.9 1,660.4 1,667.9 1,925.1 2,243.3 2,519.0 2,650.1

414.8 637.0 320.8 508.3 1,463.3 832.5 693.3 730.5 646.2 1,645.7 -46.8 1,176.4 -266.9 -271.9 874.6 413.4

4,186.6 4,518.1 4,233.4 5,362.2 7,027.7 6,818.2 7,832.7 9,878.1 12,820.8 14,864.0 11,788.9 12,535.9 12,553.2 14,745.0 17,492.1 17,769.1

Source: BOK, National Accounts, 1987. Note:

Data are based on the new SNA method.

Sncludes gas, water, and construction.

336

Susan M. Collins and Won-Am Park

Table A1.5

Total Population

Year ~

1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

Population and Employment (in thousands of persons)

~

~~

~~~~

Population Aged 14 and Over

Economically Active Population

Total Employed

Unemployment Ratio

15,085 15,502 15,937 16,367 16,764 17,166 17,639 18,253 18,984 19,724 20,438 21,148 21,833 22,549 23,336 24,024 24,678 25,335 25,969 26,531 27,130 27,793 28,489 28,225'

8,343 8,449 8,859 9,071 9,295 9,647 9,888 10,199 10,542 1 1,058 11,600 12,080 12,340 13,061 13,440 13,932 14,206 14,454 14,710 15,080 15,128 14,984 15,554 16,116

7,662 7,799 8,206 8,423 8,717 9,155 9,414 9,745 10,066 10,559 11,139 11.586 11.830 12,556 12,929 13,490 13,664 13,706 14,048 14,424 14,515 14,417 14,935 15,505

8.2% 7.7 7.4 7. I 6.2 5. I 4.8 4.5 4.5 4.5 4.0 4. I 4.1 3.9 3.8 3.2 3.8 5.2 4.5 4.4 4.1 3.8 4.0 3.8

~~

27,262 27.984 28,705 29,436 30,131 30,838 31,544 32,241 32,883 33,505 34,103 34,692 35,281 35,849 36,412 36,%9 37,534 38,124 38,723 39,326 39,929 40,513 4 I ,056 41,569

Source: BOK, Economic Sfarisrics Yearbook. various issues. '15 years old and over.

Table A2.1

Long term (%)

Public loan private loan Bank loan Bonds outstandinga IMF facilities Foreign banks “A“’ Medium term

(a) Trade credit breign banks “ A ” b Cash loan and others Short fennC

6) private Banks (Foreign banks “A’)’ Total Foreign Debt

(continued)

External Debt (i millions of U.S. dollars) 61

62

63

4 (4.8)

10 (11.2)

52 (33. I)

4

-

79 (95.2) 79 -

10

-

79 (88.8) 79 -

33 19

-

83 (52.9) 83

-

22 (14.0) 22 -

64

65

66

67

121 (58.7)

288 (73.4)

486 (75.3)

892 (74.4)

113 162

191 282

30 1 506 40

52

60

68

-

-

-

-

-

-

9

13

13

45

-

-

82 (39.8) 82

-

97 (24.7) 97

(-)

(-)

157

206

( 6 )

392

75

1,376 (76.4)

1,702 (75.8)

2,332 (79.8)

2,381 (78.9)

3,417 (80.3)

4,541 (76.5)

5.732 (67.9)

588 1,006 75

906 1,210 153

1,330 1,338 155

1,713 1,578 I26

2,495 2,491

-

-

-

63

8

-

2,063 1,928 399 19 132

446 815 70 -

-

45

33

-

-

465

19 262

-

-

-

116 (3.3)

139 (3.3)

153 (2.6)

302 (3.

93

111

114

136

150

285

-

-

-

3

3

17

701 (16.4)

1,239 (20.9)

2,409 (28.5)

-

-

218

230

170

-

-

-

-

-

-

-

-

-

-

89 (7.4)

194 (10.8)

373 (16.6)

479 (16.4)

640

72 17

114 80 (8)

193 180 (16)

246 233 (29)

397 243 (40)

489 212 (89)

413 826 (103)

1,158

1.800

2,245

2,922

3,589

4,257

5,933

8,443

7 6 6 (1.9) (10.3) 6

74

111 (3.8)

3

2

73

170 (7

-

5

72

230 (12.8)

-

3

71

218 (18.2)

-

-

70

93 (14.4)

-

(1.5)

-

69

0 6 (-)

(-)

645

1,199

2

(17.8)

1,251 (242)

’zgble A2.1

Long term (%)

Public loan F’rivate loan Bank loan Bonds outstanding’ IMF facilities Foreign banks “A”b

(continued) 76

77

78

79

80

81

82

83

84

85

86

7,067 (67.2)

8,584 (67.9)

11,189 (75.5)

14.270 (70.3)

17,040 (62.7)

21,145 (65.2)

23,685 (63.9)

26,353 (65.3)

29,612 (68.8)

33,859 (72.4)

33,568 (75.4)

3,134 2,998 483 93 359

3,666 3,812 602 163 341

10,292 6,155 6,867 641 1,354 1,044

11,056 5,905 9,314 1,491 1,570 1,276

11,376 5,742 10,157 3,576 1,508 1,500

11,250 5,836 9,199 4.060 1,545 1,678

-

-

4,340 5,045 620 25 1 263 670

5,251 5,603 1,980 242 138 1,056

6,505 6,177 2,309 288 713 1.048

7,862 6,440 4,174 388 1,246 1,035

Medium term (%)

408 (3.9)

350 (2.8)

483 (3.3)

561 (2.8)

754 (2.8)

Trade credit Foreign banks “A”b Cash loan and others

386

335

480

522

22

15

3

39

576 137 41

564 354 143

3,045 (28.9)

3,715 (29.3)

3,151 (21.2)

5,456 (26.9)

9,376 (34.5)

1,499 1,546 (364)

1,828 1,887 (792)

1,041

2,110 (558)

2,251 3,205 (805)

10,520

12,649

14,823

20,287

Short termc (%)

Rivate

Banks (Foreign banks “A’)’

Total Foreign Debt

-

-

-

-

1,061

1,686 (3.8)

1,910 (4.7)

2,016 (4.7)

2,171 (4.6)

462 34 1 168

414 837 659

392 801 823

328 863 980

10,227 (31.5)

12,427 (33.5)

12,115 (30.0)

1 1,425

4,264 5,112 (1,685)

4,193 6,034 (1,633)

4,159 8,268 (1,993)

4,997 7,118 (1,474)

4,126 7,299 (1,798)

3,640 7,092 (1,737)

3,289 5,967 (2,105)

27,170

32,433

37,083

40,378

43,053

46,762

44,510

(3.3)

Note: dashes = not available.

Yncludes FRCD. bForeign currency funds borrowed by branch offices of foreign banks from their headquarters or other branches. ‘Includes trade credit and bank refinance.

9,342 6,310 5,322 401 1,259 1,05I 97 1 (2.6)

(26.5)

10,732 (23.0)

252 1,213 22 I 9,256 (20.8)

Usage of Debt (in millions of U.S. dollars)

Table A2.2

External Assets

Net Debt

Year

Change in Gross External Debt (1)

1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

186 253 554 601 445 677 667 668 1,676 2,510 2,077 2,129 2,174 5,464 6,883 5,263 4,650 3,295 2,675 3,709

Current Account Deficit (2)

104 192

440 549 622 848 371 309 2,023 1,887 314 - 12 1,085 4,151 5,321 4,646 2,650 1.606 1,373 887

breign Direct Investment (3)

5 13 15 7 25 37 61 158 163 69 106

102 100

127 97 105 101 101

171 250

Errors and Omissions (4)

Change in Foreign Reserves (5)

-4

99 111 34 162 57 - 42 172 355 - 39 495 1,410 1,346 631 77 1 863 320 93 - 74 740 99

- 23

20 8 5

- 13 - 30

- 19 - 28 122 24 1 32 312 329 370 41 1 1.2% 942 894 880

Exports on Credit (6)

Other (7)

- 18

- 155 7

- 14 93 124 70 244 - 57 350 677 - 181 940

-40 45 - 125 - 264 - 153 93 - 135 - 288 - 70 20 568 - 78 16 - 12 - 162 160

43

- 322 653

Memo: (2 + 4 + 5) + 1 1.070 1.107 0.892 1.196 1.537 1.171 0.769 0.966 1.167 0.998 0.946 0.642 0.846 0.914 0.934 0.933 0.665 0.670 0.722 0.519

340

Susan M. Collins and Won-Am Park

Table A2.3

Korean Long-Term Debt Service (in millions of U.S. $)

A. Interest Payment, Long-Tern Year

Total

Public

Private

Bank

Bond

IMF

1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

154.0 194.3 234.9 346.7 410.2 568.5 755.8 991.8 1,419.0 1.739.4 2,225.8 1,967.1 2,312.9 2,301.7 2,6 16.8

54.6 62.8 56.4 92.2 131.9 179.2 240.3 280.5 375. I 441.0 614.9 582.8 677.9 638.2 782.9

86.4 114.3 148.4 178.7

13.0 17.2 29.5 64.6 56.0 75.7 107.0 173.0 334.0 457.0

0.0 0.0 0.3 1.6 2.4 13.6 4.0 18.0 19.0 24.0 35.0 24.0 58.0 138.0 277.0

0.0 0.0 0.3 9.6 10.3 24.7 25.0 15.0 17.0 36.0 135.0 120.0 133.0 141.0 125.0

2W.6 275.3 379.5 505.3 673.9 781.4 800.9 632.3 665.0 561.5 579.9

640.0 608.0 779.0 823.0 852.0

B. Amortization

C . Service

Year

Total

Public

Private

Bank

1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

211.3 329.5 313.5 318.4 473.6 671.5 1,160.8 1,458.8 1,274.8 1,619.1 1,782.6 2,074.2 2,706.2 3,029.8 5,102.0

14.0 21.0 34.4 45.6 73.9 105.3 144.0 178.2 262.4 333.2 388.0 493.5 659.9 704.3 1,006.5

171.3 220.7 252.7 238.3 332.2 427.3 679.8 1,019.6 829.4 983.9 1,043.6 1,128.7 1,108.3 1,127.5 1,525.5

26.0 79.7 26.4 34.5 44.5 114.4 224.0 101.0 144.0 210.0 282.0 386.0 587.0 791.0 1,983.0

Bond

IMF

0.0

0.0 8.1

0.0 0.0 0.0

0.0 0.0 0.0 35.0 6.0 44.0 30.0 23.0 83.0 40.0 322.0

0.0 0.0 23.0 24.5 113.0 125.0 33.0 48.0 39.0 43.0 268.0 367.0 265.0

Total 365.3 523.8 548.4 665.1 883.6 1,240.0 1,916.6 2,450.6 2,693.8 3,358.5 4,008.4 4,041.3 5,019.1 5,331.5 7,718.8

341

KoredData Appendix Allocation of Private Debt, 1959-84 (in millions of U.S. dollars)

Table A2.4

~

Industry Agriculture, mining, and forestry Manufacturing Textiles Chemicals Nonmetallic minerals Basic metals Transport equipment Electrical machinery Other manufacturing Infrastructure Electricity Transport and storage Other services Total

~~

~

Net*

Percentage of Total

186 8,243 1,215 2,078 670 2,480 788 309 461 6,230 2,723 2,049 1,024 14,659

1.3 56.2 8.3 14.2 4.5 16.9 5.4 2.1 3.1 42.5 18.6 13.8 6.6 100.0

Source: Ministry of Finance.

*Excluding repayment.

Table A2.5

Allocation of Government Lkbt, 1959-84 (in millions of U.S. dollars)

Industry

Gross

Percentage of Total

Agriculture, forestry, and fishery Mining Manufacturing Metals Machinery and equipment Electricity, gas, and water Construction Transportation, storage, and communication Finance, insurance, and real estate Public and private services (national defense and public administration) Total

2,086 44 312 111 127 3,630 2,184

15.0 0.3 2.7 0.8 0.9 26.1 15.7

1,087 32 239 71 91 3,285 1,780

16.3 0.3 2.2 0.6 0.8 29.7 16.1

2,066 1,992 1,510 1 ,Mx)

44.9 14.3 10.8 7.2

1,468 1,100 1,345 904

13.3 9.9 12.2 8.2

13,884

100.0

1 1,056

100.0

Source: Ministry of Finance.

Net

Percentage of Total

342

Susan M. Collins and Won-Am Park

Table A2.6

Summary of Foreign Economic Aid Received, 1948-83 (in millions of

us. Year

Total

1948 1949 1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983

179,593 116,509 58,706 106,542 161,327 194,170 153,925 236,707 326,705 382,893 321,272 222,204 245,393 199,245 232,310 216,446 149,331 131,441 103,261 97,018 105,856 107,264 82,636 51,217 5,089 2,146 982 1,155 1,740 948 169 224 361 236 61 30

GARIDA

ECAandSEC

U.S. dollars)

CRIK PL48Oa

AID

-

-

-

32,955 45,522 47,896 11,436 19,913 44,926 67,308 %,787 60,985 59,537 37,951 44,378 55,927 74,830 61,703 33,651

-

-

-

SUN

SKD

UNKRA

-

-

9,376

5,571 82,437 205.8 15 271,049 323,268 265,629 208,297 225,236 154,319 165,002 119,659 88,346 71,904 65,310 52,640 49,929 32,434 20,933 17,566 5,089 2,146 982 1,155 1,740 948 169 224 36 1 236 61 30

Source: BOK, Economic Statistics Yearbook, 1984 "A portion of proceeds used by the U.S. Government from sales of surplus agriculture commodities imported under the US. Public Law 480 cannot be regarded as foreign aid received, but for convenience it is included here to show the total imports under the same law.

343

KoredData Appendix Korean Summary Presentation (in millions of U.S. dollars)

Table A3.1

Capital Account Current Account

Year

1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

-440.3 -548.6 -622.5 -847.5 -371.2 -308.8 -2,022.7 - 1,886.9 -313.6 12.3 - 1,085.2 -4,151.1 -5,320.7 -4,646.0 -2,649.6

- 1,606.0 - 1,372.6 - 887.4 4,617.0

Total

Long-Term

Short-Term

Errors and Omissions

447.0 650.0 623.4 646.6 504.7 750.3 901.0 1,857.8 1,727.7 1,334.1 995.3 3,506.5 3,801.0 2,759.6 1,233.9 2,163.9 1,309.5 513.3 - 2,374.0

433.8 593.5 501.0 512.0 521.0 666.3 946.4 1,178.3 1,371.2 1,312.7 2,166.3 2,662.9 1,856.5 2,841.9 1,230.3 1,270.4 2,067.4 1,100.8 -1,981.9

13.2 56.5 122.4 134.6 - 16.3 84.0 - 45.4 679.5 356.5 21.4 -1,171.0 843.6 1,944.5 -82.3 3.6 893.5 - 757.9 -587.5 - 392.1

-20.2 -7.7 -5.1 13.1 30.1 18.8 27.9 - 121.5 -240.5 -31.7 -312.0 -328.7 -369.9 -410.6 -.1,295.5 -942.3 -894.4 -880.4 -543.5

Overall Balance

Change in Reserves

Net Financial Borrowing

-13.5 93.7 -4.2 -187.8 163.6 460.3 -1,093.8 - 150.6 1,173.6 1,314.7 -401.9 -973.3 - 1,889.6 -2,297.0 -2,711.2 -384.4 -957.5 - 1,254.5 1,699.5

59.1 143.1 34.0 -49.0 159.0 340.4 15.2 492.2 1,419.1 1,345.7 630.7 771.0 863.3 319.6 92.7 -74.1 739.9 99.0 206.6

72.6 49.4 38.2 138.8 - 10.3 - 119.9 1,109.0 642.8 245.5 31.0 1,032.6 1,744.3 2,752.9 2,616.6 2,803.0 310.3 1,697.4 1,353.5 - 1,492.9

Source: BOK, Economic Sratistics Yearbook, various issues.

Breakdown of Net Financial Borrowing, 1972-86 (in millions of U.S. dollars)

Table A3.2

Banks Loans and Credits Year

1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

Net Financial Borrowing

Use of Fund Credit

- 10.3

-

- 119.9

-

1,109.0 642.8 245.5 31.0 1,032.6 1,744.3 2,752.9 2,616.6 2,803.0 310.3 1,697.4 1,353.5 - 1,492.9

132.3 129.9 98.1 -25.7 -98.9 - 125.3 560.9

630.9 78.4 161.4 313.7 - 229.5 - 125.2

Total

-4.3 -93.0 826.2 398.2 133.9 107.4 1,274.5 1,850.1 2,144.9 2,010.5 2,776.6 153.9 1,511.9 1,397.1 - 1,034.7

Long-Term

-5.4

- 17.9 84.2 -32.8 -38.1 142.4 88.9 408.3 127.5 804.2 578.5 520.6 950.2 1,212.3 -569.0

Source: Ministry of Finance, and IMF, Balance of Payment Statistics.

Short-Term

Other

1.1 -75.1 742.0 431.0 172.0 - 35.0 1,185.6 1,441.9 2,017.3 1,206.3 2,198.1 -366.7 561.7 184.8 -465.7

-6.0 -26.9 150.5 114.7 13.4 - 50.7 - 143.0 19.5 47.1 -24.8 -51.9 -5.1 - 128.1 185.9 -333.0

344

Susan M. Collins and Won-Am Park Korean Capital Account, Alternative Presentation (in millions of U.S. dollars)

Table A3.3

Capital Account Year 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

Current Account -371.2 - 308.8

-2,022.7 1,886.9 -313.6 12.3 - 1,085.2 -4,151.1 -5,320.7 -4,646.0 -2,649.6 - 1,606.0 - 1,372.6 - 887.4 4,617.0

-

Long-Term 515.6 648.4 1,030.6 1,145.5 1,333.1 1,455.1 2,255.2 3,071.2 1,984.0 3,646.2 1,808.8 1,791.0 3,017.6 2,313.1 -2,550.9

Short-Tern - 15.2

8.9 696.6 1,110.5 528.5 - 13.6 14.6 2,285.5 3,961.8 1,124.0 2,201.7 526.8 - 196.2 -402.7 -857.8

Discrepancy

Use of Fund Credit

-6.0 -26.9 150.5 114.7 13.4 - 50.7 - 143.0 19.5 47.1 -24.8 -51.9 -5.1 - 128.1 185.9 - 333.0

~

Source Computed from tables A3 1 and A3 2 Note The “discrepancy” in the capital account is net financial borrowing

132.3 129.9 98.1 -25.7 -98.9 - 125.3 560.9 630.9 78.4 161.4 313.7 -229.5 - 125.2

Change in Reserves

Errors and Omissions

159.0 340.4 15.2 492.2 1,419.1 1.345.7 630.7 771.0 863.3 319.6 92.7 -74.1 739.9 99.0 206.6

30.1 18.8 27.9 - 121.5 - 240.5 -31.7 -312.0 - 328.7 - 369.9 -410.6 - 1,295.5 -942.3 - 894.4 - 880.4 - 543.5

Table A3.4

Exports by Type of Good, 1966-86 (in millions of U.S. dollars)

Marine

Agriculture

Mining

Manufacturing

Year

Total

Amount

% of Total

Amount

% of Total

Amount

% of Total

Amount

% of Total

1966 1%7 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

255.8 358.6 500.4 702.8 1,003.8 1,352.0 1,807.0 3,256.9 4,712.9 5,427.4 8.1 14.9 10,046.5 12.71 1.1 15,055.5 17,504.9 20,992.6 21,616.1 24,222.5 29,244.9 30,283.1 34,714.5

24.3 17.0 21.6 29.7 30.1 38.0 53.0 101.2 131.1 176.8 221.7 414.9 506.4 518.9 458.9 430.9 395.3 424.7 489.0 425.6 500.8

9.5 4.7 4.3 4.2 3.0 2.8 2.9 3. I 2.8 3.3 2.7 4. I 4.0 3.4 2.6 2.1 1.8 1.8 1.7 1.4 1.4

37.5 52.8 50.9 66.1 82.3 104.0 137.5 233.5 250.5 387.6 518.6 747.0 690.3 855.5 759.5 931.7 861.2 826.8 877.5 890.8 1,283.3

14.7 14.7 10.2 9.4 8.2 7.7 7.6 7.2 5.3 7.1 6.4 7.4 5.4 5.7 4.3 4.4 4.0 3.4 3.0 2.9 3.9

34.2 37.6 41.0 52.0 52.1 41.2 32.2 49.4 78.6 71.7 91.4 89.9 93.9 110.6 135.6 122.2 106.2 106.3 91.6 87.1 105.1

13.4 10.5 8.2 1.4 5.2 3.5 1.8 1.5 1.7 1.3 1.1 0.9 0.7 0.7 0.8 0.6 0.5 0.4 0.3 0.3 0.3

159.7 251.2 386.9 555.1 839.4 1,162.9 1,584.3 2,872.8 4,252.7 4,791.2 7,283.2 8,794.7 11,420.5 13,570.5 16,150.8 19,507.8 20,253.5 22,864.7 27,186.8 28,879.6 32,826.1

62.4 70.0 17.3 79.0 83.6 86.0 87.7 88.2 90.2 88.3 89.8 87.5 89.8 90.1 92.3 92.9 93.7 94.4 95.0 95.4 94.6

Source:

EPB, Major Statistics of the Korean Economy,

346

Susan M. Collins and Won-Am Park Imports by Type of Good (in millions of U.S. dollars)

Table A3.5

Capital Goods

Raw Materials (Export)

Crude Oil

Year

Total

Amount

% of Total

Amount

% of Total

Amount

% of Total

1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

716.4 996.2 1,462.9 1,823.6 1,984.0 2,394.3 2,522.0 4,240.3 6,851.8 7,274.4 8,773.6 10,810.5 14,971.9 20,338.6 22,292.0 26,131.4 24,250.8 26,192.2 30,631.4 31.135.7 31,583.9

171.7 310.2 533.2 593.2 589.5 685.4 762.0 1,158.8 1,848.6 1,909.2 2,427.4 3,008.1 5.080.3 6,314.0 5,125.0 6,158.2 6,232.7 7,814.7 10,106.3 1 1,081.3 11,326.3

24.0 31.1 36.4 32.5 29.7 28.6 30.2 27.3 27.0 26.2 27.7 27.8 33.9 31.0 23.0 23.6 25.7 29.8 33.0 35.6 35.9

-

-

-

-

-

-

-

-

-

-

-

-

-

-

1,452.0 2,144.0 2,427.0 2,948.0 3,444.0 3,799.0 4,587.3 4,651.7 4,801.7 6,201.1 6,823.5 8,219.9

20.0 24.4 22.5 19.7 16.9 17.0 17.6 19.2 18.3 20.2 21.9 26.0

40.6 59.4 72.8 107.6 125.0 174.0 206.0 277.0 966.0 1,271.2 1,607.0 1,926.0 2,187.0 3,100.0 5,633.0 6,375.7 6,097.3 5,576.6 5,770.6 5,588.7 3,373.3

5.7 6.0 5.0 5.9 6.3 7.3 8.2 6.5 14.1 17.5 18.3 17.8 14.6 15.2 25.3 24.4 25.1 21.3 18.8 17.9 10.7

Source: Korean Office of Customs Administration.

InvisiblesNet Wyments Receipts by Type, 1966-86 (in millions of U.S. dollars)

Table A3.6

Year

Total

1966 1967 1968 1969 1970 1971 1972 1973 1974 I975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

193.7 285.8 296.7 337.6 270.0 192.1 230.6 453.2 212.4 -46.4 348.8 743.9 1,057.8 727.2 -410.9 - 1,233.1 -505.0 -366.3 -829.8 - 1,788.2 -989.2

Foreign Travel

Transport

Insurance

Investment Income

Government Transactions

Miscellaneous

Donations

13.0 7.9 6.4 5.3 6.3 16.4 62.2 247.0 125.7 109.9 228.8 267.3 200.1 - 79.3 19.7 8.6 129.9 40.8 97.1 178.3 934.5

- 14.5 -23.1 -31.4 -22.9 -5.7 - 19.9 -2.7 7.0 - 18.0 -52.0 - 151.8 - 240.6 -278.9 -214.2 - 390.4 -413.3 -105.5 -12.0 126.8 - 164.8 -334.8

-0.5 3.6 0.7 -0.9 0.9 -2.1 -6.1 - 5.4 - 14.3 -24.4 -23.9 -23.3 -35.1 -23.1 23.1 - 50.6 -51.8 -46.3 -46.3 - 101.0 -78.3

2.3 -0.7 -5.1 -3.7 -43.2 -89.9 - 129.9 - 165.2 -237.7 -396.7 -412.4 -519.3 -644.0 -I,lOI.0 - 1,982.5 - 2,841.4 -3,037.3 -2,691.7 -3.275.6 -3,183.6 -3,207.8

127.9 206.5 237.3 259.5 238.2 212.6 213.4 152.3 128.8 97.3 144.0 182.2 217.5 313.8 164.7 167.1 274.5 349.2 421.5 415.1 550.9

11.0 21.7 34.4 50.2 32.9 30.1 38.5 94.5 84.6 80.0 349.0 826.3 ,259.5 ,548.7 ,475.2 ,504.5 2,161.8 1,515.9 1,400.2 542.4 226.5

54.4 69.9 54.5 50.1 40.7 44.9 65.2 123.0 143.2 139.4 215.2 251.4 338.8 282.2 279.3 392.0 383.2 477.6 446.5 525.4 929.8

Source: BOK, Economic Statistics Yearbook.

Long-Term Capital Flows (in millions of U.S. dollars)

Table A3.7

Loans and Investment

Year

Total

Public Loans

I974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

732.6 1.059.8 1,302.2 1,495.5 2,052.8 1,627.3 1,895.5 1,793.4 1,448.9 1,05 1.6 957.4 1,371.9 703.6

316.7 482.1 710.7 608.0 817.7 1,085.6 1,518.3 1,704.5 1,877.3 1.493.9 1,424.4 1,023.8 880.1

-

Commercial Loans

Direct Investment

616.0 804.5 842.5 1,260.0 1,929.8 1,621.8 1,415.8 1,242.5 918.4 973.2 858.4 964.1 1,619.9

124.1 61.6 85.5 104.4 100.5 126.0 96.2 105.4 100.6 101.4 170.7 250.3 477.5

Portfolio Investment 19.0

-

14.5 70.0 41.7 42.8 45.7 103.6 43.9 195.9 410.5 988.0 403.0

Repayments - 338.4 - 284.1

-406.6 - 536.0

-825.1 -1,122.7 - 1,084.7 - 1,315.3 - 1,430.3 - 1,671.9 - 1,768.4 - 1,831.6 - 2,531.9

Withdrawals

Repayment of Portfolio Investment

-4.8 -4.3 -4.4 - 10.9 -11.8 -90.9 -90.2 -3.4 -31.6 -32.9 -60.5 - 16.8 -42.5

-35.3 -5.6 -43.9 -29.4 -8.0 -77.7 -5.9 - 102.5

-

Others

Year

Total

Long-Term Trade Credit

Export on Credit

1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

213.8 118.5 69.0 - 182.8 113.5 1,035.6 - 39.0 1,048.5 -218.6 218.8 1,110.0 -271.1 -- 2,685.5

12.5 134.4 - 12.7 -50.7 149.4 42.3 53.2 -11.9 - 102.0 -47.6 -22.2 -64.3 -75.3

155.2 -6.5 14.3 -92.7 - 123.6 -69.7 -243.5 57.0 -349.9 -677.4 181.0 -940.0 -405.7

Import Prepayments

-85.7 - 132.2

-48.0 -114.1 -28.1 -5.3 -5.9 -6.6 -32.2 -26.5 -56.1

Others 46. I -9.4 153.1 92.8 135.7 1,177.1 179.4 1,008.7 239.2 950.4 983.4 759.7 -2,215.4

Bank LongTerm Credit 84.2 -32.8 -38.1 142.4 88.9 408.3 127.5 804.2 578.5 520.6 950.2 1,212.3 -569.0

Total 1,030.6 1,145.5 1,333.1 1,455. I 2,255.2 3,071.2 1,984.0 3,646.1 1.808.8 1,791 .O 3,017.6 2,313.1 2,616.1

348

Susan M. Collins and Won-Am Park Short-Term Capital Flows (in millions of U.S. dollars)

Table A3.8

Net Short-Term

Year

Total

1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

21.3 111.8 387.5 179.8 31.4 25.6 454.3 972.9 540.2 -59.9 -689.6 1,740.9 2.997.0 960.3 1,764.4 -312.0 -690.0 -693.0 1.476.0

Source: Ministry of Finance

Trade Credit

Refinance

Other

13.2 42.8 78.4 86.8 -26.0 138.3 - 34.9 580.0 341.4 346.3 -655.9 971.6 1,625.8 17.5 -228.9 775.0

11.9 52.9 88.0 31.6 - 20.0 -94.8 479.7 180.0 - 130.9 - 3.3 521.5 811.2 742.2 I , 123.6 1,139.7 - 298.0 - 296.0 -541.0 - 1,837.0

-3.8 16. I 221.1 61.4 59.4 - 17.9 9.5 212.9 329.7 - 402.9 -555.2 -41.9 629.0 - 180.8 853.3 - 270.0 371.0 207.0 275.0

-

- 298.0 -

282.0

Foreign Bank “ A ’ Account

8.3 7.7 12.7 10.9 49. I 14.5 138.8 122.2 428.0 469.5 689.4 1.085.0 239.4 636.9 -519.0 324.0 -61.0 368.0

349

KoredData Appendix Foreign Trade Indexes and Terms of Trade (1980= 100)

Table A3.9

Quantum Index

Unit Value Index

Year

Export

Import

Export

Import

Terms of Trade

1963 1964 1965 I966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

1.5

2.1 2.9 3.7 4.6 6.7 9.4 12.0 15.5 23.3 36.5 40.0 49.0 66.5 79.2 90.6 89.7 100.0 117.5 125.I 145.5 168.2 180.9 204.6

9.1 6.1 7.0 11.0 15.3 22.6 28.5 19.9 36.0 37.4 47.0 49.1 50.4 62.4 75.2 98.4 110.0 100.0

31.9 32.9 34.2 37.3 39.0 40.2 38.2 39.9 39.4 39.9 50.4 63.8 59.2 66.2 72.4 80.1 95.8 100.0 103.2 99.7 95.9 99.2 95.5 96.9

27.5 29.4 29.9 29.2 29.5 29.2 28.8 29.8 29.7 30.2 40.2 62.5 64.3 63.0 64.4 68.0 83.1 100.0 105.4 97.6 93.0 94.2 90.2 84.5

116.0 111.9 114.4 127.7 132.2 137.7 132.6 133.8 132.7 132.1 125.4 102.1 92. I 105.1 112.4 117.8 115.3 100.0 97.9 102.2 103.1 105.3 105.9 114.7

111.1

111.4 126.2 145.9 154.8 167.6

Source: BOK, Economic Statistics Yearbook. various issues.

Susan M. Collins and Won-Am Park

350

Table A4.1

Monetary Aggregates (in billions of won)

Year

Reserve Money

lo Change

MI

%Change

M2

B Change

M3"

O/o Change

1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

80.2 110.9 156.2 216.0 299.7 288.2 427.5 624.1 775.0 1,077.0 1,437.7 2,071.6 2,802.0 3,468.O 3,243.9 2,801.6 3,825.3 4,095.2 4,248.4 4,319.0 5,054.7

0.4 0.4 0.4 0.4 -0.0 0.5 0.5 0.2 0.4 0.3 0.4 0.4 0.2 -0.1 -0.1 0.4 0.1 0.0 0.0 0.2

85.1 123.0 177.9 252.0 307.6 358.0 519.4 730.3 945.7 1,181.8 1,544.0 2,112.6 2,713.8 3,274.5 3,807.0 3,982.4 5,799.3 6,783.4 6,820.7 7,557.8 8,846.8

0.4 0.4 0.4 0.2 0.2 0.5 0.4 0.3 0.2 0.3 0.4 0.2 0.2 0.2 0.0 0.5 0.2 0.0 0.1 0.2

157.0 253.8 436.6 704.6 897.8 1,084.9 1,451.8 1,980.5 2,456.5 3,150.0 4,204.8 5.874.3 7,928.7 9,877.8 12,534.5 15,671.1 19,904.2 22.938. I 24,705.6 28,565.2 33,871.0

0.6 0.7 0.6 0.3 0.2 0.3 0.4 0.2 0.3 0.3 0.4 0.3 0.2 0.3 0.3 0.3 0.2 0. I 0.2 0.2

1,278.0 1,683.8 2,391.1 3.041.1 3.903.3 5.293.7 7,515.5 10,210.6 13,379.3 17,810.8 23,243.2 30,964.6 37.647.7 45,204.1 54,763.9 69,155.O

0.3 0.4 0.3 0.3 0.4 0.4 0.4 0.3 0.3 0.3 0.3 0.2 0.2 0.2 0.3

Source:

BOK, Economic Srafistics Yearbook.

"Series revised in 1977.

351

KoredData Appendix Financial Survey (in billions of won)

Table A 4 2

Domestic Credit

Year

Foreign Assets (Net)

Total

1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

57.3 97.7 291.1 -344.7 -446.1 - 18.3 541.1 173.2 -800.1 -2,363.7 -4,677.5 -7.490.8 -9,097.9 - 11.099.0 - 14,828.0 - 13,150.7

1,370.6 1,808.2 2,479.3 3,695.3 5,045.0 6,329.4 8,192.0 11,878.5 16,603.1 24,582.4 32,908.8 42,717.9 50,950.2 61,553.0 75,379.4 88,949.1

Year

M1

M2

Quasi Money

NBFI Deposits*

Interbank Transactions

Debentures Issued

1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

85.1 123.0 177.9 252.0 307.6 358.0 519.4 730.3 945.7 1,181.8 1,544.0 2,172.6 2,713.8 3,274.5 3,807.0 3,982.4 5,799.3 6,783.4 6,820.7 7,557.8 8,846.8

157.0 253.8 436.6 704.6 897.8 1,084.9 1,451.8 1,980.5 2,456.5 3,150.0 4,204.8 5,874.3 7,928.7 9,877.8 12,534.5 15,671.1 19,904.2 22,938.1 24,705.6 28,565.2 33,871.0

71.9 130.8 258.7 452.6 590.2 726.9 932.4 1,250.2 1,510.8 1,968.2 2,660.8 3,701.7 5,214.9 6,603.3 8,727.5 11,688.7 14,104.9 16,154.7 17,884.9 21,007.4 25,024.2

189.8 239.8 398.0 565.8 751.3 1,132.2 1,700.6 2,491.3 3,533.7 5,332.0 7,559.5 11,144.4 14,949.8 20,009.0 25,574.9 34,060.6

- 16.6 -32.3 - 37.1 -45.2 -61.9 - 166.5 - 273.4 -474.5 -423.4 - 600.4 -661.2 - 849.6 - 1,209.7 - 1,363.2 - 1,548.3 - 1,763.8

19.9 24.5 49.7 64.0 63.9 123.2 214.0 265.2 391.3 544.8 673.8 673.1 895.7 1,172.5 1,199.6 1,920.1

Government - 163.9

-97.1 -93.2 -78.6 100.2 -9.3 -215.0 -446.3 -908.6 -649.3 70.0 305.5 -692.1 -848.0 -919.7 - 1,026.6

Source: BOK, Money and Banking Statistics

"Series revised in 1977.

Government Agency

Private

Other Items (Net)

M3"

32.0 34.0 43.0 20.0 110.0 110.0 210.0 240.0 240.0 270.0 370.0 470.0 570.0 570.0 570.0 570.0

1,502.5 1,871.3 2,529.5 3,753.9 4,834.8 6,228.7 8,197.0 12.084.8 17,271.7 24,961.7 32,468.8 41,942.4 51,072.3 61,831.0 75,729.1 89,405.7

- 149.9 -222.1 -379.3 -309.5 -695.6 - 1,017.4 - 1.217.6 - 1,841.1 -2,423.7 -4,407.9 -4,899.4 -4,262.5 -4,204.6 -5249.9 -5,787.5 -6,643.4

1,278.0 1,683.8 2,391.1 3,041.1 3,903.3 5,293.7 7,515.5 10,210.6 13,379.3 17,810.8 23,243.2 30,964.6 37,647.7 45,204.1 54,763.9 69,155.0

Commercial Bills Sold

CD

352

Susan M. Collins and Won-Am Wrk

~

Deposits in Banks and Nonbanks

Table A4.3

Commercial and Specialized Banks ~

Year

Total

A. In billions of won 1966 1967 1968 1969 1970 1971 1,167.4 1972 1,563.8 1973 2,158.6 1974 2,694.5 1975 3,563.6 1976 4,892.7 1977 7.156.4 1978 10,256.4 1979 13,316.4 1980 17,753.9 1981 24,593.6 1982 32,333.7 1983 38,857.3 1984 47,934.1 1985 56,597.5 1986 69,986.3

B. Shares (Percentage of Total) 1966 1967 1968 1969 1970 1971 I .o 1972 I .o 1973 I .o 1974 1.0 1975 1.o 1976 1 .o 1977 I .o 1978 1 .o 1979 1.o 1980 1.o 1981 1.o 1982 1.o 1983 1.o 1984 1.o 1985 1.o 1986 1.o 'Series revised in 1977

Total

Demand

Time and Savings

120.9 205.9 373.1 619. I 789.7 977.6 1.324.0 1,760.6 2,128.7 2,812.3 3,760.5 5,455.8 7,765.1 9,782.7 12,421.9 17,034.1 21,189.3 23,907.5 27,925.1 31,022.6 35.925.7

50.8 77.0 117.6 167.6 213.4 268.9 412.4 539.4 656.9 868.6 1.147.0 1.867.7 2,633.1 3,251.3 3,844.9 5,534.3 7,529.6 8,234.9 10,618.4 10,776.4 11,334.6

70.1 128.9 255.5 451.5 576.3 708.7 911.6 1,221.2 1,471.8 1.943.7 2,613.5 3.588.1 5.132.0 6.531.4 8,577.0 11,499.8 13,659.7 15.672.6 17,306.7 20,246.2 24,591.1

0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.8 0.7 0.7 0.7 0.7 0.6 0.6

0.2 0.3 0.2 0.2 0.2 0.2 0.3 0.3 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2

0.6 0.6 0.6

0.5 0.5

0.5 0.5

0.5 0.5 0.5

0.5

0.5 0.5 0.4 0.4 0.4 0.4 0.3

NBFI"

189.8 239.8 398.0 565.8 751.3 1,132.2 1.700.6 2,491.3 3,533.7 5,332.0 7,559.5 11,144.4 14,949.8 20,009.0 25.574.9 34,060.6

0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.2 0.3 0.3 0.3 0.3 0.4 0.4

0.5 0.5

353

KoredData Appendix

Table A4.4

Year 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1912 1973 1974 1975 1976 1971 I978 1979 1980 1981 1982 1983 1984 1985 1986

Interest Rates (average annual rate, in percentages) One-Year Time Deposits

Discounts on Commercial Bills

Curb Market Ratea

10.0 12.4 15.0 15.0 15.0 18.7 30.0 30.0 27.6 24.0 22.8 22.2 15.6 12.6 14.8 15.0 15.5 15.8 16.4 18.6 22.9 18.8

13.9 13.9 13.9 13.9 14.0 16.5 24.0 24.0 24.5 25.2 24.3 23.0 17.7 15.5 15.5 15.5 16.1 16.0 16.9 18.5 22.9 19.3 12.2 10.0 10.0-10.6 10.0-11.5 10.0-11.5

-

10.9 8.0 9.4 10.0 10.0

Source: BOK, Economic Statistics Yearbook, various issues.

"Unofficial figures.

-

52.6 59.4 58.9 58.1 56.5 56.0 51.4 49.8 46.4 39.0 33.3 40.6 41.3 40.5 38. I 41.2 42.4 44.9 35.2 30.6 25.8 24.8 24.0 23.1

354

Susan M. Collins and Won-Am Park

Table A4.5

Price Trends WPI

CPI"

Change

GNP Deflator

Change

Year

Index

(%)

Index

(%)

1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

5.8 6.6 7.2 8.6 11.6 12.8 13.9 14.8 16.0 17.1 18.7 20.3 23.1 24.7 35.1 44.4 49.8 54.3 60.6 72.0 100.0 120.4 126.0 126.3 127.2 128.3 125.5

13.7 9.1 19.4 34.9 10.3 8.6 6.5 8.1 6.9 9.4 8.6 13.8 6.9 42.1 26.5 12.2 9.0 11.6 18.8 38.9 20.4 4.7 0.2 0.7 0.9 -2.2

6.1 6.6 7.0 8.4 10.9 12.3 13.9 15.3 17.0 19.1 22.2 25.2 28.1 29.0 36.1 45.2 52.1 57.4 65.7 77.7 100.0 121.3 130.1 134.5 137.6 141.0 144.2

8.2 6.1 20.0 29.8 12.8 13.0 10.1 11.1 12.4 16.2 13.5 11.5 3.2 24.5 25.2 15.3 10.2 14.5 18.3 28.7 21.3 7.3 3.4 2.3 2.5 2.3

Manufacturing

Nonmanufacturing

0.066 0.075 0.085 0.106 0.146 0.157 0.178 0.185 0.183 0.215 0.231 0.243 0.274 0.307 0.379 0.466 0.542 0.599 0.680 0.794

0.030 0.034

Source: BOK, Economic Statistics Yearbook, various issues.

aData prior to 1967 refer to consumer prices in Seoul.

1

.ooo

1.137 1.194 1.227 1.237 1.257 1.316

0.040

0.052 0.067 0.070 0.080 0.092 0.111 0.125 0.146 0.168 0.196 0.221 0.289 0.365 0.442 0.520 0.649 0.797 1.ooo 1.162 1.247 1.302 1.372 1.43 1 1.463

Total

Change (8)

0.032 0.037 0.043 0.056 0.073 0.078 0.089 0.102 0.119 0.137 0.158 0.180 0.209 0.237 0.307 0.386 0.466 0.539 0.657 0.769 1.ooo 1.154 1.230 1.279 1.328 1.375 1.414

13.8 18.2 29.6 30.0 6.3 14.2 14.2 17.6 14.6 15.7 13.4 16.4 13.4 29.5 25.8 20.5 15.8 21.9 21.1 25.6 15.4 6.6 3.9 3.8 3.6 2.8

Consolidated Government Revenue

Table A5.1

Year

Current

Total Taxes

Income

Goods and Services

(VAT)

International

Other

Nontax

Capital and Grant

1,536.4 2,312.0 2,938.2 4,084.2 5,375.9 6,736.8 8,533.7 9,874.8 11,417.3 12,510.4 13,737.0 15,232.3

1,406.2 2,116.6 2,668.9 3,702.1 4,837.4 5,896.9 7,364.4 8,529.9 10,207.4 11,077.3 12,105.1 13,481.9

343.7 603.1 735.9 1,042.1 1,381.7 1,504.0 1,958.5 2,357.0 2,609.1 2,866.8 3,474.6 3,676.2

718.0 994.9 1,309.7 1,751.6 2,353.5 3,092.6 3,814.0 4,395.2 5,220.7 5.629.3 5,933.6 6.595.6

198.2 261.4 449.4 838.9 1,088.7 1,471.2 1,804.8 2,094.4 2,559.3 2,704.3 2,901.2 3,138.1

201.9 343.9 475.0 774.9 912.8 1,013.6 1,189.8 1,316.4 1,798.4 1,921.I 1,950.2 2,380.0

142.6 174.7 148.3 133.5 189.4 286.7 402.1 461.3 573.2 660.1 746.7 830.1

130.2 195.4 269.3 382. I 538.5 839.9 1,169.3 1,344.9 1,209.9 1,433.1 1.63 I .9 1,750.4

27.2 14.6 20.2 23.5 69.5 96.4 71.1 108.4 120.2 92.9 185.0 226.3

Total

A. In Billions of Won 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985' 1986b

1,563.6 2,326.6 2,958.4 4,107.7 5,445.4 6,833.2 8,604.8 9,983.2 11,537.5 12,603.3 13,922.0 15,458.6

B. As a Percentage of Total 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986

100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00 100.00

98.26 99.37 99.32 99.43 98.72 98.59 99.17 98.91 98.96 99.26 98.67 98.54

89.93 90.97 90.21 90.13 88.83 86.30 85.59 85.44 88.47 87.89 86.95 87.21

Source: Ministry of Finance, Government Finance Statistics.

"Figures for 1985 and 1986 are original budget.

21.98 25.92 24.88 25.37 25.37 22.01 22.76 23.61 22.61 22.75 24.96 23.78

45.92 42.76 44.27 42.64 43.22 45.26 44.32 44.03 45.25 44.67 42.62 42.67

12.68 11.24 15.19 20.42 19.99 21.53 20.97 20.98 22.18 21.46 20.84 20.30

12.91 14.78 16.06 18.87 16.76 14.83 13.83 13.19 15.59 15.24 14.01 15.40

10.14 8.25 5.56 3.61 3.92 4.86 5.46 5.41 5.67 5.24 6.17 6.16

8.33 8.40 9.10 9.30 9.89 12.29 13.59 13.47 10.49 11.37 11.72 11.32

1.74 0.63 0.68 0.57 1.28 1.41 0.83 1.09 1.04 0.74 1.33 I .46

356

Susan M. Collins and Won-Am Park

Table A5.2

Year

Consolidated Central Government Expenditure and Net Lending-Functional Classification

Total

A. In BilliiJns of Won 1975 1,765.0 1976 2,518.9 1977 3,275.3 4,408.0 1978 1979 5,990.0 1980 7,682.0 10,189.8 1981 11,639.2 1982 1983 12,200.1 1984 13,444.6 1985 14,867.0 1986 16,986.3

B. As a Percentage of Total 1975 100.00 1976 100.00 100.00 1977 1978 100.00 1979 100.00 1980 100.OO 1981 100.00 1982 100.00 1983 100.00 1984 100.00 1985 100.00 1986 100.00

General Public Service

Defense

Social Service'

230.6 235.7 296.9 419.1 547.5 655.5 907.7 1,076.6 1,236.9 1,208.0 1,400.7 1,553.5

465.2 770.5 1,008.3 1,438.1 1,597.4 2,349.1 2,849.0 3,180.1 3,402.5 3,573.4 3,957.9 4,501.6

356.0 531.3 735.4 954.5 1,347.2 1,892.I 2,894. I 3,589. I 3,647.0 4,294.3 4,299.0 5,110.0

13.07 9.36 9.06 9.51 9. I4 8.53 8.91 9.25 10.14 8.99 9.42 9.15

26.36 30.59 30.79 32.63 26.67 30.58 27.96 27.32 27.89 26.58 26.62 26.50

20.17 21.09 22.45 21.65 22.49 24.63 28.40 30.84 29.89 31.94 28.92 30.08

(Education)

224.5 349.4 470.4 605.0 863.0 1,124.4 1,465.6 1,980.5 2,188.6 2,258.1 2,462.3 2,739.7

12.72 13.87 14.36 13.73 14.41 14.63 14.38 17.02 17.94 16.80 16.56 16.13

Source: Ministry of Finance, Government Finance Statistics.

"Includes education, health, social security, welfare, housing, and community services.

Economic Service

Other

549.7 736.8 880.7 1,154.9 1,905.2 1,996.7 2,519.4 2,513.7 2,430.4 2,565.0 3,250.9 3,395.3

163.5 244.6 354.0 441.4 592.7 788.6 1,019.6 1,279.7 1,483.3 1,803.9 1,958.5 2,425.9

31.15 29.25 26.89 26.20 31.81 25.99 24.73 21.60 19.92 19.08 21.87 19.99

9.26 9.71 10.81 10.01 9.89 10.27 10.01 11.00 12.I6 13.42 13.17 14.28

Consolidated Central Government Expenditure and Net LendineEeonomic Classification

Table A5.3

Current Expenditure ~

Year

Total

A. In Billions of Won 1975 1,765.3 1976 2,518.9 1977 3,274.4 1978 4,408.0 1979 5,990.0 1980 7,682.0 1981 10,189.8 1982 11,639.2 1983 12,200.1 1984 13,444.6 1985 14,867.0 1986 16,986.3

Total

Wage and Salary

Interest Payments

Subsidies and Transfers

Capital Expenditure”

(Gross Capital Formation)

Net Lendingb

1,257.6 1,789.3 2,361.6 3,196.6 4,472.4 5,641.1 6,931.1 8,296.3 9,144.9 10,212.8 11,523.0 13,050.2

225.0 349.9 477.3 602.9 777.9 1,053.3 1,298.5 1,484.7 1,676.0 1,756.9 1,892.3 2,027.1

48.5 94.4 132.2 190.2 285.2 433.5 567.1 652.1 686.8 856.8 992.8 1,203.7

519.4 686.7 887.6 1,111.7 2,056.5 2,244.3 2,670.4 3,445.9 3,840.3 4,593.3 5.1 10.9 5,755.3

343.1 504.4 442.7 585.3 751.6 920.9 1,113.7 1,818.7 1,536.4 1,661.8 1,813.9 2,264.7

141.1 178.7 237.8 322.9 396.6 473.9 547.5 863.1 880.6 964.4 887.8 1,281.8

164.6 225.2 470.1 626. I 766.0 1,120.0 2,145.0 1,524.2 1 318.8 1,570.0 1,530.1 1,671.4

B. As a Percentage of Total 1975 100.00 71.24 1976 100.00 71.04 1977 100.00 72.12 1978 100.00 72.52 1979 100.00 74.66 1980 100.00 73.43 1981 100.00 68.02 1982 300.00 71.28 1983 100.00 74.96 1984 100.00 75.96 1985 100.00 77.51 1986 100.00 76.83

12.75 13.89 14.58 13.68 12.99 13.71 12.74 12.76 13.74 13.07 12.73 11.93

2.75 3.75 4.04 4.32 4.76 5.64

5.57 5.60 5.63 6.37 6.68 7.09

29.42 27.26 27.11 25.22 34.33 29.22 26.21 29.61 31.48 34.17 34.38 33.88

19.44 20.03 13.52 13.28 12.55 11.99

10.93 15.63 12.59 12.36 12.20 13.33

Source: Ministry of Finance, Governmenr Finance Srarisrics ‘Includes acquisition of new and existing capital assets, purchase of stock, land, and intangible assets, and capital transfers (domestic and abroad). bIncludes net acquisition of equities.

7.99 7.09 7.26 7.33 6.62 6.17 5.37 7.42 7.22 7.17 5.97 7.55

9.32 8.94 14.46 14.20 12.79 14.58 21.05 13.10 12.45 11.68 10.29 9.84

358

Susan M. Collins and Won-Am Park

Notes Chapter 1 1. Korea’s method of calculating GNP was changed in 1985. Unless noted otherwise, we have used data based on the new System of National Accounts (SNA) method. This issue is discussed further in the Data Appendix. 2. For example, see the Korea Development InstituteIHarvard studies on modernization of the Republic of Korea, which focus on the period from 1945 to 1975. The volumes are summarized in Mason et al. (1980). More recent studies of macroeconomic developments include World Bank (1987) and Dornbusch and Park ( 1987). Chapter 2 1. Other discussions of Korea’s economic history include Cole and Lyman (197 I ) and the summary in Mason et al. (1980). Chapter 3 1. Foreign bank “A” accounts are defined as foreign currency funds borrowed by the branch offices of foreign banks from their headquarters or from other branches. 2. See table 6 in Aghevli and Marquez-Ruarte (1985, 21). 3. Errors and omissions became large during 1979-84 and have since declined. While it is difficult to tell exactly what the errors are, there is a general consensus to rule out the possibility that they represent capital flight. Other alternatives are that they are the counterpart to overestimated exports-exporters were under considerable pressure to expand, particularly during 1980-84-a that they represent countertrade transactions, for example, with oil exporting countries. In any case, the most likely scenario seems to be that they arise from underestimates of the current account deficit. 4. See Hong (1979, 130-41), Krueger (1982), and Cole and Lyman (1971) for further discussion of foreign aid allocations. 5. The average interest rates are the ratio of the total interest payment to the outstanding debt, lagged one year. 6. Cole and Lyman (1971), Hong (1979), and Krueger (1982) provide additional discussion of both the loan application process and the allocation of loans. Chapter 4 1. There are two measures of labor productivity. One gives value added per employee. The other, produced by the Korea Productivity Center (KPC), measures output per production worker. The two series do not always tell the same story. Throughout most of our discussion, we report both. 2. This period is discussed in detail in Cole and Park (1983, ch. 3). See also Gurley, Patrik, and Shaw (1965) and McKinnon (1973). 3. Unfortunately, growth rates statistics for 1970 are not strictly comparable with earlier and later years, because pre-1969 data are computed with different base prices and the old SNA method. See the Data Appendix for further discussion.

359

KoredNotes

4. The parities of the major currencies relative to the dollar were adjusted during 1972 and subsequently allowed to float. 5. See Cole and Park (1983), especially pp. 158-68, for further discussion of this period. 6. See Hong (1979, 144) for additional information. 7. These figures are based on Customs Administration data. 8. Analyses of the 1979-86 experience are also given in Park (1985a, 1985b), Dornbusch and Park (1986), Nam (1984), and Aghevli and Marquez-Ruarte (1985). Haggard and Moon (1986) and Amsden (1986) contain interesting discussions of the political economy of adjustment in Korea. 9. Bank deposits increased in the first few months after the interest rate adjustment. 10. A scandal in the curb market forced two large corporations to go bankrupt. The incident triggered a contraction in the availability of curb market loans, and many firms threatened to default. 11. These figures all use value-added measures of productivity. 12. See Dornbusch and Park (1987) for discussion of the “problems” associated with Korea’s current account surpluses. 13. See Morgan Guaranty (1987) for a review of the performance of the major debtor countries.

Chapter 5 1. Korea is certainly not the only country with data problems that make the price versus volume decomposition of imports suspect. For example, considerable recent attention in the United States has been given to problems with the price deflator for capital goods imports. The matter has been discussed in various recent issues (esp. 1987, 1988) of the Survey of Current Business. 2. See Bacha (1986). Chapter 7 1. Dornbusch and Park (1986) point out that the 1983 import coefficient for investment may be misleading because it does not take into account the decumulation of inventories in the early 1980s. Inventories have been rising since 1983, and the true coefficient is likely to exceed 0.38. 2. The Denison method of estimating the sources of growth assumes that there is a linear, homogenous, aggregate production function and that the relative price of each factor of production reflects its marginal product. Using this framework, together with detailed data, the contribution of each factor to output growth is estimated. Account is taken of changes in factor quality as well as changes in the quantity of factor inputs. The remaining increases in growth are accounted for by a number of factors, including economies of scale in production and reallocation of resources. See Kim and Park (1985) for a detailed discussion of their methodology and for additional references. 3. In these figures, the total change in factor productivity is measured as the residual which cannot be explained by measured increases in factor inputs. 4. The figures used for intercountry comparisons are standardized growth rates.

360

Susan M. Collins and Won-Am Park

5. The terminal year of 1982 may give an abnormally low level for Korean inventories. 6. For example, see the World Bank's World Development Report 1986, a study of seventeen heavily indebted middle-income countries. 7. There were also substantial increases in overseas construction in other countries. As an indication of the increased construction activity, the number of Korean construction workers employed overseas rose from 0.5 percent of the total Korean labor force in 1977 to 1.2 percent in the peak year, 1982. Chapter 8 1. For example, see the discussion in the World Bank's World Development Report 1986. 2. We do not examine the 1965 episode here because of data constraints. This would be an interesting extension. 3. A number of authors, including Nam (1984) and Dombusch and Park (19861, have pointed out the recent shift toward investment in services. 4. The material in this section come from the Government of Korea (1983). The plan includes detailed projections of national income, the balance of payments, and external debt over the plan period. 5 . Yusuf (1985) argues that investment has behaved cyclically, focusing on gross investment figures. Excluding the highly cyclic inventory component, there is little evidence of a relation between real growth and investment shares. 6. See van Wijnbergen (1982) and Sundararajan and Thakur (1980). Chapter 9 I . For further discussion of trade and industrial policies, readers are referred to the following sources which focus on the period through the mid-1970s: Frank, Kim, and Westphal (1975), Westphal and Kim (1977), Hong (1979), and Nam (1981). K. S. Kim (1986), Y. C. Park (1985a, 1985c), and World Bank (1987) focus on more recent developments. 2. The changes in industrial policies are detailed in Koo (1985). Chapter 10 1. The unusual relationship between exchange rates and wages has been discussed by Dombusch and Park (1986). 2 . Lindauer (1984) compares household-based surveys of the economically active population with establishment-based wage and employment surveys. He estimates that official earnings data may account for as little as one-third of total urban employment. 3. Employee compensation is approximately 85 percent of the total labor cost to firms, with severance pay, welfare costs, and other payments required by law accounting for the remaining 15 percent. In contrast, compensation amounts to 77 percent of labor cost in Japan and 90 percent in the United States. 4. S. Kim (1982, 57-58) claims that total compensation is more or less equated across firms for particular skill levels, despite the large differences in basic wages. These differences arise because employers and employees agree on the total compensation, while the basic wage is determined as a residual after the other benefits are set to comply with labor laws.

361

KoredNotes

5. S. Kim (1982) cites one study of the shipbuilding industry in which allowances accounted for 27 percent of compensation for production workers and for from 5 to 35 percent for managers, professionals, and technicians. 6. The comparable figures for the United States and Japan are 34 percent and 13 percent, respectively. However, the percentages expecting to continue work at reduced take-home pay with no workforce reduction were 27 percent in the U.S., 41 percent in Japan, and only 16 percent in Korea. These results are discussed in S. Kim (1982, 4-6). 7. See Jones and Sakong (1980), or their summary in chapter 8 of Mason et al. (1980), for a useful discussion of the relationships between government, business, and labor through 1975. 8. No price increases were denied because of this regulation. 9. The legality of this resolution has been strongly disputed by the Federation of Korean Trade Unions. See S. Kim (1982, 79) for further discussion of the resolution. Haggart and Moon (1986) and Amsden (1986) both stress the increasingly important role played by credit control in the relationship between business and government. 10. Government employment accounts for 18 percent of total “formal” employment. 11. This experience makes it easier to understand the concern of the Koreans that the projected growth rates in excess of 10 percent for 1986 would overheat the economy. Rapid growth concentrated in manufacturing has at times created a scarcity of skilled labor, fueling wage and price inflation. 12. The period through the early 1970s has generally been characterized in terms of a Lewis labor surplus model. 13. The narrowing of wage differentials may also be a consequence of social and/or government pressure to reduce earnings dispersion. A recent World Bank (1984) study concludes that the “rigid maintenance, until recently, of wage relatives suggests a deep concern for distributive justice with social pressures ensuring that no group is left far behind” (91).

Chapter 11 1. See Cole and Park (1983) and Mason et al. (1980) for more detailed discussions of the development of the monetary and fiscal systems. 2. The bank financing from the fiscal accounts differs in its treatment of government loans to deposit money banks, credit from BOK to the Fertilizer fund, and payment carry-overs. 3. This shift continues the historical trend away from defense spending and general services toward social and economic services which Mason et al (1980, 312-13) discuss for the 1953-75 period. 4. The Korea Telecommunications Authority was removed from the Communications Fund in 1982, and the Civil Servants Pension Fund was removed from the central government accounts in 1983. These changes reduced public sector spending by over 2 percent of GNP. The implied reductions in nontax revenues were partially offset by rising tax revenues. 5. They calculate fiscal indicators following the IMF and the OECD definitions. Advantages and disadvantages of these indicators are discussed in their paper. 6. Corbo and Nam regress the IMF measure of fiscal impulse of real growth, obtaining the following results (?-statistics are reported in parentheses):

362

Susan M. Collins and Won-Am Park F I = -0.16 - 0.21AY (0.57) (3.14)

(R2=0.496)

7. See Hong (1979, 110-30) for a detailed discussion of loan allocation policies during 1953-76. 8. In addition to restructuring interest rates, authorities introduced a number of other measures to encourage deposits and to enhance the allocation of loans. In particular, the Korea Exchange Bank was created along with other new banking institutions. Local banks were authorized to extend commercial banking to the provinces, and some foreign banks were allowed to open branches. 9. See Cole and Park (1983, 158-68) for further discussion of the 1972 presidential decree. 10. Cole and Cho (1986) and World Bank (1987) each contain additional discussion of credit allocation during 1970-85. 11. World Bank (1987) provides a detailed discussion of the 1981-86 policy changes.

Chapter 12 1. Data problems and differences in demographic and other factors imply that measures of income distribution are not strictly comparable across countries. 2. See Mason et al. (1980, 419-24) for further discussion of the land reforms. 3. Mason et al. (1980) shows that between-group inequality and total inequality measures are sensitive to the treatment of business income. If business income is weighted by its share of income tax data, intergroup inequality accounted for just 4.7 percent of total inequality in 1963. This figure had declined to 0.4 percent by 1970. Even if the much larger weights from national accounts are used, intergroup inequality accounted for 20.7 percent of total inequality in 1963 and had fallen to 16.5 percent in 1970. For further discussion, see pp. 409-15 and tables 117 and 133 in Mason et al. 4. Poor households are defined as those with incomes below the poverty line (see Suh 1980). Chapter 13 1. These data, for 1983, are quoted from table 6 in Aghevli and Marquez-Ruarte (1985, 21).

References Adelman, I. 1974. Redistribution with growth: Some country experience-South Korea. In Redistribution with growth: Policies to improve income distribution in developing countries in the context of economic growth, ed. Hollis Chenery et al. London: Oxford University Press.

363

KoredReferences

Adelman, I., and C. T. Morris. 1973. Economic growth and social equity in developing countries. Stanford: Stanford University Press. Adelman, I., and S. Robinson. 1978. Income distribution policy in developing countries: A case study of Korea. Stanford: Stanford University Press. Aghevli, B., and J . Marquez-Ruarte. 1985. A case of successful adjustment: Korea’s experience during 1980-84. IMF Occasional Papers, no. 39. Ahluwalia, M. S . , and H. Chenery. 1974. The economic framework. In Redistribution with growth: Policies to improve income distribution in developing countries in the context of economic gruwth, ed. Hollis Chenery et al. London: Oxford University Press. Amsden, A. 1986. Growth and stabilization in Korea: 1962-84. Mimeo, Harvard Business School, August. . 1987. Economic backwardness in contemporary perspective: South Korea’s industrialization through learning. Mimeo, Harvard University. Bacha, E. 1986. Terms of reference for the country studies. World Development 14, no. 8 (August): 909-18. Balassa, B., ed. 1982. Development strategies in semi-industrial economies. Baltimore, MD: World Bank and Johns Hopkins University Press. . 1984a. Adjusting to external shocks: The newly industralizing developing economies in 1974-76 and 1979-81. Mimeo, World Bank. -. 1984b. Adjustment policies in developing countries: A reassessment. World Development 12(9): 955-72. Balassa, B., and F. D. McCarthy. 1984. Adjustment policies in developing countries, 1978-83: An update. World Bank Staff Working Papers, no. 675. Balassa, B., and J. Williams. 1987. Adjusting to success: Balance of payments policy in the East Asian NICS. Policy Analyses in International Economics no. 17, June. Washington, DC: Institute for International Economics. Ban, S. H., P. Y. Moon, and D. H. Perkins. 1980. Rural development. Cambridge: Harvard University Press. Bank of Korea. 1986. Economic statistics yearbook, 1986. Seoul, Korea: Bank of Korea. Baumol, W. 1896. Productivity growth, convergence and welfare: What the long-run data show. American Economic Review 76(December): 1072-85. Bloomfield, Arthur I., and J. P. Jenson. 1951. Banking reform in South Korea. New York: Federal Reserve Bank. Casse, T. 1985. The non-conventional approach to stability: The case of South Korea. Centre for Development Research Report no. 5. Copenhagen: Centre for Development Research. Chenery, H., S. Robinson, and M. Syrquin. 1986. Industrialization and growth. London: Oxford University Press. Choo, H. 1974. Probable size distribution of relative equity in Korean income distribution: A historical perspective. KDI Monograph. Seoul, Korea: Korea Development Institute. -. 1977. Probable size distribution on income in Korea: Over time and sectors. KDI Working Paper 7722. Seoul, Korea: Korea Development Institute. -. 1978. Economic growth and income distribution in Korea. KDI Working Paper 7810. Seoul, Korea: Korea Development Institute.

364

Susan M. Collins and Won-Am Park

. 1979. Hun-Guk ui Sodeuk Punbae wa Kyuljung Yoin (Income distribution in Korea and its determinants). Vol. 1. Seoul, Korea: Korea Development Institute. . 1982a. Widening urban-rural income differentials in Korea: A reexamination. KDI Working Paper 8205. Seoul, Korea: Korea Development Institute. _ _ , ed. 1982b. Hun-Guk ui Sodeuk Punbae wa Kyuljung Yoin (Income distribution in Korea and its determinants). Vol. 2. Seoul, Korea: Korea Development Institute. -. 1985. Estimation of size distribution of income and its course of change in Korea, 1982. KDI Working Paper 8515. Seoul, Korea: Korea Development Institute. Chun, Y. I. Over time changes in regional and urban-rural income differences in Korea. In Income distribution by sector and over time in east and south eastern countries, ed. H. Ushima and T. Mizoguchi. Quezon City, Philippines. Cole, D., and Y.J. Cho. 1986. The role of the financial sector in Korea’s structural adjustment. Mimeo, Harvard University. Cole, D., and P. Lyman. 1971. Korean development: The interplay of politics and economics. Cambridge, Mass.: Harvard University Press. Cole, D., and Y. C. Park. 1983. Financial development in Korea, 1945-1978. Cambridge, Mass.: Harvard University Press. Collins, S . , 1987. Comments. Brookings Papers on Economics Activity 2:445-50. Corbo, V., and S. W. Nam. 1987a. Korea’s macroeconomic prospects and major policy issues for the next decade. World Bank Report no. DRD27. Washington, DC: World Bank. _ _ . 1987b. The recent macroeconomic evolution of the Republic of Korea: An overview. World Bank Report no. DRD208. Washington, DC: World Bank. Dervis, K., and P. Petri. 1987. The macroeconomics of successful development: What are the lessons? In National Bureau of Economic Research macroeconomics annual 1987, ed. S . Fischer. Cambridge, Mass.: MIT Press. Diaz Alejandro, C. 1981. Southern Cone stabilization plans. In Economic stabilization in developing countries, ed. W. Cline and S. Weintraub. Washington, DC: Brookings Institution. Dombusch, R. 1982. PPP exchange rules and macroeconomic stability. Journal of Political Economy 90, no. I(February): 158-65. . 1985a. Policy and performance linkages between debtor LDCs and industrialized countries. Brookings Papers on Economic Activity 2:303-56. . 1985b. External debt, budget deficits and disequilibrium exchange rates. In International debt and the developing countries, ed.G. Smith and J. Cuddington. Washington, DC: World Bank. . 1986. Exchange rate policy for NICs. Mimeo, MIT, August. Dornbusch, R., and Y.C, Park. 1986. The external balance of Korea. KDI Working Paper 8605, December. Seoul, Korea: Korea Development Institute. (Paper prepared for the World Bank Conference on Structural Adjustment in a Newly Industrialized Country: Lessons for Korea, Washington, DC, June.) __ . 1987. Korean growth policy. Brookings Papers on Economic Activity 2:389-444.

365

KoredReferences

Economics Planning Board. Major statistics of Korean economy. Seoul, Korea, various years. Frank, C., K. S. Kim, and L. Westphal. 1975. Foreign trade regimes and economic development: South Korea. New York, NY: Columbia University Press and NBER. Giovannini, A. 1985. The interest elasticity of saving in developing countries: The existing evidence. World Development 11(7):29-36. Government of Korea. 1976. The fourth five-year economic and social development plan 1977-1981. Seoul, Korea: Government of Korea. -. 1983. The revised jifth jive-year economic and social development plan 1984-1986. Seoul, Korea. -. 1986. The sixth five-year economic and social development plan. Seoul, Korea. Gurley, J., H. Patrik, and E. Shaw. 1965. The financial structure in Korea. United States Operations Mission to Korea. Washington, DC. Haggard, S., and C. Moon. 1986. Industrial change and state power: The politics of stabilization in Korea. Mimeo, Harvard University. Hanson, J. A., and C. R. Neal. 1986. Interest rate policies in selected developing countries, 1970-82. Vol. 14, World Bank Industry and Finance Series. Washington, DC: World Bank. Harberger, A. 1985. Growth, industralization and economic structure: Latin America and East Asia compared. Paper prepared for conference on Explaining the Success of Industrialization in East Asia, Canberra, Australia, September. Hong, W. T. 1979. Trade distortions and employment growth in Korea. Seoul, Korea: Korea Development Institute. . 1980. Trade policy and income distribution in Korea. KDI Working Paper no. 8004. Seoul, Korea: Korea Development Institute. . 1986a. Import restriction and import liberalization in export oriented developing economy. In Light of Korean experience. Typescript. . 1986b. Export-oriented growth and trade policies in Korea. In Trade and structural change in PaciJic Asia, ed. C. Bradford and W. Branson. Chicago: IL: University of Chicago Press. Hong, W. T., and L. Krause, eds. 1981. Trade and growth of the advanced industrialized countries in the Pac$c Basin. Seoul, Korea: Korea Development Institute. IMF. World Economic Outlook. Washington, DC: International Monetary Fund, various issues. Jones, L. P., and I. L. Sakong. 1980. Government, business and entrepreneurship in economic development: The Korean case. Cambridge, Mass. : Harvard University Press. Jung, Y. D. 1982. Income redistribution in Korea: Causes for the growing concern. Journal of East and West Studies 11( 1982):29-48. Kim, C. K., ed. 1977. Planning model and macroeconomic policy issues. Seoul, Korea: Korea Development Institute. Kim, D. M. 1979. Koyong Mit Imgeum Kujo ui Pyunwha wa Sodeuk Punbae (Employment, changes in wage structure, and income distribution). In HanGuk ui Sodeuk Punbae wa Kyuljung Yoin (Income distribution in Korea

366

Susan M. Collins and Won-Am Park

and its determinants), ed. H. Choo, vol. 1 . Seoul, Korea: Korea Development Institute. Kim, I. 1984. Korea’s policy response to world debt crisis. KDI Working Paper 8409. Seoul, Korea: Korea Development Institute. Kim, J. W. 1985. Economic development and financial liberalization in Korea: Policy response and future prospects. KDI Working Paper 8514. Seoul, Korea: Korea Development Institute. Kim, K. S . 1981. Relative price change and industrial growth patterns in Korea. KDI Working Paper 8101, February. Seoul, Korea: Korea Development Institute. . 1986. The timing and sequencing of a trade liberalization policy-The Korean case. Mimeo, Korea Development Institute. Kim, K. S., and Park, J . K. 1985. Sources of economic growth in Korea: 1963-1 982. Seoul, Korea: Korea Development Institute. Kim, K. S . , and M. Roemer. 1979. Growth and structural transformations. Cambridge, Mass.: Harvard University Press. Kim, S . 1982. Employment, wages and manpower policies in Korea: The issues. KDI Working Paper 8204. Seoul, Korea: Korea Development Institute. Kincaid, R. 1983. Korea’s major adjustment effort. Finance and Development (December):20-23. Koo, B. Y. 1984. Industrial structure and foreign investment: A case study of their interrelationship for Korea. KDI Working Paper 8402, February. Seoul, Korea: Korea Development Institute. . 1985. The role of the government in Korea’s industrial development. KDI Working Paper 8523, December. Seoul, Korea: Korea Development Institute. Korea Development Institute. 1986. Korea in the year 2000. Summary report. Seoul, Korea: Korea Development Institute. Krueger, A. 0. 1981. Export-led industrial growth reconsidered. In Advanced developing countries in the Pacijic Basin, ed. W. T. Hong and R. Krause. Seoul, Korea: Korea Development Institute. . 1982. The developmental role of the foreign sector and aid. Cambridge, Mass.: Harvard University Press. . 1983. Trade and employment in developing countries: Synthesis and conclusions. Chicago, IL: University of Chicago Press. . 1987. The importance of economic policy in development: Contrasts between Korea and Turkey. National Bureau of Economic Research Working Paper no. 2195, March. Cambridge, Mass.: NBER. Krugman, P., ed. 1986a. Strategic trade policy and the new international economics. Cambridge, Mass.: MIT Press. . 1986b. Pricing to market when the exchange rate changes. National Bureau of Economic Research Working Paper no. 1926, May. Cambridge, Mass.: NBER. Kuznets, S. 1977. Economic growth and structure in the Republic of Korea. New Haven, CT: Yale University Press. Kwack, S. Y. 1985. External influences on a small developing economy and their policy implications: The case of Korea. Harvard University. Typescript. . 1987. The economic development of the Republic of Korea, 1965-81. In Models of development, ed. L. Lau. San Francisco, Calif.: ICS Press.

-

-

367

KoredReferences

Lanyi, A , , and R. Saracoglu. 1982. Interest rate policies in developing countries. IMF Occasional Paper no. 22. Washington, DC: International Monetary Fund. Lee, J. W. 1984. Hunkuk ui Imgium Punbae ui Chui (Over-time changes in wage distribution in Korea). Kyongjehak Yongu (Economic Studies) 32(1):81- 103. Lindauer, D. L. 1984. Labor market behavior in the Republic of Korea: An analysis of wages and their impact on the economy. World Bank Staff Working Paper no. 641. Washington, DC: World Bank. Maddison, A. 1985. Two crises: Latin America and Asia: 1929-38 and 1973-83. Paris: OECD. Mason, E., M. J. Kim, D. Perkins, K. S. Kim, and D. Cole. 1980. The economic and social modernization of the Republic of Korea. Cambridge, Mass.: Harvard University Press. McKinnon, R. 1973. Money and capital in economic development. Washington, DC: Brookings Institution. -. 1982. Financial repression and the liberalization problems within less developed countries. In The past and prospects for the world economic order, ed. S. Grassman et al. London: Macmillan. Morgan Guaranty. 1984. Korea: Adjustment model for the 1980s. World Financial Markets (March):l-9. -. 1987. LDC debt realities. World Financial Markets (June/July):l- 11. Nam, C. H. 1981. Trade, industrial policies and the structure of the protection in Korea. In Trade and growth in advanced developing countries in the Pacific Basin, ed. W. T. Hong and L. Krause. Seoul, Korea: Korea Development Institute. Nam, S . W. 1984. Korea’s stabilization efforts since the late 1970s. KDI Working Paper 8405. Seoul, Korea: Korea Development Institute. -. 1986. Determinants of the domestic savings ratio in Korea: a sectoral accounting approach. Mimeo, Korea Development Institute. Park, C. K., ed. 1980. Macroeconomic and industrial development in Korea. Seoul, Korea: Korea Development Institute. Park, F., and T. Castaneda 1987. Structural adjustment and the role of the labor market. KDI Working Paper 8705. Seoul, Korea: Korea Development Institute. Park, W. A. 1986. A quarterly macroeconometric model for Korea. Korea Development Review 8(2). Seoul, Korea: Korea Development Institute. Park, Y. C. 1985a. Economic stabilization and liberalization in Korea: 1980-84. In Bank of Korea, Monetary policy in a changing financial environment. Seoul, Korea: Bank of Korea. . 1985b. Financial repression, liberalization, and development in developing countries. Korea University. Typescript. -. 198%. Korea’s experience with external debt management. In International debt and the developing countries, ed. G . Smith and I. Cuddington. Washington, DC: World Bank. -. 1988. Economic growth and external balance constraints: The case of Korea. In A policy manual for the open economy, ed. R. Dornbusch. Washington, DC: World Bank. Rao, D. C. 1978. Economic growth and equity in the Republic of Korea. World Development, vol. 6:383-96.

368

Susan M. Collins and Won-Am Park

Renaud, B. 1976. Economic growth and income inequality in Korea. World Bank Staff paper no. 240. Washington, DC: World Bank. Rhee, Y. W. 1985. Instruments for export policy and administration. Lessons from the East Asian experience. World Bank Staff Working Paper no. 725. Washington, DC: World Bank. Robinson, S. 1976. A note on the U-shaped hypothesis relating income inequality and economic development. American Economic Review 66, no. 3 (June):437-40. Sachs, J. 1985. External debt and macroeconomic performance in Latin America and East Asia. Brookings Papers on Economic Activity 2523-74. Sakong, I. 1980. Kyongjae Songjang kwa Kyongjaeryuk Chipjung (Economic growth and concentration of economic power). Hankuk Gaebal Yongu (Korean Development Study) 2(Spring):345 - 5 8 . Sanderson, W., and J. G. Williamson. 1985. How should developing countries adjust to external shocks in the 198Os? World Bank Staff Working Paper no. 708. Washington, DC: World Bank. Scitovsky, T. 1986. Economic development in Taiwan and South Korea: 1965- 1981. In Models of development, ed. L. Lau. San Francisco, Calif.: ICS Press. Shinohara, M., Y. Yanagihara, and K. S. Kim. 1983. The Japanese and Korean experiences in managing development. World Bank Staff Working Paper no. 574. Washington, DC: World Bank. Steinberg, D. I. 1982. The economic development of Korea: Sui generis or generic? U.S. Agency for Economic Development Special Study no. 6. Washington, DC: U.S. Agency for Economic Development. Suh, S. M. 1980. The pattern of poverty. In Human resources and social development in Korea, ed. C. K. Park. Seoul, Korea: Korea Development Institute. Suh, S. M., and H. C. Yeon. 1986. Social welfare during the structural adjustment period in Korea. KDI Working Paper 8604. Seoul, Korea: Korea Development Institute. Suh, T. S. 1975. Import substitution and economic development in Korea. KDI Working Paper 75 19. Seoul, Korea: Korea Development Institute. Sundararajan, V., and S. Thakur. 1980. Public investment, crowding out, and growth: A dynamic model applied to India and Korea. IMF Staff Papers 27, no. 4 (December):8 14- 55. Szall, R. J. 1981. Emerging trends in income distribution in Korea and their implications for future planning. Labour and Society 6(4):325-28. van Wijnbergen, S. 1982. Stagflationary effects of monetary stabilization policies: A quantitative analysis of South Korea. Journal of Development Economics lO(2): 133-70. Virmari, A. 1985. Government policy and the development of financial markets: The case of Korea. World Bank Staff Working paper no. 747. Washington, DC: World

Bank. Westphal, L. 1984. Exports of capital goods and related services from the Republic of Korea. World Bank Staff Working paper no. 629. Washington, DC: World Bank. Westphal, L., and K. S. Kim. 1977. Industrial policy and development in Korea. World Bank Staff Working Paper no. 263. Washington, D.C.: World Bank. Westphal, L., Y. W. Rhee, and G. Pursell. 1981. Korean industrial competence:

369

KoredReferences

Where it came from. World Bank Staff Working Paper no. 469. Washington, DC: World Bank. World Bank. 1984. Korea. Development in a global context. Washington, DC: World Bank. ___ . 1986. World Bank Report 1986. Washington, DC: World Bank. -. 1987. Korea. Managing the industrial transifion. 2 vols. Washington, DC: World Bank. _ _ . World Development Report. Washington, DC: World Bank, various issues. Young, S. G. 1986. Import liberalization and industrial adjustment in Korea. KDI Working Paper 8613, December. Seoul, Korea: Korea Development Institute. Yu, Hoon. 1979. Hankuk Sahoe Kaebal Chungchaek ui Banghyang Mosaek (Trends in socio-economic policies in Korea). Korean Journal of Public Administration 17(1).

Yusuf, S . 1985. Capital accumulation and economic growth: The Korean paradigm. World Bank Staff Working Paper no. 712. Washington, DC: World Bank. Yusuf, S., and R. K. Peters. 1984. Savings behavior and its implications for domestic resource mobilization. The case of Korea. World Bank Staff Working Paper no. 628. Washington, DC: World Bank.

This Page Intentionally Left Blank

I11

The Marcos Legacy: Economic Policy and Foreign Debt in the Philippines Robert S. Dohner Ponciano Intal, Jr.

This Page Intentionally Left Blank

1

Introduction

The experience of the less developed country borrowers is filled with irony, but nowhere is this more apparent than in the Philippines. At the end of the 1970s the country seemed to have joined the third generation of rapid Asian industrializers. Economic growth had accelerated in the mid-l970s, despite the first oil shock and the recession in the industrialized countries. Investment rates were comparable to those of Korea. The structure of exports had shifted rapidly away from primary commodities, toward light manufacturing goods. Even agriculture expanded, as irrigation investments and new strains of rice turned the Philippines into a rice exporter by the end of the decade. Economic policy was managed by a group of universitytrained technocrats, who enjoyed the confidence of the country’s external creditors, and the Philippines was among the first countries to take advantage of the new, extended financing facilities of the IMF and the World Bank. The Philippines was also favored by the international banking community, and the “Philippine desk” became a path for rapid advancement within the international divisions of many commercial banks. All of this would unravel rapidly after 1980. The Philippines was hit hard by the second oil price shock, as were other LDCs. A domestic financial crisis led to the failure of a series of major companies, many of which were bailed out by the government at great expense. The balance of payments deficit widened and was financed by more rapid external borrowing. The domestic growth rate fell year after year, even as surrounding countries were beginning to recover from the world recession. Political opposition to the government of President Ferdinand Marcos grew and spread to more The opinions expressed in this study are those of the authors and not necessarily those of the National Bureau of Economic Research nor of any other sponsoring institution. Financial support from the United States Agency for International Development is gratefully acknowledged.

373

374

Robert S. Dohner and Ponciano Intal, Jr.

conservative sectors of the society. But a watershed was reached when the most prominent opposition politician, Benign0 Aquino, was assassinated as he stepped off his plane on his return from exile in August 1983. In October the Philippines announced that it could no longer meet its debt repayment obligations, the first, and so far the only, Asian country to declare a moratorium in the current debt crisis. The reputation of the country for prudent economic management and the reputations of its technocrats were shattered by the events of the 1980s. The Philippines had one of the best debt reporting and control systems of any LDC, and had carefully managed its obligations in the 1970s, lengthening maturities and refinancing on better terms. But in the 1980s it resorted increasingly to short-term borrowing, raising the vulnerability of the country to a cutoff of external funds. Much of this short-term borrowing was hidden through duplicate financing of trade transactions, or through borrowing in the offshore market by domestic banks’ foreign currency deposit units. The net position of the monetary authorities was also obscured by a deliberate overstatement of the country’s foreign exchange reserves of as much as $1.1 billion, or half of the reported total. In the end, the Philippines waited until its exchange reserves were nearly exhausted before declaring a moratorium, and the country failed to draw on the standby lines of credit that it had negotiated and paid for. But the fragility of Philippine economic growth was nowhere better illustrated than in the loans of the major state financial institutions, the Philippine National Bank and the Development Bank of the Philippines. The asset portfolios of these two institutions literally dissolved in the 1980s. By 1986 their nonperforming assets totaled over $7 billion, or almost a third of the Philippines’ total external debt. The deficits of state financial institutions, including the central bank, had become a huge drain on the resources of the government, amounting to 5 percent of GNP in 1986. The buildup of external debt in the Philippines took place in a relatively short period of time, from 1975 to 1983. During this period the Philippines, like a number of LDCs, took advantage of the availability of bank credit and low world real interest rates to sustain domestic growth in the wake of the first oil shock. All of these LDC borrowers were hit by the triple shocks of the early 1980s-higher oil and reduced commodity prices, higher world real interest rates, and recession in the industrialized countries. The Philippines, with its high dependence on imported oil and short-term debt, was hit harder than most, and the breaking point came just as the industrialized world was recovering from its prolonged recession. But the Philippine debt crisis was not, at base, due to a series of unfavorable external events. The country had developed a borrowing momentum that could not be sustained, and the external shocks merely accelerated a process that would have occurred eventually. The roots of the Philippine debt crisis lie in the economic structure and also in the political

375

Philippines/Chapter 1

structure built up in the years since independence, particularly in the period after 1972. Foreign borrowing played a crucial role in both spheres. It produced economic growth well above what the domestic economy and the domestic policy environment could have achieved, providing temporary internal and external legitimacy to an authoritarian government. External funds also played an important role in building and maintaining a new domestic political structure that Marcos established to challenge the traditional Philippine elite. However, in the end, both the economic environment and the political structure created under the martial law years (1972-81) were inimical to the ability of the Philippines to sustain and service foreign debt, and the Philippine position unraveled quickly in the more adverse environment of the 1980s. This study examines the features of the Philippine economy and Philippine politics that led to the rapid buildup of debt and the equally rapid spiral into recession and debt crisis. It also analyzes the prospects and problems faced by the current government of President Corazon Aquino in promoting economic growth and dealing with the debt burden inherited from the Marcos regime. The two are quite closely related. For just as the Philippine debt crisis was not due solely to external events, the economic problems that the Philippines now faces go well beyond its external debt burden and restricted access to foreign capital. The problems in the Philippines are the same as they have long been-how to achieve rapid and sustained economic and employment growth.

1.1 History and Background The Philippines is an archipelago composed of some 7,000 islands, of which about 1,000 are inhabited. However, the bulk of the land mass and population are on the northern island of Luzon, the southern island of Mindanao, and a cluster of central islands called the Visayas. The climate is tropical, and the country is rich in natural and marine resources. The Philippines is a major sugar producer, accounts for about 60 percent of world exports of coconut products, and is an exporter of copper and gold. The Philippines was at one time a major exporter of logs and lumber, but the supply of these has been greatly reduced by deforestation and more recent attention to conservation. The country is subject to the vagaries of the weather, and typhoons or drought can cause major disruptions in agricultural production. The Philippines has a population of 57 million, somewhat larger than Thailand and well above that of Korea. The population is ethnic Malay, although the Philippines has experienced waves of Chinese immigration and intermarriage. The Philippines was a colony of both Spain and the United States, and each played an important role in shaping the country. Spain brought Catholicism, now practiced by 80 percent of the population, and a system of

376

Robert S. Dohner and Ponciano Intal, Jr.

administration, modeled after that of Mexico, that divided the country into large estates, or encomiendus, that were given to Spanish settlers and to the church. The United States gained control in the Philippines in 1900 as a result of the Spanish-American War, and administered the colony until 1946. The United States shaped the language, educational system, and political institutions of the Philippines, but relied on the existing Philippine elite and never effectively challenged the land tenure system inherited from the Spanish colonial period. The population growth rate is about 2.5 percent per year, one of the highest in the region, and has led to considerable pressure on the land. Up until the mid-1960s this was met by extending the area of cultivation, particularly by movement of Christian settlers into the underpopulated and largely Moslem area of Mindanao.' Since the mid-1960s, increasing population has meant greater population densities, an increase in landless laborers in the rural areas, and migration to the major cities, particularly Manila. Land tenure and land inequality are powerful and difficult political issues.

1.2 Politics and Institutions The Philippine political system before 1972 can best be described as oligarchic-a small number of wealthy, landed families dominated politics, as well as the economic life of the country. The extended family was a particularly strong source of identification and status in the Philippines, and patron-client relationships linked the population to the oligarchic family in its area or region. The result was to give Philippine politics a highly personalized and regional orientation. The elite families competed among themselves in national politics, primarily for the presidency and the spoils that office could bring. (No president was re-elected until 1969, and the only presidents not to come from the elite group were Ramon Magsaysay in the 1950s and Ferdinand Marcos.) The system that resulted was a conservative one, generally protecting the interests of the elite, but the competition among elite groups allowed some democratization of the political process and some representation of the interests of regions and localities, despite the weakness of local government. The strongest political interest group after independence in 1946 was the sugar lobby, which dominated Congress in the early years of the Republic. The sugar lobby, and to a lesser extent other members of the traditional export sector, were the primary force in pressing for more liberal trade and exchange rate policies-avoiding overvaluation of the peso and limiting the degree of import protection. But the sugar industry's political power was weakened by its poor public reputation. The sugar barons were viewed as reactionary, self-serving, and already heavily favored, both by the U.S. sugar quota (which amounted to about a million tons per year at roughly twice the

377

PhilippinesKhapter 1

world price) and by almost nonexistent taxation on agricultural land and income. Challenging the sugar lobby in Congress, and eventually winning over them, was the domestic import-substituting industrial sector. This group barely existed before the 1950s, but by the end of the decade had emerged as a powerful political force. Sheltered by protection, benefitting from overvaluation of the currency, and shrouded in economic nationalism, their conflict with the traditional sector over trade and exchange rate policy formed the most important political debate of the late 1950s and 1960s. The political institutions of the Philippines were patterned after those of the United States, with a president, two houses of Congress, and a court system, each with its areas of responsibility. The presidency was in fact much stronger in the Philippines. The Congress was an arena of “elite representation, horse-trading, and corruption” (Abueva 1979, 49) that served as a training ground for presidents. Little of a programmatic nature came out of the Congress; it had effectively ceded budgetary authority to the president, But it was a strong force on matters of taxation, foreign investment, and alien (Chinese) business operation. Local governments had a very small role, having little power of taxation and being dependent on the national government for budgetary support. Presidential politics had a large patronage component. “What are we here for?” was the response of one Philippine president when questioned about corruption in his administration. The president effectively controlled the operations of the central bank’s Monetary Board, and the allocation of credit through state and private financial institutions was used as a means of rewarding business supporters (Power and Sicat 1971, 67). Macroeconomic policy had a strong electoral cycle, as incumbent presidents tried repeatedly to assure their re-election through public expenditure increases.

1.3 Role of Government The postwar period saw a tremendous rise in the importance of the Philippine government in influencing domestic activity, particularly in the 1970s. Indeed, much of the story of the Philippine debt crisis described in this study is the expansion of the national government’s economic role and the political strategy behind it. However, the starting point for the Philippine government was much smaller than that of governments in other LDCs. Government expenditure as a share of GNP in the Philippines averaged only 11 percent in the 1950s and 1960s, versus 20 percent in Thailand and Korea, and 24 percent in Malaysia. Gross investment by government was only about one-fifth of total government expenditure.2 Much of this was devoted to political patronage in the annual Public Works bill (termed “the Pork Barrel Bill” by domestic legislators), so that there was little systematic attempt to use the government as a vehicle for developmental capital formation.

378

Robert S. Dohner and Ponciano Intal, Jr.

The small scale of government reflected the ideological bent of the American colonial administration, but also coincided with the interests of the Philippine landed aristocracy and the commercial sector, the dominant political groups. Although the Philippine government had some early experience with state-owned enterprises, the prevailing orientation was toward private sector activity. Public utilities, transport, and communications were largely in private hands. Although the importance of public corporations and market intervention would grow tremendously, the Philippine government would continue to publicly maintain the primacy of the private sector. Tax revenue as a share of GNP has been relatively low in the Philippines, consistent with the small government expenditure share. But beyond this, difficulties in raising tax revenues have been persistent constraints on the mobilization of domestic resources through the public sector. The utilization of potential tax bases has been low by international standards, as has been the efficiency of collection of existing taxes.3 The division of political authority before 1972 gave the president de fact0 authority in allocating expenditure, but the Congress retained control over tax matters and resisted attempts to increase the revenue raised through the tax system. Of particular importance in the Philippines is the fact that agricultural property, and to some extent agricultural income, almost completely escaped taxation. As a result, export taxes have in part been used as substitutes for other taxes on the agricultural sector. 1.4

Economic Nationalism

Nationalism has been a persistent theme in Philippine politics and has had a large economic component. There has been a strong desire to “Filipinize” the country’s economy-to reserve land ownership, use of natural resources, and participation in many economic activities to native Filipinos. Nationalist sentiment and policy has been directed against foreign investors, but also against “aliens”-non-Filipino citizens, who are mostly Chinese. Almost from the beginning of the American colonial period there was strong pressure for independence, from Filipinos and also from political groups in the United States opposed to the retention of colonies. As early as 1916 the United States committed itself to eventual independence for its Asian colony. U.S. legislation in 1934 established a commonwealth in the Philippines, with a ten-year transition to full independence, although the process was interrupted by the Second World War. After the war, the United States sought to assure continued privileges for American citizens in an independent Philippines. Using the leverage of withholding its aid and rehabilitation funds, the United States forced the country to accept a series of constitutional amendments and policies that would assure Americans parity with Filipinos in key areas. These were contained in the U.S.-Philippines

379

PhilippinedChapter 1

Trade Agreement Act of 1946 (the Bell Act), which required that U.S. citizens be given the same rights as Filipinos to own land, exploit natural resources, and operate public utilities. The Bell Act included provisions which established free trade between the two countries, although Philippine exports of sugar, coconut oil, and cordage were still subject to U.S. quotas. The act also prohibited export taxes and required U.S. approval before the Philippines could change its exchange rate. Although the provisions of the Bell Act were softened by the Laurel-Langley Agreement of 1955, and under that agreement the parity amendments expired in 1974, the provisions forced upon the Philippines after World War I1 were a source of much resentment, as well as a limitation on Philippine policy choices. The measures that shaped Philippine trade and industrial policy-the adoption of import quotas and industrial incentives-were in part due to the limited flexibility of the Philippines in addressing its first balance of payments crisis after independence. The areas of particular emphasis in nationalist policy have included import and retail trade, natural resources and general land ownership, and processing and marketing of agricultural products. Import trade in the late 1940s was dominated by Western and Chinese firms. However, the import controls adopted in 1950 gave the Philippine authorities a powerful weapon for increasing the share of Philippine nationals. The import control legislation required that 40 percent of import licenses be allocated to new Filipino importers, with the share gradually increasing over time. By 1956 the import quota allocations to Filipinos exceeded 75 percent (Golay 1961, 321). Filipinization policy in natural resources and in public utilities was hampered by the parity amendments, which gave American investors the equivalent of Filipino status until the expiration of the Laurel-Langley Agreement in 1974. However, regulatory opposition to rate increases was used to encourage the sale of American-owned utilities, and the Philippine Supreme Court’s decision in the Quasha case (Republic v. Quasha, 17 August 1972), that property rights acquired under the parity amendments would lapse in 1974, encouraged American disinvestment in natural resource industries (Golay 1983, 142-43, 151-53). In contrast to the highly sensitive areas discussed above, Philippine industrialization policy has taken a more liberal stance toward foreign ownership. Philippine policy did not discriminate against foreign industrial investors until 1957, when foreign exchange allocation for capital goods and raw materials import was used to favor Filipino firms (Golay 1961, 259-60, 330-33). During the 1950s, foreign investment in manufacturing industries increased ~ubstantially.~ The Philippines went through an import decontrol period in the early 1960s, followed by a period of sluggish manufacturing growth and excess capacity. Under pressure from domestic manufacturers, government guidance of investment and preferences for Filipinos increased. The Industrial Incentives Act of 1967 required 60 percent Filipino ownership

380

Robert S. Dohner and Ponciano Intal, Jr.

in nonpioneer industries and a gradual transfer of ownership to Filipinos in pioneer industries. The act also established a Board of Investments (BOI) which was given considerable latitude in administering investment incentives, as well as the authority to limit investment in “overcrowded” industries. Although foreign ownership has remained a sensitive political issue, foreign direct investment has not been an important source of external capital for the Philippines. Net foreign direct investment inflows, shown in table 1.1, have been a small proportion of domestic capital formation, among the lowest in any of the ASEAN (Association of South East Asian Nations) countries. The net investment figures clearly reflect the swings in Philippine trade and investment policy. The import controls of the 1950s drew foreign investors into import-substituting industries. The sluggish growth of the manufacturing sector during the decontrol period (1961 -66) and the impending lapse of the Laurel-Langley Agreement is also evident in the reduction of foreign investment in the 1960s and early 1970s. This was a period of substantial disinvestment by American firms in mining, utilities, and other i n d ~ s t r i e s .The ~ imposition of martial law in 1972 led to greater efforts to promote foreign direct investment, but even during this period, the contribution of direct investment to total external capital inflows remained quite small.

1.5 Trade and Industrial Policy The thrust of postwar Philippine trade and industrial policy has been to encourage the development of industries serving the domestic market, through import protection and substantial investment incentives. In the process, the country has discriminated against its export sector, particularly Table 1.1

Foreign Investment in the Philippines (in millions of U.S. dollars) A. Net Foreign Direct Investment Inflows 1946-50

1951-55

1956-60

20.2

28.2

26.9

Average annual inflow Percentage: of GNP of Capital formation

.67 N.A.

.66 3.2

.45

2.8

1961-65 -

1966-72

1973-78

1979-83

-13.6

99.6

44.4

10.8

- .I6 -.76

-.I8 - 1.4

.65 2.3

.I2 .42

B. Book Value of U.S. Investment (year end)

Total stock

1950

1960

I966

1972

1979

1985

149

414

519

608

913

1.032

Suurces: A: Central bank, Annual Report, various issues; B: U.S. Department of Commerce Nore: N.A. = not available.

381

PhilippinesiChapter 1

the traditional commodity exports, through currency overvaluation and, in some cases, export taxes. Two attempts to liberalize the trade regime and encourage exports-ne in the early 1960s and the second in the 1970s-were in the end unsuccessful, in part because the objective of protecting existing domestic industry was never abandoned. A balance of payments crisis in 1949 led the Philippines to impose licensing requirements based on the degree of “essentiality” of the import. This led to the development of a domestic manufacturing sector providing “nonessential” consumer goods, as well as a group of industrialists dependent on import protection. Initially the policy was successful, spurring foreign direct investment and investment by domestic residents, and the country’s growth in the early 1950s was among the highest in the region. In addition, as mentioned above, the allocation of import licenses was a powerful tool for Filipinizing the import trade. Slowed growth toward the end of the 1950s, foreign exchange shortages, charges of corruption surrounding the allocation of licenses, and the continuing influence of the sugar industry led to a phased elimination of the import licensing system, as well as a devaluation of the peso, in the early 1960s. Although exports of the traditional sector increased, the overall experience of decontrol was disappointing. The economy continued to grow sluggishly, with the manufacturing sector remaining particularly weak, and the devaluation brought about a sharp rise in the domestic inflation rate. The period did not see the development of significant new exporting industries. The experience of the decontrol period profoundly influenced those on both sides of the trade policy debate. Excess capacity and low profits in manufacturing led to increased economic nationalism, as well as calls for government intervention on the part of the domestic industrial sector. Proponents of trade liberalization and export promotion shifted their ground after the decontrol of the 1960s and advocated export promotion as a way of developing new industries without challenging the existing system of trade protection. In the remainder of the decade there was a gradual increase in the level of trade protection, as well as the adoption of industrial incentive systems which encouraged industries that exported, but also industries that served the domestic market.

1.6 The First Marcos Administration, 1966-69 The events of this period are in many ways a striking precursor to the accumulation of external debt in the 1970s and early 1980s. The rapid rise of external debt during the first Marcos term, much of it of short maturity, culminated in a balance of payments crisis in 1970, the rescheduling of external debts, and an IMF adjustment program. Ferdinand Marcos defeated Diosdado Macapagal in his re-election bid in 1965, at a time of widespread dissatisfaction with the import decontrol

382

Robert S. Dohner and Ponciano Intal, Jr.

program. The Marcos administration immediately moved to accelerate economic growth and was more aggressive in its use of government expenditure and economic policy than previous Philippine administrations. During the initial years of Marcos’ first term, fiscal and monetary policies turned decidedly expansionary. The national government greatly raised its capital expenditures, largely concentrating on infrastructural projectsirrigation, roads, schools, and communications-in the rural areas. In all, government expenditure rose in real terms by approximately 43 percent from 1964 to 1968 (two non-election years), and the share of government expenditure in GNP rose from 11.5 to 14 percent. The rise in expenditure was financed primarily by borrowing, both domestic and external, as the national government budget shifted from a slight surplus to a deficit of 3 percent of GNP. On the monetary side, the central bank initiated what it described as “massive credit relaxation,” lowering the discount rate and greatly increasing rediscount ceilings. An industrial rehabilitation facility was established at the Development Bank of the Philippines that offered industrial loans for refinancing and the conversion of some loans into equity. Between -1965 and 1967, domestic credit increased by 40 percent, compared to a rise in nominal GNP of 18 percent. The stimulative program of the Marcos administration quickly ran up against external payments difficulties. The increase in government and private investment led to a 24 percent rise in imports in 1967 and a further increase in 1968. By 1968 the current account deficit reached 3 percent of GNP. The worsening external situation did not prevent the traditional run up in expenditure in the 1969 election year. Marcos became the first Philippine president to win re-election, in an election that was by far the most expensive and also the most violent and suspect of any up to that point. Government expenditure rose by over 25 percent in 1969, and the deficit of the national government tripled in that year. Most of the increase in expenditure was financed by the central bank; the money supply rose by 20 percent in the last four months of 1969 alone. The increase in expenditure by the government and the outlays of government corporations and financial institutions had been financed by extensive borrowing, both internal and external. Much of that borrowing had been short term. President Marcos explained to a business group in Manila in early January: “We have unfortunately financed the foreign exchange requirements of our development with credits of short maturities. I am told by my advisers that because of the increase in short-term debts, the total payment for interest and amortization this fiscal year ending June 1970 will take over half our export earnings.”6 A summary of Philippine external debt in this period is contained in table 1.2.

383

Philippines/Chapter 1 Philippine External Debt, 1965-70 (in millions of U.S. dollars)

Table 1.2

Year

F’ublic Medium & Long Term

1965 I966 1967 1968 1969 1970

286 269 281 433 480 738

Public Short

Term

Pnvate Medium &Long Term

Private Short Term

73 103 209 120 196 63

190 209 445 698 959 1,049

43 145 200 276 287

51

Total Debt

(% of

GNP)

(% of Exports)

600 624 1,079 1,450 1,912 2,137

10 9 15 I8 22 31

56 53 90 126 173 162

Source: Central bank, Management of External Debt and Investment Accounts Division. Data includes IMF obligations.

The major official creditors formed a Consultative Group for the Philippines in January 1970 and agreed to restructure the external debt in return for Philippine agreement to an IMF stabilization program. That program required that the peso either be sharply devalued or allowed to float. The Philippine government accepted the latter condition, and by year end the peso had fallen from 3.9 to 6.4 per dollar.

1.7 Stabilization and Martial Law The early 1970s was a period of economic stabilization and recovery, accompanied by a rapidly deteriorating political situation. The devaluation of 1970, tighter monetary and fiscal policies, and the external commodities boom quickly restored external balance. But the increasingly violent domestic political situation would culminate in the suspension of the constitution and the declaration of martial law in 1972. The macroeconomic outlines of this period are contained in table 1.3. The devaluation in 1970 was coupled with tighter fiscal and monetary policy as the government cut investment expenditure in 1970 and again in 1971. A rapid increase in exports moderated the fall in GNP during the stabilization. Some of this was due to the movement of existing exports into official channels, and the growth depended upon investments undertaken in the late 1960s, but the export response to the devaluation was still impressive. By 1972 GNP growth was above its previous trend and macroeconomic policy became more expansive. Current account balance was rapidly restored, and the rise in external commodity prices in 1973 resulted in an unprecedentedly large surplus. The foreign debt position of the Philippines improved markedly during this period, as moderate borrowing and rapid nominal income growth reduced the debt/GNP ratio sharply in the early 1970s. The most startling event of this period is the marked decline in real wages that occurred after 1969. Real wages for both agricultura1 and nonagricultural

384 Table 1.3

Robert S. Dohner and Ponciano Intal, Jr. Macroeconomic Indicators, 1%9-73 (annual percentage change, except where indicated)

Real GNP Money supply (MI) Budget surplus (% of GNP) Consumer prices Real wages (CPI) (index) Export volume Import volume Current account (% GNP) REER exports (index)a REER imports (index)= DebUGNP (70)

1965-68

1969

4.9 9. I -0.7 4.2 100 2.8 6.7 0.0 100

5.2 19.4 - 3.5 1.4 103 2.0 -2.1 -3.2 95 96 22

100

13

1973

1970

1971

1972

-

3.9

6.5 10.3 -0.5 15.0 91 9.6 7.0 0.0 112 127 27

5.4 24.9 -2.4 16.6 86 3.7 0.9 0.1 106 129 26

9.3 12.3 - 1.2 16.5 75 7.7 -6.4 5.0 132 144 22

- 1.2 0.2 14.8 96 14.4 -6.6 -0.7 121 126 33

"REER (real effective exchange rates) are defined as the export or import unit value divided by the GDP deflator.

workers fell by about 25 percent between 1969 and 1973. Furthermore, real wages were maintained at this level through 1980, despite a 40 percent rise in per capita GNP over the 1 9 7 0 The ~ ~ drop in real wages was at once the success and the failure of economic policy during the Marcos era. Low wage costs were the primary engine behind the rapid expansion of manufactured exports during the 1970s. At the same time the failure of real wages to increase and the sluggish growth of manufacturing employment were reflected in increasing income inequality and absolute poverty in the Philippines in the martial law regime. High recorded rates of economic growth did not translate into improvements in the lot of the Filipino masses.

1.8 Political Deterioration and Martial Law The mixed but generally positive results of the stabilization period were accompanied by rising domestic political tensions and violence. The 1969 election, in which Marcos had been returned for a second term, was a low point in the Philippine electoral process. Marcos had spent far more than any previous incumbent in seeking re-election against an opponent who was generally given little chance of success. The campaign and election were also more violent than previous elections; by one estimate, two hundred people were killed during the campaign (Shaplen 1979, 21 1). The election greatly heightened political animosities and spawned violent protests by student groups in 1970 and 1971, directed against Marcos but also against the Philippine Congress, which was widely dismissed as corrupt, inefficient, and obstructionist. Public cynicism toward the government was increased by the constitutional convention that Marcos called in 1971, which was a thinly disguised attempt to extend his hold on power beyond the eight-year maximum in the existing constitution. This period also saw the reorganiza-

385

PhilippinesKhapter 1

tion and heightened activity of the Communist Party of the Philippines (CPP). The domestic situation continued to deteriorate in 1971 and 1972. A hand grenade was thrown into a rally of the opposition Liberal Party in August 1971, severely injuring candidates for the off-year congressional election. In the year that followed there were bombings of public buildings and almost constant demonstrations. Wealthy Filipinos were kidnapped and held for ransom. The New People’s Army (the military arm of the CPP) widened its activity in Luzon, and the Moslem rebels in the South stepped up their attacks on Christian settlers. The Philippine constitution permitted the president to declare martial law in a time of national emergency, and Marcos considered doing so for some time. The pretext he eventually used for the declaration was a bombing attack on the car of his defense minister, Juan Ponce Enrile, on 22 September 1972, an attack that Enrile later admitted had been faked. The evening of the attack, citing a conspiracy of leftist and rightist groups and the Moslem secessionist movement, Marcos declared martial law. That night, hundreds of persons were arrested by the military, including opposition politicians and journalists. Radio and television stations were closed, and the country’s newspapers shut down. The Congress was dissolved, and under martial law powers, Marcos began to rule by presidential decree. The constitutional convention that was formed in 1971 continued, minus some dozen opposition leaders who had been detained, and in 1973 produced a draft constitution providing for a transition to a parliamentary form of government. The length of the transition period was left to Marcos’ discretion, and the draft constitution also gave Marcos the ability to dismiss any member of the judiciary. The new constitution was ratified in a hastily organized referendum and upheld by the Supreme Court. The martial law government moved rapidly to restore public order and, in its words, introduce a sense of discipline in Philippine life. The Philippines at the time was a heavily armed society, with local administration often in the hands of regional oligarchs and their private armies. The Philippine army confiscated nearly half a million guns from private citizens, moved to disband private armies, and integrated local police forces in the national bureaucracy (Abueva 1979, 36). After this and some heavily publicized executions, the incidence of violent crime dropped sharply. The new sense of discipline, or perhaps wariness, was evident in other forms of behavior. Tax collections rose significantly in the year after martial law, aided by the threat of severe penalties and a tax amnesty on the declaration of hidden wealth. The number of people filing income tax returns increased by a factor of four after martial law. And observers recall the period as the first in memory when Filipinos actually queued for buses. Initially martial law was met by public ambivalence. The declaration had been widely anticipated and was viewed as a power grab by Marcos. Yet

386

Robert S. Dohner and Ponciano Intal, Jr.

there was a widespread willingness to give Marcos and martial law a chance. The disorder of the early 1970s had frightened many people, and the restoration of public safety was widely appreciated. There was also the feeling shared by many that the political system in the Philippines had not served the country well; few mourned the passing of Congress. Finally, although the factors that precipitated martial law were political and security issues, the Marcos administration moved quickly to provide an economic justification for “constitutional authoritarianism.” Marcos himself declared that the conquest of mass poverty and the democratization of wealth were to be the major aims of his “New Society.” One of the first acts of the martial law government was to institute a heavily publicized land reform. Central to the acceleration of economic growth and the distribution of economic benefits promised by the regime was a greatly expanded governmental role in development. With the Congress removed from the budgetary process, Marcos sent orders to his executive departments to prepare a list of bottlenecks in each of their functional areas and to draw up proposals for investment projects. The planning mechanism was reorganized and strengthened with the formation of the National Economic and Development Authority (NEDA). Marcos increasingly staffed government bureaus with technocrats-Filipinos with advanced degrees in economics, business, and engineering-who were in turn drawn by the prospect of rapid implementation of policy by a progressive government. The result was a swift increase in government expenditure, particularly public investment, supported by somewhat higher tax collections and an increased resort to foreign funds. The growing importance of the public sector is only partially indicated by the rise in national government expenditure shown in table 1.4. Much of the increased investment was done off the books of the national government by state-owned corporations in the energy, agricultural infrastructure, and transport areas. The last line of table 1.4 shows the rapid rise of total public investment expenditures, including Table 1.4

National government expenditure National government revenue National government surplus Government investmenta

Public Sector Expenditure, Revenue, and Investment (percentage of GNP) 1970-72

1973

1974

1975

1976

1978

1980

1982

12.7

14.3

11.7

16.0

15.2

14.8

14.4

15.7

11.9

13.2

12.2

14.4

13.4

13.6

13.1

11.4

-0.8

1.6

-1.1

2.3

0.5

3.4

-1.6 4.3

-1.8

6.6

-1.2 7.2

-1.3 6.9

Source: NEDA, National Accounts Section. Bureau of the Treasuly, Cash Operations Sraremenrs ”National accounts basis, includes government corporations

-4.3 1.2

387

Philippines/Chapter 1

those of public corporations. By 1978 total public investment had risen to 7 percent of GNP, or 30 percent of total domestic capital formation. Martial law, and the administrative and policy changes that quickly followed, met with a favorable response from the foreign aid and multilateral community. The flow of official development assistance from members of the OECD Development Assistance Committee (ODA) more than doubled after 1972.' Support for an accelerated program of public investment can be seen in the World Bank's Country Economic Report on the Philippines, published in 1976: The basis for the structural changes that are expected is a substantial increase in investment, both public and private, which would move the economy towards sustained growth of incomes and employment and a more acceptable distribution of wealth. The large investment program cannot be financed out of domestic savings alone; large foreign inflows will be required. . . . Public investment in infrastructure will need to be raised to about US$l,OOO million a year by 1980 compared with the present level of about US$400 million (both at 1974 prices). With a GNP growth rate of about 7 percent per year, public investment would need to be raised to at least 5 percent of GNP compared with the present 3 percent. (15- 16, 26) There was a second force behind the growing importance of the national government which, although it had an economic component, was primarily a matter of political consolidation. As described above, Philippine politics had been dominated by a relatively small number of wealthy families. It was still possible for outsiders to enter and succeed in politics, but the entrenched power of the elites in the Philippine Congress had successfully blocked policies inimical to their interests. Marcos himself was an outsider. Although from a well-to-do family in the Ilocos region of northern Luzon, he was viewed as a parvenu by the traditional elite. With martial law, Marcos achieved a transformation of the political structure in the Philippines, successfully entrenching and consolidating his own power, and at the same time establishing his own family and that of his wife in the upper rank of the Philippine elite. The key to this was a greatly expanded national government, both as a means of centralizing authority and displacing the regional powers that had characterized Philippine politics, and as a patronage machine for rewarding supporters and punishing opponents. The use of patronage was by no means new to Philippine politics, but Marcos used it brilliantly, along with the authority that martial law had given him, to eclipse the elite that had controlled Philippine politics. Marcos undermined the political structure of the traditional families by cutting off their lines of influence and by breaking their local control. The Congress was. disbanded with the declaration of martial law, and the government seized and closed all newspapers and radio and television

388

Robert S. Dohner and Ponciano Intal, Jr.

stations, depriving Marcos’ opponents of much of their voice. The confiscation of guns, disbanding of private armies, and the integration of the local police into the Philippine army centralized local control. Local elections for governors and mayors were abolished, and local representation was reorganized as barangay (village) councils. Marcos also made a bid for populist support with promises to democratize wealth and with land reform.’ While not destroying the economic bases of his opponents, Marcos was able to effectively threaten them. Through confiscation and forced sale, Marcos acquired much of the assets of the Lopez and Jacinto families almost immediately after the declaration of martial law, and cowed other potential opponents.” Most of the remaining elite families, if not supporters of the regime, made their accommodations to it. The most immediate beneficiary of martial law was the army, which greatly expanded in size, responsibility, and emoluments. The army expanded from about 60,000 at the time martial law was declared to more than 250,000 by the end of 1975. The military budget more than quadrupled between 1972 and 1976. Military officers were given a wide range of administrative and managerial authority, and in many cases sat on the boards of state-owned corporations. Some developmental tasks, such as road building, were transferred to military commands at greatly increased cost. To assure loyalty within the army, Marcos also filled the higher ranks with officers from his home province of Ilocos Norte. The expansion of the scale of government in the 1970s and particularly the acceleration of public investment expenditures greatly increased the ability of the Marcos government to distribute patronage, both to enrich Marcos’ close associates and his own family and to assure loyalty in key sectors. Investment and construction projects were especially well-suited to distributional politics of this sort, as they were highly visible and employmentcreating expenditures. Furthermore, construction and the purchase of equipment offered opportunities for padded expenses, inflated prices, kickbacks from suppliers, and even outright diversion of funds. The availability of foreign funds and external borrowing in the 1970s facilitated this process in a variety of ways. Foreign exchange costs were a significant component of the developmental project costs, and the availability of foreign funds increased the scale of such expenditures beyond the level of foreign exchange resources that could normally have been generated by the traded goods sector. Access to credit had been a traditional tool for rewarding political supporters, and foreign borrowing increased these resources. In addition, foreign borrowing tended to concentrate credit allocation in the state; both lending by the multilateral institutions and, increasingly, lending by commercial banks depended on sovereign guarantees. Finally, in an economy with capital controls and a black market exchange rate premium of varying degree, foreign exchange resources offered a particularly attractive way to distribute, and hide, wealth.

389

Philippines/Chapter 1

The growth of government influence and patronage operations was not merely a function of increasing public expenditure. The extent to which the national government intervened in the operation of the domestic market vastly increased under martial law. Some of this intervention was to advance economic goals, such as price support and stabilization, industrial development, energy supply, and the rescue of firms in financial distress. But much of the intervention involved capturing and channeling economic rents. Monopoly and monopsony positions were created in key industries, particularly in the traditional export sector, exclusive rights were granted to particular firms and individuals, and government power was used to force the transfer of assets from one owner to another. To a real extent, the Philippines under martial law developed a rent-seeking and rent-distributing government, which over time would sap the energy of the domestic economy and which contributed significantly to the economic crisis of the 1980s.

1.9 The Philippine Economy in the 1970s The new martial law government was the fortunate beneficiary of the worldwide upswing in commodity prices of 1972-74; in the first year of martial law there was a 13 percent rise in the terms of trade, an expansion of exports, and a 9 percent growth of real GNP (table 1.5). Even with the oil shock, Philippine terms of trade improved in 1974, as copper and log prices nearly doubled from their 1970-72 average and sugar and coconut prices tripled. I ’ The fortunes of the Philippines reversed in 1975 with the collapse of international commodity prices, and by 1976 the country’s terms of trade were 29 percent below their 1970-72 average. Ironically, 1974 would mark

Macroeconomic Indicators, 1973-79 (annual percentage change, except where indicated)

Table 1.5

GNP National government budget (% of GNP) Expenditure Revenue Surplus/deficit Money supply (M1) CPI Real wages (CPI) (index)” Export volume Import volume Terms of trade (index)” Current account (% of GNP) ”1965-68 b1970-72.

= 100

1965-72

1973

1974

1975

1976

1977

1978

1979

5.0

9.3

5.6

5.8

7.4

6.3

5.8

6.9

14.3 13.2 -1.2 12.3 16.5 75 7.7 -6.4 90 5.0

11.7 12.2 0.5 24.0 34.2 61 - 10.7 17.8 91 -1.2

16.0 14.4 -1.6 14.5 6.8

15.2 13.4 -1.8 17.1 9.2 67 28.1 5.9 62 -5.8

14.9 13.0 -1.9 23.7 9.9 62 20.6 -2.8 56 -3.6

14.8 13.6 -1.2 13.4 7.3 65 3.1 18.2 62 -4.6

13.7 13.5 -0.2 11.2 16.5 69 8.5 8.9 65 -5.1

-1.1

11.3 8.4 97 5.2 3.5 87b -0.5

64 5.9 5.0 70 -5.6

390

Robert S. Dohner and Ponciano Intal, Jr.

the peak of Philippine export prices; despite the inflation of the 1970s and early 1980s, the dollar prices of Philippine exports would never recover their 1974 levels. Like other oil-importing developing countries, the Philippines was hit by both terms of trade deterioration and a slowdown in the growth of external markets. Since the Philippines exported commodities and was dependent on oil imports for most of its energy, the change in external prices had a severe effect, amounting to a real income loss of 5.6 percent of GNP.” Philippine economic policy in this period was caught up in the transition to martial law. The new government had liberalized foreign investment policy, had declared an end to rice and corn tenancy and started a land transfer program, and had emphasized tax collection and severe penalties for evaders. The four-year economic plan prepared by the Marcos administration called for a substantial rise in the government expenditure share and a doubling of the share of government spending devoted to investment. Government expenditures in real terms rose 17 percent in fiscal year 1973 (July 1972 to June 1973), while outlays for infrastructure rose 50 percent in the same period. The Philippine government made a deliberate decision after the oil shock to continue with the expenditure plan while accelerating its energy component.13The desire to maintain the momentum of martial law certainly entered into the decision. But the decision was also in line with external advice and was widely supported by the country’s economic advisers. The mood among policymakers in the wake of the first oil shock was in fact optimistic. The martial law government was just getting started and had had some initial success. More technocrats were being added to the government ministries, and a more rational, development-oriented policy approach had been announced. There was confidence in the ability of the economy to adjust to the external price changes and optimism about the Philippines’ export potential. The oil price shock was seen as something of an opportunity, a possibility for making fundamental policy reforms. l4 The Philippine government responded to the external shock with a huge increase in government expenditures; real outlays in fiscal 1975 rose by 40 percent. The rapid increase in expenditure was reflected in the current account deficit, which rose to over 5 percent of GNP. The Marcos administration sought additional aid flows and direct investment, but the gap was met primarily by external borrowing. The strategy was successful in maintaining, and even raising, the rate of domestic economic growth-the average GNP growth rate for the remainder of the decade was 6.2 percent per year, a higher sustained growth rate than the Philippines had had since the initial import substitution phase in the early 1950s. The expansion was led by domestic investment, which increased by 35 percent in real terms between 1974 and 1976. The share of domestic expenditure devoted to investment jumped to almost 30 percent and stayed at

PhilippineKhapter 1

391

that level for the remainder of the decade. As indicated in table 1.6, the primary reason for this was a huge increase in government investment expenditures, although private investment also increased in the initial years of martial law. Philippine exports grew at a rate of 13 percent per year from 1974 to 1980, placing the country in league with the most rapidly growing Asian exporters. l 5 The decade also saw a transformation in the product composition of Philippine exports. In 1970 over 90 percent of Philippine exports were primary commodities or slightly processed commodities. By 1979 the share of these products in Philippine exports had fallen below 50 percent. In their place were several nontraditional, labor-intensive, manufactured export products, the most important of which were garments, semiconductors, and integrated circuits. There was a high import content to Philippine manufactured exports, and the domestic investment boom kept capital goods imports high. So, despite the rapid growth of export earnings, the current account deficit hovered around 5 percent of GNP. With the rise in the current account deficit after the first oil shock, the country's foreign debt grew rapidly, nearly tripling between 974 and 1978 (table 1.7). The public sector did most of the borrowing and held two-thirds of the foreign debt of the nonbanking sector by the end of the decade. The Philippines borrowed increasingly from banks in the form of loans with floating interest rates. But this was true of all LDC borrowers during the 1970s, and the shifts toward commercial terms and floating rates were less pronounced in the Philippines than in most borrowing countries. The country's policymakers managed the debt carefully during the 1970s, lengthening maturities and refinancing when better terms were available. As a result, the debt service ratio (interest and amortization payments as a percent of exports) increased only slightly, reaching 21 percent by 1980. Few of the problems that the Philippines would face in the 1980s were evident in 1979. The Philippines had significantly increased its external indebtedness, but had also raised its export and GNP growth rates. At the Table 1.6

Investment Fixed investment Private Government of which: Public Corporations Construction Durable equipment Savings" Source:

Investment and Savings Shares in GNP, 1970-79 1970-72

1973

1974

1975

1976

1977

1978

1979

21.9 16.9 15.3 1.6

21.9 15.8 13.6 2.3

26.7 18.5 15.1 3.4

30.6 23.7 19.4 4.3

31.3 25.1 18.6 6.6

29.0 23.8 16.9 6.9

29.1 23.9 16.7 1.2

31.0 25.8 18.5 7.3

0.2

0.2 6.1 9.7 27.0

0.5 7.6 10.9 25.4

0.6 10.2 13.5 25.3

3.3 13.3 11.8 25.4

3.5 13.3 10.5 25.8

4.6

4.0 14.0 11.8 26.6

6.1

10.8 21.7

NEDA, National Accounts Section

"Including capital consumption allowance.

12.R

11.0 24.4

Table 1.7

Philippine External Debt (in millions of US. dollars) I970

Total external debt Nonmonetary debt Medium & long term Short term Monetary sector debt Memorandum items: DebtiGNP (%) DebtExports of goods, services Debt service ratio" Short term as % of total external debt

2,297 2.088 1.779 359 159

1974 3,755 2.726 2,395 33 1 1,029

1978

1980

1982

1983

1984

1985

1986

1987

10.694 8.189 6.932 I .257 2.505

17,252 12,318 9,770 2,548 4,934

24,677 17,601 13,141 4.460 7,076

24,816 19.468 15,412 4.056 5,348

25.418 20.21 1 15,926 4,285 5,207

26,252 21,270 17.679 339I 4,982

28,256 25,668 22,878 2,790 2,588

28,649 26,702 24,857 1.845 1.947

33.2

25.5

44.5

49.0

62.8

72.7

80.6

174 29.2

106 14.6

218 20.1

215 20.8

308 38.1

305 38.2

22.6

36.2

35.2

43.4

46.7

37.9

Source: Philippine central bank, Management of External Debt and Investment Accounts Division, and central bank, Financial Plan Data Center. "Total interest payments plus amortization of total medium- and long-term debt as a percentage of exports of goods and services bAfter rescheduling.

81.7

92.9

84.1

317 43.4

332 36.Yb

328 34.0b

311 35.3b

37.3

32.7

19. I

13.2

393

PhilippineKhapter 1

end of 1979, the Philippines had a debt/GNP ratio comparable to that of Korea. Its debt service ratio was higher than Korea’s, but was well below that of most Latin American borrowers.

1.10 The Second Oil Price Shock and Its Aftermath The Philippine economic situation deteriorated rapidly after the second oil price shock in 1980. The government again tried to counter the growing domestic recession by raising expenditure, and announced an ambitious program of energy and industrial investment. As a result, the public sector deficit rose sharply, from 3 percent of GNP to 5.4 percent, and the current account deficit widened to 8 percent of GNP (table 1.8). Despite the sharp jump in government investment expenditure, the Philippines was not able to ride out the second oil shock. Real growth rates dropped each year after 1979. Here the Philippines was in sharp contrast to neighboring Asian countries, which, although most suffered a terms of trade shock during the same period, were much more successful in restoring rates of economic growth and in generating exports (table 1.9). The dollar value of Philippine exports hit a peak in 1980 and then fell at an average rate of almost 5 percent per year through 1983, the result not only of weak international prices, but also falling commodity export volumes. Slower domestic growth and higher world real interest rates severely affected major domestic firms, many of them highly leveraged. A domestic financial crisis in 1981 brought about the failure of several large firms, many of which were bailed out by the government. Industrial failures continued to proliferate, leaving the government, which had guaranteed the foreign borrowings of many of these companies, with nonperforming assets with a book value in the billions of dollars. This in turn led to an increasing fiscal burden on the national government, as it was forced to absorb the losses of the two government-owned financial institutions, as well as of several government nonfinancial corporations. Philippine external borrowing accelerated in the early 1980s, and total foreign debt nearly doubled between 1979 and 1982. Borrowing increased under the pressure of a swollen current account deficit, but capital flight also accelerated sharply in the early 1980s, and may have reached 5 percent of GNP in 1981 and 1982. Net foreign direct investment inflows slowed to a trickle, as growing disinvestment offset direct investment inflows. The cautious borrowing policy of the 1970s disappeared in the early 1980s. The most abrupt change was the increasing use of short-term borrowing. This was particularly true of the public sector, which accounted for two-thirds of the increase in short-term debt outside the monetary sector. The central bank also borrowed heavily between 1980 and 1982 and encouraged other banks to do the same by providing swap arrangements.” Despite this borrowing, central bank reserves fell by $2 billion (two-thirds)

Philippine Macroeconomic Indicators (percentage of GNP)

Table 1.8

Real GNP (% change) Investment share GNP Government fixed investment National government budget Expenditure Revenue Surplus/deficit Consolidated nonfinancial public sector Investment Surplusideficit Current account balance MI (% change) Inflation rate (CPI) Nore: N.A.

=

not available.

1980

1981

1982

1983

1984

1985

1986

1987

1988

5.0 30.7 6.9

3.4 30.7 8.0

1.9 28.8 7.2

1.1 27.5 6.1

-7.1 19.2 4.1

-4.1 14.3 3.7

2.0 13.2 N.A.

5.9 15.4 N.A.

6.7 18.2 N.A.

14.4 13.1 -1.3

15.8 11.8 -4.0

15.7 11.4 -4.3

14.0 12.0 -2.0

12.7 10.8 -1.9

13.5 11.5 -1.9

17.8 12.8 -5.0

8.2 -3.0 -5.4 19.6 17.6

10.4 -5.7 -5.4 4.4 12.4

9.0 -5.4 -8.1 -0.1 10.4

8.2 -3.2 -8.1 38.3 10.0

6.9 -2.8 -3.5 3.5 50.3

6.4 -2.2 0.0 6.5 24.9

6.2 -3.6 3.3 19.0 0.7

17.0 14.6 -2.4

N.A. N.A. -1.6 22.2 3.8

N.A. N.A. N.A. N.A. N.A. -1.9 13.9 8.7

395

Philippines/Chapter 1 Comparative Growth Rates for Selected Asian Countries, 1974-84

Table 1.9

GDP Growth Rates 1974-79 Philippines Indonesia Malaysia Thailand Korea

6.5 6.9 7.1 7.8 9.7

1980

1981

5.2 9.9 7.8 5.8 -3.0

3.9 7.9 7.1 6.3 7.4

1982

Terms of Trade Shocka 1983

I984

I979-82

0.9 4.2 6.3 5.8 10.9

-6.0 5.8 7.6 6.2 8.6

-5.8 +35.9 - 8.9 -6.3 -3.8

1.6 -6.4 13.7 12.9 7.2

-1.4

2.9 2.2 5.6 4.1 5.7

Growth of Dollar Export Earnings Philippines Indonesia Malaysia Thailand Korea

11.9 15.8 21.3 15.6 29.5

28.0 43.0 19.5 28.7 15.6

7.6 11.9 -8.6 9.8 20.8

-7.1 -14.5 2.4 -2.0 4.0

11.4

17.7 -1.0 10.8

Source: Philippines: NEDA and the central bank. Others: Asian Development Bank, Key Indicators of Developing Members Countries. and IMF. International Financial Statistics. "Terms of trade shock equals percentage change in terms of trade multiplied by the share of merchandise imports in GNP.

from the end of 1980 to mid-1983, although this was not known at the time. By 1982 the share of short-term debt, including monetary sector debt, in total debt rose to 47 percent, a much higher share than in other LDC debtors. The Philippines first considered declaring a moratorium in late 1982. However, Marcos demurred, apparently unwilling to have the Philippines compared to Latin American debtors. When a moratorium was finally declared in October 1983, Philippine foreign exchange reserves were nearly exhausted. The adjustment period that followed was severe. Domestic industry was limited by the extreme shortage of foreign exchange, and capacity utilization rates below 40 percent were common. Investment fell by more than half. Per capita incomes fell back to their level of the mid-l970s, erasing the gains from the rapid growth period. And inflation soared to over 50 percent per year, only to be rapidly reduced through monetary policy so severe it forced many firms and several financial institutions to the wall. Although many factors were responsible for the election defeat of President Marcos in 1986, the wrenching adjustment process was an important contributor. By the end of 1986 the legacy of the martial law economic policy would be 1974 income levels, a foreign debt almost equal to GNP, and a fragile new democracy. The crucial question for the Philippines is how things could have changed so rapidly. How could a country that substantially raised its investment and growth rates, had transformed its export structure toward manufactured goods, and was often mentioned as one of the next generation of East Asian tiger economies, collapse so quickly in the space of four years?

396

Robert S. Dohner and Ponciano Intal, Jr.

1.11 The Role of External Shocks

The Philippines’ real income position and its ability to sustain its level of foreign indebtedness were diminished by the two oil price shocks and the accompanying industrial country recessions. The second oil price shock had a much more severe effect than the first. This time, income loss from the change in the terms of trade was larger, equivalent to 6.9 percent of GNP. In addition, the terms of trade deterioration was coupled with a rise in real interest rates of almost 12 percentage points, adding another 3.5 percent of GNP to the income loss. l 8 Thus, the external shock totaled about 10 percent of GNP, which was among the largest for major LDC debtors.

1.12 Philippine Policy and the Debt Crisis Although the Philippines was hit more severely by external shocks than most debtor countries, the debt crisis that occurred was not simply a result of the second oil price shock and the rise in world real interest rates. The Philippines had developed a borrowing momentum that could not be sustained, and the country would have eventually come to an external crisis even if the shocks of the 1980s had not been there to hurry along the process. There were two fundamental economic difficulties. First, the Philippines failed to develop self-sustaining growth that would have eased the burden of servicing its external debt. Second, the country failed to shift resources toward the traded goods sector, as was required both by its increasing debt burden and by its declining terms of trade. In more concrete terms, the problems were poor returns from investments, difficulties in mobilizing domestic resources to fund investment, and the maintenance of a trade regime that did not sufficiently encourage exports. In addition, the Marcos government created a political-economic environment that discouraged independent investment, led to capital flight, and eventually crippled much of the productive economy. 1.12.1 Investment Efficiency In retrospect there were several weaknesses in the economic growth that the Philippines achieved in the 1970s. The first was its heavy dependence on the flow of investment expenditure as a source of aggregate demand. As indicated in table 1.10, the expansion in the rate of fixed investment accounted for more than 40 percent of the increase in real domestic output between 1974 and 1979, while the rise in construction expenditure alone contributed almost 30 percent. Much of the increase in investment came from the public sector, despite a very small initial public sector investment share.

397

Table 1.10

PhilippineKhapter 1 Sources of Philippine Real GDP Growth, 1974-79

Personal consumption Government consumption Gross fixed investment of which: Construction Exports Imports GDP Memo: Private fixed investment Government fixed investment

Growth Rate

Contribution to GDP Increment

Share of 1974 GDP

5.0 3.5 13.3 21.4 9.2 8.5 6.5

52.2 5.3 41.8 27.7 23.6 - 26.5 100.0

69.0 10.5 17.7 6.2 15.7 19.4 100.0

10.4 24.8

25.9 15.9

14.8 2.9

Source: NEDA. National Accounts Section

There is nothing inherently wrong with expanded investment as a source of aggregate demand growth. But if it is to be the basis of a higher rate of secular income growth, the investment level must be maintained, and it must generate sufficiently high growth in other sectors of the economy. Although difficult to assemble and somewhat sketchy, the evidence here is that the efficiency of investment in the Philippines was lower than in surrounding countries and, in turn, the failure of investment to pay out was an important contributor to the debt crisis of the early 1980s. The simplest measure of the efficiency of investment in the aggregate is the ratio of the resulting growth rate to the amount of investment that takes place. This is normally represented by its reciprocal, the ratio of the share of investment in GDP to the GDP growth rate, known as the incremental capital output ratio, or ICOR. The lower a country’s ICOR, the smaller is the increase in the capital stock necessary to produce a given increase in output and, therefore, in a sense, the more efficient is investment. l 9 In table 1.11 we compare investment ratios, GDP growth rates, and ICORs for the Philippines and selected Asian developing countries. As is clear from the table, the Philippines has achieved a high rate of domestic investment, but has been less successful in translating that investment into economic growth. The table also shows an increase in the ICOR from the pre-oil shock period. This was not unique to the Philippines, but occurred in all countries listed in the table except Thailand. A significant part of this increase was caused by a shift of investment toward more capital-intensive industries in the 1970s, in part caused by higher oil prices.*’ We investigate in chapters 3 and 4 some of the reasons for the higher ICOR in the Philippines. Here we will simply stress its importance for the real income position of the country. Had the Philippines had the median ICOR of the countries in the table (Malaysia’s), its growth rate from 1974 to 1980 would have been 8.3 percent per year, and real GDP in 1980 would have been almost 12 percent higher.

398

Robert S. Dohner and Ponciano Intal, Jr.

Table 1.11

Comparative InvestmentlGrowth Rates Investment Rate”

Philippines Indonesia Malaysia Thailand Korea

GDP Growth Rate

Ratio (ICOR)

1967-72

1974-80

1967-72

1974-80

1967-72

1974-80

20.5 12.8 18.2 24.3 25.2

29.3 20.0 25.8 26.1 29.8

5.27 8.20 N.A. 6.46 10.0

6.26 7.42 7.26 7.48 7.67

3.88 1.56 N.A. 3.77 2.53

4.68 2.70

3.55 3.48 3.89

Sources: Philippines: NEDA, National Accounts Section. Others: IMF, International Financial Stafistics. Note: N.A. = not available.

%toss domestic capital formation as a percentage of GDP.

1.12.2 Resource Mobilization The second weakness of Philippine economic policy was the continued dependence on foreign borrowing to fund domestic investment. To maintain economic growth in the face of external recession, the Philippines increased government expenditure and borrowed abroad in 1975. But the current account deficit never narrowed and was still about 5 percent of GNP in 1979. By 1982, after the second oil shock, the deficit had risen to over 8 percent of GNP. In large part this continued deficit came from the inability of the Philippine government to close it budgetary gap. The expansion in the size and expenditure of the national government was not matched by a corresponding increase in revenue generation (see table 1.4). This was true despite a number of external program requirements to raise the government revenue share and repeated tax measures enacted by the Philippines. While the increase in the government budget deficit shown above in table 1.5 is not dramatic, it covers only the national government. Much of what took place on the fiscal side in the Philippines was the movement of government expenditure and government borrowing to the accounts of state-owned corporations. No consolidated figures exist before 1978, although the increase in the activities of the state-owned firms may be judged from the investment expenditures given in table 1.4. By 1978 the consolidated budget deficit of the nonfinancial public sector had reached 3 percent of GNP, and this deficit ballooned in the 1980s. 1.12.3 Trade and Exchange Rate Regime What now appears crucial to sustaining a large external debt is a concomitant expansion of export capacity. Here the Philippine record is highly mixed. Nontraditional manufactured exports grew rapidly and increased their share of total exports from 6 percent in 1970 to 50 percent by 1980. But the overall growth of exports was insufficient given the high investment, high foreign borrowing strategy the country pursued. The share of merchandise exports in GDP was nearly the same at the end of the decade

399

Philippineslchapter 1

as it was at the beginning. This was in sharp contrast to neighboring countries, where significant export deepening took place (see table 3.2 below). Both fiscal policy and trade and exchange rate policy played key roles in the continued borrowing of the Philippines and in the country’s inability to respond quickly enough to the abrupt change in world product and capital markets in the 1980s. But in addition to these two, there were other features of the Philippine business and policy environment that exacerbated the problems that the country had in maintaining growth and avoiding debt difficulties. These include ‘‘crony capitalism’ ’-government intervention and monopolization of domestic industry-as well as weakness of the financial system and weaknesses in the system of debt management and control over capital outflows.

1.12.4 The Martial Law Business Environment: Crony Capitalism An integral part of the operation of the Philippine economy during the period of rapid debt buildup was the development of crony capitalism-the fostering, through a variety of means, of a small group of Philippine businessmen, including the president and his family. This included the standard measures of awarding government contracts, padding expenses, and providing kickbacks. But crony capitalism went well beyond simple graft. The most important aspect was the creation of monopolies, either through direct intervention to control an industry or through granting exemptions or exclusive privileges to favored individuals. The corporate empires of the cronies were built on a mixture of corporate extortion and high financial leveraging. Outright expropriation was done only at the outset of martial law. Later, less visible pressure was brought to bear on profitable firms to sell out to Marcos family members or to cronies. The cronies borrowed heavily, either receiving funds directly from government-owned financial institutions or borrowing from the private market on the strength of their association with Marcos. With the decisively changed financial atmosphere in the 1980s, both in the Philippines and externally, the crony groups proved extremely vulnerable. Crony capitalism took a significant toll on the behavior of the private sector not associated with the Marcos government. Businessmen became less willing to invest and expand in the Philippines for fear of attracting attention and instead moved their money outside the country. By the early 1980s this movement had become a flood.

1.12.5 Financial Sector and Debt Management The Philippine financial sector played a number of supporting roles in the buildup to a debt crisis in the Philippines. The first was the failure of the system to mobilize sufficient resources in financial form. Despite the high rate of investment in the Philippines, the country had one of the lowest rates

400

Robert S. Dohner and Ponciano Intal, Jr.

of financial mobilization among East Asian countries. The second was the high degree of government participation in the financial sector and the extent to which financial flows were channeled to projects or individuals favored by the martial law government. But it was a series of financial crises starting in 1981 that accelerated the collapse of the Philippine economy and made the stabilization period more severe than it otherwise would have been. A collapse in the commercial paper market in 1981 led to the first round of business failures and the beginnings of the government’s expanding bailouts of the private sector. During the recession of 1984-85, credit to key sectors almost disappeared, forcing the collapse of many private firms and growers. By 1986, after years of gyrations, financial institutions in the Philippines were almost unwilling to do any intermediation. The debt management system of the Philippines also contributed to the crisis that the country faced in 1983. This system worked well in the 1970s, screening foreign borrowing requests, limiting total external borrowing, and refinancing existing loans when better terms were available. However, the debt management system broke down badly in the 1980s, leaving the country with one of the highest percentages of short-term borrowing among all LDCs and little or no foreign exchange reserves. The inability to control capital flight, particularly in the early 1980s, accelerated the speed with which the Philippine debt crisis amved.

1.13 Conclusion The following chapters take a more detailed look at each of the issues discussed in section 1.12. Public sector expenditure and revenue mobilization, including that of the government corporate sector, are examined in chapter 2. In chapter 3 we deal with the trade and industrial policy regime in the Philippines and investigate further the sluggish performance of much of the country’s industry. Crony capitalism and the .effect that it had on domestic economic performance is the subject of chapter 4. In chapter 5 we examine the financial system and its role in the crisis. The debt management system and the growth of capital flight are covered in chapter 6. After the debt moratorium was declared in October 1983, the Philippines went through a successful, but very severe, stabilization period. In chapter 7 we look at that adjustment and the negotiations that the Philippines had with the IMF and its external creditors. Finally, in chapter 8 we examine the first three years of the Aquino government and the prospects for the Philippines.

401

2

Philippines/Chapter2

Government Expenditure and Revenues

As described in the previous chapter, the public sector in the Philippines has historically been rather small. Government expenditure averaged 10 to 11 percent of GNP, with total government revenues somewhat less than this amount. The developmental role of the public sector was minimal; most government outlays were for recurrent expenses, and only about 2 percent of GNP was spent on public investment. Under relative neglect, much of Philippine infrastructure deteriorated in the generation after independence. This was particularly true of the road network and of almost all infrastructure outside of Manila and Central Luzon. To some extent this small scale of government was the product of ideology inherited from the United States, but funding had always been a more concrete obstacle to an increased public sector role in the Philippines. Resistance to revenue measures in Congress limited the expansion of government expenditure. Sluggish economic growth during the import decontrol period in the early 1960s led to more support in the Philippines for an activist government. Ferdinand Marcos drew on this sentiment in his election campaign in 1965, and he greatly increased infrastructure investments during his first term. But this first term was only a precursor for what was to happen after 1972. Martial law transformed the character of the state, greatly increasing its expenditure, regulatory, and allocational role. Although security concerns were the pretext for the declaration of martial law in 1972, more rapid economic growth quickly became the regime's claim to legitimacy. Greatly increased public expenditure, particularly public investment, was the centerpiece of the government program. With Congress no longer functioning and Marcos able to rule by decree, the way was cleared to both raise domestic revenue in support of public expenditure and solicit outside sources of finance. The previous inactivity and small scale of the Philippine government made the country an attractive candidate for external project support under a modernizing government. As described below, the country was very successful in raising external funds for investment. Philippine statistics make it difficult to document the growth of the public sector in the early martial law years. Public finance data before 1975 cover only spending obligations, not actual outlays. Since the budget was subject to frequent revision, obligations and outlays often differed substantially. Budget obligations by category for 1969 through 1975 are shown in table 2.1. Actual outlays of the national government for 1975 onward are given in table 2.2. Since government expenditure was rising in the early years of martial law, obligations ran ahead of actual outlays. But the actual growth of expenditure

'

402 Table 2.1

Robert S. Dohner and Ponciano Intal, Jr. National Government Budget, Obligation Basis, 1%9-75 (percentage of GNP)

Current expenditure Administration General government Justice, police National defense Economic services Agriculture, resources Commerce, industry Transport, communications Other development Social services Debt service Subsidies, transfers Capital expenditure Administration Economic services Agriculture, resources Commerce, industry Transport, communications Other development Social services Total expenditures

FY 1969

FY 1970

FY 1972

FY 1974

FY 1975

8.9 2.5 0.9 0.6 1.1 1.7 0.5 0.2 0.8 0.2 3.3 0.8 0.6 1.6 0.1 1.3 0.3 0.8 0.0 0.2 0.2 10.6

8.2 2.6

0.5

7.8 2.5 0.8 0.7

1.1

1 .o

1.5 0.5 0.1 0.6 0.4 3.2 0.6 0.2 1.8 0.1 1.5 0.1 1.1 0.0 0.2 0.2 9.9

1.3 0.4 0.1 0.6 0.2 3.0 0.6 0.3 2.2 0.1 2.0 0.5 0.8 0.0 0.7 0. I 10.0

7.9 2.4 0.6 0.1 1.7 1.9 0.6 0.2 0.7 0.3 2.4 0.6 0.7 5.6 0.1 5.3 2.0 1.4 0.3 1.6 0.2 13.6

10.2 3.1 0.7 0.2 2.2 1.9 0.8 0.1 0.7 0.3 2.6

1 .o

0.8

1.7 5.0 0. I 4.5 1.3 2. I 0.2 1 .o

0.4 15. I

Source: World Bank (1976, 547-48). Nore: Fiscal year from July I to June 30.

was substantial. National accounts figures show a 15 percent per year growth in government expenditure from 1972 to 1975, with an almost 30 percent per year growth in total fixed investment outlays, including those by government corporations. By 1976 total government expenditure had risen to 15 percent of GNP and investment expenditures had climbed to 6.6 percent of GNP (table 2.3). The change in priorities is clear from shifts in the obligational budget in the early years of martial law. Within current expenditures there is a huge increase in commitments for defense, which rose from 13 percent of the budget in FY 1972 to almost 22 percent by FY 1975 (see table 2.1).2 Subsidies also increased rapidly as the government sought to limit the domestic price rises of grains, petroleum products, and fertilizer after the first oil shock.3 The more activist role of the government is also reflected in a shift in current expenditures away from social welfare (education and health) and toward economic services, particularly agriculture and transportation. The shift in investment priorities in the early years of martial law is shown in table 2.1. Increased investment expenditures went into transportation (roads, highways, and ports), imgation, and power and rural electrification.

Table 2.2

National Government Expenditures (percentage of GNP)

Expenditures Current Personnel Maintenance, operations Interest Transfers to local government Transfers to GCs Capital Infrastructure Other capital outlays Equity and net lending Equity contributions Net lending (Assistance to GFIs) Revenues Budger surplusldef cit

I975

1976

1977

I978

1979

1980

1981

1982

1983

1984

1985

1986

1987

16.0 12.8

1.4 1.3 0.1

14.8 10.9 4.2 4.1 0.6 0.5 0.4 2.6 2.1 0.4 1.4 1.3 0.1

13.7 9.5 3.8 3.7 0.8 0.6 0.2 2.3 2.0 0.3 1.9 1.6 0.4

14.4 9.3 3.5 3.7 0.9 0.6 0.2 3.1 2.8 0.4 2.0 1.7 0.3

14.4 -1.6

13.4 -1.8

13.0 -1.9

13.6 -1.2

13.5 -0.2

13.1 -1.3

15.8 8.7 3.5 3.7 0.8 0.5 0.2 4.2 3.3 0.9 3.0 2.7 0.3 0.2 11.8 -4.0

15.7 9.2 3.2 3.7 1.1 0.7 0.6 3.0 2.2 0.7

0.6 0.5 0.1

14.9 11.6 10.3 4.1 0.6 0.5 0.2 1.8 1.5 0.3 1.5 1.5 0.0

14.0 9.1 3.7 3.1 1.3 0.7 0.3 2.8 1.8 0.9 2.1 1.5 0.6 0.0 12.0 -2.0

12.7 8.1 3.2 2.4 2.0 0.5 0.1 1.9 1.2 0.7 2.7 1.9 0.7 1.6 10.8 -1.9

13.5 9.1 3.9 2.2 2.5 0.6 0.2 1.7 0.9 0.7 2.8 2.4 0.3 1.9 11.5 -1.9

17.8 10.8 4.0 2.4

0.2 2.6

15.2 11.8 10.4 5.3 0.5 0.5 0.3 2.0

17.0 13.5 4.6 2.7 5.2 0.6 0.3 1.8 1.0 0.8 1.6 0.6 1.0 0.1 14.6 -2.4

0.1

Source: Philippines Bureau of the Treasury, Cash Operarions Statement. Note: GCs are government corporations. GFIs are government financial institutions.

3.5 2.8 0.7 0.4 11.4 -4.3

3.5 0.6 0.3 1.9 1.3 0.6 4.4 2.0 2.4 2.6 12.8 -5.0

Table 2.3

Fixed investment: Government General government Public corporations Private Total Memo: Total government expenditure

Fixed Investment (percentage of GNP) 1970

1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

1.2

2.1

2.3

3.4

4.3

6.6

6.9

7.2

7.3

6.9

8.0

7.2

6.1

4.1

3.7

N.A.

N.A.

1.1

1.7

2.1

2.9

3.8

3.3

3.4

2.6

3.3

2.4

2.0

2.5

2.8

1.4

2.0

N.A.

N.A.

0.1 15.9 17.2

0.5 14.4 16.5

0.2

0.6 19.4 23.7

3.3 18.5 25.1

3.5 16.9 23.8

4.6 16.7 23.9

4.0 18.5 25.8

4.5 18.8 25.7

1X.1

3.3 19.0 25.1

2.7 16.0 20.1

1.7 11.5 15.1

N.A. N.A.

26.1

4.6 18.5 25.6

N.A. N.A.

15.8

0.5 15.1 18.5

6.0

13.6

13.2

14.6

9.9

11.5

10.9

12.3

16.0

15.2

14.9

14.8

13.7

14.4

15.8

15.7

14.0

12.7

13.5

17.8

17.0

Source: NEDA. National Accounts Section

405

Philippinesichapter 2

The most dramatic change in the budgetary process came in the emphasis on foreign funding. The Philippine government improved its planning and proposal mechanism and solicited funds from multilateral and bilateral sources. The Philippines met with considerable success in this effort; official aid flows more than doubled after 1972, and development loans rose from a total during the 1960s of $338 million to $1.6 billion from 1970 to May 1976.4 Much of these funds flowed through publicly owned corporations. To increase the ability of these corporations to borrow, the government raised their equity base through capital infusions, and this is reflected in the sharp rise in equity contribution in the middle 1 9 7 0 ~ . ~

2.1 After the First Oil Shock, 1975-79 The first oil shock and the subsequent collapse of commodity prices in 1975 had little effect on the Philippine government’s expenditure plans. Martial law politics were tied up in a more activist and expanded national government, and the Philippines found sufficient external support to continue with its public expenditure program. The share of national government expenditures in GNP actually peaked in 1975 and remained near 15 percent for the rest of the decade. While total expenditures stabilized, public investment continued to increase relative to GNP through a striking realignment of shares of current and capital expenditures. By 1979 national government current expenditures as a share of GNP had fallen by three percentage points (see table 2.2). In addition, investment expenditure by publicly owned corporations, not reflected in the published budget, continued to increase. By the end of the decade most of the government’s infrastructure investment program was conducted through these corporations. The growth of government corporations is discussed in more detail below, but their importance in public investment expenditures is easily seen from national accounts data. From half a percent of GNP in 1974 and 1975, investment by public corporations had risen to 4 percent of GNP by 1978 and 1979 (see table 2.3). Funding for these corporations came primarily from foreign loans and from equity contributions and net lending from the national government, which took up an increasing share of national government capital outlays. Data on the sectoral allocation of public investments comes from two sources. The first is data from NEDA on its infrastructure program (table 2.4). This covers the bulk, but not all of public investment. The second breakdown comes from estimates made by a World Bank mission in 1983 and is shown in table 2.5. Both sources show an increasing concentration of public investment in infrastructure during the 1970s. Transportation and irrigation diminish somewhat as a share of investment and GNP, while investment in the energy sector increases greatly in importance. By the late 1970s half of public fixed investment was in energy projects.

406

Table 2.4

Robert S. Dohner and Ponciano Intal, Jr. NEDA Infrastructure Program (in millions of pesos)

Total infrastructure Agriculture Water supply Transportation Power Social services National government Government corporations

1975

I976

1977

1978

1979

1980

1981

1982

4.0 0.7 0.1 1.6 0.9 0.2 2.8 1.2

4.9 0.4

5.2 0.5 0.1 1.5 2.4 0.2 2.6 2.6

4.1 0.5 0.2 1.0 1.9 0.2 2.0 2.2

5.0 0.8 0.1 1.0 2.6 0.1 2.0 3.0

4.4 0.6 0.2 0.8 2.4 0.2 1.7 2.7

4.6 0.6 0.3 1.0 2.3 0.2 1.9 2.7

4.9 0.6 0.4 1.3 2.3 0.2 2.1 2.8

0.2 1.6 1.5 0.4 3.0 1.9

Source: World Bank (1979), table 5.12, and World Bank (1984b), table 1.12a.

Table 2.5

Public Fixed Investment by Sector, 1978-83 (percentage of GNP)

Economic services Agriculture, forestq Industry, trade, tourism Energy Transportation Water supply Other infrastructure Social services Defense General administration Local government Total public fixed investment of which: National government Government corporations

1978

1979

1980

198I

1982

1983a

5.7 0.8

5.5 0.6 0. I 3.3 0.9 0.2 0.5 0.5 0.2 0.1 0.2

8.0 0.7 0.6 2.9 2.2 0.5 I .o 0.6 0.2 0.2 0.2 9.2

7.3 0.9 0.6 2.6 2.0 0.5 0.7

6.5

7.0 0.9 0.9 2.5 1.8 0.3 0.6 0.6 0.1 0.1 0.2 8.1

0.1 0.2 0.2 8.2

6.8 0.9 0.7 2.8 1.4 0.5 0.5 0.7 0.1 0.1 0.2 7.9

2.3 4.0

3.2 4.7

4.2 4.8

3.0 5.0

2.6 5.1

0. I

3. I 0.9 0. I 0.6 0.4 0.2 0.1 0.2 6.6 2.4 4.0

0.5

Source: World Bank (1984b). table 1.16. "Figures for I983 are estimated.

The Philippine energy investment program was an ambitious plan to substitute domestic and other energy sources for imported oil, provide for additional capacity for economic growth, and extend electrification to rural areas. The Philippines, being mountainous and volcanic, offered a variety of possible sources. Hydroelectric energy was developed and by 1979 accounted for 21 percent of total electricity generation. By the 1980s the Philippines had one of the world's most extensive systems of geothermal energy. A small domestic oil field was developed, which in 1979 produced 10 percent of the total oil supply, and Philippine coal reserves offered large potential for development. However, the bulk of Philippine energy investments went into electricity generation. The institution through which these investments were made was the National Power Corporation (NPC), a government-owned corporation, which accounted for almost 80 percent

407

PhilippinesKhapter 2

of the energy investment program by the end of the 1970s. Between 1977 and 1982, NPC’s installed generating capacity rose from 1,038 to 4,323 megawatts. One particular investment, the Bataan nuclear power project, is worth describing in more detail, both because of its size and foreign loan exposure, and because of the controversy that surrounds it. The Philippine government made the decision to build a nuclear power plant in 1973, and by early 1974 intensive negotiations were underway with General Electric for its construction. General Electric’s proposal was the construction of two 600 megawatt reactors for a total price of approximately $500 million. The Westinghouse Corporation entered the competition for the plant at a late stage, but hired Herminio Disini, a presidential associate, to arrange a meeting with President Marcos and his cabinet in May 1974. At that meeting Westinghouse made an offer to build a complex similar to the GE proposal for about the same price. At the end of May, Marcos instructed NPC to sign a contract with Westinghouse, despite the fact that Westinghouse had not yet submitted a detailed proposal.’ By the time the contract was signed in February 1976 the proposal from Westinghouse was to build a single 626 megawatt reactor for a price of $1.1 billion, or roughly quadruple the cost per megawatt of the original GE proposal. The project was supported by a loan and additional loan guarantees from the Export Import Bank of the United States. The Westinghouse plant had a checkered history. The site approval process went on longer than anticipated because the plant was near an active volcano, there was vociferous domestic opposition to the plant, and construction was halted for eighteen months after the Three Mile Island accident in the United States in 1979. The renegotiated contract with Westinghouse, including inflation, delay costs, and capitalized interest, was for $2.2 billion, making the nuclear power plant alone responsible for almost one-tenth of Philippine external debt. Although almost finished, the plant never opened. Shortly after the Soviet Union’s Chernobyl accident in 1986, President Aquino announced that the plant would not be put into operation.’ In addition to NPC, two other government-owned corporations have played an important role in the energy investment program. The Philippine National Oil Company (PNOC) has been responsible for domestic oil and coal development, as well as the geothermal energy program. The National Electrification Administration (NEA) extended loans to rural cooperatives to finance extensions of the electricity distribution system and to construct small-scale electricity generating facilities. By the end of the decade total expenditure by the national government and public corporations exceeded 18 percent of GNP. Despite continued efforts, described below, to raise tax collections, national government revenues remained at about 13.5 percent of GNP, leaving a substantial gap to be closed by borrowing.’ Foreign loans provided much of the funding for the

408

Robert S. Dohner and Ponciano Intal, Jr.

investment projects that the Philippines undertook; in fact, foreign-assisted projects made up three-quarters of Philippine infrastructural investment by 1979. But foreign project loans had requirements for domestic counterpart funds, funds that the national government (or a public corporation) would put up for its participation in the project. By 1978 the Philippine government was having difficulty generating sufficient peso funds to match the counterpart requirements of foreign-assisted projects. Part of the response of the government to revenue and funding shortfalls was to squeeze the current expenditure portion of the budget. In many cases this meant a reduction in operating and maintenance expenditures for the existing capital stock, despite its increase as a result of the investment program.

2.2 After the Second Oil Shock, 1980-83 A second leap in public expenditure occurred in the wake of the second oil price shock. After a reduction in 1979, real national government expenditure rose by 13 percent in 1980 and by an additional 12 percent in 1981. Once again the Philippine authorities had chosen to counteract the external shock with expansionary policies. However, the domestic recession was worse this time than it had been in the 1970s, and the Philippine industrial sector was particularly hard hit. Government revenues also dropped unexpectedly sharply, falling by 2 percent of GNP between 1979 and 1981. As a result, the national government budget, which was nearly balanced in 1979, was in deficit by over 4 percent of GNP in 1981. AS in the 1970s, the increase in outlays was almost entirely a rise in government investment expenditure (see table 2.2). Real capital spending of the national government nearly doubled in two years. Government-owned corporations also increased their investment expenditure, so that total public sector fixed investment rose from 6.5 percent of GNP in 1979 to over 9 percent in 1981 (see table 2.5). The investment program was similar to that of the 1970s. The largest share of public investment went to energy, with transportation a close second. The Philippine government had prepared an ambitious ten-year energy plan, which was published in January 1980. President Marcos then ordered an accelerated energy program to compress most of the goals of that plan into five years. The NPC developed an accelerated program for conversion of oil-fired plants to other energy sources, against the criticism of the World Bank and Asian Development Bank, and was proposing new projects as late as 1983.'' But the 1980s also marked a shift in Philippine public investment strategy toward a greater emphasis on industrial development. One aspect of this policy was continued investment in, and expansion of, the country's export processing zones. The second was the Philippine government's Major Industrial Projects (MIPS). The MIPS were an elaborate and expensive

409

PhilippinesKhapter 2

thrust into secondary import substitution through heavy industrialization. The goals were to deepen Philippine industrial structure, reduce the heavy dependence of domestic industry on imported materials, and establish growth poles away from Manila. The eleven projects are listed in table 2.6. The decision to go ahead with the MIPS was made in 1979, after several years of discussion in the Philippines. The minister of industry, Roberto Ongpin, was a strong proponent of the projects and pushed actively for them within the cabinet. In deciding in favor of Ongpin and the MIPS, the Philippine cabinet was influenced by the heavy industrialization thrust of policy in many East Asian countries during this period. Korea and Taiwan had adopted extensive efforts to encourage heavy industry, but neighboring countries in Southeast Asia-Malaysia, Thailand, and Singapore-were also supporting investments in petrochemicals, aluminum smelting, pulp and paper, and other heavy industries." The government's lead agency for this industrialization effort was the reactivated National Development Corporation (NDC), which was attached to the Ministry of Trade and Industry, headed by Ongpin. The projected cost of the original proposals was $6 billion, but after critical external reviews, the projects were scaled back and the estimated Table 2.6

Major Industrial Projects: Schedules, Costs, Financing (in millions of U.S. dollars) Financing Start of Project/ Operations

Completed projects: Copper smelter Diesel engine manufacturing Cement industry coal conversion Phosphatic fertilizer Coconut-based chemicals Started. but deferred projects: Heavy engineering Integrated steel plant

Private Equity Project Cost"

1980183 1980183

344 9

1981183 1981184 1982184

37 513 116

Subtotal Deferred projects: Aluminum smelter Integrated pulp and paper Petrochemical complex Alcohol-gasoline facility Total

1982 b

Government Equity

63

Foreign

37 4

60

60

40 4

23 1,806 2,848 -

650 250

1

,ooo

4,748

Source: World Bank (1984b), table 4. I . "Project cost includes fixed assets and interest during construction bSite development started in 1983.

Domestic

Foreign Loans

244 5

37 413

52 23

I61 290 -

60 -

-

85

1.639 2,413 -

410

Robert S. Dohner and Ponciano Intal, Jr.

cost was reduced to about $4 billion. The financing strategy that the Philippines adopted was to encourage foreign equity participation to cover as much of the costs as possible and to rely on foreign loans for the remainder, to minimize the use of domestic resources in the projects (IMF 1984b). All but one of the projects was to have some government participation and ownership. The IMF, World Bank, and other outside agencies were critical of the MIPS, as is carefully phrased in their reports on the Philippines. The Philippine government was committed under the first Structural Adjustment Program with the World Bank to rationalize industrial investment incentives and reduce their bias toward capital intensity. Yet the MIPS represented a huge investment in highly capital-intensive industry. In a country in which the average investment per industrial worker was about $20,000, the cost of completing five of the initial projects was estimated to be $3.4 billion, or $500,000 per job created (Callison 1981, 24). The financial and operational history of the MIPS is also shown in table 2.6. Much of the program was overtaken by events in the 1980s and the difficulties in raising external finance. In addition, foreign equity participation never lived up to expectations, resulting in a larger government participation. Five of the projects were completed, two were started but deferred in 1983 as a result of the debt moratorium, and the remaining four were deferred at an earlier stage, before contracting took place. The expenditure on the seven projects that went forward amounted to $2.8 billion, almost entirely financed through foreign loans. In addition to increased investment expenditure on its own account, national government equity contributions and net lending rose sharply during this period, from 2 percent of GNP in 1980 to 3.5 percent by 1982. Some of this increase was government contribution to the planned investment program. However, two additional requirements emerged during these years. The first was the greatly expanded program of the Ministry of Human Settlements, headed by Marcos’ wife, Imelda. Only limited information exists on the activities of the Ministry. Originally its efforts were directed toward housing construction and finance, but these rapidly expanded so that its work paralleled much of the efforts of other ministries. Much of the activity of the Ministry and most of its expenditure were done through public corporations and their subsidiaries. Nineteen government corporations were attached to the Ministry for Human Settlements, with at least thirty-six additional subsidiaries one level down.I2 The Ministry and its attached corporations were active in housing, food distribution, area development, finance, energy, public utilities, hotels, industry, cultural affairs, and health services. Through the Home Development Mutual Fund (the PAG-IBIG fund), the Ministry levied a payroll tax in support of housing. The Human Settlements Development Corporation alone received P. 1.1 billion in transfers from the national government, making it the ninth largest corporate recipient. l 3

411

PhilippinesKhapter 2

But there were other, more important reasons for the rapid rise in government equity contributions and net lending. The domestic financial crisis in 1981 led to the failure of numerous large and highly leveraged companies, many of them owned by presidential cronies. The Philippine government organized rescue operations for these firms, acting through the NDC and through government financial institutions, supporting the resulting deficits with equity transfers from the government. The greatly increased investment program of other public corporations and their almost nonexistent ability to generate funds internally led to increased equity transfers from the national government to meet the peso counterpart requirements of project loans as well as to provide finance. As the world and domestic recessions continued into 1982 and external debt continued to mount rapidly, the Philippine government tried to change course. National government capital expenditure on its own account fell by 26 percent in real terms in 1982. But other claims on the government continued to increase. Control over the investment programs of state-owned corporations was weak, and their investment expenditure continued to climb, rising to 5 percent of GNP (see table 2.5). As a result of this and the corporate rescue operations mounted by the government, equity contributions and net lending rose by 17 percent in real terms in 1982, enough to keep the share of national government expenditure in GNP almost constant. The shift in government priorities in the 1970s and later, the almost desperate attempts to keep capital inflows going, show up clearly in the continuing pressure on outlays for current operations. Government current expenditures, after dropping in the late 1970s, held at a little over 9 percent of GNP until 1983 (see table 2.2). However, this figure is deceptive, since interest payments and transfers took up a much larger share of the total. Real wages of government employees fell between 1979 and 1983, while operations and maintenance expenditures were 14 percent lower in real terms in 1983 than in 1978, despite a greatly increased government capital stock (World Bank 1984b, 13). The Philippines continued to initiate new projects with foreign financing during this period, even as it had increasing difficulty in providing the required level of counterpart funds for existing projects. The reaction was to delay the implementation of existing projects and to squeeze current expenditures, particularly for operations and maintenance, in order to sustain the momentum of the investment program and maintain foreign capital inflow. By the early 1980s, operations and maintenance expenditures had been reduced to such an extent that the capital stock was deteriorating prematurely; the Philippines was in effect consuming its existing capital stock in order to maintain investment. Ironically, despite the abolition of Congress and the decree-making powers of martial law, Philippine public expenditure continued to be constrained by the inability to raise government revenues. This was particularly true of the early 1980s, forcing the delay in existing projects, the compression of current expenditures, and the increasing resort to foreign loans, in many

412

Robert S. Dohner and Ponciano Intal, Jr.

cases at short term. The sluggishness of domestic revenue generation is the next topic we examine.

2.3 National Government Revenues The level of taxation in the Philippines has historically been low, consistent with the country's relatively small government expenditure. International comparisons of tax effort place the Philippines near the bottom among less developed countries, both in terms of taxes collected as a share of GNP and the rates of effective taxation of the existing tax base.14 Before martial law, the Philippine Congress had been a consistent obstacle to increased taxation or to tax reform measures. Dominated by the wealthy landowning class, the Congress viewed limited taxation as consistent with its own interests and as a way of limiting the power of the e x e c ~ t i v e . ' ~ There was widely expressed hope under martial law that the national government would become more effective in mobilizing domestic tax revenue as a means of financing public development expenditure. Recommendations for raising revenue generation were contained in reports by the International Labour Office mission (ILO 1974) and in an extensive country report prepared by the World Bank in 1975 (World Bank 1976). There were ambitious targets in the 1974-77 development plan to raise domestic revenue generation to 17 percent of GNP, and revenue targets were conditions included in the extended facility drawing from the IMF in 1976. The government had considerable success in raising revenues during the first few years of martial law. The administration quickly moved on several reform proposals that had long been pending in Congress. One of the first presidential decrees was a tariff reform program that reduced the number of rate categories, set a minimum tariff rate of 10 percent, and eliminated duty exemptions for public bodies. Other decrees revised the individual income tax and the system of property taxation. l6 The tariff revisions and the elimination of exemptions, coupled with the rise in external prices, resulted in a large increase in tariff revenues, nearly doubling the share of import taxes in GNP. Taxation of exports also increased during this early period. The export taxes that had been included in the 1970 stabilization program as temporary measures were made permanent in 1973. In early 1974 the government sought to capture some of the windfall gains from the rise in international commodity prices. A base price was set at 80 percent of the February 1974 price for a variety of commodity exports, and taxes of 20 to 30 percent were applied to the excess of current prices above this base. The government also announced more vigorous enforcement of existing direct taxes on individuals and corporations, along with threats of severe penalties for tax evaders. The number of corporations and individuals filing tax returns increased sharply in 1973, resulting in much higher tax collections.

413

PhilippineKhapter 2

The result of these efforts was an increase in the revenue mobilization of the national government from 10 percent of GNP in the early 1970s to over 14 percent in 1975 (table 2.7). In the years that followed there were tax packages with revenue-enhancing measures introduced almost every year, including significant taxes on crude oil and petroleum products, revisions in domestic sales taxes, and some increases in business taxes. Even with these continued efforts and the targets that had been set in the development plan and in external loans, the share of tax revenue in GNP remained constant until 1980. Despite all of its running, the government had just managed to stay in place and had come nowhere near its revenue targets. The reasons behind this sluggishness of revenues point out some of the difficulties of the Philippine tax system and some fundamental problems of the martial law regime. The increase in revenues up to 1975 had been heavily dependent on increased taxes on international trade, and the share of international trade taxes in total tax revenue rose from one-third to almost one-half by 1975. The premium duty system, which was designed to capture price windfalls, led to greatly reduced collections from export taxes in 1975-77, when world commodity prices fell and the taxable premium disappeared. Other early martial law tax measures represented only temporary gains. The tax amnesty brought in a significant amount of revenue, but the initial caution that had inspired additional filings in 1973 subsided, returning direct tax collections to more normal levels.” Part of the revenue problem of the Philippines was the high reliance on indirect taxes-taxes on trade and on domestic sales-that had low income elasticities and therefore required continuing additional measures to keep up with the growth of GNP. But in addition to low elasticities, the Philippines also faced problems of erosion of the tax base and difficulties in administering the existing tax system. The corporate income tax is a case in point. Despite the increasing importance of the organized corporate sector, collections from direct taxes on corporations fell sharply as a percent of GNP in the remainder of the 1970s. Here the problem was mainly the erosion of the existing tax base. Fiscal incentives for industrialization had been a feature of Philippine policy since independence, but the Investment Incentives Act of 1967 and the Export Incentives Act of 1970 provided more extensive fiscal tools to channel resources. Each act allowed accelerated depreciation, tax deductions for expansion reinvestments, tax credits for domestic capital equipment, and exemptions from selected business taxes. During the martial law period these incentives were liberalized by presidential decree and were extended to other industries, including agriculture. In addition to these general incentives, there were a large number of special incentives for particular industries granted by legislation before, and by presidential decree after, martial law. The most important special incentives covered cottage industry, chemical fertilizers, mining, textiles, shipping and shipbuilding, tourism, and

’*

Table 2.7

National Government Revenues (percentage of GNP) 1970

1972

1973

1974

1975

1976

1978

1979

1980

1981

1982

1983

1984

1985

1986

1987

Income and profit

2.5

L6

2

3 . '

3.0 -

2.7 -

3.1 -

2.8 -

2.8 -

2.6 -

2.5 -

2.3 -

2.3 -

3. I -

3.1 -

3.1 -

Personal Tax amnesty Corporate Other

0.7 0.0 1.7 0.2

0.8 0.0 1.6 0.2

0.8 0.8 1.9 0.1

0.6 0.3 2.1 0.1

0.8 0.2 2.0 0. I

1.1 0.0 1.6 0.0

1.9 0.0 I .2 0.0

1.5 0.0 1.3 0.0

1.3 0.0 1.5 0.0

I .3 0.0 1.2 0.0

I .2 0.0 1.3 0.0

1 .O

N.A. 0.0 N.A. N.A.

N.A.

0.0 1.3 0.0

0.9 0.0 1.5 0.0

N.A. N.A.

1.1 0.0 I .4 0.6

-

-

3.8

3.s

4.9

5.7 -

-

-

4.6 -

4.4 -

4.3 -

4.4 -

3.7 -

3.7 -

4.3 -

3.3 -

3.1 -

2.9 -

3.7 -

3.3 I .4 1 .o

4.2 0.5 0.0

4.2 0.2 0.0

4.0 0.3 0.0

4.2 0.2 0.0

3.6 0.1 0.0

3.6 0.1 0.0

4.3 0.1 0.0

3.0 0.3 0.1

2.7 0.2 0.2

2.7 0.1 0.1

3.7 0.0 0.0

-

-

-

3.4 -

3.7 -

4.5 -

4.1 -

3.8 -

3.7 -

3.5 -

3.6 -

3.8 -

4.3 -

5. I -

1.2 2.2

I.4 2.3

1.9 2.6

1.9 2.2

1.8 2.1

1.7 2.0

1.5 2.0

1.4 2.2

1.4 2.3

1.7 2.6

I .9 3.2

0.3 -

0.3 -

0.3 -

10.2 I .3

10.6 2.2

12.2 2.4

International trade

2.7

Import duties Export taxes Sales taxes imports

1.5 0.3 1.0

1.9 0.9 1.0

2.0 0.6 0.9

2.8

-

-

-

Domestic goodsiservices

2.6

2

1.9

2.3

2.5 -

1.2 1.4

1.1 1.2

0.9 1.0

1.1

1 .O

1.2

1.5

-

-

-

-

General sales Excise tax

1.1

1.0

0.0

-

Other taxes

yL70.70.7

0.5 -

0.7 -

0.4 -

0.3 -

0.3 -

0.3 -

0.3 -

0.3 -

0.3 -

Total tax revenue Nontax revenue

8.4 0.0

9.3 0.0

9.8 0.0

10.9 0.0

11.7 2.7

11.4 2. I

11.6

2. I

11.9 1.6

11.5 1.6

10.4 1.5

10. I 1.3

10.4 1.6

9.5 I .3

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

8.4

9.3

9.8

10.9

14.4

13.4

13.6

13.5

13.1

11.8

11.4

12.0

10.8

11.5

12.8

Total revenue

-

Source: Philippine Bureau of the Treasury, Cash Operations Sraremenrs. Note: N.A. = not available. Components may not add to totals due to rounding.

-

14.6

415

Philippines/Chapter 2

overseas construction. Many of these industries were of particular importance to industrial groups or individuals associated with the martial law regime. In addition to the special industrial incentives, there were a large number of presidential decrees that benefitted particular firms, granting either exemptions from import duties on products or exemptions from certain taxes. The result was a corporate tax system that was highly complex, with an array of deductions and loopholes, and a low effective rate of collection. A World Bank mission in 1979 calculated that nearly 45 percent of taxable corporations were exempt from paying taxes and that deductions claimed by corporations amounted to about 70 percent of gross income (World Bank 1979, 12). The Philippine tax system flags after 1980, just at a time when countercyclical and corporate rescue outlays were swelling the expenditure side of the budget. Total national government revenue fell from 13.1 percent of GNP in 1980 to 11.4 percent in 1982, recovered slightly in 1983, and then fell sharply in 1984. Almost every category of taxes fell as a percentage of GNP. Part of this reduction is easily explained. The weak economy and corporate distress in the early 1980s depressed collections from a number of taxes. But in addition, tax collections fell as a result of several policy changes. As a part of the first structural adjustment loan (SAL) from the World Bank, the Philippines initiated a tariff reduction and trade liberalization program starting in 1981. As a result, tariff collections fell from 24.5 percent of total imports in 1980 to 18.6 percent in 1982, and then to 14.5 percent by 1984.19 The effect of the tariff reductions was counteracted by an import surcharge that reached 10 percent by 1984 and also by a foreign exchange tax of 10 percent that was briefly in force between June and October 1984. The substantial drop in individual income tax collections was more disturbing. Nominal individual income tax collections fell in 1982 and 1983, and their share in GNP fell sharply. Between 1979 and 1984 the individual income tax registered a buoyancy of only 0.3, extraordinarily low for this type of tax.” There were two reasons for the drop in income tax collections. First, there was a major revision of individual income tax rules in 1982 that substituted separate schedules for different types of income for the global income tax system that preceded it. Other reforms adopted a modified gross income tax system for individual income, eliminating many deductions and reducing tax rates. These reforms were adopted in order to simplify the tax system and increase the effectiveness of tax administration. While the reforms were designed to be revenue neutral, in fact they resulted in a substantial drop in individual tax collections. The reforms also shifted the tax burden away from the wealthiest taxpayers. A study by the Philippine’s Bureau of Internal

416

Robert S. Dohner and Ponciano Intal, Jr.

Revenue (BIR) of taxpayers with over P. 1 million in income found the average tax rate declined from 48 to 32 percent for those who relied entirely on compensation income, and from 29 to 9 percent for those with multiple schedules, between 1981 and 1982.21 The second reason for the drop in personal income tax collections is an apparent increase in the amount of tax evasion. Little evidence exists on the trend of tax evasion, although the view of increasing evasion is widely held. The level of tax evasion was quite large, however. An IMF mission conservatively estimated that the individual income tax receipts in 1984 should have exceeded P. 10 billion, based on current rates and the distribution of household income, while actual collections were only P. 4.5 billion.22

2.4 Failure of Taxation Efforts Despite the professed intent in Philippine development plans and external pressure for increased revenue generation through the tax system, the Philippines was not successful in raising government revenues. The fall in tax collections during the early 1980s was an important factor in the country’s balance of payments crisis and external debt difficulties. In part, the failure to raise the revenue share reflects the structure of the Philippine tax system, Much of the tax system depended on indirect taxes on foreign and domestic trade that had low income elasticities. As a result, the government had to make continuing tax introductions and modifications just to maintain the share of these taxes in GNP. But this explanation cannot be sufficient, for revisions in the tax structure itself formed an important part of the plans and of external conditions. The Philippines had on several occasions pledged to broaden the base of the tax system and to shift the emphasis toward direct taxes, away from taxes on international trade. The reasons for the lagging tax effort were more fundamental. To a large extent the revenue efforts of the Marcos government were vitiated by a growing array of incentives, exemptions, and special privileges that steadily eroded the tax base. Much of the investment that was undertaken during the 1970s was done under Board of Investment incentives, and a growing number of industries benefitted from special industry incentives programs. Despite the early efforts at removing tariff and tax exemptions for government corporations, these crept back into the tax system so that by 1983 the value of government corporate tax exemptions was over P. 1.5 billion, or almost 4 percent of tax revenue.23 Exemptions also went to private or quasi-private enterprises engaged in commercial activity, displacing taxpaying concerns. One example involves the Philippine Veterans Investment and Development Company (PHIVIDEC), which was given tax free importation of finished products such as tires and appliances, which it then resold

417

Philippines/Chapter2

to favored companies. Another example was Nivico, a private concern, which was allowed to import televisions tax and duty free to support a program of extending television use to the countryside. Nivico in fact sold the televisions in Manila and other major markets. The Philippines also developed a number of special tax funds with revenues earmarked for a particular use. Some of these, such as the petroleum industry special fund which supported energy investments, were little different from general revenue taxation. However, others were more in the nature of private funds and competed with the tax system in raising revenues. An example of the latter was the Coconut Industry Development Fund, which was supported by a levy on coconut production. The coconut levy had a variety of goals, including the maintenance of domestic price ceilings and the capture of windfall gains in the mid-1970s. By the late 1970s it became a privately administered fund for the benefit of the industry, effectively controlled by Eduardo Cojuangco, a private businessman and a friend of Marcos. The levy was initially P. 150 per metric ton (about 12 percent of the price of copra) and eventually P. 1,000 per metric ton (32 percent of the copra price). Revenues from the fund were used to purchase the United Coconut Planters Bank and much of the country’s coconut milling capacity through the United Coconut Oil Mills (UNICOM), as well as some expenditures for replanting and various benefits for coconut farmers. But much of the revenue collected cannot be accounted for; the fund was privately administered and not subject to audit. Estimates of the cumulative collections from 1973 to 1983 range as high as P, 9.7 billion, about one quarter of 1983 tax collections (Montes 1986, 44).24 A second example of a special tax that competed with the domestic tax system was the PAG-IBIG fund attached to the Ministry of Human Settlements. This was a compulsory savings scheme to provide housing for its members and was supported by a payroll tax of 3 percent. Collections under the fund were P. 100 million in 1981, 600 million in 1982, and 1 . 1 billion in 1983.25 While these special levies were at times effective in raising revenue, they ranged from being beyond the government’s control (as in the case of the coconut levy) to being unavailable for other pressing budgetary needs (as was the case with the oil industry special levy, which built up a surplus at a time when nonenergy projects were being delayed for lack of peso counterpart funds). And the existence and growth of these funds limited the expansion of the general revenue of the national government. Beyond this erosion of taxing power, the revenue experience of the Marcos government illustrates some of the weakness of the martial law regime and the deterioration in its later years. While Marcos was able to dislodge the traditional elite from power with the declaration of martial law and the dissolving of Congress, the victory was never a complete one. The

418

Robert S. Dohner and Ponciano Intal, Jr.

early dissipation of the land reform program in the 1970s was an indication that Marcos could or would not press his challenge to the landed elite too far. The hold that the Marcos government had on power may have been too weak to allow a significant increase in taxation on the wealthy. Instead, he sought to create a competitive elite, using the rent mobilization powers of government to favor his own interests and those of his associates. Many of the policy measures benefitting Marcos and his associates involved special privilege or tax and tariff exemptions. Thus the ability to raise revenue was compromised by the erosion of taxation that this policy created.26 The failure of the revenue effort also illustrates the loss of momentum of the Marcos administration as a development-oriented and reforming government. The decree-making power of the martial law government was used initially to implement reform measures that had been widely discussed, but blocked by Philippine politics. After the initial successes in the 1970s, the character of the martial law government shifted toward concern with regime maintenance and enrichment, and the decree-making power was used in more particularistic and preferential ways. Special exemptions for individual industries and, in some cases, individual firms proliferated, as Marcos associates were fostered. Rules were enforced in different ways depending upon the individual or case involved. In tax administration, this meant the shift from enhanced participation and self-assessment in the early Marcos years to the (perhaps traditional) feeling that the rich and influential were exempt from taxation. This was echoed in the perceptions of BIR employees, who were reluctant to pursue large taxpayers for fear of who their sponsors might be.27 As a result the compliance and self-enforcement necessary to make an income tax system function weakened considerably, and widespread evasion was the result.

2.5

Public Corporations

Focussing on the budget of the national government provides only a partial understanding of the Philippine fiscal position. During martial law there was a tremendous growth in the number and importance of government-owned corporations. These entities became the primary vehicle for infrastructural and industrial investment programs. And, although the accounts of these firms are not carried on the books of the national government, the corporations came to have an increasing share of the public sector deficit and of total public external borrowing. During the 1980s the continued investment of these corporations, coupled with their heavy losses, provided a huge drain on the national budget at a time when policymakers were belatedly trying to adjust to external shocks and the mounting debt burden. Although many government corporations were audited by the government’s Commission on Audit, there was no attempt until quite recently to put

419

PhilippinedChapter 2

together consolidated figures for the government corporate sector. In fact, the report of a mission of the World Bank in 1983 was the first attempt to achieve a consolidation of their accounts (World Bank 1984b). While governmentowned corporations existed before the martial law period in the financial, public services, and industrial promotion areas, their role in the Philippine economy and in the public sector was relatively small. At the beginning of martial law there were about 30 government-owned corporations. By 1984 there were 96 parent companies, with at least 149 subsidiaries.28 The increasing importance of these firms may be gauged from a few summary figures. Investment by government-owned corporations in 1975 amounted to 0.4 percent of GNP and 12 percent of public sector investment. By 1982 investment of public corporations was over 5 percent of GNP and 60 percent of total investment of the public sector. By 1982 the deficit of public sector corporations had grown to over 2 percent of GNP, and their external debt was almost three-fifths of the public sector total. Data on the sectoral breakdown of government corporations are shown in tables 2.8 and 2.9. State-owned financial institutions dominate in terms of gross value added. However, the bulk of the investment and the use of external funds was accounted for by nonfinancial public corporations. The most important of these were in the energy sector, but government corporations played a crucial role in agriculture (mostly in irrigation investments), transportation, and later, in manufacturing. Table 2.9 presents Table 2.8

Gross Value Added (GVA) of Government Corporations by Sector (in percentages) Shares of Total Government Colporation GVA

Agriculture Manufacturing Electricity Transportation Finance Othersa All government corporations

1975

1978

1982

1984

10.8 4.4 2.9 1.7 79.9 0.3

3.5 5. I 8.3 5.0 76.1 2.0 100

2.3 5.3 10.6 2.7 71.6 1.5

5.1 6.0 30.7 4.3 50.0 3.9

IM)

100

100

~

Shares of Total Sectoral Value Added Agriculture Manufacturing Electricity Transportation Finance All Industry

1.1 0.5 9.2 I .0 58.6 3.0

Source: Manasan and Buenaventura (1986), tables 3, 4.

"Mining, construction, trade, other services.

0.5 0.8 33.1 3.4 66.6 3.7

0.5 1.1

46.1 2.2 95.5 5.3

0.7 0.9 85.9 2.5 84.1 3.7

420

Robert S. Dohner and Ponciano Intal, Jr.

Table 2.9

Companies Agriculture NFA NIA Industry EPZA NDC Energy NEA NPC PNOC Transportation LRTA MMTC PNR PPA Water supply LWUA MWSS Housing HSDC NHA

Fifteen Major Nonfinancial Government Corporations Cash generation as % of Capital Expenditures. 1980-84

Number of Employees, 1984"

Capital Expenditures, 1980- 84 (million pesos)

11,195 7.382

8.905

1,705 4,670

478 10.718

- 10.9

3.6

6,445

915 11,523 13,513

3.453 29,859 7.053

-11.4 8.8 31.6

5,716 N.A. 34.969

34 2,136 6,222 2.033

2.675 I06 469 1,844

-11.5

-

11.3 -43.5 51.0

329 687 1.157

687 4,497

898 4,152

-9.5 18.1

1.55 I 3.983

2,556 2,531

1.477 2.369

3.5 2.1

206 977

641

Debt Outstanding 1984' (million pesos)

3.6 -8.0

8.286 2,981

-

Sources: Presidential Commission on Government Reorganization ( 1985b); and World Bank (1986). vol. 3. tables 5-6, 5-7. Nufe: N.A. means data were not available. and a dash indicates that the amount was negligible

"Includes subsidiaries of government firms, but not firms acquired from the private sector. Company Names: NFA: National Food Authority; NIA: National Irrigation Administration; EPZA: Export Processing Zone Authority; NDC: National Development Corporation; NEA: National Electrification Administration; NPC: National Power Corporation; PNOC: Philippine National Oil Company; LRTA: Light Rail Transit Authority; MMTC: Metro Manila Transit Corporation; PNR: Philippine National Railways; PPA: Philippine Ports Authonty; LWUA: Local Water Utilities Administration; MWSS: Metropolitan Waterworks and Sewerage System; HSDC: Human Settlements Development Corporation; and NHA: National Housing Authority.

more detailed information on the largest of the public sector corporations. Within this group the NPC stands out, both for the size of its investments and the size of its losses during the early 1980s. As in other countries, the Philippines had a variety of reasons for using the corporate form for major infrastructure investments. Government corporations offered a more flexible organizational vehicle for many objectives. Since they were outside the national government, they were not subject to the same restrictions nor to the same pressures of the budgetary cycle. Corporations were not bound by civil service requirements and could, and did, pay higher salaries than the government. Public corporations were also an effective way of raising salaries for key civil servants or for rewarding

421

PhilippinesiChapter2

political allies and military officers, either through seats on corporate boards or through corporate hiring of individuals who were then released to work at ministries on a more or less permanent basis. The public corporations were outside the regular lines of authority of the national government and had substantial independence. They could borrow in their own name and could, to a considerable degree, determine their own budgets. Oversight and control by the government was often lax. Each corporation was attached to a government ministry, and the relevant minister had a seat on the corporate board. But in practice this arrangement did not provide for effective supervision. Cabinet ministers often sat on multiple corporate boards. In addition, a presidential decree or letter of instruction could overrule any minister or policy review process, and those who had direct access to President Marcos used it extensively to create functional autonomy for their corporations. In certain situations, notably in the public corporations involved in the sugar and the coconut industry, the heads of the corporations had cabinet rank and sole responsibility for matters affecting their industry and were able to win out in key battles with other cabinet ministers.29 Although nominally subject to government audit, many of the corporations resisted, and accounts are difficult or impossible to come by. This independence and relative obscurity was a decided advantage for government corporations that intervened in domestic markets, particularly where substantial rents were involved. In both sugar and coconuts huge sums of money were collected from producer levies, In addition to sugar and coconuts, there were interventions, through government firms, in food marketing and distribution and in overseas labor services. Governmentowned corporations also became the favored approach for the projects of Mrs. Marcos, for heart and kidney research and treatment, for international film exhibitions, and for the various activities of the Ministry of Human Settlements. During martial law, the number of government corporations greatly proliferated, in many instances far beyond the natural monopoly or public good rationale that justified the initial corporate entities. Thus by 1985 the Philippines had public sector corporations formed for any or all purposes, ranging from banks, nuclear plant, real estate, racehorses and gamecocks, gambling casinos and lottery houses, poultry farms and tomato paste, to a dizzying array of Centers concerned with culture, music, science, health, artists, and all known fields of human endeavor including the meaning of life. (Briones 1985, 2) Based on the obligational budgets in table 2.4 and national accounts investment breakdowns, the investment expenditures of the government corporate sector increased sharply beginning in 1976. Estimates of the accounts of the nonfinancial public corporate sector are available starting in 1978 (table 2.10). These show a level of investment by public corporations in excess of 4 percent of GNP at that time. Also evident from the table is the

422

Robert S. Dohner and Ponciano Intal, Jr. Public Nonfinancial Corporations, Cash Operations (in billions of pesos)

Table 2.10

Year

Investment

1978 1979 1980 I981 1982 1983 1984 1985 1986

10.4 12.0 16.5 16.8 18.8 16.5 16.2 9. I

1.5

(% of GNP)

Government Contribution

Cash Generation

Surplus/ deficit

(92 of GNP)

2.5 4.2 5.3 8.4 10.2 7.9 5.6 5.3 11.4

0.9 0.8 0.2 0.1 - 0.3 2.7 2.7 4.2 0.3

-4.1 -5.4 -6.6 - 8.0 -7.0 -8.1 -8.1 -6.7 2.6

(-2.3) ( - 2.5) ( - 2.5) (-2.7) (-2.1) (-2.1) (-1.5) (-1.1) (0.4)

Source: Philippines, Government Corporate Monitoring Committee

large jump in corporate investment that took place in 198 I as the Philippines moved to offset the domestic recession, accelerate its energy program, and start on the MIPS. Between 1981 and 1983, over 65 percent of public fixed investment was done by government-owned corporations. The fourth column in table 2.10 shows the cash generation of the public corporations, or the surplus after meeting current operating and interest expense that was available to fund investment. Cash generation by Philippine public corporations has been very low, averaging only 6 percent of the annual investment of the corporations covered in the table. To some extent this performance represents the long gestation times characteristic of many of the infrastructural investments, coupled with a rapidly expanding public capital stock. But it also represents low profitability of operation of existing capital equipment. In some cases, tariffs for public services were set too low to achieve targeted rates of return. This was particularly true of the electricity charges of the NPC, which were judged in a 1979 study to be 30 percent below the long-run marginal cost of electricity. Low rates of collection of existing tariffs have plagued the National Irrigation Administration, leading to minimal cash generation in that agency. Finally, those government corporations that have engaged in lending operations, such as the National Electrification Administration’s loans to rural cooperatives or the Local Water Utilities Administration’s loan to local water authorities, have had large arrears in repayments, resulting in substantial losses in those agencies. Despite recognition of the problem and pressure from the IMF and World Bank to raise public corporate revenue mobilization, increasing the cash generation of public corporations was made quite difficult by the inflation and exchange depreciation of the 1980s. Although there were repeated increases in rates, revenues of government corporations little more than kept pace with increases in domestic costs and debt service costs on foreign loans. (The improvement in cash generation shown in table 2.10 for 1983 was largely the result of an increase in arrears by the NPC on external loans.)

423

Philippines/Chapter 2

Low internal cash generation and high rates of investment resulted in large demands for external funds by the government corporate sector. A large part of the burden was borne by contributions of the national government, and the remainder through external borrowing. Contributions from the national government took the form of operating subsidies, equity contributions to the corporations, and loans. Each has a distinct accounting implication, but in practice the difference among them was blurred. There are no easily accessible data on interest payments from government corporations to the national government, and there have been numerous instances in which outstanding loans were converted to equity. Public corporations have generally not paid dividends to the national government (Manasan and Buenaventura 1986, 6-7). National government contributions to public corporations are shown in table 2.11. By the 1980s as much as 20 percent of government expenditure, and in several cases all of the national government deficit, were accounted for by these contributions. These are explicit contributions from the national government to public corporations; they do not include the implicit subsidies offered to these firms by tax and import tariff exemptions. In 1983 tariff and tax exemptions to government-owned corporations were estimated to have been worth P. 1.5 billion, or 20 percent of explicit budgetary contributions to the corporations. Although these exemptions were reduced by presidential decrees in 1984, many were reinstated during the following months.30 More extensive data on the financing of investment are available for the fifteen largest public corporations and are shown in table 2.12. Internal cash generation financed less than 10 percent of investment over the 1978-84 period. National government contributions made up 45 percent of the additional funding requirement, and the remainder came from external loans. The domestic capital market was not tapped as a source of funding for government corporations. In fact, they slightly reduced their indebtedness to Table 2.11

National Government Contributions to Public Corporations (in millions of pesos and percentages) Share of

Year

Current Transfers

Equity

Net Lending

Total

1975 1976 1977 1978 1979 1980 1981 1982 1983 1984

285 392 245 632 478 505 564 889 586 429

522 1.804 2,252 2,245 3,391 4,739 7,862 8.419 4,821 9,819

I22 100 45 238 853 675 929 2,218 2,393 10,086

929 2,296 2,543 3,115 4,722 5,919 9,355 11,526 7,800 20,334

Source: Manasan and Buenaventura (1986), table 8.

Government Expenditures 5. I 11.2 11.2 11.9 15.8 15.5

19.5 21.9 14.7 30.5

Government Deficit 66 98 89 144

1,380 175 77 80 105 207

GNP 0.8 I .7 I .7 I .8 2.2 2.2 3. I 3.4 2.1 3.9

424

Robert S. Dohner and Ponciano Intal, Jr.

Table 2.12

Years

Sources of Financing for Fifteen Major Nonfinancial Corporations

Capital Expenditures

Internal Surplus"

Total Financing

National Government Contributions

Net Domestic Borrowing

Net Foreign Borrowing

(in millions of pesos)

2,236 3.335 4,300 7.169 8,378 6,402 5.664

1978 1979 1980 1981 1982 1983 1984

7,281 9,518 11,079 15,293 15,028 19,448 15,282

1.636 2,917 1,083 705 - 1,985 1,889 2,831

All

92,929

-~ 9,076 83.853 37,484 -

5.645 6,601 9,996 14,588 17,013 17,559 12,451

-3.320

2,512 5,519 5,680 7,933 7,934 10,557 10,108

-3,863

50.243

907 -2,253 16

-514 70 I 600

(as percentages of GNP)

1978 1979 1980 1981 1982 1983 1984 All

4. I 4.4 4.2 5.0 4.5 5.1 2.9 4.2

Source: Manasan and Buenaventura

0.9 1.3 0.4 0.2 -0.6 0.5 0.5 0.4

3.2 3.0 3.8 4.8 5. I 4.6 2.4 3.8

I .3 1.5 1.6 2.4 2.5 1.7 1.1

1.7

0.5 - 1.0

0.0 -0.2 0.2 0.2 -0.6 -0.2

1.4 2.5 2. I 2.6 2.4 2.8 I .9 2.3

(1986). table 14

"Including change in cash balances.

the private sector over this period. Domestic counterpart funds for investment projects were made up entirely from contributions from the national government, which explains the correlation between size of national government contributions and external borrowing that has been noted for the Philippines.31 Public corporations in the Philippines played a crucial role in the early 1980s. The combination of their large funding requirements and the difficulty in bringing their outlays under control gave public expenditure a momentum that could not be reversed when there was a critical need for the Philippines to adjust to the worsened external environment. Monitoring and control of public enterprises was made more difficult by the fact that the available budgetary information covered only the national government; there were no integrated accounts of expenditure and funding requirements of public corporations. After the Philippine debt crisis, at the prodding of the IMF, the government put together data on the major public corporations. This data, from which the tables in this section have been compiled, make possible the construction of accounts for the nonfinancial consolidated public sector going back 1978. These figures give a much clearer indication of the fiscal stance of the Philippine government. Before we examine these

425

Philippines/Chapter 2

consolidated accounts, we review briefly the two remaining components of the nonfinancial public sector, local governments and the social security institutions.

2.6 Local Governments Local governments have not played an important role in the Philippines. Government functions and authority have been concentrated in Manila, and this centralization was greatly strengthened under martial law. Local government expenditure has been constrained by severe revenue limitations. Property taxes are the primary source of revenue, but tax rates are low and, since no process of cadastral surveying exists, many properties are not on the tax rolls. The expansion of activity by the communist New People’s Army (NPA) has reduced tax collections in many localities, and in some areas the NPA operates as the only government, levying taxes of its own. In addition to local sources, about 40 percent of the revenue of local governments comes directly from the national government through revenue sharing and as additional aid and allotments. Total revenues of local governments from all sources have hovered around 2 percent of GNP (table 2.13). Local government expenditure has been almost entirely for current operations. Capital expenditures of local governments, on roads and local facilities, have made up only about 15 percent of total local government expenditure. Borrowing by local governments is possible, but in general strictly limited. In total, local governments have run balanced budgets or small surpluses.

2.7

Social Security Institutions

The Philippines has two government-sponsored employment security institutions. The larger, the Social Security System (SSS), covers nine million workers in the private sector, while the Government Service Insurance System (GSIS) covers about one million government workers. Total revenue collection by these bodies has averaged just under 2 percent of GNP, while their surpluses (after expenses and net lending to members) have been about 0.8 percent of GNP in recent years. About one-third of this surplus has been invested in government securities, and most of the remainder has been invested in the Philippine National Bank or the Development Bank of the Philippines. In recent years the social security institutions, particularly GSIS, have made equity investments at the direction of the government, often in companies in financial difficulties. As a result, the character of the GSIS portfolio has deteriorated substantially, and the institution now owns some of the most prominent assets slated for privatization by the Aquino government, including the Manila Hotel and Philippine Airlines.

Table 2.13

Consolidated Income and Expenditures of Loeal Governments, 1975-85 (percentage of GNP) 1975

Total revenue Tax revenue Nontax revenue Aid and allotments from national government Expenditures Cument Capital Surplusldeficit

1.9 0.7 0.5 0.8

1976

1977

0.7

1979

1980

1981

1982

1983

1984

1.8

1.4

0.6 0.5

0.5 0.4

1985

1986

1.9

1.8

1.6

1.8

0.7 0.6

0.7 0.5

0.7 0.5

0.6 0.5

0.6 0.5

0.7

0.6

0.7

0.6

0.7

0.8

0.8

0.6

-

-

-

-

-

-

-

1.7

1.7

1.7

1.4

L_4 1.3

-1.7 -2.0 0.6 0.4

1978

-

-

-

1.9 -

-1.8

-1.9

1.8

1.7

19 . 0.6 0.5

14 . 0.5 0.4

1987

1.4 0.5 0.4

0.5 0.3

0.6

0.6

0.6

-

-

1.3

1.6

1.6

1.5

16 . 1.4

1.5

1.5

1.5

1.2

1.4 1.3

0.3 -

0.3

0.3

0.2

0.2

0.2

0.2

0.2

0.2

0.2

0.l

0.l

0.l

0.0

-0.0

0.0

0.0

0.1

0.0

0.1

0.1

0.1

0.0

0.1

0.0

0.0

1.6

1.5

Source: Philippines, Ministry of Finance, Office of Local Governments. Nore: Components may not add to totals due to rounding.

1.2

427

PhilippinedChapter 2

2.8 Government Financial Institutions and the Central Bank With the material in the previous sections and tables we are now in a position to assemble budget figures for the consolidated nonfinancial public sector in the Philippines. This consolidation is done in table 2.16 below. However, in the Philippines even this degree of consolidation misses an important part of the fiscal story, since it excludes the operating balances of the state-owned financial institutions and the central bank. In the 1980s these bodies were used for fiscal operations, particularly for the rescue of failing private corporations, many of which were owned by Marcos cronies. In addition, public financial institutions were forced to make good on loan guarantees extended to private borrowers who could not meet their external debt service obligations. The losses of the state-owned financial institutions and the central bank reached major proportions in the 1980s, rivaling the deficits of the entire nonfinancial public sector. Only very limited data exist on the balances of the public financial institutions, covering only the 1983-86 period. These are shown in tables 2.14 and 2.15. Table 2.14 shows the balances of the three major financial institutions, the Philippine National Bank (PNB), the Development Bank of the Philippines (DBP), and the Philippine Export and Foreign Loan Guarantee Corporation (PhilGuarantee). Both PNB and DBP suffered major losses as a result of the devaluations in 1983-84 and the following domestic recession. For PNB the losses arose primarily on domestic operations, particularly from the collapse of the sugar industry in 1985. The Development Bank of the Philippines was used extensively as a corporate rescue agent in the 1980s. The period saw a sharp rise in DBP’s industrial loans and investments, many of them extended at the behest of the government to financially strapped firms (Lamberte 1984, 16- 17). DBP was also used to support domestic commercial banks associated with the Marcos government. Equity investments totaling P. 267 million were made in Associated Bank and Pilipinas bank. These investments were in turn rediscounted with the central bank, in effect channeling central bank funds to these two private banks at a cost far below that charged by the central bank on its emergency advances (1984, 17). By the end of 1983 DBP had equity investments totalling P. 9.2 billion, or 17 percent of its assets, the bulk of which were in banks, hotels, mining, textile manufacturing, steel, and construction. A remarkable memorandum from the director of DBP to Marcos in 1983 lists the loans and investments made at the government’s behest. These totaled P. 28.2 billion ($2.5 billion at 1983 exchange rates), or over five times DBP’s capital (Tengco 1983, 1). In 1982 and 1983 support for the deficits of the state-owned financial institutions came primarily from the central bank. The reserve money targets of the IMF program limited the use of central bank credit after 1983, and the burden of supporting the deficits of these financial institutions fell on the

428

Robert S. Dohner and Ponciano Intal, Jr. Deficits of Major Government Financial Institutions' (in billions of pesos)

Table 2.14

1983

1984

1985

1986

Domestic surplus/deficitb PNB DBP PhilGuarantee

N.A. N.A. 0.9

-2.2

- 1.8

-4.8 -3.9 -0.9

-

-0.3 -0.1

-6.6 -5.8 -0.8 -

-

Foreign interest and net foreign paymentsC PNB DBP PhilGuarantee

N.A. N.A. -3.1 -0.2

-6.3 -2.5 -3.6 -0.2

-11.9 -2.2 -8.7 - 1.0

-1.1 - 1.8 -4.5 - 1.4

-8.5 1.6) -4.3 -3.9 -0.3

- 18.5 (-3.1) -8.0

Overall deficit (% of GNP) PNB DBP PhilGuarantee

-5.3 (-1.4j -2.9 -2.2 -0.2

(-

-

- 12.4

2.0) -5.7 -5.4

(-

9.5

- 1.0

-1.4

Transfers from national government PNB DBP PhilGuarantee

0.2 N.A. N.A. N.A.

1.3 I .o 5.5 0.8

10.7 1.5 8.3 0.9

15.5 6.4 8.0 1.1

Memo: Foreign debt service after first reschedulingd PNB DBP PhilGuarantee

N.A. N.A. 10.1 0.2

6.6 2.1 3.1 0.2

16.1 5.0 10.I I .o

9.1 2.9 5.2

~

~~

~~~

~

1 .o

~

Source: IMF (1986a. 24). table 8; and IMF (1987). Nore: N.A. means data were not available, and a dash indicates the amount was negligible.

*Philippine National Bank, Development Bank of the Philippines, and Philippine Export and Foreign Loan Guarantee Corporation. bDomestic operations, net payments on domestic guarantees, captial expenditures, and sales of assets. 'After 1985 debt rescheduling. dlnterest and principal repayments on foreign loans plus gross advances on foreign loan guarantees.

national government, as is shown from the rapidly rising transfers in table 2.14. By 1985 these institutions had become a fiscal nightmare, and their annual losses amounted to more than 3 percent of GNP. What table 2.14 does not adequately show is the capital loss and continuing obligation shouldered by the government as a result of the rapidly deteriorating loan portfolios of PNB and DBP. At the end of 1983, the DBP had external liabilities of approximately $1.5 billion, plus an additional $1 billion in outstanding foreign loan guarantee^.^^ Even larger losses were suffered by the Philippine central bank during this period; in the four years shown in table 2.15, central bank losses averaged 3.5 percent of GNP. Some of the reasons for this loss are indicated in the table. The central bank entered a number of forward cover and swap contracts in the early 1980s. When the exchange rate depreciated in 1983 and 1984, the central bank was forced to accept the resulting losses. Much of the forward cover was extended to public corporations, particularly the

429

PhilippineKhapter 2 Central Bank Net Income, 1983-86 (in billions of pesos)

Table 2.15

-

1983 Net interest Payments (Domestic) (Foreign) Receipts Forward cover profits Swap profits/losses Total surplus/deficit (percent of GNP)

-2.0 -7.2 -0.7 6.5 5.2 -5.0 -6.8 - 13.8 (3.6)

-

1984 -8.3 - 16.4

-5.3 -11.1 8.1 -5.3 - 14.0

-27.6 (5.2)

1985

1986

15.6

N.A. N.A. N.A. N.A. N.A. N.A. N.A. - 16.9 (2.8)

-

- 24.6 - 12.6 - 12.0

9.0 - 7.6

7.0 - 15.5

(2.6)

Deficits Mount Due to ‘ J o b ’ Bills,” Manila Chronicle, 15 September 1986, p. I . Forward cover, swaps, and total-IMF (1987). Sources: Net interest--“CB

Note; N.A.

=

not available.

PNOC.33 Swap contracts were entered with domestic commercial banks and their foreign currency deposit units, as a way of encouraging further international borrowing. Losses under both of these contracts were posted in every year, except in 1985 when the peso appreciated and the central bank earned a profit on its swaps. Much of the loss of the central bank during this period came from increased net interest payments in 1984 and 1985. In order to reduce the domestic money supply, the central bank sold its own bills in 1984 and again in 1986, sometimes at rates in excess of 40 percent per annurn. Outstanding central bank securities, net of repurchase agreements, reached almost P. 40 billion ($2 billion) by September 1985. Additional losses came from the external liabilities of the central bank. Net foreign assets of the central bank turned negative in 1982, and by the end of 1984 the net external liabilities of the central bank had reached P. 35 billion ($1.8 billion), almost 6 percent of GNP. The most difficult effect to gauge is the weakening of the central bank portfolio from the emergency advances to financial institutions and, indirectly, to troubled nonfinancial firms in the private sector. Assistance to financial institutions increased sharply at the end of 1984 during the adjustment crisis and again in early 1986 during the post-election boycott of government-associated banks. The central bank was also exposed to the two troubled government financial institutions, PNB and DBP, having indirectly supported some of their rescue operations of private firms. Central bank emergency loans and overdrafts to financial institutions reached P. 11 billion (2 percent of GNP) at the end of 1984. Of this, P. 1 billion was to specialized government financial institutions, a category made up almost entirely by DBP. These and other issues surrounding the Philippine financial system are discussed in more detail in chapter 5.

430

Robert S. Dohner and Ponciano Intal, Jr.

2.9 Consolidated Budget We can combine the accounts of the various units of the public sector, netting out the transfers among them, to arrive at a consolidated public sector budget. Two consolidations are shown in table 2.16. The first is of the nonfinancial public sector, which includes the national and local governments, social security institutions, and public nonfinancial corporations. The second consolidation, available only for 1983 through 1986, adds in the major public financial institutions and the central bank. Several things are apparent from the consolidations. The first is that the relatively low deficit of the national government between 1978 and 1980 is deceptive. Most of the public sector deficit was contained in the accounts of public sector corporations, although this information was not readily available at the time, and the consolidated nonfinancial public sector deficit was about three times that of the national government. Although the consolidated figures are larger, the jump in the deficit in 1981 and 1982 was not as dramatic as that of the national government. What is also apparent is the momentum of the public sector deficit in the critical 1982-83 period, despite the efforts of the national government to change fiscal course. Continued investments by public corporations in 1983 added to the deficit, while the huge increase in losses of the public financial sector continued the deficits after 1983. While the deficit figures contained in table 2.16 are respectable, they are not extraordinarily large. The nonfinancial public sector deficit never reaches 6 percent of GNP, while even with the losses of public financial institutions, the combined deficit peaks at just over 8 percent of GNP. While public sector deficits play a role in the drift of the Philippines into debt crisis, they are not the sole, nor perhaps the most important, explanation. The problems caused by the public sector deficits in the Philippines are more a matter of timing and content than size. The Philippines was unfortunate to have had a sharp increase in the deficit of the public sector at precisely the time when external signals called for a scaling back of public borrowing to limit external borrowing. A more fundamental difficulty that the Philippines had was the increasing weakness of public investment expenditure. Analyses of sustainable foreign borrowing make the distinction between borrowing for investment and borrowing for consumption. Although the Philippines maintained high and increasing rates of public investment, in fact what was purchased with that investment expenditure made it little different from consumption. Much of public sector investment in the 1980s was loans and equity contributions to failing private sector corporations, absorbing the losses of those enterprises. Certainly not all, but much of the Philippine investment effort was ill advised, in assets that never paid out or, in some cases, never materialized. Public sector investment took on a momentum of its own and

Table 2.16

Public Sector Balances (percentage of GNP)

National government Revenue Expenditure Current Capital Total equity, net lending Aid to GCs Aid to GFIs Savings Surplus/deficit Local government Revenue Aid from national government Expenditure Current Capital Savings Surpluddeficit Social security Investment Savings Surpluddeficit Government corporations Investment Contributions from national government Cash generation Surpluddeficit (continued)

1978

1979

1980

1981

1982

1983

1984

1985

1986

13.6 14.8 10.9 2.6 1.4 1.4 0.0 2.7 - 1.2

13.5 13.7 9.5 2.3 1.9 1.9 0.0 4.1 -0.2

13. I 14.4 9.3 3.1 2.0 2.0 0.1 3.9 - 1.3

11.8 15.8 8.7 4.2

3.0 2.8 0.2 3.1 -4.0

11.4 15.7 9.2 3.0 3.5 3.0 0.4 2.2 -4.3

12.0 14.0 9.1 2.7 2.1 2.1 0.0 2.9 -2.0

10.8 12.7 8.1 1.9 2.7 1.1 I .6 2.7 - 1.9

11.5 13.5 9.2 1.7 2.8 0.9 1.9 2.4 - 1.9

12.8 17.8 10.8 1.9 4.4 1.9 2.6 2.0 - 5.0

1.9 0.6 1.8 1.6 0.2 0.2 0.0

1.8 0.7 1.7 1.5 0.2 0.3 0.1

1.6 0.6 1.6 1.4 0.2 0.3 0.0

1.8 0.7 1.7 1.5 0.2 0.3 0.1

1.9 0.8 1.7 1.5 0.2 0.3 0.1

1.8 0.8 1.7 1.5 0.2 0.3 0.1

1.4 0.6 1.4 1.2 0.2 0.2 0.0

1.4 0.6 1.4 1.3 0.1 0.2 0.1

1.4 0.6 1.4 1.3 0.1 0.1 0.0

0.4 1.o 0.6

0.3 1.1 0.8

0.3 1.o 0.8

0.4 1.2 0.8

0.3 1.2 0.9

0.3 1.0 0.8

0.2 0.8 0.6

0.2 1 .o 0.8

0.2 1.o 0.8

4.3 1.4 0.5 -2.3

4.8 1.9 0.4 -2.5

4.5 2.0 0.1 - 2.5

5.4

5.0 3.0 -0.1 -2.1

4.9 2.1 0.7 -2.1

3.1 1.1 0.5 1.5

2.7 0.9 0.7 -1.1

1.5 1.8 0.0 0.4

2.8 0.0 -2.7

Table 2.16

(continued) ~

Consolidated nonfinancial public sectoP Investment Savings Surplus/deficit Public finuncial sector Surplus of GFIs Net income of central bank Contributions from national government Surplus/deficit Consolidated public sector Surplus/deficitb

1978

1979

1980

1981

1982

1983

1984

1985

1986

7.5 4.5 -2.9

7.5 5.8 -1.7

8.2 5.2 -3.0

10.4 4.7 -5.7

9.0 3.6 -5.4

8.2

5.0 -3.2

6.9 4.1 -2.8

6.4 4.3 - 2.2

6.2 3.2 -3.6

-1.4 -3.6 0.0 -5.0

-1.6 -5.2 1.6 -5.3

-3.1 -2.6 1.8 -3.9

- 2.0 -2.8 2.7 -2.1

-8.2

-8.3

-6.1

-5.8

Source; Tables 2.1, 2.7. 2.10, 2.13, and the Philippines, Government Corporate Monitoring Committee. Note: GCs are government corporations and GFIs are government financial institutions.

'National government, local government, social security institutions, and nonfinancial public sector corporations, net of interagency transfers and investments. bConsolidation of nonfinancial public sector and public sector financial institutions, net of interagency transfers and investments.

433

PhilippineKhapter 3

became more of a way of assuring foreign currency inflows than a means of creating capital stock. New projects were started as late as 1983, despite delays and stretch-outs of existing projects due to the inability of the government to come up with counterpart funds. Current expenditure, particularly operations and maintenance expenditures, were cut back to sustain investment, in some cases prematurely retiring the existing capital stock. Finally, the Philippines shifted the public sector deficit from the national government to public corporations and later to government financial institutions and the central bank, using the borrowing ability of each to keep the system afloat, until the process could no longer be sustained.

3

Trade Policy, Industrial Policy, and the Exchange Rate

Trade and industrialization policy have been the vortex of Philippine economic debate. Trade policy has been more extensively argued in the Philippines than has any other economic policy, starting with the outcry over the administration of the import control program in the early 1950s and extending through the current debates on import liberalization. This prominence is reflected in research on the Philippine economy, and there is now an extensive literature on Philippine trade and industrial policy. Trade policy issues are also central to our analysis of the Philippine debt crisis. In comparative studies of LDC borrowers, the extent to which exports grew appears to play a key role in determining whether or not countries were forced to reschedule.2 In the Philippines in particular, trade and industrial policy were powerful forces behind the slide of the economy into crisis in the 1980s. Despite the importance given to industrial and trade policy, Philippine industrial and trade performance has been largely disappointing. The initial period of import substitution led to rapid economic growth in the early 1950s. However, in what has now become a classic pattern of import substitution, growth slowed as the industries that were created reached the limits of the domestic market and as their high dependence on imports of capital goods and intermediates meant that the growth of the economy as a whole was limited by recurrent balance of payments crises. The Philippines went through an import decontrol program in the early 1960s, but with disappointing results. Economic growth remained sluggish, particularly in the manufacturing sector, and the country failed to develop significant new export industries.

434

Robert S. Dohner and Ponciano Intal, Jr.

For much of the 1970s it appeared that the Philippines had achieved a decisive break with its past. The Marcos martial law government was outwardly strongly export promoting and actively encouraged investment, both domestic and foreign, in export industries. Philippine export growth in the seventies was quite respectable, averaging 8.6 percent per year in volume. But what was more striking was the transformation in the export structure of the country that took place over this period. In the 1960s the Philippines was overwhelmingly dependent on commodities+oconut products, sugar, forest products, copper, and gold-for its export earning, as we point out in table 3.1. At the beginning of the 1970s these four product categories alone accounted for 76 percent of Philippine exports; by the end of the decade, their share had fallen to 46 percent. In their place was a group of nontraditional products, particularly electrical components and garments. For most of the decade, manufactured exports had increased by over 30 percent per year (de Vries 1980, 5). But despite this success in diversification and the rapid growth of manufactured exports, in aggregate Philippine exports were still problematic. In table 3.2 we present a comparison of export shares in GDP for the Philippines and neighboring Asian countries. While the Philippines starts out in a large group with a low export to GDP share, the ratio never changes over the course of the decade, in contrast to the other countries listed in the table. In fact, the sluggishness of the Philippine export share is worse than is indicated in table 3.2. The structural transformation that took place during the decade shifted Philippine exports toward goods characterized by low margins of domestic value added and high import requirements. As a result, the growth in domestic value added, or alternately, the net foreign exchange generating capacity of the export sector, was much less than suggested by the growth of exports. A rough correction for this effect is shown in the Table 3.1

Share in Total Merchandise Exports (percentages)

Exports

1965

1970

1975

1980

1984

1986

1987

Coconut products Sugar products Logs, lumber & plywood Copper and gold

34.0 18.5 22.7 5.9 -

18.6 17.2 24.2 16.2 -

20.3 26.9 9.4

14.0 11.4 6.6

9.7 2.2 5.2 4.8

-

9.2 I .0 3.9 4.8

Suhlotal Pineapple and bananas Electncal equipment and components Garments Other nontraditional manufactures Others

76.2

13.5

81.1

12.6 69.2

13.5 6.1 4.1 4.1

45.5

28.4

21.9

I .8

2.5

5.0

3.6

4.4

5.3

N.A.

-

-

-

-

2.0 4.4

11.6 8.6

24.7 11.1

18.7 15.5

19.6 19.2

5.8 11.3

10.9 10.4

11.5 7.9

16.2 14.5

21.9 9.5

21.3 17.3

N.A.

Source: NEDA, Philippine Statistical Yearbook, 1987; and IMF (1988b). Note: N.A.indicates data were not available, and a dash means that the amount was negligible

19.0

24.9

435

PhilippineKhapter 3

Table 3.2

Merchandise Export Shares in GDP (in percentages) 1970

1974

1976

1978

1980

1982

1984

1986

Philippines Indonesia Korea Malaysia Thailand

15.1 12.7 10.1 41.3 10.5

18.6 28.8 24.1 43.8 18.1

14.2 22.9 27.0 47.5 17.9

14.2 22.6 25.4 46.7 17.5

16.4 33.0 28.1 54.0 19.3

12.6 21.9 28.8 44.6 18.6

16.7 24.5 30.8 48.3 20.2

15.6 19.1 34.5 49.3 20.6

Philippines Memo items: Exports net of consignment imports Service exports"

15.1 2.6

17.7 3.5

12.6 3.0

12.4 4.6

14.0 4.7

10.3 5.9

12.5 6.3

N.A. N.A.

Sources: Philippines: Central bank and National Census and Statistical Office. Others: Asian Development Bank, Key Indicarors of Developing Member Countries. 1986. Note: N.A.

=

not available.

'Net of interest receipts and government service exports (including U S . base receipts).

second to last row of the table, which compares Philippine merchandise exports, net of consignment imports, to GDP. This ratio declines significantly over the decade. As a result of this export sector narrowing, the Philippines was left in a more difficult position from which to adjust to an adverse external shock. The last line in table 3.2 illustrates a factor that sets the Philippines apart from most LDC debtors. Service exports were an important and increasing source of foreign exchange earning over the 1970s and 1980s. Most of this came from overseas labor earnings in construction, nursing, and domestic work, and this source increased in the 1980s as more Filipinos sought overseas employment. Of course, Philippine trade policy affected the entire allocation of domestic resources and not just the export sector. Trade policy and also investment incentives have had a particularly strong impact on the Philippine manufacturing sector, a sector whose history has been puzzling and ultimately disappointing. In contrast to other countries in East and Southeast Asia, the manufacturing sector has not been a leading sector in the growth of the economy (table 3.3). Philippine manufacturing grew rapidly in the early 1950s under the impetus of import protection. But once the initial burst of import substitution played out, output growth slowed considerably. Nor was the sector revived by the decontrol episode in the early 1960s. Manufacturing growth fell behind that of the economy as a whole, profits were low, and excess capacity and competition led to rising nationalist sentiments. The manufacturing sector recovered in the latter part of the decade, and in the early 1970s it appeared that the Philippines might have broken the dependence of the sector on the growth of the domestic economy, with a rapid growth of

436

Table 3.3

Robert S. Dohner and Ponciano Intal, Jr. Comparative GDP and Manufacturing Growth Philippines

Real GDP Growth 1965- 70 1970-75 1975-80 1980-83 Manufacturing value added 1965- 70 1970-75 I975 - 80 1980-83 Manufacturing employment 1963-70 1970-75 1975-80 1980-83

Indonesia

Malaysia

Thailand

Korea

5.0 6.2 6.2 2.6

6.2 7.6 7.9 5.0"

N.A. 7.4 8.6 6.4

8.6 6.3 7.6 5.4

10.4 9.5 7.5 8.0

6.9 6.9 6.0 2.7

7.8 9.6 15.0 N.A.

13.9 12.3 11.3 N.A.

10.2 10.3 10.5 6.0

22. I 19.7 14.8 7.9

2.9 2.9 4.1 3.5

4.9 4.8 4.2 13.4"

16.3 6.6 0.0 0.5

5.1 4.5 5.7

10.7 15.2 6.2 3.3

1 .o

Sources: World Bank (1979). table I .2: Asian Development Bank, Key lndicarors of Developing Member Countries; and IME International Financial Sfafistics. Note: N.A. = not available.

"1980-82

nontraditional manufactured exports and a growth rate for the sector well in excess of that of domestic GDP (table 3.4). But the growth of manufacturing was not sustained, and a closer look reveals serious weaknesses in the sector. The most glaring weakness is its productivity performance. A recent study of manufacturing productivity concluded that total factor productivity in the sectorfell by 1.23 percent per year over the 1970s. The decline was more rapid in the last half of the decade, when total factor productivity fell by almost 2 percent per year.3 The poor productivity performance for manufacturing is an obvious candidate for explaining a problem mentioned in the introductory chapter, the high investment requirement of Philippine output growth. In comparison to other developing countries, both inside and outside the region, the Philippines invested more to grow less. The second difficulty with the manufacturing sector was its inability to generate significant employment, in spite of the rapid growth of laborintensive manufactured exports. The share of manufacturing in total Philippine employment actually peaked at 12 percent in 1956, and has remained just under that level in the period since. Again, this is in contrast to other countries in the region, where the manufacturing sector has played a lead in employment growth (see table 3.3). Finally, the output expansion in the manufacturing sector as a whole proved extremely fragile. The rate of output growth in manufacturing hit a peak in 1977 and then declined steadily through 1985. Manufacturing and the investments that had been made in the 1970s proved extremely

437

Philippines/Chapter 3

Table 3.4 Sector Agriculture Manufacturing Services Domestic product"

Philippine Sectoral Growth Rates 1950-55

1955-60

1960-65

1965-70

1970-75

1975-80

1980-85

1985-87

7.1 12. I 9.0

2.9 7.7 5.1

4.8 4.5 4.5

3.5 6. I 4.8

4.3 6.9 5.7

5.3 6.0 5.7

0.5 -2.7 0.8

2.0 2.8 4.7

7.9

4.6

4.8

5.0

6.2

6.2

- 1.2

3.3

Source: NEDA, National Income Accounts of the Philippines.

*Net domestic product 1950-70, gross domestic product 1970-85

vulnerable to the rise in oil prices and real interest rates in the early 1980s and to the accompanying international recession. Many of the industries that were developed in the 1970s saw massive drops in output in the 1980s. Thus an analysis of the impact of trade and industrial policy has a large amount of ground to cover. Two critical and puzzling questions stand out. The first is the failure to shift resources toward the traded goods sector and the apparent export shallowing of the Philippine economy, despite the transformation of export structure that took place. The second is the inability to generate faster output and employment growth in manufacturing, and the weakness of the sector during the 1980s. Four elements of Philippine trade, exchange rate, and industrial policy provide the explanation. The first is the structure of trade protection and incentives, which encouraged indiscriminate and inefficient import substitution. The second is the country's exchange rate policy, which has been characterized by delayed responses to payments disequilibrium, thereby hurting export industries, dampening export diversification, and discouraging backward linkages. Third, direct interventions by the government in the traditional agricultural export sector during the 1970s not only penalized that sector but also exacerbated its problems in the face of a highly volatile world price environment. And fourth, there was an expansion of governmentcontrolled and government-associated corporations and with it, a tendency toward centralization and politicization of economic decision making during the 1970s. This chapter examines the first two factors; the next chapter discusses the last two.

3.1 Trade Policy and the Structure of Protection Philippine trade policy and the industrial policy that grew out of it were more products of the response to crisis and the limited policy choices that the Philippines had under the Bell Act than they were conscious attempts to promote developmental goals. But once the trade policy regime was established in the 1950s, it quickly became the cornerstone of the country's

438

Robert S. Dohner and Ponciano Intal, Jr.

industrialization policy and rapidly developed a constituency concerned with its preservation. The Philippines had suffered extensive damage during the Second World War, and in the late 1940s there were huge expenditure demands for reconstruction, which led to a rapid increase in imports. In addition, after the substantial wartime inflation, the prewar exchange rate of 2 pesos/dollar was unrealistically high, further increasing import demand. Exports recovered somewhat, aided by a rise in export prices, but payments balance was maintained largely through transfers from the United States. These payments for relief, veteran’s pensions, and military expenditure totaled almost $300 million in the period 1946-50 and covered over half of Philippine import^.^ A large increase in public spending during the 1948 election campaign and the impending reduction of U.S. aid payments led to a balance of payments crisis. In December 1949 the eleven-month-old central bank initiated foreign exchange controls. Foreign exchange controls were followed by legislation in 1950 that required licenses for all imports, allocated by an Import Control Board. Commodities were grouped by degree of “essentiality.” The most stringent controls were imposed on nonessential consumer goods (i.e., luxuries and consumer durables) and the least stringent on essential consumer goods (e.g., pharmaceuticals, milk products) and essential producer goods (most machinery, fertilizer, fuels and lubricants). Domestic prices for the most heavily restricted imports rose sharply, leading to investment and production in a variety of consumer goods industries. While the balance of payments crisis was the initial motive for import controls, industrial promotion rapidly became the rationale of the protective system as a domestic import-substituting industrial sector developed and economic growth, led by manufacturing, accelerated in the early 1950s. The industrialists who emerged would become a powerful lobby for import restriction in future policy debates, and the import-substituting strategy established in the 1950s would persist, largely unchanged, until the 1980s. The Philippine experience with import controls was typical of that of a number of LDCs. Domestic entrepreneurs responded quickly to profitable opportunities created by import restrictions, and manufacturing output grew rapidly. The share of consumer goods in imports dropped sharply, as the import pattern shifted toward raw materials and capital goods. However, the rapid growth of manufacturing in the early 1950s did not persist, and output growth became much more sluggish in the latter half of the decade. Philippine tariff policy follows a parallel history. In the years immediately after independence, the Philippines was severely limited in its ability to impose and collect tariffs. Seventy percent of Philippine imports came from the United States, and the country was bound under the Bell Act to a very gradual schedule of implementing tariffs on these imports. The transition period was substantially shortened with the revised trade agreement between

439

Philippines/Chapter 3

the United States and the Philippines in 1955 (the Laurel-Langley Agreement), which permitted 25 percent of Philippine tariffs to be applied to imports from the United States in 1956, 50 percent in 1959, and 75 percent in 1962. The Philippines introduced a revised system of tariffs in 1957 that duplicated the protective system established by the import control program, with the highest duties on nonessential consumer goods. While the tariffs were redundant as protective measures, they did succeed in siphoning off some of the rents that had been created by the import control program. The import control program itself had a rocky history. The initial surge in import prices led to the adoption of price controls for a variety of basic commodities. The import premium, as measured by the ratio of the black market to the official exchange rate, averaged 60 percent and reached 100 percent by the end of the decade. Charges of corruption and mismanagement dogged the import licensing process, even after authority for import control had been shifted to the central bank. By the end of the decade, pressure from the sugar bloc in Congress, continuing dissatisfaction with the corruption of the control system, slowing economic growth, and growing evasion of the exchange control system by exporters led the Philippines to dismantle the system of import control. Decontrol took place over a three-year period; by 1962 the peso had been devalued to P. 3.9 per dollar and licensing requirements for imports had been removed. The impetus behind decontrol was largely reaction to the graft involved in the import control system and the feeling that exporters had been able to evade foreign exchange surrender requirements, rather than a conscious attempt to diminish the extent of protection of domestic industry and shift resources toward the exporting sector. In order to protect domestic industry from the effects of decontrol, the Philippine government raised duties on roughly 700 items, and further increases in tariffs followed (Baldwin 1975, 57). In addition, the share of Philippine duties applicable to imports from the United States under the Laurel-Langley agreement jumped from 50 percent to 75 percent in 1962, significantly raising duties on the large majority of Philippine imports. Even with the rise in tariff rates, the elimination of import licensing significantly lowered the degree of protection afforded to import-competing industries. Despite the rise of f.0.b. peso import prices by 101 percent from 1959 to 1962, the wholesale price index of imported goods rose only 22 percent. Measurements done by Baldwin (1975, 58, 100) indicate that the smallest price increases occurred for the most heavily restricted goods. However, despite the change in relative prices, decontrol does not appear to have led to industrial restructuring nor to the emergence of new industries geared toward exports. Manufacturing output growth slowed significantly in the wake of decontrol, as indicated in table 3.4. The slowdown was spread across all sectors of manufacturing, and there is little evidence of any structural change in this p e r i ~ d Manufacturing .~ exports increased, but not

440

Robert S. Dohner and Ponciano Intal, Jr.

dramatically, and not enough to raise the sector’s growth rate. Manufacturing profits in general fell after 1961. Industry groups, particularly the Philippine Chamber of Industry, became more vocal in their complaints about excess capacity and competition and in their demands for support of existing Philippine Protests against foreign investment increased, and economic nationalism became a more potent force in the 1960s. In contrast, the fortunes of the traditional export sector increased dramatically in the decontrol period; from 1959 to 1963 the wholesale price index for export goods increased by 20 percent relative to the index for local manufactures. The output response of the traditional sector was swift, although measuring the effect is complicated by the underreporting of exports in the late 1950s and by the increase in the U.S. sugar quota in 1962. Even with corrections for these effects, exports increased by about 50 percent in dollar terms from 1959 to 1966.’ Afterward, the rate of export growth slowed considerably, averaging only 1.5 percent between 1966 and 1969. An additional effect of the rise in the relative price of traditional exports was to switch agricultural land from food production to the production of export crops. The result was a rapid rise in domestic prices of foodstuffs, in addition to the inflation resulting from the exchange rate depreciation. This in turn led to a general drop in real wages. The food price inflation and the decline in urban real wages were the most politically sensitive of the decontrol outcomes, and they spawned domestic protests, particularly in 1963. The prevailing opinion of the decontrol period in the Philippines was one of disappointment. Filipinos viewed the early 1960s as a period of relative stagnation, excess capacity, rising prices, and falling real wages. The failure of the decontrol period to spur economic growth weakened the ground of economic liberals, and would shift their attention to export promotion, without trying to change the existing incentives for resource allocation. Finally, the decontrol experience produced a more favorable view toward economic planning and government market interventions. An outgrowth of this sentiment was the enactment of a comprehensive system of industrial incentives in 1967, as well as a financial facility for distressed firms set up at DBP in 1966. The Philippine government also intervened more actively in allocating resources among industries after the decontrol episode. The Board of Investments (BOI) that was established by the Industrial Incentives Act of 1967 had substantial discretion in administering the incentives, as well as the authority to limit investments in industries with excess capacity. The latter half of the 1960s saw more rapid growth and a recovery of the manufacturing sector, produced by the expansionary policies of the Marcos government. As the balance of payments worsened, there was a gradual reintroduction of exchange controls. But the Philippines did not return to comprehensive import controls before, or during, the payments crisis of

Philippines/Chapter 3

441

1970. The IMF stabilization program called for a depreciation of the peso, and the Philippine economy recovered rapidly in the early 1970s. One of the first economic policy measures of the martial law government was a tariff reform program in 1973, which was designed to limit the number of tariff rate categories and lower the dispersion among tariff rates. This was a tariff reform in only a limited sense. The 1973 tariff program increased the average tariff rate by 3 to 4 percent, and maintained the cascading structure of the Philippine tariff system (ILO 1974, 113). The tariffs established at this time remained until the early 1980s. The structure of protection, as indicated by the sector or end-user estimates of effective rates of protection and by the average effective exchange rate, is presented in table 3.5. Because of the limited ability to impose or collect tariffs under the terms of the Bell Act, the Philippines developed a system of domestic sales taxation, differentiated by product and, in some cases, by domestic or foreign origin. The estimates of the effective Table 3.5

Philippine Protection A. Effective Protection Rate

All sectors Exports Sugar Nonexportables Import competing Import noncompetingb Agriculture and primary Manufacturing Capital goods Intermediate goods Inputs for construction Consumer goods

I965

I974

48

36 4

28

37 I48 9 44

35 67 3 36 25 33 31 42

- 19"

12

183 59' 83' I04 51 16 27 55 70

18

23 16 77

B. Average Effective Exchange Rate (pesos/dollar)

Traditional exports New exports Essential consumer imports Nonessential consumer imports

1950-59

1960-69

1970-80

2.00 2.30 2.06 3.64

3.46 3.70 3.91 10.55

6.60 7.99 8.12 25.48

Sources: Power and Sicat (1971). Tan (1979), Baldwin (1975). and Senga (1983). "Excluding sugar bManufacturing sector only 'Import noncompeting industries are those in which imports amount to less than 10 percent of domestic production. 'kighted average of Power and Sicat's estimates for agriculture (17 percent), forestry ( - 26 percent), and mining ( - 17 percent).

442

Robert S. Dohner and Ponciano Intal, Jr.

rates of protection, which were drawn from Tan (1979) and Power and Sicat (1971), take into consideration the differential protective effect of these domestic product taxes. An alternative estimate of sector protection is the effective exchange rate, the number of units of domestic currency actually paid or received per dollar in a given transaction. These were computed by Baldwin (1975) and Senga (1983), and include the effects of multiple exchange rates, tariffs and export taxes, discriminatory domestic sales taxes, subsidized borrowing rates, and margin deposit requirements. Both measures are shown in table 3.5. During the 1960s and 1970s the highest effective rates of protection were given to the import-substituting consumer goods industries (primarily the nonessential consumer goods according to the central bank's classification system). The table also indicates the much lower protection given to exports and to the agricultural and primary sectors. The estimates in table 3.5 show that, despite the revisions in the tariff code in 1973, the general pattern in the structure of protection remained the same during the 1970s. This conclusion is strengthened when the effects of tariff exemptions during the 1970s are included. Soon after the 1973 tariff reform, duty exemption privileges were granted to a number of public institutions and private enterprises. During the martial law period these exemptions became more widespread. The ratio of estimated duty exemptions to actual duties paid on all imports reached at least 22 percent during the latter 1970s from the 9 percent level during 1973-74 (Alburo and Shepherd 1986, 60). These duty exemptions were mostly on capital and raw material imports. Thus, the effective rate of protection overstates the protection given to capital goods industries and understates the actual protection given to consumer goods industries. Despite the professed aim of the martial law government to revamp the system of incentives and encourage exports, Philippine trade policy remained firmly committed to encouraging production for the domestic market. Among countries in Southeast Asia, only Indonesia provided a higher effective rate of protection to imports relative to that given to exports (table 3.6). This orientation of trade policy was reinforced in the 1970s by nontariff barriers. Nontariff barriers had been the principal means of industrial protection during the 1950s, but were deemphasized during the 1960s as part of the decontrol program. The 1970s and 1980s saw their increasing use, for reasons of balance of payments adjustment, industry protection, support of local content programs, safeguarding of public health and national security, and centralizing of importation.' Nontariff barriers paralleled the structure of protection of the tariff system; by 1977, 62 percent of all consumer good items were regulated, as compared to 24 percent of mineral fuels and lubricants, 23 percent of intermediate goods, and 17 percent of capital

443

Philippinestchapter 3 Effective Rates of Protection, Selected Southeast Asian Countries

Table 3.6 Country

Year

Exportables

Importables

Total Manufacturing

Philippines Indonesia Malaysia Singapore Thailand

I974 1975 1978 1979 1980

4 - 13 6

61 98 53

44 39 34 3 65

4 24

1

89

Source: Findlay and Gamaut (1986, xix). Note: Effective rates of protection weighted by value added. Industries are classified by whether net exports

are positive or negative.

goods.’ Although tariff rates stayed constant after the 1973 tariff revision, the increase in nontariff measures raised the average level of effective protection given to domestic industry in the 1970s and increased the variability of protection across industries.

3.2 Fiscal Incentives and Export Promotion The granting of fiscal incentives has been an important tool of industrial promotion in the Philippines since independence. Several laws granting fiscal incentives for investments in priority industries were enacted, with some modifications over time in their nature, mechanics, and industry coverage. The incentives initially took the form of exemptions from domestic taxes, but broadened over time to include exemptions from customs duties (especially on imported capital equipment and parts), tax deductions, and tax credits. At the same time, the granting of fiscal incentives, initially indiscriminate, became somewhat more defined with the passage of the Investment Incentives Act of 1967 and the subsequent establishment of the BOI. Nonetheless, there was a high degree of continuity among the incentive recipients; most of the firms that benefitted during the 1960s were the same ones that benefitted from fiscal incentives during the 1950s (Power and Sicat 1971, 79-82). The Investment Incentives Act of 1967 and the Export Incentives Act of 1970 provided the framework for fiscal incentives during the 1970s. The Industrial Incentives Act distinguished two priority sectors: “preferred” industries, where existing capacity was considered to be smaller than what the domestic market and likely export potential could support, and “pioneer” industries, which would introduce new products or processes to the Philippines. Investments in preferred industries by registered firms benefitted from accelerated depreciation, tax exemptions on imported capital equipment, and tax credits on the purchase of domestic capital equipment. Pioneer industry investments were eligible for the same benefits, as well as exemptions from all internal revenue taxes except the corporate income tax.

444

Robert S. Dohner and Ponciano Intal, Jr.

In addition, firms in pioneer industries could be wholly foreign-owned, while the maximum foreign ownership for registered firms in preferred industries was 40 percent. The act also allowed exporters access to inputs at world market prices by providing a tax credit for import taxes paid. The act established the BOI, attached to the Ministry of Trade and Industry, to identify priority sectors through its Investment Priorities Plan, process applications, and administer the incentives. In addition, the BOI was given authority to limit investment in “overcrowded industries,” an outgrowth of the complaints over excess capacity and competition during the decontrol period. Support for exporters was liberalized and extended by the Export Incentives Act of 1970, which gave registered exporting firms tax and duty free imports of capital equipment, tax deductions for various business development expenses, and duty drawbacks on imported intermediate goods. The act, and a later revision by presidential decree, also provided a subsidy through tax deductions to domestic employment and procurement by exporting firms for a five-year period. This subsidy was calculated by formula and did not fully compensate firms for purchasing inputs domestically at higher than world market prices. The incentives under the act were later supplemented with presidential decrees establishing export processing zones and bonded warehouses, and with a preferential rediscounting facility at the central bank to encourage export finance. The industrial incentives legislation embodied a two-pronged approach that would characterize Philippine industrial policy until the 1980s. The export promotion features of both the Industrial Incentives and Export Incentives Acts provided subsidy to exporters and, more importantly, allowed them to acquire inputs at world market prices. However, the adoption of these measures was in addition to, rather than instead of, policies that encouraged production for the domestic market. The Philippines retained the existing system of trade protection for domestic industry and supplemented it with industrial incentives and, in many cases, further protection. Thus, industrial incentives extended the system of import substitution measures, while simultaneously seeking to counteract their discouraging effect on exports. Two additional features of the industrial incentives were important in the Philippines. Rather than an open-ended encouragement of industrial activity, the availability of incentives depended on industry capacity measures-what policymakers thought the domestic market and likely export potential could support. Once capacity in an industry reached that level, the industry was removed from the Investment Priorities Plan and no further incentives were given to new entrants or to expansion. In addition, in 1970 thirty industries, including cement and textiles, were designated as overcrowded, and neither new investment nor expansion could take place in these industries without B01 approval. lo

445

Philippinesichapter 3

This focus on capacity carried over to the design and implementation of the incentives themselves. Incentives were given for the establishment of firms and new capacity, rather than for the profitable operation of these firms. Although the legislation included some incentives for expansion and replacement, these were deemphasized in favor of new investment. And once firms were established, the incentive system and restrictions on investment and entry tended to protect them from competition. The second feature of the industrial incentives was their effect in encouraging the use of capital. The most important of the incentives were duty exemptions on imported capital equipment and accelerated depreciation of investment, both of which lowered the cost of capital to participating firms. The capital-cheapening effect of the full range of BOI incentives has been estimated to have been between 39 and 42 percent." In addition, firms that were registered with the BOI had preferred access to low interest rate funds from state financial institutions, which further reduced their cost of employing capital. The bias toward capital intensity was also a characteristic of the drafting of the Investment Priorities Plan and the implementation of the incentives program. In the 1970s, Philippine policymakers decided on a strategy of industrial deepening, encouraging the development of domestic industrial intermediates producers and, later, capital equipment producers. The largest industrial recipients of BOI incentives during the 1970s were copper smelting and refining (36 percent of total benefits in 1977), pulp and paper, chemicals and chemical products, and synthetic textile fibers. Intermediate industries are in general capital intensive, and of the total distribution of industrial and export incentives, almost two-thirds went to industries with above average capital intensity (World Bank 1980, 31 -32). Two additional programs followed this strategy of industrial deepening. The first was the Progressive Manufacturing Program (PMP), which was originally adopted for automobiles in the mid-1970s and later extended to trucks, motorcycles, and consumer electronics. The PMP was designed to force the use of locally manufactured intermediates by progressively reducing the allocation of foreign exchange for imports. In exchange, participating firms received protection from foreign competition and from new domestic entrants. Despite the small size of the Philippine domestic market, each program attracted a number of participants, five for automobiles and nine for the truck program.13 The second program was the Major Industrial Projects (MIPS), described in the previous chapter on government expenditure. Evaluating the overall importance of the industrial incentives is a difficult task. In aggregate they were relatively small, amounting to about 0.6 percent of GNP in 1978 and 0.8 percent in 1985. The incentives were heavily concentrated in manufacturing, where they were about 3 percent of value added and 10 percent of profits. However, for individual firms the incentives

446

Robert S. Dohner and Ponciano Intal, Jr.

were often quite large and important. A survey of 164 BOI registered firms in 1985 found that BOI incentives represented 30 percent of value added and 100 percent for firms in the metal products industry.14 In addition, the BOI provided subsidy without expenditure in a variety of ways. As mentioned above, BOI incentives were often supplemented by access to low cost funds from state financial institutions. In addition, many of the nontariff barriers that were introduced during the 1970s were designed to protect registered firms in the industrial incentives programs and were administered by the BOI.

3.3 Reform in the 1980s After years of analysis and growing criticism, both within and outside the Philippines, the system of trade protection and industrial incentives began to change in the early 1980s. Under the first World Bank structural adjustment loan (SAL), the Philippines agreed to a tariff reform and import liberalization program. The tariff reform program called for a substantial reduction in rates, as well as a more uniform structure of tariffs, to be phased in between 1981 and 1985. In tandem with the tariff reform, the Philippines agreed to an import liberalization program designed to eliminate licensing requirements for imports of most consumer goods, as well as some intermediate and capital goods. The tariff reform program was completed on schedule and changed the character of Philippine tariff protection substantially. The average tariff rate declined from 42 percent in 1980 to 28 percent in 1985.15 The effective rate of protection for manufacturing declined from 4 4 percent in the 1974 tariff system to 36 percent in 1985 (see table 3.5). The disparity in effective rates of protection among exports, import-competing manufactures, and importnoncompeting manufactures was reduced. However, the data also indicate that the gap between the primary sector (agriculture, forestry, and mining) and the manufacturing sector widened because of the more than proportionate decline in the average effective rate of protection for the primary sector. In contrast to the tariff reform program, import liberalization proceeded much more fitfully. Firms whose products were removed from import licensing requirements in many cases appealed successfully to President Marcos, who later issued presidential decrees reestablishing import control for many products that had been removed from the list. By 1986 the Philippines had made little or no progress on import liberalization. The Philippine government also agreed to reforms in its investment incentives program under the two SALs. The Investment Incentives Act of 1983 eliminated a number of benefits that were based on the use of capital and substituted instead tax credits based on value added and net local content. As a result, the capital-cheapening effect of the incentives program was substantially reduced.

447

Philippines/Chapter 3

These changes in trade and industrial policy came late in the game for the Philippines and were quickly superceded by other events. The tariff reduction program was partially reversed by import surcharges that were imposed starting in 1982 for balance of payments purposes. After the debt moratorium in October 1983, the Philippines reverted to foreign exchange allocation, effectively establishing the same system of protection by degree of essentiality that had long characterized Philippine policy. And in a more fundamental sense, the Philippine attempt to even the balance of incentives across industries was overshadowed by growing exchange rate overvaluation in the early 1980s, a subject to which we now turn. 3.4

Exchange Rate Policy

The history of Philippine exchange rate policy is characterized by consistently delayed responses to overvaluation, followed by large, discrete changes in the exchange rate. In almost every case, devaluation came after the emergence of a balance of payments crisis, and often required substantial external pressure. As a result, the Philippines has failed to achieve a real exchange rate that was both stable and at a level that encouraged the development of Philippine exports and profitable import substitution. Despite a much higher wartime inflation than in the United States, the Philippines maintained its pre-World War I1 exchange rate after independence. The government responded to the payments crisis in 1949 by imposing import and foreign exchange controls during the 1950% and pursued a conservative monetary and fiscal policy that resulted in a level consumer price index for the decade of the 1950s. The peso was devalued in 1962 as part of a decontrol program which was brought about by balance of payments difficulties and stagnating economic growth at the end of the 1950s. This marked the first Philippine devaluation since 1903. Expansionary monetary and fiscal policy during the last half of the 1960s led to a balance of payments crisis in 1969-70. In response, the government initially imposed foreign exchange rationing; eventually, the peso was floated in 1970 and it subsequently depreciated by 51 percent. The Philippines officially followed a managed float during the 1970s, in which the government would intervene in the foreign exchange market solely to smooth out fluctuations around the trend. However, except for a 7 percent depreciation in 1975, the changes in the peso-dollar rate were very slight and averaged only a 1.6 percent annual rate of depreciation between 1973 and 1980. Under pressure from a widening current account deficit in the 1980s, and increasing difficulties in raising external funds, the Philippines began gradual devaluations of the currency: 5.2 percent in 1981 and 8.1 percent in 1982. In the crisis year 1983 there were much larger depreciations and a fall in the peso exchange rate from P. 9.2 per dollar at the end of 1982 to P. 19.8 per dollar at the end of 1984.

448

Robert S. Dohner and Ponciano Intal, Jr.

It is of course the real, and not the nominal, exchange rate that governs allocation of resources. The decontrol and 1962 devaluation produced a substantial real depreciation that was gradually eroded, but only partly reversed, in the 1960s. During the 1970s and early 1980s the Philippines experienced a number of large changes in relative prices, making a description of the course of “the” real exchange rate difficult. Table 3.7 presents several relevant definitions of the real exchange rate over this period. Although the series in table 3.7 vary, they tell a roughly similar story. The devaluation of 1970 created a significant depreciation of the real exchange rate that was extended by the international commodity price boom between Table 3.7

Philippines Relative Price Indexes fTradedi PNontraded

Year

(1)

REER Export Markets (2)

REER Asian Competitors (3)

Terms of Trade (4)

PExportsi PGDP (5)

Manufacturing Real Wage (6)

94 95 90 118

N.A. I08 I07 I03

109 100

105 100

88

I25 159 I25

87 77 86

78 71 78 82 69

I02 95 98 I05 96

81 86 85

90 74 91

~~

1967 1968 1969 1970

85 88 92 97

73 73 74 109

1971 1972 1973 1974 1975

100 100 110 115

1976 1977 1978 1979 1980

107 I05 102

I02

101

101

97

99

104 101 104 97 92

1981 1982 1983 1984 1985

93 91 93 103 98

96 91 109 109 96

93 93 109 I09 93

60

1986 1987

92 90

I I7 123

100

102

60 65

112

79 80 82 Ill

127 123

107

I06

100

100 108

Ill 100 I I3 115

I08 96 I04

101

I08

I00 I07

121 1 I9

59 61 60 56

85

91

83

N.A. N.A. N.A. N.A. N.A.

88 91

N.A. N.A.

101

Notes: N.A. = not available. (1) Traded goods prices are a weighted average of gross value added deflators for agriculture and forestry,

(2) (3) (4) (5)

(6)

mining, and tradable manufactures. Nontraded prices are a weighted average of deflators for construction, electricity and gas, and services. Weights are 1972 value addeds. REER is real effective exchange rate. Dollar wholesale prices in major Philippine markets divided by dollar prices in the Philippines. Markets are the U S . . Japan, Germany, Netherlands, and Korea. Increase i s real depreciation. REER is real effective exchange rate. Weighted average of consumer price indexes in dollars in Korea, Indonesia, Malaysia, Thailand, and Singapore, divided by Philippine CPI in dollars. Weights are 1980 exports. Increase is real depreciation. Export unit value divided by import unit value. Gross value added deflators from national accounts. Basic manufacturing wage divided by GDP deflator. Wage series discontinued in 1981.

449

PhilippinesiChapter 3

1972 and 1974. After 1974 there is a gradual fall in the competitiveness of Philippine traded goods production, one that shows up most heavily in the commodity-price-dominated relative price of exports in column (4). What stands out from the table is the fact that the real exchange rate did not increase sharply during most of the 1970s, despite the rapid domestic economic growth and the high rate of foreign borrowing during this period. The reason that the Philippines avoided the significant currency overvaluation that characterized many of the LDC borrowers is that the external borrowing was accompanied by an equivalent rise in investment expenditure, the equipment portion of which was almost entirely imported. Therefore, the domestic demand impact of the increase in Philippine expenditure was smaller than in other L D C S . ’ ~ The real appreciation that occurred in the Philippines occurred relatively late, in the period after 1979, with a sharp appreciation of the real exchange rate measures in the three years to 1982. This was the period in which a variety of events raised the domestic inflation rate-the unsuccessful attempt at economic stimulus in 1980 and 1981 and the domestic financial crisis in early 1981. The real appreciation of the peso also coincides with the recovery of the U.S. dollar from its 1978 trough. Since the Philippines maintained an almost stable exchange rate with the dollar, the peso in large part followed the course of the U.S. real exchange rate. While the swings in the real exchange rate are not dramatic, what is striking from the table is the secular fall in two important measures. The first is the substantial drop in real wages, starting with the 1970 stabilization episode and continuing through the decade. Philippine wage data is not particularly accurate, but the fall in real wages is difficult to reconcile with the apparent growth of real GNP during the decade.” The second and contrary indicator is the almost unbroken fall in the country’s terms of trade, starting in the 1960s. Even at their peak in 1974, Philippine terms of trade were below their 1969 level; by 1981, Philippine terms of trade had declined by 50 percent. Thus, the backdrop against which these real exchange rate changes took place increases their importance. With the kind of adjustments the Philippine economy was forced to make, the real appreciation that took place after 1979 was a major blunder. A number of reasons explain the tendency of the Philippine government to delay exchange rate adjustments to payments imbalances during the postWorld War I1 period. The decision to maintain the prewar exchange rate was in fact out of Philippine hands; the country had agreed to forgo exchange rate changes without the approval of the United States, in exchange for U.S. war rehabilitation funds. The historical precedent of having maintained a fixed exchange rate from the beginning of the century until 1962 may have also created a reluctance to devalue. But two other reasons played the more important role in Philippine reluctance to adjust the exchange rate until external payments crisis made it

450

Robert S. Dohner and Ponciano Intal, Jr.

inevitable. The first was the identification in the minds of Philippine policymakers of the exchange rate as a nominal magnitude associated with domestic inflation, but not with the allocation of resources (at least outside of agriculture) nor with the development of new industries. The second reason was the development of a strong domestic constituency opposed to devaluation, and the failure to develop a constituency for devaluation that was both influential and legitimate. With a constant exchange rate and conservative macroeconomic policy the Philippines achieved a nearly stable price level during the 1950s. When the peso was devalued in 1962, the increase in earnings from sugar and coconuts led to a shift of land away from food crops, resulting in higher domestic food prices and a higher rate of inflation. The domestic inflation rate subsided in the late 1960s, despite expansionary policy, in part due to the increase in rice yields from the development of new high yield varieties. In the early 1970s, the combination of devaluation, higher external prices for export crops and rice, and extremely bad weather in 1972 was responsible for the acceleration of inflation. The link between exchange rate changes and domestic food prices, both through direct import costs and the diversion of land into export crops, was viewed as the key determinant of domestic inflation in the Philippines by the middle 1970s, both in the minds of policymakers and in analysis of price change in the Phi1ippines.l8 This has led both to a reluctance to devalue and, when devaluation took place, to taxation of export producers, often used as a way of funding domestic price stabilization schemes. In addition, the lack of development of new manufacturing export industries during the early 1960s created a skepticism about the ability of exchange rate changes to create new traded goods industries. Perhaps most important for describing exchange rate policy in the Philippines was the lack of a strong domestic voice arguing in favor of lower exchange rates and the avoidance of overvaluation. In the first decades after independence, those who would benefit from devaluation had little public legitimacy. Devaluation would have further increased the income and political power of the sugar interests, who were resented by many Filipinos for their ostentatious life style, and who already benefitted from the U.S. sugar quota premium and the Bell Act proscription on export taxes. Similarly, the perceived existence of foreigners in agricultural trading made a peso devaluation (which was thought to benefit the foreign traders more than the farmers) inconsistent with the nationalist, pro-Filipino aspirations of the time. By the 1970s, the sugar bloc, the most important pressure group among the traditional exporting industries, had declined in political and economic power because of the termination of the U.S. sugar quota in the early 1970s and the increased control of the industry by the government and its appointed sugar administrator, Eduardo Cojuangco. The nontraditional garment and

451

PhilippinedChapter 3

semiconductor exports were growing so fast during the 1970s that the exchange rate did not appear to be a critical bottleneck. In addition, investors in these industries were either foreign or new domestic industrialists without a significant power base. The only group which consistently called for a more realistic exchange rate during the decade was a small group of government technocrats and academics, as well as the multilateral agencies, and their voice did not carry sufficient weight in domestic politics. The establishment of import-substituting industries in the 1950s led to the emergence of an important pressure group that was opposed to peso devaluations. With heavy tariff and nontariff protection, the output of many industries became essentially nontraded, and peso devaluation would have lowered the prices of these goods relative to tradables. Because their imported raw materials and machinery were fully traded, with complete pass-through of exchange rate changes, a peso devaluation would have resulted in a profit squeeze for these industries. This is what had happened in 1962; manufacturing profits in general, and especially those of industries highly favored by the 1950s control system, declined substantially with decontrol and peso devaluation (Power and Sicat 1971, 43). As a result, the potential of the exchange rate in shifting resources, developing new industries, and spurring adjustment was never exploited. In the 1970s and especially the early 1980s, the Philippines allowed a real appreciation of the peso, despite external signals that a major devaluation was in order. Philippine trade and exchange rate policy never addressed the problem raised by the substantial secular deterioration in the country’s terms of trade and its buildup of external debt.

3.5 The Structure of Philippine Industry We turn now from a discussion of policy to outcomes. Trade policy and the structure of incentives since the 1950s encouraged the establishment of a largely import-substituting manufacturing sector. Import substitution was concentrated in consumer goods during the 1950s. As a result, the share of consumer imports to total imports declined from 37 percent in 1949 to about 23 percent during the early 1950s, and dropped further to about 14 percent by 1960 (Power and Sicat 1971, 39). Table 3.8 presents the ratio of imports to domestic supply in selected manufacturing industries. The table indicates that domestic substitution of food imports was largely completed during the 1950s. Import controls also accelerated import substitution in wearing apparel, and publishing and printing. Among producer goods, import controls led to the substitution of domestic production for imports of textiles, paper and paper products, and nonmetallic mineral products during the 1950s. The pace of import substitution slackened during the 1960s, particularly for intermediate goods industries such as textiles and basic metals. This

452

Robert S. Dohner and Ponciano Intal, Jr.

Table 3.8

Share of Imports in Domestic Supply (percentages)

Industry Food manufactures Dairy products Textiles Wearing apparel Paper & paper products Publishing & printing Rubber Basic industrial chemicals Other chemicals Petroleum & coal products Cement Other nonmetallic minerals Basic metals Fabricated metal products Nonelectnual machinery Electrical machinery Transport equipment

1948i49

1961

I969

1974

1979

32

9 46 13 49

N.A. 39 23

N.A 21

86 32 98 35 N.A. 90 b

b

N.A. 63

24 21

N.A 45 40 3 34 14 13 70 28 6 14 20 52 28 84 52 57

N.A. N.A. N.A. N.A. N.A.

10 11

33

54 31 75 51 62

1

25 I1 15

80 37 6 I 23 51

28 84 45 59

18

0 23 12 12 58 22 12 I 15

37 19 70 55 52

1983 2 26 22 2 27 10

25 57 20 10

0 10 30 30 50 56 48

Source: NEDA, Philippine Inpur-Ourpur Tables. various years. Data for 1948149 from Baldwin (1975) Nore: N.A. = not available.

"Included in food manufacturing. bIncluded in basic industrial chemicals. 'Included in cement.

reflects the impact of the import decontrol and peso devaluation, which increased potential product competition from abroad at the same time that (imported) input costs increased. Import substitution picked up again during the 1970s, particularly for textiles, basic metals, and fabricated metal products. This second round of import substitution was the result of the increased use of nontariff barriers. Nontariff barriers specifically for the purpose of industry protection were imposed on imports of textile fabrics, synthetic yams, fibers and paper board products, synthetic resins, and liquid caustic soda among others. What is clear from table 3.8 is the extent to which the import substitution strategy has dominated the development of Philippine industry. Some sectors have progressed farther than others in replacing imports with domestic production, but by 1983 domestic output provided more than 40 percent of domestic supply in all sectors listed, and 70 percent or more in most. This across-the-board representation of Philippine industry carries on within individual industries. For example, Philippine textile plants are unspecialized, integrated operations, designed to serve the full range of domestic market demand. This lack of specialization and the relatively small Philippine market has meant higher costs and, in the case of textiles, lower quality and a higher yam fault rate. l9

453

Philippines/Chapter 3

3.6 Growth of Exports In contrast to the broad range of production for the domestic market, exports have been much more concentrated within certain industries, as is evident from table 3.9. The vast majority of Philippine manufactured exports have depended either on domestic natural resources (wood products, furniture, and within chemicals, coconut oil) or labor-intensive manufactured exports within apparel and electrical machinery. Philippine merchandise exports, broken down into major commodities, are shown in table 3.10. As is clear from the table, export earnings hit a peak in 1980 when favorable crop yields coincided with high international prices, but then declined into the 1980s. The share of merchandise exports in GDP shows an uneven pattern, as it was affected by real exchange rate changes and changes in the terms of trade, but little change over the entire period. Behind the merchandise export total are two divergent trends. There was an extremely rapid growth in exports of manufactured products after 1972, in both nominal and real terms. In contrast, the growth of earnings from the traditional export sector was more sluggish, and total dollar earnings

Table 3.9

Ratio of Exports to Total Output

Industry

1961

1969

1914

I979

1983

Other food manufactures Beverages Tobacco manufactures Textile Wearing apparel Leather & leather products Lumber, plywood & veneer Other wood products Furniture & fixtures Paper & paper products Rubber & plastic products Basic industrial chemicals Other chemical products Petroleum & coal products Cement Other nonmetallic minerals Basic metals Fabricated metal products Nonelectrical machinery Electrical machinery & parts Transport equipment

N.A. 0.2 4.5 2.0

N.A. 1.2 5.5 I .5 18.7

8.5 1.3

15.1 0.4 0. I 12.9 29.3 54.6 35.8 58.9 32.8 1.7 4.2 24.0 2.1 I .6 5.2 5.3 11.7 1.9 2.9 51.1 4.5

12.9 1.1 0. I 15.3 37.0 54.9 26.6 20. I 52.3 3.4 2.9 13.9 2.4 4. I 1.6 4.1 2.6 16.2 3.1 70.4 5.8

0.0 -

-

13.0

15.5 43.9 3.9 1.7 1.4 6.3 2.2 7.1 1.7 1.1 8.3 0.6 1.8 2.5 2. I

0.9

7.2 b

1.0

6.1 9.3 11.1 27.5 N.A. 34.0 2.8 1 .o 6.1 I .6 1.2 21.2 7.4 6.7 0.6 4.0 1.7 0.3

Source: NEDA, Philippine Inpur-Ourpur Tables, 1961, 1969. 1974, 1979, and 1983. Note. N.A. means data were not available, and a dash indicates the amount was negligible.

"Is included in lumber, plywood & veneer. bIs included in basic industrial chemicals.

L'IL L'E9 P'ZI 8'EZ (S.91) OZL'S 860'1 611'1 ZP9'E

Ill 601 PZZ 0 EPZ IL 18E 195 19E'I 1861

1.08 1'19 P'8 9'ZZ (9'SI) ZP8'P 8Pl Po6 986'Z LOP 06 19Z 9z I 0z 80 I EEE OLP 960'1 9861

S'6S 1'19 1'6 8'SZ

("PI) 6Z9'P 619 9S0'1 lS8'Z 61V P8 ZW 6E E61 191 8ZE

opp 961'1 5861

6'09 2'85 8'1 6'62 (L.91) 16E'S 009 6ZE'I 9EI'E 81P s11 E8E 88 591 ZlZ EPS 069 019'1 P86 I

0'06 6'8P

S'8 S'8E

9'66 P'9E E'8 1.1s

1'66 P'IE 1'8 8'95

0861

Z861

Z18 EP6' 1

181 6S6'Z

EIZ S8S

LET

06s

(Z'tl) ZZP'E 9ZE ESZ 910'1 L62 0sz PES SP 1 PZE

(P.91) 881's 00s IL9 601'2 ESP SPS 891'1 26 OZP

(9'ZI) IZO'S 6ES 1 LSP'Z 6ZP ZIE 989 61 062 96E WE E9S SE6' I

ooo'

(SlSllOp

8161

0'88 6'LI L'L 1'Zl

(S'PI) P6Z'Z 001

1P 1IP 9 11 ZIZ S9E L9 I

szz 519 9zz Z9P L99' 1 SL61

y n JO suo!(l!uc u!)

0'001 9.8 1.L

c '18 (I'EI) 901'1 Z Z 56

S8 161 WZ 991 9zz LIZ E8 1ZZ 016 Z161

dl!porumo~dq

~ i d x au!dd!i!qd a

OI'E JlclU

455

Philippines/Chapter 3

declined sharply after 1980. The combination of both of these factors was responsible for the marked shift in the structure of Philippine exports in the 1970s and continuing into the 1980s. The rapid growth of nontraditional manufactures exports appears to confirm the success of the Philippine strategy of export promotion, and was one of the primary reasons for the optimism concerning the Philippine economy among international agencies and commercial banks during the 1970s. Several factors were behind this rapid export growth. The first was the provision of investment incentives and incentives for domestic value added contained in the Industrial Incentives Act of 1967, the Export Incentives Act of 1970, and later amendments to these two acts. While not unimportant, the effect of the direct incentives to export was relatively small. A World Bank mission calculated the value of export incentives as constituting between 3 and 9 percent of the exports of registered firms between 1973 and 1977.20 A later estimate put the effect of the incentives for domestic employment and procurement at between 5 and 10 percent of value added, depending on the input structure and profitability of the firm.” This conclusion is strengthened by the fact that much of the export took place through bonded warehouses, which were ineligible for the incentives. Much more important were the provisions of these two acts, supplemented by the establishment of export processing zones and the authorization of bonded warehouses, which allowed exporters to purchase inputs at world market prices. It was this access to inputs, plus the changes in real exchange rates during the 1970s, that made the rapid growth of nontraditional exports possible (see table 3.7). Even more important for the kind of export industries that developed was the fall in real wages of approximately 25 percent that took place in the early 1970s. By the mid-l970s, Philippine wages for garment workers were roughly half that of Korea and a quarter of those in Hong Kong.22 Philippine manufactured exports are highly concentrated in a few products. Electrical components and garments together make up about 60 percent of the total of manufactured exports. Within electrical components, exports are concentrated in the assembly of semiconductor devices and microcircuits, and within garments, on finishing of consigned imports of women’s wear and garments for infants and children. Each industry depends heavily on imported inputs. Estimates of domestic value added vary, but are about 25 percent for garments and 15 percent for electronic components. These ratios appear to have remained constant in the Philippines, unlike other less developed country exporters which have succeeded in raising their value-added margin through subcontracting. The weak performance of traditional export commodities counteracted the rise in manufactured exports. The growth in export receipts from the traditional commodity export sector during the 1970s was entirely due to inflation; a quantity index formed using 1978 weights varies over the decade,

456

Robert S. Dohner and Ponciano Intal, Jr.

but ends up at the same level in 1980 as in 1972 (see table 3.10). Over the next four years the quantity of Philippine traditional exports declined by 39 percent, with a little over half of the decline coming in 1984. The importance of the stagnation and decline of the traditional export sector is hard to overstate; had the quantity of these exports grown at the same rate as real GDP and all other things remained equal, Philippine exports would have been 53 percent, or $2.8 billion higher in 1984. There are various explanations for the poor performance of the traditional export sector. The Philippines was hit by the low growth in international demand for commodities after the first oil shock and the accompanying weakness in their prices. Two features peculiar to the Philippines were also responsible for reductions in export quantity. At the beginning of the 1970s, exports of forest products, mainly logs, made up a quarter of Philippine traditional exports. However, the deforestation that had occurred in the Philippines led to increased concern for conservation and increasingly severe restrictions on log exports. By 1980, recorded log exports had dropped by 90 percent. 23 The second feature is the dependence of Philippine sugar exports on U.S. quota allocations. The Philippines is a higher than average cost producer of sugar, and historically almost all Philippine exports have gone to the United States under quota. High sugar prices and the dismantling of the U.S. sugar quota in the 1970s led to a diversification of destinations, but the importance of the U.S. market was reestablished with the reinstitution of U.S. import quotas in 1981. And, as total U.S. imports have been reduced in the 1980s, Philippine exports have been correspondingly squeezed. But much of the responsibility for the performance of the traditional export sector must be laid to Philippine policies. Philippine trade policy has consistently discriminated against the sector, promoting import-substituting manufacturing and then nontraditional export products. As a consequence, the commodity export sector has suffered negative effective protection, as outlined above. The 1970s saw the institution of additional policies that discriminated against traditional exports. After the 1970 devaluation of the peso, Philippine authorities introduced taxes of 4 to 10 percent on most commodity exports. Although supposedly temporary, these measures were never repealed and became a permanent feature of martial law policy.24 Coconuts, the most important of the traditional products, had the largest taxation. A levy on the first sale of copra was introduced in 1973, initially to finance a price stabilization fund for domestic coconut products, but later to support a Coconut Industry Development Fund. The amount of the levy varied somewhat over the next ten years, but averaged about 20 percent of the export price, and eventually exceeded 30 percent. Thus, at a time when world prices were turning against commodity producers, the Philippines raised the effective taxation of the commodity sector. Beyond this, government and quasi-governmental intervention

457

PhilippinedChapter 3

increased dramatically during the martial law period in the most important sectors, coconuts and sugar. By the end of the decade, each was under the control of a Marcos crony, who had a monoply over purchases from the industry. These interventions are described in more detail in the following chapter, but the siphoning off of rents from sugar and coconuts, in addition to the taxes that were collected, further weakened the incentives for producers in those industries. We are now in a position to answer one of the questions posed at the beginning of this chapter, that is, why there was such a sluggish performance of exports relative to gross domestic product, while at the same time there was a dramatic change in the structure of Philippine exports and a rapid growth in exports of manufactures. Behind this lay the fact that Philippine export promotion had a very narrow base. What Philippine export promotion measures did was allow producers to obtain imported inputs at world market prices, leading to the development of export reprocessing based on imported materials and the low wages of Philippine labor. The retention and augmentation of the system of protection for manufacturing firms producing for the domestic market meant that value-added margins of these export producers would stay very thin; the higher cost and lower quality of domestic materials precluded the growth of domestic sourcing. The high degree of protection of the domestic market also tended to limit export products to industries where materials transport costs were low and labor input requirements high. Garments and electronic components fit those requirements perfectly, and export growth was highly concentrated in these two sectors. Thus, Philippine export growth was intensive rather than extensive. The rapid transformation of the structure of Philippine exports is misleading since it was based as much on the low growth of traditional export products as it was on the rapid growth of nontraditional exports. Unfavorable external conditions were partly responsible, but these were greatly exacerbated by Philippine policies which explicitly and implicitly taxed the traditional sector. The rapid shift in export structure, propelled from both ends, resulted in the narrowing of the value-added base of the country’s exports since the domestic content of the nontraditional manufactures was much lower than that of Philippine agricultural and mineral commodities. Thus, as exports of manufactured goods expanded, imports of materials increased pari passu. The resulting structure was one in which a greater increase in gross exports was necessary to generate a net increase in foreign exchange earnings when external funding faltered.

3.7 Trade Policy and Manufacturing Performance The remaining question raised at the beginning of this chapter concerns the weak performance of Philippine manufacturing in the 1970s and early 1980s. Productivity growth for the sector as a whole was negative throughout the

458

Robert S. Dohner and Ponciano Intal, Jr.

1970s, with a particularly sharp fall after 1975. Employment generation was also low, despite the growth of labor-intensive manufactured exports. Finally, the growth rate of manufacturing output started to decline after 1977 and largely collapsed in the 1980s, again despite the continued growth of manufactured exports. As described above, Philippine trade policy continued to protect the domestic manufacturing sector during the 1970s through existing tariffs and an increasing use of nontariff barriers, despite the adoption of export promotion. The domestic manufacturing sector remained closed to foreign competition, as well as being insulated from the exporting segment of the industry. As a result, the growth of the sector was largely determined by domestic demand and by industrial incentives policy. The industrial policy that was in fact adopted by the Philippine authorities encouraged the development of intermediate industries as a part of a process of industrial deepening. This policy thrust was reinforced by the shift in domestic demand toward investment and the particularly rapid growth of construction, activities with an especially high demand for materials. The outcome was a pronounced shift in manufacturing output growth toward intermediate industries, as well as capital goods, during the last half of the 1970s (table 3.11).

Growth Rates of Real Manufacturing Production

Table 3.11

(in percentage change per year) 1975-80

1970-75 Fastest growing Beverages Machinery Plastics Apparel Furniture Food

23.0 23.0 22.2 16.3 14.5

13.3

Total manufacturing Memo items: Consumer goods Intermediates Capital goods Food Petroleum products Other K-intensive" L-intensiveb

Nonferrous metals Industrial chemicals Footwear Paper products Wood products Glass Transport equipment

32.9 28. I 27.6 23.7 21.9 21.3 21.2

6.8

5.4

13.1 3.3 6.2 13.3 10.8 6.8 8.8

2. I 9.5 12.4

I .2 - 2.0

13.8 7.3

Source: Hooley (1985). using data from the annual surveys of manufacturing adjusted to reflect firms with twenty or more workers. Nore: Growth rates are valued at 1972 prices.

'Capital-intensive industries include paper, chemicals. nonmetallic minerals, glass, iron and steel. and nonferrous metals. bLabor-intensive industries are the remaining industries, except food and petroleum products

459

Philippines/Chapter 3

As is clear from table 3.11, output growth was also concentrated in industries with above average capital intensity. This was partly the result of the shift in demand toward intermediates, described above, as well as the nature of industrial incentives and government-sponsored finance, which lowered the cost of capital. Again, this took place despite the rapid growth of labor-intensive manufactured exports. The shift toward capital-intensive industries provides an explanation for the low growth payout of Philippine investment (or, alternately, the country’s high ICOR) since capital-intensive industries require more investment per unit of output. In addition, the shift toward these industries helps explain the low employment generation of Philippine manufacturing. Shifts in the industry composition of manufacturing output also lie behind much of the poor productivity performance of Philippine manufacturing. During the 1970s, most manufacturing industries registered positive total factor productivity growth. For the decade as a whole, within industry productivity growth added 0.34 percent per year to total factor productivity in manufacturing, while interindustry shifts lowered manufacturing productivity by 1.57 percent per year.25 Other factors lay behind the poor performance of the manufacturing sector. During the martial law period there was increasing government intervention in the industry, as well as increased activity by publicly owned corporations. From 1965 to 1980 the share of government corporations in total assets of nonfinancial corporations doubled to 27 percent (Hooley 1985, 28).26 The importance of this in explaining the low productivity growth of the sector has been stressed by Hooley, who found total factor productivity in government corporations to be only 56 percent of that of privately owned corporations (29-30). In addition to direct government intervention, firms owned by Marcos associates played an increasing role in domestic activity, largely through acquisition. The impact of government intervention and crony capitalism is discussed in more detail in the next chapter, in which we argue that both played a role in the deteriorating performance of domestic industry. Trade, exchange rate, and incentives policies played a crucial role in the slide of the Philippines toward debt crisis. Although deceptively successful in developing manufactured exports and shifting total export composition, Philippine policy failed to respond to the worsening external environment and the country’s foreign debt accumulation. The depletion of forestry resources and the secular decline in the terms of trade should have led to a real exchange rate depreciation that would have provided across-the-board encouragement to exports and efficient import substitution. Instead the Philippines allowed a gradual appreciation of the real exchange rate, encouraged manufactured exports while maintaining and strengthening the protection of domestic industry, and increasingly taxed the traditional export sector. The result was a significant hollowing of the country’s export base, greatly reducing its foreign exchange earning capacity.

460

Robert S. Dohner and Ponciano Intal, Jr.

The combination of trade protection, investment incentives, and domestic growth propelled by investment and construction took its toll on Philippine industry. Output shifted toward more capital-intensive, lower productivity industries, limiting the growth that the Philippines got out of its investment and foreign borrowing. But these industries were also dependent on the continued momentum of borrowing and investment and proved extremely vulnerable to the recession of the early 1980s. The sharp declines in many of these industries worsened the recession in the early 1980s, and many of these firms ended up in the hands of the government, either through rescue operations or the assumption of guaranteed external loan obligations.

4

Government Interventions and Rent Seeking

In the popular imagination the legacy of the Marcos administration was the accumulation of vast wealth by Ferdinand Marcos, his family members, and various individuals, or “cronies,” who were closely associated with him. Corruption and the accumulation of wealth through government did not originate with Marcos, nor was it unusual in the Philippines as opposed to other countries in or outside the region. But the scale on which corruption and the generation of rents took place in the Philippines under Marcos was at a qualitatively different level. What observers in the Philippines referred to as crony capitalism, and what less charitable observers outside the country referred to as “government by kleptocracy,” was of such a scale as to have macroeconomic consequences, and plays its own important role in the slide of the Philippines into crisis. The use of government power to generate and distribute wealth, what economists have termed “government rent seeking” had three critical consequences. First, quasi-governmental control and monopolization of the two principal commodity export crops, sugar and coconuts, was responsible for much of the sluggish growth of traditional and total exports. Second, the particularistic way in which the government issued regulations and granted access to credit to favored firms and the way in which crony business empires were built weakened and demoralized the private, nonassociated business sector and encouraged capital flight. Finally, when the crony empires dissolved in the 1980s, the government was left with a huge burden of failed assets, called loan guarantees, and unmet domestic payment obligations, creating a fiscal problem of major dimensions. We start our story with two of the most insidious interventions, those in sugar and coconuts.

461

PhilippineKhapter 4

4.1 Government Interventions in Sugar and Coconuts The exports of the Philippines have historically been dominated by coconut and sugar products, although logs and lumber were very important during the 1960s. Coconuts and sugar accounted for about three-fifths of total exports during the 1950s and about one-half during the 1960s. Their export share dropped, however, to one-third by the late 1970s and to about one-fifth by the early 1980s. This drop was partly a consequence of the sharp rise of nontraditional manufactured exports during the period, but also reflected sluggish growth in output and weak international prices for the two crops. Both industries had grown rapidly in the ten years before martial law, and there were optimistic projections and investment commitments in both sectors. But the world market turned out to be far less favorable than foreseen, forcing severe problems in each industry. In addition, the organization and trading arrangements changed decisively during martial law, in ways that weakened both sectors and impeded adjustment.

4.1.1 Sugar Before the 1970s, the sugar industry benefitted enormously from its access to the protected U.S. market, which offered prices substantially higher than the world market. The absence of export taxation in the Philippines meant that much of the U.S. sugar price premium was captured by Philippine producers. Government involvement in the industry was limited to the allocation of the quota for export to the United States, although the Philippine government forced sugar producers to satisfy a domestic sales requirement in order to export, and this kept domestic prices below the U.S. level. Because of the privileged access to the U.S. market, the small number of sugar producers, and the highly skewed distribution of sugar land ownership, several sugar planters and millers became enormously rich. Using their wealth, they also diversified into other areas of the Philippine economy. Their economic prominence translated into political influence and, for a few of them, tremendous political clout during the pre-martial law days. They formed part of what Marcos called the Philippine oligarchy, who he attacked shortly after the martial law declaration and tried to replace with his own cronies. The conflict within the ruling elite provided the sociopolitical subtext to the government interventions in the economy in general and in the sugar industry in particular during the 1970s. Two events in the early 1960s greatly benefitted the Philippine sugar industry. The first was the devaluation and import decontrol that took place between 1960 and 1962, which shifted price incentives in favor of the export sector. The second was the suspension of Cuba’s quota for sugar exports to the U.S. market after Castro nationalized the industry in 1960, and its assignment to other producing countries.

462

Robert S. Dohner and Ponciano Intal, Jr.

Philippine land area planted to sugar increased by 80 percent between 1960 and 1972. The Philippine government also supported the expansion of sugar milling capacity through financial support from the Philippine National Bank. The number of operating sugar mills rose by 40 percent between 1967 and 1974. The 1970s began a period of upheaval for the industry. During the commodities boom of the early 1970s the world price of sugar rose sharply from under 7 cents to over 30 cents a pound in 1974. As important for the Philippines was that the world market price rose above the controlled price in the U.S. market. During that year, the U.S. Congress failed to renew the Sugar Act of 1948 and the U.S. quota system came to an end. The termination of the U.S. sugar import quota forced the Philippine sugar industry to operate within a far more volatile world market. The price of sugar declined dramatically in 1975 and 1976, rose sharply in 1980, and then declined precipitously during the early 1980s. Beyond the volatility in prices, developments in the world sugar market during the latter 1970s and early 1980s tended to depress the secular trend in world sugar prices. These included the decline in per capita sugar consumption in a number of developed countries, the significant inroads into the sweetener market, particularly for industrial purposes, of sugar substitutes such as high fructose corn syrup, the substantial productivity improvements in sugar beet production, and European Economic Community price support policy which turned the EC into a net sugar exporter.’ In order to capture some of the windfall from the 1970 devaluation, the Philippine government imposed export taxes on commodity exports, with the highest rates on sugar, copra, and log exports. In 1974 the government imposed premium taxes on sugar and other commodity exports to capture some of the international commodity price rise. Government control of sugar trading started that year when, after the increase in world sugar prices and the growing scarcity of sugar in the domestic market, the government ordered PNB, the major financier of the sugar industry through crop loans to planters and investment loans to millers, to purchase the sugar crop. The bank’s subsidiary, the Philippine Exchange Company (PHILEX), was given responsibility for all sugar exporting. The rationale behind the takeover of sugar trading was to stabilize domestic sugar prices and prevent private hoarding. It drew on the historical bias in the Philippines against middlemen (who tended to be Chinese Filipinos) in the agriculture sector and the popular feeling that profits from agriculture had often gone to the “monopolistic” traders and not to the farmers. In taking over sugar trading the government declared its “single agency” concept, wherein a single trading agency would replace “a system of excessive dependence on individual selling efforts, coursed traditionally through brokers or middlemen [in order to have] better control of supply and more efficient marketing (Marcos 1979, 114-15).

463

PhilippinesiChapter 4

The decision enabled PNB to capture large rents from the export price surge of 1974, since the prices paid to farmers were substantially below prices realized from sugar sales. At this point PHILEX overreached itself. Expecting a continued rise in world sugar prices, the agency held back its sales from the world market, accumulating inventories in the Philippines. When world prices dropped, PHILEX was forced to sell the already deteriorating stocks of sugar at a substantial loss in 1977. And, in order to maintain the domestic purchase price, PHILEX was forced to borrow heavily. In 1977 control over sugar trading was transferred from PNB to the recently created Philippine Sugar Commission (Philsucom) headed by Roberto Benedicto, a fraternity brother of Marcos’ from his law school days.3 The trading arm of Philsucom, the National Sugar Trading Corporation (NASUTRA), was given sole authority to trade sugar domestically and internationally, as well as to set purchase prices for milled and unmilled sugar. Under Philsucom’s direction, the Republic Planters Bank, controlled by Benedicto, became the principal private financial institution for the sugar industry. Although the government takeover of sugar trading had ostensibly been undertaken to limit monopoly rents in the industry, it in fact substituted its own monopoly for the one that had allegedly existed before. The net effect of the government’s direct interventions in the sugar industry during the 1970s and early 1980s was that the price that sugar producers received declined as a share of the world market price. Before 1974 the domestic sugar producers received a price for sugar roughly 60 percent higher than the world market price, due to premium enjoyed under the U.S. sugar import program. Although domestic sales, accounting for about 35 percent of Philippine production, were below the U.S. price, they still were higher than the world market price. After 1974 there was a dramatic shift toward effective taxation of the sugar industry. Retail prices of sugar were highly visible and politically sensitive, and the Philippine government sought to keep them below world market levels. During the 1974-82 period, domestic prices for sugar averaged only 69 percent of world market prices (Nelson and Agcaoili 1983,23). In addition, prices paid by NASUTRA were in most years well below realized prices on the agency’s international sales. A University of the Philippines workshop study estimated that the difference amounted to P. 5.4billion between 1974 and 1984, or 17 percent of the revenue that NASUTRA r e a l i ~ e d If . ~ we use the International Sugar Agreement price as the relevant border price for the Philippines, there was a net income transfer to the sugar farmers averaging 38 percent of the value of output during 1960-7 1; during 1972-82 there was a net income loss to sugar farmers averaging 30 percent (table 4.1). NASUTRA and its parent agency, Philsucom, expanded the government’s reach in the sugar industry by acquiring and operating the leading enterprises

464

Robert S. Dohner and Ponciano Intal, Jr. Nominal Rate of Protection and Transfers: Sugar

Table 4.1

Transfers" (P. million)

Nominal Rate of Protcction (%)

Year

ISA

1960-64" 196S-69b I970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

57 170 49 35 -8 - 27 - 63 - 36 -5

Ratio of Transfer to Value of Output ('70)

XUP

ISA

XUP

ISA

-9

44.9 475.9 458.0 41 1.4 - 195.6 - 909.4 -6.576.7 - 3,146. I -255.8 303.3 497.0 773.7 - 5,801.9 - 1,266.6 2.789.3

-60.2 -132.1 -437.2 -411.4 - 434.6 -761.9 -4,185.2 -6,685.4 - 1,278.9 - 808.9 3.0 48.9 - 1.498.0 -3.217.5 - 532.5

18.0 61.5 32.4 25.6 -9.7 - 38. I -161.1 - 56.9 -5.3 8.7 4.8 18.1 - 117.9 - 22.3 42.2

- 17 - 24 - 20 -

18

- 24 - 52 - 55

- 21

10

- 19

18 22 - 54 - 18 73

7 37 - 23 - 37 -7

XUP

- 10.8 - 18.2 -30.9 -25.6 -21.5 -32.0 - 106.3 - 120.8 -26.3 -23.2 6.3 26.9 - 30.4 - 58.5 -8.1

Source: lntal and Power (1987) aTransfers are equal to the difference between the actual price received by the producer and the border price. multiplied by the volume of output. 'Annual averages. ISA = border price based on the International Sugar Agreement daily price XUP = export unit value.

in transport, bulk storage, and handling of sugar and sugarcane, and by establishing new sugar refineries and operating sugar centrals (Intal and Power 1987, 47-48). The construction of sugar mills was a source of corruption and wealth generation. Before the declaration of martial law, a Senate Blue Ribbon Committee challenged the Marcos government on the construction of sugar mills by Japanese firms that had no previous experience in mill construction, and the fact that the cost of these mills substantially exceeded the cost of construction of mills of equivalent capacity in other countries (Canlas et al. 1984, 88). Despite the crisis due to mounting sugar inventories in 1975 and 1976 and problems already existing from excess milling capacity, the Philippine government allowed the establishment of new sugar centrals by politically favored individuals, including the country's sugar administrator, Roberto Benedicto. The profits in these sugar mills usually occurred during their construction through overpricing of the project accompanied by kickbacks (Wideman 1976, 54-55). The investments in turn were financed by loans or guarantees from PNB. During the period after 1974, productivity in the domestic sugar industry stagnated. Output growth during the 1960s and 1970s was through extensive means-expansion in sugar hectarage and the establishment of sugar mills.

465

PhilippineKhapter 4

As sugar growing was extended to less suitable land, the growth in output was accompanied by a decline in sugar yield per hectare. It was only with the reduction in sugar hectarage during the late 1970s and early 1980s that there was some improvement in farm productivity; nonetheless, farm productivity and processing efficiency during the late 1970s and early 1980s remained below that of the early 1960s. After four years of low levels, world sugar prices jumped sharply in 1980, leading to a dramatic recovery in the Philippine sugar industry. At the height of the price cycle, NASUTRA signed contracts to deliver 565,000 tons of sugar per year (about half of Philippine exports) between 1981 and 1984, at a price of 23.5 cents per pound. This proved to be a particularly fortunate move, as world prices plummeted in the 1980s, dropping to just over 5 cents per pound in 1984. In 1981, the United States reinstated its sugar import quota, giving the Philippines a quota of 342,000 tons for the 1982 crop year. Production costs in the Philippines have been estimated to be 14- 15 cents per pound, well above the world market price in the 1980s. The combination of the long-term contracts signed by NASUTRA and U.S. quota sales at about 18 cents per pound sustained the industry in the early years of the decade. But the U.S. import quota shrank significantly in succeeding crop years, and the long-term contracts NASUTRA had signed expired in 1984. The next year was a particularly disastrous one. World sugar prices fell further, and Philippine production fell by 16 percent. NASUTRA was unable to maintain the domestic producer support price for sugar and failed to pay many producers for the 1985 crop year. This in turn led to a collapse in the sugar industry support system and widespread malnutrition in some sugar producing areas such as Negros Occidental. In 1986 the Philippine sugar crop fell precipitously to a level of 35 percent below that of 1984. The Philippines was forced to import sugar in order to meet its (now much smaller) U.S. import quota. The troubles of the Philippine sugar industry reflect in part the adjustment problems of an increasingly less efficient producer, which no longer had privileged access to a large and protected export market. However, the nature of the government interventions and the manner of their implementation aggravated the structural adjustment problems of the Philippine sugar industry. The failure of the Philippine government in its interventions in the sugar industry during the 1970s was due fundamentally to its focus on the control of the sugar trade and on industry expansion, rather than on industry rationalization and an increase in farm productivity, processing efficiency, and crop or product diversification. The bullish government assumption about the long-term trends in the world sugar market proved to be far too optimistic. What aggravated the adjustment problem in the Philippine sugar industry was that the two government interventions provided opportunities for rent seeking and political control. The PHILEX price speculation fiasco, export taxation, and the relative inefficiency of the marketing operations of

466

Robert S. Dohner and Ponciano Intal, Jr.

the government sugar trading agency exacted a heavy burden on the sugar producers. 4.1.2

Coconuts

Government intervention in coconut pricing and marketing has been significant only since the 1970s. Nonetheless, the interventions have been more controversial than those in the sugar industry for several reasons. First, the interventions affected a much larger proportion of Filipino farmers than the interventions in sugar. Despite the historical prominence of sugar in Philippine public policy, sugar farms account for only about 1 percent of the total number of farms and about 4 percent of the total farm area. In contrast, coconut farms account for nearly one-fifth of all farms and nearly one-quarter of farm area. Second, the interventions created a parastatal but legally private bureaucracy that disposed of large funds and was outside the purview of government auditing regulations. Finally, although designed to make long-term investments and assist in restructuring and crop replanting in the industry, the funds collected were used to effect the vertical integration of the coconut industry, to establish monopoly control, and ultimately to enrich the Philippine defense minister and a crony of the Marcos government. Government intervention in the coconut industry came initially out of pressure from the Coconut Producers Federation (COCOFED), the largest organization of planters. COCOFED pressed for and got the Philippine Congress to pass Republic Act 6260 in 1971 creating the Coconut Investment Fund and the Coconut Investment Company to administer it. The aims of the Coconut Investment Fund were to establish regional banks in partnership with the farmers and the central bank, to mobilize bank loans for long-term investment in coconut marketing and processing, and to encourage manpower development (ILMS 198 1). In the vertical integration program pushed by COCOFED, the farmers would gain control of the trading and processing subsectors of the coconut industry. The Coconut Investment Company was to be funded from a small levy on the first sale of copra, and a part of the levy was to go to the support of the federation as the primary representative of the industry. The proposals of COCOFED were essentially reactions to the apparent neglect of the sector by the government and antipathy toward the middlemen who controlled processing and marketing. Before the 1970s, the government barely acknowledged the coconut industry and did not try to promote coconut exports, nor were there programs of research or agricultural extension in the industry. In addition, a substantial share of copra financing and trading and coconut oil manufacturing was controlled by Chinese Filipinos and foreigners. The later interventions in the industry were able to draw on sentiment against these two groups; the use of the coconut levy during the 1970s to establish a bank, buy coconut mills, and establish

467

Philippines/Chapter4

coconut marketing centers were efforts at “de-alienization” of the coconut trading and processing sectors (Intal and Power 1987). The next initiative came from the martial law government in 1973. Early in the year the government had established the Philippine Coconut Authority to implement policy in the industry. A small crop in the Philippines and for other coconut producers led to a shortage of copra and high world prices. The sharp rise in the world price of coconut oil led to domestic scarcity of coconut oil, resulting in acute shortages of cooking oil, laundry soap, margarine, and other coconut-based consumer products. In response, the government imposed price controls on domestic coconut-based products and created the Coconut Consumer Stabilization Fund (CCSF), which the Coconut Authority used to compensate product manufacturers who were caught between spiralling input costs and the price ceilings imposed by the Price Control Council. The CCSF was funded through a levy of 150 pesos per metric ton, or roughly 12 percent of world prices. Although the stabilization fund was supposed to be a temporary measure, its aims and those of the Coconut Investment Fund were merged in 1974 by a presidential decree that allowed the stabilization fund to set aside part of its revenues to fund investment, extension, and research and development in the industry. Later in 1974 a second presidential decree created the Coconut Industry Development Fund, to be funded out of the accumulated levies of the stabilization fund as well as the levy on copra sales, now raised to 200 pesos per metric ton. One of the first acts of the development fund was to finance the establishment and operation of a new hybrid coconut seednut farm, which grew a higher-yielding MalaysiadIvory Coast variety, and was owned by Eduardo Cojuangco. The seednut farm was to be the basis of a long-term replanting program for the industry. In December 1974 the Philippine Coconut Authority’s governing board was reorganized and was now made up of the chairman and president of PNB and five members from COCOFED. In 1975 the Authority approved the use of funds collected from levies on coconut sales to purchase a bank, to be owned by the farmers, that would provide finance to the coconut industry. The Coconut Authority bought the major interest in the ailing First United Bank, which was renamed the United Coconut Planters Bank (UCPB). The bank’s president was Eduardo Cojuangco, who also had a 7 percent ownership share, and its board chairman was Juan Ponce Enrile, Marcos’ defense minister. The accumulated funds of the stabilization fund and half the collections from the investment fund were deposited in UCPB without interest, spurring rapid growth of the bank. A presidential decree announced in 1978 allowed the COCOFED and UCPB to use the funds from the coconut levies to make investments in coconut milling and to purchase existing mills on behalf of the coconut farmers. After several mills had been acquired, UCPB organized the United Coconut Oil Mills (UNICOM) in 1979, which became the vehicle for

468

Robert S. Dohner and Ponciano Intal, Jr.

downstream investment from the coconut levy funds. The bargaining position of UCPB and UNICOM in negotiations for the purchase of existing mills was strengthened by a presidential decree which limited price subsidy restitution payments to mills “owned by the farmers” (Hawes 1987, 72-73). The effort to purchase oil mills also gained from the severe financial losses experienced by many millers in 1979, when a fall in copra production pushed the capacity utilization rate of oil mills below 50 percent. By 1980 UNICOM owned thirteen coconut oil mills representing 80 percent of the country’s coconut oil milling capacity, and managed two more, bringing its total to over 90 percent (Ocampo 1980, 45). The taxation of coconut production and the control that UNICOM exerted over coconut milling created tremendous opportunities for siphoning off income from the industry. Despite the number of agencies involved, actual control of the industry was vested in very few people. One of the directors of COCOFED, Eduardo Cojuangco, was also head of the Philippine Coconut Authority, president of UCPB and UNICOM, as well as owner of the hybrid seednut farm. The collections under the Coconut Consumers Stabilization Fund and the Coconut Industry Development Fund were never subject to audit. One study done by the Philippine planning ministry, NEDA, put the total levy collections at P. 10 billion (about $1 billion at 1982 exchange rates), of which only P. 2.1 billion was spent to reimburse coconut products producers (NEDA 1985, IV-48). A life insurance scheme and a variety of scholarship and other assistance funds were set up for coconut farmers and their families, but actual disbursements under these programs were very small. The bulk of the funds went to Cojuangco’s hybrid seednut farm, making him the richest crony in the Philippines. In addition to the taxes that were levied on coconut producers, the control that UNICOM established over coconut oil milling and the restrictions that the Philippine government placed on direct export of copra created a monopsony buyer of copra within the Philippines, further depressing the returns that coconut farmers got from their crop. UNICOM appears to have paid between 9 and 15 percent below the price it would have paid under competitive conditions for copra supplies (Clarete and Roumasset 1983, 34). The power that UNICOM exercised over the industry was demonstrated in 1981 when the coconut levy was suspended at the insistence of Finance Minister Cesar Virata. UNICOM refused to buy copra or sell coconut oil and within five days Marcos restored the levy.5 As a result of the levies and the control over processing facilities, the effective taxation of coconut producers increased dramatically during martial law. The combined effects of the levy and purchase arrangements on the incomes of coconut farmers have been estimated by Intal and Power (1987) and are shown in table 4.2. The domestic copra producer price as a percent of the border price decreased during the 1960s and fell precipitously after 1979. Transfers from (forgone income of) coconut farmers averaged 5

469

PhilippinesKhapter 4

Table 4.2

Year

Nominal Protection Rate and Transfers: Coconuts

Nominal Rate of Protection (70)

1960-Mb 1965-69b 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

I .8 -4.2 - 20 -11 - 16 9 - 18 - 33 -5 - 24 - 14 0 - 40 - 34 - 35

Transfers" (P. million) - 1.5

- 37.4 -315.7 - 164.6 - 185.9 185.4 -914.5 -911.8 - 125. I - 1,264.5 -913.7 -33.8 -3,217.5 - 2,208.5 -2,083.0

Share of Transfers to Total Value of output (%) 1.1 -4.6 - 24.3 - 12.7 - 19.2 7.9 -21.7 -49.4 -5.1 -32.1 - 15.8 -0.3 -67.7 -50.4 -54.6

Suurce: lntal and Power (1987).

"Transfers are equal to the difference between the actual price received by the producer and the border price, adjusted to the farm level, multiplied by the volume of output. bAnnual averages.

percent during 1961-71, 19 percent during 1972-78, and 43 percent during 1979-82. As in the case of sugar, the coconut industry was hampered by low productivity growth and the absence of additional lands to open up for cultivation. Yields per hectare declined during the 1960s and rose only modestly during the 1970s, despite higher world prices for coconut products.6 At the same time that the rate of effective taxation increased in the 1980s, average yields dropped significantly, and by 1984-85 total production of coconuts was 32 percent below its 1978-80 average.' Philippine coconut yields have been hampered by the growing senility of the stock of bearing trees, a problem that the replanting program was supposed to address.' In more general terms, the world market for coconut oil, the major product, weakened as a result of competition from soybean oil and palm oiL9 In this second instance, the Philippine government increased its taxation of a sector whose external terms of trade had deteriorated. 4.2

Crony Capitalism and Rent Seeking

The use of government power to distribute wealth went well beyond the traditional agricultural sector. The 1970s saw the virtual institutionalization of cronyism and rent seeking in the Philippines. While the associates of Marcos and the extent of their business operations were reasonably well

470

Robert S. Dohner and Ponciano Intal, Jr.

known, evidence that has come to light since the overthrow of the Marcos government reveals how extensive the interests of the first family and their relatives were in the operations of crony firms, and the extent to which they also gained from the associates they fostered. Underpinning cronyism and rent seeking during the 1970s were the centralization of economic decision making and the distortion of policies to suit particular firms or individuals. Under martial law, Marcos had almost unlimited discretionary power. Legislation could be accomplished through presidential decree; in many instances this was nothing more than a scribbled acquiesence and signature on a request that had been sent to the president. In some cases decrees were not made public until well after their issuance.” Access to Marcos became the ultimate determinant of policy; often policies that had been established by the ministries, with Marcos’ concurrence, were overruled by a later presidential decree. l 1 The dramatis personae of Philippine cronyism during the 1970s have become internationally known, particularly since the fall of Marcos from power in 1986. Apart from Ferdinand and Imelda Marcos themselves (who, as recent revelations show, extensively used cronies as agents), the most well known include Roberto Benedicto (Marcos’ sugar czar), Benjamin and Alfredo Romualdez (Mrs. Marcos’ brothers), Herminio Disini, Rodolfo Cuenca, Ricardo Silverio, Antonio Floirendo, and Eduardo Cojuangco. A few others, such as Jose Campos, acted as agents for the Marcos family. The use of government power to marshal and distribute wealth worked in several ways. The first might be termed standard graft-the allocation of government contracts and access to credit from public financial institutions to favored individuals in return for some interest or kickback from the operations. While this was certainly not unusual to the Marcos administration, the growth of the economy during the 1970s and the rapid expansion of the public sector increased the possibilities from this source. Thus for example, the Construction and Development Corporation of the Philippines (CDCP), run by Rodolfo Cuenca, received most of the public works and large construction projects of the martial law government. Roberto Benedicto’s firm, Integral Factors Corporation, became the exclusive agent of the Government Service Insurance System (GSIS) and had a virtual monopoly as the insurance broker for government properties. Jose Campos was granted the exclusive contract to provide all medical supplies to the Ministry of Health. The generation of rent through kickbacks on overpriced projects and contracts was not uncommon before the 1970s, but the magnitude increased substantially during the 1970s. The best known and most controversial case was the Bataan nuclear power plant project. Marcos chose the more expensive Westinghouse proposal brokered by Herminio Disini which gave Disini, and allegedly the Marcos family, substantial commissions, as well as the additional subcontracts in the construction of the plant that Disini also

471

Philippineslchapter 4

received. A recent report detailed the process of siphoning off Japanese aid “in the form of commissions or rebates from the purchase of equipment as well as technical and advisory fees for the implementation of development projects funded by yen credits from Japan’s OECF.”I2 The same report indicates that the actual prioritization of development projects funded with Japanese aid depended in part on the willingness of the Japanese companies implementing the projects to either pad costs or reduce profit margins, with the difference being remitted to the agents of the top Philippine government officials. Rent seeking through the overpricing of contracts is illustrated by the commission fee of 7.5 percent of the cost of shipment-instead of the standard 2.5 percent broker’s fee-charged against the Philippine National Oil Company (PNOC) on its oil importations, with the proceeds from the difference allegedly going to the foreign bank accounts of the top PNOC official (Veloso 1986). Similarly, the typical procedure in sugar mill investments was that the foreign supplier would price the investment package to include an allowance for kickbacks to the Filipino proponents and/or government officials. These investments in turn were financed primarily by loans granted or guaranteed by PNB (Wideman 1976). Wideman quotes a senior Japanese government official that Japanese businessmen in the Philippines spent on average 12 percent of contract prices on kickbacks (1976, 55). Historically, access to credit at below market rates was one of the spoils of government office in the Philippines. This continued under martial law, but the extent of foreign borrowing that took place and the expansion in size of government financial institutions and financial institutions associated with the government, greatly increased the resources available through this source, The two major government financial institutions, PNB and DBP, were heavily tapped by the cronies. “Behest loans,” loans granted at the request of the government or its agencies, dominated the loan portfolios of the two banks. In principle, behest loans were supposed to reflect the government’s priority areas where investors were given implicit interest rate subsidy given the lower interest rate charged by PNB and DBP relative to the prevailing market rate. Apart from the Marinduque Mining Corporation which was the largest borrower, the major borrowers from PNB were Benedicto, Silverio, and Cuenca (Quiambao 1986). The loan portfolio of DBP is less concentrated than PNB’s; nonetheless, apart from mining and cement firms, firms of cronies like Cuena, Disini, Benedicto, and Dewey Dee and relatives like the Martels figured prominently among the major borrowers (Tengco 1983). The result was that profits in many of the investment projects undertaken during the Marcos years were made at the investment and construction stage and not from the profitable operation of the facilities constructed. Public loans, or publicly guaranteed loans, removed the financial discipline in initiating and operating investment projects. This encouraged shoddy

472

Robert S. Dohner and Ponciano Intal, Jr.

construction, as in the case of the nuclear power plant, and in some cases the substitution of used equipment when new had been paid for. It also removed much of the market discipline, and led to overinvestment and excess capacity in such assets as hotels, sugar mills, and cement plants. But cronyism in the Philippines went well beyond the simple graft described above. Much of the generation and distribution of rents was done through the creation of monopoly positions or through differential and particularistic application of the law and regulation. The martial law period saw extensive interventions of this type. The monopoly positions created in the sugar and coconut industries were two of the most important, however, there were many other instances of government-mandated monopolies which generated rents for cronies or relatives. One of the most lucrative was the gambling monopoly, especially the profitable casinos and jai alai stadium, granted to Benjamin Romualdez. Rodolfo Cuenca’s Galleon Shipping Corporation was the only Philippine flag carrier permitted to operate container ships from America’s West Coast to the Philippines. A monopoly on meat importation from Australia and New Zealand was given to a private group which worked along with the Bureau of Animal Industries. The coconut-chemical plant set up by Cojuangco was given the sole right to import alkyl benzene, an input in making detergents, as well as the exclusive right to import products that would compete with the plant’s output. In another example of a firm being given exclusive rights to import the competing product, Peroxide Philippines was the only firm allowed to import peroxide (Sicat 1986, 29, 31; Canlas et al. 1984, 74). In some instances these monopolies were exercised directly by government agencies-the National Grains Authority, later renamed the National Food Authority, was given exclusive rights in wheat importation and domestic flour distribution (Sicat 1986, 23-25). Even Imelda Marcos’ Cultural Center of the Philippines earned substantial income as the country’s sole distributor of pornographic movies. In other instances, exemptions from taxes or duties or other differential application of the law created competitive advantages that conferred substantial benefits and, in some cases, monopolies, to the receiving firm. For example, Disini’s flagship firm became the largest and dominant seller of cigarette filters in the country when the government set a tariff rate of 100 percent on the raw material imports of competing foreign-owned firms, while Disini’s firm faced a tariff rate of only 10 percent (Sacerdoti 1983, 50). One of the reasons for the emergence of Ricardo Silverio’s firm, the Delta Motors Corporation, the sole assembler and distributor of Toyota automobiles in the Philippines, as the industry leader was that competing car assembly firms were allowed to offset through their exports of manufactured components only 15 percent of the local-content requirement for cars assembled and sold in the country, while Silverio’s firm was exempted from the rule (McDougald 1987, 211). Roberto Benedicto’s firm, Nivico, was

473

Philippines/Chapter 4

allowed tax free importation of unassembled black and white TV sets, ostensibly for distribution to the rural population; however, Nivico’s television sets were readily available in Manila. Antonio Floirendo’s fast rise to fortune occurred when his firm,TADECO, was able to secure the use of large tracts of land that were ideal for bananas from the Davao Penal Colony. As part of the plantation development agreement with the penal colony, Floirendo’s firm hired live-out prisoners who were paid much less than the prevailing wage rate (David, Barker, and Palacpac 1984). During the martial law years the distinction between public policy and private action for economic gain was often blurred. The case of the Bataan Shipyard and Engineering Company (BASECO) serves as an illustration. l 3 BASECO, a private corporation, was an offshoot of the president’s directive to privatize the ship repair and building industry as part of the national policy of relying on private enterprise as a catalyst for development. BASECO acquired, although never paid for, the assets of a government firm, the National Shipyard and Steel Corporation. A presidential decree required the government’s Maritime Industry Authority to draw up a shipbuilding program which, a confidential memorandum to the president from his brother-in-law stated, “would then be a source of ship orders for BASECO” (Espinosa 1986, 6). And BASECO did land millions of pesos worth of contracts with the Bureau of Public Works and the Philippine Navy. Marcos took a special interest in the corporation; he transferred the title for Engineering Island from the National Development Corporation to BASECO and ordered the Bureau of Public Works to improve the facilities at Engineering Island and Mariveles. He also intervened in BASECO’s application for loans. The investigation by the Presidential Commission on Good Government (PCGG), established by Corazon Aquino in 1986, explains Marcos’ active interventions: the firm was probably owned by him, and the published owners were largely front men. The use of government power to effect the transfer of assets from private titleholders to members of the Marcos “inner circle” at minimal or no cost was also a characteristic of crony capitalism. This included the forced sales of assets that were justified as ways to weaken the country’s oligarchs, but were as much matters of personal vengeance. This is best exemplified by the case of the Lopez family, whose controlling interests in the Manila Electric Company and newspaper publishing facilities were taken over by Benjamin Romualdez, and whose radio and television facilities were turned over to Roberto B e n e d i ~ t o . ’Another ~ prominent businessman who stood up to Marcos, Fernando Jacinto, had his business in steel smelting and processing (which had substantial government exposure) effectively nationalized and managed by military officers (Mijares 1976, 192). Various methods were used to effect asset transfer including: (1) the automatic foreclosure of mortgages on properties used as collateral in obtaining loans from government financing institutions; (2) the granting of equity shares to

474

Robert S. Dohner and Ponciano Intal, Jr.

Marcos or his cronies by businesses in undertakings requiring presidential approval; and (3) the setting aside of public lands for the ownership and business use of favored individuals (193-95). The business empires of the most prominent cronies were built through acquisitions of existing firms. In many cases, these sales were less than voluntary and occurred at below market prices, since Marcos, and cronies acting with his support, could bring substantial pressure to bear. In the case of a utility company, it might be the assurance that no rate increases would be forthcoming for the existing owners; in others, the firms would be threatened with labor troubles or closure for health and sanitary reasons. And in some cases, the recognition that government policy was skewed in favor of certain firms in the industry would cause existing firms to sell out, hoping to salvage some of the value of their assets by selling early. The rise of the cronies took place in a very short time, a period of just four to six years. With access to credit and the backing of the government, their business organizations expanded extremely rapidly in nearly all sectors of the economy. Cronies controlled wholly or owned a substantial share of businesses in such areas as agricultural export (sugar and coconut milling and trading, bananas), banking and finance, broadcasting and print media, construction, communications, car and truck manufacturing and distribution, gambling, mining, logging, electricity generation and distribution, pharmaceuticals, transportation, tobacco and beverages, real estate, machinery distribution, shipping and ship repair, and oil and coal exploration. Among the major cronies and relatives, Eduardo Cojuangco controlled or had substantial shares in around seventy firms, and Roberto Benedicto in some fifty firms. Benjamin Romualdez controlled or had substantial shares in around fifty firms, and Herminio Disini in fifty-one firms.I5 None of these practices outlined above were unique to the Philippines. Government corruption and favoritism, kickbacks from investment projects, grants of exclusive privilege, and pressure on asset holders to sell out to firms close to the current government, have all been features at various times of other countries. These practices were more extensive and more remunerative in the Philippines than elsewhere. But there still remains the question of why these practices seemed to have had such a debilitating effect on the Philippine economy, given their existence in other, more successful, economies. In fact, maintenance of below market interest rates, the allocation of bank credit to favored firms, and the encouragement of business concentration have been features in some of the most successful economies in the region. Arguably, crony capitalism was little different from the growth of zaibatsu in interwar Japan and the growth of the chaebol in postwar Korea. In those two countries economic concentration and government intervention on behalf of favored firms appear to have been powerful engines of growth, if not of economic equality. This was in fact the way in which the crony system was justified in the Philippines on the few occasions when the Marcos government was challenged on the issue of favoritism. The cronies,

475

PhilippinesKhapter 4

the government explained, were an energetic and entrepreneurial group that would do for the Philippines what similar industrialists had done in Korea and Japan. l 6 There were numerous differences between the Philippine experience with industrial concentration and that of Japan and Korea, but what was most fundamental was the inability to maintain and foster market competition. In both Japan and Korea there was a substantial degree of overlap in the industrial base of the large industrial groupings, so that in most markets there were several firms, each from a different industrial group. It was true that credit, foreign exchange, allowable investment, and other opportunities were channeled to firms in the large groups, but the amounts were primarily determined on a performance basis. Those firms that were the most successful exporters or who had the largest share of a particular market were rewarded with the largest allocation. While there was an ethos of fostering national firms, there were limits on the amount of discretion that bureaucrats could exercise and rarely, if ever, were individual firms or groups given a particular benefit at the expense of the others. Exporting was a priority in both Japan and Korea, each maintained relatively open trading systems, and there was little that the government could do explicitly to assure the financial success of a firm in the export market. In the Philippines, in contrast, there was a far greater degree of governmental discretion, as well as extensive application of the law and regulation in particularistic form. Almost anything could be arranged by presidential decree, and the president in many cases had a direct financial interest in the success of particular ventures. Unlike in Korea and Japan, there was little or no industrial overlap among crony enterprises; individual monopoly positions were created and awarded to single firms. Thus, there was little competition in individual industries. l7 Finally, the Philippine trading system was less open than that of Japan or Korea, and the country was less successful in channeling entrepreneurship into the export sector. Philippine cronies tended to focus their energies in nontraded goods sectors and the more protected industries. Among nontraded goods industries, the cronies had prominent shares in power, communications, construction, finance, transportation, and distribution services. They also had large shares in heavily protected industries such as automobile assembly and chemicals. The cronies were involved in export industries, but primarily in rent-gathering resource activities such as logging, mining rights, and control of sugar and coconut trading and processing. Because the nontradable sector tends to grow only in response to growth in national income, crony enterprise growth took place through the acquisition of existing firms rather than the establishment of new ones, and since the domestic market was protected from competition, income was almost assured from the exercise of monopoly positions.18 Thus the emphasis of the crony entrepreneurs was on the appropriation of rents rather than the generation of profits. Growth and efficient operation of firms was a

476

Robert S. Dohner and Ponciano Intal, Jr.

less attractive way to wealth than was extraction of surplus from the existing economy. By the end of the 1970s the scale of government favoritism and cronyism in the Philippines was such as to constitute a serious drag on the domestic economy. In some cases, such as sugar and coconuts, the taxes on producers were so great that they debilitated the underlying industry. In industry the effect was the demoralization of the private, nonassociated business sector. Since the cronies built their business empires extensively by acquiring firms in a range of industries, there was a reluctance on the part of domestic businessmen to become too large or too profitable and thus draw the attention of someone close to the Marcos government. The term used in the Philippines to describe the reaction of cronies to profitable firms was “saliva c a p i t a l i ~ m . ” ’There ~ was also a reluctance to enter new lines of business, since a firm could not be sure whether its competitor firms in the new activity were connected with the government or not. Businessmen, when they could, acted to protect their firms from the acquisitive impulse of the Marcos government and its cronies. Firms sought foreign joint venture partners on the assumption that the government would be less likely to move against the assets of a foreign firm. But even foreign-owned firms faced pressure to sell out to the cronies. Procter and Gamble, Lever Brothers, and Cargill were forced to sell their coconut mills to UNICOM. Foreign companies responded when they could by participating in U.S. or multilateral financial programs. Union Oil Company came under heavy pressure to sell its operations, but dissuaded the government by its participation in U.S. Overseas Private Investment Corporation (OPIC) insurance and its threats to force the United States to undertake a public investigation. Other companies used participation with the World Bank’s International Finance Company to discourage a takeover. The other way in which firms responded was by moving their assets beyond the reach of the government through capital flight. The issue of capital flight is dealt with below in chapter 6. As we argue in that chapter, the real increase in capital flight came well before the assassination of Benign0 Aquino, starting as early as 1980. As a result, less investment was undertaken by firms that were motivated by efficiency and profit, and more of domestic investment and activity shifted toward the government and those who could count on benefitting from government action. The failure of the economy to respond to the sharp increases in government investment expenditure during the early 1980s, in contrast to the response of the private sector after the first oil shock, was one indication of the weakening that had taken place.

*’

4.3 The Fiscal Burden of Crony Capitalism The rapid growth of the crony business empires was based on access to credit. Marcos associates borrowed from government financial institutions

477

PhilippinesKhapter 4

and, in some cases, directly from foreign lenders. But in addition, several of the cronies, Disini and Cuenca in particular, borrowed short-term funds from the domestic commercial paper market. In going to the domestic market, these individuals could draw on the strength of their names and on the widespread perception that they were close to the Marcos government. Since they were expanding rapidly, their operations were highly leveraged, leaving them vulnerable to a downturn in domestic business conditions or to financial shocks. Both occurred in 1981. Despite the attempts of the Marcos government to counteract the world recession, the rate of output growth slid in 1981. In addition, the sudden departure of Dewey Dee, a Chinese businessman who had borrowed heavily in the commercial paper market, led to a domestic financial crisis and the collapse of the short-term money market upon which many firms were dependent. During the Dewey Dee crisis, described in more detail in the next chapter, the central bank extended emergency funding to the financial institutions that had been caught in the liquidity crisis. Faced with the bankruptcy of numerous large firms, the Philippine government established an industrial rescue fund for corporations that had been affected. The industrial rescue fund was originally set at P. 1.5 billion, but the limit was later raised to P. 5 billion as the extent of the corporate distress became evident. As a result, the government corporate equity position increased dramatically beginning with this episode. The Philippine government, either directly or through public financial institutions, converted loans into equity and assumed the foreign obligations of the rescued firms, most of which carried public guarantees. Not all the firms rescued were those that belonged to Marcos cronies; the largest industrial failure of this period was the Marinduque Mining Corporation, which had built a nickel refinery with substantial government participation and which was hit by higher energy prices and the downturn in world metals markets. But many of the firms that were bailed out were those of the most prominent and aggressive cronies. One of the largest firms was CDCP, owned by Rodolfo Cuenca. In two letters of instruction in February 1981, President Marcos ordered PNB, GSIS, and two other state financial institutions to take over the company. Debts totaling P. 3.9 billion ($490 million) were converted into equity, and the government injected an additional P. 1.1 billion in new funds, giving the government a total interest of P. 5.1 billion in the firm.21 Management of CDCP was transferred to the National Development Corporation (NDC), but Cuenca was allowed to maintain a minority share.22 Ricardo Silverio’s Delta Motors Corporation was also forced under in the domestic recession and financial crisis. Debts to PNB of P. 1 billion were turned into equity, and an additional P. 150 million was provided, giving PNB a 70 percent share in the company. The extensive corporate empire of Herminio Disini fell apart as a result of the crisis. His Atrium Capital Corporation was the financial institution most

478

Robert S. Dohner and Ponciano Intal, Jr.

heavily affected by the crisis and received substantial support from the central bank. Disini’s business operations were highly leveraged: his principal firm, Philippine Tobacco Filters, had a 1O:l debvequity ratio. Of the 5 1 companies in the Herdis Group in 1981, 7 were ultimately retained by Disini, 18 were taken over by government agencies (17 by the NDC), and the remainder were sold, reorganized, or wound up (Ibon 1983b, 7-8). Total government exposure to the Herdis Group was estimated at P. 4.6 billion.23 Disini left the Philippines shortly thereafter and retired in Austria. The recession and financial crisis affected the cronies involved in the industrial sector. The cronies whose business interests were primarily in the agricultural sector-Eduardo Cojuangco, Roberto Benedicto, and Antonio Fiorendo-survived with most of their holdings intact. Both Cojuangco and Benedicto were cushioned by their control of trading in their industries and, in addition, had greater access to bank credit-Cojuangco from the UCPB, which received the coconut levy proceeds, and Benedicto from PNB, the traditional source of credit for the sugar industry, and his own Traders Royal Bank, where the casino earnings were deposited. Government financial institutions that extended credit to domestic corporations ended up with ownership positions in a variety of domestic firms. PNB and its subsidiary, the National Investment and Development Corporation, had large stakes in CDCP, Delta Motors, Pilipinas Bank, and full ownership of a passenger bus company. DBP had equity holdings of over one billion pesos in Marinduque Mining, several textile manufacturers, and Philippine Blooming Mills, and had major equity stakes in banks, hotels, pulp and paper mills, and cement plants.24 Total equity holdings of DBP jumped from 11.6 percent of its assets in 1980 to 18.0 percent in 1982, and totaled P. 7.9 billion (about $930 million).25 The social insurance fund for public workers, the GSIS, gained controlling shares in two banks. But it was NDC, headed by the minister of trade and industry, that became the major holdedmanager of distressed firms, with a list of eight-two firms at the end of 1982, spanning basic metals, textiles, pulp and paper, fertilizer, banking, chemicals, and mining (Montelibano 1983). The financial extension of the government to the firms owned by presidential cronies and others exacted a heavy fiscal toll. Exactly how much additional government expenditure took place as a result of the financial crisis and string of corporate failures is difficult to determine. Adding up extensions of funds to the most publicized firms gives a figure of at least P. 3.8 billion, roughly 1.3 percent of GNP and 11 percent of government revenues. Another way to approach this question is to look at the increase in national government equity contributions to the institutions involved in the rescue operations. These are shown in table 4.3. Using the 1980 figure as a base gives an additional equity contribution totaling P. 3.1 billion in 1981 and 1982, which does not count the contributions of other agencies of government, notably GSIS.

479

Table 4.3

PhilippinesiChapter 4 National Government Equity Contributions (in millions of pesos, cash basis)

Central bank Development Bank of the Philippines Philippine National Bank National Development Corporation Total Percent of budget

1980

1981

1982

1983

38 85 72 618 813 2. I

583 300 210 1,813 2.766 5.8

325 607

25 0 150

160

893 1.985 3.8

406

58 I 1.1

Source: Philippines, Government Corporate Monitoring Committee, unpublished data

The rescue operations were strongly criticized within the private sector not closely associated with the Marcos government as rewarding inefficiency, bailing out cronies of Marcos’ family, and unfairly encroaching upon the province of private firms.26 The government defended its actions by saying that the firms rescued represented the largest employers in the country and to have let them fail would have meant tremendous economic cost. Although the losses to the Marcos cronies were not as harsh as could have been applied, they did lose control over the bulk of their companies, were forced to sell others, and in some cases were barred from reentering the industry. The episode does appear to have been a victory for the technocrats over the cronies in the industrial sector, but at considerable economic cost. The most visible cost was the huge increase in government outlays required for the rescue operations. But much of the effect was housed in the deteriorating portfolios of state-owned financial institutions, examined in detail in the next chapter. By the mid- 1980s, the almost complete disintegration of the portfolios of state-owned financial institutions would become the most serious fiscal problem of the Philippine government.

4.4 Cronyism in Philippine Politics and the Economy The impact of government corruption, rent seeking, and favoritism is more easily described than quantified. It is not sufficient nor is it accurate to say that the slide of the Philippines into debt crisis was simply the result of wholesale plunder by Marcos and his associates that drove the economy into the ground. But neither is it sufficient to describe corruption in the Philippines as a marginal effect of economic activity, analogous to a turnover tax on economic activity. The intervention, monopolization, and acquisition of the martial law government and Marcos cronies changed the nature of the economy in the Philippines, weakened the efficiency and profit motivation of its actors, and postponed or made more difficult economic adjustments to an increasing adverse international environment. The strongest and most visible effect was on the two most important export crops, sugar and coconuts. But in a wide variety of industries

480

Robert S. Dohner and Ponciano Intal, Jr.

cronyism discouraged activity by other businessmen and investors, and encouraged capital flight. The operations of the cronies, particularly their growth through acquisition, accentuated the tendency in the Philippines for high financial leveraging and dependence on short-term borrowing, both of which increased the vulnerability of the economy to domestic, as well as international, financial crises. And when the implicit government guarantee that Marcos cronies enjoyed was called in the early 1980s, the excessive risk and bad management that the cronies had undertaken was socialized and became an enormous fiscal burden. While the economic importance of crony capitalism was considerable, it cannot be understood simply as a matter of economics and wealth accumulation. Government intervention, monopolization, and cronyism was also fundamentally a way of establishing and maintaining the political power of the martial law regime. Marcos and martial law represented a direct challenge to the traditional elite that had dominated Philippine politics in this century, and the neutralization of that elite was a crucial challenge for the regime. The seizure of the assets of the Lopez and Jacinto families and the initial activity in land reform were as much measures to threaten potential opponents, as they were economic and distribution policies. The nationalization of sugar trading and its transfer to a trusted crony were important in establishing control over an industry that had traditionally been the source of wealth and influence in the Philippines. In the same vein, interventions in major sectors of the economy, either directly by the state, or through actions which favored associates of the government, created powerful incentives for cooperation with the martial law regime, as well as powerful means to threaten those who did not. Furthermore, wealth and patronage had always been a key to Philippine politics, and the early generation of rents was seen by Marcos in those terms. In later years, perhaps, wealth would become more of an end in itself. In addition, once created, the cronies had their own autonomy and influence and in some ways may have controlled their creator. This political transformation in the Philippines was as much at the center of martial law policy as was the rapid development of the economy. This required a more powerful and centralized government in the Philippines, and this was facilitated by the willingness of foreign donors, multilateral institutions, and commercial lenders to assist a government that had pledged to undertake economic reform and increased developmental investment. It was also facilitated by the preference of all three for publicly guaranteed obligations, which further channeled resources through the state. The debate within development economics of the merits of employing domestic savings or foreign resources for investment has largely ignored the institutional aspects of this choice, but this made an important difference in the Philippines.

481

PhilippineKhapter 5

Financial resources were key to the martial law regime and to Philippine cronyism. The financial vulnerability of the cronies also brought about their downfall, as domestic financial crisis led to corporate failure and then to government rescue at great cost. Financial markets and issues are the next subject to which we turn.

5

The Philippine Financial System and the Debt Crisis

Financial markets played a central role in the events leading up to the Philippine debt crisis of 1983 and the difficulties of the adjustment period that followed. A crisis in the domestic commercial paper market touched off the first round of corporate and financial institution failures, which led to fiscal rescue operations by the Philippine government. By 1984 losses within the government-owned financial institutions became a tremendous drain on fiscal resources, complicating both the achievement of external balance and the fostering of recovery in the country. This chapter examines the financial system in more detail, considering both its contribution to increasing foreign indebtedness in the Philippines and its contributions to Philippine macroeconomic difficulties in the 1980s.

5.1 Financial Institutions and Markets The following provides a brief tour of the financial system in the Philippines. The aim here is not to be exhaustive, but to provide an introduction to the important players in the debt story. 5.1.1

Capital Market

As is the case in other LDCs, the capital or securities market is not well developed in the Philippines and has provided an almost insignificant share of total funds raised for private investment. There were 184 companies listed on the Manila and Makati stock exchanges in 1983, and the total capitalized value of listed shares amounted to $800 million, or roughly 2 percent of Philippine GDP.2 Corporate bond issues, while not unknown, have been insignificant. The size of the primary corporate security market can also be judged from the low number of public offerings, averaging roughly thirty per year (World Bank and IMF 1980, 23). There has been a much larger volume of public securities issued, but there have been only limited private holdings and almost no secondary trading.

482

Robert S. Dohner and Ponciano Intal, Jr.

The Philippine government has instead opted to hold down its financing costs by selling government securities to captive purchasers-commercial banks, who can use government securities for various lending and liquidity requirements, social security institutions, and the central bank-at rates well below market rates. There are a number of explanations for the limited development of the Philippine capital market. Perhaps most important is the fact that interest rate controls resulted in the subsidization of bank credit to prime commercial borrowers, the firms most likely to issue primary securities. Lack of government support for secondary trading in public securities also hindered the growth of the capital market. Potential investors have been discouraged by the speculative nature of most stocks in the Philippine exchanges and by the tendency of the market toward manipulation. The supply of primary securities has been limited by the tax advantages of loan financing, the reluctance of many family-owned firms to relinquish any control, and by the disclosure requirements (of interest both to investors and tax authorities) for listing on the exchanges. The results of the stunted development of securities markets in the Philippines has been a predominance of loan financing of business activities and, as a consequence, very high debvequity ratios. This, and the fact (taken up below) that most firms are dependent on the continued rollover of short-term loans, made the corporate sector particularly vulnerable to the unusual occurrence of recession with high interest rates in the early 1980s.

5.1.2 Financial Institutions in the Loans Market Commercial banks are the predominant financial institutions in the Philippines, holding roughly three-fifths of total financial system assets, as is shown in table 5.1. In 1985 there were thirty commercial banks operating in the country. The government-owned Philippine National Bank (PNB) was by far the largest, with approximately 30 percent of commercial bank asset^.^ Four of the remaining banks were foreign owned, accounting for about 15 percent of bank assets. The importance of commercial banks, while high, is not out of line with other countries in the region nor with other countries at a similar level of development. What is unusual about the Philippines is the large number of commercial banks and the relatively small size of many of them. Most of the commercial banks in operation today were established between 1950 and 1965. During this period the central bank encouraged the entry of new banking firms, and capital requirements for forming a bank were minimal. Many of the newly formed industrial groups found it in their interest to add a bank to their holdings, and a total of twenty-seven banks were incorporated during this period. The result has been a number of small and, in many cases, family-managed banks.

Philippines/Chapter 5

483

’hble 5.1

Philippine Financial System, 1983 Total Assets (billions of pesos)

Central bank Financial system Banking institutions Commercial banks PNB Thrift banks Rural banks Other government banks DBP Land Bank Philippine Amanah Bank Nonbank financial institutions Investment houses Finance companies Investment companies Securities dealers Pawnshops Fund managers Nonstock SLAs Other financial institutions Private insurance companies Special nonbanks GSlS

sss Offshorebanking units Total’

130.4 354.6 326.0 235.0 70.5 16.1 9.5 65.3 56.5 8.5 0.3 28.6 7.2 11.8 6.2 0.7 0.5 1.5 0.7 63.5 13.7 49.7 14.7 16.3 4.4 418.0

Shares (percent) -

84.8 78.0 56.2 16.9 3.9 2.3 15.6 13.5 2.0 0.1 6.8 1.7 2.8 1.5 0.2 0. I

0.4 0.2 15.2 3.3 11.9 3.5 3.9

-

Number of Institutions 1 1,122 34 136 949 3

1,474 14 336 65 124 701 12 74 179 136 I

21

100.0

Source; Nomura Research Institute, “A Capital Market Study of the Philippines” (Manila: Asian Development Bank, 1984). cited in Lamberte (1985, 4). ‘Excluding the central bank and offshore banking units.

After 1965 central bank policy changed, raising minimum capital requirements and effectively denying new bank applications. Financial reforms in the early 1970s raised capital requirements again and encouraged banks to obtain foreign equity partners to both raise capital and strengthen bank management. Several Philippine banks entered joint venture arrangements, although many of the foreign partners sold their stakes by the end of the decade. The most important characteristic of commercial bank portfolios is the high proportion of short-term lending. Although medium- and long-term lending increased significantly during the 1970s, loans of one year or less maturity still accounted for almost 80 percent of commercial bank loans in 1980.4 In addition to commercial banks, there are several categories of banks with more restrictive deposit or loan portfolios. Within thrift banks, savings and loan associations and mortgage banks mobilize smaller deposits for

484

Robert S. Dohner and Ponciano Intal, Jr.

mortgage lending and consumer finance. Together they make up about 3 percent of financial system assets. Rural banks were created in 1952 to channel credit to the agricultural sector. Although there are almost one thousand rural banks, they hold only 2 percent of the assets of the financial system. They have been dependent on central bank support for funding and have recently had severe problems in arrearages. Development banks provide longer term credit for industry. The state-owned Development Bank of the Philippines (DBP) was by far the largest of these banks, holding 14 percent of financial system assets and extending almost half of long-term red it.^ Funding for DBP came primarily from borrowings, either from multilateral organizations (the Asian Development Bank and the World Bank), government entities (the central bank, national government, and the social security institutions), or from foreign commercial bank loans. Deposits have represented only about one-fifth of DBP’s liabilities, and roughly half of these have come from deposits of the national government. In addition to extending industrial loans, DBP has also guaranteed foreign loans to the private sector. At the end of 1983, DBP guarantees equalled P. 15 billion, or over 25 percent of total assets (Lamberte 1984, 20).6 DBP also acquired equity interests in domestic firms; these investments rose rapidly during the early 1980s as the institution was used as a rescue agent for distressed industrial firms and banks. The financial crisis and the resulting slowdown in the Philippine economy have forced DBP to honor many of its guarantees, in exchange for which it took over equity interest in the distressed firms. By 1985 as much as 70 percent of its portfolio was nonperforming, presenting a major drain on fiscal resources. Private development banks serve the same role of extending long-term credit to industry, but are much smaller than DBP, holding about 1 percent of the assets of the financial system. They have been dependent on funding from the DBP and borrowings from the World Bank and other multilateral development banks. Two social security institutions, the Government Service Insurance System (GSIS) and the Social Security System (SSS) covering private workers, play an important role in funds mobilization and potentially in the provision of long-term credit. The proceeds of these institutions have largely been invested in public securities and in DBP and PNB notes. GSIS, however, has made substantial equity investments, generally in industries facing financial difficulties. Thus, GSIS owns several of the hotels that were constructed during the 1970s, as well as Philippine Airlines. Two additional financial institutions are important for our story, largely because of their operations in the money market. Investment houses were originally established to provide long-term funds and underwriting services to industry. Instead their operations have centered in the money market, raising funds by borrowing in that market and extending short-term loans to

485

Philippines/Chapter 5

domestic firms. Finance companies extend credit to consumers and to businesses through discounting commercial paper, factoring, and leasing. The larger finance companies have acquired quasi-bank licenses and have been active participants in the money market.

5.1.3 The Money Market In contrast to the relatively inactive capital market, the Philippines has developed a sophisticated and extensive market in short-term instruments. The money market provided investors with profitable opportunities for short-term funds, and the market became a major source of finance for Philippine corporations outside the first tier of borrowers. The money market developed in the mid-1960s when a few investment houses began buying and selling short-dated obligations of banks and prime corporations. The investment houses were later joined by banks and finance companies, and this market grew rapidly in the late 1960s and early 1970s. Interbank call loans-loans made between banks to adjust reserve positions-account for the largest activity in the money market. However, markets also developed to supply investment opportunities for firms and individuals with surplus short-term funds. Termed “deposit substitutes,” these consisted of promissory notes of commercial banks and other financial institutions, repurchase agreements, and certificates of assignment and participation involving other assets. Interest rates in this market were substantially above bank deposit rates; effective rates ran as high as 30 percent per year, compared to 8- 11 percent for time deposits (World Bank and IMF 1980, 26). Deposit substitutes grew rapidly in volume, and by 1975 they had exceeded the total value of time and savings deposits and were equivalent to 50 percent of M2 (currency plus bank deposits). In addition, commercial paper was traded in the money market, either directly or resold by investment houses as deposit substitutes (repurchase agreements or certificates of assignment and participation). Directly marketed commercial paper made up about 5 percent of total money market instruments, but by 1980 remarketed commercial paper formed the basis for an additional 15 percent of all deposit substitutes (Licuanan 1986, 102, 123). While the money market provided attractive investment opportunities for individuals and firms with surplus funds, the commercial paper segment of the market and the lending activities of investment houses and finance companies offered an opportunity for smaller and less well known firms or for rapidly growing firms to obtain access to credit. The credit obtained was more expensive than bank credit, but it was available to many firms which had been closed out of bank lending or could not obtain sufficient bank funds for their activities. But, like bank loans, funds sourced from the commercial paper market and its quasi-banking institutions, were short-term funds. Only about 5 percent of money-market transactions had maturities over forty-five

486

Robert S. Dohner and Ponciano Intal, Jr.

days (Licuanan 1986, 9- 10). Firms tapping this market did so regularly and depended on rollovers to fund continuing operations. Commercial banks, investment houses, and finance companies with quasi-banking licenses have used the money market as a source of funds for their operations. Although commercial banks are the largest participants, money-market borrowings make up a relatively small share of their total sources of funds. In contrast, investment houses and finance companies have been heavily dependent on the market as a source of funds. Money-market borrowings accounted for about 80 percent of the funding of investment houses and from 65 to 80 percent of that of finance companies with quasi-bank licenses (Licuanan 1986, 43, 66). The initial growth of the money market took place when the market was essentially unregulated. During the 1970s the central bank, through various measures, extended its regulation to transactions in the money market and sought to curb the growth of the market. Amendments to the Central Bank Act placed nonbank financial intermediaries under central bank regulation. Central bank circulars defined a new class of activity called quasi banking, which covered borrowing from twenty or more lenders, by issuing deposit substitutes. An interest rate ceiling of 17 percent was established on deposit substitutes in 1976, reserve requirements were established, and a transactions tax of 35 percent on deposit substitute interest payments was imposed. The effect of these regulations was to slow, but not halt, the growth of deposit instruments in the money market, and the ratio of deposit substitutes to GNP fell from a peak of 8.5 percent of GNP in 1975 to about 4.5 percent by 1980.’ Actions were taken by some intermediaries to sidestep and, in some cases, evade, the regulations that were in place. Institutions accepting deposits from fewer than twenty investors were not covered by quasibanking regulation. The central bank definition of deposit substitutes covered certificates of assignment and participation issued with recourse to the intermediary. After the mid-l970s, the securities covered by these certificates were issued directly to the public without recourse, although they were accompanied by the postdated checks of the financial institutions, ostensibly in their capacity as paying agent for the issuer of the security. The volume of outstanding commercial paper also increased after regulation of deposit substitutes. Commercial paper issuances were regulated by the Securities and Exchange Commission, which proved to be lax in its oversight. Thus, the effect of the regulation was not only to slow the growth of the money market, but also to push the money market toward riskier transactions. 5.1.4

Foreign Exchange Markets

The central bank of the Philippines has exercised strict, if not always successful, control over resident foreign exchange transactions. Exporters and other recipients of current foreign exchange income are required to surrender their proceeds in exchange for pesos. Despite these restrictions, a

487

PhilippineKhapter 5

black market in foreign exchange has coexisted with the official market, fed by tourists, overseas workers, and exporters, and has served as a source of foreign exchange for restricted imports and for capital flight. In 1976 the central bank moved to encourage offshore banking in foreign currencies within the Philippines. The motives were several. The first was the apparent success that Singapore had in increasing domestic financial activity with its offshore banking sector. The second was the desire to send more foreign exchange through official channels within the Philippines. A final aim was to mobilize foreign exchange resources for the Philippine economy. To develop the market, the central bank exempted offshore banking units (OBUs) from reserve requirements, local taxes, and fees, and permitted them to extend foreign currency loans to any enterprise from deposits raised outside the country. Low rates of taxation were applied to the income of OBUs. In addition, domestic commercial banks were allowed to establish foreign currency deposit units (FCDUs) with similar privileges and the ability to accept foreign currency deposits from domestic residents. The assets of OBUs and FCDUs grew rapidly in the last years of the 1970s, and by 1980 their total nonbank placements amounted to over $4 billion (P. 32 billion), or roughly one-third of gross domestic credit of the deposit banking system. Onshore lending requires the approval of the central bank, but almost all the placements of these offshore units were made to borrowers within the Philippines.'

5.2 Policy and Financial Markets 5.2.1 Interest Rate Controls Interest rate controls have been a persistent feature of Philippine monetary policy up to at least 1983. Controls have been in place for loan, deposit, and rediscount rates almost from the beginning of the postwar period. The motivation in setting interest rates has been the encouragement of investment and the channeling of funds to priority sectors. Little attention was paid to mobilization of domestic savings in financial form and, as a result, substantial excess demands for credit have been a recurring feature of the Philippines. Ceiling rates on loans were established by the Usury Law of 1916, which set a maximum rate for secured loans of 12 percent and 14 percent for unsecured loans. No adjustments were made for the term of the loan, and these ceilings held until the middle 1970s. Deposit rates were initially set in 1956 at 2 percent, with 2.5 percent offered for one-year deposits. These rates were adjusted upward slowly; by 1970, one-year deposits yielded 7 percent. Monetary policy during the 1950s was largely directed toward defending the exchange rate. The inflation rate was negligible during the first half of

488

Robert S. Dohner and Ponciano Intal, Jr.

the decade and averaged only 2.4 percent per year from 1956 to 1962, the year of decontrol. As a result, despite the low rates on deposits, a substantial degree of financial deepening took place during this period. The ratio of M2 to GNP rose from 19 to 26 percent from 1956 to 1963. The presence of excess reserves and the low level of borrowing from the central bank suggests that interest rate ceilings on loans were not binding, at least until the mid- 1960s. The loan rate ceilings became limiting after 1965, as the economy picked up during the first Marcos administration, the manufacturing sector recovered from the decontrol episode, and the government borrowed more heavily. By 1970 the inflation rate had also increased substantially to over 10 percent per year. After 1974 the central bank adjusted interest rate ceilings more rapidly and began to provide higher ceiling rates to encourage long-term loans. Real loan rates increased, as table 5.2 indicates, but real deposit rates were negative throughout the period. The persistence of disequilibrium in the credit market may be judged from the sketchy evidence

Table 5.2

Low interest rate period 1956- 69 1970 1971 1972 1973 Transition period 1974 1975 1976 1977 1978 1979 1980 Average Floating rate period 1981 1982 1983 1984 1985 1986 1987 Average

Philippine Real Interest Rates Average Money Market

Savings Deposits

Time Deposits

0.31 -8.85 - 15.90 -2.22 - 10.50

-0.61 -7.85 - 14.90 -1.22 -9.58

-28.16 -0.78 -2.23 -2.93 -0.20 -7.51 -8.60

-24.66 2.72 0.77 0.07 2.72 -4.51 -3.60

- 16.59

-7.20

-3.78

-0.93

-0.79 1.28 - 16.41 -38.79 5.14 8.32 -2.92

5.02 5.72 -11.76 - 17.87 16.28 12.00 2.38 1.68

-6.32

Secured Short- term Loans

CPI Inflation'

8.28 -2.85 -9.90 3.78 -4.50

3.72 14.85 21.90 8.22 16.50

-22.16 5.22 2.77 2.07 4.72 -2.51 -3.60

34.16 6.78 9.23 9.93 7.20 16.51 17.60

- 1.93

14.49

5.02 5.72 -9.50 -23.19 15.11 12.70 3.51

5.42 8.64 -4.82 -11.25 22.56 17.23 5.67

10.58 8.49 26.07 50.83 5.66 -0.30 7.45

1.34

6.21

15.54

8.23 3.71 2.66 3.44 -3.62 -4.33

Source: Lamberte (1985). table 111.4. Data for 1985-86 from Central bank, Philippine Financial Sfatisrics. floating rate period, CPI inflation reported for December to December of each year.

489

Philippines/Chapter 5

on effective interest rates, which were well above legal ceilings, and by the rapid growth of the money market, where interest rates were also high.' Interest rate regulation had several effects in the Philippines. The ceilings on loan rates and the excess demand for credit that developed effectively limited commercial bank credit to prime corporate borrowers or to firms affiliated with a bank. Other firms, or firms that could not obtain sufficient bank credit, were forced to go to alternative markets where funds were available at much higher interest rates. The structure of rate ceilings also discouraged long-term lending. Short-term lending was safer than lending long, and the addition of booking fees raised the effective return on short-term loans over that of long-term loans. In addition, only short-term loans could be rediscounted with the central bank. The relative incentives for short- and long-term lending were clearly reflected in bank portfolios. In the early 1970s almost 96 percent of commercial bank loans had maturities of one year or less (Lamberte 1985, 30)." Higher ceilings for long-term loans encouraged the growth of term lending during the remainder of the decade, but the availability of long-term finance remained a problem for industry. Interest rate restrictions kept deposit rates well below loan rates, which provided banks with substantial margins on loan operations. The available spreads encouraged the entry of new banking operations during the two decades before 1965. After 1965, when the central bank limited entry, the spreads inherent in the interest rate restrictions generated substantial rents to the owners of banks in the Philippines. The favorable spreads were reinforced by central bank discounting policy that kept rediscount rates quite low. This encouraged banks to rely on the central bank as a source of funding. For the central bank this meant that discount policy was less a matter of monetary control than a means of allocating credit to priority sectors.

5.2.2

Credit Allocation

Credit allocation and financial specialization is the second characteristic of Philippine financial policy. In 1959 the central bank began more active intervention to channel credit to priority sectors. This credit allocation policy was approached in three ways: (1) through priority rediscounting windows with the central bank; (2) through explicit requirements for the allocation of bank funds; and (3) through the establishment of specialized financial institutions with narrowly defined lending missions. Agricultural loans were the first beneficiaries of preferential rediscounting, followed in the 1960s by export loans and selected industrial loans. Agrarian reform loans and loans to small-scale industries were added in the 1970s, and a host of preferential rediscount windows were opened in the early 1980s. Rediscounts were limited to short-term securities until the 1980s and generally carried restrictions on the maximum rate that could be charged to the borrower.

490

Robert S. Dohner and Ponciano Intal, Jr.

A presidential decree in 1975 directed banks to allocate 25 percent of their net loanable funds to agricultural lending, divided between agricultural production loans and loans financing the transfer of land under the agrarian reform program. As an alternative to agrarian reform lending, banks could instead purchase government securities carrying a coupon rate of 9 percent, and most banks chose securities purchase in order to meet this requirement. Perceived credit needs of priority sectors were also addressed by creating specialized financial institutions. Rural banks, created in the early 1950s to support the agricultural sector, were followed by development banks for industry, the National Cottage Industry Bank for small-scale industry, the Philippine Amanah Bank for Moslem areas, and the Land Bank for agrarian reform. In nearly all cases these institutions were heavily dependent upon central bank deposits and credit for their funding, and in many cases had very precisely defined restrictions on whom they could lend to and for what purpose. To support these allocative mechanisms the central bank raised reserve requirements on deposits, particularly time and savings deposits, so that by the 1970s the Philippines had the highest reserve requirement in the region. This, and the fact that rediscount rates were set well below prevailing loan rates, encouraged the dependence of Philippine financial institutions on central bank credit and the channeling of an increasing portion of total finance through the central bank or other government agencies. Patterns of ownership and control also influence the allocation of credit. Almost all banks are owned by, or closely associated with, a family-owned industrial group in the Philippines. This reflects the historical concentration of wealth in the Philippines and the fact that most banks now in operation were formed during the period from 1946 to 1965, when the central bank actively encouraged the entry of new banking firms. The result was a financial structure in many ways similar to that of the interwar Japanese zaibatsu, where each of the major industrial groups owned or controlled its own bank and most banks were connected with an industrial group.” Ownership of a bank provided a way of mobilizing deposits and, therefore, credit in support of other group operations. As a former central bank governor put it, “the average Filipino banker is in the business not for banking profits; he uses his bank for allied businesses.”’2 In a situation in which deposit and loan rates were below market clearing levels and well below rates in parallel markets, this resulted in the capture of substantial rents, and supported and perpetuated the concentration of wealth and economic power in the Philippines. During the martial law regime government control and government influence increased substantially in the banking sector. PNB, the country’s largest commercial bank, is a government-owned bank. In addition, as a result of the financial crises of the 198Os, the Marcos government acquired four commercial banks from the private sector, accounting for about 4 percent of total bank a s s e t ~ . ’ ~

491

PhilippineKhapter 5

Two banks, the Republican Planters Bank and the United Coconut Planters Bank (UCPB), are best described as quasi-official banks because of their involvement in the sugar and coconut industries, both heavily regulated by the government. Each was controlled by a close associate of Marcos, Roberto Benedicto in the case of Republic Planters and Eduardo Cojuangco for UCPB. Funding for both of these banks has come from export taxes on sugar and coconuts and, for coconuts, a levy on domestic production and milling. At least five other commercial banks, including Traders Royal Bank and Allied Banking Corporation, have been termed “political banks” with close ties to the Marcos government. Although the precise extent to which they were favored by the Marcos regime is difficult to determine, several are highly dependent on central bank funds, and there is the widespread impression that their rapid growth was made possible by their government connections. l 4 The combination of the assets of PNB, DBP, and the assets of commercial banks closely tied to the Marcos government, gives a total of over half the assets of the commercial banking system (Canlas et al. 1984, 68). The sizable direct and indirect participation of the government in the Philippine financial system gave the regime a large voice in the allocation of credit within the economy, for public and, in some cases, for private purposes.” The final characteristic of Philippine financial markets is their turbulent postwar history.16 Various financial institutions have been in trouble at one time or another, including commercial banks, which have experienced several runs and occasional failures. The failure of Continental Bank in 1974 was particularly serious because it led to a run on the entire banking system, which was finally stopped when the central bank stepped in with emergency loans and assurances that it would cover liquidity problems. A more serious crisis came with the departure of Dewey Dee in 1981, since it led to the virtual collapse of one segment of the money market and runs on investment houses, finance companies, and some banks. The crisis was halted with central bank intervention, but the effects of the crisis persisted for several years. Crises have arisen not so much from lack of capital as from financing of long-term investments from short-term funds and from loans to directors and related individuals of banks (Patrick and Moreno 1985, 326-28). 5.2.3 Financial Performance Of particular interest here is the ability of the financial system to mobilize domestic savings and make them available to investors, both over time and in comparison to other countries of similar income levels. A standard measure for LDCs, where the banking system dominates the organized financial market, is the ratio of M2-currency plus bank deposits (demand, time, and savings)-to gross domestic product (GDP). This ratio provides a rough measure of financial “depth,” the extent to which financial resources are held relative to the value of domestic production.

492

Robert S. Dohner and Ponciano Intal, Jr.

While this is the standard measure for cross-country comparisons, its use for the Philippines is problematic. The reason is the growth of the market for deposit substitutes, which serve a function similar to time deposits, but are not counted in M2. In what follows we also use M3, defined in the Philippines as M2 plus deposit substitutes, and examine both ratios with respect to GDP. Figure 5.1 shows the ratio of M2 to GDP for several East Asian developing countries plotted against the level of per capita GNP for 1960, 1970, and 1980. As is clear from this figure, the extent of money and bank deposit holdings is related to the level of per capita income and ordinarily increases along with per capita GNP. The Philippines is something of an anomaly in the comparison, although financial deepening is more apparent when M3 is used instead of M2. What is most striking about the Philippines is the stagnation of these ratios during the 1970s, a period in which per capita incomes were rising much faster than in the previous decade and when the savings ratio increased. In addition, Philippine financial mobilization is somewhat low for countries with similar per capita income levels.

200

600

1000

1 400

Per Capita GNP (US

1800

$)

Fig. 5.1 Financial deepening in East Asian countries (1960, 1970, 1980) Note: Ind = Indonesia (1970, 1980); Kor = Korea; Ma1 = Philippines (M3); and Th = Thailand. Source: IMF, International Financial Statistics.

=

Malaysia; Ph = Philippines; Ph3

493

PhilippineKhapter 5

A more detailed look at financial ratios for the Philippines in the 1970s is provided in table 5.3. Ratios of broad money to GDP increased steadily during the early 1970s, peaking in 1978. All indicators of financial asset mobilization fall in the succeeding two years, a period that coincided with the acceleration of inflation and the drop in the share of savings in GNP. What is also apparent from the table is the steady decline in narrow money to GDP, a factor that limited monetary finance of government budget deficits, During the 1970s there was a huge increase in both private and public investment expenditures at a time when the loanable resources of the financial system were increasing only moderately. The result was a very large increase in foreign borrowing during this period, a time when the ratio of foreign debt to GNP increased most rapidly. For private firms there were two advantages of foreign currency financing. First, borrowing in foreign currency was in many instances cheaper than raising funds domestically. Assuming an external interest rate of approximately 10 percent and adding the DBP annual guarantee fee of 3 percent, foreign loans were left considerably cheaper than domestic loans whose effective rates were about 19 percent (World Bank and IMF 1980, 34).17In cases in which the foreign borrowing was intermediated, this lower rate was usually passed on to the ultimate borrower (Patrick and Moreno 1985, 357). The second advantage of foreign borrowing was that funds were available for longer terms than were domestic currency loans. Philippine firms required central bank approval for foreign borrowings, but export-oriented firms and firms registered with the Bureau of Industry found approval relatively easy to obtain. Foreign currency borrowing did carry the additional risk of exchange rate changes. However, the peso had depreciated against the dollar by only 3 percent per year between 1973 and 1976 and was almost constant between 1976 and 1980. Furthermore, many firms were able to arrange swap agreements, through banks, with the central bank, that provided forward exchange at favorable rates. Complete information on the level of outstanding swap and forward exchange cover is difficult to obtain. Figures for swap arrangements

Money Holdings in the Philippines as a Percentage of GNP, 1970-80

Table 5.3

1970-72

1973-75

1976

1977

1978

1979

1980

M2

11.1 21.7

M3

N .A .

9.4 17.1 24.8

9.0 18.6 26.7

9.7 21.2 28.7

9.6 22.8 29.3

8.6 20.8 26.3

8.5 21.0 25.6

MI

Source: N.A. MI M2 M3

Central bank, Annual Report and Philippine Financial Statistics. various issues. = not available. = currency plus demand deposits. = MI plus time and savings deposits. = M2 plus deposit substitutes.

494

Robert S. Dohner and Ponciano Intal, Jr.

conducted with commercial banks for 1978 to 1980 are shown in table 5.4. Swap arrangements grew rapidly during this period, and by 1980 they rival direct borrowings from the central bank as a source of funding for Philippine commercial banks. The granting of swap arrangements was a controversial issue in the Philippines, and charges of favoritism surrounded their allocation. The central bank governor at this time, Gregorio Licaros, personally approved all swap transactions in excess of $1 million, and banks differed greatly in their use of swap transactions. The increasing amount of medium- and long-term lending that took place during the 1970s and the access to longer term funds from foreign sources lowered the risks of debt financing, at least for those firms that had access to commercial bank credit and had hedged their foreign exchange risk. However, many firms greatly increased their reliance on debt during the decade, making them more vulnerable to economic shocks. Over the ten years from 1970 to 1980, the debuequity ratio of the top one thousand corporations in the Philippines increased from 1.6 to 4.0.'' In addition, the reliance of domestic firms on rollovers of short-term credit from the money market also increased, making them more vulnerable to liquidity problems. This was particularly true for the rapidly expanding firms of Marcos cronies, who borrowed heavily in this market.

5.3 Financial Markets in the 1980s Financial markets in the Philippines were subject to a variety of strains, including market crises and substantial reversals of policy. By the midpoint of the decade the financial system was thoroughly intertwined in the country's fiscal problems, and the reluctance of the banking sector to perform its lending function had become a substantial impediment to the recovery of economic growth in the Philippines.

5.3.1 Financial Market Liberalization During the 1980s the Philippines undertook a major financial reform that freed interest rates from administrative control and drastically reduced the

Table 5.4

Bank Availments of Central Bank Swaps'

Foreign exchange futures bought (millions of pesos) (In millions of U.S. dollars) Share of total external liabilities (%) Memo item: Central bank credit to commercial banks (millions of pesos)

1978

1979

1980

7,375 I ,Ooo 5.8

13,477 1.818 8.7

20,902 2.750 11.1

18.348

25.660

13,476

Source: Patrick and Moreno (1985). table 15.7, and Philippine central bank, Management of External Debt

and Investment Accounts Department (MEDIAD). Original swap data from the central bank. "Outstanding balances, December 3 1

495

Philippines/Chapter 5

functional specialization among financial institutions. The reforms were the outgrowth of two studies of the Philippine financial system commissioned by the central bank, one by a joint World Bank-IMF mission (World Bank and IMF 1980), and the second by a former official of the Mexican central bank. Both reports focussed on the lack of long-term funds within the Philippine financial system and the specialization and fragmentation of financial institutions that had developed. The World Bank-IMF report argued for a shift in central bank operations away from credit allocation and toward a lender of last resort facility. Both reports argued for an increase in capitalization of financial institutions and for the ability of large commercial banks to engage in underwriting and equity investments. Almost all of these proposals were implemented over the next three years. The interest rate reform was achieved in stages. Ceilings on long-term loans and all types of deposits were lifted in July 1981, and ceilings on short-term loans were finally removed in January 1983. Specialization among types of banks was greatly reduced. With the exception of underwriting and investment in equities, each type of bank was eligible to perform all types of bank activities. Minimum capital requirements were set or raised for thrift banks, rural banks, and quasi banks. In addition, a new category of commercial bank was created, an expanded commercial bank or “universal bank” (popularly known as a unibank) with a minimum capital requirement of P. 500 million, five times that of an ordinary commercial bank. Unibanks were permitted to engage in underwriting and to purchase up to 35 percent of the equity of nonallied businesses. PNB became the first unibank, followed by seven other commercial banks. Other elements of the reform package were not as successful. Reserve requirements on deposits were reduced from 20 to 19 percent in July 1981 and were scheduled to be reduced in one-percentage-point steps to 16 percent. However, financial and balance of payments crises intervened, leading to subsequent rises in reserve requirements to 24 percent by early 1984. Under the reform program, the central bank was to assume more of a stabilization and lender of last resort role, reducing its provision of subsidized credit through priority rediscount windows. This was finally achieved, but only in 1985. In the interim, the number of rediscount windows multiplied to the point at which virtually any activity was eligible for rediscounting. The selective rediscounting policy had by this time lost its selectivity, requiring more rationing of credit through these windows. The effects of the financial liberalization program are obscured by the financial and balance of payments crises that characterized the early 1980s. The lifting of interest rate ceilings led to moderate rises in deposit and loan rates in 1981 and 1982, apparently the result of oligopolistic coordination among the country’s banks. The limited movement of interest rates continued until the central bank and Treasury began actively competing for savings in 1984, forcing banks to raise their deposit rates. The Philippines

496

Robert S. Dohner and Ponciano Intal, Jr.

did not liberalize the capital account at this point, so that the country did not face the inflow of speculative funds that appreciated the real exchange rate in Argentina when a similar liberalization took place. The fall in the domestic inflation rate in 1981 and 1982 did mean higher real interest rates, and a rapid growth of bank deposits followed. Although the ratio of broad money to GNP increased during this period, much of the growth of time and savings deposits was at the expense of narrow money as indicated in table 5.5. The share of long-term loans (one-year or more maturity) in banks' portfolios also increased during this period, from 22 percent in 1980 to 31 percent by 1983. 5.3.2

The Dewey Dee Crisis and Aftermath

The first economic shock of the 1980s for the Philippines was an external one-the rise in oil prices and real interest rates-that sharply reduced the real income of the country. The second shock took place in early 1981 in Philippine financial markets. Failures of financial institutions and liquidity crises were by no means unknown; furthermore, the Philippines weathered a run on Consolidated Bank just a few days before the 1981 financial crisis. But the crisis that followed was more serious and set in train a series of events that would unhinge the budget and seriously weaken the financial system. In January 1981 Dewey Dee, a respected Philippine-Chinese entrepreneur, disappeared from the Philippines leaving approximately P. 635 million ($80 million) in unpaid debts incurred by his textile company, Consolidated Manufacturing. l 9 Dee and his companies had borrowed from several Philippine banks, including DBP, but had also run up substantial debt in the money market. The initial confusion surrounding the full extent of his obligations sparked a run on several investment houses and a general flight of capital to larger, non-Chinese banks. Investors refused to roll over existing commercial paper holdings and in many cases preterminated existing money market investments, which required the central bank to announce that it would stand behind the obligations of the affected financial

Table 5.5

Liquidity Ratios (to GNP), 1980-87

MI Time deposits M2 Deposit substitutes M3 Memo item: M3 GS"

+

1980

1981

1982

1983

1984

1985

1986

1987

8.5

7.7 13.9 21.6 5.4 27.0 27.0

7.0 16.5 23.5 4.9 28.4 28.4

8.6 16.7 25.3 4.5 29.8 34.8

6.4 14.5 20.9 2.1 23.0 29.1

6.0 14.8 20.8

6.9 15.1 22.0 0.8 22.8 37.8

7.4 14.7 22.0 0.5 22.5 41.0

12.4 21.0 4.7 25.6 25.6

Source: Central bank, Philippine Financial Statistics.

"Government securities held by the nonbank public.

1.4

22.3 31.6

497

PhilippineKhapter 5

institutions, including investment houses and finance companies. More than forty financial institutions turned out to be exposed, and a larger number of these required central bank support.20 The central bank advanced a little over P. 1.6 billion (about 8 percent of total loans and advances) to investment houses and finance companies during the crisis.21 The cental bank in turn sold Certificates of Indebtedness to recapture the liquidity that had been created by the advances. In the weeks that followed a variety of abuses came to light. The regulation of the commercial paper market by the Securities and Exchange Commission had been quite lax, and paper was issued in many instances without the approval of the SEC, sometimes by firms not licensed to deal in commercial paper. In some cases, commercial paper was printed on ordinary typing paper. Public confidence did not return as the crisis passed and, by the end of the year, deposit substitute holdings of nonbank financial institutions had dropped by 24 percent to P. 2.7 billion. Suspicion extended beyond the financial institutions to the firms that had been active borrowers in the money market, and the commercial paper market virtually dried up. Many of the corporate borrowers turned out to be in extremely weak financial condition, and the Dewey Dee crisis brought many of them down. The period was one well suited to financial difficulties. The Philippine economy had slowed considerably after the second oil shock, and real interest rates rose sharply in 1980 and 1981, as interest rate ceilings were raised and as inflation rates receded after the external shock. Without access to long-term funds, many corporations had borrowed short in order to fund long-term obligations, and in some cases firms had been borrowing to cover current losses. Particularly affected were the conglomerates owned by several of the Marcos cronies. As described in the previous chapter, these firms were highly leveraged, and several were heavily involved in money-market borrowings. Herminio Disini's investment company, Atrium Capital Corporation, was in the center of the liquidity crisis. During the crisis, the central bank advanced P. 1.2 billion to Atrium. The firm was later merged with Asia Pacific Finance into International Corporate Bank (both owned by Disini) and taken over by DBP. Ricardo Silverio's Philippine Underwriters Finance Corporation (Philfinance) was also unable to meet its obligations and was later accused of wholesale violations of the securities laws. During the crisis, the central bank had supplied funds to financial institutions at penalty rates, 24 percent for six-month money, rising by two percentage points with each rollover. But the firms that the financial institutions had lent money to were illiquid or insolvent, and collateral covering the loans were missing or vastly overstated so that the institutions, and in some cases the banks they were tied to, were unable to make repayment to the central bank.22

498

Robert S. Dohner and Ponciano Intal, Jr.

The response of the government was to undertake reorganization plans for the largest financial and nonfinancial corporations affected. For nonfinancial corporations these involved the conversion of existing government loans into equity, forcing the sale of some corporations or assets, and providing additional government funds. An industrial rescue fund was established, initially of P. 1.5 billion, but the limit was later raised to P. 5 billion. The first firms to draw on the fund were Consolidated Manufacturing (Dewey Dee’s firm) and Alfa Textiles. They were followed by a host of firms owned by Cuenca, Disini, and Silverio, and also by Marinduque Mining, a non-crony firm. The two large investment firms that had closed in July 1981 were merged with related commercial banks-Atrium, as described above, with Interbank, and Bancom with Union Bank. Both banks were taken over by the government. Two additional banks were taken over in early 1982, bringing to six the number of private banks the government had acquired.23 These acquired banks held just 8 percent of bank assets, but their share of central bank credit rose from 11.5 percent in 1980 to 24.3 percent in 1983. Even this is an underestimate, since the central bank replenished the funds that had been supplied by DBP. By the end of 1984, the government’s exposure to these banks was P. 6.7 billion ($330 million), as detailed in table 5.6. This episode in the early 1980s was not the first time that the government had rescued ailing corporations. After the boom of hotel construction for the 1975 IMF-World Bank meetings, DBP and GSIS had taken over ownership of a number of hotels that were losing money. In 1978 GSIS took over majority ownership of Philippine Airlines. But the rescue operations mounted in 1981 and 1982 were far more extensive. By 1983, PNB had P. 2 billion in equity holdings and DBP had P. 7 billion in equity in 122 corporations (Montelibano 1983).

5.4 Portfolio Deterioration of Government Financial Institutions As explained above, the Philippine government has long played a prominent role in the financial sector, in contrast to its limited participation in other sectors of the economy. The financial sector has also been an area of

Table 5.6

Government Support to Acquired Commercial Banks (end of 1984, in millions of pesos) Government equity Government deposits Parent advances Central bank advances

Total

1,808 1,221 1,199 2,429 6,651

Source: Philippine government, “Report A: Role of Government Financial Institutions,” 4

July 1985, p. 50.

499

PhilippineKhapter 5

political patronage-access to cheap credit was seen as one of the spoils of winning elections in the Philippines. Thus the element of political influence in the allocation of credit by government financial institutions during the late 1970s and early 1980s was not a new phenomenon in the Philippines, nor in fact is it unusual among state-owned financial institutions in other countries. What was striking about the Philippines in the 1980s was the massive deterioration of portfolios of government financial institutions and the tremendous drain they placed both on the budget and on economic recovery. The two major public financial institutions with greatly deteriorated portfolios were PNB, including its subsidiary, the National Investment and Development Corporation (NIDC), and DBP. In addition, weak or insolvent assets characterized the investment portfolios of the two social insurance institutions, SSS and GSIS. The remaining institution was the Philippine Export and Foreign Loan Guarantee Corporation (PhilGuarantee), which primarily guaranteed loans made by foreign banks, but also extended guarantees to loans made by Philippine institutions. Neither PNB nor DBP had been very successful in deposit mobilization. PNB did better, with its extensive network of branches that covered most of the country, but private deposits made up only about 25 percent of PNB’s liabilities (government deposits have accounted for an additional 18 percent) compared to about 54 percent for private commercial banks (Tan 1984, 66). DBP generated less than 10 percent of its funds from private domestic sources, and instead relied almost entirely on government deposits, investments of the social security institutions, central bank credit, and foreign loans. By 1982 foreign loans made up almost 40 percent of its liabilities. In addition, as a result of guarantees extended, DBP had contingent liabilities of $847 million in foreign currency, or 18 percent of total assets.24 All government-owned banks have been heavily reliant on central bank rediscounting as sources of funds, and that dependence increased markedly during the early 1980s (table 5.7). The problems that DBP had in the 1980s came from several sources. Much of the bank’s loan portfolio was in large-scale, capital-intensive industry-mining, cement, textiles, metals, synthetic materials-that had been built up during the 1970s and was hit by low commodity prices, high Table 5.7

Government chartered

The Use of Central Bank Credit by Government Banks (as percentage of credit to all commercial banks)

I960

1910

1980

1983

1984

19.8

6.0

35.7 20.4 11.3

0.0

0.0

11.5

31.6 10.5 18.9 24.3

21.2 7.7 11.6 12.3

PNB DBP Government acquired

Source: Presidential Commission on Government Reorganization (1985a.34). table 12.

500

Robert S. Dohner and Ponciano Intal, Jr.

energy costs, high interest rates, and the real depreciation of the currency in 1983. A prominent example was Marinduque Mining, to which DBP had a P. 6.3 billion exposure in 1983 (including guarantees), or 11 percent of total assets. Political influence was an important determinant in many of these loans, as DBP lent heavily to projects organized by favored individuals of the martial law regime. Possibilities for corruption abounded in such arrangements, particularly if foreign loans and loan guarantees were involved. A borrower would contract for imported equipment at inflated prices (or in some instances would purchase used equipment invoiced as new), and in return the equipment supplier would make a kickback to a foreign bank account owned by the borrower.25 In such a situation, the borrower might have little interest in the ultimate profitability of the investment; the money was made by the investment going forward. Later, in the early 1980s, when many of the firms to which loans and guarantees had been made were failing, these loans were converted to equity and additional funds were extended through DBP and other government financial institutions. From 1981 to 1984 the assets acquired by DBP from firms unable to service their loans rose from P. 1.3 billion to P. 8.3 billion.26 A memorandum from the central bank governor to President Marcos in August 1983 described 87 problem accounts of P. 5 billion or more. Of these, 44 accounts amounting to P. 28.2 billion ( $ 2 billion) were classified as “behest” accounts, financed at the request of the government (Laya 1983). Under the weight of its deteriorating portfolio, DBP’s income fell steadily. Loan and guarantee payments as a percent of total outstanding loans fell from 16.6 percent in 1980 to 6.7 percent in 1983 (Lamberte 1984, 2 2 ) . By the end of 1983 DBP was insolvent (Laya 1983, 1 ) . In order to hide this fact from foreign creditors, DBP apparently doctored its financial statements by listing noncash income as profits.*’ Less information is available about the operations of PNB. PNB had lent extensively to the sugar industry during the 1960s and 1970s for the construction of sugar mills. However, the price of sugar fell sharply from its 1974 peak, and the number of sugar mills constructed greatly exceeded the requirements of the industry. PNB acquired many of these mills in the late 1970s. At this time PNB’s subsidiary, NIDC, was set up to manage these acquired assets. In addition to DBP, PNB also had a large exposure to Marinduque Mining. And, like DBP, PNB had extended loans to a number of favored associates of the martial law regime and participated in rescue operations in the early 1980s. Major beneficiaries were Delta Motors (Ricardo Silverio) and the Construction and Development Corporation of the Philippines (Rodolfo Cuenca). PNB also acquired majority ownership of Pilipinas Bank in 1980 when that bank failed. Total earnings as a proportion of assets for PNB averaged 10.6 percent from 1980 to 1983, well below the prevailing loan rate of over 18 percent (Tan 1984, 67-68). And by 1984 the government’s

501

PhilippineKhapter 5

Commission on Audit estimated PNB's overdue loans at P. 21.9 billion, 42 percent of all loans outstanding (1985, 2 ) . The portfolios of the two government social security institutions had also become heavily entwined in these troubled assets. Total assets of the two institutions were P. 28.5 billion in 1983, or 7.5 percent of GNP. More than 65 percent of these assets were invested in PNB and DBP. In addition, the government employees insurance system, GSIS, made direct investments in Philippine Airlines, two hotels, CDCP, and Philippine Cellophane Film (Disini), as well as loans without interest to the Human Settlements Development Corporation. Of its P. 3.9 billion (23 percent of its assets) invested in securities, most have never paid dividends.28 The private sector retirement institution, SSS, had less invested directly in troubled companies, but did have part ownership (along with the government's Land Bank) in Union Bank. The portfolio problems of government financial institutions continued to mount during the adjustment period. By 1986 the total exposure of government financial institutions to nonperforming assets reached P. 1 13 billion, or a value equivalent to almost 20 percent of GNP. Estimates of the distribution of nonperforming assets are shown in table 5.8. Of the P. 113 billion total exposure, slightly over 30 billion came from additional outlays of the institutions to maintain or rehabilitate acquired assets.29 Although the national government made equity contributions to PNB and DBP during the early 1980s, much of the support for these institutions came from the central bank and was outside the budget. Central bank contributions were cut off under the 1984 IMF program, and the budgetary drain resulting from the financial institution losses began to soar. By 1986, capital infusions into the financial institutions accounted for 17 percent of government expenditures and 3 percent of GNP (table 5.9).

5.5 Financial Markets and the Debt Crisis Philippine financial markets played several important contributing roles in the development of a debt crisis in the phi lip pine^.^' Interest rate controls Table 5.8

Nonperforming Assots of Government Financial Institutions, 1986 (in billions of pesos)

DBP PNB NlDC PhilGuarantee Total

Total Exposure

Total Claimsa

60.5 40.9 6.0 5.9 -

72.4 51.9 13.0 5.9 -

113.3

143.2

~

Source Ministry of Finance, cited in Business Day, 8 September 1986, p 2

"Total exposure plus unbooked interest and other charges

Robert S. Dohner and Ponciano Intal, Jr.

502 Table 5.9

Budgetary Contributions to Government Financial Institutions (in billions of pesos)

Amount Percentage of government expenditure Percentage of GNP

I984

1985

1986

8

10 13 1.7

19 17

12

1.5

3.1

Source: Philippines, Government Corporate Monitoring Committee.

discouraged financial mobilization in the period of rapid economic growth, limiting the ability of the economy to generate savings and make them available to borrowers through the banking system.31 In addition to the low interest rates offered to depositors, Philippine financial regulation increased the cost of bank intermediation, resulting in high effective rates to borrowers. Reserve requirements were high, and the requirement to loan to agriculture in fact resulted in the channeling of low interest rate funds to the public sector. Both reduced the yield on a large portion of bank portfolios, raising the cost of supplying funds to other borrowers. As a result of these two policies and other efforts to direct credit to priority sectors, effective rates of interest on commercial bank loans were substantial and exceeded the cost of foreign borrowing through much of the 1970s. Controlled interest rates and high effective loan rates were characteristics that the Philippines shared with many less developed countries. What the Philippines added was innovative, as well as reckless, domestic financial institutions that made the financial system more unstable and vulnerable to a tightening of credit or downturn in the domestic economy. An active market in deposit substitutes developed at the initiative of several investment houses, offering better rates to depositors and access to credit to many smaller borrowers. Attempts by the central bank to impose reserve requirements led this market to evolve in ways that made it more unstable. The way in which the banking system developed, with many small banks serving the interests of their industrial or family group, the tendency to lend to directors, stockholders, and related individuals, and relatively lax banking and securities regulation led to substantial abuse. Add to this the rapid growth of several conglomerates based on their proximity to the martial law government and high financial leveraging, and one had a highly volatile mixture waiting for an event like the departure of Dewey Dee. The large degree of government involvement in financial markets, through ownership of major financial institutions, extensive regulations channeling credit to priority sectors, and the leverage that the government had over nominally independent banks, encouraged the flow of funds at low cost to politically favored, and ultimately very low return, investments. The effective control of the government over the flow of financial resources was

503

PhilippinesKhapter 6

increased by the large use of foreign borrowing and the preference of external lenders for government entities and government guarantees. Heavy state involvement in the financial sector encouraged the government to use public financial institutions for fiscal purposes. The corporate rescue operations of the early 1980s were conducted primarily through PNB, DBP, GSIS, and indirectly through the central bank. This had the effect of hiding the true cost of the rescue operations, as well as substantially weakening the portfolios of the major financial institutions. By 1985 and 1986 the fiscal problem would be primarily a public financial institution problem, as we will explain in detail in chapter 7. Philippine financial markets contributed to the debt crisis; they also made adjustment to the crisis more difficult, as government demands for financial resources nearly starved the private sector in the adjustment episode. In chapter 7 we take up the adjustment period in detail, but before that we examine the one remaining piece of the Philippine debt crisis story, the accumulation and management of the external debt.

6

External Debt and Debt Management

The 1970s opened with an external debt crisis in the Philippines that was in some ways similar to the current crisis. Expansionary policy during the first Marcos administration, coupled with heavy external borrowing on short term, led to the crisis. The debt/GNP ratio for the Philippines rose from 10 percent in 1965 to 22 percent by 1969, and debt maturing within the next year amounted to one quarter of total external debt. In tandem with an IMF stabilization program and a float of the peso, the Philippines negotiated longer maturities for much of the outstanding short-term debt of the public sector. The crisis led to a number of changes in external debt policy. Republic Act 6142 (November 1970) established a comprehensive system of control and information for foreign borrowing. Under the system all external borrowing by the public or the private sector, except short-term borrowing by the commercial banking sector, required prior approval of the Monetary Board. The Management of External Debts and Investment Accounts Department (MEDIAD) was set up within the central bank to screen applications for external borrowing. Applicants were required to submit information on the proposed project and its likely returns, as well as details of the financing involved. The central bank could, and did, set minimum requirements for

504

Robert S. Dohner and Ponciano Intal, Jr.

maturities and grace periods, as well as maximum interest rates and spreads on foreign loans. All transactions on external loans-drawings, interest and principal payments, and refinancings-had to be reported to MEDIAD, which maintained statistics on Philippine external debt. In addition to the approval and reporting requirements, the act established a statutory limitation on debt service, and therefore implicitly, on total external indebtedness. Debt service payments were not to exceed 20 percent of average foreign exchange inflows of the three preceding years. In 1970 the Philippines was above this level, but the statutory limitation was achieved by 1972. The monitoring and control system covered all nonmonetary debt, defined to include all direct borrowings of the nonbanking sector and medium- and long-term funds borrowed by the banking sector and then lent on to the nonbanking sector. Short-term liabilities of the banking sector were not under the control of MEDIAD and were not generally included in Philippine debt statistics prepared before the 1980s.’ In fact, external borrowing by the banking sector has generally been encouraged through the swap facilities made available to commercial banks by the central bank and through the Foreign Currency Deposit System. The Foreign Currency Deposit System was established in 1970 to attract offshore funds, including those held by Philippine residents, to the country and to develop the Philippines as a financial center. Subsequent modifications of the rules governing the system in 1976 led to rapid growth in the rest of the decade. Under the Foreign Currency Deposit System domestic banks were allowed to create separate accounting units called foreign currency deposit units (FCDUs), which could accept deposits in foreign currency from residents or nonresidents. Secret accounts were permissible, and deposits could be withdrawn and transferred out of the country without restriction. FCDUs could make foreign currency loans to residents or nonresidents and could make peso loans by exchanging foreign currency with the central bank under its swap f a ~ i l i t i e s .FCDUs ~ were considered part of the Philippine banking system, and their liabilities were included in (banking system) external debt. This was the case even if the deposits came from Philippine residents. The liabilities of FCDUs grew rapidly in the late 1970s and early 1980s to a level of just over $8 billion in 1983.4 In addition to the Foreign Currency Deposit System, which was restricted to domestic and foreign-owned commercial banks operating in the country, provisions were made for foreign commercial banks to set up offshore banking units (OBUs) in the Philippines. These units could accept foreign currency deposits from external sources and make loans to nonresidents of the Philippines, with minimal restrictions by the central bank. Lending to Philippine residents in foreign currency was also permissible, but had to be approved by the central bank as a regular external borrowing. OBUs are treated as foreign commercial banks by the Philippines, and their loans to

505

PhilippineKhapter 6

domestic residents, but not their liabilities, are included in Philippine external debt.

6.1 External Borrowing in the 1970s Summary figures on Philippine external debt are presented in table 6.1. The impact of the successful adjustment period in the early 1970s is clearly seen in the debt figures in the table. The current account improved dramatically during these years and registered a surplus of 5 percent of GNP in 1973. Nonmonetary external debt grew modestly, and both the debt/GNP and debt/export ratios fell dramatically. Maturities on nonmonetary debt lengthened, although the share of short-term debt in the total external debt of the Philippines increased as the banking sector resumed normal operations and banking sector liabilities grew. The next four years, 1974-78, saw a surge in foreign borrowing in the Philippines, a consequence of the first oil shock and the acceleration of the country’s development program. Outstanding nonmonetary debt tripled during this period. The growth in real terms was substantially less because of inflation, but both the debt/GNP and the debt/export ratios increased considerably. By the end of the 1970s, indicators of Philippine foreign debt had reached high but certainly not alarming levels. The debt/GNP ratio in the Philippines in 1979 was comparable to that of Korea and Indonesia and somewhat below the Latin American average. The Philippine debuexport ratio was well above Korea’s, but well below most of Latin America (see table 6.6 below). Several changes in the characteristics of Philippine debt took place during the last half of the 1970s. The first was a marked shift in external borrowing toward the public sector. After falling in the early 1970s, the share of public sector debt rose sharply, reaching two-thirds by the end of the decade (table 6.2). As was the case for other LDC borrowers during the 1970s, there was a shift in the Philippines toward commercial sources of finance and toward debt with floating interest rates. Between 1976 and 1980 the share of banks and other financial institutions in the Philippine credit mix rose to just over half, shown in table 6.3. But the Philippines put considerable effort into diversifying its sources of finance. One source was the international bond market, which the Philippines tapped for about 10 percent of its mediumand long-term funding. The Philippines also assiduously courted multilateral and bilateral sources during the 1970s, receiving an increasing share of funding from these two. Major official sources for the Philippines were the World Bank, the Asian Development Bank, and the U.S. and Japanese governments. Creditor breakdowns for earlier years in the 1970s are available for medium- and long-term debt of the public sector and show the same patterns (table 6.4). Aid from official sources was largely coursed

Table 6.1

Philippine External Debt (in millions of US. dollars) 1970

Nonmonetary debt Medium & long term IMF obligations Other Short term Monetary Fector deht Total debt Total debt As 9 of GNP Ar % of exports Monetary rector debt db 7 c of totdl Short-term debt As 7~of total deht AF 9 of nonmonetary debt GNP US$

2.138 1,779 I08 1.671 359 159 2.297

33.2 I74 6.9 22.6 16.8 6.9 I 4

1972 2.221 1,901 I45 1.756 320 51 I 2.732

1973

1974

1975

1976

1978

1979

2.306 2.025 I39 1.886 28 I

2.726 2.395 131 2.264 33 I 1,029 3.755

3.422 2,985 238 2.147 438 1.517 4.939

5.120 4.405 450 3.955 7 15 I.64X 6.768

8.189 6.932 626 6.306 1.257 1.505 10.694

9.899 8.086 718 7.368 1.813 3.453 13.352

580 2.886

18.7

27 0 II 4 20. I

25.5 106 27.4

31.3 I54 30.7

37.5 I96 24.3

44 5 218 23.4

45.2 213 25 9

30.4 14.4 8.381

29.8 12.2 10.685

36.2 12.1 14.71 I

39.6 12.8 15.789

34.9 14.0 18.037

35.2 15.4 24.033

39.4 18.3 29.553

32.6

I88

1980 Nonmonctary debt Medium & long term IMF obligations Other Short term Monetary sector debt Total debt Total debt As % of GNP As % of exports Monetary sector debt as % of total Short-term debt As % of total debt As % of nonmonetary debt GNP US$

12.318 9,770 936 8.834 2,548 4.934 17.252

1981

1982

1983

1984

1985

1986

1987

14.990

17.601 13.141 908 12,233 4,460 7.076 24.671

19.468 15.412 1.013 14.399 4.056 5.348 24.816

20.21 I 15.926 844 15.082 4,285 5.207 25.418

21 2 7 0 17.679 1,115 16.564 3.59 I 4.982 26.252

25.668 22.878 1,243 2 1.635 2.790 2.588 28.256

26.702 24.857 1.183

1 1.326

I .030 10,296 3,664 5.903 20.893

1.845 I .947 28.649

49.0 215 28.6

54.4 242 28.3

62.8 308 28 7

72.7 305 21.6

80.6 317 20.5

81.7 332 19.0

94. I 327 9.2

84. I 311 6.8

43 4 20.7 35,218

45.8 24.4 38,435

46.7 25.3 39.278

37.9 20.8 34.136

37.3 21.2 31,517

32.7 16.9 32.124

19.0 10.9 30,021

13.2 6.9 34,056

Source: Philippine central bank. MEDIAD and Financial Plan Data Center.

Table 6.2

Private sector debt Medium & long term Short term Public sector debt Medium & long term IMF obligations Short term Total nonmonetary debt Percentage shares. Nonmonetary debt Private Public Medium- & long-term debt Private Public Short-term debt Private Public

Philippine Nonmonetary External Debt, Public and Private Sector (in millions of U S . dollars and percentages) 1970

1972

1973

1974

1975

1976

1978

1979

919 I75

983 297

1,041 262

1,213 320

1.411 409

1.812 568

2,091 936

2,072 1,247

75 1 I08 184

762 145 23

845 I39 19

1.048 131 II

I .3 I6 238 28

2.120 450 147

4,192 626 321

5.211 718 566

2,137

2,210

2,306

2,723

3,402

5.097

8.166

9.814

51.2 48.8

57.9 42. I

56.5 43.5

56.3 43.7

53.5 46.5

46.7 53.3

37. I 62.9

33.8 66.2

55 0 45 0

56.3 43.7

55.2 44.8

53.6 46.4

51.7 48.3

46. I 53.9

33.3 66.7

28.4 71.6

48.9 51 I

92.8 7.2

93.2 6.8

96.7 3.3

93.6 6.4

79.4 20.6

74.5 25.5

68.8 31.2

Private sector debt Medium & long term Short term Public sector debt Medium & long term IMF obligations Short term Total nonmonetary debt Percentuge shares: Nonmonetary debt Private Public Medium- & long-term debt Private Public Short-term debt Private Public Source:

1980

1981

1982

1983

1984

1985

1986

1987

2.454 1.601

2.761 1,802

3.229 2,123

3.125 1,949

2.71 I 2.092

2,643 2.405

3.692 2.199

4.020 I .259

6.292 936 947 12.230

7.443 1,030 1.863 14,899

8.874 YO8 2,337 17,471

10.951 1.013 2.107 19.145

12,173 844 2.193 20.013

13,009

17.943 I .243 59 I 25.668

19.654 1,183 586 26.702

1.115

1.186 20.358

33.2 66.8

30.6 69.4

30.6 69.4

26.5 73.5

24.0 76.0

24.8 75.2

23.0 77.0

19.8 80.2

28. I 71.9

27 I 72 9

26.7 73.3

22.2 77.8

18.2 81.8

16.9 83.1

17.1 82.9

17.0 83.0

62.8 37.2

49 2 50.8

47.6 52.4

48. I 51.9

48.8 51.2

67.0 33.0

78.8 21.2

68.2 31.8

Central hank, MEDIAD

Note: The classification system for nonmonetary debt differs slightly starting in 1986. Figures for 1985 and earlier are approximately. hut

not strictly, comparable

Table 6.3

Nonmonetary Medium- and Long-term Debt Characteristics (percentage shares) 1976

Total medium- and long-term debt Total external debt Creditors: Mukikdteral institutions Governments Bonds Banks, other financial institutions Suppliers credits Maturity: 1-5 years 5- 12 years Over 12 years Interest Rate: Fixed Floating Borrower: Public Private, public guarantee Private, nonguarantecd

3,932

1977 5.024

1978 6,283

1979 7,283

1980 8,746

1981

1982

1983

1984

1985

10,204

12,103

14.076

14,884

15,652

10.4 7. I 9.4

11.9 12.0 10.0

13.4 12 7 12.4

15.2 10.5 12.2

16.2 9.4 11.0

18.9 10.5 9.9

19.1 10.2 8.3

21.2 9.3 7.7

22.4 9.4 6.7

22.5

41.3 31.8

42. I 23.9

42.8 18.7

50.3 11.8

51.0 12.5

50.4

52.0 10.4

51.2 10.7

51.1 10.3

7.2 53.3 39.4

6. I 58.2 35.7

4.1 57.3 38.6

2.5 56.7 40.8

I .9 56.9 41.2

53.7 45.1

1.4 51.5 47. I

1.3 47.5 51.2

3.0 44.5 52 5

82.8 17.2

74.9 25. I

73.1 26.9

68.4 31.6

63.4 36.6

61.0 39.0

56.6 43.4

60.I 39.9

60.7 39.3

58.8 39.5

53.9 14. I 32.0

56.9

66.7 7. I 26.2

71.6 6.9 21.6

71.9 6.8 21.3

72.9 5.9 21.1

73.3 7.3 19.4

77.8 7.3 14.9

81.8 6.3 11.9

83.1 4.9 12.0

11.0

32.0

10.3 1.1

Source: Central bank, MEDIAD. Data on total external debt from the central bank, Financial Plan Data Center

1985

1986

1987

26,252

28,256

28,649

5.9

17.1 10.9 na

16.6 13.9 na

17.6 18.2 na

50.9 9.5

58.2 12.4

57.3 10.9

53.1 8.2

11.1

47.8 52.1

511

Table 6.4

Philippines/Chapter 6 Public Sector Medium- and Long-Term Debt by Creditor (in millions of U.S. dollars and percentages) 1970

Total medium- & long-term debt Public sector Percent shares: Multilateral Bilateral Bonds Financial institutions Supplier credits

1,491 572 21.0 19.4 I .9 46.7 10.9

1975 2,789 1,377 24.6 39.1 I .6 29.5 5.1

1979 7,230 5,159 22.9 21.9 15.2 35.2 4.9

1980 8,918 6,464 23.2 19.7 13.9 38.7 4.5

1981

1982

10,338 7,578

11,992 8,836

26.5 19.3 11.2 38.9 4.0

26.0 17.5 9.9 42.9 3.7

Source; World Bank (1984a). table A.8 Nore: This table is based on an older data series which has since been revised. Comparable breakdowns from revised data are not available.

through the public sector, and thus has a much larger share among the creditors shown in table 6.4. As we point out in table 6.3, the proportion of Philippine medium- and long-term debt at floating interest rates doubled to 37 percent by the end of the decade. However, the use of floating rate debt was not as extensive in the Philippines as it was in other medium income countries, and at 40 percent in the early 1980s it was well below that of Latin American debtors. The Philippines was also able to make heavier use of bilateral and multilateral loans than were Latin American debtors, where the share of debt to official creditors averaged 18 p e r ~ e n t . ~ Philippine authorities paid close attention to debt management during most of the 1970s. Short-term debt as a proportion of nonmonetary debt was reduced steadily until the mid- 1970s, particularly short-term debt of the public sector. Within medium- and long-term debt, there was a continuing shift toward longer maturities (see table 6.3). The Philippines also used the opportunities presented in 1978 and 1979 to refinance a portion of existing loans and lengthen maturities. By the late 1970s, the average maturity on fixed term credits was just over 13 years.6 Although it is difficult to measure or prove, the Philippines probably received better terms on foreign borrowing through the minimum terms and queuing that were enforced through the approval process. In order to limit the number of approaches to the market, the central bank borrowed in large amounts in its own name for smaller and less well-known Philippine applicants. Through this Consolidated Foreign Borrowing Program. the central bank borrowed $2.4 billion between 1978 and 1983. About two-thirds of the funds raised went to public sector entities (including DBP) and the rest to private sector applicant^.^ Although the loans carried the

512

Robert S. Dohner and Ponciano Intal, Jr.

guarantee of the central bank, the foreign exchange risk was borne by the ultimate recipients of the loans. Although no single indicator adequately describes the burden and risk associated with a given level of external debt, the most common one used is the ratio of debt service payments to export earnings. Despite the rapid accumulation of external debt during the 1970s, there was only slight movement of the Philippine debt service ratio (defined here as interest on nonmonetary debt, plus principal repayments on medium- and long-term debt divided by exports of goods and services). The debt service ratio declined from a level of 30 percent in 1970 to below 20 percent by 1973 (table 6.5). The ratio varied somewhat, but maintained a level of about 18 percent between 1975 and 1980. International comparisons, using a definition of the debt service ratio that does not include service exports, are shown in table 6.6. Here the Philippines shows up with high debt ratios relative to the other Asian countries, but generally moderate ratios when comparisons are made to Latin American countries or other LDCs outside the region. At the end of the 1970s the Philippines was a high external debt country but, on macro indicators, not a worrisome one. A second debt service ratio, unique to the Philippines, is important in discussing Philippine debt. This ratio was defined by Republic Act 6142 in 1970 and influenced, and to some extent constrained, Philippine behavior. In addition, it may have misled domestic policymakers and external creditors about the debt risk of the Philippines. The statute limited total debt service payments to less than 20 percent of total foreign exchange receipts. The numerator in this statutory debt service ratio included principal and interest payments on medium- and long-term debt, IMF obligations, and short-term debt of fixed maturity. The definition excluded debt service on revolving credits and on short-term banking system liabilities. The denominator included all external receipts on a cash basis, including capital inflows, SDR allocations, and monetization of gold. In the original ratio, foreign exchange receipts were to be the average of the three preceding years, but this was later amended by presidential decree to refer to just the preceding year. The statutory debt service ratio was unusual in that it included capital inflows as well as current earnings in the denominator. What this meant was that additional foreign borrowing could actually lower the statutory ratio in the following year, if the interest rate plus rate of amortization was below the current level of the statutory ratio. In addition, the statutory ratio left out short-term revolving credits, which became a major part of debt accumulation in the 1980s. The statutory ratio tracked the conventional debt service ratio closely during the 1970s, as shown in table 6.5, but the two measures diverged greatly during the 1980s. Between 1980 and 1982, when the conventional debt service ratio almost doubled, the statutory ratio increased by a mere 4 percent.

Table 6.5

Medium- & long-term debt Principal Interest Total debt service Debt service ratio As Z of exports of goodsandservices Statutorily defined Interest as % of debt service

Debt Service on Philippine External Debt (in millions of U.S. dollars and percentages) 1970

1972

1973

1974

I975

1977

1978

271 115 386

264 120 384

378 125 503

369 I52 521

355 234 589

378 236 614

545 440 985

1979

1980

1981

1982

1983

1984

634 626 1,260

692 975 1,667

794 1,374 2,168

1,060 1,990 3,050

1,119 1,985 3,104

1,225 2,257 3,482

1985

713 2,208 2,921

1986

890 2,046 2,936

1987

1,027 2,226 3,253

29.2 34

26.4 19

19.9 20

14.6 20

18.4 17

14.5 13.7

20.1 18

20.1 18.6

20.8 18.7

25.2 19.7

38.1 19.4

38.2 19.1

43.4 na

36.9 na

34.0 na

35.3 na

29.8

31.3

24.9

29.2

39.7

38.4

44.7

49.7

58.5

63.4

65.2

63.9

64.8

75.6

69.7

68.4

Source: Central bank. MEDIAD and Financial Plan Data Center.

514

Robert S. Dohner and Ponciano Intal, Jr.

Table 6.6

Comparative Debt Ratios, 1979 and 1986 1979

1986

Total Debt Service as % of Exports

Total DebVGNP

Egypt Israel Algeria Peru Venezuela Pakistan Morocco Philippines Chile Korea Bangladesh Mexico

62.8 61.2 49.1 47.2 43 4 40.7 40.3 32.4 31.9 31.6 29.6 29.4

NOLDC~"

21.2 19.4 19.4 19.0 15.1 12.9

Thailand Turkey Malaysia Argentina India

Egypt

76.9 43.8 35.4 29.4 27.3 26.6 24.5 24.3 20.7 20.4 20.1 16.8 16.4

NOLDCS~ Indonesia Bangladesh Thailand Malaysia

15.0 14.5 11.2 8.2 5.5

Mexico Brazil Chile Morocco Argentina Algeria Pe N Philippines Pakistan Venezuela Turkey Korea

Debt Service on MLT Debt" as % of Exports

MLT Debtd/ GNP Ratio Chile Malaysia Mexico Philippines Israel Venezuela Egypt Argentina Peru Indonesia Bangladesh Turkey Brazil Korea Pakistan Thailand Algeria India ~~~

120.1 77.0 16.1 72.2 72. I 66.9 58.8 51.7 50.5 49.7 47.5 42.3 37.6 36.1 36.0 35.2 24.8 15.1

Argentina Algeria Mexico Brazil Venezuela Chile Indonesia Turkey Israel Pakistan Thailand Bangladesh India Korea Egypt Philippines Peru Malaysia

64.1

54.8 51.5 41.8 37.4 37. I 33. I 32.4 27.5 27.2 25.4 25. I 24.6 24.4 23.8 21.3 20.5 20.0

~~

Source Campbell (1982). dnd World Bank, World Debt Tables 1980 and World Developmenr Report 1981.

1988 "Medium- and long-term debt bNonoil cxporting less developed countries

6.2 Foreign Borrowing and Debt Management in the 1980s The relatively sanguine position of the Philippines at the end of the 1970s changed extremely rapidly during the early 1980s.The current account deficit widened sharply as the second oil price shock, the fall in commodity prices, and the rise in world real interest rates hit the country. Once again the Philippines adopted a countercyclical policy, increasing the public investment program to maintain domestic incomes. The budget deficit swelled further as a a result of rescue operations mounted for domestic firms in the wake of the Dewey Dee crisis and the failure of crony and capital-intensive firms. This was also a period of increasing opposition to the Marcos regime. A number of protests and bombings occurred in 1980, and a wave of strikes took place after the lifting of martial law in early 1981. Capital flight, a continual phenomenon in the 1970s, increased dramatically in 1981. Errors and omissions in the balance of payments turned sharply negative in that year, a $500 million dollar turnaround from 1980 equivalent to 35 percent of the capital account.

515

PhilippinedChapter 6

The combined result of these events was a very rapid increase in external debt during a period of much higher real interest rates. The debt/GNP ratio rose from 35 percent in 1980 to 56 percent in 1983. The debt service ratio, which had remained almost constant during the last half of the 1970s, skyrocketed from 1980 to 1983; it jumped from 21 to 38 percent, making the Philippines an obvious problem debtor. Four factors explain the extraordinary rise of the debt service ratio during the early 1980s. The first is the increase in world interest rates. In 1980 47 percent of Philippine nonmonetary debt was short-term debt or debt contracted at floating interest rates.' As world interest rates rose, so did debt service. One rough measure of this effect is interest payments during the year divided by the average level of outstanding debt during the year; this rose from 5.7 percent in 1979 to 9.5 percent in 1982. The second reason behind the sharp rise in the debt service ratio was a reduction in export earnings. Between 1981 and 1982 the dollar value of goods and services exports fell by 7 percent after having risen by an average of 14 percent per year from 1974 to 1980. Export receipts increased only slightly during 1983 and then fell again during 1984. The decline was sharpest for merchandise exports, which fell by 14 percent between 1980 and 1983. Service receipts, which accounted for almost one-third of exports, leveled off in 1981. The Philippines was hit by the appreciation of the U.S. dollar, which weakened international commodity prices in dollar terms. There was little relief on the debt side, since three-quarters of Philippine external debt was in dollars and most of the rest in Japanese yen, also a relatively strong currency (table 6.7). Table 6.7

US. dollar Deutsche mark Japanese yen Pound sterling French franc Others Total Percentages: U.S. dollar Deutsche mark Japanese yen Pound sterling French franc Others

Philippines Fixed-term, Nonmonetary External Debt by Currency of Repayment (in millions of U.S. dollars) 1979

1980

I98 1

1982

1983

1984a

1985b

6,016.5 194.3 1,320.7 106.3 252.7 224.4

7.412.7 163.7 1,558.5 98.7 238.2 237.8

8,895.1 138.2 1,824.4 102 176.7 235.9

10,343.5 164.5 1,987.8 74.9 172.1 340.0

11.902.4 230.7 2,402.7 44.2 88.1 466.5

11,931.3 218.0 2,630.3 77.1 133.7 737.6

13,163.9 183.2 2,458.9 50.6 80.0 561.4

8,114.9

9,709.6

11,372.3

13,082.8

15,134.6

15,728.0

16,501.0

74.1 2.4 16.3 1.3 3. I 2.8

76.3 1.7 16.1 1 .o 2.5 2.4

78.2 1.2 16.0 0.9 I .6 2. I

79.1 1.3 15.2 0.6 1.3 2.6

78.6 1.5 15.9 0.3 0.6 3.1

75.9 1.4 16.7 0.5 0.9 4.7

79.8 1.1 14.9 0.3 0.5 3.4

Source: Central bank, MEDIAD. "Excludes short-term loans ($144 million total in 1984) bExcludes short-term loans, and excludes Central Bank assumed obligations amounting to $266 million.

516

Robert S. Dohner and Ponciano Intal, Jr.

But Philippine external borrowing increased to a much greater extent than indicated by the worsening of the current account deficit, as domestic residents shifted their own capital overseas. Capital flight has long been a feature of the Philippine economy, but the scale increased dramatically in the early 1980s in response to domestic political and economic uncertainty and the increasing overvaluation of the Philippine peso. In table 6.8 we present two measures of capital flight in the Philippines. The first (col. 1-5) is the gap between total external borrowing and the recorded uses of foreign exchange in the Philippine economy. The second measure (col. 6-7) is a narrower definition and includes only errors and omissions from the balance of payments.' Both measures show a sharp increase in capital outflows starting in 1981, two years before the Philippines declared a moratorium. During 1981 and 1982 the difference between total external borrowing and identified uses of foreign exchange reached $1.85 billion per year, or 3.4 percent of GNP. The measures in table 6.8 understate the true extent of capital flight in the Philippines, since much of it has occurred through manipulation of trade invoices, primarily underinvoicing of exports. Comparing recorded exports of the Philippines and imports from the Philippines as recorded by its trading partners shows a similar increase in capital flight starting in 1980 and averaging approximately $690 million a year (1.3 percent of GNP) for 1980 Table 6.8

1971 1972 1973 I974 1975 I976 1977 1978 1979 1980 1981 1982 1983 1984 1985

Philippine Capital Flight, Balance of Payments Measures (in millions of U.S. dollars)

External Debt Increase

Current Account Deficit

Other Uses of Foreign Exchange"

Capital Outflow

Cumulative Capital Outflow

Errors and Omissions

Cumulative

(1)

(2)

(3)

(4)

(5)

(6)

(7)

96 339 154 869 1,184 1,829 1,301 2,625 2,658 3,900 3,641 3,784 139 602 834

3 -9 - 536 I76 892 1,050 752 1,102 1,497 1,904 2,061 3,200 2,750 1,116 77

137 230 643 534 - 23 - 18 - 159 758 470 1,301 - 571 - 952 - 1,137

-44 118 47 159 315 797 708 165 69 1 695 2,157 1,536 - 1,474 -529 708

-44 74 121 280 595 1,392 2,100 2,865 3,556 4,251 6,408 7,944 6,470 5,941 6,649

230 136 3 86 399 - 20 -81

230 366 369 455 854 834 753 808

15

49

55 200 55 545 528 428 - 144 - 638

1,008

1,063 1.608 2,136 2,564 2,420 1,782

Source: Dohner ( l987), table 1.

"Increase in central bank reserves and banking system foreign assets, less net inflow of foreign direct investment.

517

PhilippinedChapter 6

through 1982." What is interesting about all of the measures is that capital flight accelerated in the early 1980s, well before the Aquino assassination in 1983. In fact, the measures in table 6.8 and trade invoice comparisons indicate that capital flight diminished in 1983. The exodus of capital from the Philippines occurred for a variety of reasons. Political uncertainty heightened in the early 1980s due to the increased strength of the communist New People's Army and questions about Marcos' health and presidential succession. The increasing business influence of Marcos and his cronies and their aggressive actions to take over successful domestic enterprises led businessmen to transfer funds out of the country. In addition, there were the more conventional economic motivations for capital flight-a desire to earn higher real returns on financial investment, anticipate devaluation, or avoid domestic taxation. I ' In table 6.9 we compare capital flight in the Philippines to that of other LDC debtors. Although capital flight was not the driving force in the Philippines that it was in Argentina and Venezuela, it did play an important role during the period of debt accumulation, accounting for about one-third of the total debt buildup. And the acceleration of capital flight in the early 1980s was part of what swept the Philippines into crisis. The final factor contributing to the sharp rise in the Philippine debt service burden was a surge in short-term borrowing in the late 1970s and early 1980s. During this period short-term debt grew from 15 to over 25 percent of nonmonetary debt. These loans were contracted at high interest rates and left the country vulnerable to shifts in external confidence. After declining steadily in importance during the early part of the 1970s, short-term, nonmonetary debt began to rise after 1976 (table 6.10). Most short-term debt was in the form of revolving credits, and open accounts (O/A) and documents against acceptance (D/A) used in trade finance. Acceleration of the payments schedule by international oil companies after the second oil price shock forced the Philippines to use bank financing for oil imports and was responsible for much of the growth in short-term credit.

Table 6.9

Philippines Argentina Brazil Korea Mexico Venezuela

Capital Flight in Selected LDC Debtors (in billions of U.S. dollars)

Capital Flight, 1976-82

Increase in External Debt, 1976-82

Capital Flight as 70 of Debt Accumulation

7.3 25.3 12.6 6.1 36. I 20.5

19.7 34.1 65.9 28.4 67.9 26.3

37% 74

Source: Cumby and Levich (1987). table I (using Erbe definition of capital flight).

19

21 53 78

Table 6.10

Revolving credits Private Public Short-term fixed Total short-term debt Private Public Total nonmonetary debt Private Public Short-term debt as % of total nonmonetary debt Distribution of short-term debt (%) Private Public Short-term debt as % of total sector debt Private Public Source: Central bank,

Short-term Nonmonetary Debt by Sector (in millions of U.S. dollars) 1970

1972

1973

1974

1975

1976

1978

1979

1980

1981

1982

1983

1984

1985

305 103 203 54 359

273 254 19 9 281 262 19

324 313 7 33 1 320 11

312 283 29 126 438 409 29

571 424 147 144 715 568 141

1.197 927 270 160 1,357 1,036 321

1,618 1,130 489 195 1,813 1,254 559

2,477 1,541 936 71 2,548 1,601 947

3,552 1,827 1,725 I12 3,664 1,903 1,761

3,919 1,824 2,096 74 3,993 1,897 2,096

3,968 2,041 1,926 45 4,013 2,045 1,968

4,038 2,100 1,938

208

294 249 45 26 320 272 49

4.182 2,104 2,078

2,925 1.978 948 142 3,068 1,981 1,087

2,137 1,071 1,067

2,210 1,255 956

2,306 1,303 1,003

2,723 1,533 1.190

3,403 1,820 1,582

5,099 2,380 2,719

8,166 4,028 4,139

9,814 4,392 5,423

12,230 5,455 6,775

14,898 6,386 8,512

17,471 7,234 10,237

19,145 6,976 12,169

20,013 6,703 13,310

20,358 6,686 13,672

151

11

144

16.8

14.5

12.2

12.2

12.9

14.0

16.6

18.5

20.8

24.6

22.9

21.0

20.9

15.1

42.1 51.9

84.8 15.2

93.2 6.8

96.6 3.4

93.5 6.5

79.4 20.6

76.3 23.7

69.2 30.8

62.8 37.2

51.9 48. I

47.5 52.5

51.0 49.0

50.3 49.7

64.6 35.4

14. I 19.5

21.6 5.1

20.1 1.9

20.9 0.9

22.5 1.8

23.9 5.4

25.7 7.8

28.6 10.3

29.3 14.0

29.8 20.7

26.2 20.5

29.3 16.2

31.4 15.6

29.6 8.0

MEDIAD.

-

519

PhilippineKhapter 6

However, particularly in 1981 and 1982, the growth of revolving credit was not trade related but was used to replace maturing long-term credits, particularly by public sector enterprises. Oil import financing was also used to provide credit for other purposes. By 1982 outstanding oil import credits had risen to 85 percent of annual oil imports, although the financing period of the transactions was less than six months. Both PNOC and the central bank were involved in oil import financing on a partly overlapping basis. Although both private and public short-term borrowings increased, the growth of short-term debt was concentrated in the public sector. Between 1980 and 1982, two-thirds of the increase in short-term debt came from public sector borrowings. As a result the public sector, which had accounted for as little as 3 percent of total short-term indebtedness, accounted for over half in 1982,. The vulnerability entailed in relying on short-term debt became clear in 1983. As anxieties about the Philippines built up during late 1982 and 1983, foreign banks began to reduce their exposure to the country. Between the end of 1982 and the end of 1983, short-term nonmonetary debt outstanding to the Philippines fell by 9 percent (table 6.1), forcing a rapid exhaustion of the nation’s reserves. l 2

6.3 Banking System Liabilities and International Reserves As explained above, short-term debt of the banking system was not included within the external debt monitoring and approval process of MEDIAD and the central bank. Banks were generally encouraged to borrow in foreign currency through the Foreign Currency Deposit System and through the swap facilities that the central bank provided. Over the last half of the 1970s, commercial bank external liabilities grew steadily from $1 billion in 1975 to $3.7 billion in 1980 (table 6.11). The commercial banks were continually net borrowers, but during the 1970s the central bank maintained foreign currency assets large enough to give the banking system a net asset position of about 3 percent of GNP. Foreign currency reserves of the central bank were maintained at a level of about four months’ worth of imports. In order to finance a widening payments balance in the 1980s, the central bank reduced its reserve holdings. It also incurred additional foreign currency liabilities by borrowing from the banking sector and by swapping pesos for foreign exchange. As a result the banking sector, which had shown a net asset position up to 1980, registered increasing net liabilities, reaching 7 percent of GNP and 16 percent of nonmonetary debt by 1982. Capital flight accelerated during the last part of 1982, leading to a precipitous decline in central bank reserves. By September 1983 total reserves were below $700 million, or less than 4 percent of annual imports (table 6.12).

Robert S. Dohner and Ponciano Intat, Jr.

520 Table 6.11

Gross and Net Foreign Liabilities of the Banking Sector

(in millions of U.S. dollars) Gross Liabilities

Net Liabilities

Assets

Year

Central Bank

Commercial Banks

Total

Central Bank

Commercial Banks

Total

1975 1976 I977 1978 1979 1980 1981 1982 1983 1984 1985 1986 I987

527 566 39 152 414 1,247 1,493 2,206 2.690 2.696 2,897 3,373 2.685

990 1,082 1,419 2,353 2,979 3,687 4,410 4,870 4,526 4.217 3,338 3,089 3.491

1.517 1,648 1,458 2,505 3.453 4.934 5.903 7.076 7.216 6,973 6,235 6.462 6.176

1,361 1,642 1,525 1.883 2,423 3.155 2.574 1.711 865 886 1,061 2,459 1.959

719 564 738 1.312 1.309 1.904 2,297 2,540 1,655 1,837 1.915 1,986 2,464

2,079 2,205 2,264 3,194 3,732 5,059 4,871 4,251 2,520 2,723 2,976 4.445 4,423

Source:

Central Bank - 834

1.076 - 1,486 - 1.731 - .949 - ,908 - ,081 495 ,825 ,810 ,836 -

914

726

Commercial Banks

Total

272 518 680 1,041 1.670 1,783 2.113 2,330 2,871 2,440 1.423 1.103 1,027

-562 -558 -806 -690 -279 - 125 1,032 2,825 4,696 4.250 3.259 2,017 1,753

IMF. Rerenr Economic Developments: 1985, table 14. and 1988. p. 46, table I I

The deteriorating reserve position of the central bank was not known at the time, for published reserve figures substantially overstated the true reserves of the central bank. In transactions involving the PNOC and the London branch of PNB, the central bank lent and reborrowed existing reserves, counting them twice in reserve statistics. The published reserve figures of the central bank and the later, audited figures are shown in table 6.12. The reserve overstatement varied from month to month, but averaged about $600 million. The existence of the overstatement was discovered in December 1983, after the declaration of a moratorium on principal repayments by the Philippines, and led to the resignation of the central bank governor and an audit of the central bank’s books. The central bank negotiated its last consortium loan in March 1983 for $300 million. The terms of this loan were more severe than the previous consortium loan in 1982, and a substantial portion of the loan was tied to the U.S prime rate instead of the London interbank offer rate for dollar deposits (LIBOR). At the same time in early 1983 the Philippines was borrowing frantically short term, almost at whatever spreads were quoted, despite seemingly substantial reserves, and many banks, sensing that something was wrong, cut back their short-term exposure to the country. The Aquino assassination on August 15 created a rush of anxiety about the country, and when the Philippines declared a moratorium in October, reserves were practically exhausted. The initial impression among commercial banks after the moratorium was that the rescheduling process for the Philippines would be the easiest to date and achieved very quickly. The announcement of the reserves overstatement

521

Philippinesichapter 6

Table 6.12

1980 December 1981 March June September December 1982 March June September December I983 March June September December 1984 March June September December 1985 March June September December 1986 March June September December 1987 March June September December

Central Bank Reserves (in millions of U.S. dollars) As Originally Stated

(Month’s Imports)”

3,155

4.7

2,865 2,597 2,549 2.574

After Audit

(Month’s Imports)”

3.3 3.0 3.0 3.0

2.574

3.0

2,499 2,585 2,449 2,543

2.7 2.8 2.6 2.7

2,139 2,105 2,101 1,711

2.3 2.3 2.3 1.8

2,433 2,282 1.431

2.5 2.3 1.5

1,368 1,134 682 865

I .4 1.2 0.7 0.9

869 632 47 1 886

0.9 0.7 0.5 0.9

588 1,037 1,436 1,061

0.7 1.3 1.8 1.3

1,244 1,602 1,710 2,459

1.8 2.3 2.5 3.6

2,484 2,360 2,112 1,959

3.7 3.5 3. I 2.9

Source: Central hank, Annual Report, and Philippine Financial Statistics.

‘Defined using total goods and services imports from the previous year.

shattered this optimism and poisoned and atmosphere between the Philippines and the banks and the Fund. External confidence in the Philippines had largely been confidence in the technocrats, and with the discovery of the reserves overstatement this trust vanished. The negotiation process was lengthened greatly by the audit of central bank accounts that the creditors required and by the verification that followed almost every Philippine presentation of information. l 3

522

Robert S. Dohner and Ponciano Intal, Jr.

6.4 The Debt Management System During the 1970s and 1980s the Philippines had in place one of the best debt management and information systems among LDC borrowers. Data collected by MEDIAD provided a reasonably complete view of external indebtedness, although it was somewhat weak on short-term borrowings, particularly those of the banking sector. The foreign loan approval system gave the government the tools to control the level of outstanding debt and to influence terms and maturities. The remaining area of debt management is what might be termed macro coordination, information on the demand for additional foreign capital implicit in domestic budget and investment decisions, and decisions about what amount of additional foreign finance was prudent for the country. A presidential directive in 1978 established the Investment Coordinating Committee (ICC), headed by the minister of finance and including the central bank governor, the minister of planning (NEDA), the minister of the budget, and the head of the BOI. Project proposals of over P. 300 million were examined by a subcommittee of the ICC, and a second subcommittee prepared a foreign resources budget for the country. l4 The debt management system of the Philippines was successful in the 1970s in limiting access to the external loans market and in lengthening maturities, reducing interest rates, and refinancing existing foreign loans. However, the system was not successful in preventing the sharp deterioration of the Philippine debt position in the early 1980s. In particular, it was unable to prevent a very rapid accumulation of short-term debt that raised the debt servicing burden and left the country vulnerable to shifts in domestic and external confidence. While the Philippines had all the elements of successful debt management in place, there were weaknesses in the operation of the system. The information gathering of MEDIAD concentrated on medium- and long-term debt. The monitoring of short-term debt was less complete, particularly for trade credits, and MEDIAD did not have responsibility for short-term debt of the banking sector. Short-term foreign loans to entities outside the banking sector required central bank approval, but approval was given to public sector corporations to borrow up to certain ceilings rather than approval on individual loans.15 In the early 1980s the central bank was unable to enforce those ceilings, leading to several directives and a presidential letter of instruction strengthening controls on short-term borrowing in 1983. While the institutions for macro coordination existed, they were not effective. The ICC failed to adequately screen and rank proposals to borrow; ICC approvals greatly exceeded ceilings on available foreign borrowing. In addition, the discretionary authority that existed under martial law undermined the review and control process. Presidential letters of instruction could supercede the review process and grant foreign borrowing approval,

523

Philippines/Chapter 6

and in the later years of martial law there was a greater tendency to preempt the existing review and regulatory mechanism, not just in foreign borrowing, but throughout the economic policy process. Nor was external conditionality effective in limiting the growth of external debt. Ceilings on external borrowing were a part of every IMF program that the Philippines undertook, beginning with the extended facility in 1976. In 1982, the one year in which the Philippines did not have an IMF program, the Philippine authorities set reduced limits on external borrowings, and in 1983 set limits that were lower than those contained in the IMF program. In each case the ceilings contained in the IMF program or set by the Philippine authorities were met. However, the ceilings applied only to loans of one- to twelve-year maturities (with subceilings for one to five years) and not to short-term loans. The second factor favoring short-term borrowing was the debt service ratio contained in Philippine statutes. It is difficult to determine to what extent the Philippine authorities or their external creditors were misled by the stability of the statutory debt service ratio, but there is no question that it influenced the behavior of Philippine policymakers as the ratio approached its ceiling in the early 1980s. What the statutory ratio did not include was short-term revolving credit, which was the overwhelming part of short-term nonmonetary debt. Revolving credit outstanding increased sharply during the 1980s, particularly to public sector enterprises. In addition, the ratio of revolving credit outstanding to trade flows being financed increased sharply, indicating that these credits were being used for general financial purposes and not simply for trade finance. What comes out of this review is that the character of the debt management and control process, coupled with the willingness of the Philippines to sidestep, but not violate, its limits, greatly weakened the debt position of the Philippines, making the country more vulnerable to external shocks and shifts of confidence. The Philippines in this period also illustrates a more widespread tendency not to make hard policy decisions until those decisions are effectively forced by events. Philippine policymakers considered declaring a moratorium in late 1982, but then balked after the announcement of the Mexican debt situation." Bankers from some of the major foreign banks in Manila approached the government early in 1983, advising that the Philippines declare a moratorium and ask for a rescheduling of its external debts, but the country instead decided to try to ride out its difficulties.I7 In addition to its international reserve, the central bank had a standby line of credit, arranged in the 1970s, of $450 million that it could have used, but never did. When the moratorium was finally declared on 17 October 1983, foreign exchange reserves were extremely low, and as a result the foreign exchange allocation process that followed was extremely severe.

524

7

Robert S. Dohner and Ponciano Intal, Jr.

Debt Crisis and Adjustment

The Philippines went through a particularly wrenching adjustment process in the three years following the declaration of a moratorium in October 1983. This adjustment episode is interesting for a number of reasons. The Philippines had waited until its foreign exchange reserves were almost exhausted before declaring a moratorium and had very little room for maneuver once foreign credit was cut off. Second, the Philippines resisted adjustment policies insisted upon by the IMF, taking over a year to reach a standby agreement and nineteen months to reach a rescheduling agreement with its creditor banks. Third, the program that the Philippines followed was strikingly successful in achieving adjustment, eliminating the current account deficit and domestic inflation by 1986. Finally, much of the discussion between the Philippines and the IMF were reported in the Philippine press, giving a unique insight into the negotiation over adjustment programs. The crisis and adjustment period may be divided into three phases. The first, from the declaration of the moratorium to mid-1984 is the initial reaction phase. During this period the Philippines was not under an IMF program, although negotiations with the Fund continued for a standby program for 1984 and the Fund did influence some of the policy measures that the Philippines adopted. The second phase of the adjustment process was characterized by a stringent IMF adjustment program. Although the program was signed in December, there were prior conditions that affected Philippine action in the last half of 1984. This second phase ran until the presidential election in February 1986. The last phase started with the negotiation of the nineteenth standby agreement between the Fund and the new Aquino government. The adjustment phases are determined by the presence or absence of IMF adjustment programs, and with good reason. During these years the relationship between the Fund and the Philippine government changed twice, and sharply in both instances. The Philippines had a relatively lenient IMF program in 1980-81, in which it exceeded targets on domestic credit and net international reserves by a wide margin. The relationship with the Fund began to change in 1982. The unilateral increase in the size of the Philippine industrial rescue fund led the IMF to reject a program for 1982. The seventeenth standby arrangement, which became effective in February 1983, had more restrictive limits on domestic credit and government contributions to public corporations, as well as a limit on short-term nonmonetary debt. However, the Philippines rapidly and substantially exceeded the ceilings on domestic credit and the target for the balance of payments, and the IMF terminated the agreement in the summer of 1983 after only two of the program’s four tranches had been drawn (IMF 1984a, 8, 12). Thus the Philippines declared its moratorium during one of the rare periods in which the country was not under an IMF standby arrangement.

525

Philippines/Chapter7

During the early crisis period the Philippines reversed course on policy reforms it had undertaken through IMF and World Bank programs in favor of policies that were familiar, could be implemented rapidly, and were addressed to the critical foreign exchange constraint.

7.1

Trade and Exchange Rate Policy

At the time that the Philippines suspended payments, foreign exchange reserves were extremely low, and the first task of policy was to try to manage international transactions with a minimal level of liquidity. The Philippines almost immediately instituted a foreign exchange control and allocation system. Banks were required to surrender their foreign exchange receipts to a pool at the central bank, from which the central bank made allocations for critical imports. The central bank established an allocation system that gave priority to crude oil imports, raw materials and supplies for export industries, grain imports, raw materials for certain vital domestic industries, and debt servicing for loans from multilateral organizations. The exchange policy allowed certain imports to take place outside the allocation system. Exporters could import materials on consignment as long as no external financing was required. Firms that were subsidiaries or joint ventures of foreign firms were allowed to pay for raw material imports with equity. In addition, the allocation system allowed imports of goods not on the priority list on a “no dollar” basis, if the importer could come up with the necessary exchange independently. This in fact created a dual exchange rate system, where importers receiving allocations purchased at the official rate, while others financed their imports through the black market at higher rates (table 7.1). It also weakened the exchange surrender system, and by mid-1984 the central bank estimated that it only received about 60 percent of the country’s foreign exchange receipts. The Philippines had a dual exchange rate system in a more organized and unusual sense. Faced with a black market exchange rate of almost P. 30 per dollar (versus an official rate of P. 14), the Marcos government intervened to oligopolistically organize the foreign exchange black market in the Philippines. In the words of the Minister of Trade and Industry, Roberto Ongpin, who was responsible for the operation: the largest and best known Chinese black marketeers were rounded up by the National Intelligence and Security Authority and were told that it was the government’s intention to organize a currency stabilization program. The Chinese black marketeers would be permitted to keep their profits from their operations provided that they cooperated with the government by buying and selling foreign exchange at government-dictated rates. . . . In order to maintain discipline, the participants were subject to arrest if they deviated from the established rates. (Ongpin 1986)

526

Table 7.1

Robert S. Dohner and Ponciano Intal, Jr. Official and Black Market Exchange Rates (pesos per dollar and ratio) Official Exchange Rate

Black Market Premium"

19x3 January February March April May June July August September October November December

9.287 9.464 9.606 9.869 10.032 10.355 11.002 11.002 11.002 13.702 14.002 14.002

1.001 1.027 I .074 I .046 I .027 I .070 1.009 1.095 1.168 1.111 1.345 1.350

1984 January February March April May June July August September October November December

14.002 14.002 14.002 14.002 14.002 17.402 18.002 18.002 18.002 19.148 19.959 19.859

1.484 1.263 1.094 1.277 1.370 1.195 1.092 1.082 1.077 0.999 0.967 0.984

1985 January February March April May June July August September October November December

18.979 18.256 18.478 18.484 18.480 18.473 18.581 18.605 18.616 18.704 18.737 18.896

0.937 0.931 0.951 0.982 0.982 0.983 0.977 0.976 0.975 0.970 0.969 0.961

1986 January February March April May June July August September October November December 1987 January

February March April May June July August September October November December

Official Exchange Rate

Black Market Premium"

19.042 20.461 20.781 20.505 20.500 20.552 20.454 20.432 20.509 20.437 20.436 20.491

0.953 0.945 1.047 1.073 1.073 I .039 1.015 0.997 0.995 0.999 0.997 0.995

20.504 20.525 20.563 20.505 20.473 20.456 20.450 20.439 20.601 20.706 20.817 20.815

0.995 0.994 0.992 0.982 0.983 0.978 0.978 0.984 0.985 0.984 0.998 0.967

-

aHong Kong banknote rate divided by the official exchange rate.

This operation was termed the area in Manila where most of the and existed from early 1984 until were set by Ongpin, and officers

"Binondo Central Bank," named for the black market currency trading was done, February 1986. Exchange rate guidelines of the National Intelligence and Security

PhilippineKhapter 7

527

Authority accompanied shipments of currency to Hong Kong. The organization of the black market helped the Philippines narrow the black market premium on the exchange rate in early 1984 and also helped meet IMF conditions later in the year. To support the exchange rate allocation system, the government also halted the program of import liberalization that had been adopted as a part of the World Bank's structural adjustment loans, and instead imposed additional import restrictions. Taxes on international trade were increased, both to discourage imports and to raise government revenue. An additional 3 percent surtax had been levied on all imports in January 1983. The government raised this surtax to 5 percent in November, later to 8 percent, and then to 10 percent in June 1984. The Philippines also raised taxes on exports of primary commodities to levels that ranged from 6 to 10 percent.' The Philippines had pursued a passive exchange rate policy during the 1980s. The real exchange rate of the peso remained almost constant with respect to the U.S. dollar, as gradual depreciation offset a higher domestic inflation rate. But the real effective exchange rate moved with the U.S. currency and followed the dollar up in the early 1980s, so that by 1982 it had risen 15 percent from its 1978 level. There were two devaluations in 1983, one in June from P. 10.2 to P. 11 per dollar, and a second in October to P. 14. This last devaluation resulted in a real effective depreciation of about 21 percent from the December 1982 level. The effective exchange rate change for traditional exports was somewhat less because of the increase in export taxes. For import commodities not on the priority list the real depreciation was huge; not only were there surcharges on imported commodities, but currency had to be sourced through the black market (table 7.2 and fig. 7.1). Real Effective Exchange Rates

Table 7.2

Official exchange rate Philippine CPI Partners WPI ($) Black market premium Import tariffs Traditional export taxes' Nontraditional export taxes Traditional exports Nontraditional exports Importables ~

Dec 82

June 83

Dec 83

June 84

Dec 84

June 85

Dec 85

June 86

9.171 177.6 124.39

11.002 184.4 124.80

14.002 223.9 124.76

18.002 275.2 124.71

19.760 337.7 123.90

18.465 351.3 123.08

19.032 356.8 130.46

20.580 351.4 137.84

1.035 1.31

1.020 1.32

1.350 1.34

1.195 1.39

0.984 I .39

0.983 1.34

0.961 1.34

1.026 1.34

0.960

0.960

0.943

0.879

0.943

0.943

0.943

0.943

0.997

0.997

0.996

0.996

0.996

0.996

0.996

0.996

100.0

115.9

119.3

116.4

110.9

98.9

106.4

123.5

100.0 100.0

115.9 115.1

121.3 162.1

126.9 155.6

112.8 113.9

100.6 97.8

108.2 102.9

125.6 127.4

~~~

Source Lamberte et al (1985). table IV 3, updated by the authors

'Traditional export taxes include 3(18- 14)/18 for economic stabilization tax

528

Robert S. Dohner and Ponciano Intal, Jr.

160

140

1 20

100

80 12.82

6.83

12.83

6.84

I 12.84

1

6.85

I 12.85

I

6.86

Fig. 7.1 Real effective exchange rates (December 1982 = 100) Note: M = importables; Xnt = nontraditional exports; and Xtr = traditional exports.

7.2 Government Budget In 1981 and 1982 the consolidated public sector budget deficit reached 5.5 percent of GNP, with the deficit of the national government over 4 percent of GNP. This marked the peak of Philippine budget deficits, and the following year saw a significant narrowing of the deficit of the public sector. Revenues were increased by the import surcharge, the additional taxes on exports, and increased excise taxes on petroleum and other products. The result was a rise in the revenue share of GNP to 12.0 percent in 1983 from 11.4 percent in 1982. National government expenditures fell by 8 percent in real terms during 1983. Nonwage, noninterest expenditures (primarily operations and maintenance expenditures) were reduced by 13 percent in real terms, and the capital expenditure of the national government was trimmed. But the biggest decline came in equity transfers and net lending to government corporations, which fell from 3.5 percent of GNP in 1982 to 2.1 percent in 1983. As a result of these expenditure cuts and the increased revenue effort, the deficit of the national government fell to 2 percent of GNP for 1983.

529

Philippines/Chapter7

The deficits of the public corporations were more difficult to control. Despite the deferral of five of the Major Industrial Projects during the year, investment expenditures by public corporations stayed at 5 percent of GNP, even as the national government was reducing its contributions to public enterprises. Part of the resulting financing requirement was met through internal funds. Cash generation of the public corporations rose during the year because of higher charges on public services, but also because the National Power Corporation ran up external arrears toward the end of the year. However, the borrowing requirement of the public corporate sector remained at 2.1 percent of GNP. Almost the entire deficit was financed through foreign borrowing.2 Only at the end of the year did the Finance Ministry exert more direct control over the operations of the nonfinancial corporate sector. Tariffs and taxes were raised on water, power, petroleum products, and transportation, and the investment programs of the corporations were substantially reduced. In addition, the government imposed more effective ceilings on foreign borrowing by corporations, including short-term borrowing.

7.3 Monetary Policy Monetary policy also became more restrictive at the end of 1982, and this continued through the first half of 1983. Reserve money in June 1983 was only 2 percent higher than it had been in June 1982.3 During 1983 the central bank took further actions to limit the growth of reserve money and to reduce the money multiplier. The Philippines completed its financial liberalization in January 1983 when the last limits on short-term interest rates were removed. At the same time, the central bank announced limits on drawings from the subsidized rediscount windows and for the first time tied the rates of those rediscounts to market interest rates.4 The central bank raised reserve requirements from 18 to 23 percent during the last half of the year, reversing the decline in reserve requirements that had taken place under the financial liberalization program. Despite these efforts, there was massive growth in the monetary base in the last half of 1983. From June to December of that year reserve money increased by 72 percent to a level of P. 28 billion (table 7.3). The increase in reserve requirements and greater public holding of currency restrained the growth of the money supply, but MI still increased by 40 percent over the course of 1983. As a result of this explosion in money, the central bank largely accommodated the devaluations that took place during 1983. In December market interest rates were only two percentage points over their June levels and inflation was running at 50 percent. Four factors accounted for the rise in reserve money over the latter part of 1983. The first was an increase in central bank credit to the national

Table 7.3

Factors Affecting Reserve Money' (in billions of pesos) 1982

Reserve money Net foreign assets Net domestic assets Credit to public sector National government Assistance to financial institutions Regular rediscounting CB securities, RPs, RRPs Other items net Swap differential Forward cover differential

1983

1984

Dec

Mar

Jun

Sept

18.6 -35.5 54.1 8.0 9.9 3.1 15.9 -4.6 31.7 1.2 3.8

17.2 -37.4 54.6

16.1 -45.1 61.2 8.3 9.0 3.0 13.1 -4.7 41.5 2.6 6.1

17.6 -52.9 70.5 9.6 9.5 2.7 12.1 -4.4 50.5 3.5 7.8

8.0 9.7 3.5 15.8 -4.7 32.0 1.7 4.7

Dec 27.7

-44.1 72.4 12.5 12.0 5.0 12.2 -4.4 47.1 8.0 8.8

1985

Mar

Jun

Sept

24.8 -45.0 69.8 13.9 14.2 3.8 11.1 - 10.9 51.9 9.8 9.4

27.1 -49.2 76.3 15.7 16.1 6.7 10.6 -9.9 53.2 11.9 10.4

28.7 -51.8 80.5 13.6 14.0 13.4 9.6 -11.9 55.8 18.2 12.4

Dec 34.2 - 47 .O

81.2 11.0 11.4 10.9 8.4 - 13.9 64.8 22.0 14.1

1986

Jun

Dec

Mar

Jun

Dec

31.6 -45.4 77.0 14.7 13.6 12.5 7.3 -37.3 79.8 15.3 17.9

38.0 -80.1 118.1 20.9 16.3 13.7 8.3 -32.2 107.4 14.2 21.7

40.4 -81.1 121.5 28.1 23.6 19.1

38.0 -78.8 116.8 17.6 12.8 13.9 8.5 -46.9 123.7 13.7 22.1

50.0 -86.9 136.9 12.9 8.7 13.7 6.9 - 29.9 133.3 13.2 22.4

9.0 -49.5 114.8 14.2 21.8

Source: IMF (1984b; 1985c; 1986b; 1988b). Note: RPs are repurchase agreements and RRPs are reverse repurchase agreements.

"Monetary authorities basisincludes Treasury IMF accounts. Foreign assets and liabilities converted into pesos at constant exchange rate of P. 18.002 per U.S. dollar

531

PhilippineKhapter 7

government, which rose by 3 billion pesos (45 percent of the government’s annual deficit). The other sources of monetary expansion were less conventional. The largest increase in reserve money came from the losses incurred on outstanding swap and forward cover contracts when the exchange rate depreciated. These losses added P. 8 billion to domestic assets, equivalent to two-thirds of the increase in reserve money, although a substantial portion of these losses was blocked to their recipients by the central bank. Additional growth in reserve money came from the accumulation of external payments arrears. The buildup of arrears blunted the contractionary effect of reserves outflow, and this was not offset by the collection of peso deposits on amounts in arrears. The external crisis weakened confidence in the already fragile financial system of the Philippines, particularly in nonbank financial institutions. The central bank expanded emergency loans to financial institutions by 2 billion pesos during the last half of 1983, most of it to thrift banks, which added to the growth of reserve money during the last part of the year.

7.4 Domestic Inflation and Price Intervention Policy The domestic inflation rate accelerated dramatically during the last half of 1983, fueled by devaluation, the institution of import controls, money supply growth, and consumer hoarding of commodities such as rice and cooking oil. By the beginning of 1984 consumer prices were rising at a rate of 60 percent per year, and this rate persisted throughout 1984 (fig. 7.2). The Philippine government had set domestic price ceilings for rice, sugar, and a number of other consumer commodities since the beginning of the martial law p e r i ~ d In . ~ addition, the government regulated prices for transportation and domestic sales of petroleum products.6 These prices were adjusted at frequent intervals after June 1983 to reflect higher import and domestic costs, but the discrete nature of the price ceiling adjustments led to anticipatory buying and shortages of consumer items during 1983 and 1984, and an acceleration of the consumer price index at the time each adjustment was made. Philippine actions in the initial crisis period increased the difficulty that the country had in adjusting to the cutoff of foreign funds. The very low level of foreign exchange reserves available to the country dominated initial policy, leading to exchange and tax policies that discriminated against exports. The government had difficulty reducing the deficit due to its ineffective control over public corporations. But the loss of control over the money supply in the last half of 1983 was the most damaging. The effect of the 1983 devaluation was lost through price level increases, and by mid-1984 inflation was proceeding at Latin American rates.

532

7.5

Robert S. Dohner and Ponciano Intal, Jr.

An IMF Standby Agreement

7.5.1 Negotiations Negotiations between the Philippines and the International Monetary Fund for a new standby to replace the cancelled 1983 agreement were disrupted by the Aquino assassination in August, but resumed in September. Events moved rapidly in October. A devaluation was one of the conditions of the Fund which the Philippines carried out on October 5. On October 17 the Philippines declared a moratorium on principal repayments on its external debt, and a ten-bank advisory group was formed, headed by Manufacturer's Hanover. On November 10 the Philippines submitted a letter of intent to the IMF. The program contained in the letter of intent was approved by the Fund's managing director and sent to the Fund's executive board for approval. Unable to rush the approval process forward, the Philippines applied to the United States for a bridge loan until disbursements from the IMF program came through, but the U.S. demurred, insisting that it too would wait for the executive board's approval.

533

PhilippineKhapter 7

The IMF program and the rescheduling negotiations were derailed by the discovery in early December of the Philippine reserves overstatement. The banks broke off negotiations on rescheduling, and the IMF immediately sent a team to Manila to examine the books of the central bank, shelving the agreement that had been reached. While the practice of padding reserve figures was not unusual among LDCs, it had a devastating effect on the relations between the Philippines and both the IMF and the country’s creditor banks. The amounts involved were large; the overstatement had run as high as $1.1 billion, or SO percent of stated international reserves. But the effect was far greater, because it cast doubt on the integrity and honesty of the Philippine technocrats, a group that the IMF had considered allies in promoting policy reforms in the Philippines. Jaime Laya, the central bank governor, was most directly affected, but the reputations of all of the technocrats were tarnished, with the possible exception of Cesar Virata, the finance minister. Negotiations became much more testy, and nearly every factual submission by Philippine negotiators was subject to question. The IMF insisted on an outside audit of the books of the central bank which was performed by Sycip, Gorres, and Velayo, a large, local accounting firm. The IMF and the World Bank sent teams during 1984 to examine Philippine economic statistics. The negotiating banks informed Finance Minister Virata that they would rather not see Governor Laya at the table when negotiations resumed. Laya was appointed minister of education, culture, and sports in January, and Jose Fernandez, head of the Far East Bank, was appointed governor of the central bank. The negotiations on a new standby agreement were prolonged and difficult, and it was not until December 1984 that a new agreement with the IMF was signed. During that time the Philippines had to extend the original ninety-day moratorium several times since agreement with the banks hinged on an agreement with the Fund. Reaching an agreement with the IMF was a critical matter. The Philippines had to deal with an extreme shortage of foreign exchange after the moratorium was declared. Reserves were low, and the centralization of foreign exchange receipts that the central bank had tried to establish was not very successful. The dollar pool that the central bank created had a balance of only $800 million in the first week of January, of which $455 million was earmarked for oil imports. Even importers of goods on the priority list were having trouble getting allocations from the central bank. As the negotiations with the Fund wore on, the Philippines finally arranged additional financing from other sources: the U.S. Export Import Bank and the Commodity Credit Corporation provided additional funds early in 1984, as did the Asian Development Bank. In all, the Philippines was able to raise $370 million from these agencies. This time the IMF took a much tougher stand than it had in previous standby consultations. Negotiations took place not only on the content of the

534

Robert S. Dohner and Ponciano Intal, Jr.

program, but also on the prior actions that the Philippines would have to take before the IMF would agree to a program. What the Fund particularly wanted was a reduction in the money supply after the huge increase of the last part of 1983. At this time the inflation rate was already running at 60 percent per year, and the Fund worried that the Philippines would develop a high inflation-devaluation cycle similar to that of many Latin American countries. The IMF originally set as a target a reduction of reserve money from P. 28 billion at the end of December 1983 to P. 23 billion by the end of March.’ The central bank had some success in reducing liquidity through open market operations in the first quarter of 1984, although some of the reduction was seasonal. Outstanding central bank securities and reverse repurchase agreements increased by P. 6.5 billion during the quarter (see table 7.3). However, parliamentary elections were held on May 14, and advances to the government, largely used to finance the election, increased sharply in the second quarter, reversing the gains the central bank had previously made. The central bank also increased its emergency loans to financial institutions during the quarter. The fragility of the financial system was an obstacle to liquidity reduction during much of the year. The system had been battered by the Dewey Dee crisis in 1981 and by the prolonged domestic recession. When reserve requirements were raised by five percentage points in the last half of 1983, reserve deficiencies rose to over 20 percent of required reserves. Half of the reserve deficiencies were those of the government-owned PNB. These were erased early in 1984 when the central bank allowed PNB to exchange uncollectible promissory notes issued by sugar mills for bonds issued by the newly created Philippine Sugar Corporation, and then made those bonds eligible as reserves. The central bank had made large emergency loans to thrift banks and investment houses during the last part of 1983. In 1984 emergency loans and overdrafts to commercial banks and DBP increased sharply (table 7.4). In July, a run on the Banco Filipino, the country’s largest savings bank, forced the bank to close its doors.’ At this point the central bank had extended almost one billion pesos to Banco Filipino. The central bank at first refused to supply additional emergency assistance but, after a presidential directive, extended an additional P. 3 billion to the savings bank. In negotiations with the IMF, Virata and Fernandez argued that further reductions in the money supply would create more extreme difficulties in the financial system. There were other issues that were important in the negotiations on prior action for a standby program. The Fund wanted a further devaluation of the peso to restore the erosion of real exchange rates that had taken place since the previous devaluation, and a floating of the exchange rate to allow the resumption of bank foreign exchange trading and the narrowing of the black market premium. The Fund also wanted further cuts in the investment

535

PhilippineKhapter 7

Table 7.4

Central Bank Emergency Loans to Financial Institutions (in millions of pesos) 1982

1983

Emergency loans Commercial banks Thrift banks Financecompanies Investment houses Rural banks Special government banks Overdrafts Commercial banks Thrift banks Finance companies Investment houses Total

1984

Dee

Jun

Dec

Jun

3,137 3,074 25 32

3,051 2,989 25 28

-

-

3.787 2,753 226 205 396 7 200 1,170 487 252 29 402 4,957

-

6 2 2

-

9

2 2 -

-

-

-

-

3,139

3,053

1985

I986

Dec

Mar

Dec

Mar

Sept

3,432 1,397 234 283 499 7

3,732 1,565 358 288 497 12

3,950 1,839 285 306 497 11

3,996 1,942 248 302 486 6

3.961 1,919 245 297 482 6

3,874 1,871 236 306 443 6

1,012 3,219 2,524

1.012 7,189 3,444 3,530 9 206 10,921

1,012 8,255 4,416 3,625 8 206 12,205

1,012 9.752 6,009 3,530 8 205 13,748

1,012

1,012 9,960 6.247 3,500 8 205 13,834

480

9 206 6,651

15,181

11,443 3,525 8 205 19,142

____

Source IMF (1985~).table 47, and the central bank, Quarterly Economic and Financia[ Report. September 1986.

programs of the state-owned corporations, an increase in taxes, and a shifting of tax burdens away from international trade and toward domestic transactions. On June 5 the central bank devalued the peso to a level of P. 18 per dollar, and at the same time instituted two additional trade taxes. The first was a 10 percent tax on foreign exchange for nonmerchandise transactions. The second was a windfall tax on traditional exports, called the economic stabilization tax.’ The effect of the two taxes was to establish a multiple exchange rate system. Nontraditional exports and merchandise imports for which exchange was allocated took place at P. 18 per dollar, nonmerchandise imports at P. 19.80 per dollar, and traditional exports at an effective rate of P. 16.80. These additional measures became further sources of disagreement between the IMF and the Philippines, and their removal became a prior condition of the agreement. In September and October 1984 the Philippines took a number of actions that paved the way for an eventual agreement with the Fund. The economic stabilization tax on exports, imposed along with the June devaluation, was removed on September 22. In September the central bank also allowed commercial banks to retain half of their foreign exchange receipts, and the exchange allocation system was abolished in October. The import surcharge was reduced to 5 percent, and the 10 percent excise tax on foreign exchange for nonmerchandise imports was replaced by a uniform 1 percent tax on all foreign exchange transactions. Finally, on October 15 the exchange rate was allowed to float. In August and October the Philippine government increased taxes on petroleum products and crude oil, coke and coal, and other products subject

536

Robert S. Dohner and Ponciano Intal, Jr.

to excise taxes. Interest exemptions were removed and, in a major victory for the Fund, the Philippine government agreed to remove tax and tariff exemptions from government-owned corporations and to reduce the tax exemptions given to manufacturing firms under BOI incentives. The estimated impact of these measures on the 1984 budget was P. 2.6 billion, or 0.5 percent of GNP. In March the Monetary Board gave the central bank permission to sell short-term, non-reserve-eligible notes. The central bank aggressively marketed these bills, called “Jobo bills” after the nickname of the central bank governor, starting in mid-August, at rates that exceeded 40 percent. Job0 bills were purchased by banks as a safe alternative to domestic loans and by the nonbank public. Through the sale of these securities, the central bank succeeded in reducing reserve money by approximately 12 percent between early August and the end of October, clearing the way for an agreement with the Fund. A letter of intent was submitted October 3 1, and the standby arrangement was approved by the executive board of the IMF on 14 December 1984. 7.5.2 The Adjustment Program The standby arrangement worked out with the IMF was exceptionally stringent. The agreement was for $650 million, spread over eighteen months and divided into seven tranches. Program reviews were set for each quarter, and some of the conditions covering the remainder of the program were to be set at later reviews. The IMF agreement cleared the way for rescheduling agreements with the commercial banks and with official creditors, but the program itself brought little additional funding. The initial drawings under the agreement were to be used to clear up arrears that the Philippines had accumulated. The program and the assumptions on which it was based are outlined in tables 7.5 and 7.6. The most immediate indication of the severity of the adjustment program is given by the GNP assumptions. The expected drop in GNP for 1984 was 6 percent, with zero growth assumed in 1985, implying a fall in per capita income of 11 percent over two years. In fact, these turned out to be too optimistic; real output fell in both years, leading to a 15 percent reduction in per capita income from 1983 to 1985. The monetary targets proved to be the most difficult to meet. The targeted measure was reserve money, currency plus bank reserves. lo The reserve money figure had already reached a peak of P. 34 billion by early August 1984, and the December 1984 ceiling required a substantial reduction. The growth of reserve money was to be limited to 23 percent over the course of 1985. The monetary targets also included substantial reductions in banking system credit to the public sector, ruling out monetary financing of the government deficit during 1985.

Table 7.5

Philippines 1985-86 Standby Program Rrforrnance Criteria 31 December 1984

Ceiling Money and credit (billion pesos) Reserve money Net banking system credit to national government Bank credit to public sector Central hank credit to PNB External payments (million US$) Net international reserves (Boor) External payments arrears Short-term external debt New borrowing approvals 1 - 12 years maturity 1-5 years maturity Public sector (billion pesos) Combined deficit, 13 monitored public corporations

Actual

31 March 1985

Ceiling

31 December 1985

Actual

Ceiling

Actual

32.0

34.2

31.0

30.5

39.5

38.0

17.9 21.4 5.0

17.0 25.6 5.0

17.9 27.1 5. I

19.0 29.0 5. I

13.0 20.7 5.3

12.4 19.2 5.0

-4,719 2,144 9,649

-3,886 2,690 9,998

1,800 300

512 171

13.0

Source: IMF (1984a; 1985a; 1986a) and Montes (1987).

11.6

--4,351

3,161 9,649

-4,150 2,642 9,757

0.9

Ceiling

38.9

Actual

40.4

10.0 5.3

14.2 8.2

- 1,910

-- 1,841

0 9.649

0 9.458

0 9.649

9.088

2,255

1,232 250

2,255 400

1,246 25 1

400

2.4

31 March 1986

-1,615

-. 1,512

0

538

Table 7.6

Robert S. Dohner and Ponciano Intal, Jr.

Philippines 1985-86 Standby Program Assumptions 1983

External objectives Current account deficit (US$ billion) (percent of GNP) Exports (percent change $ value) Imports (percent change $ value) Trade balance (US$ billion) Savings and investment Gross domestic investment Total savings Gross national savings Foreign savings Money (percent increase, end period) M3 Reserve money GNP and prices (percent increase) Real GNP CPI (De-Dec) CPI (average) Source:

1984

1985

1986

Actual

Projected

Actual

Projected

Actual

Projected

Actual

2.8 8.1 -0.3 -2.3 -2.5

1.5 5.2 5.9 -22.7 -0.5

1.1 3.5 7.7 - 18.9 -0.7

1.1

4.1 10.0 - 1.6 0. I

0.1 0.2 - 14.1 - 15.8 -0.5

0.5 2.3 11.0 3.4 0.6

1.0 -3.5 4.6 - 1.3 -0.2

27.5 27.5 19.4 8. I

22.0 22.0 16.8 5.2

19.2 19.2 15.7 3.5

22.5 22.5 18.4 4. I

16.3 16.3 16.1 0.2

23.0 23.0 20.7 2.3

15.4 15.4 18.9 -3.5

I9 49

15 10

7.3 20.6

13 II

9.6 13.6

12

12.8 31.6

0 10-15 20-25

-3.8 5.7 24.9

I 8- 10

0.9 26. I 10

-6 4-45 45-50

-6.8 50.8 50.4

10

10

-

2.0 -0.3 0.8

IMF (1984a; 1985a; 1986a) and Montes (1987).

Using the IMF assumptions on the course of domestic prices and real GNP, the program limits implied a reduction in real reserve money outstanding by the end of 1985 of 7 to 14 percent of their end of 1982 levels, which was before the jump in money supply took place. The ratio of reserve money to nominal GNP was assumed to rise only 7 to 15 percent from its end of 1982 level and be well below its end of 1983 peak. Broad money (M3) was projected to grow by 10 percent in 1984 and 13 percent in 1985. This implied a 20 to 30 percent rise in velocity from the end of 1982, again, before the huge monetary expansion took place. Both the IMF and the Philippines recognized that the government financial institutions (PNB, DBP, and PhilGuarantee) would run substantial deficits in both 1984 and 1985, even after rescheduling. There was no real scope for internal actions at DBP or PhilGuarantee, but some possibilities for reduced losses at PNB. One of the performance criteria under the program limited central bank credit to PNB to a near constant amount over the life of the program. The program required the elimination of all external arrears by the end of 1985, and an increase in net international reserves of the central bank of $2.8 billion over the course of the year. The program also set limits on new medium- and long-term external borrowing and ruled out an increase in

539

PhilippinesXhapter 7

short-term debt for the entire program. The growth in exports and reduction in imports was assumed to be sufficient to generate a trade surplus by 1985. The program called for a rapid reduction of the public sector deficit, from 3.2 percent of GNP in 1983 (already a substantial reduction from 1982) to 1.4 percent during 1985. This was after the inclusion of government support to public financial institutions, whose shortfalls were expected to rise to 2 percent of GNP in both 1984 and 1985 (IMF 1984c, 31). The Philippine letter of intent outlined a program of tax increases for 1985 that was estimated would yield revenues of 2.1 percent of GNP (IMF 1984c, 63)." Continued sharp reductions in expenditure were also to take place. The Philippines agreed to cut the investment program of public corporations in half from 1983 to 1985. For the first time the standby program covered the deficits of public corporations. The combined deficit of the thirteen largest public nonfinancial corporations was to be limited to P. 13 billion (about 2.5 percent of GNP) for 1984 and to P. 2.4 billion for the first quarter of 1985 (1.5 percent of GNP), with further quarterly limits to be set at the time of the first program review. The Philippine government raised price ceilings on controlled commodities and increased administered prices for petroleum products and transport several times after the crisis. By the end of 1984 retail prices for petroleum products were more than double their September 1983 levels and electric power tariffs had been raised by 70 percent. During the negotiations with the IMF, the Philippines agreed to eliminate price ceilings for basic consumer commodities. By the end of 1984, this had been done for all commodities except rice, and rice followed during 1985. The Philippine government raised minimum wages for public and private employees on several occasions during 1983 and 1984, although the increases did not match the rate of inflation. In addition, there were widespread complaints that firms were not abiding by minimum wage floors. Data on actual wages paid are scanty for the Philippines, but suggest a fall in real wages for unskilled manufacturing workers of about 5 percent during 1984. Actual compensation per employee in the national government was only 16.5 percent higher in 1984 than in 1983, despite a 50 percent rise in consumer prices. As part of the IMF program, the Philippines agreed to target wage increases for public sector workers at 22 percent during 1985, about the same as the estimated increase in prices. The IMF program was unique in containing structural reform measures, measures that in some cases challenged the political base of the martial law regime. These fell in the areas of tax reform, the rehabilitation of financial institutions, and the dismantling of public and quasi-public monopolies in the agricultural sector. On tax matters, the program required a shift in tax orientation away from international trade and toward domestic transactions. This entailed the reduction or elimination of the variety of trade and foreign exchange taxes and surcharges that the Philippines had imposed since

540

Robert S. Dohner and Ponciano Intal, Jr.

October 1983 and the raising of taxes on domestic sales of a number of commodities. The Fund also obtained agreement to reduce the wide range of tax exemptions available to manufacturing firms through BOI incentives. The Philippines agreed to eliminate tax and duty exemptions for public corporations, although the requirement left an escape clause for the Philippine government in which they could review and possibly reinstate the exemptions. By the time that the standby agreement was signed, the IMF had become more aware of the instability of the domestic financial system. In the agreement, it endorsed a central bank plan to strengthen the financial system by encouraging mergers. The agreement limited the credit of the central bank to PNB explicitly, and limited credit to financial institutions generally through strict limits on the expansion of reserve money. The Philippine government also agreed to prepare rehabilitation plans for governmentowned financial institutions. The most politically controversial part of the standby program was the requirement to disband government marketing and price restrictions in the agricultural sector. The monopoly on importation and marketing of grains by the National Food Authority (NFA) was the most important of the purely public controls on agricultural transactions. But the production levies, export restrictions, and other marketing controls on the sugar and coconut industries were far more important in the Philippines. Although controls in these industries had been given the imprimatur of public policy, the institutions and the funds generated by the controls were in private hands, those of two of the most important cronies of the martial law regime. Each control scheme acted as a substantial tax on production in the two most important traditional export sectors. The IMF program, in tandem with a World Bank loan for the agricultural sector, required that the Philippine government dismantle these controls and allow producers to trade at market prices through proposals to be worked out during subsequent program reviews. The Marcos government generally resisted the structural measures affecting the most important monopolies, dissolving institutions and announcing, but not effectively implementing, policy reforms. The monopoly on flour distribution was removed from the NFA, but it retained a monopoly on wheat importation, with the pledge to gradually open it up to the private sector. The monopoly on exports of coconut products given to UNICOM and three other mills was abolished, but the ban on copra exports introduced in 1984 was maintained, as were the producer levies. In the sugar industry, Philsucom was reconstituted with a larger representation of planters, and NASUTRA was replaced by a new body owned by planters and millers. Reform of the sugar industry was made more difficult by the extremely low world sugar price and the large, unpaid NASUTRA obligations to planters. Regulation of the domestic sugar industry was tightened, and producer support through higher domestic prices was reintroduced.

541

PhilippinesKhapter 7

7.6 Rescheduling Agreements with Banks and Official Creditors 7.6.1

Negotiations

When the Philippines declared a ninety-day moratorium on 17 October 1983, the initial expectation among foreign bankers was that the rescheduling exercise for the Philippines would be quickly accomplished. Instead the negotiations for the Philippines proved to be the most prolonged, and in some ways the most difficult, of any of the LDC borrowers. Nineteen months would pass before the Philippines finally signed a rescheduling agreement with its commercial creditors, and the moratorium was extended an additional six times. The reasons for the delay were varied. In part they reflected the inability of the Philippines to meet IMF prior conditions for the establishment of the standby program on which the financing agreement hinged. But Philippine negotiations with the banks were also held up by disagreements among the creditor banks on issues over which the Philippines had little control. Agreements with the IMF and with the bank advisory committee were nearly concluded by the end of 1983, before the discovery that the central bank had been overstating its foreign exchange reserves prior to the declaration of the standstill. l 2 The Bank Advisory Committee had agreed to a proposal to reschedule commercial bank debt falling due before 30 June 1985 and to extend $1.6 billion in new money to the Philippines, when the reserves overstatement was discovered. This disclosure did more than push the negotiations back to their starting point. The trust that external creditors had in the Philippine technocrats vanished, and the emphasis in both bank and Fund negotiations shifted to prior conditions, frequent review, and “backloading” -pushing the largest funds releases to the end of a program, after intermediate conditions had been met. The reserves disclosure and the subsequent discovery of the extraordinary increase in the money supply in the latter part of 1983 also increased the importance of the IMF negotiations for a revised standby agreement, and for much of 1984 the commercial bank negotiations receded to the background. Negotiations with the bank advisory committee resumed in March 1984, only to be postponed until June, pending the completion of an IMF standby agreement. By the end of October a new Philippine letter of intent was drafted and sent to the IMF for approval. With the outlines of an IMF program in place, negotiations with the banks resumed in November. The original financial package, negotiated in the latter part of 1983, called for a rescheduling of $9.1 billion in commercial and official loans maturing between the declaration of the moratorium and the end of 1984. In addition, there was a new money component of $3.3 billion, evenly divided between official and commercial bank creditors, as well as the IMF standby arrangement of $650 million. The $1.6 billion new money proposal for commercial banks represented about 10 percent of their outstanding exposure

542

Robert S. Dohner and Ponciano Intal, Jr.

to the Philippines, a relatively large new commitment. With the delay in reaching a new agreement with the IMF in 1984 came increasing doubts among commercial banks of the willingness of the Marcos government to carry through adjustment measures, and an increasing reluctance to extend money to the Philippines. By the summer of 1984 the financial programming included all of 1985, with the proposal of $1.6 billion from the commercial banks extending over two years, and now including $2.7 billion in additional money from bilateral and official creditors. Once the Philippines reached an agreement with the IMF and the letter of intent was sent, negotiations with the official creditors proceeded quickly. These were concluded on 20 December 1984, and the outcome is described in table 7.7. The restructuring covered 100 percent of principal and 60 percent of interest falling due between 1 January 1985 and June 1986. The remaining 40 percent of interest became payable on a staggered basis, ending September 1987. Due but unpaid principal and interest on obligations up to the end of 1984 (about $200 million) became payable over eighteen months. In addition to restructuring existing obligations, the agreement included long-term new money from official sources of $2.2 billion, slightly below the figure discussed in midsummer and early autumn. However, the creditor bank negotiations continued to be difficult and extended on into 1985.

Complications in the Negotiations

7.6.2

The primary reason for the delay in negotiating an agreement with the commercial banks was the difficulty that the Philippines had in reaching an agreement with the IMF. However, there were three additional issues that complicated the negotiation process with the banks, only one of which directly involved the Philippines.

Table 7.7

Philippine Debt Restructuring Agreement, May 1985

Commercial bank rescheduling

($ billion)

Amount

Debt falling due New money Trade facility Total

5.88 0.925 2.97 9.69

Maturity (grace)

20 December 1984 Source: Central bank.

$ billion

0.725

IH 1Y' 4 1%

10 (5 yrsj 9 (4 YES)

Principal Amount Official Creditors

Interest Spread

Interest Amount I 100

$ billion

B

0.258

60

543

PhilippineslChapter 7

The first issue arose immediately after the Philippines declared a moratorium on 17 October 1983. The Manila branch of Citibank had accepted approximately $600 million in interbank deposits in its foreign currency deposit unit. When the Philippines declared a moratorium these deposits were blocked, and Citibank refused to make good the deposits from its home office. Several major creditors were affected by the Citibank action, including Wells Fargo of San Francisco, which filed suit against Citibank in New York to recover $2 million in deposits in the Citibank unit. Two independent issues were involved in the Citibank FCDU dispute. The first was the status of the interbank deposits that other creditors had placed with Citibank’s Manila office, and the terms under which those deposits could be withdrawn. The second issue concerned the responsibility for that foreign exchange exposure to the Philippines-whether Citibank, or its depositors, would have to put up the pro rata share for those deposits in any new money agreement for the Philippines. The dispute was resolved in stages. In April, Citibarik received permission from the central bank to remit 46 percent of the FCDU deposits to their owners.13 The dispute over the remaining deposits and the exposure issue continued until November, when a compromise was reached. Citibank converted the remaining interbank deposits to four-year time deposits and accepted new money exposure for the 46 percent portion that it had remitted in April. The second issue involved approximately $57 million in loans to Planters Products, Inc. (PPI), the largest importedmarketer of fertilizers in the Philippines. The company was in the private sector, and the loans did not carry a government guarantee. It was the insistence of PPI’s creditors that the Philippines assume the obligations of the firm that formed the basis of the dispute. The Philippine government administered price ceilings for fertilizer, a program started when world fertilizer prices rose sharply in 1973. Losses on fertilizer imports and on domestic production of fertilizer were made up by a subsidy program for firms in the industry, a program of which PPI had been the largest beneficiary. The company encountered severe financial difficulties after October 1983 because of restricted supplies of foreign exchange for imports and because the government fell behind in its subsidy payments. PPI’s creditors argued that the firm was in effect a state-controlled corporation and the government’s failure to provide the subsidies to which PPI was entitled was the reason for its financial difficulties. The creditors demanded that the Philippine government assume the loans before they would participate in the commercial bank rescheduling. The Philippines finally worked out a compromise on PPI. The loans would remain in the private sector. However, Finance Minister Virata signed a letter of undertaking in May 1985 stating that the central bank would satisfy PPI’s subsidy claim and that a rehabilitation program for the company would

544

Robert S. Dohner and Ponciano Intal, Jr.

be initiated. While this compromise allowed the bank rescheduling to go forward, it did not lay to rest the PPI issue, which reappeared in the negotiations in 1987, The last issue that almost derailed the negotiations was the reluctance of a major creditor bank to participate. The National Commercial Bank of Saudi Arabia had roughly $150 million in trade financing exposure to the Philippines. The bank initially refused to participate in the rescheduling program, holding out for a separate agreement with the Philippines. l4 Manufacturer’s Hanover, the lead bank in the advisory committee, refused to sign an accord without the participation of National Commercial. This issue, along with the dispute over PPI, held up the commercial bank agreement for several months until National Commercial finally agreed in April 1985 to participate in the program. The Philippine government finally signed its restructuring agreement with the bank advisory committee on 20 May 1985, over a year and a half after the country had declared its first moratorium. The terms of that restructuring agreement are also outlined in table 7.7. The agreement had three parts. The first was a rescheduling of $5.88 billion in debt falling due between 17 October 1983 and 31 December 1986. Of this restructured debt, about $3.4 billion, or 57 percent, was short-term debt. The restructuring agreement included different provisions for debt that was owed or guaranteed by the public sector, debt of the private financial sector, and nonguaranteed debt owed by the private corporate sector. The outlines of the programs and the amounts involved are shown in table 7.8. Of these, the procedures for the restructuring of private corporate sector debt were the most complex. Creditors and their corporate debtors were left to work out arrangements for repayment of existing loans. Repayments could be made on the original schedule or they could be restructured. In addition, the central bank offered an option for foreign exchange cover for restructured private corporate debt. A final option, similar to the FICORCA program in Mexico, offered forward exchange protection and credit assistance for corporate borrowers in financial distress. l5 For all of the options under the private corporate borrowers program, counterpart payments were made to the central bank on the agreed schedule and the central bank assumed the corresponding external obligation, payable on the same schedule as restructured public sector debt. In each case, the commercial risk of the loan was borne by the original lender. l6 The agreement also included a new money commitment by the commercial banks of $925 million, or 7.5 percent of their outstanding commitments on the date of the moratorium. This facility carried an interest rate spread of 1% percent over LIBOR, although a creditor bank could, at its option, receive its domestic prime rate plus 1% percent. The last part of the commercial bank agreement was a $2.97 billion revolving trade facility. This corresponded to the amount of trade financing

545

PhilippineKhapter 7

Table 7.8

Commercial Bank Financing Program, May 1985 Amount (million $)

Rescheduling: Public sector debt Medium, long term Short term

Total Private financial debt Medium, long term Short term Total Private corporate debt Medium, long term Short term Total Total rescheduled New money facility Revolving trade facility Total

Maturity (grace)

Facility Fee (%)

I%

2,059 1,183 3,242

1%

1% <2

16 1,594 1,610 448 585 1,033 5,885 925 2,974 9,784

Spread over LIBOR

10 (5 yrs)" 10 (5 yrs)"

9 (5 Y") 30 June 87'

1%s' 1 %a

15/ab 1%

Source: Central bank. "Once counterpart deposit has been made and the central bank has assumed the external obligation. bOr 1% over prime, at creditor bank's option. 'Facility exists through 31 December 1986. final maturity date 30 June 1987. d k r annum.

outstanding at the time of the moratorium, and each commercial bank agreed to maintain a level of exposure that was the same as that on 17 October 1983. The trade facility also included central bank overdrafts run up before the moratorium. Availments from the trade facility could be used to repay current and past due trade obligations. Any unutilized portion of the facility was to be placed on deposit with the central bank. The trade facility carried an interest spread of 1'/4 percent over LIBOR and a facility fee of Yx percent. The new money commitment, at $925 billion, was well below the original proposals of $1.6 billion in additional money from the commercial banks. The total financial commitment of the commercial banks in the Philippine program was roughly the same, with the smaller amount of new money balanced by the larger rescheduled amounts. Much of the difference was absorbed in the trade facility, which greatly exceeded the need of the Philippines for trade finance. l7 The terms that the Philippines received in this first rescheduling were roughly comparable to those negotiated with other LDCs at that time (Chile in November 1984, Costa Rica in May 1985, and Argentina in August 1985), although they were less favorable than the terms that Mexico received in March and August 1985." But, reflecting the suspicion with which the commercial banks now held the Marcos regime, most of the drawings under

546

Robert S. Dohner and Ponciano Intal, Jr.

the new money facility were pushed toward the end of the program. The first drawing, to be made after the Philippines passed the second IMF program review, was for $400 million, almost all of which went to settle outstanding arrears. Subsequent drawings were scheduled for $100 million, $175 million, and a final drawing of $350 million after the Philippines had successfully completed its IMF review at the end of March 1986.

7.7 Evaluating the Adjustment Period Measured purely in terms of stabilization, the Philippine adjustment program in 1984 and 1985 was a great success. The country was able to meet its external balance requirements under the IMF program. External arrears were eliminated by the end of 1985, and net international reserves rose faster than the program required. The current account deficit narrowed sharply and in 1985 was virtually balanced, a much greater change than had been anticipated in the program (see table 7.6). The extent of the Philippine adjustment may be gauged from the shift in the noninterest current account shown in table 7.9. In just two years this balance increased by 9 percent of GNP. The Philippines was also successful in rapidly bringing inflation under control. l 9 The consumer price index was rising at rates of 50 to 70 percent throughout 1984 (see fig. 7.2). But by May 1985 the inflation rate over the previous six months had been reduced to an annual rate of less than 10 percent. Over the entire year 1985 (December to December) consumer prices increased by 5.7 percent, the lowest inflation rate since the 1960s. During 1986 consumer prices actually fell. Thus the Philippines, which had been borrowing abroad in the amount of 8 percent of GNP, was successful in making the adjustment to the severing of its access to the world capital market. However, this stabilization was purchased at a tremendous cost of output and income and with considerable damage to the financial system and many industries. Real GNP by 1985 had fallen 10.4 percent from its 1983 level. Per capita incomes fell by about 15 percent, which erased almost all of the gains since the first oil shock. The effect was particularly severe on investment, which fell by 50 percent during this two-year period. By January 1985 the unemployment rate had

Table 7.9

Balance % of GNP

Noninterest Current Account Balance (in millions of U.S. dollars) 1982

1983

1984

1985

1986

1987

- 1,575

- 1,139

820 2.6

1.827 5.8

2,813 9.3

1,687 4.9

-4.0

-3.3

Source: Central bank, and NEDA, National Accounts Section

547

Philippines/Chapter 7

risen to 14 percent and the underemployment rate was estimated at 45 percent. Although increases in legislated minimum wages did not nearly compensate for the rise in prices after 1983, there was widespread noncompliance, leading to increased, although still illegal, strikes in 1985. The recession was unevenly distributed across sectors, as is shown in table 7.10. Conditions were generally less severe in the rural areas. Agricultural output continued to grow through the period after 1983, as devaluation increased production incentives and real incomes in the rural areas. The sugar industry, however, is a glaring exception to this characterization. The collapse of international sugar prices, the expiration of long-term supply contracts signed at favorable prices, and the inability of NASUTRA to make payments to sugar millers and producers, left many producers and workers Table 7.10

Real GDP by Sector and Manufacturing Industry Real Gross Value Added (total percentage change)

GDP market prices Agriculture Industry Mining Manufacturing Construction Electricity, gas Services Transport, communications Trade Finance & housing Other services Total Manufacturing Miscellaneous manufacturing Basic metals Publishing Beverages Leather Footwear Electrical machinery Food Rubber Tobacco Petroleum products Furniture Wood & cork Paper Chemicals Textiles Nonmetallic minerals Metal products Machinery (nonelectrical) Transport equipment Source: NEDA, National Accounts Section

1983-85

1985-87

-9.6 5.6 - 19.9 - 10.1 - 14.2 -44.8 4.1 -9.8 -5.9

6.7 4.1 5.7 - 14.0 8.0 -6.9 33.1 9.5 6.0 7.7 34.3 4.0 8.0 -3.1 4.0 18.3 1.5 - 2.9 16.1 30.8 11.0 8.5 - 36.6 6.7 28.4 -22.4 18.4 -9.0 37.1 9.3 8.4 15.4 3.7

1.o

-37.1 - 6.5 - 14.2 32.0 13.8 7.3 3.7 1.5

-0.5 -4.0

-6.9

- 13.0 - 15.5

- 17.2 -23.2 -23.9 -26.0 - 26.7 - 29.5 -35.3 -35.5 -50.1 -81.9

548

Robert S. Dohner and Ponciano Intal, Jr.

destitute and led to widespread famine in the province of Negros Occidental. Output in the mining sector also dropped sharply during this period, again because of weak external prices. The costs of the stabilization program itself were felt most heavily in the urban sector-by the construction and manufacturing industries and by the urban wage earner. The sharp fall in domestic investment led to a collapse of the construction industry. Within manufacturing, the industries that were hardest hit were the industries that the system of trade protection had worked so hard to protect-import-substituting industries producing intermediates or durable goods. In some cases, such as in transport equipment, major portions of the manufacturing sector went under during the stabilization episode. Export-oriented industries, such as footwear, electrical machinery, and much of miscellaneous manufacturing, fared better, although even here the recession had its costs. Labor force data for the Philippines is sketchy, but indicates that industrial employment fell by about 9 percent between 1983 and 1985. Legislated minimum wages for nonagricultural workers fell in real terms each year after 1981. Actual compensation paid also fell in real terms during the 1980s; data from the National Wages Council estimates the drop for unskilled workers in Manila at 5 percent in 1984.'' In the sections that follow we look more closely at how the external and inflation adjustments were achieved and the reasons for their high output costs.

7.7.1 External Adjustment Between 1983 and 1985 the dollar value of merchandise imports fell by 32 percent. It was this compression of imports that was responsible for the elimination of the current account deficit. Imports were reduced through lower incomes and, in particular, through the reduction of domestic investment. Capital goods imports fell by 54 percent over the same period. However, the severe shortage of foreign exchange, high black market currency premiums, and the tightening of quantitative restrictions on imports were also responsible for some of the drop. Exports present a mixed picture, rising by more than expected in 1984 and then falling sharplly in 1985. Over the entire 1983-85 period the dollar value of merchandise exports fell by 7.5 percent. Exports of the major traditional commodities-sugar, coconut products, lumber, gold, and copper concentrates-fell in volume as well as value during the period. This was the result of weak external prices, shifts in domestic production incentives away from traditional commodities, and in some cases, poor weather. Nontraditional manufacturers exports did reasonably well, with very strong gains in 1984 and somewhat weaker performance in 1985. Although exchange and tax policy was designed to shield nontraditional exports, there is still evidence that this sector was hurt by the stabilization

549

PhilippinesKhapter 7

period. Consignment imports were allowed without restriction, and some firms with foreign partners were able to finance imports through equity contributions from their parent firms. However, domestic credit was extremely tight, and there were shortages of foreign exchange even for goods on the priority allocation list. Indirect evidence of the effect of the disruption of the domestic economy on the manufacturing export sector comes from Philippine performance in its largest export market, the United States. The Philippines lost market share to other exporters during this period in most of its important nontraditional exports (table 7.11). While the Philippines reduced import levels, it had less success in shifting the balance of incentives in favor of the production of tradable goods. The peso had appreciated steadily in real terms from 1978 to 1982, measured both by comparison with the currencies of the Philippines’ trading partners and by the ratio of traded to nontraded goods prices domestically. The devaluations of 1983 and 1984 reversed this process, but did not succeed in decisively shifting the real exchange rate in the Philippines. Although the devaluations were large, they were mostly offset by high rates of inflation in the country. In addition, the disparities in protection rates across sectors widened considerably during this early period, increasing the protection of much of the domestic industrial sector. The peso actually appreciated at the end of 1984 and in early 1985, eliminating the gains in competitiveness that had taken place in the two previous years (see fig. 7.1). The real depreciation that took place during 1986 was mainly due to the fall of the dollar; the Philippine peso experienced little change against the currencies of its Asian competitors. In summary, the Philippines was successful in dealing with the break in external funding and the severe foreign exchange constraint of the adjustment period. The way in which this was done, however, increased the domestic output cost of the adjustment. What is more important is that the Philippines failed during this period to make the relative price adjustment Table 7.11

Share of Philippine Exports in U.S. Imports (percentage)

Refined sugar Lumber, shaped Copper ore, etc. Coconut oil Transistors, valves, etc. Electronic microcircuits Electronic components Furniture, parts thereof Clothing Footwear Source: U N

1980

1981

1982

1983

1984

1985

1986

8.7 11.5 58.4 83.1 9.9 12.2

5.9 14.1 50.5 91.7 11.9 14.8 1.1 5.4 3.7 1.6

8.1

8.4 27.7 91.3 12.7 15.5 2.2 3.9 3.3 i .0

10.6 13.5 17.3 86.6 12.3 14.7 4.3 3.4 3.3 0.7

10.9 11.9 24.6 79.7 10.5 12.0 6.0 3.0 3. I 0.6

13.2 10.4 0.0 64.6 9.3 11.6 3.9 2.4 3.0 0.3

12.6 8.5 0.0 90.1 7.2 9.5 2.2 2.0 2.7 0.3

1.1

4.4 3.5 I .3

Commodity Trade Statistics, series D

550

Robert S. Dohner and Ponciano Intal, Jr.

that would facilitate a resumption of economic growth in the face of the terms of trade deterioration that the economy experienced and the restricted access to international capital markets that the rest of the decade would entail. The adjustment episode halted and partially reversed the movement toward trade liberalization in the 1980s and did little to change real exchange rates. 7.7.2 Domestic Absorption The current account balance reflects dependence on foreign savings as well as the balance in trade and service flows. Thus, the elimination of the current account deficit in the Philippines was also matched by an equivalent reduction in reliance on foreign capital to support domestic expenditure. Here we examine how this reduction was achieved. We divide the economy into public, private, and foreign sectors, each of which can draw capital from, or provide excess savings to, the other sectors. The current account deficit of the Philippines is just the net demand for additional savings from the public and private sectors combined. This is shown in table 7.12, which is based on national income statistics. During the adjustment period, 1983 to 1985, both the private and the public sector reduced their dependence on external finance. The adjustment was considerably larger in the private sector, where the net deficit position (investment minus savings) was reduced by 6.2 percent of GNP, compared to a reduction of 1.8 percent of GNP for the public sector. The extent of the adjustment by the private sector is even more apparent from the gross figures on investment and savings. The change in the net balance of the private sector was achieved in the midst of rapidly falling private sector savings. The Table 7.12

SavingdInvestmentBalances (percentage of GNP)

Private sector Investment Savings Personal Corporate” Capital consumptionb Surpluddeficit Public sectof Investment Savings Surplus/deficit Net foreign resourcesicurrent account

1980

1981

1982

1983

1984

1985

23.73 20.75 5.98 5.49 9.28 -2.99

22.67 21.88 6.51 5.27 10.10 -0.80

21.61 18.02 3.25 4.44 10.33 -3.59

21.43 16.57 1.97 4.27 10.33 -4.87

15.05 12.33 -0.80 2.67 10.46 -2.72

12.57 13.95 -0.53 3.50 10.98 1.38

6.94 4.98 -1.96 4.95

8.04 3.79 -4.25 5.05

7.16 3.20 -3.96 7.55

6.09 4.00 -2.09 6.96

4.11

3.69 3.37 -0.33 -1.05

Source: NEDA, National Accounts Section.

“Includes savings by government-owned corporations. blncluding statistical discrepancy. ‘National government plus government corporations.

4.02 -0.08 2.75

551

PhilippinesKhapter 7

reduction in private sector investment is enormous-over 10 percent of GNP since the early 1980s. Since depreciation remained roughly constant at about 10 percent of GNP, net investment by the private sector practically disappeared during this period. Both individuals and corporations sharply reduced their savings after 1982. A number of factors led to the drop in savings: a reduction of corporate incomes during the domestic recession, an increase in capital flight that occurred in 1982 and 1983, and the desire to cushion consumption expenditure over the adjustment period. The Philippines is unusual among countries that went through adjustment during this period in that real consumption expenditure actually increased steadily throughout the process. (Although this was not enough to sustain real per capita consumption expenditure, which fell by 3.3 percent from 1983 to 1985.) 7.7.3

Monetary Policy and Inflation

The Philippines was highly successful in reducing the domestic inflation rate, in contrast to many other countries that have tried to adjust to a debt crisis. The domestic inflation rate dropped as quickly as it had risen in the Philippines; by 1985 consumer prices were rising at their lowest rate in two decades, and prices actually fell during 1986. Three factors account for the rapid reduction of the inflation rate: a tight monetary policy, the stabilization and unification of the exchange rate, and the absence of indexing and other features that add momentum to inflation. The various indicators of monetary growth in table 7.13 indicate the tightness of monetary policy in 1984 and 1985 after the huge growth that took place in 1983. The growth rate of nominal money in circulation, both M1 and M3, was low by historical standards, and the real money supply dropped sharply in 1984. By December 1984 the real values of all three aggregates were at their lowest points for the decade. Velocity measures for money in circulation increased sharply in 1984, in line with the tightening of policy. The course of reserve money was quite different, reflecting the substantial fall in the money multiplier that took place between 1982 and 1985. This was the result of increased currency holdings by the public, as confidence in financial institutions diminished, and the substantial increases in reserve requirements of 1983 and 1984. Interest rates lagged behind the increase in inflation in 1983 and its retreat in 1985. As a result, real interest rates, though negative during most of 1984, were at historically high levels through most of 1985 and 1986 (figure 7.3). The high real interest rates of these two years were the product not only of slow monetary growth, but also of the heavy public sector demands on available credit, a point which is discussed below. The second reason for the reduction in inflation was the steadying of the exchange rate after October 1984 and the narrowing of the black market premium on foreign exchange, as bank trading in foreign exchange resumed.

552

Robert S. Dohner and Ponciano Intal, Jr.

Table 7.13

Money Supply and Growth Rates (billion pesos and percentages)

Money supply Reserve money Narrow money M3 Growth rates Reserve money Narrow money M3 Real money growth ratesa Reserve money Narrow money M3 Velocity Reserve money Narrow money M3

1980

1981

1982

1983

1984

1985

1986

1987

16.19 22.54 67.80

17.80 23.52 82.08

18.64 23.50 95.30

27.72 32.49 113.00

33.45 33.63 121.20

37.99 35.83 132.80

49.98 42.66 149.80

59.53 52.38 160.50

10.7 19.6 18.2

9.9 4.3 21.1

4.8 -0.1 16.1

48.7 38.3 18.6

20.6 3.5 7.3

13.6 6.5 9.6

31.6 19.1 12.8

19.1 22.8 7.1

-6.1 1.5 0.3

-0.6 -5.6 9.5

- 3.4 -7.9 7.0

17.9 9.7 -5.9

- 20.0

7.5 0.8 3.7

32.0 19.4 13.2

11.4 14.9 0.2

90.8 82.2 110.9

94.8 90.4 105.1

100.0 100.0 100.0

76.1 81.8 95.4

87.5 109.6 123.4

87.0 116.3 127.2

67.9 100.3 115.8

64.2 92.0 121.8

-31.4 -28.9

Source: Central bank, Annual Report and Philippine Financial Statistics.

“Calculated using December CPI.

3c 2c

1c C -1 0 -

20 30

~

40

~

-

~

50 60

- 70 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 I I I I I I I I I I I I I I I I I I I 1 1 1 1 1 1 1 1 1 1

83

I

84

Fig. 7.3 Real interest rates Note: Treasury bill rate less CPI inflation rate.

I I

85

I I

86

I I

553

PhilippinesiChapter 7

Monthly values of the exchange rate and the black market premium are shown in table 7.1 above. The black market premium was finally eliminated with the floating of the exchange rate in October 1984, and the exchange rate hit a peak the next month, In November the central bank reduced its purchases of foreign exchange in the market, and from November 1984 to June 1985 the exchange rate appreciated against the dollar, rising by a total of 7 percent before starting to depreciate again. However, the nominal appreciation of the peso resulted in an appreciation of the real exchange rate of about 1 1 percent (see table 7.2 and fig. 7.1). This reversal of the real exchange depreciation immediately became an issue with the IMF during and after the first standby review. The Philippine government argued that the limits on growth of reserve money prevented them from intervening in the foreign exchange market to prevent the peso’s rise. The Fund in turn argued that remaining restrictions on bank holdings of foreign exchange had prevented them from purchasing and in fact had turned some banks into unwilling net sellers (IMF 1985a, 1 1 - 12). The final reason for the sharp reduction in inflation in the Philippines is the absence of many of the institutional features that tend to perpetuate inflation in other countries. Price ceilings and administratively set prices were adjusted rapidly during this period to changes in the exchange rate and increases in external prices. In addition, by the end of 1984 most consumer goods price ceilings had been removed. Thus there was no buildup of price increases that had not yet been passed through to consumers. Nor were wages a source of inflationary pressure. Only about 10 percent of the labor force was unionized, and unions remained relatively weak under martial law.21 Strikes were banned in vital industries, and prenotification was required for strikes in other industries. The government set minimum wages for agricultural and nonagricultural workers, with upward adjustments for cost of living increases. However, the adjustments were not automatic, and in the early 1980s real minimum wages fell significantly. In addition, distressed firms could apply to the National Wages Council for an exemption from the minimum wage rules, and noncompliance with minimum wage requirements was widespread. Finally, and most importantly, there seems to be a remarkable ability among the Philippine populace to tolerate real wage declines. Real wages fell significantly after the decontrol episode in the early 1960s and after both oil price shocks. This characteristic of the Philippine inflation process-rapid rises in the inflation rate followed by equally rapid declines-is evident from the first oil shock and from other inflation episodes in the country’s postwar history. With this kind of inflation process, the tight monetary policy that was imposed in 1984 and maintained through 1985 and into 1986, despite the very low rate of inflation at the end, was a case of misplaced worry and significant overkill. While the Philippines definitely had inflation wrung out of the economy, the restrictive monetary policy of the period, coupled with

554

Robert S. Dohner and Ponciano Intal, Jr.

other characteristics of financial institution policy, resulted in the virtual drying up of private domestic credit, as well as huge increases in its cost when available. The extreme domestic credit stringency drove many firms to the wall, both nonfinancial and financial, and left the financial system unwilling to take on anything but the most minimal lending risk. The 1984-85 period saw continuing expansionary influences on the money supply. These included the expiration of the remaining forward contracts on foreign exchange, an almost doubling of net credit to the public sector, increasing assistance to financial institutions in distress, and mounting interest payments on outstanding central bank securities. However, in the last half of 1984 and throughout 1985 the central bank maintained a very tight monetary policy. Reserve money, the monetary component subject to ceiling in the IMF adjustment program, increased by only 11 percent from the end of 1984 to the end of 1985. When reserve deficiencies and reserve-eligible government securities are included, the base for deposit expansion increased by only 8 percent during the year. To achieve a tight overall policy, the central bank sold large amounts of its own securities to the banking system and directly to the public. As discussed above in section 7.5.1, the sale of Job0 bills began in earnest in August 1984. These had short maturity, large denomination, and flexible interest rates, and defined a floor rate of interest for fully secured loans and other financial obligations. Between June and December 1984 the stock of outstanding central bank securities rose from P. 10 billion to P. 14 billion, or to 15 percent of the domestic credit extension of the central bank (see table 7.3).** The central bank continued to sell its own securities in 1985. In addition, starting in the first quarter of that year it entered into a number of reverse repurchase agreements with commercial banks. Under a reverse repurchase agreement, the central bank would sell government securities to a bank with an agreement to repurchase them at an agreed upon price at a later date (typically thirty days). By September 1985, outstanding central bank securities and reverse repurchase agreements had risen to over P. 38 billion (28 percent of the domestic credit extension of the central bank) before diminishing toward the end of the year. Credit tightened dramatically in 1984 and 1985 as a result of the stricter monetary policy of the central bank and also due to an unprecedented fall in intermediation by the financial sector. The real money stock fell by 31 percent between the end of 1983 and the end of 1985, while real M3, a measure of funds available from banks and quasibanks, fell by 26 percent. The fall in domestic intermediation had a number of causes. The first was uncertainty about the safety of the financial system, which led to increased demand for currency by the public. Diminished confidence particularly affected nonbank institutions active in the money market-investment companies and finance companies-as deposit-substitute holdings dropped

555

Philippinesichapter7

by over two-thirds in real terms. The financial system was also affected by significant disintermediation, as the central bank and the Treasury sold their securities directly to the public. Between the end of 1983 and the end of 1985, central bank and national government securities held by the public increased from P. 19 billion to P. 56 billion, reaching a level corresponding to 42 percent of M3. Commercial banks responded to the credit tightness and increased uncertainty in the financial markets in much the same way as did the public, by shifting into safer and more liquid assets. In table 7.14 we show components of commercial bank asset portfolios in the Philippines. Between December 1983 and September 1985 there was a significant increase in bank holdings of liquid assets, particularly in their holding of central bank securities and reverse repurchase agreements. At the same time, there was a shift out of other assets, particularly loans and discounts, which fell over the period in nominal terms and by 44 percent in real terms. The result of this shift in bank asset portfolios, combined with the direct sale of government securities to the public, was a significant redirection of available credit toward the public sector and away from the private sector. Domestic credit to the private sector fell by 20 percent in nominal terms and by 50 percent in real terms between the end of 1983 and the end of 1985. Here, as in the case of expenditure, most of the adjustment was done by the private sector. Some fall in private credit demand would have been inevitable given the reduction in expenditure that the stabilization program entailed and the foreign exchange constraints that limited imports of necessary inputs and equipment for many industries. However, the reductions in real credit Table 7.14

Commercial Bank Asset Portfolios (billion pesos)

Liquid assets Central bank obligations Total loans Loans & discounts Total assets Percentage shares Liquid assets Central bank obligations Total loans Loans & discounts Memo item: real loans & discounts

Jun 83

Dec 83

Jun 84

Dec 84

Jun 85

Sept 85

66. I 16.0 132.5 94.6 232.8

57.4 16.2 151.8 109.2 248.2

76.3 23.5 158.4 117.8 278.9

95.3 31.2 153.4 116.3 303.5

106.1 40.0 137.4 104.5 293.0

104.4 31.3 129.6 97.4 287.9

28.4 6.9 56.9 40.6

23. I 6.5 61.2 44.0

27.4 8.4 56.8 42.2

31.4 10.3 50.5 38.3

36.2 13.7 46.9 35.7

36.3 13.0 45.0 33.8

100.0

95.1

83.5

67.2

58.0

53.4

Source: Business Day. quarterly surveys, data from published bank Statements of Condition. Note:

Liquid assets: Cash, checks, due from the cenhlil bank and other banks, secured trade accounts, bonds. Total loans: Loans and discounts, interbank loans, agricultudagrarian reform bonds, bills purchased, customer liabilities. Real loans: Deflated by monthly consumer price index.

556

Robert S. Dohner and Ponciano Intal, Jr.

availability were far greater than output reductions in the economy as a whole and in manufacturing, and there were numerous complaints in the Philippines about the availability of credit as well as its cost. This was true even in industries that were not as market limited as manufacturing, such as export industries and agriculture. Commercial bank credit to agriculture declined by 36 percent in nominal terms in 1984 and stayed constant in 1985 (Montes 1987, 29). For industrial forms the result of the stringent credit measures was widespread bankruptcy, particularly among firms in the import-substituting manufacturing sectors, although data on the extent of insolvencies is very sketchy. Perhaps one-quarter to one-third of all firms faced debt servicing problems.23 In many cases, newer, technologically efficient firms with high debvequity ratios went bankrupt, while less efficient but less extended firms survived (World Bank 1987, 10). Among surviving firms there has been a significant deterioration in financial ratios, particularly in sugar milling, mining, construction, and cement. The stabilization episode was also an extremely difficult one for financial institutions. Sharply increased interest rates in 1984, after a period in which the government had been urging banks to lend on longer term, caught many banks in a term squeeze as funding costs increased faster than the earnings of loan portfolios. In addition, the high reserve requirement (which yielded 4 percent interest) plus the 25 percent lending to agriculture requirement (much of which was met through government bonds yielding 9 percent) plus a tax on gross bank receipts resulted in a very high rate of taxation of intermediation when inflation rates rose. This raised break-even bank spreads, causing difficulties for banks and their loan customers.24 The domestic recession and high interest rates turned many loans into nonperformers. Finally, the general financial uncertainty and instability led to substantial withdrawals from several institutions, including investment houses and finance companies, one major savings bank, and several smaller banks. Twenty-three financial institutions failed in 1984 and another fiftyeight failed in 1985 (IMF 1986b, 34). Central bank emergency assistance increased sharply on several occasions during 1984-86, threatening the country’s monetary targets. The system for delivery of credit to the rural sector essentially broke down. The bank responsible for financing of the sugar industry, Republic Planters, became insolvent. Over 100 of the roughly 1,000 rural banks became insolvent, and arrearages of rural banks as a proportion of central bank rediscounts to them reached 83 percent in 1985.25 Ironically, what may have saved many financial institutions was the extensive central bank borrowing at high interest rates that allowed banks to greatly increase the yield and liquidity of their portfolio by buying securities. This eventually involved large transfer of interest payments from the central bank and the passing of the brunt of the adjustment to the bank’s traditional customers.

557

PhilippinedChapter 8

7.8 Conclusions on the Adjustment The crisis and adjustment period in the Philippines reflects a striking about-face in Philippine relations with the IMF and its external creditors. After a period of relatively lenient standby programs, regularly exceeded by the Philippines and then renewed, the Philippines was placed under an extremely stringent program that was successfully carried out in what would prove to be the last months of Marcos’ hold on power. This period demonstrates that institutions have memories, and these are reflected in their programs. The events of 1983-the speed with which the 1983 program targets were exceeded, the overstatement of international reserves, and the tremendous increase in the monetary base-all served to transform the IMF from “doting parent to vengeful god” (Montes 1987, 21). This resulted in prior conditions and an IMF program that were unusually harsh.26 The stringency of the adjustment that was actually carried out also reflected the unwillingness of creditors to supply additional funds to the Philippines and resulting capital inflows that were less than the IMF program. The Philippines compounded its problems by waiting until its foreign exchange reserves were almost exhausted, by failing to draw on standby credits that it had negotiated, and by delaying its declaration of a moratorium until after the central bank had run up substantial overdrafts. After years of failing to meet conditions in IMF programs, the Philippines met or more than met the conditions in this one. The assiduousness with which the Philippines pursued these targets was in part due to the limited options that the country had; foreign exchange was extremely tight, and the release of funds under both the IMF and commercial bank programs, other than for settling arrears, was quite slow and subject to frequent reviews. But Philippine adherence to program conditions also reflected the increasing domestic opposition to the Marcos regime after the Aquino assassination and the need of the government for external support.” The stabilization program as it was carried out represented a tremendous gamble for the regime, as it tried to simultaneously meet the external conditions while protecting its supporters, forcing a greater burden of adjustment on the private, nonassociated sector (Montes 1987, 21-27). The adjustment program that the Philippines carried out demonstrates the power of monetary restraint in an economy without institutional inflation mechanisms. The adjustment saw a tremendous shift in the noninterest current account and the elimination of both the current account deficit and inflation. In contrast to many debtor countries in Latin America, stabilization in the Philippines was rapidly achieved. But the stabilization was accomplished through reduction in income and a particularly large reduction in investment expenditure, at severe cost to the domestic economy and population. Little was achieved to lay the foundations for recovery and growth. The Philippines reduced imports, but did not shift resources toward

558

Robert S. Dohner and Ponciano Intal, Jr.

the tradable goods sector. Despite the large devaluations, the real exchange rate at the end of 1985 was nearly unchanged from its level in the early 1980s. Beyond the income loss and unemployment, the severity of the credit squeeze traumatized the financial and industrial sectors, forcing many firms under and undermining the willingness of the banking system to take on lending risk, factors that would later delay the recovery of confidence and investment.

The Aquino Government and Prospects for the Economy The “New Society” proclaimed by Marcos in the early 1970s failed to produce its promised improvements in the welfare of Filipinos, and, as martial law continued, opposition to the Marcos government grew. The oil shock and subsequent fall in commodity prices hurt incomes, particularly in the rural areas. Although measured economic growth was substantial, regional income disparities increased, and over the decade the proportion of the population living in poverty remained high.’ Continuing arrests and human rights violations by the military brought protests, particularly from the Catholic church. But the most crucial loss of regime support came in those groups that had provided the initial constituency for martial law. The patronage machinery on which the political foundations of martial law rested, especially the particularistic interventions and the generation and distribution of monopoly rents, tended over time to narrow the base of political support. By 1985 Marcos’ base was dangerously thin. The early support for the government within the rural sector faded as the momentum of land reform dissipated and as military abuses and corruption increased. Even though agricultural terms of trade were falling, the government increased its taxation of important crops through export and producer levies and through monopolization of trading and processing activities. The increasing incidence of rural poverty aided the insurgency, and the communist NPA took over effective administration, including taxation, in several areas. Although the military had been the main beneficiary of martial law, opposition within the armed forces developed over promotions and over the deployment of forces against the Moslem and communist insurgencies. Before 1972, the Philippine armed forces had been small, relatively professional, and reasonably effective. Martial law greatly zxpanded the size

559

PhilippinedChapter 8

of the military, politicized its command structure, and greatly increased the opportunities for personal gain. Loyalty and close association to the president, or to his wife, became the primary criterion for advancement within the services, and Marcos filled the highest posts with officers from his home province, Ilocos Norte, skipping over other officers in the ranks. Elite forces were created to protect the presidential palace, and military resources generally were concentrated in and around Manila. The path to advancement was through the capital; less favored personnel engaged in the war against the communist and Moslem insurgencies, mostly in badly led and ineffective operations. Soldiers in the field were poorly equipped and supported. In many cases they were forced to live off the population, adding to popular grievances toward the military. The business sector benefitted under martial law from restraints on labor organizing, the rapid rate of public investment, and the continuing protection of the manufacturing sector. Yet by 1985 the private business sector had almost completely deserted Marcos, prominent industrialists were outspoken critics of the regime, and white collar workers in the financial district in Makati were protesting against the government. The particularistic interventions that had favored associates of the regime, the creation of monopoly positions, and the aggressive takeovers through threat of ruin alienated increasing numbers of businessmen. Many sought accommodation with Marcos; more transferred capital beyond the government’s reach. The corporate bailouts of the early 1980s spurred protests from some in the business sector, and there was cautious, although increasing, opposition to the government.’ Marcos’ deteriorating health, the undefined succession mechanism, and the possibility that his death would lead to a seizure of power by his chief of staff, General Fabian Ver, or by his wife, Imelda, added to business unease. In order to diffuse some of the internal opposition to him and pressure from the United States, Marcos instituted political reforms in a carefully controlled fashion. Martial law was officially lifted in 1981, although Marcos retained his ability to rule by decree. Presidential elections were held in that year, although the opposition refused to field a candidate in an election that Marcos controlled. A legislature, the Batasang Pambansa, was created and elections were held in 1983. The elections were also carefully controlled, although opposition parties did field successful candidates. The Batasang was subordinate to the president; Marcos could overrule any act of the legislature and could dissolve the body if he wished. But it was the assassination of Benigno Aquino on his return from exile in the United States that changed the political atmosphere in the Philippines and brought continuing and widespread public protests and opposition to the Marcos regime. Political violence was by no means unknown in the Philippines and had characterized the local electoral process. But the killing of Aquino, the most prominent of the opposition figures, as he was being

560

Robert S. Dohner and Ponciano Intal, Jr.

escorted by military guard from the airplane on which he arrived, was such a shocking transgression that it irrevocably altered the public view of the regime. For the first time, the Manila upper and middle class felt physically threatened by the Marcos government. In the end Marcos also lost his external support. The attitudes of the IMF toward the Philippines changed markedly after the reserves overstatement discovery and the explosion in the money supply in the last half of 1983. The Fund imposed stringent conditions in the stabilization program, including a drastically reduced deficit target for 1985. The Fund also pressed hard (although unsuccessfully) for reforms in the sugar and coconut industry, threatening vital remaining Marcos constituencies. The World Bank took a somewhat more sympathetic view, but its funding declined sharply after 1983. More important for the maintenance of power in the Philippines was the shift in attitudes of the United States. U.S. relations with the Philippines had improved considerably with the inauguration of President Ronald Reagan in 1981. Vice President George Bush attended Marcos’ inauguration after the 1981 presidential election, Marcos visited the United States in 1982, and Reagan had planned a visit to the Philippines in 1983. The Aquino assassination and the growing protests in the Philippines and the United States against the Marcos government played a role in shifting the U.S. position. But it was the growing strength of the NPA that was in the end decisive, convincing even conservatives in the U.S. government that Marcos had become a liability and that fundamental reforms would have to take place in the Philippines. Despite mounting domestic and international opposition, Marcos remained a master of Philippine politics, constantly keeping his opponents off balance and exploiting the divisions among the opposition parties. In November 1985, in an attempt to diffuse pressure from the United States, Marcos called a snap election for 7 February 1986, counting on the divisions within the opposition and his control of the electoral process to assure an easy victory. He erred on both counts. The two major opposition groups were able to unite on Corazon Aquino, the wife of the slain opposition leader, as a presidential candidate, when Salvador Laurel, the leader of the other group, agreed to join her ticket as the vice presidential and id ate.^ The Marcos political machine, which had served so well in the previous twenty years, either deserted him or proved unable to deliver. Marcos won the official election count, but only through blatant and extensive fraud that was evident to the numerous foreign observers who had gathered in the Philippines for the election. In the weeks that followed, the opposition refused to accept the official election results and publicly called for Marcos to step down. Opposition party members organized protests and called for a boycott of financial institutions that were associated with the Marcos government, leading to large deposit withdrawals from several government banks.

561

PhilippinesiChapter 8

The breaking point came on February 22, when Marcos’ defense minister, Juan Ponce Enrile, along with General Fidel Ramos and a small number of troops, set up a command post at the Manila army base Camp Crame and declared their support for Corazon Aquino. The failure of Marcos and his chief of staff, General Ver, to quickly move against the rebels, the mobilization of huge numbers of civilians to surround and physically block loyalist troops from entering Camp Crame, and an eventual flood of defections from government forces left Marcos, by the time of his inauguration on February 25, a virtual prisoner in Malacanang p a l a ~ e He .~ left the Philippines the next day on a U.S. Air Force jet for Guam and then to exile in Hawaii. On 26 February 1986, Corazon Aquino was suddenly president of the Philippines, in an unexpected and uneasy alliance with the military. The presidential election and its immediate aftermath had a significant impact on economic policy. The fiscal and monetary discipline that the Philippines had exercised under its standby arrangement with the IMF was abandoned to the cause of electoral survival. At the end of the third quarter in 1985, the Philippines was well within targets for the money supply, but between October and mid-February the money supply grew by over 50 percent, reaching a level of P. 44 billion, well above the March 1986 program target of P. 39 billion. The government deficit ballooned to P. 10 billion in the first quarter, ten times the deficit of the first quarter of 1985 and more than the program target for the entire year. After the election, the central bank sought to rein in the money supply to meet the IMF program requirements. The central bank’s efforts were complicated by the opposition-inspired boycott of government-associated financial institutions which severely affected several commercial banks, particularly the Philippine National Bank, and required a sharp increase in emergency assistance. But the central bank persisted in selling its own securities, and by the end of June the money supply had been reduced by 13 percent to P. 38.0 billion. The result was a further domestic financial shock-Treasury bill rates rose from 16 percent in January to 24 percent in February, at a time when inflation was decelerating. Real interest rates on commercial bank loans reached 30 percent in the same period. The stringent monetary actions prevented the injection of funds during the election campaign from translating into higher inflation or currency depreciation, but those gains had a high price. Real investment fell by an additional 15 percent in 1986, and the episode further unsettled an already skittish financial system.

8.1 Challenges Facing the New Government Despite the mood of national euphoria that accompanied Corazon Aquino’s assumption of the presidency, the problems that her government

562

Robert S. Dohner and Ponciano Intal, Jr.

faced were extremely difficult ones, involving both political and economic reconstruction. The Marcos regime had corrupted and nearly destroyed key Philippine institutions. The military had become highly politicized, intimately involved in government, and accustomed to the patronage flows that martial law had created. Public esteem for the armed forces had almost disappeared. So had the effectiveness of the military as a force in countering the communist insurgency; in 1986 the NPA was far better organized and more capable of mounting operations than were the Armed Forces of the Philippines. However, despite its weakness in the field, the military remained a potent force in Manila politics, and its crucial role in precipitating the collapse of the Marcos government convinced many in the defense establishment that their position was coequal with that of the new president. The Philippine judiciary had also suffered under martial law. The ability of Marcos to dismiss judges at any level bent the judiciary into cooperation. Judicial decisions were increasingly based on influence or on wealth during the latter years of the regime. Other democratic institutions had been greatly weakened. The 1973 constitution had been molded to assure Marcos’ hold on government and had been repeatedly amended by him to suit immediate needs. The Batasang Pambansa, the legislature that Marcos created, had no substantive authority and was controlled by members of Marcos’ party, the Kilusang Bagong Lipunan (KBL). Political power and almost all decision making had been centralized in Manila under martial law; local governments had in some cases been superceded by the military, and the administrative ability of provincial and local governments weakened. Political issues and events dominated President Aquino’s first two years in power. The new government moved immediately to dismantle the political structures that Marcos left behind. Aquino dissolved the Marcos-controlled legislature. She dismissed thousands of local governors and mayors, replacing them with officers in charge, who would serve until elections could be held. The new government also moved quickly to recapture the assets that Marcos and his cronies had accumulated during the previous administration. Aquino’s first executive order established the Presidential Commission on Good Government (PCGG) to pursue legal action against Marcos and his associates outside the Philippines, and to identify and sequester their assets in the Philippines. Next was the need to provide a legal framework for the new government, and to begin to rebuild the institutions that had deteriorated or been destroyed by martial law. In March 1986 Aquino declared a “Freedom Constitution” that gave her the decree-making powers of her predecessor, although for a limited period, included a bill of rights, and called for a convention to draft a new constitution. The convention produced a draft constitution that was approved in a national referendum in February 1987. The constitution provided for a legislature similar to the U.S. Congress, and

563

PhilippineKhapter 8

congressional elections were held in May 1987. Local elections for mayors, governors, and other officials were held in January 1988, completing a process of four national referendums in just under two years. The other political challenge the Aquino government faced was assuring its own survival. Early coup attempts were mounted by Marcos loyalists. Later, more serious attempts were mounted by members of the armed forces, first in November 1986 and then in August 1987, in a coup that almost toppled the government. The military grievances against the government were several. The first concerned the national reconciliation policy that Aquino had adopted to deal with the Moslem and communist insurgents. The low pay and lack of support and equipment for military personnel also fueled discontent. Furthermore, military officers viewed themselves as political participants; some of the same individuals who had first deserted Marcos in February 1986 were the ones who led the coup attempt in August 1987. The partnership between the military and the Aquino government remains uneasy, and the military itself is highly factionalized. The Aquino government moved to satisfy some of the demands of the military after the August coup attempt, adopting a firmer stance toward the communist insurgency, increasing military pay, and removing from the cabinet members of whom the military had been especially critical. But Aquino was also forced to co-opt the military leadership, appointing and promoting on the basis of loyalty to the new government, and this has increased resentment in some quarters and set back the process of professionalizing the armed forces. While the NPA is a more serious threat to the security of the country, the military remains a more immediate threat to the elected government. The Aquino government also faced serious economic problems, as well as exaggerated expectations about what a new government could accomplish in the economic sphere. The recession of the previous three years had led to a sharp fall in per capita income, erasing the fruits of the previous eight years of growth. Open unemployment in Manila had doubled to 22 percent, while rates of underemployment were estimated at 40 percent. The importsubstituting industrial sector had been decimated by the recession, with production cuts in some industries as high as 80 percent. Real investment dropped by over half between 1983 and 1986, but substantial excess capacity and continued financial and political uncertainty limited its recovery. The Philippines also faced the burden of the external debt inherited from the previous administration. At the end of 1985 total external debt of the Philippines was $26.3 billion. Although the total had only increased by 6 percent in the previous three years, the ratio of Philippine external debt to GNP had jumped from 63 to 82 percent, the result of declining economic activity and the real depreciation of the peso. Net external borrowing was almost zero for 1986, but the total external debt jumped again to $28.3 billion. This was the result of the appreciation of the yen against the dollar during the year and the fact that about 15 percent of Philippine external debt

564

Robert S. Dohner and Ponciano Intal, Jr.

was denominated in yen. By the end of 1986, the debtiGNP ratio had risen to 94 percent, placing the Philippines among the most heavily indebted less developed countries. The 1985 rescheduling had reduced the immediate debt service burden, but in 1986 external debt service still absorbed 34 percent of total exports of goods and services. And the limited capital inflow under the restructuring agreement meant that, with interest payments and a current account surplus, the Philippines on net was transferring resources to the rest of the world amounting to 9 percent of GNP. Furthermore, the debt rescheduling agreement was to expire at the end of 1986, which meant that the new administration had to turn almost immediately to a new round of negotiations with the country’s external creditors. As important, and more complicated, were the fiscal problems facing the Aquino government. These were especially crucial, for the fiscal constraints severely limited the government’s room to maneuver and its ability to respond to strident domestic demands. The Marcos government had reduced the national government’s budget deficit to 1.9 percent of GNP in 1985 (although it missed the IMF target of 1.5 percent). But the reduction in the national government deficit is deceptive. The revenue effort of the government, the share of total government revenue in GNP, had declined markedly during the 1980s, increasing the required expenditure cuts to close the budget gap. In addition, the requirements for debt service and for support of state-owned corporations and financial institutions increased dramatically from 1 1 percent of expenditure in 1982 to 35 percent in 1985.5 The cuts came largely in government capital expenditures, which fell by more than 50 percent in real terms. But the Marcos government had also drastically reduced real outlays for operations and maintenance from levels that were already low relative to the public sector capital stock. Not only was the government not investing, it was consuming its current capital stock by neglecting maintenance. Public employment continued to expand during the stabilization period and was 11 percent (70,000 workers) higher in 1985 than in 1983. Real wages of public sector workers fell, and there were critical pressures for wage increases for teachers and soldiers. Finally, the hemorrhaging of the budgets of the government financial institutions continued to worsen in the first year of the Aquino government, as remaining government guarantees were called and as the interest burden of central bank securities increased. It was recognized early on that transfers to financial institutions were going to increase in 1986; in fact they almost doubled, reaching P. 24.2 billion, or 23 percent of national government expenditure and 4 percent of GNP.6 The economic landscape facing the new government was not entirely bleak; there were some significant advantages that assisted the new government in its first year in office. The first of these advantages was that the Marcos government had successfully carried out the stabilization

565

Philippinesichapter 8

program required by the IMF. The current account deficit had been closed, and the aggressive sale of central bank securities had reduced the inflation rate to below historical levels by the end of 1985. Thus the costs of stabilization had already been paid by the time the Aquino government came into office. The stabilization program had its weaknesses. It was more severe and costly than the situation required, and it was also pure stabilization-nothing in the program provided structural adjustments that would encourage and sustain a recovery in the future. But the achievement should not be underestimated, and its importance to the succeeding government was real. The second advantage for the Aquino government was that, for the first time in many years, external events worked in its favor. Philippine terms of trade rose by 16 percent in 1986, the largest increase since the 1973 commodity price boom. Almost all of this improvement came from the sharp fall in world oil prices, which reduced Philippine petroleum imports by $580 million, or 2 percent of GNP. Commodity export prices were little changed until the last quarter of 1986, when coconut prices rose sharply. World commodity prices rose strongly in 1987, and Philippine terms of trade rose by an additional 9 percent. International interest rates also slid during 1986. By the end of the year, LIBOR rates had fallen over two percentage points, saving the Philippines $200 million in interest payments. The Aquino government was the beneficiary of substantial international goodwill which it was able to translate into additional concessional finance. Official aid flows for 1986 came to $1.4 billion, over twice their level in 1985, although not as large an increase as the Aquino government had originally hoped for. Additional grant funds came primarily from the United States, which increased its Economic Support Fund contributions by $100 million, bringing the total to $300 million for the year. Major additional loan funds came from the Asian Development Bank and Japan. The Philippines was also able to secure additional commitments for 1987 and beyond, from the Japanese, the Asian Development Bank, and from the World Bank for a $300 million economic recovery loan. The successful stabilization program and the end of the Marcos government’s predation on the economy greatly diminished outflows through capital flight. Privately owned capital began to flow back into the Philippines in 1986 into real estate investments, the Philippine stock market, and campaigns for the congressional elections in 1987. As a result, pesos sold at a premium in the black market in Hong Kong, as residents sought to avoid the official market in Manila (see table 7.1). Personal transfers to the Philippines also increased sharply in 1986. What did not occur was a significant increase in foreign direct investment, despite the government’s desire for additional foreign capital. Inward foreign direct investment flows increased from their 1985 level, but remained well below their 1981-83 average.

566

Robert S. Dohner and Ponciano Intal, Jr.

The final advantage that the new government had when it entered office was the general revulsion against the excesses and plunder of the Marcos government. Aquino had campaigned on a platform of reducing the role of government in business activities, doing away with particularistic grants of privilege, and returning acquired companies, as well as many govemmentestablished corporations, to the private sector. The widespread desire to purge the country of the institutions of the previous government, the desire for economic reform and recovery, and in addition, the disarray of the sectors that had depended on government protection or largess, created an opportunity for sweeping economic reforms. Although economic policy was overshadowed by political reconstruction and political crises, the government was able to push through fundamental policy changes that had long been advocated by external critics of the Marcos government.

8.2 Economic Policy The Aquino government faced three types of economic policy problems. The first was short-run economic policy to achieve recovery after almost three years of recession. The second was the establishment of broad outlines of economic policy for longer term growth, while the third was a renegotiation of the Philippine external debt with official and private creditors.’ 8.2.1

Recovery

The Philippine stabilization episode was unusual in that domestic consumption expenditure did not drop, but instead continued to slowly rise as households drew down their savings. By 1986 there was little reserve left to consumption that could be drawn on to support recovery. Investment had been decimated during the recession and dropped further in 1986, a reflection of both the high degree of political and economic uncertainty and large excess capacity in most industries. The government immediately drew up a plan of increased public expenditure to boost the economy, primarily through a Community Employment and Development Program (CEDP) for labor-intensive projects in rural areas. The targets under the previous IMF program for the Philippines had been exceeded in the pre-election spree of the Marcos government, and the government was unable to draw on the second-to-last tranche of the standby agreement. Instead of a waiver, the government and the Fund negotiated a new standby program for the country. The atmosphere in the negotiations was far more cordial than that of the previous standby, and the resulting program was more liberal than its predecessor. The program allowed a 30 percent growth in base money from March 1986 to March 1987, the first increase in the government investment program since the crisis, and a rise in

567

PhilippineKhapter 8

the national government deficit from 1.9 percent of GNP in 1985 to 4.4 percent in 1986.8 The new program, submitted in September and approved in October, was for $262 million over eighteen months, with an additional $296 million available immediately under the Compensatory Finance Facility (IMF 1986~).Although real government expenditure did increase in 1986, organizing and implementing expenditure programs was difficult for the new government. The CEDP employment program was slow in getting started, and at the end of the year disbursements lagged well behind targets. Public sector investment also came in below targets due to a shortfall in expenditure by government corporations. 8.2.2 Policy Reforms The Aquino government had much greater success in implementing broad economic policy reforms. Although these took place under the shadow of political events, and not all reforms were achieved, by mid-1988 the policy environment in the Philippines had been fundamentally altered. Economic policy was more neutral in rewarding lines of economic activity, more transparent than it had been since the 1950s, and compared favorably with that of any other East Asian country. Understanding the success of the Aquino government in shifting economic policy requires some discussion of the ideological bases of the new administration. Although the opposition to the Marcos government in 1986 covered the entire ideological range, the 1986 presidential election was contested by only a limited segment of the opposition. The two groups that united to field Corazon Aquino and Salvador Laurel as candidates were from the center to center-right of the political spectrum, a group that has been characterized as “conservative reformists” (Lande 1986, 124). This group has provided the core of the new government, despite its appearance of being ideologically fractious. The conservative reformist nature of the cabinet was epitomized by President Aquino’s choices for the two key economics portfolios. The finance minister, Jaime Ongpin, was chairman of Benguet Mining Corporation and had also headed the citizens’ election commission, NAMFREL, in the presidential election. The central bank governor, Jose Fernandez, was the sole holdover from the previous government and had been appointed by Marcos, under pressure, after the reserves overstatement discovery. As central bank governor, Fernandez had been responsible for much of the success, and the severity, of the stabilization program. Other cabinet ministers represented a wider range of opinions. Furthest to the left was Labor Minister August0 Sanchez, whose pronouncements about labor justice and workers rights quickly made him the Aquino government’s bCte noire among business groups, bankers, and U.S. officials.’ More centrist, but also outspoken, was Planning Minister Solita Monsod. Monsod was strongly critical of foreign commercial banks for many of the loans that

568

Robert S. Dohner and Ponciano Intal, Jr.

had been extended to the Marcos government, and argued for a case by case review and “selective repudiation” when funds had been knowingly lent for corrupt or questionable projects. Within the cabinet, she has pushed for trade liberalization, land reform, and a more growth- and foreign-resourceoriented economic policy. The last key player on economic matters was Joker Arroyo, who held the powerful office of presidential executive secretary, or “little president” as it is known in the Philippines. Arroyo pushed for a more nationalist economic policy and resisted efforts to sell off state-owned or -acquired corporations, using his control over the flow of paper to the president to sidetrack measures he opposed. Arroyo was also a harsh critic of the military and a focal point for armed forces criticism of the government. He left the government after the coup attempt of August 1987, in an arrangement that also included the resignation of Finance Minister Ongpin. l o Although the Aquino government did not have the benefit of a transition period, it was able to draw on a group of University of the Philippines academics, who had been involved in critical analysis of the Philippine economy in the latter years of the Marcos regime.” This group quickly prepared a policy agenda for the new government that became known as the “yellow book” (Alburo et al. 1986). The economic policies proposed in the yellow book were essentially conservative, market-oriented policies, many of which had long been advocated by the multilateral institutions, the United States, and other outside observers of the Philippine economy.’* The authors of the yellow book argued for greatly reduced government intervention in the economy, an emphasis on alleviating poverty and achieving distributive justice through land reform, and a reorientation of development policy toward rural areas. To support the rural-based development strategy, the yellow book recommended an outward-oriented trade policy, the elimination of quantitative restrictions on imports, the elimination of taxes on traditional exports, and the avoidance of an overvalued exchange rate. The authors argued that debt servicing should be subordinated to the goals of economic recovery and achieving growth, but they favored seeking additional external resources to sustain growth rather than reducing debt servicing costs through unilateral action. l 3 Other recommendations included a shift in the basis of taxation toward direct taxes, the reduction of taxes on financial intermediation, and the privatization of much of the government corporate sector. The recommendations of the yellow book became the basis for economic policy under the Aquino government, although implementation of some proposals has proved difficult. l4 The Aquino government was able to move quickly to dismantle the particularistic and monopolistic interventions of the martial law government. In the coconut sector, it removed the export levy and the prohibitions on direct exports of copra and new investments in milling; as a result, the spread between world and farmgate copra prices narrowed sharply. In the

569

PhilippinesiChapter 8

sugar industry, the Aquino government abolished the Philippine Sugar Marketing Authority (PHILSUMA) and its control over domestic and export sales. A new regulatory administration was set up to allocate quotas for domestic sales and sales to the U.S. market, with the quota certificates freely tradable among producers. The government eliminated other monopolies, including those covering imports of livestock and most fertilizer. Fiscal Reforms Within a few months of coming to power, the new government agreed on and began to implement an extensive tax reform program, with surprisingly little opposition. The program had several objectives-to raise revenues, improve the responsiveness of the tax system to economic growth, and to lower the taxation of poor families. Additional revenue came from increases in taxes on cigarettes and beer, the removal of tax and duty exemptions from publicly owned corporations, and the adoption of a uniform corporate income tax rate with limitations on allowable deductions. The elasticity of the tax system was improved by converting excise taxes to an ad valorem basis, by making tax rates more uniform and removing exemptions, and by plans to strengthen collections. Although income taxes were eliminated altogether for families below the poverty line, and all export taxes, except those on logs, were removed, the program produced a net revenue gain estimated to be P. 15 billion in 1987, or 18 percent of total tax r e v e n u e ~ . ' ~ The reform proposal also included measures to strengthen tax administration. The government approved plans to introduce a general value added tax of 10 percent in 1988 to replace the poorly performing sales tax system. Tariff collections by the Bureau of Customs had been well below applicable duties on Philippine imports. After considerable internal struggle, the Philippines engaged the services of a Swiss firm, SociCtC GCnCrale de Surveillance, to do preshipment checks on Philippine imports from selected ports to improve customs collections. The Aquino government also introduced tax amnesty provisions to encourage payment of back taxes. The elimination of monopolies and the tax reforms were primarily administrative measures that could be carried out quickly once the new government came to power. Other problems carried over from the previous government were less easily solved. The Aquino government inherited approximately 285 government corporations and subsidiaries, many of them creatures of the Marcos government, as well as a huge burden of external liabilities and nonperforming domestic assets. In one of her first acts in office, Aquino established a Presidential Commission on Government Reorganization (PCGR) to study and make recommendations for reforms of government corporations, as well as public financial institutions. l6 Under the recommendations of the commission, 125 government corporations were to be sold, 37 retained in their current form, and the remainder abolished,

570

Robert S. Dohner and Ponciano Intal, Jr.

consolidated or converted to regular government status (Philippines 1987, 8). Along with the reorganization proposals, the Aquino government adopted policies to rationalize and increase the transparency of government support for the corporations. Government corporations were required to pay dividends on government equity and interest on outstanding government net lending. In addition, the Philippines promised the Fund and the World Bank that it would clear up the arrears that had developed among public sector corporations and between the corporations and the national government. The government took several steps to deal with the problems of the National Power Corporation, the largest of the public sector corporations. The principal difficulty of the NPC was the nearly completed but nonoperational nuclear power plant, and the $2.1 billion external obligation associated with it. Various concerns had been raised about the quality of the construction in the plant and the geological safety of the site. Based on these concerns, Aquino decided against operating the plant. The nuclear power plant and its debt obligations were transferred to the national government to assist the rationalization of NPC’s operations. The government also announced other reforms in the power sector, eliminating the tax and duty exemptions enjoyed by NPC and declaring that it would gradually eliminate the subsidized provision of electricity to Manila customers. l 7 Financial Institution Reform

The position of the two major government financial institutions, the Development Bank of the Philippines and the Philippine National Bank, was particularly acute. Almost 90 percent of the assets of DBP were nonperforming, as were well over half of the assets of PNB; in total, the nonperforming accounts came to $7 billion. l 8 Government budgetary support for the two institutions alone swelled to P. 16 billion in 1986, 15 percent of the budget, and 2.6 percent of GNP. National government aid to nonfinancial public sector corporations added an additional P. 11 billion. The drain on the government budget posed by the nonperforming assets of the financial institutions was an immediate concern of both the IMF and the World Bank, and became a qualitative performance requirement in the program that the new government negotiated with the IMF. The Philippine government developed a rationalization program for DBP and PNB in 1986, with the assistance of the World Bank and support from a $150 million economic recovery loan. Under the program, both institutions transferred their nonperforming assets to the national government, which also assumed the corresponding liabilities. The book value of the assets transferred came to P. 85 million for DBP ($4.2 million) and P. 72 billion ($3.6 billion) for PNB. Both institutions made substantial cuts in staff to reduce costs, and PNB closed most of its overseas branches. The reform measures that the Philippines adopted with respect to PNB and DBP were designed to place both institutions on an equal footing with

571

Philippines/Chapter 8

private commercial banks. The obligations of the two institutions would no longer carry government guarantees nor would the two have access to special credit or tax privileges. Public sector deposits would no longer be available on the scale or at the low cost they previously enjoyed and, just like private commercial banks, both would be subject to external audit. In addition, the government agreed that social projects with below market rates of return would be funded through explicit budgetary support, rather than at government behest through its financial institutions. The transfer of the assets of PNB and DBP to the national government and the government's assumption of the liabilities of the two institutions greatly reduced the share of the government in the domestic financial market.'' Under the restructuring program, the government pledged to operate PNB as an ordinary commercial bank, eventually offering shares for sale to the private sector as its profitability improved. The emphasis of DBP was to shift from large-scale, industrial sector loans to the provision of credit for agriculture, housing, and small and medium-sized firms. The rationalization programs were largely successful; by 1988 both institutions were profitable. The Philippine government also pledged to sell off the six commercial banks that it had acquired. Privatization

One of President Aquino's campaign pledges was to return the corporations acquired by the government to the private sector and to sell off government corporations that duplicated private sector activities. The new government established an interagency Committee on Privatization headed by the finance minister and, as an implementing arm,the Asset Privatization Trust (APT) headed by David Sycip, a prominent local businessman. The APT was given a five-year timetable to dispose of the acquired assets; all sales were to be done in cash, with bids collected through auction. The privatization program in the Philippine has proceeded slowly and has revealed deep ideological splits within the government. There has been widespread dissatisfaction among investors over the pace of the program, and recurring disagreements within the government over the assets that have been included or left out of the sales list. There has also been no single agency responsible for carrying out privatization. The APT has been responsible for the sale of nonperforming assets transferred to it, and these have come almost entirely from the two major government financial institutions. In contrast, the reorganization and privatization of nonfinancial government corporations has been left to the corporations themselves, and many have been reluctant participants. At the end of 1987, the APT had accepted bids for over forty assets for a total of P. 3 . 3 billion ($160 million).20 In these sales, the APT has been able to realize about 20 percent of book value.2' But as yet none of the government corporations slated for privatization has been sold, and the status of many of the most attractive

572

Robert S. Dohner and Ponciano Intal, Jr.

assets that the government holds is still uncertain. Despite Aquino’s reaffirmation of her government’s commitment to proceed with its privatization program, the program remains in doubt.” Trade and Industrial Policies

The Marcos government had initiated a tariff reform and trade liberalization program in the early 1980s under a structural adjustment program with the World Bank. The tariff reductions were successfully completed by 1985, but little progress was made in reducing quantitative restrictions and licensing requirements. Both the tariff reform and trade liberalization programs were superceded by the balance of payments crisis in 1983, as tariff surcharges were applied and the central bank imposed foreign exchange allocation. Both the tariff surcharge and the central bank allocation of foreign exchange were removed under the IMF program in 1985. At the end of 1985, the Philippines adopted a second program of trade liberalization, covering 1,232 items subject to import licensing, but this program was interrupted by the presidential elections and by the change of government in early 1986. Trade liberalization was among the policy proposals contained in the yellow book and was a priority of both the IMF and the World Bank in negotiations with the Philippines in 1986. Within the cabinet, trade liberalization was pushed strongly by Planning Minister Monsod. However, trade liberalization quickly became the most controversial economic policy issue facing the Aquino government. Liberalization was strongly resisted by industrial groups, represented in the cabinet by the minister of trade and industry, Jose Concepcion, as well as by several leftist groups. The government decided to continue with the 1985 program of trade liberalization, replacing the import licensing requirements with tariffs of equal protective effect. By April 1988 all 1,232 items had been liberalized, and the government was committed to liberalizing an additional 104 items by mid- 1989. On the remaining items subject to licensing, the government decided to retain limits on 114 for health and security reasons, and deferred decision on the remaining 455 items subject to quantitative restraint. This last list contains many of the most sensitive products-finished consumer durables and industrial intermediates-and the IMF has continued to press for their liberalization. The Marcos government also revamped the industrial incentive system in 1983 as a part of the structural adjustment program. The revised investment incentives made the encouragement of exports a priority, deemphasized the measured capacity concept that had been used to define overcrowded industries, and simplified and reduced the number of incentives. Investment incentives were replaced with performance-based incentives covering profits, wages paid, and the extent of domestic procurement of inputs. As a result, the capital-cheapening effect of the incentive system was greatly r e d ~ c e d . ’ ~

573

PhilippineKhapter 8

The Aquino government, after much internal debate and delay, finally issued its own set of investment incentives in the Omnibus Investment Code of July 1987. Most of the incentives system remained unchanged from the 1983 revisions. The most significant modification was the replacement of performance-based incentives on net local content and net value added with income tax holidays. In addition, the government appears to have increased the scope for discretion in administering the incentives by reviving the notion of measured capacity and reinstituting an investment priorities plan. The nature of the incentive program is still under discussion with the World Bank, and a further study of the incentive system is one of the conditions of the Bank’s economic recovery loan to the Philippines. The policy reform momentum in the Philippines has, at least for the time being, waned, and policy debates over the next few years are likely to involve holding actions on reform measures that have already been taken. The Aquino government has moved much farther in some areas than in others. The fewest results have been achieved in the highly visible privatization program, which became for many investors a barometric indicator of the attitudes of the Aquino government. The Aquino government has had more success in policies establishing the economic framework, or environment, in the Philippines. And despite varying success, one should not understate the degree of reform that has been achieved. Government intervention in many domestic markets has been greatly reduced or eliminated, monopolies and exclusive privilege have been abolished, and publicly owned firms are now on a footing similar to that of private companies. A comprehensive tax reform has been undertaken in a very short period of time. And the trade and industrial incentive reforms to which the Marcos government had committed itself have been taken up and largely completed. In the 1980s the Philippines has gone from being the ASEAN country with the most stringent import protection to a country with restrictions comparable to other East Asian countries and low by international standards, as we point out in table 8. l . A recent World Bank mission concluded that “as a result of cumulative policy adjustments since 1980, the Philippine regulatory and incentive structure has fewer distortions than at any time since 1950, and is now comparably neutral with other East Asian countries’ ’.24

Land Reform There is one further policy issue that is important both for the economic development of the Philippines and for its political stability, and that is the issue of land reform. Land reform has long been an issue in Philippine politics. In the period since independence there have been several efforts at reform, mostly confined to legislation and the creation of agencies responsible for land reform efforts. But land reform has largely been undertaken for political reasons, either to mobilize rural support or to diffuse

574

Robert S. Dohner and Ponciano Intal, Jr. Comparative Trade Protection, East Asian Countries

Table 8.1

Average Nominal Tariff

Percent of Import Items Restricted

43

37 17

Philippines 1980 1986

29

Indonesia 1980 1987

28 23

na

25

31

22

12

I6

Korea 1980 1985

Malaysia 1980

I985

12 14

<5

31 34

<5

<5

Thailand 1980 1985

<5

Source: World Bank (1987), vol. 2, p. 54, table 2.5

rural-based conflict, and once those aims were fulfilled reform quickly subsided, being neither actively funded nor strongly pushed from the center. A case in point is the limited, although politically successful, land reform and resettlement efforts during the Magsaysay administration in the late 1950s that was instrumental in sapping the Huk rebellion, a communist guerilla movement that grew out of the resistance to the Japanese occupation. Ironically, the most extensive, although still quite limited, land reform was carried out at the beginning of martial law, covering rice and corn lands. As on previous occasions, land reform had its political objectives: rallying rural support for Marcos’ New Society and threatening the existing elite. Eventually, however, land reform was used as a way of enriching presidential associates and Marcos himself, for the martial law period also saw the most extensive land seizure and extralegal manipulation of land claims in Philippine history.25 Economic factors in the 1970s and 1980s have added to the pressure of rural land tenancy and, increasingly, of rural landlessness. Up until the mid-l960s, a growing population was absorbed by an extension of land under cultivation, particularly with resettlement in the southern, and Moslem, island of Mindanao. However, the Philippines reached the limit of further land expansion, and land densities have grown steadily since that time. The period has also seen some shifts within the agricultural sector to high value crops, generally crops that require less labor input. In addition, the development of high yielding rice strains in the Philippines put greater emphasis on non-labor inputs. Add to this the fall in world prices of Philippine commodities, the monopolization and increasing effective taxa-

575

Philippines/Chapter 8

tion of the agricultural sector, and the failure of the manufacturing sector to generate employment opportunities, and the result was increasing rural poverty and landlessness, as well as a dropping urban real wage. Poverty, the abuses of the military in local areas, and the ceding of large tracts of land to cronies or multinational exporters, greatly increased the strength of the NPA in the rural areas and the importance of agrarian reform as a domestic political and security issue. Land reform held a central part in President Aquino’s campaign and government platforms and is a responsibility set down in the constitution that was drafted in 1986. However, very little was done on agrarian reform in her first year in office until a protest by peasants outside Malacanang Palace in January 1987, in which police fired on and killed several protesters, brought the land reform issue immediately to the fore. An interagency group was created to develop a program and, after opposition within the president’s office watered down several drafts, the Comprehensive Agrarian Reform Program (CARP) was released in late May. The program would start by completing the agrarian reform Marcos had started in rice and corn lands, followed by reform of government-owned and sequestered land, before going on to redistribution of sugar and coconut lands. Landlords would be paid the market value of their land, 10 percent in cash and the remainder in government bonds, while tenants would purchase the land over a thirty-year period. Aquino did not implement the land reform program by executive decree when she had the opportunity, but instead passed the program on to Congress to determine the maximum area that current landlords could retain and the timing of the program. After much debate, the Philippine Congress passed the CARP, which President Aquino signed on 10 June 1988. The CARP extends over ten years. Current landowners are allowed to retain five hectares of land, plus three hectares for each child over fifteen years of age. The program is expected to cost P. 332 billion ($15.8 billion) over its ten-year life, of which the bulk is for support services and infrastructure. Implementation is likely to be a difficult task. In addition, three-quarters of all privately owned agricultural land is in plots of less than ten hectares and likely to be excluded from CARP. While the effect of CARP on rural poverty and landlessness is likely to be modest, the political benefits from implementing the program could be huge. 8.2.3 Debt Renegotiation Economic recovery and policy reforms were the first two challenges facing the new Philippine government; the remaining one was debt renegotiation. In contrast to relations with multilateral institutions, the change of government brought more politically charged and acrimonious relations with the Philippines’ commercial bank creditors. The conflicts with the banks have centered

576

Robert S. Dohner and Ponciano Intal, Jr.

on the terms of Philippine rescheduling and debt service, and the status of loans made by the banks to the previous government. Many of the loans made to the Marcos government had been to support public investment projects, or privately organized but government-favored projects, that were either of questionable viability or involved substantial misappropriation of funds. The suspicion of many in the new government was that the banks were cognizant of the uses to which the funds were being put and, therefore, bore some of the responsibility for questionable and now nonperforming loans made to the previous government. The calls for case-by-case evaluation of commercial bank loans, with repudiation in egregious cases, came almost immediately after the formation of the Aquino government and were most closely associated with the planning minister, Monsod. The finance minister, Ongpin, and the central bank governor, Fernandez, were opposed to any repudiation, and they won the argument within the government in 1986. President Aquino announced in mid-year that the Philippines would honor its foreign obligations, even in cases of questionable loans. But the bitterest bone of contention between the Philippines and its creditor banks concerned the terms of the second rescheduling. Philippine negotiators were strongly influenced by the outcome of the bank negotiations with Mexico in September 1986 that had provided $6 billion in new money, an interest spread over LIBOR of percentage points, and a contingency fund of $500 million if Mexican growth fell short of targets. Reacting to the Mexican outcome, the Philippines came to the negotiations in October with the demand that the Philippines be granted an interest rate spread of over LIBOR. The Philippine negotiators had two arguments for the reduced spread. The first was that the terms of the initial rescheduling had been deliberately harsh in order to punish the Marcos administration. The Aquino government, they argued, was committed to the refoms that the external creditors had long stressed and deserved a better program. The second argument was that the Philippines had successfully followed its adjustment program, eliminating its current account deficit, inflation, and external arrears. The Philippines did not need and was not asking for additional new money and, the negotiators argued, the country deserved better terms than Latin American debtors, who had been less successful in adjusting. Most of the members of the bank advisory committee were willing to reduce the spreads of the original rescheduling agreement, but Citibank was initially adamant against any reduction in interest rates on rescheduled Philippine debt.26 The negotiations with the bank advisory committee broke down in November over this issue, but resumed in January when Citibank agreed to the negotiating position of the other advisory committee banks. The banks and the Philippine negotiators still remained divided over terms when the negotiations resumed. As a compromise measure, Ongpin proposed

2

577

PhilippinedChapter 8

that the banks receive a higher interest rate spread if they were willing to take the spread in the form of Philippine Investment Notes (PINs). These notes were dollar denominated and sold at discount, but their terms (six-year maturity, no interest payment) made their conversion into pesos for participation in the Philippine debt equity conversion scheme particularly attractive. The initial proposal was rejected by the advisory committee, but a revised version, making the acceptance of PINs optional on the part of the banks and guaranteeing a return of percent, was accepted and formed a part of the second Philippine rescheduling.27 The Philippine negotiators and the bank advisory committee reached agreement on a multiyear rescheduling on 27 March 1987 (table 8.2). The agreement restructured $3.6 billion in debt falling due between 1987 and 1992, as well as the $5.9 billion previously restructured in the 1985 agreement. The amortization period on restructured debt was lengthened to seventeen years, with seven and one-half years' grace. The trade facility, which previously had been extended to mid-1987, was further extended to June 1991. There was no additional new money in the 1987 agreement. The $925 million new money from the previous agreement was not rescheduled, but the facility was repriced along with the restructured amounts. In the agreement, the Philippine negotiators settled for an interest rate spread that was slightly higher than the Mexicans had received in their negotiations in 1986, as well as an arrangement that allowed commercial banks to book higher earnings through the acceptance of PINs in lieu of

5

Table 8.2

Philippine Debt Restructuring Agreements

Commercial Banks

Second Round, July 1987

($ billion)

Amount

Maturity (grace)

Previously restructured

5.88

17(7.5yrs)

Debt falling due New money First restructuring Second Revolving trade credit Total

3.58

17(7.5yrs)

Interest Spread

Amount ($ billion)

5.88 71

First round (20 December 1984) Second round (22 January 1987)

0.925

8

0.00 2.97 12.26

Official Creditors

3

Amount

Maturity (grace)

Interest Spread

10(5yrs)

12

Wyrs)

1:

I!

2.97 9.69

4

Amount

$ Billion

%

$ Billion

%

Period

0.725 0.704

100 100

0.258 0.289

60

Jan 85/June 86 Jan 87/June 88

Source: Central bank, Financial Plan Data Center. 'Repricing.

First Round, May 1985

70

578

Robert S. Dohner and Ponciano Intal, Jr.

cash. The interest rate spread on restructured debt, the trade facility, and the 1985 new money agreement was seven-eighths over LIBOR.28 The negotiators also agreed that if the Philippines were to later ask for new money, it would also seek to reschedule the 1985 new money facility, accepting a 1 percent spread. The agreement on the interest rate was, however, made retroactive to 1 January 1987. The 1987 agreement significantly extended the period for amortizing principal on Philippine external debt and reduced the interest spread by three-quarters of a point. As a result, the debt service ratio for the Philippines was reduced from 48 percent before rescheduling in 1987 to 34 percent after. However, the failure of the negotiators to achieve an interest rate spread as low or lower than that received by the Mexicans was a stumbling block to approval of the agreement in the Philippine cabinet. The terms became a major issue in the Philippines when, shortly after the agreement was concluded, Argentina reached an agreement with its commercial creditors for an interest rate spread of the same as the Mexicans had received and below that of the Philippines. The results of the Argentine negotiations were greeted in the Philippines with dismay. Finance Minister Ongpin declared his intention to reopen the negotiations with the banks over the issue of the interest rate spread, but was eventually dissuaded by bank promises to take more interest in the form of PINS. The issue of the interest rate spread and the treatment of the Philippines relative to other LDC creditors remained the source of much domestic opposition to the 1987 agreement and a focal point of domestic opposition to Ongpin and Fernandez. But the arguement over spreads was almost overshadowed by the reemergence of the Planters Products (PPl) issue. Barclays Bank, a major PPI creditor and a member of the bank advisory committee, refused to sign the rescheduling agreement worked out in April, and renewed the demand that the Philippine government assume the $57 million dollar external liability of the fertilizer company. The Philippine negotiators first tried to work out a compromise but in the end President Aquino agreed to assume PPI’s obligations to rescue the agreement and preserve its retroactivity to January 1. The PPI compromise and the whole debt rescheduling package became a major domestic political issue when, in an address to the newly assembled Philippine Congress in July, President Aquino charged that the banks had blackmailed the country into the agreement. In the months that followed, the external debt became the principal issue of the Philippine Senate. Bills were introduced to repudiate part of the debt and to limit debt service to a portion of foreign exchange receipts. Ongpin and Fernandez were called to testify before Congress to explain and justify the terms of the rescheduling agreement.

2,

579

PhilippineKhapter 8

8.3 Economic Recovery After more than two years of falling output, the Philippine economy began to recover in 1986. For the year as a whole, GNP growth was 2.0 percent, insufficient to prevent a third year of declining per capita income. In 1987, real output grew by almost 6 percent, which was below the official target of 6.5 percent but above most private forecasts. The recovery in 1986 was led by exports and by increased government expenditure, the product of election spending early in the year and the subsequent agreement with the IMF to allow a larger budget deficit for 1986. Consumption demand was weak, as households rebuilt private savings, and investment dropped by an additional 15 percent to a level 60 percent below that of 1983. The strongest growth was in the agricultural sector. The increase in world prices for copra and coconut oil late in the year, the lifting of monopoly restrictions in the sector, and the end to the ban on copra exports resulted in a doubling of farmgate prices for coconuts. This greatly increased rural incomes and sales of commodities in rural areas. Export earnings rose by almost 5 percent during the year, with large volume increases of coconut products, fish, and prawns. The recovery strengthened in 1987, with growth for the year estimated at 5.9 percent. Government expenditure rose more strongly than in 1986 through increases in public sector wages and a gearing up of the employment generation program in rural areas. Consumption demand rose by 5.5 percent with a recovery in the demand for durable goods. For the first time since 1983, there was growth in the industrial sector, where output increased by 8 percent, based largely on sales to the domestic market. Two particularly welcome signs were the growth in exports and the recovery in domestic investment. Merchandise export earnings in 1987 totalled $5.7 billion, up by 18 percent from 1986. Particularly strong growth came from electronics exports, as the world market recovered; garments, as the United States, Canada, and the EC increased their bilateral import quotas from the Philippines; and fresh fish, as agricultural producers in the sugar areas diversified into prawn production. Domestic investment grew 28 percent in 1987 after three successive years of sharp declines. Most of the increase in investment in 1987 came from domestic sources; foreign investment increased, but still remained below the dollar levels of 1981 through 1983. The economic growth rate continued to accelerate in 1988, with growth estimated at 7.5 percent for the first three quarters, until typhoons late in the year reduced growth for 1988 as a whole to 6.9 percent. Investment provided the momentum in 1988, rising 26 percent in real terms. Foreign investment increased particularly rapidly, and foreign investment inflows increased by 120 percent in the first eight months of the year.

580

Robert S. Dohner and Ponciano Intal, Jr.

8.4 Prospects for the Philippines Although the Philippines has made a strong start at economic recovery and policy reform, sustained economic growth is by no means assured. Difficult challenges for the Philippines remain, both political and economic. Much of the early attention of the government was directed to its own survival. The coup attempt organized by military officers in August 1987 was extremely serious and nearly succeeded in toppling the government. In the period since the attempted coup, the situation improved considerably. The coup plotters were captured and jailed, and Aquino has taken actions, described above, to more closely bind the military to the civilian government. Although the threat of an overthrow has not disappeared, time is now on Aquino’s side. Congress has been established and its members are now taking a more active role in the policy process. Political parties are reforming and jockeying with each other for position. The passage of time has served to focus ambition on the country’s democratic institutions, and the next presidential election in 1992, which President Aquino has said she will not contest, is no longer so far away. A less immediate, but ultimately more difficult challenge is that posed by the communist insurgency. The NPA has an estimated twenty to twenty-four thousand regular forces and constitutes the best military force in the country. In comparison, the Armed Forces of the Philippines has a combat strength of about fifty-five thousand, although many units suffer from poor training, support, and morale (Kessler 1988, 10). Rural inequality has nurtured the NPA, and the defeat of the guerrillas will require that the government carry out both a rural development program and a land reform program that address questions of economic inequality in the rural areas. Recent arrests of NPA leaders have weakened the movement, but its hold on many areas of the Philippines remains. 8.4.1

Economic Prospects

The challenge for the Philippines is the same as it has always been, achieving and sustaining rapid economic growth in the longer term. In the last part of the 1980s and the 1990s this will be a much more difficult task than in earlier decades, since the Philippines now has the burden of servicing a much larger external debt, and because obtaining external capital will be much harder than before. The necessity of achieving economic growth is particularly acute because of the momentum of population growth in the Philippines, at 2.5-2.7 percent per year, the highest in the region. The post-1983 stabilization program in the Philippines cut per capita incomes, effectively wiping out the gains from the high growth, heavy foreign borrowing period. The current Philippine government has an ambitious target of 6.5 percent growth per year, which would return the

581

PhilippinesKhapter 8

Philippines to 1983 income levels by 1992. Those outside the government, including the IMF, have adopted more modest, but still optimistic, projections of 6 percent real growth per year. Achieving even this rate of growth is by no means certain and will involve policy reform and perseverance, external cooperation, and some degree of luck. The most obvious difficulties that the Philippines will face are the burden of servicing its external debt and limitations on its ability to borrow to achieve higher growth. The recent official and commercial bank reschedulings resulted in a significant reduction of scheduled amortization payments on external debt, as shown in table 8.3. However, projected interest payments still amount to 6 percent of national income, and the Philippines faces the burden of meeting the remaining amortization payments over the next five years. The debt crisis affects the Philippines and other LDCs not just through the difficulty in shouldering the current debt burden; it also affects their ability to grow out from under that burden. The severely limited access to additional external finance is for most of these countries a constraint on the rate of investment and growth. Difficulties in financing an emerging current account deficit did not affect the Philippines immediately, primarily because of the unexpectedly large current account surplus in 1986. But the country has now reached a point where the external financing constraint is a real one, and additional external funds are necessary to sustain growth. The difficulty lies in the way in which the current account deficit was closed during the stabilization period. Table 8.3

Philippines Projected Debt Service, Government Projections (in billions of U.S. dollars)

Debt service before rescheduling Principal Interest Debt service after rescheduling Principal of which: IMF Interest Interest as % of GNP Debt service ratio" Before rescheduling After rescheduling Memo Items: GNP (billion S) Exports goods and services

1987

1988

1989

I990

1991

1992

4.40 2.27 2.13 2.94 0.98 0.31 1.96 6.0

5.44 3.13 2.31 3.03 0.88 0.16 2.15 6.0

4.59 2.27 2.32 3.01 0.82 0.18 2.18 5.5

5.22 2.80 2.41 3.43

5.43 2.85 2.58 3.70 1.25 0.31 2.45 5.0

5.17 2.49 2.68 3.66 1.07 0.15 2.58 4.7

1.13

0.33 2.30 5.2

49.4 33.0

56.9 31.7

44.7 29.3

46.5 30.6

44.4 30.3

38.1 26.9

32.79 8.90

36.05 9.55

39.94 10.27

44.38 11.22

49.21 12.23

54.41 13.57

Source: Central bank. Notes: Reschedulings include monetary and nonmonetary debt. Interest rescheduled by the Paris Club.

"Ratio of interest and principal payments to exports of goods and services.

582

Robert S. Dohner and Ponciano Intal, Jr.

Between 1983 and 1986, Philippine merchandise export earnings fell by 3 percent; thus, the entire closure of the current account deficit was due to a reduction in imports. After years of import substitution, Philippine imports are heavily concentrated in capital goods and imports of fuels and raw materials; only about 16 percent of the country’s imports are made up of consumer goods (table 8.4). The Philippines has been relatively successful in reducing its dependency on imported oil, substituting coal and other sources of domestic energy. But over a medium-tern planning range, the demand for oil remains quite price inelastic, and the oil import bill is largely determined by swings in the world price of oil. Two of the other categories are closely tied to domestic economic activity-capital goods imports to the level of domestic investment and raw materials imports to industrial output. The behavior of each of these import components during the stabilization episode is shown in table 8.5. Capital goods imports as a percentage of GNP dropped rapidly, reflecting the sharp reduction in domestic fixed investment that took place during the period. Although the drop in the share of industrial output in GNP was not huge, those industries that depended on imported raw materials were particularly hard hit. Finally, slack domestic demand, plus the weakness in world oil prices, was responsible for the drop in oil imports. In contrast, consumer goods imports remained constant at about 5 percent of GNP, despite their being low priority imports. Table 8.4

Philippine Import Composition (percentage)

Capital goods Mineral Fuels Nonoil raw materials Consumer goods

1970-75

1976-80

35.7 13.1 36.4 14.7

29.1 24.7 29.4 16.9

~

Source NEDA, Philippine Staristical Yearbook, 1987

Table 8.5

Philippine Imports by End-Use Category (percentage of GNP)

Capital goods Nonoil raw materials Mineral fuels Consumer goods Total Memo Items (% of GNP): Exports Fixed investment Industrial output

1981

1982

1983

1984

1985

1986

1987

5.0 5 .O 6.4 4.3 20.7

4.5 5.3 5.4 4.4 19.5

5.0 5.7 6.2 5.0 21.9

3.6 4.6 5.2 5.8 19.3

2.5 4.0 4.5 4.9 15.9

2.8 5.5 2.9 5.4 16.6

3.5 7.0 3.6 5.5 19.6

14.9 26.1 36.8

12.8 25.7 36.5

14.7 25.1 36.4

17.1 20.1 35.4

14.4 15.1 33.6

15.9 13.0 32.6

16.7 14.0 32.6

Source: NEDA, Philippine Staristical Yearbook. 1987.

583

PhilippinesXhapter 8

The fact that exports did not contribute to closing the current account deficit and that the reduction in imports took place in capital and intermediate goods explains why the current account balance has worsened sharply as the economy recovered. Even very large rates of growth of exports are unlikely to change this outcome. The reason for this is twofold. The categories of exports most likely to increase rapidly, electronics and garment exports, have imported input contents of 65-80 percent, and thus their expansion will lead to a partially offsetting rise in imports. Other categories of exports that are candidates for expansion will require substantial new investments, raising, at least for a time, imports of capital goods. Nor is there likely to be much help from the service sector. The Philippines is unusual among the LDCs studied in this NBER research project in having substantial service export earnings. Despite net interest payments amounting to 6 percent of GNP, the country ran a small surplus on service account in 1985 and 1986.29 Net service receipts are expected to diminish and become negative by 1992 as the growth of the economy raises service payments and because the prospect of weak growth in Middle Eastern countries limits expected earnings from construction and Filipino labor overseas. As a result, medium-term projections for the Philippines all give current account deficits of 2-2.6 percent of GNP. Current account projections prepared by the Philippine government and the IMF are shown in table 8.6. The government projection assumes an economic growth rate of 6.5 percent per year between 1987 and 1992, and thus a more rapid rate of growth of imports and exports than does the IMF projection which assumes a domestic growth rate of 6 percent per year. Even with growth of export earnings of 13-14 percent per year, the current account deficit rises to 2.6 percent of GNP by the end of the forecast period. The current account is not the only source of a need for external finance. In addition, the Philippines will have to either make or reschedule amortization payments on its external debt falling due, plus achieve a balance of payments surplus sufficient to allow reserves to grow along with imports. In some cases, financial commitments have already been identified, either in the form of loans that are being drawn down in tranches, commitments for external aid, or rescheduling that has already been agreed upon. But even with these three, there remains required finance that has not as yet been identified. Projections of future financing requirements are less forecasts than they are planning exercises. One such projection for the Philippines is shown in table 8.7. For table 8.7 we use the current account projections of table 8.6, along with scheduled amortization payments on medium- and long-term external debt before agreed upon rescheduling. The overall balance of payments is assumed large enough to meet servicing requirements on

584

Robert S. Dohner and Ponciano Intal, Jr.

Table 8.6

Philippine Current Account Projections, 1987-92 (in billions of U.S. dollars)" 1987

1988

1989

1990

1991

1992

- 1.4

Philippine government Trade balance Exports Imports Services and transfers Interest payments Current account Percent of GNP

- 1.0

- 1.2

- 1.5

6.8 -8.0 0.5 -2.4 -0.7 - 2.0

-1.4 7.7 -9.1 0.6 - 2.6 -0.9 -2.1

- 1.5

5.7 -6.7 0.5 -2.2 -0.5 -1.6

8.7 - 10.2 0.4 -2.8 - 1.2 -2.6

9.8 -11.3 0.2 - 3.0 -1.2 -2.5

IMF Trade balance Exports Imports Services and transfers Interest payments Current account Percent of GNP

-1.0 5.7 -6.7 0.5 -2.2 -0.5 - 1.6

- 1.1

- 1.4

- 1.5

- 1.5

6.6 -7.7 0.5 -2.3 -0.5 - 1.4

7.2 -8.6 0.4 - 2.3 - 1.0 -2.5

8.2 -9.6 0.2 -2.5 - 1.2 -2.8

9.3 - 10.8 0.0 -2.7 - 1.4 -3.0

11.2 - 12.6

1987-92b

14.3 13.3

0.0 -3.1 - 1.4 -2.6 -1.4 10.5 -11.9 0.0 -2.8 - 1.4 -2.7

13.0 12.1

Source: Philippine projections are from the central bank. IMF projections are from IMF (19881, p. 27, table 11.

"The central bank projections were prepared in November 1988, while the IMF projections were prepared in May 1988. 'Average growth rate Table 8.7

Philippines Balance of Payments Finance Projections (in millions of U.S. dollars) 1988

1989

1990

1991

I992

- 743 - 1,854

- 1,402

- 593

687 2,541 617 395 134 1,451 619 832 0

-1,151 -874 1,512 2.386 376 350 -662 1,961 993 968 446

- 1,019

98 I 2,195 205 356 - 265 1,456 599 857 0

-874 -778 1,684 2,462 537 324 --1,972 2,763 1,883 880 1,293

- 1,240

- 1,214

1,565 2,584 326 363 - 175 2,345 1,439

1,620 2.213 330 34 I - 660 1.944 1.230 714 1,230

1987 Current account Medium- & long-term loans Inflow outflowa Direct investment (net) Other items (net)' Net international reserves' Rescheduling Paris Club Commercial banks Financing gap

- 539

906

925

Source: Central bank.

"Before rescheduling. bSbort-tenn capital, monetization of gold, revaluation, and errors and omissions. 'Minus sign indicates an increment.

monetary debt and to allow reserves to grow sufficiently to maintain a level equivalent to three months of imports. Current account financing makes up the largest component of the demand for additional finance, but not the only significant component; the financing gap in this projection, even after likely rescheduling by official (Paris Club) creditors, averages just under $1 billion per year between 1989 and 1992.

585

PhilippineKhapter 8

In early 1989 the Philippines has just begun its third round of foreign debt rescheduling. An agreement has been reached with the IMF for a $1.3 billion extended finance facility. As a part of this program the Philippines has agreed to reduce the national government’s budget deficit through a reduction of subsidies and improved tax collections. The country has now begun negotiations with its official and commercial bank creditors. The Aquino government is asking for new money in this round and has proposed a figure of $1.6 billion from the commercial banks. The commercial bank negotiations are likely to be difficult this time, just as they were in 1987. This is due not only to the increasing reluctance of commercial banks to provide new money in recent debt rescheduling, but also because the funding available to the Philippines from the Multilateral Assistance Initiative, Japanese foreign aid, and a possible renegotiation of the Military Bases Agreement with the United States are all uncertain. Finally, and ironically, the recent U.S. proposals on LDC debt made by Treasury Secretary Brady have increased the reluctance of commercial banks to provide additional funds before the status of their existing loans under the proposals are determined.30 For its part, the Philippines has decided to force the issue by proposing a substantial increase in foreign exchange reserves in 1989, requiring an immediate need for new money.

8.4.2 Export Prospects Achieving sustained economic growth and at the same time limiting the current account deficit will require a substantial rise in Philippine exports. But financing of import growth is not the only role that exports will play in the Philippine recovery. Over the course of the stabilization period, both domestic investment rates and domestic saving rates declined. Restoring a higher rate of domestic investment will require a substantial increase in private and public saving rates in the Philippines, and this in turn will mean that the domestic market will grow less rapidly than output and income. To make all of this possible, exports must be a leading sector in Philippine economic expansion over the near to medium term. A final reason for the importance of exports in Philippine recovery and growth is the lack of success the country has had in achieving growth through production for the domestic market. Rates of employment growth and productivity growth in import-substituting industries have been distressingly low, and, at least for manufactured goods, export industries have provided the only source of dynamism in the last ten years. As we have pointed out in previous chapters, the structure of Philippine exports changed dramatically in the 1970s. Traditional primary commodity exports-sugar, coconuts, logs and lumber, and minerals-were responsible for about 90 percent of export earnings in 1970, but only 50 percent by the end of the decade. Although declining in importance, these commodity exports still maintained substantial influence over total export revenue.

586

Robert S. Dohner and Ponciano Intal, Jr.

Slumping international commodity prices and declining Philippine export volumes of these primary commodities were responsible for a 15 percent decline in the dollar value of merchandise exports from 1981 to 1986, despite a continued growth in nontraditional export volumes and earnings. The influence of primary commodities in total Philippine exports has been greatly reduced to a current share of about one-quarter, although they still account for a larger share of net export earnings due to their higher domestic content. For a variety of reasons, the prospects for sustained growth in Philippine primary export earnings are not good, although some near term growth can be expected as export taxes and the systems of monopoly regulation are removed. In the sugar industry, Philippine costs of production are estimated to be about 12 cents per pound, compared to a current world price of about 8 cents per pound. The Philippines benefitted up to 1984 from favorable long-term supply contracts it had negotiated when sugar prices were high. Since those contracts have expired, Philippine sugar exports have gone entirely to the United States under the U.S. sugar import quota, where sales prices have been about 18 cents per pound.31 Thus even a substantial recovery of world sugar prices would not significantly increase Philippine export earnings. The prospects for future sales to the United States are also bleak; the U.S. has reduced overall sugar import levels substantially since quotas were reintroduced in 1982 and announced substantial reductions for 1988. For the Philippines this meant a reduction in quota from 240,000 tons in 1987 to 148,000 in 1988. The U.S. drought in 1988 brought an increase in sugar imports late in the year, but the longer term outlook remains unchanged. In the past two decades the Philippines has increased its output of coconut products by extending the area of cultivation. Limits on cultivatable land have been reached, and increasing output now depends on increasing yields. Philippine coconut production has been hampered by low yields and aging trees, and total coconut production has fallen well below the peaks reached between 1978 and 1981. A program of replanting higher yield varieties would increase coconut output, but only after a gestation period of several years. In addition, since the Philippines accounts for about two-thirds of world exports of coconuts, and coconut products face competition from other vegetable oils, the prospects for increasing earnings through additional exports are quite low. Philippine mining exports have some prospect for growth if world copper prices remain high. Philippine production costs are slightly above world market averages, making the Philippines a marginal supplier although much of the country’s mining capacity closed during the financial crisis. Gold is a byproduct of copper mining in the Philippines, and higher world gold prices have already raised mining earnings. The prospects for expanding manufactured goods exports from the Philippines are brighter, although not assured. Philippine manufactured

587

Philippines/Chapter 8

goods exports are dominated by two industries, electronic components and garments, which together account for 72 percent of nontraditional exports and 53 percent of total exports. As a result of the protection given to Philippine domestic industry, both the electronics and the garment industries have developed as virtual enclaves, largely foreign owned, dependent upon imported inputs, and with few linkages to the Philippine economy other than the purchase of labor. Exports of electronics components declined with the slump in the world semiconductor market in 1985 and 1986, but increased strongly in 1987 and 1988. Philippine electronics exports are diversified, with the largest share going to other ASEAN countries, but future growth will depend on the growth of the world market and the continued maintenance of competitive real wages. Philippine garment exports go almost exclusively to the United States, and thus the prospects for growth depend considerably on the degree of protection provided to the American industry. Despite this fact, Philippine garment exports have the potential for substantial growth. The Philippines has not been as aggressive as other countries in increasing its bilateral quotas, although it did receive higher U.S. and European quotas in 1987. More importantly, the Philippines has not taken full advantage of the quotas that it has. Quota utilization by the Philippines was the lowest among ASEAN countries in the 1980s and fell below 50 percent in 1984 and 1985.32 Other factors work in favor of the Philippines in the longer term. The country has a small share of world exports of textiles and clothing, and both its product composition and export markets are highly concentrated, leaving open the possibility for diversification. Finally, after years of falling real wages, the Philippines now has a sizable cost advantage over other garment exporters in the region; only China has a productivity adjusted wage rate lower than that of the phi lip pine^.^^ The Philippines also has the opportunity to increase net export earnings from the garment industry, and perhaps also from the electronics industry, by developing backward linkages to the domestic economy as have other countries in the East Asian region. Successfully substituting domestic for foreign inputs will require that the Philippines carry through with its intention to reduce protection of the domestic industrial sector, and will also require that incentives for this type of indirect exporting match those for sourcing components externally. For example, subsectors of the domestic textile industry, in particular spinning, knitted fabric, denim, and sewing thread, appear to be competitive on world markets if the textile industry were to have access to inputs at world prices.34 Diversifying export industries may be the most successful strategy that the Philippines can pursue in increasing exports over the next five to ten years. The prospects for developing new industries are especially promising in the rural areas, in resource-based or labor-intensive industries. Already the Philippines has seen dramatic increases in exports of prawns to Japan and

588

Robert S. Dohner and Ponciano Intal, Jr.

other East Asian markets, and, before the land reform controversy erupted in mid-1987, producers in sugar growing areas were making substantial investments in prawn beds. Other possibilities for export crop diversification include coffee, cacao, pineapple and other fruits and vegetables, black pepper, and ramie. To a greater extent than growth from existing industries, growth through export diversification will require a more aggressive and sustained array of policy incentives. Diversification of exports from the rural areas will require complementary inputs of capital and rural infrastructure, a settlement of the land tenure issue, and maintenance of security and order in several of the rural areas. Both agricultural and export diversification will also require a shift in the balance of production incentives to encourage export activities, including the maintenance of an appropriately valued exchange rate. 8.4.3 Financial Issues In addition to overcoming the external financing constraint, achieving sustained economic growth in the Philippines will require solving several outstanding domestic financial sector problems. Financial markets in the Philippines have been buffeted by a series of financial crises, starting in 1981 and extending through the stabilization episode. The burden of credit stringency fell largely on the private sector, and among firms that survived there has been a significant deterioration in financial ratios. By one estimate, nearly half of all firms are facing, or are likely to face, significant financial d i f f i ~ u l t i e s .These ~ ~ show up most often as problems in servicing existing loan obligations or difficulties in obtaining adequate working capital. Furthermore, until the financial ratios of these corporations improve, they are unlikely to be able to raise additional funds for investment or expansion as the Philippine economy recovers. Carrying out programs of financial restructuring and recapitalization for the affected firms will be necessary for them to participate in an expansion of the Philippine economy, and given the number of firms involved, these restructuring programs may be required before significant growth can take place. This presents a number of problems for the Philippines, since there has been little experience with financial restructurings of this kind, and since the available expertise is quite limited. The necessary expertise may be available through foreign technical assistance, but the restructurings may also require legal and procedural changes to write down the assets of current creditors and recapitalize the firms. The second and more general financial issue concerns the mobilization of domestic credit and the terms upon which it is made available to borrowers. The financial trauma of recent years in the Philippines has wiped out the progress that the Philippines made in financial deepening in the late 1970s and early 1980s. It has also erased what had been gradual progress toward extension of credit for longer terms; after several gyrations of interest rates,

589

PhilippineKhapter 8

banks have been reluctant to perform any term intermediation. Furthermore, although interest rates were freed from administrative control as part of the financial deregulation in the early 1980s, spreads between deposit and loan rates in the Philippines are very high, averaging 16 percentage points, compared to 3-4 percent for neighboring countries.36 Much of the spread between deposit and loan rates results from the implicit taxation that the Philippines imposes on financial intermediation through high reserve requirements, the gross receipts tax, and the requirement that 25 percent of bank lending go to agriculture or eligible government securities. The high taxation of bank intermediation and the resulting large spreads are issues that the Philippine government is currently studying and may soon address. To increase domestic savings made available through the financial system, and to lengthen the terms on which they are made available, the government must persuade the public that in the future there will be low and steady rates of inflation and safety in financial institutions. Monetary discipline and avoiding the use of central bank credit for fiscal purposes should achieve the first. The maintenance of the integrity of financial institutions will require closer and more effective supervision than the Philippine has so far known, more independent and professional management in many institutions, and perhaps larger and more diversified financial institutions. The long history that the Philippine has had of financial scandal and mismanagement must be laid to rest. These same prescriptions should encourage the development of both longer term lending and longer term deposits. In addition, increasing use of floating rate loans and the mobilization of assets from public and private insurance companies might also be used to encourage term credit. 8.4.4 Sustaining Growth

The Aquino government has made considerable strides in its first two years in office. The most egregious monopolies have been eliminated, curbs have been placed on government interventions and particularistic application of law, the system of industrial incentives and protection has been reshaped in more neutral form, and reforms have been undertaken in the tax and financial systems. The Philippines has already begun to reap some of the benefits of these actions, particularly in the agricultural sector, in an external environment that has recently turned more favorable. However, further actions will need to be taken, and the Philippines will have to establish a reputation for consistency in policy in order to sustain an economic recovery. Several more general challenges to policy are likely to appear in the coming years. The first, a waning of the momentum behind policy reform, is already beginning to occur. The window of opportunity presented by the indignation against the Marcos government has narrowed as the current government has become more established. This is most clearly seen in the flagging effort behind the government’s privatization program. Resistance to privatization and to trade liberalization has also grown recently due to a

590

Robert S. Dohner and Ponciano Intal, Jr.

revival of Philippine nationalist sentiment and a reluctance to open the economy up to foreigners. The Philippine Congress has already passed several bills raising tariffs, although these have been vetoed by President Aquino. Part of the problem for the Philippines is the ambiguity surrounding Marcos’ administration and his departure from the Philippines. There has been a tendency to attribute the poor performance of the country’s economy entirely to martial law and to cronyism, and to conclude that the end to the Marcos regime is sufficient to turn the economy around. There is also ambiguity as to the policy lessons that can be drawn from the martial law period. The rhetoric of the Marcos administration, and much of the public perception, was that this was an export promotion regime open to, and encouraging, foreign direct investment, and therefore that export promotion and international openness failed in the Philippines. In fact, as argued above, the Marcos government continued, and in many ways strengthened, the system of protection of the domestic manufacturing industry-the Philippines received the smallest gross inflows of foreign direct investment in the region and, at the same time, experienced substantial gross outflows, for the same reasons that domestic businessmen were moving their money overseas. Finally, and ironically, Marcos was a parvenu, an outsider who challenged the traditional political system in the Philippines, but a challenger whose administration went badly wrong. The “People’s Power Revolution” of February 1986 was in many ways a restoration of the former political order and institutions, with a strong possibility that the Philippines will resume the log-rolling patronage and obstruction to change that characterized the system before martial law. There was no consensus among the opposition to the Marcos government on the outlines of a broad economic policy, other than the desire to be rid of the most invidious of the institutions of martial law, and the battle over policy direction will continue to go on within the government and the country. The arguments that are marshalled, and in many cases the outcomes, will be influenced by external policy, particularly that of the United States, the multilateral institutions, and the major creditor banks. The Philippines is a small country, and politics and policy are to a large extent personalized. These external actors were quite conscious of their role in supporting the technocrats under martial law, and ultimately placed too much trust in their ideology and their influence. Perhaps because of the disappointments that flowed from reliance on the technocrats, the external actors have not always been cognizant of their role in strengthening or weakening certain positions within the current Philippine government. Nowhere has this been more apparent than in the 1987 commercial bank rescheduling exercise. Despite the fact that the Philippines had successfully stabilized the economy, deposed the man who was universally seen as the fundamental obstacle to reform, initiated substantial policy changes of the

591

PhilippinesXhapter 8

type that had long been argued for, and was not asking for new money, the banks gave the Philippine negotiators an extraordinarily difficult time. Their heavy-handed treatment of the Planters Product issue, holding the entire agreement hostage to the government's assumption of the firm's liabilities, their failure to come down further on the interest rate spread, and their subsequent delivery of the Philippine interest rate demand to the Argentines, led to the discrediting of Jaime Ongpin and Jose Fernandez, the introduction of a raft of bills in the Congress on debt limitation or repudiation, and the departure from the government of Ongpin, who had been their strongest ally. Much has been said about the difference in the interest rate spread between the Philippine and the Argentine rescheduling, amounting to only $5 million in annual interest payments for the Philippines. But the argument cuts both ways. The banks gained only $5 million, having squandered the precedent in the Argentine negotiation, and lost Jaime Ongpin. Corruption, or more generally, the generation and distribution of economic rents, has been an enduring feature of the Philippine economy. The Aquino government has not been immune to this charge. Two cabinet ministers were dismissed early in the administration after allegations that they were using their ministries for profit. Members of President Aquino's family, particularly her younger brother, have used their ties to her to increase their influence and wealth. There have been charges that the activities of the Presidential Commission on Good Government, set up to recover the assets of Marcos and his cronies, have diverted funds or transferred ownership to commission members or their relatives, and there have been numerous stories about low-level graft in the country even after the change in government. 37 Opposition politicians, particularly the former defense minister Enrile, have charged the government with developing a new set of cronies, as bad or worse than those of the Marcos government. Such charges vastly overstate the case. Corruption in the current government is nowhere near the organized and extensive draining of the economy during the Marcos administration. The issue for the Philippines is not so much whether there is corruption, but on what scale corruption takes place, and how much damage it does to the operation of the economy. The key for fostering economic growth is to limit the scope of possible corruption and the amount of activity that can profitably be expended in seeking kickbacks, privilege, or other economic rents. In the Philippine case this means decentralizing and localizing corruption, so that it is not run on an almost economy-wide basis as it was under M a r ~ o s . ~It' also means changing the nature of the rules and limiting the scope for discretion, so that entrepreneurial energy is channeled into production and not the competition for rents. This is the real importance of government reorganization and privatization in the Philippines. It is not so much an issue of whether the private sector is inherently better than the public sector, but more an issue of removing the

592

Robert S. Dohner and Ponciano Intal, Jr.

operation of industry and economic outcomes from those who control policy. Other policy reforms, import liberalization and the reform of the incentive structure, are important here because they shift the ground rules under which firms operate more toward neutrality and universality, and so direct energy into activities that are truly profitable for the society. So far, most attention has been placed on the pace of liberalization and privatization. More important are the terms on which these take place and the ground rules that are established. Finally, to maintain the policy reforms that have been established and to achieve stability in economic policy, the solutions to the difficult economic problems that the country faces must, be seen to be in Philippine hands. Economic ideology in the Philippines places the United States and U.S. influenced institutions in an ambiguous role. On one hand, the United States is seen as the source of Philippine problems4olonial dominance, exploitation by foreigners-but the United States is also viewed as the ultimate savior and guarantor of the Philippine situation. There has been a strong sense of deus ex machina in Philippine thinking about policy, a feeling that the United States would intervene to save them, when in fact the United States has treated the Philippines carelessly for most of postwar history. The Philippine desire for an external solution to its current difficulties was seen in the expectation, immediately following the February revolution, that the Philippines would be the recipient of a flood of foreign aid and direct investment, as well as the tremendous interest in a “Marshall Plan” for the Philippines when such a proposal was introduced in the U.S. Congress at the end of 1987. While a program will definitely go forward in what has now been renamed the Multilateral Assistance Initiative, the amount of additional funds will be far less than the $10 billion originally suggested. It is also evident in the number of “solutions” to debt difficulties that have been proposed-the recovery of the Marcos fortune, the bargaining demand that the United States take over the Philippine debt in exchange for a new bases agreement, even the government-sponsored search for the “Yamashita treasure,” a vast fortune in gold rumored to have been buried by Japanese soldiers. But the economic problems of the Philippines go beyond debt and beyond its relations with the United States and the rest of the outside world. If the Philippines is to achieve rates of economic growth that would allow it to rejoin the income ranks of its neighbors, it will require fundamental reorientation of domestic economic policy toward a more pragmatic, instrumental, and growth-oriented stance. Much intellectual effort has been expended in the search for the policy recipe that was responsible for the rapid growth of Japan and other East Asian countries, as if simply repeating the steps would duplicate their outcomes. But the success of these countries came primarily from their almost single-minded devotion to growth based on exports, and because they turned the outside world to their advantage.

593

Philippines/Notes

In contrast, much of Philippine policy, and certainly much of Philippine nationalism, has been defensive in character, designed to insulate and protect the economy from the outside world and the dangers perceived there. What the Philippines needs to develop is a more aggressive and self-confident nationalism, one that manipulates and takes advantage of the opportunities that the outside world offers-an “inward culture and an outward economy” rather than the reverse (Intal 1987). In fact, the situation in which the Philippines finds itself today is not so different from the situation characterizing many of the industrializing East Asian countries before their rapid growth took place, although none had the foreign indebtedness that the Philippines now shoulders. Japan, Hong Kong, and Singapore each had to deal with an unfavorable economic event that drastically limited their options and forced them to focus on export growth. For Taiwan and Korea it was the imminent reduction in U.S. aid which had supported domestic investment and large current account deficits that forced the shift in policies. This is not to say that adversity always leads to success, or that the problems that the Philippines now faces are no more difficult than those faced by the East Asian exporters. Nor will Philippine prospects be unaffected by events and policies in the outside world. External markets that stay open to LDC exports, rather than markets that close down at the first sign of exporting success, will have a tremendous influence over whether economic policy in those less developed countries encourages production at world prices for world markets. A commercial bank lending process that recognizes and rewards successful policy reform and economic growth, rather than a concerted lending process that provides the minimum in financing that a country can get by with, will also greatly influence the incentives for persevering in reform. But whether the Philippines will ultimately be successful will be primarily determined by the actions that the country takes and the commitment that it can forge to policies that achieve growth.

Notes Chapter 1 1 . The tensions created by this migration are the primary force behind the Moslem insurgency which has troubled the Philippines for the past two decades. 2. Golay (1961, 202), Table 39. 3 . On the problems of collection of Philippine taxes, see Golay (1961, 186-91). Tait, Gratz, and Eichengreen (1979) provide a comparison to other countries, and rank the Philippines near the bottom in tax effort.

594

Robert S. Dohner and Ponciano Intal, Jr.

4. U.S. direct investment in the Philippines increased from $149 million in 1950 to $415 million by the end of 1961. U.S. direct investment in manufacturing increased ninefold to account for half of the total by 1961 (Golay 1983, 140). 5 . Golay (1983) has compiled a list of sixteen major assets sales to Filipinos

between 1962 and 1972 totalling $246 million (152-53, table 6). The source for the table is a memorandum written by the American Chamber of Commerce in the Philippines. 6. Speech before the Rotary Club of Manila, 8 January 1970, as reported in Central Bank News Digest, 10 February 1970, p. 6. Quoted in Baldwin (1975, 76). 7. The explanation of the real wage decline and its reconciliation with the national accounts data has become a lively topic for research and discussion. Various sources of wage data exist, and while the year-to-year movements are not always consistent, they all show a substantial decline during the early 1970s. La1 (1 983) has argued that the real wage decline can be explained by a fall in the price of (labor-intensive) nontradables relative to (capital-intensive) traditional exports. The reconciliation issue is addressed somewhat inconclusively by Oshima, de Borja, and Paz (1986). They found that between 1970 and 1980 real wages fell by about 10 percent and, at the same time, national income per person employed rose by about 25 percent. They argue against an increase in hours per person employed as an explanation for the income growth. Nor does there appear to be evidence of a substantial shift in income shares. One possibility that they raise is that the growth of national income may have been substantially overstated, “perhaps even by onehalf” (10). 8. Average ODA receipts for 1969-71 were $79 million per year. From 1972 to 1975 they averaged $181 million (OECD, Geographical Distribution of Finuncial Flows to Developing Countries, various issues). U.S. military assistance to the Philippines also doubled during the period, from $60 million in fiscal years 1970-72 to $119 million during fiscal years 1973-75 (Abueva 1979, 39). 9. The Marcos land reform was limited in scope and has been widely criticized. However, it went further and was more successful than any previous land reform. It also had high public visibility. Land reform was limited to corn and rice land, and did not include sugar and coconut lands. 10. Marcos took Manila Electric Company from the Lopez family, as well as their holdings of television stations. From the Jacinto family he seized a steel complex which became the National Steel Company. Among the initial arrests under martial law was Eugenio Lopez, Jr. The story circulated in the Philippines was that the senior Lopez was told that to see his son again he would have to sign over his fortune to a specially created organization, the Marcos Foundation (Race 1975, 5). 11. The sharp rise in the price of coconut products in 1974 was associated with the particularly large reductions in Philippine exports that year due to bad weather. The price of copra rose by 140 percent between 1973 and 1974, a period in which world exports of copra fell from 1.05 million metric tons to .53 million metric tons. Philippine exports fell from .73 to .29 million metric tons in the same period. Due to the Philippine’s large share in the world coconut market, Filipinos occasionally talk of monopoly power in coconuts. This one-year example lends some support. Even so, the value of Philippine copra exports fell by 16 percent in 1974, suggesting that it is not a very profitable monopoly. Data from UNCTAD, Yearbook qf Commodity Trade Statistics, 1979.

595

Philippines/Notes

12. There is no single measure for the income effect of the terms of trade shock. The measure that we prefer focusses on the additional domestic resources necessary to maintain a given level of imports. Philippine terms of trade in 1976 were 29 percent below their 1970-72 average. The average import share of GNP in 1970-72 was 19.2 percent. (.29)(.192) = .0557. 13. Since the Philippines was engaged in a conflict with Moslem secessionists in Mindanao, supplies of oil to the country were to be reduced by 25 percent under the terms of the OPEC embargo. In 1974 Marcos declared a target of reducing oil imports from 90 percent of energy supply to 70 percent in five years’ time. 14. Interview with Gerard0 Sicat, former Director-General, National Economic and Development Authority, Washington, D.C., June 1986. 15. Over the same period the rates of export volume growth for other countries in the region were: Korea, 16.4 percent; Malaysia, 10.2 percent; Singapore, 16.6 percent; and Thailand, 17.0 percent. 16. In contrast to other countries in this study, exports of services make up a significant part (one-third) of total Philippine exports. These include the rental on U.S. military bases, but also the supply of overseas workers and construction services. Dollar earnings from exports of goods and services stayed roughly constant from 1980 to 1984. 17. In a swap arrangement a bank would borrow abroad in foreign currency and then exchange the proceeds with the central bank for pesos. The central bank in turn agreed to sell foreign currency to the bank at a set exchange rate in the future, so the bank could pay back the loan. 18. The year 1978 was taken as the starting point for the terms of trade calculation. Between 1978 and 1982 import prices rose 33.1 percent relative to Philippine export prices. This multipled times the 1979 import share of GNP of 20.8 percent equals 6.89 percent. The average interest rate paid on Philippine external debt rose from 5.21 percent in 1979 to 8.73 percent in 1982. Rates of export price inflation used to calculate the real interest rate were 5.20 percent (1975-79 average) and -2.96 percent (1979-85 average). The change in the real interest rate was multiplied times the share of gross external debt, less foreign assets of the monetary system, in average 1979-80 GNP. These types of calculations are highly sensitive to the endpoints. The endpoints were chosen as reasonable approximations and produce neither the highest nor the lowest income changes possible. 19. The ratio described is the increment in the capital stock (investment) divided by the increment in GDP. Divide numerator and denominator by the level of GDP to get the share of investment in GDP divided by the growth rate of GDP, the conventional ICOR measurement. The ICOR is an ex post and far from perfect measure of the efficiency of investment. There are questions of the timing of returns to investment, net versus gross additions to the capital stock, and short-run fluctuations in output. Thus ICOR measurements are sensitive to the time periods over which they are constructed, and are more useful when measured over longer periods. Furthermore, ICORs neglect other inputs, and therefore do not really measure the profitability of investment. However, despite these difficulties, ICORs represent a rough, but useful, measure of the effectiveness of countries in generating growth from their investible resources, and that approach is used here. 20. The rise in energy prices after 1973 made investments to reduce oil consumption profitable, even if they did not add to domestic final goods output, and

596

Robert S. Dohner and Ponciano Intal, Jr.

also made investments in (generally capital-intensive) domestic energy production profitable.

Chapter 2 1. In addition, before 1976, the Philippine fiscal year ran from July to June. In 1976, the fiscal year was shifted to January to December. 2. The rise in defense expenditures is somewhat misleading since many of the local police forces were absorbed within the military, and the share of justice and police in the budget falls accordingly. The military did increase its power and influence, as well as its size, during this period, but the budget figures overstate the increase in defense as a budget activity. 3. There was also a transfer of P. 1,300, about 11 percent of current expenditures, made to the Philippine Coconut Authority, which purchased First United Bank (renamed the United Coconut Planters Bank) in that year. Domestic prices for coconut products were subsidized with a producers levy. 4. Average ODA receipts for 1969-71 were $79 million per year. From 1972 to 1975 they averaged $1 8 1 million (OECD, Geographical Distribution of Financial Flows to Developing Countries, various issues). Data on development loans is from Magno (1976), table 6. The formation of the Consultative Committee for the Philippines in 1970 was also instrumental in focussing attention and organizing resources for the Philippines. 5. These transfers were P. 640 million in 1975 and P. 1.9 billion in 1976. Equity and net lending to government corporations again grew substantially at the end of the decade. See the line on equity contributions in table 2.2. 6. Half of the capacity addition came through new plant construction; the remaining half came from the purchase of existing oil-fired power plants from the privately owned Manila Electric Company. 7. The details of the Marcos decision and Disini’s role in it are the subject of much controversy. Jesus Vergara, the former head of Westinghouse’s Philippine sales affiliate, told investigators from the Aquino government that Westinghouse paid Disini a commission of at least $50 million, and that Disini gave Marcos about $30 million of that. Westinghouse denies that money went to Marcos, but admits giving a commission of $17 million to Disini. A U.S. Justice Department and Securities and Exchange Commission probe was launched after the contract was signed, but a grand jury in 1978 voted not to indict Westinghouse under the Foreign Corrupt Practices Act. The SEC’s enforcement division was reportedly considering in mid-1986 whether to start a new probe of the Westinghouse contract. See Dumaine (1986). 8. The nuclear power plant has figured prominently in Philippine proposals for “selective repudiation” of its foreign debt, and the Aquino government has recently filed a suit against Westinghouse over the plant. 9. Cash generation by public enterprises was relatively small, but the two government social security institutions provided savings of about 1 percent to GNP. See table 2.16. 10. For a discussion of the NPC proposal, see World Bank (1982), vol. 1, pp. 48-50. 11. For a discussion see Oshima (1983). 12. Presidential Commission on Government Reorganization (1985b), annex tables A, A.l.

597

Philippines/Notes

13. Ibid. Table 7, p. 15. 14. This is borne out in several comparative tax effort studies, referred to by Manasan (1982, 261 -62). Tait, Gratz, and Eichengreen (1979) rank the Philippines 41st out of 47 countries in their sample in terms of tax collection from existing tax bases. 15. See Golay (1961), chap. 9 for a discussion of Congress and tax policy. 16. See Kintanar (1976) for a discussion of the tax reforms in the early years of martial law. 17. The tax amnesty and early martial law period did add a large number of persons to the taxable rolls, explaining part of the secular rise in individual income tax collections. 18. In 1977 the benefits of BOI registered firms under the Investment Incentives Act were estimated at P. 926.8 million (a little over 5 percent of total tax collections). Of this, about half represented deductions from taxable income and the rest exemption from indirect taxes (World Bank 1980, tables 2.6, 2.7). 19. IMF (1985b, 81), table 33. 20. Buoyancy and elasticity are two measures of the responsiveness of tax collections to income changes. Elasticity measures the response of the existing tax structure to changes of income. Buoyancy is a measure of the ex post change in tax collections with a change in income, including the effects of any changes in rules or rates. 21. Taxpayers with over P. 1 million in income accounted for 23 percent of taxable individual income and 61 percent of tax collected in 1978. See Diokno (1986), table 2. 22. IMF (1985b, 31), table 13. The estimation procedure used only urban household income, and is described in pages 26-30 of that study. 23. IMF (1985b, 85), table 35. Two presidential decrees in 1984 eliminated most tariff and tax exemptions, but in the next few months a large number of the previous exemptions were reinstated. 24. An article in Business Day, “Coconut Fund Nearing P. 6 billion,” 1 September 1980, gave a figure of P. 5.8 billion at that time. 25. Presidential Commission on Government Reorganization (1985a, 65). Mrs. Marcos, who was minister for human settlements, was also known for soliciting “donations” from corporations and institutions to support the various projects that she undertook. 26. Manuel Montes and Emmanuel de Dios make this argument. See Montes (1986, 45-49) and de Dios and Montes (1985). 27. BIR employees repeatedly made this point to the JMF mission. See IMF (1985h, 100, 103-4). 28. This count, and in general the statistics available for government-owned corporations, excludes private companies that were acquired by the state due to bankruptcy or default and therefore differs from the privatization list. 29. The most important battles were the attempts of the finance minister, Cesar Virata, to remove the coconut levy and circumscribe the trading monopoly of the National Sugar Trading Corporation (NASUTRA) in 1981. Marcos suspended the levy in 1981, but reinstated it after pressure from Eduardo Cojuangco. The Philippine cabinet overruled the proposal on NASUTRA in September 1981. See “Philippine Technocrats Lose Round in Fight to Depoliticize the Economy,” Wall Street Journal, 29 October 1981, p. 34. 30. See IMF (1985b, 85), table 35, and the discussion (84-86). 31. See Amatong et al. (1985), who state “the study concludes that [national government] contributions expand rather than reduce the need for borrowing” (54). 32. Lamberte (1984), tables 1 , 8.

598

Robert S. Dohner and Ponciano Intal, Jr.

33. Since most of the forward contracts were drawn with public corporations, the losses were in a sense transferred from one public sector agency to another. However, the existence of the contracts may have meant a greater exchange rate exposure of the public sector, and thus greater losses, than it would have suffered had the operations of corporations like PNOC been uncovered.

Chapter 3 1. The major references on Philippine trade policy are Power and Sicat (1971), Baldwin (1975), Bautista, Power, et al. (1979), Bautista (1981), Alburo and Shepherd (1986), and Power and Medalla (1986). 2. See, for instance, Sachs (1985). 3. Hooley (1985, 24, 26), tables 6, 7. Industry output data show large year-to-year variations in the Philippines, apparently due to differences in the degree of reporting, so conclusions about productivity in the Philippines must be somewhat tentative. 4. Baldwin (1975, lo), table 1-4. 5. See Alburo and Shepherd (1986), especially pp. 96-98. 6. Power and Sicat (1971, 43) and Alburo and Shepherd (1986, 91). 7. George L. Hicks, “Philippine Foreign Trade, 1950-1965: Basic Data and Major Characteristics” and “Supplementary Data and Interpretations, 1954- 1966,” (National Planning Association, Center for Development Planning, Washington, D.C., 1967; mimeo), cited in Baldwin (1975, 64),fn. 18. 8. The product groups for which imports were regulated specifically for industry protection included textile fabrics, synthetic yarns, fibers and threads, basic iron and steel products, newsprint, used and new tires, synthetic resins, liquid caustic soda, cellophane and paper and paperboard products. 9. See de Dios (1986). 10. This policy was enforced through the Board of Investment’s control over imports of capital goods. 11. Gregorio (1979, 217) and summary by Bautista, Power, et al. (1979, 33). 12. Interview with Cesar Virata, Washington, D.C., November 1986. 13. World Bank (1986), vol. 2, p. 145. 14. World Bank (1987), vol. 2, pp. 94, 98, and table 3.1 1. 15. Power and Medalla (1986, 24), table 3. 16. In addition, the Philippines did not have speculative capital inflows, such as occurred in Chile. 17. See chapter I , n. 7. 18. See Clunies-Ross (1966) and Bautista (1976) for analyses of Philippine inflation, and the review by Tan (1980, 226-35). Food and beverages accounted for 60 percent of the consumer price index in the Philippines as late as 1978. 19. World Bank (1987), vol. 1, p. 17. 20. World Bank (1980, 5 3 , table 2.9. 21. Alburo and Shepherd (1986, 107), table 3.1. 22. American Apparel Manufacturers Association, ‘‘Apparel Manufacturing Strategies, 1984,” quoted in World Bank (1986), vol. 2, p. 160, table 5.10. 23. This overstates the actual decline in log exports, since undeclared and illegal exports of logs increased substantially during the 1970s.

599

Philippines/Notes

24. During the international commodities boom in the early 1970s, the Philippines introduced premium taxes of 20-30 percent on the excess of commodity prices over their February 1974 level. These collections disappeared when international prices fell below the base level. 25. Hooley (1985, 26), table 7. 26. Much of this was in energy and mining, as well as manufacturing

Chapter 4 1. The Philippine sugar quota to the U.S. market rose from 864,000 metric tons to 1.3 million metric tons in 1961. The Philippines gained additional quota amounts when other producers failed to meet their additional quotas. 2. See Brown (1986). 3. Benedicto was also regional director of Marcos’ political party, the New Society Movement, and had been ambassador to Japan in the late 1960s and early 1970s, when the Philippines contracted with Japanese firms to construct several new sugar mills. 4. Canlas et al. (1984, 81), table 13. The revenue estimates are after subtraction of a 10 percent allowance for trading costs. 5. “End of An Experiment,” Far Eastern Economic Review, 16 October 1981, p. 125. 6. David, Barker, and Palacpac (1984) show a decline in yield per hectare of 4.3 percent per year during the 1960s, and a gain of 0.8 percent per year during the 1970s. 7. Philippine Statistical Yearbook 1987, p. 270, table 5.1. 8. During 1970-82, the number of nuts per bearing tree averaged 39 per year as against 42 nuts per year during the 1960s (NEDA 1985, table 1.13). 9. The share of coconut oil relative to the sum total of world production of soybean oil, palm oil, and coconut oil declined from about one-quarter in 1968 to about one-eighth in 1983; in contrast, the share of palm oil rose from about one- sixth in 1968 to about one-quarter in 1983 (Ibid., table 3.10). 10. Canlas et al. (1984), appendix 1, lists 688 presidential decrees and 283 letters of instruction that involved government intervention in the economy. 11. See for instance, Sicat (1986, 28-29, 36-38). 12. See Coronel (1986, 3). 13. See Espinosa (1986) for a discussion of the BASECO case. 14. See, for instance, Mijares (1976, 197-202). 15. There are various sources on the distribution of crony assets. See, for instance, Doherty (1982), Canoy (1980), and the Manila Times, 16 May 1986. Extensive, but controversial, information may become available if the investigations of the Presidential Commission on Good Government are ever made public. 16. See Villegas (1986, 163-64). Imelda Marcos was once asked why so many relatives and associates of the Marcos government had profited so greatly during the martial law years. She replied, in a famous quotation, “some are smarter than others.” This became the title of an underground analysis of cronyism in the Philippines that was ultimately published as Doherty (1982). 17. This is not to say that there was no Competition among the cronies. Although instances are hard to come by, there was an attempt by Eduardo Cojuangco to take over the monopoly trading in sugar-the province of Roberto Benedictein the

600

Robert S. Dohner and Ponciano Intal, Jr.

1980s when NASUTRA had financial difficulties. But the competition was generally over exclusive rights in an industry, rather than between or among firms in an industry. 18. Although, even without competition, several of the cronies went bankrupt in the early 1980s. 19. As Benjamin Diokno, an academic and now an official in the Budget Ministry, put it, “In advanced countries, the government intervenes when firms are in distress. We did some of that, but [the Marcos government] tended to intervene when firms were profitable” (Interview, Manila, 25 August 1986). The government gained control of the export of labor services when that industry grew in size in the early 1980s, and had plans, before the 1986 election, to intervene in the videocassette rental industry. 20. Interview with Fred Whiting, President, Sime Darby, Philippines, Manila, 8 December 1986. Sime Darby had been worried about Cojuangco during the last years of the martial law government, but found, after Corazon Aquino came to power, that Marcos’ son-in-law, Freddy Araneta, had intended a takeover of part of Sime Darby’s operations. 21. In order to provide a scale against which these amounts may be measured, one billion pesos in 1981 was roughly one-third of 1 percent of GNP, and just under 1 percent of gross domestic credit of the banking system. 22. Business Day (Manila), 24 February 1983. Cited in Ibon (1983a, 7). 23. Ibon (1983a), based on news reports. 24. Lamberte (1984), annex A. 25. Lamberte (1984, tables 4 and 6. 26. The most publicly outspoken critic was Jaime Ongpin, brother of the director of the National Development Corporation and, until October 1987, finance minister in the Aquino government, who termed these actions “the most disgraceful waste of public funds since independence.” See his “A Matter of Decree” and “A Matter of Principle” (Manila; mimeo, ca. March 1983).

Chapter 5 1. More general and complete descriptions of the Philippine financial system are found in World Bank and IMF (1980) and Pascual (1984). 2. Asian Development Bank (1985, 37), table 4.1. For comparison, the figure for Thailand was 4.5 percent, Korea, 5.4 percent, and Indonesia, 0.1 percent. 3. PNB Bank is now a much smaller bank, since its nonperforming assets have been transferred to the Asset Privatization Trust (AFT) as a part of the restructuring of government financial institutions carried out by the Aquino government. The current intention of the government is to ultimately convert PNB into a privately owned commercial bank. 4. Lamberte et al. (1985, 30), table 3.8. 5. As was the case with PNB, the transfer of DBP’s nonperforming assets (almost 90 percent of its loan portfolio by 1987) to the APT has left it a much smaller institution. 6. These guarantees were not carried on the balance sheet of the institution. 7. The financial crisis which occurred in the money market in 1981 and the continuing difficulties of financial institutions in that market led to a greatly reduced

601

Philippines/Notes

role for deposit substitutes, which accounted for only 6 percent of total liquidity by 1985. 8. The fact that the lending by foreign currency deposit units was almost entirely to domestic residents was not known by the Philippine’s external creditors. Thus, the net external debt of the monetary sector was almost equal to its gross debt, and much higher than external creditors thought. This was the surprise that was contained in the Philippine announcement that the country’s external debt, including monetary debt, was over $25 billion, instead of the published $19 billion figure that included only nonmonetary debt. 9. In 1971-72, when ceilings limited commercial bank loan rates to 12-14 percent, studies by the National Economic Council (NEDAs forerunner) and Victor Barrios show a range of 14-29 percent (Tan 1980, 205). The World Bank and IMF (1980) study reported effective rates in the money market of up to 30 percent. 10. Average for 1971-73. 11. Precise ownership and control patterns are difficult to establish. Estimates range from two-thirds (World Bank and IMF 1980, 8) to 23 out of 28 banks existing in 1980 (Doherty 1980, chap. 5), quoted in Patrick and Moreno (1985, 316). 12. Gregorio Licaros, then central bank governor, quoted in the Far Eastern Economic Review, 7 April 1978, p. 80. 13. The acquired banks were Associated Bank (acquired by DBP), Commercial Bank of Manila (GSIS), International Corporate Bank (NDS), and Union Bank (SSS and Land Bank). The government owned two additional banks, Republic Planters Bank, acquired in 1978, and Pilipinas Bank, acquired by PNB in 1980. 14. For a discussion of the “political banks,” see Patrick and Moreno (1985) and Canlas et al. (1984, 65-72). 15. In fact this understates the influence of the government in the financial system. As one Philippine banker put it, “Look, everyone made behest loans. You would receive a phone call asking you to make a loan. If you refused you would have labor problems, or the Metro-Manila government would close you down on a health or safety violation.” (anonymous interview in Manila, 19 December 1986). 16. Robert Emery, writing of the 1960s, commented, “the Philippines has probably had more financial scandals, or financial institutions in distress, than any other Southeast Asian country” (1970, 482). 17. LIBOR ranged from 6 to 12 during 1974-79. A study by the Ministry of Industry in the Philippines (cited in the World Bank and IMF 1980 study) also found foreign borrowing significantly less expensive than domestic borrowing for Philippine firms. 18. From Business Day, Top 1000 Corporations in the Philippines, cited in Lamberte (1989, table 3.10. 19. Dee reportedly suffered huge losses in commodity speculation. He later surfaced in Canada, where he applied for asylum. 20. Asian Finance, 15 September 1981, p. 88. 21. The vast majority of central bank advances went to two institutions. Atrium Capital, a member of the Herdis (Disini) group, received over P. 1 billion, while Bancom, an institution that had founded the deposit substitute market, received P. 400 million. 22. Technically the commercial paper of these firms had been sold to the public without recourse to the financial institution, in large part to evade the regulation of the central bank. In fact, the instruments were in many cases accompanied by the postdated checks of the financial institutions, and many institutions allowed

602

Robert S. Dohner and Ponciano Intal, Jr.

pretermination of the investments. When the central bank stepped in, the default risk was removed from the public and placed on the institutions or, ultimately, on the central bank. 23. They were Republic Planters Bank (1978), Pilipinas Bank (1980), International Corporate Bank (1981), Union Bank (1981), Associated Bank (1982), and Commercial Bank of Manila (1982). 24. Development Bank of the Philippines, Annual Report, 1982. 25. The World Bank reports evidence that the Philippines paid significantly more for capital equipment (World Bank 1986, vol. 2, p. 133). Japanese companies were reported to have cooperated in these schemes, and that has been an alleged reason for the reluctance of the Japanese government to cooperate in the Aquino government’s investigations of hidden wealth. See “Kickback for Japan” Far Eastern Economic Review, 24 April 1986, pp. 42-43, and “Marcos Papers Show Japanese Sought to Avoid Payments,” Washington Post, 22 March 1986, p. A l . 26. Aide-mCmoire from DBP President Cesar Zalamea to President Aquino, 11 March 1986. Cited in “DBP Bankrupt since Dec. 1983,” Manila Times, 29 March 1986, p. 1. 27. “DBP Bankrupt Since Dec. 1983,” Manila Times. DBP president Zalamea is quoted in the aide-memoire as saying, “at a time when large borrowings were possible from both the domestic and international markets, DBP did not hesitate to draw such funds to cover cash losses and pursue lending programs.” 28. “COA Bares Bad GSIS Investments,” Business Day (Manila), 10 September 1986, p. 15. 29. “Government May Suffer P. 70 billion Loss,” Business Day (Manila), 8 September 1986, p. 2. The 30 billion peso figure is the difference between booked and total exposure to the nonperforming assets. 30. See Fry (1986) for a summary of the consequences of financial policy in the Philippines that makes many of the points listed here. 31. The importance of interest rate controls, or “financial repression,” in reducing savings is controversial. Fry (1978) found a significant effect of real interest rate levels on savings mobilization in a group of Southeast Asian countries that included the Philippines, a finding contradicted by Giovannini (1983). For the Philippines, the periods of substantial growth of financial intermediation, the 1950s and the early 1980s, were periods when interest rate controls were absent or nonbinding. Controls also influenced the form in which savings were mobilized, as savings flowed into deposit substitutes and money market instruments in the 1970s.

Chapter 6 I . The Monetary Board is chaired by the minister of finance, and includes the governor of the central bank and the minister of planning. 2. The inclusion of gross short-term external debt of the banking sector is the reason for the large jump in published figures of Philippine external debt that appears in 1983, from approximately $19 billion to $25 billion. 3. In addition, FCDUs were required to maintain foreign exchange cover equivalent to foreign currency liabilities, and faced limitations on maturities of their lending. The relevant central bank circulars governing FCDUs are 343 (April 1972) and 547 (November 1976). Liabilities of these units are often referred to as 343/547 deposits in Philippine statistics. 4. Central bank, Annual Report, vol. 1, various issues. 5. IMF, World Economic Outlook, April 1986, p. 243, table A47. 6. Data from MEDIAD.

603

PhilippinesINotes

7. IMF (1984b, 129), table 56. 8. Gross liabilities of the banking sector were an additional $5 billion, or 40 percent of nonmonetary debt. However, in 1980 the Philippine banking system still had a small net asset position. 9. See Cuddington (1986) and Cumby and Levich (1987) for a discussion of measurement concepts for capital flight. This section on capital flight draws on Dohner (1987), where the measures used are derived. Cumby and Levich also present data for the Philippines and reach similar conclusions about the rise in capital flight in the early 1980s, well before the Aquino assassination. 10. Dohner (1987), table 4. 11. Dohner (1987) estimates regressions equations for several measures of Philippine capital flight. The results confirm the importance of domestic and foreign interest rates and the real exchange rate in determining capital flight. 12. Figures on short-term debt collected by MEDIAD and the central bank’s Financial Plan Data Center differ, due to the latter’s more inclusive definition. Thus the figures in table 6.10 do not show a decline. 13. Other than the reserves overstatement, little else was uncovered about Philippine statistics. The employment, unemployment, and wage data are subject to question, as is the true extent of savings and investment in the Philippine economy. But no additional examples of data falsification arose. 14. The first was the Subcommittee on Major National Projects (SMNP) and the second was the Foreign Resources Subcommittee. The World Bank (1984a), chap. 4, has a good description and evaluation of the debt management system in the Philippines. 15. Export-oriented firms were also given approval to borrow short term for pre-export financing, raw materials, supplies, and spare parts. 16. Interview with Cesar Virata, Washington, D.C., October 1986. 17. Interview with Robert Hess, Chase Manhattan Bank Manila, 25 November 1986.

Chapter 7 1. In most cases export taxes were increased from 4 to 6 percent, but taxes for coconut products were doubled, and taxes were introduced for the first time on coffee and tuna exports. 2. There was an increase in central bank credit to public sector enterprises during the year, but this was a relatively small part of total financing. 3. Reserve money corresponds to the conventional definition of high-powered money4urrency in circulation, banks’ holdings of cash, and bank reserves with the central bank. In the Philippines certain government securities are also eligible as reserves. A more precise measurement of high-powered money is provided by what the IMF calls “base money ”-reserve money, plus reserve-eligible securities, plus banks’ reserve deficiencies. It is only recently that longer series for base money have become available, and most policy discussions and program limits were phrased in terms of reserve money. 4. Rates were phrased in terms of so many percentage points below the Manila Reference Rate (MRR). The MRR is the weighted average of the interest rate paid by the ten largest commercial banks on deposit substitutes or promissory notes of ninety day maturity.

604

Robert S. Dohner and Ponciano Intal, Jr.

5. The thirteen commodities controlled by the Price Stabilization Council included rice, corn grits, chicken, eggs, pork, sugar, milk, canned fish, and some school supplies. Together these commodities made up 24 percent of the Consumer Price Index. Rice alone accounted for 11.6 percent. 6. Price interventions by the government took the form of price ceilings and differential taxes (for petroleum products). For some commodities such as rice, government procurement policies set price floors. However, Philippine price intervention policies did not involve major subsidization by the government. There were producer levies in the sugar and coconut industries that supported consumer prices for sugar and cooking oil below world market levels. In addition, the government would occasionally (generally before elections) import rice and other consumer goods to insure sufficient domestic supply. 7. Reserve money became an additional and more prominent monetary target due to the suspicion cast on net domestic assets by the reserves overstatement, and by disputes on what should be included in domestic assets. Since net domestic assets is a balance sheet counterpart to net international reserves, an overstatement of reserves should also have resulted in an understatement of net domestic assets. The reserve money figure gained prominence because it was more easily and more accurately measured. 8. The crisis started after leaflets were spread across Manila in which it was warned that various banks were unsound, urging people to take their money out. 9. This tax was 30 percent of the difference between the commodity price at the new exchange rate and its price at the previous exchange rate of P. 14 per dollar. At an exchange rate of P. 18 per dollar, this implied an effective exchange rate for the exporter of P. 16.8. 10. In fact adjustments were included to incorporate changes in reserve-eligible securities held by banks and in reserve deficiencies, so that the targeted variable was in fact base money-total high-powered money, or support for deposit expansion. 11. The Philippine letter of intent was published by Business Day (Manila), 3, 4, and 5 January 1985. 12. The bank advisory committee was headed by Manufacturer’s Hanover, with the Bank of Tokyo as deputy. It included Bank of America, Banque National de Paris, Barclays, Chase Manhattan, Citibank, Morgan Guaranty Trust, and the Bank of Montreal. Later, Dresdner Bank and Fuji Bank were added to the advisory committee. 13. This matched the foreign exchange assets of Citibank’s Manila branch that were held outside the Philippines. Thus it did not involve a principal repayment from the Philippines. 14. As in the case of the Citibank deposits, the issues at stake went beyond the Philippines. National Commercial Bank’s exposure to the Philippines was dwarfed by its $4 billion lending to Korea. It was concern about a possible standstill and rescheduling in Korea, more than in the Philippines, that apparently motivated the bank’s refusal. 15. The Philippine commercial bank restructuring package is described in more detail in Bucheit (1985), Hutton (1985), and Villaruel and Reyes (1985). For information about the FICORCA program, see chapter 8 of Edward Buffie’s study of Mexico in volume 2 of the Sachs series on developing country debt.

605

Philippines/Notes

16. Only the program for restructuring private financial sector debt had a provision for the Philippine government’s assumption of private debts. The major dispute over government assumption of private sector debts came in the PPI case, described above. 17. The trade facility was also large because the Philippines had used trade finance to cover part of their balance of payments deficit through the central bank oil facility in the period prior to the moratorium. 18. Watson et al. (1986), table 40. 19. In this section, unless otherwise indicated, we measure the rate of inflation by the change in prices over the previous six months, expressed as an annual rate. 20. Data from IMF (1985c), annex 1. This figure probably underestimates the drop in real wages that took place. 21. This and the following discussion draw on Tidalgo and Esguerra (1984, 75 - 82). 22. Domestic credit extension of the central bank is defined here as all other net domestic assets of the central bank, ie., net domestic assets plus the (positive) stock of outstanding CB securities. 23. World Bank (1987), vol. 1, pp. 10-11, n. 19. See also Montes (1987, 28- 32). 24. The World Bank argues that these raised bank spreads by seven percentage points in 1985 (World Bank 1986, vol. 2, pp. 25-26). 25. Philippines, Ad Hoc Study Group (1986, 23). 26. See “Philippines Reels Under Severe Restraints,” Asian Finance, 15 September 1986, pp. 118-19. IMF Managing Director Jacques de Larosiere reportedly overruled staff recommendations and insisted on tighter conditions in the 1984-85 program because of the overstatement of reserves (Fur Eastern Economic Review, 26 July 1984, p. 49). 27. Montes (1987) makes this argument.

Chapter 8 1. See World Bank (1985, 8-19). 2. Business interest groups, such as the Makati Business Club, were established during this period. 3. Cardinal Jaime Sin persuaded the two candidates to unite on a single ticket and Laurel to accept the vice presidential spot. 4. The mass mobilizations of civilians around the rebel army camp gave the government transition its name, the “People’s Power Revolution.” The story of these events is told in Johnson (1987). 5 . Philippines, Department of Budget and Management (1987, 3). 6. Ibid., 14. The P. 24 billion is larger than that of table 2.16 because it includes transfers to the central bank. 7. For further discussion and analysis of the Aquino government’s economic performance, see Dohner (1988). 8. Philippines (1986), tables 1, 2. The large increase in the deficit of the national government is deceptive because it included transfers of liabilities from NPC and the government financial institutions. 9. Sanchez was later dropped from the cabinet during a reorganization in late 1986.

606

Robert S. Dohner and Ponciano Intal, Jr.

10. Arroyo and Ongpin had fought bitterly within the cabinet, and the departure of Ongpin was a condition for Arroyo’s resignation. Ongpin was replaced by another businessman, Vicente Jaime, who had been appointed head of the Philippine National Bank by Aquino. Several weeks after he left the cabinet, Ongpin was found dead, the victim of an apparent suicide. 11. Their 1984 report (Canlas et al. 1984) was strikingly critical of government economic policies. On the influence of University of the Philippine economists in the new government, see Rigoberto Tiglao, “UP Economists Gain Influence,” Business Day (Manila), 19 May 1986. 12. In general orientation, particularly in the emphasis on rural-based development policy, the yellow book proposals closely resemble those of the 1973 International Labour Office study mission headed by Gustav Ranis (ILO 1974). 13. The yellow book did argue for repudiation in cases where loans were made for fraudulent purposes with the knowing participation of the lender. 14. Compare for instance, NEDAs “Policy Agenda for People-Powered Development” (1986). 15. Not all of this is a net improvement in the government’s budget position since some of the increased revenues come from the withdrawal of tax exemptions of government corporations, which would be offset, at least in the short term, by increased government contributions to those firms. The estimated yield from removing tax exemptions was P. 4 billion in 1987 (Philippines, Department of Budget and Management 1987, 25). 16. These efforts were supported by funds and technical assistance from the World Bank. 17. Tax exemptions were later reintroduced for NPC, and reforms in power pricing have been very slow. 18. The valuation of nonperforming assets varies by the measure applied; the broadest, which produces the $7 billion figure, includes unbooked interest and penalty charges. Total booked exposure of the nonperforming assets is estimated to be $5.5 billion. 19. As a result of the restructuring, the share of PNB in commercial bank assets was reduced from 28 percent in 1985 to 13 percent in 1987. 20. Business World (Manila), 8 December 1987, p. 3. 21. Far Eastern Economic Review, 5 November 1987, p. 84. 22. On the Philippine privatization program, see Haggard (1988) and “Manila’s Hardy Privateers,” Far Eastern Economic Review, 7 July 1988, pp. 88-93. 23. World Bank (1987), vol. 1, p. 38, quoting a study done by Rosario Manasan. 24. World Bank (1987), vol. 1, p. iii. 25. See Koppel (1987) for a discussion of the Marcos land reform and land manipulation experience. 26. Citibank was apparently supported in this position by the Bank of Montreal (Wall Street Journal, 26 January 1987, p. 24). The Philippines reacted angrily to the intransigence of Citibank in the negotiations, and there were calls in the country at the end of 1986 for a reduction in the operating authority of the Citibank branch in Manila. 27. Under the proposal, PINs applied to the spread only; the base rate (LIBOR) would continue to be paid in cash. The accounting implications of PINs were ambiguous, and may have required that banks write down their holdings of Philippine debt. To date, no PINs have been issued.

607

Philippines/References

28. Interest rate spreads on restructured debt of the private financial sector, where the lender retained the commercial risk, were higher, but could not exceed one and three-eighths percentage points over LIBOR. The repricing of the 1985 new money facility carried additional restrictions. The Philippines agreed to prepay 4 percent ($37 million) of the facility before 1991, and also agreed that if the country missed any of its principal payments, the spread would revert to I percent. 29. A peculiarity of the Philippine balance of payments statistics is that U.S. Economic Support Funds, which are an outgrowth of the current agreement on US. military bases, are included as service receipts. The Philippine government regards these as the equivalent of rental payments, while the U.S. government considers them aid. These amounted to $62 million in 1985, $300 million in 1986. Even after netting these out, the Philippines had small surpluses on service account in 1985 and 1986. 30. These and other issues surrounding the 1989 debt negotiations are discussed in Dohner (1989). 31. In 1987 the Philippines was forced to import sugar on the world market in order to meet its export quota to the United States, so there is still some prospect of increased net export earnings as the sector recovers. 32. World Bank (1987), vol. 2. p. 266. 33. A study done by the American Apparel Manufacturers Association in 1984, “Apparel Manufacturing Strategies,” gives the following productivity-adjusted wage comparisons for the industry: United States, 6.50; Hong Kong, 2.56; Taiwan, 2.33; Korea, 1.67; Jamaica, 1.36; Costa Rica, 1.33; Egypt, 1.10; Philippines, 1.OO; Haiti, .80; and China, .75. 34. World Bank (1987), vol. 2, p. 276. 35. World Bank (1987), vol. 2, p. 127. 36. World Bank (1986), vol. 1, p. 25. 37. As Customs Commissioner Salvador Mison told his staff in November 1987, “Don’t take bribes. Accept only ‘tokens of appreciation’.” Quoted in Claudia Rosett, “Aquinos Embattled Democracy,” Wall Street Journal, 2 March 1988, p. 21. 38. This is similar to the notion, developed by Albert Hirschman, of “traittaking” technologies that are robust to corruption. He compares truck transport with rail transport in Nigeria, and concludes that decentralization and the ability to make fast, on-the-spot decisions gives trucking firms an advantage over the rails. See Hirschman (1967, 140-44).

References Abueva, Jose V. 1979. Ideology and practice in the “New Society.” In Marcos and martial law in the Philippines, ed. David Rosenberg, 32-84. Ithaca, NY: Cornell University Press. Alburo, Florian, and Geoffrey Shepherd. 1986. Trade liberalization: The Philippine experience. World Bank study on timing and sequencing of trade liberalization. Manila: Philippine Institute for Development Studies, mimeo, May.

608

Robert S. Dohner and Ponciano Intal, Jr.

Alburo, Florian, et al. 1986. Economic recovery and long-run growth: An agenda for reforms. Manila: Philippine Institute for Development Studies, May 1. Amatong, Juanita, Gil Beltran, and Emy Boncodin. 1985. Explicit budgetary contributions to public enterprises. Manila: Philippine Institute for Development Studies, mimeo. Ascher, William. 1980. Political and administrative bases for economic policymaking in the Philippines. Washington: World Bank, memorandum. Reprinted as: And after Marcos? Asia Record (December 1980): B 1-4. Asian Development Bank. 1985. Capital market development in selected member countries of the Asian Development Bank. Manila: Asian Development Bank. Baldwin, Robert. 1975. Foreign trade regimes and economic development: The Philippines. New York: Columbia University Press for the National Bureau of Economic Research. Bautista, Romeo. 1976. Inflation in the Philippines: 1955- 1974. In Philippine economic problems in perspective, ed. J. Encamacion et al. Quezon City: School of Economics, University of the Philippines. -. 1981. The 1981- 1985 tariff changes and effective protection of manufacturing industries. Journal of Philippine Development 3: 1-2. Bautista, Romeo, et al. 1979. Industrial promotion policies in the Philippines. Manila: Philippine Institute for Development Studies. Bresnan, John, ed. 1986. Crisis in the Philippines: The Marcos era and beyond. Princeton: Princeton University Press. Briones, Leonor. 1985. Philippine public enterprises in the 1980s: Problems and issues. PIDS Development Research News 3(4):1-6. Broad, Robin. 1983. Behind Philippine policy making: The role of the World Bank and the International Monetary Fund. Ph.D. diss., Woodrow Wilson School, Princeton University. Brown, J. 1986. The international sugar industry: Developments and prospects. Draft Working Paper. Washington, D.C.: World Bank, September. Bucheit, Lee. 1985. Details of the Philippine debt repayment program. International Financial Law Review (December): 15- 17. Callison, C. Stuart. 1981. Economic assessment of the New Society and key problems and issues facing the Republic of the Philippines. Manila: U.S. Agency for International Development, October I . Campbell, Burnham 0. 1982. Petrodollar recycling, 1973-1980. Part 11: Debt problems. Economic Staff Paper no. 12. Manila: Asian Development Bank. Canlas, Dante, Emmanuel de Dios, Raul Fabella, Felipe Medalla, Solita Monsod, Manuel Montes, Edita Tan, and Rose Linda Tidalgo. 1984. An analysis of the Philippine economic crisis: A workshop report. Quezon City: University of the Philippines, mimeo. Canoy, Reuben. 1980. The counterjeit revolution. Manila: Philippine Editions Publishers. Clarete, Ramon, and James Roumasset. 1983. An analysis of the economic policies affecting the Philippine coconut industry. Working Paper 83-08. Manila: Philippine Institute for Development Studies. Clunies-Ross, James. 1966. Understanding Philippine inflation. Philippine Economic Journal 5. Coronel, S . 1986. How yen aid was stolen. Manila Times, 29 March, pp. I , 3.

609

Philippines/References

Cuddington, John T. 1986. Capital Jlight: Estimates, issues, and explanations. Princeton Studies in International Finance no. 58. Princeton, NJ: Department of Economics, Princeton University. Cumby, Robert, and Richard Levich. 1987. On the definition and magnitude of recent capital flight. NBER Working Paper no. 2275. Cambridge, Mass.: National Bureau of Economic Research. David, Cristina. 1983. Economic policies and Philippine agriculture. Working Paper no. 83-02. Manila: Philippine Institute for Development Studies. David, Cristina, R. Barker, and A. Palacpac. 1984. The nature of productivity growth in Philippine agriculture, 1948- 1982. International Rice Research Institute, Agricultural Economics Department Paper no. 84-22. Los Banos, Laguna. de Dios, Emmanuel S., and Manuel Montes. 1985. A perspective on the Philippine economy. Quezon City: University of the Philippines, School of Economics, December, mimeo. de Dios, Loreli. 1986. Nontariff measures affecting Philippine imports. Special Paper Series, PIDS-Tariff Commission Project on Trade Liberalization. Manila: Philippine Institute for Development Studies. de Vries, Barend. 1980. The transition towards more rapid and labor intensive industrial development: The case of the Philippines. World Bank Staff Working Paper no. 424. Washington, D.C.: World Bank. Diokno, Benjamin. 1986. Revenue mobilization and responsiveness of Philippine income taxes: Implications for fiscal policy. Discussion Paper 8605. University of the Philippines, School of Economics. Doherty, John F. 1980. A study of interlocking directorates among financial, commercial, manufacturing, and service enterprises in the Philippines. Manila: Anteneo de Manila University, mimeo. -. 1982. Who controls the Philippine economy: Some need not try as hard as others. In Cronies and enemies: The current Philippine scene, ed. Belinda Aquino. Philippine Studies Occasional Paper no. 5. Honolulu: Center for Asian and Pacific Studies, University of Hawaii. Dohner, Robert S. 1987. Capital flight in the Philippines: 1971-1986. Medford, Mass.: Tufts University, mimeo. . 1988. Aquino and the economy: An assessment of the first three years. Pilipinas 11 (fall): 1-33. . 1989. Philippine external debt: Burdens, possibilities, and prospects. New York: Asia Society. Dumaine, Brian. 1986. The $2.2 billion nuclear fiasco. Fortune, 1 September, pp. 39-46. Emery, Robert. 1970. Financial institutions in Southeast Asia. New York: Praeger. Espinosa, Roberto. 1986. How Marco set up BASECO. Manila Chronicle, 25 July, p. 1, 6. Findlay, C. and Ross Gamaut. 1986. The political economy of manufacturing protection: Experiences of A S E M and Australia. Sydney: Allen and Unwin. Fry, Maxwell. 1978. Money and capital or financial deepening in economic development? Journal of Money Credit, and Banking 10:464-75. . 1986. Financial structure, financial regulation, and financial reform in the Philippines and Thailand: 1960-1984. In Financial policy and reform in Pacific Basin countries, ed. Hang-Sheng Cheng. Lexington, Mass.: Lexington Books.

610

Robert S. Dohner and Ponciano Intal, Jr.

Giovannini, Alberto. 1983. The interest elasticity of savings in developing countries: The existing evidence. World Development 11(7):601-7. Golay, Frank H. 1961. The Philippines: Public policy and national economic development. Ithaca, NY: Cornell University Press. Golay, Frank H. 1983. Taming the American multinationals. In The Philippine economy and the United States, ed. Norman G. Owen, 131-76. Ann Arbor: University of Michigan Center for South and Southeast Asia Studies. Gregorio, Rosario. 1979. An economic analysis of the effects of Philippine fiscal incentives for industrial promotion. In Industrial promotion policies in the Philippines, ed. R. Bautista et al. Manila: Philippine Institute for Development Studies. Haggard, Stephan. 1988. The Philippines: Picking up after Marcos. In The promise of privatization, ed. Raymond Vernon, 91-121. New York: Council on Foreign Relations. Hawes, Gary. 1987. The Philippine state and the Marcos regime: The politics of export. Ithaca, NY: Cornell University Press. Hirschman, Albert. 1967. Development projects observed. Washington, D.C.: Brookings Institution. Hooley, Richard. 1985. Productivity growth in Philippine manufacturing: Retrospect and future prospects. Monograph Series no. 9. Manila: Philippine Institute for Development Studies. Hutton, C. Peabody. 1985. Restructuring Philippine private corporate debt. International Financial Law Review (December): 14- 15. Ibon Facts and Figures. 1983a. The state in distressed business. Vol. 118 (15 July). Manila: Ibon Databank Philippines. _ _ . 1983b. The biggies close up. Manila: Ibon Databank Philippines, 123 (30 September). ILMS. See Institute of Labor and Manpower Studies. ILO. See International Labour Office. IMF. 1984a. Philippines: Staff report for the Article IV consultation. EBS/84/114. Washington, D.C.: IMF, 29 May. . 1984b. Philippines: Recent economic developments. 8 June. -. 1984c. Philippines: Request for standby arrangement. EBS/84/226. 5 November. -. 1985a. Philippines: First review under standby arrangement. EBS/85/109. 1 May. _ _ . 1985b. The tax and customs system of the Philippines. Washington, D.C.: IMF Fiscal Affairs Department, 25 June. __ . 1 9 8 5 ~Philippines: . Recent economic developments. 30 August. . 1986a. Philippines: Staff report for the 1986 Article IV consultation. EBS/86/222. 19 September. . 1986b. Philippines: Recent economic developments. SM/86/249. 1 October, and supplement, 2 October. _ _ . 1986c. Survey. 3 November, p. 351. _ _ . 1987. Philippines: First review under standby arrangement. EBS/87/59. 12 March. . 1988a. Philippines: Staff report for the 1988 article IV consultation. SM/88/113. 25 May. ~

611

PhilippineslReferences

_ _ . 1988b. Philippines: Recent economic developments. SM/88/121. 6 June.

Institute of Labor and Manpower Studies (ILMS). 1981. The government programs in the coconut industry: Why pays, who benefits? Research Paper no. 1. Manila: ILMS . Intal, Ponciano. 1987. Fostering nationalism. Manila Chronicle, 17 December, p. 17. Intal, Ponciano, Jr., and John Power. 1987. Political economy of agricultural pricing policies: The Philippines. Final report to the World Bank, September. International Labour Office (ILO). 1974. Sharing in development: A programme of employment, equity and growth for the Philippines. Geneva: ILO. Johnson, Bryan. 1987. The four days of courage. New York: Free Press. Kessler, Richard J. 1988. The Philippines under Corazon Aquino: An assessment of the first two years and the challenges ahead. New York: Asia Society, February. Kintanar, Agustin, Jr. 1976. Recent fiscal reforms in the Philippines. Professional Chair Lectures, Monograph no. 17. Quezon City: University of the Philippines Press. Koppel, Bruce. 1987. Agrarian problems and agrarian reform: Opportunity or irony? In Rebuilding a nation: Philippine challenges and American policy, ed. Carl Lande. Washington, D.C.: Washington Institute Press. Lal, Deepak. 1983. Real wages and exchange rates in the Philippines: 1956-78. World Bank Staff Working Paper no. 604. Washington, D.C.: World Bank. Lamberte, Mario. 1984. The Development Bank of the Philippines and the financial crisis: A descriptive analysis. Staff Paper no. 84-07. Manila: Philippine Institute for Development Studies. . 1985. Financial liberalization and the internal structure of capital markets: The Philippine Case. Staff Paper no. 85-07. Manila: Philippine Institute for Development Studies. Lamberte, Mario, Rosario Manasan, Roberto Mariano, Erlinda Medalla, Manuel Montes, and Celia Reyes. 1985. A review and appraisal of the government response to the 1983-84 balance-of-payments crisis. Monograph Series no. 8. Manila: Philippine Institute for Development Studies. Lande, Carl. 1986. The political crisis. In Crisis in the Philippines: The Marcos era and beyond, ed. John Bresnan. Princeton: Princeton University Press. . ed. 1987. Rebuilding a nation: Philippine challenges and American policy. Washington, D.C.: Washington Institute Press. Laya, Jaime C. 1982. A crisis of confidence. Manila: Central bank of the Philippines. . 1983. Development Bank of the Philippines. Memorandum to President Ferdinand Marcos, 11 August. Licuanan, Victoria S. 1986. An analysis of the institutional framework of the Philippine short-term financial markets. Manila: Philippine Institute for Development Studies. McDougald, C. 1987. The Marcos file. San Francisco: San Francisco Publishers. Magno, Cora P. 1976. Notes on development loan assistance to the Philippines, CY 1956-76. Philippine Economic Journal 3:313-28. Manasan, Rosario G. 1982. Public Finance. In Survey of Philippine development research, vol. 2. Manila: Philippine Institute for Development Studies. Manasan, Rosario G., and Corazon Buenaventura. 1986. A macroeconomic overview of public enterprise in the Philippines, 1975-1984. Staff Paper no. 86-03. Manila: Philippine Institute for Development Studies.

612

Robert S. Dohner and Ponciano Intal, Jr.

Marcos, Ferdinand. 1979. Notes on the New Society. In The democratic revolution in the Philippines, by F. Marcos. Englewood Cliffs, NJ: Prentice Hall International. Mijares, Primitivo. 1976. The conjugal dictatorship of Ferdinand and Imelda Marcos. San Francisco: Union Square Publications. Montelibano, Teodoro. 1983. Government priorities. Business Day (Manila), 2 December, p. 8. Montes, Manuel F. 1986. Financing development: The Politico-economic constraints on fiscal policy in the Philippines. Quezon City: University of the Philippines, School of Economics, mimeo. _ _ . 1987. Macroeconomic adjustment in the Philippines, 1983-85. Working Paper no. 87-01. Manila: Philippine Institute for Development Studies. National Economic and Development Authority (NEDA). 1985. A study of the cucunut industry. Manila: NEDA. NEDA. See National Economic and Development Authority. Nelson, Gerald, and Mercedita Agcaoili. 1983. Impact of government policies on Philippine sugar. Working Paper no. 83-04. Manila: Philippine Institute for Development Studies. Ocampo, Sheilah. 1980. Building a farmer’s empire. Fur Eastern Economic Review, 6 June. Ongpin, Roberto. 1986. The Binondo central bank. Business Day (Manila), 25 April, p. 6. Oshima, Harry. 1983. Problems of heavy industrialization in Asian development. Philippine Review of Economics and Business 20:1-47. Oshima, Harry, E. de Borja, and W. Paz. 1986. Rising national income per worker and falling real wages in the Philippines in the 1970s. Quezon City: School of Economics, University of the Philippines, mimeo, December. Pascual, Alfredo. 1984. Financial institutions and markets in the Philippines. In Financial institutions and markets in Southeast Asia, ed. Michael Scully. New York: St. Martins Press. Patrick, Hugh, and Honorata Moreno. 1985. Philippine private domestic commercial banking, 1946-80, in the light of Japanese experience. In Jupan and the developing countries, ed. Kazushi Ohkawa and Gustav Ranis. Oxford: Basil Blackwell. Presidential Commission on Government Reorganization. 1985a. Draft study on government financial institutions. Mimeo, 15 October. -. 1985b. Rationalization of the government corporate sector. Sector Study H. Mimeo, 27 November. Philippines. 1986. Memorandum on economic policy. Mimeo, 18 September. -. 1987. Economic memorandum. Mimeo, 10 March. Philippines. Ad Hoc Study Group on the Rural Financial System. 1986. Rediscount policy and the arrearages problem. Manila. Mimeo. Philippines. Commission on Audit. 1985. Annual audit report on the Philippine National Bank year ended Dec. 31, 1984. Mimeo, 24 May. Philippines. Department of Budget and Management. 1987. The budget and external assistance requirements. Mimeo, 1 June. Power, John H., and Erlinda Medalla. 1986. Trade liberalization in the Philippines: Assessment of progress and agenda for future reform. PIDS-Tariff Commission

613

Philippines/References

Project on Trade Liberalization. Staff Paper no. 86-01, Manila: Philippine Institute for Development Studies. Power, John H., and Gerardo Sicat. 1971. The Philippines: Industrialization and trade policies. London: Oxford University Press for the OECD. Quiambao, D. 1986. PNB ‘Financed’ Benedicto empire. Philippine Daily Inquirer, 22 May, pp. 1, 6. Race, Jeffrey. 1975. Whither the Philippines? New York: Institute of Current World Affairs, 30 November. Sacerdoti, Guy. 1983. Favouritism still in favour. Far Eastern Economic Review, 30 June, pp. 50-51. Sachs, Jeffrey D. 1985. External debt and macroeconomic performance in Latin America and East Asia. Brookings Papers on Economic Activity 2523-64. Senga, K. 1983. A note on industrial policies and incentive structures in the Philippines: 1949- 1980. Philippine Review of Economics and Business 20. Shaplen, Robert. 1979. A turning wheel. New York: Random House. Sicat, Gerardo. 1986. A historical and current perspective of Philippine economic problems. Monograph Series no. 1 1. Manila: Philippine Institute for Development Studies. Tait, Alan M., W. Gratz, and B. Eichengreen. 1979. International comparisons of taxation for selected developing countries: 1972-76. IMF Staff Papers 26: 123-56. Washington, D.C.: IMF. Tan, Edita. 1980. Monetary policy. In Survey of Philippine development research, vol. I , 170-257. Manila: Philippine Institute for Development Studies. . 1984. Development finance and state banking: A survey of experience. Staff Paper no. 84-04. Manila: Philippine Institute for Development Studies. Tan, Norma. 1979. The structure of protection and resource flows in the Philippines. In Industrial promotion policies in the Philippines, ed. R. Bautista et al. Manila: Philippine Institute for Development Studies. Tengco, Jose R. 1983. Action program for the rehabilitation of DBP. Memorandum to President Ferdinand Marcos. Manila: Development Bank of the Philippines, 25 November. Thompson, Mark, and Gregory W. Slayton. 1984. Credit arrangements between the IMF and the Republic of the Philippines: 1970-1983. Manila, Mimeo, 5 November. Condensed version published as: An essay on credit arrangements between the IMF and the Republic of the Philippines: 1970-1983. Philippine Review of Economics and Business 2259-81. Tidalgo, Rosa Linda, and Emmanuel Esguerra. 1984. Philippine employment in the 1970s. Manila: Philippine Institute for Development Studies. Veloso, R. 1986. Velasco: Marcos’ vicar. Manila Times, 26 March. Villaruel, Teresita, and L. Reyes. 1985. The Philippine debt restructuring program. CB Review 12, no. 2 (February):4-7. Villegas, Bernardo. 1986. The economic crisis. In Crisis in the Philippines, ed. John Bresnan. Princeton: Princeton University Press. Watson, Maxwell, Russell Kincaid, Caroline Atkinson, Eliot Kalter, and David Folkerts-Landau. 1986. International capital markets: Developments and prospects. IMF Occasional Paper no. 43. Washington, D.C.: IMF. Wideman, Bernard. 1976. The dizzy pace of expansion. Far Eastern Economic Review, 26 November, pp. 54-55.

614

Robert S. Dohner and Ponciano Intal, Jr.

World Bank. 1976. The Philippines: Priorities and prospects for development. Country Economic Report. Washington, D.C.: World Bank. -. 1979. The Philippines: Domestic and external resources for development. Report no. 2674-PH. 12 November. -. 1980. Philippines: Industrial development strategy and policies. World Bank Country Study. __ . 1982. Philippines energy sector survey. Report no. 3199a-PH. 12 February. -. 1984a. The Philippines: A review of external debt. Report no. 4912-PH. 2 November. _ _ . 1984b. The Philippines: Public expenditures and their financing. Report no. 4919-PH. 25 November. -. 1985. The Philippines: Recent trends in poverty, employment and wages. __ . 1986. The Philippines: A framework for economic recovery. Vol. 1-3. Report no. 6350-PH. -. 1987. The Philippines: Issues and policies in the industrial sector. Vol. 1-3. Report no. 6706-PH. World Bank and IMF. 1980. The Philippines: Aspects of the jnancial sector. Washington, D.C.: World Bank.

IV

Debt, Adjustment, and Growth: Turkey Merih Celiisun Dani Rodrik

Policy Phases and Adjustment Patterns

1

Turkish Economic Development: An Overview

This monograph is an analysis of the interactions between external debt and internal adjustment in Turkey since the early 1970s. As an oil-importing, middle-income economy, Turkey experienced a series of external shocks after 1973. Correspondingly, it went through a cycle of foreign borrowing and a sequence of sharply altered policy phases. Two important questions arise from Turkey’s experience in this period. First, what were the underlying reasons for Turkey’s debt debacle in the late 1370s? Our interest here centers on the puzzling fact that Turkey entered its debt crisis considerabIy earlier than most other middle-income countries. Secondly, how can we interpret the adjustment experience of the Turkish economy in the early 1980s? This question is of obvious comparative interest as it has become commonplace by now to point to Turkey’s example as a successful case of “life after debt.” In what follows we will attempt to provide synthetic answers to both of these questions. Since this requires a fairly broad view, however, our monograph can also be read as an analytical macroeconomic history of the Turkish economy during the last two and a half decades. We should stress at the outset that our emphasis is on macroeconomic phenomena. This is the natural consequence of the fact that external debt represents the central focus of our narrative. The accumulation and servicing of foreign debt are both linked tightly to the relation between national output and aggregate expenditures. Consequently we will pay only limited attention to many aspects of the microeconomic and institutional structure which have no doubt played important roles in the performance of the Turkish economy since 1973. This limitation notwithstanding, we hope to make clear in our account that Turkey’s relationship with external debt has set the terms for the economy’s growth: debt has acted in turn as an opportunity and a constraint for growth. 617

618

Merih Celksun and Dani Rodrik

To provide context and perspective for the assessments undertaken in the remainder of the monograph, we present in this chapter an overview of Turkey’s development in the earlier periods with broad references to its political, institutional, and structural characteristics. Some basic background data on the Turkish economy are presented in table 1.1. To the reader unfamiliar with the Turkish economy, a quick glance at this table may be useful prior to the historical discussion to follow.’ The table brings out the semi-industrial nature of the Turkish economy in 1985 (see the shares of industry in GDP and total employment) and presents the relevant growth rates from 1953 to 1985. We conclude this introductory chapter with an outline of the policy chronology for the post-1973 period and a plan of the monograph. 1.1 Historical Background

1.1.1

Institution-building Prior to the 1950s

The Republic of Turkey was established in 1923, after the War of Independence following the total collapse of the Ottoman Empire in the late 1910s. Although the newly founded state had a claim on the historical heritage of an empire, its leaders immediately turned to contemporary Western methods and institutions in the early phases of political reconstrucTable 1.1

Turkey: Basic Data

A. Major Indicators, 1985 Population Employment Unemployment rate GNP GNP per capita Gini coefficient

49.8 15.9 12.6 $ 53.0 $ 1,064.0 0.51

B. Sectoral Structure, 1985

GDPa

Agriculture Industry Services

C. Growth Rate (% per year) GNP Agriculture Manufacturing GNP per capita Prices GNP deflator Agriculture deflator Terms of foreign trade

million million percent (excluding labor surplus in agriculture) billion (current prices) (current prices) (1973). 0.50 (1978). 0.525 (1983) Employment

58.9% 22.9 28.2

18.4% 30.6 51.0 1953163 4.8 3.2 8.5 2.0 10.6 10.2

63173 6.7 2.3 10. I

4.0 9.6 10.5

77179 -

6.8 6.8 7.9 4.7

1.2 2.8 1.7 -0.9

3.6 2.4 5.7 12

21.6 23.3 -5.6

56.5 41.8 -3.7

47.4 43.8 -4.5

~

Source: SPO; central bank; Cellsun (1983); and for Gini coefficients, 1978-83, Celisun (1986b).

‘GDP at factor cost.

79/85

73/77 -

619

TurkeyXhapter I

tion. The first fifteen years of the new republican regime, led by the founding president M. Kemal Ataturk, was characterized by deep social and cultural reforms, including, most importantly, the adoption of secular principles in the political life of the country. At the outset, the national leadership faced colossal tasks in the reconstruction of a war-torn and long-neglected economy. Throughout the 1920s, the government grappled with difficulties in two sets of economic policy: (1) the renegotiation and servicing of a huge external debt; and (2) the dismantling of the remaining portions of the so-called capitulations inherited from the Ottoman era. As a balancing factor for political favors received in earlier times, the capitulations granted foreign powers the rights to collect tax revenue and fix import tariffs, which effectively limited domestic policy initiatives to redesign the foreign trade and fiscal regimes for an improved management of the national economy. The externally imposed tax and tariff constraints were largely removed by 1929. To correct the disappointing economic performance in the 1920s and offset the adverse impact of the world economic depression, Turkey instituted a new set of economic policies in the early 1930s, which placed a heavy emphasis on import-substituting industrialization. Turkey’s government-led industrial drive in the 1930s was quite successful in resource mobilization, and generated growth and considerable structural change in output. Major investment projects were implemented within the framework of the first industrial plan in the 1934-38 period. In the mid-thirties, the government (along with the ruling bureaucratic elite) formulated an official ideological position, called etatism (statism). This position sought a middle way between Soviet-style comprehensive planning and a Western-style market economy system. Etatism assigned a leading role to the public sector in savings generation and in carrying out key entrepreneurial functions in industrial development and technological improvement.’ The political appeal of this ideology eventually led to the evolution of a particular form of a mixed economy system in Turkey, which imparted a considerable antimarket bias in the foreign trade regime and the financial system until the liberalization episode of the 1980s. The attempts to implement a second industrial plan during 1938-44 were disrupted by national defense concerns connected with World War 11. During the war, Turkey’s political energies were consumed in maintaining a quasi-neutral stance with a tilt toward the allied powers. Turkey could not escape the devastating economic effects of the external environment and faced severe commodity shortages, black markets, and high inflation in the early 1940s. In the immediate postwar years, two major factors shaped domestic policy and economic performance. First, Turkey obtained access to Marshall Plan aid to Europe and U.S. bilateral assistance programs, which were partly based on defense considerations. The conditions of these foreign aid

620

Merih Cellsun and Dani Rodrik

programs required, however, a shift in economic priorities away from industrial development and toward primary production, as called for by the newly emerging perceptions of the optimal division of labor in Europe. Second, against the backdrop of rising domestic discontent with one-party rule, the government (under the second head of state, ismet inonii) initiated a change toward a multiparty parliamentary system. Given the intense preoccupation with political changes, a draft five-year plan was aborted and industrialization objectives were pushed aside in the late 1 9 4 0 ~ .Following ~ a major exchange rate adjustment in 1946, government policies began to favor agricultural expansion and free enterprise. Average annual GNP growth rates have been estimated as 7.4 percent, 1.2 percent, and 7.9 percent for the periods 1923-38, 1938-48, and 1948-53, re~pectively.~ 1.1.2

1950-60: Democratic Party Rule

In the context of rising domestic dissatisfaction, Turkey’s switch to a multiparty parliamentary framework eventually culminated in the defeat of honu’s Republican People’s Party and a victory for the newly formed Democratic Party (DP) in the May 1950 general elections. The mass basis of the DP was rural conservatives. On economic matters, the DP’s plan was not to plan.’ The Turkish economy expanded rapidly in the early 1950s with the help of a steep rise in agriculture output (due to favorable weather conditions and the extension of farm land to low-yield areas) and primary exports (partly due to the world trade boom connected with the Korean War). However, after the massive crop failure of 1954, the economy entered into a phase of foreign exchange stringency and reduced GNP growth, averaging around 4 percent per year during 1953-58. External debt management and domestic policy increasingly became ad hoc, with a growing reliance on short-term foreign borrowing and trade arrears (much like the scenario to be observed in 1976-77). The central bank financing of public enterprise deficits and agricultural support purchases resulted in high inflation, which eventually led to the reluctant introduction of an IMF-designed stabilization and devaluation program in mid-1958. This program was supported by a sizable package of external financial assistance and debt consolidation under a multilateral agreement. To the students of Turkey’s recent history, a balanced reassessment of the Democratic Party administrations represents, in our view, a continuing research challenge, On the one hand, DP rule in the 1950s stimulated broader political participation and improved the political status of the rural population. On the other hand, the DP governments became increasingly repressive in the face of mounting economic difficulties and rising political dissent by the urban elite. The end result was a tragic one for the top party leaders and came in the form of a complete military takeover in May 1960.

621

TurkeyXhapter 1

Military rule was transitional and ended quickly after the adoption of a socially progressive constitution in 1961, which provided more checks and balances in the overall political process. 1.1.3 1963-73: Experience with Economywide Planning From the standpoint of development policy, a notable feature of the 1961 constitution was the requirement of formal economywide planning through five-year plans and annual programs, the preparation of which was entrusted to the newly established State Planning Organization (SPO). In the formal sense, the planning techniques emphasized the consolidated treatment of government accounts, balanced macroeconomic projections, sector-level consistency studies, and improved methods of project selection. While providing compulsory guidelines for the public sector, the plans have been indicative for the private sector, relying on continually modified mixtures of specific incentives. With the introduction of economywide planning, the style and effectiveness of development policy improved considerably from 1962 onward, exhibiting a greater concern for noninflationary resource mobilization and industrialization. Despite the sluggish growth of agricultural output, annual GNP growth for 1963-73 averaged around 6.7 percent as compared with 4.8 percent in 1953-63. During the first and second plans (1963-67 and 1968-72), the policy emphasis on domestic savings performance paid off quite well, resulting in economywide marginal savings ratios of 32 and 26, respectively, in these two consecutive plan periods. The share of the public sector in total domestic savings was about 45-50 percent, reflecting the significant role of the government in major development programs. However, the pursuit of development in a planned fashion did not fundamentally change the restrictive and largely ad hoc character of trade policies, which discriminated against exports. The annual import programs (containing devices such as quotas, licences, import deposits, and tariffs) served as important policy instruments under the successive five-year plans. These instruments were used more for limiting imports to foreign exchange availability than for evolving a selective and increasingly competitive import-substitution pattern in the economy.6 Under the prolonged maintenance of an increasingly overvalued fixed exchange rate regime, the strains on the external balance intensified in 1969. To prevent the emergence of a payments crisis, the administration of Suleyman Demirel was persuaded to introduce an IMF-supported stabilization program in August 1970, involving a maxi-devaluation. The liberalization objectives of this program were largely abandoned after the partial intervention of the military in March 1971. Concerned with the rising political violence on the left, the chiefs of the armed forces intervened in 1971 mainly to avoid a complete takeover by the younger officers. Their proposed policy remedies were the adoption of some

622

Merih Cellsun and Dani Rodrik

restrictive clauses in the 1961 constitution, and legislation of socioeconomic reforms to restore confidence in the future of the nation. The reforms suggested by the military were ill-defined, however, resulting essentially in the passage of the so-called agricultural reform legislation, which sidestepped the basic task of designing an operationally feasible land reform program for the rural sector. Although it was not a complete takeover, the military intervention in 1971 produced a highly destabilized political structure in the post-1971 period. In the setting of a hospitable world economic environment and a trade boom in the early 1970s, the 1970 devaluation contributed favorably to export and GNP expansion from 1971 to 1973. The most notable development in this period was the surge in remittances from Turkish workers abroad, whose emigration had accelerated in the late 1960s. In light of an unprecedented rise in foreign exchange reserves, an ambitious third plan (1973-77) was adopted with far-reaching objectives of importsubstituting industrialization in capital-intensive sectors. The planned growth process was then seen as part of a large national effort to broaden the productive structure of the country for a more effective integration with the European Economic Community (EEC). 1.1.4 Turkey-EEC Relations Notwithstanding the heavy use of a restrictive trade regime, an important long-term policy choice was made in 1963-73 as regards integration with the EEC. In September 1963, Turkey and the EEC signed the Association Agreement, which envisaged two consecutive stages (preparatory and transitional) before Turkey’s eventual accession to a full member status. Upon the completion of the preparatory stage at the end of the 1970s, the Additional Protocol was signed in November 1970, which became effective in January 1973. This protocol specified the ground rules for the transitional stage, which projected the establishment of a customs union before the full membership stage. In the Additional Protocol, Turkey agreed to remove gradually tariff and nontariff barriers for EEC manufactured exports according to two timetables (over 12- and 22-year periods, as differentiated by products). In turn, the EEC removed tariff barriers for Turkish manufactured exports, except for particular product categories such as cotton yams, textiles, and processed food items in which Turkey had a comparative advantage. The selective trade advantages granted to Turkish agricultural exports rapidly eroded after the EEC’s subsequent agreements with other Mediterranean countries. The EEC also agreed, in principle, to allow free movement for Turkish labor by 1986. Until the emergence of a severe foreign exchange stringency in 1977-78. Turkey carried out tariff reductions as scheduled in the Additional Protocol. Following the 1978-86 period of somewhat cold and strained relations

623

TurkeyKhapter 1

(partly due to Turkey’s internal political difficulties), Turkey formally applied for full membership in the European Community in mid-1987.

1.2 Structural Peculiarities and the Mixed Economy System Before we embark upon the analysis of the post-1973 phases in detail, it would also be useful to draw attention to some of the peculiar structural properties of Turkey’s growth and the major characteristics of its mixed economy system. 1.2.1 Structural Transition: A Cross-Country Comparison Turkey’s historical growth process produced a considerable change in its economic structure. However, in some ways, this structural change lagged behind the norms (or standards) predicted for countries at Turkey’s population size and income level. The leads and lags in structural transition are helpful in identifying the areas where special policy and nonpolicy factors were at work in the growth process. A convenient analytical framework for such a comparative analysis of structural trends is the cross-country regressions of Chenery and Syrquin (1979, which bring out the average (or normal) patterns of development over time. Chenery and Syrquin examine the observed shifts in development characteristics as functions of per capita income for various country groupings differentiated by population sizes andlor net resource inflows. By making use of Chenery and Syrquin’s regressions (for large countries), an earlier study by Cellsun (1983) compared Turkey’s actual structural change with that of an average country with size (measured by population) and per capita income similar to Turkey for the benchmark years 1953, 1963, 1973, and 1978. Table 1.2 presents a selective summary of the results of this study.7 The data shown in this table point to a number of peculiarities in the form of notable deviations of actual values from predicted cross-country norms during the 1953-78 period. Turkey performed substantially below the predicted norms in the accumulation of capital and restructuring of domestic demand and production in the 1950s and early 1960s. However, unlike the preceding ten years, the 1963-73 decade saw (under economywide planning) a considerable catching up in the accumulation and industrialization processes. In turn, external trade ratios were far below the cross-country norms throughout 1953-78. This clearly indicates the inward-orientation of Turkey’s past development strategy. This finding is supported by a further analysis of the sources of growth, carried out for the manufacturing sector and summarized in table 1.3. Compared with other industrializing countries, the demand side effects of external trade have been very small in the expansion of Turkish manufacturing in comparison with domestic demand.

624

Merih Celisun and Dani Rodrik

Table 1.2

Resource Accumulation and Allocational Processes, 1953-78* Actual

A. Accumulation (lo of GDP) I . lnvestment a. Savings b. Gross investment c. Capital inflow' 2. Government Revenue a. Government* b. Taxd B. Allocation (% of GDP) 1. Domestic demand a. Gross investment b. Public consumption c. Private consumption d. Food consumption 2. Production (value added at factor cost) a. Primary b. Industry' c. Utilities d. Services 3. Trade a. Imports b. Exports c. Primary exports d. Manufacturing exports e. Service exports C. Labor allocation 1. lo Share of: a. Primary labor b. Industry labor c. Utilities & service labor

Predictedb

1953

1963

1973

1978

1953

1963

1973

1978

11.5 14.8 3.3

12.0 15.6 3.6

14.8 18.3 3.5

17.8 21.6 3.8

16.3 18.0 2.8

18.0 19.6 1.6

19.6 21.0 I .4

20.5 21.8 1.3

16.0 12.6

16.6 13.9

21.1 18.5

24.4 20.2

15.6 15.0

16.8 15.8

18.7 17.8

19.7 18.6

14.8 11.0 77.6 41.4

15.6 76.6 38.9

18.3 14.1 71.2 28.9

21.6 12.7 69.4 27.4

18.0 12.2 71.5 33.9

19.6 13.6 70. I 30.4

21.0 13.7 68.9 26.7

21.8 13.8 68.2 25.1

45.9 12.8 5.7 35.6

40.1 19.2 8.8 31.8

29.7 22.9 10.7 25.7

26.9 25.0 11.0 37. I

34.4 22.6 6.2 36.2

30.2 23.5 7. I 38.8

24.5 26.8 7.5 40.5

22.0 28.4 7.8 41.2

11.3 8.0 7.4 0.6

9.6 6.0 4.0

I .o I .o

11.4 7.9 4.1 1.8 2.0

9.9 6.1 3.3 1.2 1.6

17.0 17.3 10.8 4.4

16.2 15.2 9.7 4.8

15.6 14.8 8.3 6.0

15.3 14.6 7.6 6.6

77.6 10.1 12.3

64.8 13.6 21.6

60.9 15.3 23.8

56.9 15.8 17.4

53.8 17.9 28.2

47.2 21.7 30.8

44.9 22.9 31.9

-

79.2 7.4 13.4

11.5

~

Source: Cellsun (1983). a

Definitions and measurements follow Chenery and Syrquin (1975, 180-87). Predicted shares may not add to appropriate totals; from basic regressions in Chenery and Syrquin (1975). Capital inflow is net imports of goods and nonfactor services.

Central plus local government revenue, excluding savings bonds and public factor income. Tax revenue in(:ludes SEE corporate taxes. Includes manufacturing and construction.

Another structural peculiarity pertains to the pattern of sectoral allocation of labor in the Turkish economy. Despite the rapid growth of manufacturing, the share of industry (including construction, as in table 1.2) in total employment has remained significantly below cross-country standards. The counterpart of this development has been the very slow pace of reduction in the share of primary labor in total employment. In the Turkish setting, where mining has a negligible share in employment, primary labor largely corresponds to agricultural employment. This structural peculiarity, com-

TurkeyEhapter 1

625

Table 1.3

Sources of Manufacturing Gross Output Increase Over Time: An Intercountry Comparison" (in percentages) Avereage Manufacturing Output Growth Rate (% per year)

Japan 1914-35 1935-55 1955-60 1960-65 1965-70 Israel 1958-65 1965- 68 Korea 1955-63 1963-70 1970-73 Mexico 1950-60 1960-70 1970-75 Norway 1953-61 1961-69 Taiwan 1956-61 1961-66 1966-71 Turkey 1953-63 1963-68 1968-73 1973-77b 1978-81 ~~

Domestic Demand Expansion

Export Expansion

Import Substitution

Change in 1-0 Coefficients

5.5 2.8 12.6 10.8 16.5

69.9 70.9 76.6 82.3 74.2

33.6 -7.1 11.9 21.7 17.6

4.7 15.5 -3.4 -0.3 - 1.4

-8.2 20.7 15.2 -3.7 9.6

13.6 9.4

58.9 68.7

26.2 54.8

9.8 -27.7

5.2 4.2

10.4 18.9 23.8

57.3 70. I 39.0

11.5 30.4 61.6

42.2 -0.6 -2.5

-11.0 0.1 1.8

7.0 8.6 7.2

71.8 86.1 81.5

3.0 4.0 7.7

10.9 11.0 2.6

- 1.0

5.1 5.3

64.3 50.6

37.3 58.8

11.2 16.6 21.1

34.8 49.2 34.9

6.4 9.9 9.4 8.0 -3.0

81.0 75.2 76.2 100.4 - 36.7

14.4 8.2

- 19.4

14.4 10.0

27.5 44.5 57.0

25.4 1.7 3.8

12.3 4.6 4.3

2.2 4.5 10.7 - 1.0 81.5

9.1 10.4 0.6

7.7 9.9 14.6 0.0

- 143.9

- 1.0

~

15.9

- 1.5

~~~

Sourre:

Kubo and Robinson (1979); Lewis and Urata (1983) for Turkey, 1973-81.

'The sources of growth contributions (based on share-total method) in columns 2-5 are measured as percentages of the increment in manufacturing gross output during the indicated periods, and add up to 100 percent. bAll sectors.

bined with more normal patterns of sectoral production, leads to large income differentials among sectors and widens the overall income inequality (as elaborated further in chap. 5).' Our review of Turkey's structural trends suggests that a basic concern of development policy in the 1970s should have been the attainment of a more normal trade orientation in the growth process, even in the absence of the external shocks of the mid-1970s. With a more vigorous exploitation of Turkey's comparative advantage, such a policy shift would have generated a larger labor absorption in nonprimary sectors. Furthermore, the prolonged maintenance of inward-orientation produced rigidities in the output structure,

626

Merih Celasun and Dani Rodrik

which would require large real devaluations and wage cuts to accommodate the trade-liberalization objectives of the post- 1980 adjustment program. 1.2.2 Turkey’s Mixed Economy System Prior to 1980 An analysis of macroeconomic adjustment and debt in the post-1973 period also requires an awareness of the scope and structure of the public sector and the financial system in Turkey’s mixed economy framework. In section 1.1.1 we noted the emergence of etatism as Turkey’s systemic response to the developmental challenges of the interwar period. Under the political conditions of the time, an ovemding concern with self-sufficiency led to the establishment of government-owned and operated public enterprises, not only in social overhead and services sectors, but also in mining and practically all branches of the manufacturing sector. Although they were established as a temporary vehicle for industry-led growth, public enterprises have nevertheless become firmly entrenched in the national economy. After the switch to a multiparty political system in 1950, they served as institutional tools for job creation, regional development, and other extra-market interventions on social grounds. During 1963-77, the share of the public sector in the value added of large manufacturing (including firms with more than ten workers) was around 45 percent. Similarly, from 1963 to 1977, the public sector accounted for about 45 percent of total fixed investment in manufacturing and 50-55 percent of total investment in the economy. In the post-1970 period, however, the economy started to witness an increasingly more vigorous involvement of the private sector in industrial activity. The entrenched position of the public sector in the growth process created strains in the financial system. Public enterprises often required large budgetary transfers, funding from the social security institutions, and deficit financing from the central bank. In such a milieu, the financial system exhibited a highly segmented pattern of growth, requiring financial resources to flow through administered channels. Because of strict controls over interest rates, the financial sector expanded mainly through the rapid growth of its monetary component (including central bank and deposit banks) on the basis of an unusually large credit usage by public administration and public enterprises.’ Table 1.4 provides data on the structural properties of Turkey’s financial system for the benchmark years 1970, 1975, and 1980.’’ Besides showing the predominant share of the monetary system within the overall financial sector, the data also bring out the relatively small volumes of the equity and bond issues by the real sectors, reflecting the virtual absence of a capital market in Turkey in the pre-1980 period. With the limited scope for equity and bond financing, private firms relied on deposit banks and their own resources for capital formation. This situation led to the evolution of sellers’ markets for bank credits, large spreads in interest rates, and strong

627

TurkeyKhapter 1

Table 1.4

Turkey’s Financial System, 1970-80

Total assets of financial institutions (Sof GNP) Distribution of assets 1. Monetary system (Central bank and deposit banks) 2. Investment banks 3. Social insurance institutions 4. Other institutions Total Net issues of domestic nonfinancial (real sectors) 1. Total (% of GNP) 2. By sector (%) a. Public sector Administration Enterprises b. Private firms c. Households and others” Total 3. By type (%) a. Equities b. Debt issue Bonds Nonbondsh Total 4. Held by (%) a. Financial system b. Monetary system

1970

1975

1980

76.1

80. I

66.6

68.4 12.4 14.4 4.8 100.0

73.3 11.5 11.8 3.4 100.0

84.7 6.3 6.4 2.6 100.0

59.7

57.8

47.8

48.2 27.3 20.9 37.3 14.5 100.0

42.9 23.6 19.3 47.6 9.5 100.0

55.1 31.8 23.3 37.3 7.6 100.0

8.9

11.2

6.9

11.3 79.8 100.0

8.6 80.2 100.0

8.6 86.6 100.0

87.1 57.0

83.9 59.7

85.8 71.1

Source: Akyuz and Ersel (1984, annex 2) for financial assets; Akyuz (1984, tables 4.1 and 4.4) for other

data. a

Includes agricultural producers and nonprofit organizations. Credits, advances, etc.

preference for a restrictive trade regime to sustain high-cost industries established for home markets. Thus, the simultaneous use of three major institutional instruments, namely (1) public enterprises, (2) a restrictive trade regime, and (3) financial repression, produced a highly compartmentalized mixed economy system. Prior to 1980, the domestic policy debate centered on the relative sizes of the public and private sectors, rather than on the improvement of the integrative price and planning mechanisms within the Turkish economy. Finally, two remarks on our data presentation are in order. First, the public sector concept in our macroeconomic discussions is a broad one, covering institutional components such as the central government, local government, state economic enterprises, and various adjunct entities. In this context, the public sector deficit refers to a wider aggregate than the deficit in the so-called consolidated budget of the central government. Second, it should be noted that the interchangeable use of the terms “public enterprises’’ and “state economic enterprises (SEES)” is not strictly correct under Turkey’s

628

Merih Celiisun and Dani Rodrik

legal arrangements. I’ The SEEs constitute the largest subset of public enterprises and are supposed to function under commercial business principles. The non-SEE enterprises are viewed more as public agencies involved in the production of public goods in the conventional sense. In turn, the SEEs have nonfinancial (operational) and financial subcategories. Since our overall concern in the present study is the aggregative analysis of debt and adjustment, we often eschew these institutional distinctions and crudely use the terms SEEs and public enterprises interchangeably, unless noted otherwise.

1.3 Policy Chronology and A Road Map It is possible to delineate in retrospect three major policy phases subsequent to the first oil shock of 1973. These phases correspond roughly to 1973-77, 1978-79, and post-1980. By way of introduction, we present a brief overview of these three phases. 1.3.1 Phase 1 (1973-77): Debt Accumulation and Postponed Adjustment The perceived need to accelerate and deepen the industrialization effort formed the background to the official planning process throughout much of the 1970s. Thus, Turkey’s response to the external shocks of the mid-1970s became one of postponed internal adjustment, which turned out to be feasible through reserve decumulation initially, and heavy short-term borrowing subsequently, up until mid- 1977. The largely unnoticed buildup of price distortions in Phase 1 caused not only a stagnation in exports but also a rapid rise in the import-intensity of domestic production. Moreover, the government’s external debt “strategy” promoted overborrowing on the part of the private sector. The latter precipitated a severe payments crisis by mid- 1977, when practically all reserves became depleted and available bank lines were terminated. 1.3.2 Phase 2 (1978-79): Foreign Exchange Crisis and Inflationary Response During this phase, Turkey secured important debt reschedulings but lacked adequate flows of fresh foreign credits to halt a rapidly worsening position in the balance of payments. With insufficient domestic policy remedies and uncurbed monetary expansion, the real sector adjusted to import compression via an inflationary process of output contraction. 1.3.3 Phase 3 (from 1980 on): Stabilization and Outward-Oriented Adjustment With the introduction of a mix of stabilization and liberalization policies in January 1980 and thereafter, the economy embarked upon a new adjustment path with a greater reliance on export expansion and market

629

TurkeyKhapter 2

forces. Turkey received sizable volumes of new lending and debt relief in 1980-84. The new policy stance produced an export-led recovery and acceptable degree of creditworthiness by 1982-83, just as most of the LDC debtors were entering a deep crisis phase in their development process. After the termination of debt relief in 1984, Turkey began to face an increase in its external debt service. This strained the fiscal position and required a large rise in domestic borrowing at high real rates of interest. 1.4

Plan of the Monograph

Our monograph is organized in two parts. Following the broad retrospective provided on Turkish economic development in this chapter, in part 1 (chapters 2 to 5) we examine the aggregate performance and adjustment patterns from 1973 to 1986. These chapters constitute an analytical chronology of the policy phases outlined above. In part 2 (chapters 6 to 10) we focus on selected aspects of internal adjustment and external debt. Chapter 6 presents the principal findings of a multisector general equilibrium analysis and evaluates the interactions among external borrowing, trade liberalization, and exchange rate policy. Chapter 7 explores in greater detail the sources of Turkey’s export boom in the post-1980 period. In chapters 8 and 9 we assess public finance and external debt management, respectively. In chapter 10, we recapitulate our conclusions and discuss the prospects for the future of debt management in Turkey. An appendix contains a political chronology, as well as supplementary tables on subjects covered in the main text of the monograph.

2

Economic Boom and Debt Crisis, 1973-77

For the Turkish economy, the 1970s were the best of times and the worst of times. The decade witnessed an unprecedented spurt of investment and growth until about 1977, accompanied by what looked like a steady improvement in income distribution. That was followed by a crash which was equally unprecedented. From mid-1977 on, Turkey found itself in a monumental debt crisis which took several years of intricate negotiations with creditors and a long series of rescheduling agreements to resolve. Growth suffered heavily, with two years of real contraction at the end of the decade, and income distribution turned sharply against urban workers and the peasantry.

630

Merih Celisun and Dani Rodrik

This chapter and the next are devoted to providing an analytical overview of this boom-and-bust experience. The present chapter is concerned with the period immediately preceding the debt crisis of 1977, providing an interpretation of the economic boom as well as an explanation for the ultimate crisis. The period of forced adjustment between the onset of the crisis in mid-1977 and the reform package of January 1980-a period of vast importance despite its short duration-is the subject of chapter 3. With hindsight, it is not too difficult to provide a broad interpretation of the Turkish experience prior to 1977. The early years of the decade were a time of great optimism as the perennial foreign exchange constraint appeared to have been permanently relaxed, thanks largely to a rapid rise in workers’ remittances. As table 2 . I shows, the current account was actually in surplus for two years in a row in 1972 and 1973. Partly as a consequence, the public sector went on an investment binge shortly thereafter and encouraged the private sector to follow suit. As the share of investment rose from 18.1 percent of GNP (in 1973) to 25.0 percent (in 1977), the real growth rate of the economy reached its zenith at 8.9 percent (in 1975 and 1976). There were two problems, however. First, all of this was taking place in the context of the fourfold rise in world oil prices. Second, the government succumbed to all of the usual policy pitfalls: price distortions, including overvalued exchange rates, and large public sector deficits. These helped swing the current account sharply into deficit, moving it from a surplus of $534 million in 1973 to a deficit of $3,431 million in 1977. The current account deficits were financed by external borrowing, much of it of short-term maturity. As foreign lenders started getting jittery at the beginning of 1977, the stage was set for a debt crisis. Capital flows slowed down to a trickle, and Table 2.1

Year 1972 1973 1974 1975 1976 1977 1974-77 average 1978 1979 1980 1978-80 average

Macroeconomic Performance of Turkey During the 1970s

Real GDP Growth (%) 6.0 4. I 8.8 8.9 8.9 4.9

Current Account Balance (million %)

( % of GDP)

18.0 20.5 29.9 10.1 15.6 24.1

47 534 - 662 - 1,889 -2,286 -3,431

20. I 18.1 20.7 22.5 24.1 25.0

Inflation Rate (WPU (%)

Investment

7.3

19.9

- 2,067

23.2

4.3 -0.6 - 1.0

52.6 63.9 107.2

- 1,595 -

1,203 -3,304

18.5 18.3 21.4

0.9

74.6

-2,034

19.4

Sources: State Institute of Statistics (SIS),SPO, and the central bank of Turkey

631

TurkeyIChapter 2

the central bank’s depleted reserves forced it into arrears on payments to foreign banks, governments, and export suppliers. The consequent foreign exchange shortages led to a forced reduction of the current account deficit via administrative means, the collapse of investment and growth, and an upsurge of inflation (see table 2.1). The next few years witnessed a series of debt renegotiations with creditors. What were the sources of this debt debacle? Conventional wisdom stresses the adverse external environment and the short-term nature of the liabilities incurred during 1973-77. But more must have been at work. Until the debt crisis of 1982 came along, Turkey’s debt problems were among the most severe experienced by the postwar international system, and its debt reschedulings were the largest undertaken to date. As table 2.2 reveals, Turkey alone accounted for 69.0 percent of the total volume of debt renegotiated by developing countries in the 1978-80 period. These facts point to a peculiar aspect of the Turkish experience. Unlike practically all other newly industrializing countries experiencing debt difficulties, Turkey got into trouble after the first oil shock, rather than the second one. This suggests prima facie that the usual explanations of the crisis in terms of a combination of external shocks with a number of key inappropriate domestic policies-such as overvalued exchange rates and a lax monetary and fiscal stance-will go at best only part of the way in explaining its origins.’ Unless it can be demonstrated that the shocks were particularly severe and/or the policies particularly excessive, the analyst has to look for additional reasons for Turkey’s precocious debt crisis. We argue in this chapter that the key to the puzzle is Turkey’s dependence between 1975 and 1977 on a form of foreign borrowing with intrinsically destabilizing features. To attract capital inflows, the authorities relied disproportionately on the “convertible Turkish lira deposit” scheme, whose Table 2.2

Debt Renegotiations, 1975-84: Turkey’s Share in the Total Amount of Renegotiated Debt

Turkey’s Share in Total

Year

Number of Countries Renegotiating

All LDcs

Turkey

(%)

1975 1976 1977 1978 1979 1980 1981 1982 1983 1984

2 3 3 3 7 6 13 9 21 23

373 1,350 373 2,195 6,564 5,323 2.157 2,382 51.08Y 116.220

-

-

-

-

-

-

1,612 3,898 4,200 100

73.4 59.4 78.9 3.6

Sources: World Bank, WorldDevelopmenr Reporr 1985. p. 28, except for Turkey, for which the same sources as those in table 2.1 have been used. The LDC totals given by the WDR have been adjusted to account for discrepancies between the Turkish series used here and those given in the WDR.

632

Merih Cellsun and Dani Rodrik

key feature was that it protected domestic borrowers from all exchange risk. As will be explained below, this scheme had the fatal flaw of engendering an ever-expanding spiral of overborrowing by the private sector. Even though the counterpart to the current account deficits was, in an accounting sense, an investment boom by the public sector, the boom was sustainable only to the extent that foreign banks were willing to increase their exposure to Turkey at an ever-increasing pace. Once foreign banks slowed their net lending, the edifice collapsed. Hence it was primarily the dynamics of the debt accumulation process itself that was responsible for the early onset of the crisis. Fiscal, monetary, and exchange rate policies of the authorities would likely have gotten the country into trouble eventually. But the borrowing “strategy” in place ensured that this would come sooner rather than later.

2.1 External Shocks and Policies: International Comparisons To put Turkey’s experience in the mid-1970s in its proper perspective, it is useful to start with a brief comparative look. Was Turkey subjected to larger external shocks than other developing countries? There is some evidence that the oil shock of 1973-74 had somewhat more severe consequences on the balance-of-payments position of Turkey than in most other similarly placed countries. In table 2.3 we summarize Bela Balassa’s (1984) findings with respect to the magnitude of the shocks experienced during 1974-76 by twenty-four oil-importing countries, among which Turkey is included. Balassa’s calculations cover the terms-of-trade effect, as well as the reduction in export demand due to the recession in industrial countries.2 It appears that Turkey fared relatively badly on the first score, largely because Turkey’s exports scarcely benefited from the offsetting price rises common to many commodities besides oil. Moreover, Balassa’s calculations understate the magnitude of the shocks experienced by Turkey. These do not take into account the fall in workers’ remittances consequent upon the reduction of economic activity in Germany and other countries in which the Turkish Gastarbeiter were concentrated. To

Table 2.3

Balance-of-Payments Effects of External Shocks, 1974-76 (as a percentage of GNP)

24 Developing Countrics” Turkey

Terms-of-Trade Effect

Export Volume Effect

- 4.2

-0.7

-

11

-0.4

-1.7

~

~

Sourcer Balassa (1984) and own calculations (see text)

“Includes Turkey

Remittance Effect

Total -4.9 - 9.2

633

TurkeyiChapter 2

correct for the omission, we have followed Balassa's procedure to estimate the balance-of-payments effect of the derived reduction in workers' remittances. The estimate reported in table 2.3 is based on a conservative procedure which assumes the following: (a) in the absence of the reduction in foreign activity, remittances would have continued to grow at the trend rate observed during 1972-74; and (b) with the reduction in foreign activity but no significant additional policy distortions, the level of remittances would have remained at the 1974 level throughout 1974-76.3 This procedure adds another - 1.7 percent (of GNP) to the external shocks faced by Turkey, bringing the total to -9.2 percent for the 1974-76 period. By contrast, the average shock for Balassa's sample of oil importers amounted to -4.9 percent. The effect of this difference should not be exaggerated, however. Even with the remittance effect added for Turkey, Balassa's figures show nine countries (out of twenty-four) with more severe shocks, including Yugoslavia, Philippines, Portugal, Israel, and Korea. Turning to economic policies, the next question is whether policies in Turkey in the immediate aftermath of the first oil shock were considerably more distortionary than in other countries. We will analyze these policies in more detail below, but for the moment the tentative answer has to be: not really. This can be seen by concentrating on three aspects of policy which have borne the brunt of criticism: exchange rate policy, budget deficits, and pricing of domestic energy. In table 2.4 we summarize evidence on these policies for Turkey and a sample of other developing countries. While Turkey appears to have been hardly a paragon of virtue in these respects, it

Table 2.4

Policy Comparisons, 1973-77

Argentina Brazil Chile Mexico Korea Philippines Indonesia Perub Turkey ~~

Domestic Energy Price as a % of World Price. I977

Appreciation of Real Exchange Rate, 1973-76

Government Budget Deficit. 1975-76 (% of GNP)

Regular Gasoline

Residual Fuel Oil

5.5 -6.0 3.0 0.5 19.4 15.6 54.6 4.4 9.9

7.8 0.3 2.3 4.8 I .7 I .5 4.3 3.5 1.7

213 385 233 I25 n.a. 200 173" 238 141

61 112 213 n.a. ma. 137 ma. 42 97

~

Sources: Exchange rates from Morgan Guaranty, World Financial Markers; budget deficits from World Bank, Wurld Tables; energy prices. unless othenvise noted. from Fallen-Bailey and Byer (1979). "Calculated using data for premium gasoline from U .S. Department of Energy, lnrernatiunal E n e r p Annual. bPeru experienced debt-servicing problems in this period. and renegotiated in 1978- 79.

634

Merih Cellsun and Dani Rodrik

nonetheless was not a particularly promiscuous offender of economic rationality either. The proximate cause for Turkey’s debt crisis can be observed in the rapid deterioration of its current account during the period. In table 2.5 we compare Turkey with other developing countries in that respect. Here we see the relatively quick turnaround of the Turkish current account from a surplus of 2.4 percent of GNP in 1973 to a deficit of 5.1 percent in 1975, slightly higher than for all developing countries. This more rapid deterioration is consistent with the possibility that external shocks were indeed more severe in the Turkish case than on average. However, the distinguishing aspect of the Turkish performance comes after 1975: whereas other oil-importing countries managed to reduce their deficits to 2.1 percent of GNP by 1977, Turkey’s deficit continued to grow and reached 7.1 percent. The initial deterioration can be accounted for by the oil shock, but the trend after 1975 requires additional explanations. What lay behind these deficits, and were they large enough to have brought about the crisis? A preview of the arguments contained in the next two sections would go as follows. First, the counterpart to these deficits was an increased investment effort, mainly by the public sector. Hence, external borrowing was used primarily for investment purposes and not for consumption. Secondly, while consumption and investment decisions in the economy were considerably distorted by inappropriate pricing policies, mainly an overvalued exchange rate, these alone would not have brought about the crisis. What probably tipped the balance was the dynamics of the debt process itself. To prevent private sector crowding-out and to ensure foreign exchange availability for its own needs, the government subsidized private sector foreign borrowing by providing blanket protection against foreign exchange risk. As we shall show below, this type of external financing contained the germs of its own destruction. The implicit subsidy on foreign borrowing was larger the greater the likelihood of a crisis; in turn, the crisis became more likely as borrowing skyrocketed. Hence, while the underlying cause of the deteriorating external balance has to be located in the public sector investment drive, what precipitated the debt crisis per se was private sector borrowing behavior, itself in turn conditioned by government policy.

Table 2.5

All LDCs Oil-importing LDCs Turkey

Current Account Balances, 1972-77 (as a percentage of GNP) 1972

1973

I974

1975

1976

1977

1.7 1.5 0.3

- 1.3 - 1.1

- 2.3

-4.2 -4.3 -5.1

-2.8 -2.6 -5.4

- 2.6 -2.1 -7.1

-

2.4

-3.9 -2.2

Sources: World Bank, World Developmenr Reporr 1985, p. 17; and central bank of Turkey

635

TurkeyIChapter 2

2.2 Public Investment, Current Account Deficits, and Debt The political scenery of the 1970s was replete with instability and volatility, and no economic account of the period is complete without at least lip service to this fact. After 1973 Turkey was governed by a series of coalition governments of varying political outlooks. Following the defeat of the right-wing Justice Party in the 1973 elections, Demirel was replaced as premier by Bulent Ecevit who led an awkward coalition between his left-of-center Republican People’s Party and the Islamist National Salvation Party. In March 1975, a new coalition of right-wing parties brought Demirel to power once again. This coalition lasted until the general elections of June 1977 which proved indecisive. After an unsuccessful try by Ecevit, Demirel was then able to resuscitate his previous coalition, which lasted however only until January 1978. Ecevit’s minority government which replaced it collapsed in turn in October 1979, enabling Demirel to return to power once again. The lack of decisive central authority during those years is frequently alleged to have been the main source of inadequate economic policymaking. While this is no doubt true, it should not cloud the fact that a series of weak governments of varying political ilk still managed to undertake an impressive and sustained investment boom. Table 2.6 documents the steady increase in the investment ratio after 1973, rising from 18.1 percent of GNP to 25.0 percent in 1977. That the investment effort was spearheaded by the public sector is equally clear. Public investment almost doubled from 7.0 to 13.1 percent of GNP, while the private sector investment rate remained roughly constant in the 10- 12 percent range. Table 2.6

Investment-Savings Balance and Growth of Real Expenditures, 1973-77

investment Private Public Domestic savings Private Public Foreign savings Sectoral savings-investment balances Private Public Total Growth of real expenditures Private Public Total

1973

1974

1975

1976

1977

18.1 7.0 20.3 11.6 8.8 -2.2

20.7 10.0 10.8 18.4 11.0 7.4 2.3

22.5 10.3 12.2 17.4 8.5 9.0 5.0

24.7 13.1 11.6 19.3 11.2 8. I 5.4

25.0 11.9 13. I 18.0 11.7 6.4 6.9

0.5 1.8 2.3

I .o -3.4 - 2.4

- 1.8

- 1.9

-3.2 -5.0

-3.5 -5.4

3.1 1.4 4.5

7.3 10.3 7.9

11.1

7.4 20.2 10.0

9.6 12.5

10.2

-0.2 -6.7 -6.9 2.7 9.0 4.2

Source: SPO.

Note: Investment-savings balance data reported as a percentage of GNP. Growth of real expenditures excludes expenditures on inventories and is reported in percentages.

636

Merih Cellsun and Dani Rodrik

The structure of investment reflected the underlying economic philosophy of the various governments in power, with industry and infrastructure receiving emphasis. In table 2.7 we show the distribution of investment by major sectors during 1973-77 and compare the breakdown with the earlier 1968-72 period. The only important difference is the greater emphasis on transportation projects after 1973. In the later period, investment in transportation accounted for no less than a quarter of total public sector investment. It is tempting to speculate about whether the ultimate outcome would have been much different had a greater share of investment been allocated to tradables sectors. In all likelihood, the microeconomics of project selection played only a secondary role in precipitating the debt crisis. First, it is difficult to argue that the social rates of return to infrastructure projects in a country like Turkey are systematically lower than in, say, agriculture. Second, as the Turkish experience of the 1980s shows, it would not have been too difficult to generate an export boom with an unchanged economic structure, once the appropriate macroeconomic environment was established. The deliberate expansion of investment in this period was not accompanied by policies that would ensure a commensurate level of domestic resource mobilization; this is perhaps where the fragility of the governments of the time most clearly exhibits itself. As shown in table 2.6, the aggregate saving rate actually fell between 1973 and 1977, reflecting the consumptionstimulating influence of the growing overvaluation of the exchange rate (on which more later). The private sector’s contribution to the public sector’s savings-investment gap was nil, as the former became incapable after 1974 of generating enough resources even for its own investments. The private sector balance was in deficit at the level of 2 percent of GNP during 1975-76 until the foreign exchange crisis of 1977 brought private expenditures crashing down. Table 2.7

Structure of Fixed Investment, 1968-72 and 1973-77 (in percentages)

1968-72

Agriculture Industry Manufacturing Mining Energy Transportation & communication Health and education Housing Others Total

Public

Private

13.0 42.4 21.8 4.9 15.7 21 .O 9.6 3.6 10.4 100.0

9. I 34.7 31.9 1.4 1.4 10.3 0.3 38.5 7.1

Source: World Bank (1980). tables 8 and 9.

100.0

I973 - 77 Total

Public

Private

Total

11.1

10 0 44.5 24.0 6.6 13.9 25.0 6.4 2.3

13.4 36. I 35. I 0.6 0.4 15.5 0.2 32.6 2.2

11.6 36.7 26.3 3.6 6.8 21.6 4.6 17.9 7.6

100.0

100.0

38.9 26.7 3.3 8.9 16.0 6.3 20. I 7.6 100.0

11.8 100.0

637

TurkeyXhapter 2

As the boom in investment would suggest, the remarkable aspect of the foreign borrowing experience in this period was the use of these foreign funds for investment, rather than consumption. The figures in table 2.6 amply attest to this. Net use of foreign savings rose in this period from - 2.2 percent of GNP in 1973 to 6.9 percent in 1977.4 This amounts to a rise in the external borrowing ratio of 9.1 percentage points. Of this increase, 6.9 percentage points (or 76 percent) are accounted for by the rise in investment, and only 2.3 percentage points by the rise in cons~mption.~ The role of the public sector is also clear: 92 percent of the increase in the net foreign savings ratio is accounted for by the deterioration of the public sector balance, and only 8 percent by the decrease in net private savings. This is clearly a case of foreign borrowing to finance public investment. Whether it amounted to overborrowing or not is an important question which will be addressed below. The deterioration of the public sector finances can be observed more clearly in the trends of the public sector borrowing requirement (PSBR). In table 2.8 we adjust the public sector savings-investment gap calculated from the national accounts (and displayed in table 2.6) with additional financial items to amve at an aggregate PSBR.6 The results show the rise in the PSBR over the 1973-77 period to be equally dramatic: from 2.0 percent of GNP to 10.6 p e r ~ e n t . ~ In table 2.8 we also present the available evidence on the modes of financing of these deficits. Despite a somewhat large "other" category for certain years, the inescapable conclusion is that foreign borrowing did not

Table 2.8

The PSBR and its Financing, 1973-77 (as a percentage of GNP)

Public disposable income Consumption Savings Public investment Public savings-investment gap Public-private capital transfers Inventory revaluation fund (SEES) Increase in accounts payable, net PSBR Financing: External borrowing, net Domestic borrowing Long term Short term (Treasury bills) Central bank, net Other"

1973

1974

1975

1976

1977

20.7 11.9 8.8 7.0 1.8 -2.5 1.3 n.a.

18.4 11.0 7.4 10.8 -3.4 -0.8 - 1.0 n.a.

20.9 11.9 9.0 12.2 -3.2 - 1.3 -0.6 - 1.0

20.6 12.5 8.1 11.6 -3.5

19.7 13.3 6.4 13. I -6.1 -2.5 -2.5

2.0

5.1

6. I

6.6

10.6

0.3 1.4 I .4 - 0.0 1.9 2.6

0.8 1.5 I .5

0.5 2. I I .4 0.7 6.6

~

0.6 0.6 0.9 -0.3 0.2 0.6

-0.2 1.1

0.8 0.3 3.6 0.6

- 1.4 - 1.5 -0.2

0.0 4.5

-0.2

Source: SPO, IMF. and the central bank of Turkey. "Includes changes in holdings of deposits and currency, SEE arrears, and errors and omissions

1.1

1.3

638

Merih Cellsun and Dani Rodrik

play an important role in financing the public sector directly.' Foreign borrowing by the public sector remained well below 1 percent of GNP throughout the period. The single largest source of financing was instead the central bank, which provided more than half of the funds needed over the period 1974-77. Now this might seem surprising in view of the close links drawn above between the public sector imbalance and foreign borrowing. The apparent contradiction is resolved by looking at the financing of the public sector in general-equilibrium rather than partial-equilibrium terms. Foreign borrowing did indeed finance the public sector, but it did so indirectly via the intermediation of the banking sector, and of the central bank in particular. To see how the system worked, we have to turn to an analysis of the nature of the external liabilities incurred during this period. Since the current account deficit is the mirror image of the domestic savings-investment imbalance, we first take a look at the financing of the former. Table 2.9 shows that the cumulative current account deficit from 1974 to 1977 amounted to $7.5 billion. Around 17 percent of this deficit was financed by running down the reserves of the central bank, and 8 1 percent by borrowing, with foreign direct investment playing an insignificant role. By far the most important item among the financing entries, however, is short-term borrowing, which accounted for more than half of the cumulative deficit. As will be discussed below, short-term borrowing was typically channeled through the central bank, and did not constitute a liability of the consolidated government or of the SEES. Long-term borrowing, most of which did constitute a liability of the public sector, made up only 22 percent of the cumulative deficit.

Table 2.9

Financing the Current Account, 1973-77 (million $)

Current account balance Nondebt financing Foreign direct investment Change in reserves' Counterpart to valuation changes Net foreign borrowing Long term IMF Implied short term

1973

1974

1975

1976

I977

I974 - 77

534 625 79 - 704

-662 457 33 424

- 1,648

-2,029 94

- 3,140

54

375 27 349

-7,479 1,429 (19.1%) 184 (2.5%) 1,256 (16.8%)

n.a. 91 349 0 - 258

n.a. 205 197 0 -8

- 40 1,145 I48 248 749

30 1,935 509 143 1,283

-1

- 1 1 (-0.1%)

2,765 782 18 1,965

6.050 (80.9%) 1,636 (21.9%) 409 (5.5%) 3,989 (53.3%)

~

503 114 429

10

Sources: Current-account and nondebt financing figures are from the central bank of Turkey; long-term debt flows are from the World Bank, World Debt Tables, various issues; net borrowing from the IMF is from the central bank of Turkey and the IMF, International Financial Statistics; short-term borrowing has been calculated as the residual, and includes errors and omissions and arrears besides the usual forms of short-term lending. Note: Numbers in parentheses are the percentage distribution of the current account financing.

"A positive (negative) number denotes decrease (increase) in reserves.

639

TurkeyIChapter 2

A better idea of the predominant role of short-term debt can be obtained from table 2.10 in which Turkey’s outstanding external debt is disaggregated by maturity and type of liability. Notice that while these stock statistics tell broadly the same story as the flow statistics of table 2.9, the two sets of figures are not perfectly consistent with each other as flows do not exactly match the difference between stocks. Beside the usual statistical problems with recording and omissions, the discrepancies are also due to changes in valuation as cross-rates fluctuate. Hence, since a considerable share of Turkey’s short-term debt is denominated in deutsche marks, the dollar value of Turkey’s external debt rises when the deutsche mark appreciates (as it did after 1975, for example). From the perspective of debt management, such fluctuations create special problems, some of which we will discuss below. Leaving these valuation problems aside, table 2.10 documents the phenomenal rise in short-term liabilities after 1974. Within the span of three years, short-term liabilities rose from a meager 6.4 to 54.0 percent of total external debt. To put this rise in perspective, note that the comparable figure for all oil-importing developing countries barely budged from 14.1 to 15.3

Table 2.10

External Debt and its Composition, 1973-77 (million $) 1973

Total external debt Long-term debt Public and publicly guaranteed Private nonguaranteed IMF Short-term debt Convertible TL deposits (of which: overdue) Banker’s credits Overdrafts Dresdner Bank scheme Acceptance credits Petroleum credits Commercial arrears Third-party reimbursement claims Suppliers’ credits (“cash against goods” scheme) Other Memo: Share of short-term debt in total (%) Turkeyb Nonoil LDCs

3,263 2,984 2,867 115 0

279 225 (0) 9 0 0

1974 3,494 3,271 3,126 146 0

223 I45 (0) 0 7 0 0

I975 4,723 3,325 3,165 160

243 1,155 999 (0)

12 32 0 0

0

0

0

0

0

0

1976 7,280 3,838 3,590 248 39 1 3,051 1,781 (0) 413 I24 48 0 0 234

1977 11,280 4,779 4,300 479 408 6,093 2,267 (241) 384 240 I73 560 359 1.712 (204)

45

71

8.9 14.1

6.4 14. I

112

24.5 14.3

(234) 45 I

41.9 14.6

(1508) 398

54.0 15.3

Sources: World Bank, IMF, and the central bank of Turkey. Data for short-term debt of nonoil LDCs are from IMF, Recent Multilateral Debt Resrructurings with Officinl and Bank Creditors, December 1983, table 1.

aFigure for banker’s credits for 1973 is included under convertible TL deposits for that year. bLiabilities to the IMF have not been included under short-term debt.

640

Merih Celgsun and Dani Rodrik

percent. The overwhelming majority of this short-term debt (around 85 percent) was, directly or indirectly, a liability of the central bank. The hallmark of the foreign debt experience in this period was an innovative form of borrowing called the ‘‘convertible Turkish lira deposits” (CTLDs) scheme. As the figures in table 2.10 show, this type of short-term debt dominates in volume all the rest. At the end of 1977, CTLDs amounted to 37.2 percent of all short-term liabilities. This share rises to 48.9 percent if we exclude payments arrears and overdue CTLDs. At the end of 1976, before the impending crisis slowed the flow, they accounted for 58.3 percent of all short-term debt. Hence, the CTLD scheme played a determining role at the margin in financing the current account deficits of the period. What were the CTLDs? In May 1975 the pressure from the balance of payments led the government to resuscitate an old scheme whereby nonresidents could open deposit accounts with Turkish commercial banks.’ Principal and interest payments on these deposits were guaranteed by the government (effectively by the central bank) against all foreign exchange risk arising from devaluations of the Turkish lira. The interest rate ceiling on such deposits was initially set at 1.75 percentage points above the Euromarket rate for the corresponding currency. Upon the opening of the CTLD account, the recipient bank would turn over the foreign currency to the central bank and have its account with the central bank credited with the Turkish lira equivalent of the deposit. The local bank could then extend loans denominated in liras to domestic firms. As interest and principal payments on CTLDs became due, the foreign lender would recover the original deposit plus interest, both in foreign currency. The central bank would in turn cover any loss experienced by the local bank due to the depreciation of the Turkish lira since the opening of the CTLD account. The CTLD scheme was a wonderful system whereby the public sector could finance its deficits, while the private sector, far from being crowded out, ended up itself as the beneficiary of a credit explosion. To see this in greater detail, it pays to follow how CTLD funds were cycled (table 2.11). For the sake of concreteness, consider the consequences for the banking sector of an increase in foreign liabilities of D M l O in the form of a CTLD. Table 2.11

The CTLD Scheme Assets

Step 1 Commercial bank Step 2 Commercial bank Central bank Step 3 Central bank

Liabilities

Foreign exchange reserve of DMlOO

Deposit of DMlOO

Loan of TL528 Reserves of DMlOO

Deposit of DMlOO Increase in base money of TL528

TL528 domestic credit to government

Increase in base money of TL528

641

TurkeyKhapter 2

In step 1 of the process, the domestic bank holds a deposit liability of DM100, balanced by its holdings of DMlOO in foreign exchange. Next, the commercial bank turns the DMlOO over to the central bank, in return for which it is credited the Turkish lira equivalent of DMlOO, say TL528 (at the rate prevailing at the end of 1975). Abstracting from reserve requirements, the commercial bank can in turn use this to make loans to its customers. In step 2, there has been an increase in base money arising from the credit extended to the commercial bank by the central bank. In the final step, consider what happens as the public sector borrows from the central bank to finance its investment drive. The increased public sector expenditure eventually turns up in the form of imports, requiring foreign currency to be supplied by the central bank. When the process is complete, the money supply has increased (by 528 times the money multiplier minus one), the private sector enjoys new credits, and the government has found the foreign exchange with which to finance its investment. lo There are two noteworthy aspects of this type of borrowing. First, the government's ambitious investment program was being funded by accumulating liabilities of short-term maturity. Indeed, most of the CTLDs were of one-year maturity, and the rollover rate appears to have been less than 40 percent (Brennan 1976). These funds were being used to finance investment projects with considerably longer gestation lags. " Hence such foreign borrowing was inherently risky given the maturity transformation involved. Second, the level of such borrowing was determined not by the central bank or the public sector at large, but by the private sector, even though the major part of the resources thus mobilized ended up being used by the public sector. Essentially, the public sector investment drive acted as a powerful vacuum into which all foreign exchange brought in by the private sector would be quickly sucked. This played a crucial role in precipitating the debt crisis. The reason is that the incentives provided to the private sector by the CTLD scheme were fundamentally destabilizing. The argument here requires a closer look at the operation of the CTLDs during the 1975-77 period, to which we now turn.

2.3 The CTLD Scheme in Practice The CTLD scheme is typically portrayed as having provided exchange guarantees to foreign lenders. While this is true, it is only part of the story. The return to foreign commercial banks taking part in the scheme consisted of the relevant Euromarket interest rate, i*, plus the spread, (+,of about 1.75 percent plus whatever front-end fee, 6,they could extract. The nominal return to the foreign lender in foreign currency terms, q*, can be written as: (2.1)

q*

=

i* +

(+

+ 4.

642

Merih Celisun and Dani Rodrik

The lenders had no reason to concern themselves with devaluations of the lira, save for possible effects on the liquidity of the country. What has been less recognized is that the primary role of the exchange guarantee was to provide the eventual domestic users of the CTLD credits with an interest rate subsidy. Remember that the local commercial banks used the domestic currency counterpart of the CTLDs to make loans to the private sector. From the perspective of the domestic firms (i.e., in domestic currency terms), the nominal cost of funds, q, borrowed in this fashion consisted of the nominal return to the foreign lender, q*, plus a margin, p,-of perhaps 0.5 percent-acquired by the intermediating local bank: q = q*

+ p.

Notice how the domestic borrower has been insulated from exchange risk, as the expected rate of depreciation of the lira appears nowhere in this formula. The interest rate effect of this scheme from the perspective of domestic borrowers can be conceptually separated into two components. First, there is the effect of allowing the private sector to borrow abroad, where no such possibility existed before. Given the prior restrictions, this is just like removing a tax on foreign borrowing, with the initial rate of the tax set at the price-equivalent level of the restriction. It is reasonable to assume that the “effective” domestic interst rate-i.e., taking into account domestic credit rationing and curb markets-exceeded foreign interest rates (adjusted by the expected rate of currency depreciation) in the presence of borrowing restrictions; in other words, the implicit tax rate can be assumed to have been positive. This effect would naturally boost foreign borrowing. However, the CTLD scheme had an additional element of subsidy deriving from the exchange guarantee. The subsidy consisted of the difference between the cost faced by the domestic borrowers, q, and the true opportunity cost of foreign funds, the latter of which was made up of the sum of three elements: the rate of return to the foreign lenders in foreign currency, q*, the intermediating margin of the local banks, p , and the expected rate of depreciation of the domestic currency, 2.From equation (2.2), the implied subsidy, s, can be seen to equal simply the expected rate of depreciation of the home currency: (2.3)

s

= (q*

+ p + 2)

-

q = 2.

This makes obvious sense since the system insulated borrowers from exchange rate movements. Some estimates of the magnitude of these implicit subsidies are presented in table 2.12. To recapitulate, there are two important lessons in all this. First, the distinguishing characteristic of the scheme was that it acted as an implicit subsidy on foreign borrowing by domestic firms. As the evidence in table 2.12 indicates, the magnitude of the subsidies involved was hardly negligible, amounting to 1.1 percent of GNP by the end of 1976. Secondly,

643

TurkeyIChapter 2

Table 2.12

Estimates of the Subsidy Component of the CTLD Scheme

1975 Expected depreciation of TL against": DM SFr

US$

Year-end 1976

1977

22.1% 17.6 10.0

30.5% 42.9 16.6

49.8% 60.4 29.9

22.1% 17.6 10.0 20.4

30.5% 42.9 16.6 32.6

49.8 60.4 29.9 51.1

Estimated interest rate subsidy for borrowing denominated in:

DM SFr

US$ Weighted averageb Implicit subsidy payments' Total (million $) Share of GNPd ( 8 )

203.8 0.5

580.6 1.1

1,158.4 1.7

Sources: Tables 2.10 and 2.13: and IMF, International Financial Statistics.

'Actual rate of depreciation during the following year. bWeights are the shares of different currencies in table 2.13 'Calculated as the average subsidy rate multiplied by the outstanding stock of CTLD liabilities at year end. '%e denominator used here is the following year's GNP since the subsidy amounts are calculated for year-end figures.

the level of such subsidization was not fixed and depended on the state of expectations. Anything which fueled, say, domestic inflationary expectations, in turn giving rise to expectations of greater depreciation, would also raise the subsidy element, reducing the ex ante cost of foreign borrowing. We will analyze the implications of this for the debt accumulation process below. But first some details on the operation of the CTLDs. It will not come as a surprise after the account above that it was private sector firms which took the lead in attracting CTLDs to the domestic banking sector. Typically, a Turkish firm would locate a foreign bank willing to make the deposit, and would be the beneficiary of the credit extended by the domestic bank using the counterpart funds. With a spread of 1.75 percent and front-end fees running around 4 percent, there was in fact little difficulty at first in attracting foreign lenders.'* In addition, some of the inflows were engineered through capital flight: many entrepreneurs bought foreign exchange in the black market and channeled these funds back in via selected foreign banks. By the time of the collapse, more than two hundred foreign banks had been lured. Detailed information about these transactions is scarce. There is, however, one survey which covers the operations of six of the largest intermediary local banks. Information from this survey is summarized in table 2.13. The table covers a total of 547 separate deposits made by foreign banks, amounting to $517.7 million (at year-end 1977 exchange rates), which is 23 percent of all CTLD liabilities at the end of 1977.

644

Merih Cellsun and Dani Rodrik

Table 2.13

Summary Information on a Sample of CTLDs Deposits

Amount (thousand $)” Share in sample (%) Number of deposits Size of average deposit (thousand $)

DM

SFr

US$

373,525 72.2 422 885

116,950 22.6

26,597 5. I 24 1,108

100

1,170

NLG 581 0. I I -

Total

517,653 100.0 547 946

Source: From the appendix to Yalqin Dogan. IMF Kiskurindu Turkiw. 1946-3980. 2d ed. (Istanbul: Tekin Yayinevi, 1986). The information in the appendix is attributed to a study by h e r Goren.

‘At year-end 1977 exchange rates. Key: Dm = Deutsche mark; SFr = Swiss franc; US$ = U.S. dollar; NLG = Dutch guilder.

Two aspects of the evidence in table 2.13 are particularly noteworthy. First, the sheer number of separate transactions involved is itself mindboggling. Judging from the sample, the CTLD scheme must have involved no less than 2,000 different loans of around $1 million each. While the number of foreign banks involved was substantially lower, the diffuseness of the process provides an important clue to the forthcoming crisis: foreign lenders had little idea, until it was too late, of the total amounts involved. One of the bankers involved would later express his bafflement as follows: “We began toting things up and I was quite surprised at the exposure. For instance, we would find out that a London bank was in for $100 million. It was just too astounding. Some nights I would wake up in a cold sweat” (Bleakley 1978, 50). Secondly, the currency composition of the CTLD liabilities also deserves comment. As table 2.13 shows, only about 5 percent of the deposits were denominated in dollars, with the rest split between the deutsche mark (72 percent) and the Swiss franc (23 percent). This is surprising at first sight given the fact that American banks played a predominant role in the CTLD scheme; according to one account, some sixty U.S. banks were involved (Bleakley 1978). Yet this was the natural consequence of the incentives provided by the borrowing subsidies. From the Turkish borrower’s perspective, borrowing costs were minimized in those markets where the nominal interest rates were lowest, irrespective of potential changes in cross rates (recall equation [2.2]). Since in 1975-77 these were the Euromarkets for the deutsche mark and the Swiss franc, Turkish borrowers ovenvhelmingly chose these currencies. The lower rates on loans denominated in these currencies of course reflected their expected appreciation against the dollar. This in turn meant that the implicit subsidy was larger for borrowing in DM and SFr. This is in fact borne out by the estimates of the subsidies presented in table 2.12. Throughout the 1975-77 period, the subsidy for borrowing denominated in these currencies was about twice that for borrowing denominated in dollars.

645

TurkeyIChapter 2

The fluctuations in cross rates suggests an additional problem with the CTLD scheme which was probably unanticipated when the program was first launched in May 1975. Even in the best of all possible worlds, had Turkey managed to maintain its peg against the dollar, the ex post subsidy on foreign borrowing would still have been positive (and sizable) as a consequence of the depreciation of the dollar against third currencies. The central bank would suffer losses due to the exchange guarantee even though no formal devaluation of the lira had taken place. As it happened, there were devaluations, and their impact was magnified by these cross-rate changes. For the Turkish authorities, this was a rude introduction to the world of floating rates. Besides the public finance aspect of the CTLDs, i.e., the fact that the implicit subsidy payments by the central bank had to be financed somehow, the scheme had devastating behavioral consequences. For the scheme not only subsidized foreign borrowing, it also made the level of subsidization directly proportional to the expected rate of depreciation of the lira and, hence, to the magnitude of the current exchange rate disequilibrium. The combination gave rise to a potentially explosive scenario: the CTLD scheme would engender overborrowing as long as the lira was expected to depreciate against some major currency; the overborrowing would then cause the present exchange rate to become (more) overvalued; this in turn would fuel expectations of further depreciation, further overborrowing, and so on until foreign bankers would discover the transversality condition and refuse to play along. This story highlights the critical role of exchange rate policy in the process of debt accumulation. Once the CTLD system got under way, the authorities were caught in a bind. Validating devaluationary expectations in order to set the current account straight would give rise to large exchange losses under the guarantee; refusing to do so would render the current account less sustainable by fostering further private sector borrowing. Exchange rate policy of the period strived to maintain, unsuccessfully, a middle road between the Scylla of large transfers to domestic firms and the Charybdis of growing current account deficits. Figure 2.1 shows the trends in the real effective exchange rate during the 1970s. Between 1975 and 1977, the real value of the lira was maintained at a roughly constant level, even though this constituted a real appreciation of about 10 percent relative to the 1973 level. The oil shock of 1973-74 had rendered a step increase in competitivenessi.e., a real depreciation-imperative, a fact to which exchange rate policy remained impervious. How overvalued was the Turkish lira? Table 2.14 provides two sets of estimates of the extent of overvaluation against the dollar during the 1974-77 period. The first of these is simply the black-market premium on the dollar. The second set of estimates is derived from a computable general equilibrium model used by Kemal Dervi? and Sherman Robinson (1978). In

646

Merih Celkun and Dani Rodrik

1970 Fig. 2.1

1972

1974

1976

1978

1980

Real effective exchange rate, 1970:I-1980:I (1972:I = 100)

Table 2.14

Measures of Overvaluation of the Turkish Lira ~~

Official exchange rate (TLI$)” Black-market rate (A)b “F4uilibrium rate” (B)C Overvaluation (%): (A) (B)

1974

1975

1976

13.93

14.44 15.76 17.80

16.05 17.64

14.35 14.70

3.0

9.1

5.5

23.3

1977 18.00

20.30

2 I .02 28.20

9.9 26.5

16.8 56.7

Sources: Dervi? and Robinson (1978). table 3.2; Pick’s World Currencv Yeorbook, 1984; and IMF, Internarionol Financiul Statisfirs.

aPeriod average. bCalculated as the geometric average of twelve end-of-month rates. ‘From DerviS and Robinson (1978). See discussion in text.

this model, the equilibrium exchange rate is defined as the rate that would achieve the current account path consistent with a “normal” level of reserves and foreign borrowing. While neither measure needs to be taken literally, they both tell the same story of increasing overvaluation after 1974. By 1977 a conservative estimate would be that the Turkish lira was overvalued by at least 20 percent. The growing overvaluation had the consequence of raising the subsidy on foreign borrowing beyond any reasonable level. As table 2.12 shows, the

647

TurkeyIChapter 2

average subsidy rose from around 20 percentage points at the end of 1975 to more than 50 points at the end of 1977. The latter figure need not be taken too seriously-except for an indication of the ex post transfers made to the borrowers-since few foreign banks were foolhardy enough to continue establishing CTLD accounts past mid-1977. l 3 Still, an interest rate subsidy of 33 percentage points (at year-end 1976) must have presented an inordinate inducement to borrow as much as possible, as quickly as possible. It is no wonder that many borrowers were soon willing to put up stupendously high front-end fees: these fees are reported to have risen eventually to more than 20 percent (Kafaoglu 1986).14 The fact that the process could continue until mid-1977 is testimony to how oblivious foreign bankers were to the Stiglitz and Weiss (1981) notion of equilibrium credit rationing.

2.4 Exchange Rate Policy and Debt Dynamics with CTLDs: A Model To get a better grip on how exchange rate policy and foreign borrowing interacted under the CTLD scheme, it is instructive to look at a bare-bones model of the current account. The current account deficit, B , can be expressed as the difference between national expenditures and income, the latter of which equals national product minus interest payments on foreign debt. Abstracting from changes in reserves and direct foreign investment, B is also identically equal to the rate of accumulation of foreign debt, F . We can then express F as the sum of interest payments and a component which depends negatively on the domestic real interest rate and the real exchange rate: (2.4)

F = B = q*F - ar - /3(e

-

p)

+ g,

where q* = effective foreign rate of interest, assumed to be fixed; r = domestic real interest rate; e = logarithm of the nominal exchange rate and the price of traded goods; P = logarithm of the price of nontraded goods; g = a shift factor. The current account deficit depends negatively on the real interest rate since an increase in the latter reduces domestic expenditures (consumption and investment). The negative sign on the real exchange rate can be motivated either by the existence of excess capacity in the traded sector or by the negative real-balances effect of depreciations on expenditures. The real interest rate, in turn, is the difference between the relevant nominal rate-which we shall denote generally as i for now-and the expected rate of domestic inflation, he.

(2.5)

r=i-+

648

Merih Cellsun and Dani Rodrik

where, by appeal to rational expectations, ire is assumed to equal the actual rate of inflation, itself a weighted average of the increases in the prices of traded and nontraded goods: (2.6)

.ir

= pi

+

(1 - p)j.

The Turkish authorities controlled the nominal exchange rate, e, and manipulated it with the current account deficit in mind. Their policy can be summarized in the following manner: (2.7)

e

=

AB.

The greater the current deficit, the larger the rate of depreciation of the exchange rate, with the rate stabilized only when the target for the current account-here zero-is reached. This formalization does not do too much injustice to the actual exchange rate policy of the time, which consisted of a series of small adjustments. For simplicity, we will abstract from the developments in the market for home goods and set p equal to zero. By doing this we are neglecting some important issues involved in the so-called capital-inflows problem: an autonomous capital inflow will tend to appreciate the real exchange rate as long as some of its proceeds are spent on nontraded goods. This adds a further layer of complications to the dynamics of debt accumulation being considered here. But since these complications are readily understood, we leave an explicit treatment of the home goods sector to the appendix to this chapter. As a benchmark case, consider first the adjustment process of the economy with a fixed real interest rate, i.e., in the absence of the CTLD scheme. The model can be visualized with the help of figures 2.2 and 2.3. Figure 2.2 displays the FF schedule, which is defined as the combination of e and F that leaves the current account in balance (i.e., F = 0). The FF locus is upward-sloping since a higher stock of debt implies larger interest payments and hence requires a more depreciated currency to equilibrate the current account. In the medium to long term, the economy has to locate itself somewhere on the FF schedule. The adjustment process of the economy out of this equilibrium is shown in figure 2.3. Making use of the policy rule expressed in (2.7), differentiation of (2.4) yields: dFfdF = q* -

PA.

Stability of the process requires that (2.8) be negative, i.e., that A > q*/P. Unless exchange rate policy is sufficiently responsive to the deficits, foreign debt may keep on growing, fueled by the servicing of the existing liabilities. The stable case is demonstrated in figure 2.3. A shock to the economy in the form of an increase in g-no harm in thinking of this as government spending-shifts the FF schedule up to F ' F ' , since at any level of debt a higher e is now required for long-run

649

TurkeyIChapter 2

Fig. 2.2 The FF schedule

equilibrium. The point on F'F' at which the economy eventually settles depends on the adjustment process. The higher the speed of adjustment of the exchange rate (and the lower q*), the steeper the slope of dF/dF and the lower the eventual levels of the exchange rate Z and of the debt stock F . Unless A is infinite, i.e., the exchange rate is adjusted instantaneously, > Fo and 2 > eo. Now consider the effects of the CTLD scheme. As explained above, the scheme served to fix domestic nominal rates at the level of foreign rates (plus the intermediary banks' margin, which we ignore). This rendered the effective real interest rate solely a (negative) function of the (expected) rate of inflation, and through (2.6), of the rate of depreciation of the currency:

F

(2.9)

r = q * - + e =

q* - p t .

One consequence of the policy is the magnification of the effect of any shock on the current account. Hence, upon the increase in g , the current account deficit on impact, B(O), was previously simply Ag. The same calculation now yields

(2.10)

B ( 0 ) = [ l / ( l - apA)]Ag.

650

Merih Cellsun and Dani Rodrik

Fig. 2.3 External debt accumulation

Stability requires that (1 - (YFA) > 0 (see eq. [2.11] below). Further, since all three parameters involved in this expression are positive, we must also have (1 - apA) < 1 . Therefore, [l/(l - apA)]is greater than unity, which is the magnification effect mentioned above. The larger initial current account deficit in the presence of CTLDs is due to the greater spending encouraged by inflationary expectations and fixed nominal interest rates whenever the current account is adversely affected. Hence, in addition to the direct effect of the autonomous increase in g, we have an increase in private sector expenditures that is brought about by the instantaneous fall in real interest rates. The adjustment process is also faster with the CTLD scheme in place. This can be seen once again by differentiating (2.4), and using the definition of r from (2.9):

(2.11)

dF/dF

= [l/(l -

a p A ) ](q* -

PA).

651

TurkeylChapter 2

Stability will require the term in square brackets to be positive. With this requirement met, the same reasoning as in the previous paragraph can be used to show that this expression is larger than that in (2.8). In other words, the deficits are now reduced faster, since, with the stability of the process assured, any cut in current deficits has the added beneficial effect of further reducing expenditures via the dampened inflationary expectations. Figure 2.4 compares the dynamics of debt accumulation with and without CTLDs. Note that the eventual stock of foreign debt (as well as Z) are identical in the two cases. During the adjustment process, however, the current account deficits are always larger with CTLDs than without. The apparent contradiction is resolved by noting that the long-run level of debt is reached faster in the first case. In other words, debt is accumulated more rapidly with the CTLD scheme in place. How much more rapidly depends on the expenditure elasticity with respect to interest rates, a,and the strength of the linkage between the exchange rate and the price level, p. The argument so far is predicated on an exchange rate policy-here represented by A-devised so that the process of debt accumulation eventually settles at some stable long-run level F . With the CTLD scheme in place, the stability requirement in fact becomes more stringent. In addition to the previous condition that A > q*/p, we now have ( 1 - apX) > 0, which requires A < l / c ~ p .This ’ ~ new upper limit on the responsiveness of exchange rate policy corresponds to the fact that a “too high” A will cause too much of a borrowing binge via inflationary expectations. In the limit, as

F

i

Fig. 2.4 Debt dynamics with CTLDs

652

Merih Celasun and Dani Rodrik

A goes to infinity, foreign borrowing creates the possibility of infinite capital gains. Hence, the difficult task for the authorities would have been to adjust the exchange rate so as to satisfy both requirements, i.e., to settle on a pace of depreciation which was neither too slow nor too rapid. Notice that the band of stability represented by l / a p > A > q*/p might have been quite a narrow one. In fact, no such A would have existed in the case where q*lp is larger than l / ( ~ p . ' ~ The more rapid accumulation of debt under the CTLD scheme and the narrower range of stable policies in turn create further problems. Turkish borrowers might have anticipated that the CTLD scheme, like all things too good to be true, would have to come to an end. The end could be precipitated either by the government or, as it eventually happened, by foreign lenders who finally got their sums right. Cut off from new inflows, the government would then have no choice but to undertake larger depreciations to bring the current account under control. Indeed, such expectations on the part of Turkish borrowers could prove self-fulfilling in the sense of bringing a collapse of lending where none need have occurred. To see this, suppose that at some time 7 during the debt-accumulation process, borrowers come to believe that there is a subjective probability, 6, that foreign lenders will autonomously reduce their rate of lending. If the possibility materializes, the authorities will have to increase the rate of depreciation from A to This possibility raises the expected inflation rate from pA to p[(1 - 6)A 6x1 and correspondingly lowers the real interest rate. For sufficiently high 6 and/or such expectations can be self-fulfilling by rendering the adjustment process unstable. From 7 onwards, we now have:

x.

+

(2.12)

dFldF

x,

= {1/(1

x,

-

a p [ ( l - 6)A

+ 6x])}(q*- PA)

For large 6 or the expression in the curly brackets could turn negative, making dF/dF positive. This possibility is illustrated in figure 2.5. The crisis is now made inevitable since the borrowing process no longer has a natural brake, and debt will keep on accumulating until something gives. The self-fulfilling nature of expectations here is the consequence of the nature of the CTLDs: the higher the probability of an eventual debt crisis, the cheaper it was to borrow.

2.5 Recapitulation Did the experience of the 1974-77 period amount to overborrowing? The discussion above should leave no doubt that it did. With spread and front-end fees included, the marginal cost of funds (denominated in DM), was at least 20-25 percent toward the end of 1976. This amounts to a real interest rate burden of around 16-21 percent. It would be hard to believe that the ongoing investment drive in Turkey justified borrowing on such terms. The

653

TurkeyKhapter 2

i

FO Fig. 2.5 Self-fulfilling crisis

authorities not only greatly subsidized foreign borrowing, but they also did so in a way that seriously hampered the dynamic stability of the process. To be sure, the CTLD scheme allowed a domestic credit explosion to finance a rapid increase in private sector expenditures alongside the public sector investment drive, and helped engineer unprecedented rates of economic growth. But the edifice was constructed on inherently shaky foundations. The CTLD episode lasted for roughly two years, between May 1975 and July 1977. Given the problems already discussed, even this might be considered as too long a time frame. Why did the foreign banks not pull out earlier? Some clues to the answer have been given above. Banks seem to have had little knowledge of the rapid rise of Turkey’s aggregate foreign liabilities, and do not appear to have analyzed the behavioral consequences of the scheme in any great detail. One account suggests that banks would “test” the CTLDs by frequently withdrawing their money, to redeposit it again promptly (Brennan 1976, 84). With a sufficiently large number of banks all doing the same, the information content of this strategy must have been close to nil. Individual banks simply imitated their competitors’

654

Merih Cellsun and Dani Rodrik

behavior with the assurance that, were a crisis to come, they would be caught in good company. How about the government authorities? Here a number of factors must have been responsible for the implementation of the scheme. Perhaps foremost was the desire to maintain, for political as well as sound economic reasons, the investment effort. As has been aptly stressed by Boratav (1986): [tlhis was a period when the rivalry and confrontation between the two major parties for the allegiance of the main social groups took acute and, at times, violent forms. The alternating governments-the Republican People’s Party (RPP) in 1974 and the Justice Party (JP) during 1975-77-could, therefore, not afford to suspend the well-rooted distributional and allocational mechanisms of populism. And . . . these mechanisms could produce the intended results only in an expanding economy. (2) The ready availability of foreign exchange through the CTLD scheme must have served to create the illusion of a soft budget constraint on public investment. Hence, after 1973 the realized levels of public sector investment (as a share of public sector income) consistently exceeded the planned levels under the annual programs.” Second, lack of experience with floating rates must have obscured the ill-effects of cross-rate changes discussed earlier. Third, a lack of recognition of the overvaluation of the current exchange rate also would have played a role in minimizing the risks involved. Besides, the margin of indeterminacy noted above with respect to outcomes might have provided some ground for optimism. Finally, we might add the most cynical explanation of all: it was likely that the eventual mess would have to be cleared up by the next government, which is precisely what happened. The CTLDs were eventually consolidated and rescheduled in an agreement signed in August 1979. Principal repayments on these liabilities began again in 1984, for the first time since July 1977. The burden of servicing the CTLD overhang would prove to be a substantial one in a period of vast real depreciations of the Turkish lira. Who paid the subsidies implicit in the CTLD scheme? Technically, the burden was the Treasury’s, but the central bank was the effective source of payments as the Treasury never compensated the bank. Lacking resources of its own, the central bank had to generate funds somehow. As will be discussed in the following chapters, it did so partly through the inflation tax and partly by shifting the burden onto future generations via renewed external borrowing.

Appendix Consider the effect of foreign borrowing on home-goods prices. Since borrowing allows a higher level of domestic expenditures, it puts upward pressure on the prices of such goods. This is the familiar problem of real

655

TurkeyIChapter 3

appreciation of the exchange rate in the presence of capital inflows. To represent this process, let the rate of increase of home-good prices, p , be proportional to the excess of expenditures over income, as captured by the current account deficit B: p = OB.

(A2.1)

Now the expected rate of inflation has an additional component coming from the dynamics of p : (A2.2)

7i

=

pi

+

(1 - p)&

Using this, we can calculate the impact effect of Ag on the current account as follows: (A2.3)

b(O)=-{1/(1

- a[pA

+

( 1 - p)O])}Ag,

which is greater due to expectations of higher inflation. The dynamics of debt are in turn determined by: (A2.4)

dF/dF = {1/(1 - a[pA + (1 - p)O])}(q*-

P(A -

0)).

Once again, the stable region for A is smaller. In other words, incorporating the dynamics of the nontraded sector makes instability more likely and exchange rate management more problematic.

3

Crisis Without Adjustment, 1978-79

The debt crisis developing in mid-1977 threw Turkey into a period of forced adjustment. As foreign exchange sources dried up, external balance became for the first time in many years a genuinely binding constraint, requiring an adjustment in the relation between income and absorption in the economy. Until the reform package of January 1980, the policies employed by the authorities were unsuccessful in extricating the Turkish economy from the crisis. In view of the foreign exchange constraint, some belt-tightening had become inescapable. However, the governments in power during this period-and there were many-compounded the problems by their refusal to implement vigorous adjustment measures. The investment boom collapsed, economic growth came crashing down, inflation rose to unprecedented heights, and income distribution worsened disastrously. The only positive development was the beginning of a series of debt reschedulings with official

656

Merih Celasun and Dani Rodrik

and private creditors, to be discussed in chapter 9, which eased the debt-service burden considerably. In the present chapter we provide an interpretation of the immediate aftermath of the debt crisis. Our argument will be that a collapse in investment and an inflationary spiral were the key mechanisms enabling the required adjustment to the external constraint to be achieved. In the absence of sufficient reductions in nominal expenditures and of real exchange rate depreciations, absorption could be brought in line with the available resources only by engineering inflation. Inflation, in turn, had severe consequences on income distribution, as real wages and the agricultural terms of trade eroded. Meanwhile, it created windfall gains for the entrepreneurs who had borrowed prior to the crisis, by effectively validating their expectations. The lack of serious adjustment measures had a high cost in terms of both foregone output and distributional objectives. What follows can be read as a cautionary tale regarding the consequences of ignoring accounting identities.

3.1 Government Policies and Inflation The immediate consequence of the debt crisis was a foreign exchange shortage, as foreign inflows slowed to a trickle. That meant that the current account deficits had to come down. As table 2.1 showed, the deficit of $1.6 billion in 1978 was less than half that of 1977 ($3.4 billion), and a further cut to $1.2 billion took place in 1979. How were these cuts achieved? Policy itself was of little help. The IMF was called on to administer a series of stabilization programs, and two sets of adjustment measures were announced, one in early 1978 and the other in 1979. These programs aimed at reducing the public sector deficits and improving the foreign exchange outlook. They included devaluations, increases in prices of SEES, and credit ceilings. But both programs, as well as the two corresponding IMF standby arrangements, proved unsuccessful. Until January 1980, the various adjustment measures undertaken by the authorities can be aptly described as having been too little, too late. The reduction in government spending was only half-hearted, and exchange rate policy, albeit more active, lagged behind rising inflation. The governments in power were too conscious of political support to administer radical shock treatment and too divided to implement any feasible alternative. The first casualty of the shortage of foreign exchange was investment, which dropped sharply as the availability of imported inputs and capital goods became problematic. In fact, the primary vehicle for reducing dependence on foreign inflows was this reduction in capital formation. As table 3.1 shows, the aggregate investment ratio collapsed from 25.0 percent of GNP in 1977 to 18.5 percent in 1978, and then to 18.3 percent in 1979. The public sector investment ratio declined to levels not experienced since

'

657

TurkeylChapter 3

Table 3.1

Investment-SavingsBalance and Growth of Real Expenditures, 1978-80

Investment Private Public Domestic savings Private Public Foreign savings Sectoral savings-investment balances Private Public Total Growth of real expenditures Private Public Total

1978

1979

1980

18.5 9.1 9.5 15.9 10.6 5.3 2.6

18.3 8.8 9.5 16.2 13.5 2.7 2. I

21.4 9.9 11.5 15.9 10.6 5.3 5.5

1.5 -4.2 -2.7

4.7 -6.8 -2.1

0.7 -6.2 -5.5

0.0 -2.9 -0.7

-3.4 4.0 - 1.6

- 1.3

3.6 -0.1

Source: SPO. Nofe: Investment-savings balance data reported as a percentage of GNP. Growth of real expenditures excludes expenditures on inventories and is reported in percentages.

1974. But public savings also declined as a result of the adverse consequences of inflation on the tax revenues of the government. As a result, the public sector savings-investment balance improved in 1978 by only 2.5 percentage points, only to deteriorate in 1979 to a level below that of 1977. The public sector borrowing requirement remained at 7.9 and 8.9 percent (of GNP) in 1978 and 1979, respectively. This was lower than the record 10.6 percent of 1977, but was worse than the performance during 1974-76 (see table 2.6). The bulk of the burden fell on the private sector, in which the savings-investment balance turned positive for the first time since 1975 and rose to 4.7 percent of GNP in 1979. The lack of serious adjustment on the part of the public sector is also visible from the figures on growth of real expenditures. The investment collapse of 1978 was associated with a real contraction of public expenditures in 1978, but such expenditures resumed their growth in 1979 and thereafter, albeit at much reduced levels compared to the mid-1970s (see tables 2.6 and 3.1). By contrast, the real expenditures of the private sector stagnated in 1978 and then fell for a number of consecutive years. By 1979 the public sector had regained its 1977 level of expenditures; the private sector would have to wait until 1984 for the same to happen. The various governments’ lack of resolve in expenditure reduction was equally evident in their expenditure-switching policies. A series of nominal devaluations of increasing magnitude was undertaken, but these proved ineffective in arresting the deterioration in competitiveness. As figure 2.1 showed, after a small depreciation in 1978, the real exchange rate appreciated

658

Merih Cellsun and Dani Rodrik

sharply in 1979 as nominal adjustments failed to keep pace with inflation. The authorities also experimented with other schemes. An attempt was made to increase workers’ remittances by giving them more favorable exchange rates. The effective subsidy rate on exports was increased by expanding tax rebates and allowing some retention of foreign exchange by exporters. While export receipts and remittances rose in response to these incentives, much of the adjustment burden still fell on imports. In the end, foreign exchange was rationed not by price, but administratively. The authorities progressively reduced travel allowances for Turkish tourists going abroad, and allocated the scarce supplies according to the perceived needs of various importers. The black-market premium on the dollar rose as high as 91.4 percent at the end of the first quarter of 1979.2 Widespread shortages of imported commodities developed. The consequence of all this was a rise in the rate of inflation. In the presence of an effective ceiling on external borrowing opportunities and a lack of sufficient restraint on nominal expenditures, real absorption could be brought in line with the available resources only by creating inflation. In this sense, it was inflation which equilibrated the open-economy national income identity, by closing the gap between ex ante demand and supply. Normally, inflation would have had adverse consequences on the external balance by, among other things, appreciating the real exchange rate. While, as pointed out above, the appreciation did take place, its effects on the current account were insulated by direct controls on the allocation of foreign exchange and by rationing. In the meantime, rising prices undercut the real purchasing power of large segments of the population with “sticky” nominal incomes (on which more later). Inflation here was not the problem but the “solution.” Another way of observing this is through the perspective of the flow of funds. A given credit expansion in the economy gives rise to a larger expansion of monetary aggregates, and hence greater inflation, when leakage through the balance of payments is excluded. The figures from the balance sheet of the monetary authorities (table 3.2) show clearly that that is indeed what h a ~ p e n e d In . ~ the 1974-77 period, the accumulating external liabilities of the central bank (including CTLDs) discussed above amounted to a staggering 116.2 percent decline in net foreign assets (in annual terms). With the central bank cut off from foreign funds after 1977, no such decline took

Table 3.2

Annual Average Growth Rates (in percentages) End-1974 to End-1977 Net foreign assets Domestic assets Base money

End-1977 to End-1979

- 116.2

2.1

60.6 35.7

40.8 46.1

659

TurkeyIChapter3

place in 1977-79. The consequence was that a smaller rate of increase in domestic credit in 1977-79, principally to the public sector, had a larger eventual effect on base money.4 This also provides an answer to the puzzle arising from the existence of an inflationary tide in a period in which the PSBR and public recourse to the central bank were modest compared to the heights reached in 1977. Inflation, then, was part and parcel of the adjustment process which achieved the drastic reductions in the current account deficits of 1978 and 1979. This process of adjustment via inflation is reminiscent of Albert Hirschman’s analysis of the politics of inflation in the Latin American context: “Where the state defers readily to all the successive demands made on it by one group or one government department after another, inflation has the function of denying part of what the state, in its weakness, has granted. . . . [Tlhe state might be said to hand over to inflation the disagreeable job of saying no” (1985, 73).5 In the same vein, the Turkish governments of 1978-79 handed over to inflation the disagreeable job of cutting real expenditures. The close association between the current account and inflation can be seen from the evidence on prices portrayed in figure 3.1. Hence, it is clear that inflation started to take off in the second half of 1977, just as new borrowing opportunities were becoming extinct as a result of the developing crisis. This was the period when foreign exchange shortages first made themselves felt, with a consequent scarcity of imported inputs. The rate of increase in wholesale prices reached 52.6 percent in 1978, 63.9 percent in 1979, and culminated in 107.2 percent in 1980 before it was brought under control (see table 2.1). The peak of inflation in 1980 came at the same time as the relaxation of the external constraint, which could be viewed as contradicting the thesis here. But the high rate of inflation in 1980 was largely a validation of price increases which had already effectively taken place in black markets, and which became reflected in official statistics once price controls were lifted in January 1980.

3.2

The Distributional Implications of Adjustment via Inflation

Figure 3.1 shows the relatively large divergence in price trends for industrial and raw materials on the one hand, and food items on the other. The first category was affected by the foreign exchange shortage to a greater extent since its import content is much higher. As very little food is imported, food prices were determined by domestic agricultural conditions plus domestic support payments by the government. These support payments were in turn eroded by inflation, with the consequence that the industrial and raw materials index rose more rapidly than the food index. As these trends suggest, the unwitting changes in the structure of relative prices caused by inflation had serious consequences for income distribution.

660

Merih Cellsun and Dani Rodrik

I I I

4 I\

::

1977

'

/! 1 %

INDUSTRIAL RAW MATERIAL INDEX

I :GENERAL

1978

1979

Fig. 3.1 Wholesale prices (increase over previous year, 1963= 100) Source: OECD Economic Surveys, Turkey, April 1980.

rJ

I

:

,!

1980

661

TurkeyKhapter 3

In an economy like Turkey’s, there are two key relative prices determining the relative share of various groups in national income. The first of these is the agricultural sector’s terms of trade with the rest of the economy, which is a rough but good indicator of the rural-urban split. The second is the real wage, which is an indicator of distribution within urban areas between workers and others. With neither of these indicators was the Turkish experience after 1977 a happy one. Figure 3.2 displays the trends in both of these relative prices, together with the trends in aggregate real income and expenditure for purposes of comparison. The evidence on the agricultural terms of trade shows the consequence of the price trends discussed briefly above. After a cumulative improvement of 24 percent between 1970 and 1977, the terms of trade declined sharply thereafter. Within the time span of two years (1978-79), a reduction of 23 percent brought the level to below where it had been in 1970. There was a further fall of around 4 percent in 1980, the effects of inflation being partially offset in that year by a large real depreciation of the currency.6 The fall in real wages was even more drastic. While the nominal wage data used in constructing the real wage series are problematic in that they do not include bonuses and fringe benefits, the evidence portrayed in figure 3.2 is sufficiently telling. Once again, the collapse comes right after 1977 and is the consequence of nominal wage contracts which did not fully anticipate the coming inflation. Between 1972 and 1977 real wages had increased by 32 percent. After 1977 real wages fell almost in the same proportion as the rising inflation: by 12 percent in 1978, 14 percent in 1979, and a stupefying

2

1972 73

74

75

76

77

78

79

80

81

82

83

84

Fig. 3.2 Real income, real expenditures, real wages, and agricultural terms of trade (1970 = 100)

662

Merih Celasun and Dani Rodrik

25 percent in 1980, bringing the cumulative reduction in three years to 43 percent. The worsening of income distribution in this period was the consequence of inflationary pressures being reflected in contract prices (such as wages) only with a lag whereas other prices could adjust freely. ironically, the resulting relative-price structure would be maintained and consolidated in the post- 1980 period of adjustment. As will be discussed at greater length in the following two chapters, the inflationary episode of 1977-80 had disturbing effects which outlasted this period.

3.3 Concluding Remarks Was there an alternative? Given the necessary retrenchment on the current account, it was inevitable after 1977 that real expenditures would have had to be cut somewhat. Instead, the authorities acted as if the crisis might go away if ignored, setting off an inflationary spiral which eroded the real incomes of the poorest segments of Turkish society. A series of bold measures of expenditure reduction and expenditure switching early on in the game might have enabled the economy to avoid some of the worst excesses of adjustment via inflation. In the event, letting inflation do the job of cutting real expenditures proved a very costly method compared to the obvious alternative of reducing nominal spending itself. In any case, the policies of the 1978-79 period did not seriously tackle any of the underlying problems of the economy, and they were incapable of promoting recovery. The latter would have to wait for the 1980s.

4

Stabilization and Adjustment Policies, 1980- 85

As described in the earlier chapters, Turkey became the first major developing country debtor to face a deep payments crisis in the post-1973 period. Because of the poorly managed macroeconomic environment and massive short-term borrowing, Turkey’s debt rout arrived early in mid- 1977 before the second oil shock of the late 1970s. A heavy reliance on import compression in combination with unrestrained nominal expenditures resulted in an inflationary slowdown of growth during 1978-79. The accompanying shortages in commodity supplies produced wide public discontent. The unsuccessful implementation of the IMF standby arrangements also strained

663

TurkeyIChapter 4

relations with the Fund and the international financial community. Against the background of deteriorating economic performance and increasing social and political tensions, Turkey had parliamentary by-elections in October 1979, which resulted in the resignation of Ecevit’s cabinet and the formation of Demirel’s minority government. Faced with a thoroughly destabilized macroeconomic picture. Demirel’s minority government introduced a new policy package in January 1980. After the resumption of growth, together with a sizable inflation reduction in 1981, the policymaking process, with the full support of the IMF and World Bank, increasingly focused on export-oriented adjustment issues and liberalization reforms in the economy. Turkey gained an acceptable degree of creditworthiness by 1982-83, just as most of the major LDC debtors were entering a disruptive crisis phase in their development. The Turkish recovery has been accompanied, indeed partly caused by, an export boom which has taken even the most optimistic observers by surprise. The relative success of the Turkish adjustment policies since 1980 has been undeniable. In the aftermath of the Mexican debt-service moratorium in August 1982, the IMF and World Bank have stressed trade and financial liberalization in their high-conditionality programs for LDCs with debt-servicing difficulties. Together with the well-known cases of the export-oriented East Asian economies, Turkey’s recent adjustment experience has been showcased by these multilateral institutions as a successful application of their liberalization-focused approach to the management of the LDC debt crisis. The current policy debate on the LDC debt crisis is concerned with the efficacy of the official international approach to a number of crucial points. As aptly argued by Sachs (1986) and Dornbusch (1985), fiscal correction problems are very serious in the LDC debtors, as the bulk of external debt is held and serviced by the public sector. The trade-liberalizing measures involving sizable depreciation of the exchange rate often tend to overburden the budget-correction process and destabilize the internal balance. As noted by Sachs (1985), in his comparative analysis of the Latin American and East Asian cases, the political-economy context also matters in the maintenance of realistic exchange rate and commercial policies in the pursuit of export-led growth processes. Nonetheless, in recent years distributional aspects and social costs have invited very little formal concern in launching adjustment programs. In the context of the contemporary policy debate on the LDCs, a balanced review of the recent Turkish experience, as well as other country-specific case studies, may provide useful points for generalized assessments. Turkey’s outward-oriented experiment in the 1980s is still a continuing one and contains imperfections as well as strengths in the way conditions for equitable and sustainable growth in the long run are being established. Hence, Turkey’s policymakers may also benefit from expost assessments of the policy and adjustment patterns observed in the post-1980 period.

664

Merih Cellsun and Dani Rodrik

A number of earlier reports and papers by official agencies and individual authors offer useful accounts of the Turkish economic recovery in the 1 9 8 0 ~While . ~ drawing on the data and commentaries of the earlier studies, our analysis strives to focus on the overall pattern of policy sequence and interdependencies, and to bring out the key linkages in the adjustment process. Rather than treating policy elements and actual outturn in one unified chapter, it appears more convenient to discuss the nature of policies in the present chapter (with occasional references to performance data), and analyze the anatomy of the adjustment process in chapter 5. 4.1

Special Factors Affecting Economic Performance

Before proceeding with the discussion of policy aspects, certain special factors that have played a role in Turkey’s recent economic recovery should be pointed out. The purpose of this is not to downgrade the domestic policy effort, but to provide a more balanced picture for intercountry comparisons. Among a host of special factors prevailing in the policy setting, the most salient ones are the following. Domestic Political Context. Although initiated by Demirel’s government in January 1980, the implementation of the new stabilization program received not only continuity but also additional political clout under the military rule from September 1980 to November 1983. The military regime retained Turgut Ozal, the principal technocrat behind the new program, as deputy prime minister during 1980-82. With the return of Ozal as the newly elected prime minister in late 1983, the liberalizing measures were further extended and strengthened. Besides ensuring continuity in the policy process, the interim military arrangements facilitated legislative and administrative changes pertaining to the structural components of the program. Furthermore, the presence of the military in the political arena made it possible to attain downward flexibility in real wages and agricultural prices-which we will stress in the next chapter as a key aspect of the adjustment process-and to avoid open distributional conflicts in policy implementation. Sizable Debt Relief and New Lending. The post-1980 policy experiment benefited from the debt relief, balance-of-payments assistance, and policy support of the major bilateral creditors (mainly the OECD countries) and multilateral lending institutions. As we will discuss in greater detail in chapter 9, the debt relief granted through the OECD Aid Consortium reached $4.6 billion in 1980-85. Not only the size, but also the timing of the external assistance was beneficial to Turkey’s economic recovery. During the difficult stage of 1980-83, the cumulative net resource transfer (excluding the minor items connected with foreign direct investment) was nearly a positive $2 billion, which obviated the need to generate surpluses in the noninterest current account. The effective policy dialogue with the IMF

665

TurkeyKhapter 4

and World Bank has facilitated debt relief agreements and concessional bilateral lending in the 1980-85 period. Special Market Conditions in the Middle East. The trading opportunities in the Middle East assumed special political characteristics with the Iran-Iraq military conflict in the Persian Gulf. To offset the cyclically unfavorable export conditions in the OECD region, a comprehensive effort was made by Turkey to penetrate the Middle Eastern and North African markets. The marginal share of the Middle East in the expansion of merchandise exports in 1980-83 (from $2.9 billion in 1980 to $5.7 billion in 1983) was 68 percent. The export drive to this region was complemented by the rapid rise in construction projects, the value of which reached $12 billion in 1981 from $3 billion in 1979. In turn, the incremental contribution of the Middle East has lessened after 1983 with the reemergence of the OECD region (mainly the EEC countries) as the predominant trading area for Turkey, accounting for around 55 percent of both exports and imports in 1985.3 The econometric work reported in chapter 7 is suggestive of the important role played by Middle Eastern demand in Turkish exports after 1980. 4.2

Latin American Debtors and the Turkish Case: A Digression

To place the policy review for Turkey in a more relevant cross-country context, attention may also be drawn to a basic similarity, as well as to a number of differences, between Turkey and major Latin American debtors in their crisis management in the 1980s. One basic similarity pertains to the structural nature of external capital in the pre-crisis growth process. At least until the mid-l970s, Turkey and major Latin American debtors (e.g., Mexico, Brazil, and Argentina) pursued a development strategy that centered on a growing home market and used external capital to supplement domestic savings in investment. The prolonged maintenance of heavy protection and the exhaustion of relatively easy import-substitution possibilities eventually yielded a productive structure which rigidly depended on imported inputs almost in fixed proportion^.^ As observed by Dornbusch (1986, 138), these heavily indebted countries have been structural importers of capital, facing a wide range of short-run structural impediments in restoring external balance. Unlike the cases of flexible open economies, the rapid return to external balance from a large deficit position cannot be accomplished just by cutting down overspending, at least not without substantial losses in output. In coping with structural rigidities in the adjustment process, the Turkish experience in the 1980s differed, however, from the Latin American cases in four key respects. First, Turkey could secure sizable new lending in the early years of the recovery effort, and thus bolstered its foreign exchange position to resume the needed growth of imports in overcoming the structural impediments to a more outward-oriented stance in development policy.

666

Merih Celisun and Dani Rodrik

Second, Turkey’s public finances were not overburdened by external debt servicing in 1980-83, largely due to the debt relief granted by its major creditors through the OECD Aid Consortium. Third, unlike in major Latin American economies, the Turkish policymakers did not have to deal with indexed contract prices, and therefore could attain a credible price stabilization-through an orthodox shock treatment-in the earlier years of the program. Fourth, the dependence on primary exports was considerably less in Turkey than in the Latin American region, which suffered as a whole substantially from the sharp fall in primary commodity prices in the post- 1982 p e r i ~ d . ~

4.3 Broad Strategy and Policy Objectives In retrospect, it seems clear that the new economic team under Ozal by late 1979 had made a thorough assessment of the earlier policy trials, the current situation, and possible future actions. As discussed in chapter 3, it was evident that immediate corrective actions were needed to regain control over monetary expansion and to relieve commodity shortages. The evaluation of past policies pointed to the adverse consequences of aborting trade-liberalization objectives in the earlier stabilization episodes, and of subsequent appreciation of the real exchange rate which produced an anti-export bias in the growth process. To restore creditworthiness, it was necessary to attain a rapid rise in foreign exchange earnings through domestic efforts. The resumption of growth seemed to be essential in gaining public confidence in the policymaking process. Drawing lessons from past experiences and recognizing the limited scope for further import compression in the country’s current stage of industrialization, the economic team persuaded Demirel’s government to introduce a bold and comprehensive set of policy measures on 24 January 1980. Apparently, these measures went further than the proposals and requirements of the IMF.6 The package of January 1980 was specific on policy measures, but not explicit on the magnitude and sequence of objectives sought in the future performance of the economy. In retrospect, it can be stated that the policy objectives were basically twofold: ( I ) to attain, as quickly as possible, an acceptable degree of price stability combined with export-led (output) recovery, and (2) to achieve a greater outward-orientation of the economy through a sequential liberalization and structural adjustment. The new strategy was more of an approach and a style, rather than a blueprint for future actions. Besides featuring a strong commitment to flexible pricing, this approach emphasized gradual changes in institutional mechanisms for the development of a unified market economy. The gradualist character of the overall strategy was stressed by Ozal in a 1982 interview as follows:

667

TurkeyIChapter 4

Change has to be gradual; we try to have what I call dynamic programming, but, in certain areas, change has to be step by step. For example, on January 24 [1980] we did not free interest rates. Six months later we freed them. But real freedom came at the beginning of last year [1981], when the banks started to fight each other. The same applied with foreign exchange. This year [1982] we hope to change the line protection scheme, which we couldn’t change immediately because people were so used to it. (Euromoney 1982) In the discussion of the broad strategy, a further observation relates to the policymakers’ preference for the simultaneous pursuit of macroeconomic stabilization and export-led recovery in Turkey’s post- 1980 adjustment effort. This point is important for our subsequent discussions, requiring preliminary remarks at this juncture. Although committed to a rapid disinflation in the early stage, the policymakers were well aware of the potentially adverse consequences of a prolonged recession in the Turkish economy. Too rapid a squeeze in the real sector could have discouraged the manufacturing-based export initiative, which required complementary actions in a wide range of sectors. An early success in export promotion was perceived to be essential to restore creditworthiness, establish the credibility of liberalization measures, and extend penetration in foreign markets for a sustained export drive in the future. After attaining a reduction in the annual rate of inflation from over 100 percent in mid-1980 to around 30 percent in mid-1981, the policymakers began to emphasize export-led output expansion also to avoid higher unemployment in the urban sector, which had experienced a sharp fall in real wages in 1979 and 1980. In fact, to reinforce the growth process, public investment was increased (about 9 percent) in real terms in 1981, partly to offset the continuing decline in private investment, despite the advice to the contrary by the international financial agencies. Against the background of continual upward adjustments in the exchange rate and SEE prices, domestic inflation settled around a 30 to 35 percent core rate in the post-1981 period, which saw a fairly steady expansion of output through 1986 (as will be discussed in chapter 5). In turn, the persistence of inflation around this rate (with a substantial amount of variability) reduced the information content of relative price changes, possibly retarding deeper allocational restructuring in the economy.

4.4 Policy Mix and Sequence After this discussion of the initial conditions and overall policy trends, we now take a closer look at the pattern and content of policy measures introduced in early 1980 and extended thereafter. The time frame for our

668

Merih CelAsun and Dani Rodrik

present analysis is the 1980-85 period. For convenience in presentation, the entire set of measures is loosely referred to as the post-1980 program. The time horizon is extended to 1986 in chapter 8, in which we deal with public sector finances. While benefiting from hindsight, an ex post classification of the wide range of measures taken in 1980-85 involves three sets of methodological problems. First, the policy measures had mutually interdependent effects on the observed economic performance. Second, their effects spilled over time in a non-uniform fashion. Third, some of the more important measures were qualitative in character, the effects of which were expected to be seen in the longer run. Thus, a precise mapping of policy instruments onto policy objectives, on the basis of a well-defined set of criteria, is neither possible nor critically necessary in the context of our principal concern with the basic pattern of Turkish policies in this period. Having stated these caveats, in this section we present a crude policy taxonomy which is shown in table 4.1. As can be seen from the table, the main policy measures implemented from 1980 to 1985 can be classified in three main categories: (1) measures basically aiming at stabilization and export-led recovery, (2) liberalization measures, and (3) supportive fiscal and institutional actions. Four successive policy stages are identified within the 1980-85 period, which roughly take place in 1980, 1981-82, 1983, and 1984-8s. In the table, the implementation subperiods for specific policy measures are marked by the sign (x), where the occasionally used sign (?) indicates the doubtful nature or weak application of the corresponding measure. Although a further breakdown of the policy measures and subperiods is possible, the level of disaggregation adopted in table 4.1 appears to be sufficient to identify the seven main characteristics of the policy mix and sequencing as follows. (1) Because of the severity of internal and external imbalances, the program emphasized macroeconomic stabilization in the first year with a heavy reliance on price shocks (mainly exchange rate devaluations and increases in the price of public enterprise products), which complemented monetary and budget restraint. The brunt of adjustment in the budget was on the SEE subsidies and social expenditures. (2) With an early emphasis on export promotion, a clear signal was given to producers that output recovery would be induced by export expansion. The maxi-devaluation of January 1980 was followed by frequent minidevaluations through May 1981. From May 1981 onward (to the end of 1983), the exchange rate was adjusted daily against a currency basket. Besides devaluations, export incentives included tax rebates, credit subsidies, an exchange retention scheme, and duty-free imports for the production of exportables.

669 Table 4.1

TurkeyKhapter 4

Mix and Sequence of Policy Measures, 1980-85 1980 (1)

A. Stabilization and export-led recovery I . Price shocks a. Maxi-devaluation b. Interest rate shock c. SEE Price hikes 2. Flexible pricing of exchange rate, interest rates (partial), and industnal products (partial) 3. Supportive incomes policy for downward flexibility of real wages and agricultural prices 4. Tight monetary stance 5 . Budgetary restraint 6. Export incentives (exceeding 20% total subsidy on eligible manufactures) 7. Debt relief and net new lending B . Liberalization measures 1. Flexible pricing (same as A.2): deregulation of product markets (partial), decontrol of interest rates (partial) 2. Import liberalization a. Removal of quotas, retention of licensing b. Introduction of a Prohibited List c. Major reductions in licensing d. Realignment of tariffs 3. Partial decontrol of external financial flows 4. SEE reform legislation 5 . Financial development a. Capital Market Board b. Framework for stock exchange c. Bank supervision system 6. Encouragement of foreign direct investment C . Supportive fiscal actions and institutional changes I . Restructuring public investment 2. Tax incentives for financial intermediation 3. Income tax reform 4 . Introduction of VAT system 5. External debt management (improved debt reporting system) 6. Sector-specific actions (planning and pricing schemes for energy and agriculture) 7. Extrabudgetary funds

1981-82

1983

1984-85

(2)

(3)

(4)

X

X

X

X

X

X X

X

?

X

X

X

X

X

X

X

X

X

X X X X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X

X X

Source: The authors’ ex post classification of major policy measures

(3) The initial stabilization effort was strengthened by two additional factors: namely, a supportive incomes policy, and sizable debt relief and new borrowing. The incomes policy worked in implicit ways through restrictions on labor union activity, delays in wage adjustments, real reductions in the salaries of government employees, and restrained nominal increases in agricultural support prices. The increased foreign exchange availability

670

Merih Cellsun and Dani Rodrik

contributed to the elimination of hoarding of essential commodities, enhancing the workability of the new program. (4) The overall liberalization process proceeded in four distinguishable steps. The sequence was as follows: First, industrial product markets were deregulated (with the initial exceptions of fertilizers, coal, electricity, sugar, and certain oil products). Second, interest rates were to a large extent deregulated in mid-1980. This step consisted of decontrol of bank deposit rates and free setting of nonpreferential lending rates. Third, a preliminary step in import liberalization came in mid-1981 in the form of the removal of the quota list from the import regime. With the elimination of quotas, the import policy relied on licensing and prohibited imports as restrictive devices until 1984. Fourth, the import regime was further liberalized and a partial decontrol of external financial flows was undertaken in one unified major step at the end of 1983 and early 1984. (5) Financial sector reforms, SEE reorganization, and encouragement of foreign direct investment (FDI) constituted other supplementary measures with varying degrees of strength and success. Besides the deregulation of interest rates in mid-1980, the major policy initiatives for the financial sector included the establishment of the Capital Market Board (1981), introduction of a new framework for the stock exchange (1983), legislation of a new banking law and regulatory system (1983-85), and creation of an interbank money market (1986). (6) On the SEE front, the main emphasis was on price flexibility, a hiring freeze, and real wage reductions. The reform legislation for the SEEs came rather late (1983). The new legislation delineated the particular SEEs that would function on the basis of market criteria. Privatization studies for the SEEs were initiated in 1985-86. Policy measures related to FDI were mainly directed toward a more flexible and simplified application of the existing legislation on FDI, which contained highly liberal provisions.8 (7) Finally, the remainder of the policy measures shown in table 4.1 are classified as supportive fiscal actions and institutional changes. This category includes rationalization of public investment, tax incentives for exports and financial intermediation, income tax reform, adoption of the value-added tax (VAT) system, and sector-specific actions for energy and agriculture (mainly involving improved planning and pricing schemes). Income tax reform was helpful in creating more realistic tax brackets and reducing marginal rates for wage and salary earners. After showing an initial rise (to about 20 percent) in 1981, the tax/GNP ratio dropped (to 16.6 percent) in 1984. Because of the lagging tax performance, the overall budget correction was achieved instead through the SEE price hikes and restrained current expenditures. As summarized above, Turkey’s liberalization drive proceeded in successive stages. In its basic outline, it followed the general pattern suggested in the literature, which argues in favor of the following sequence: fiscal correction, deregulation of product and financial markets, liberalization

671

TurkeyIChapter 4

of the current account, and decontrol of the external financial flows in the balance of payments.’ Turkey’s liberalization effort should not be viewed, however, as a completed process. As of late 1987, it still contained a number of incomplete aspects such as: a complicated system of preferential credits, fragmented financial markets, political interference in SEES, and arbitrarily handled import permissions and levies. Despite the remaining imperfections, these reforms constitute in their entirety a credible policy package, which broadened the role of market forces in determining the direction of the Turkish economy. Some of the key aspects of policy pertaining to relative prices are reviewed in sections 4.6 and 4.7, following some remarks on the role of the IMF and World Bank in the next section.

4.5 IMF and World Bank Conditionality The implementation of the 1980-85 policy measures outlined in section 4.4 benefited from the balance-of-payments assistance and policy support of the major creditors and international financial organizations. A three-year standby arrangement concluded in June 1980 with the IMF (SDR 1,200 million, 625 percent of quota) was followed by a one-year standby in June 1983 (SDR 225 million), which was later cancelled and replaced by a final one-year arrangement with the Ozal government in April 1984. In turn, as shown in table 4.2, the World Bank provided five structural adjustment loans (SALs) totaling about $1.6 billion in support for liberalization reforms and rationalization programs in the energy, agricultural, and financial sectors in addition to regular project lending. The smooth and effective policy cooperation with the IMF and World Bank facilitated multilateral debt relief agreements and concessional bilateral lending in the 1980-85 period. In their work with Turkey, the IMF and World Bank collaborated closely, with their conditionalities determining, to an important extent, the contents and modalities of the policy frame sketched in table 4.1. The overall concern of the IMF with the management of aggregate demand, the payments

Table 4.2

Loan

World Bank Structural Adjustment Loans (SALs) for Turkey

Date of Approval

Amount (million $)

Disbursements to 31 December 1984

3/25/80 11/18/80 51 12181 5/27/82 5/23/83 6/14/84

200.00 75.00 300.00 304.50 300.80 376.00

200.00 75.00 300.00 304.50 300.80 250.00

SAL 1 Supplement SAL 2 SAL 3 SAL 4 SAL 5 Source: Yagci et al. (1985)

672

Merih Celasun and Dani Rodrik

regime, and the exchange rate was complemented by the efforts of the World Bank focusing on trade liberalization, resource mobilization, financial development, public investment planning, SEE reorganization, and sectorspecific issues. The IMF relied on the standard performance criteria, emphasizing interest rate reform, ceilings on net domestic assets of the central bank, subceilings on central bank credits to the SEES, limits on contracting new external debt, and currency depreciation in a unified framework. The World Bank was particularly effective in its public investment review and trade policy conditionalities, which are listed in table 4.3. The policy actions envisaged in five successive SALs for the gradual liberalization of the import regime, as disclosed in table 4.3, illustrate how deeply the World Bank was involved in the policy process. We will return to the role of these Bretton Woods institutions in chapter 9.

4.6

SEE Price Hikes and Agricultural Support Prices

As indicated in the policy overview above, a key aspect of the reform package in 1980 was a restructuring of key relative prices within the Table 4.3

World Bank SAL Conditionality for ImporI Liberalization

Issues and Objectives

Issues 1 . Exchange rate policy

2. Elimination of quotas 3. Rationalization of tariffs Objectives

Improved efficiency of production and support exports by reduction in anti-export bias Policy Measures Envisaged SAL I (1980) and SAL 2 (1981) Flexible exchange rate policy after January 1980 (SAL I ) , daily adjustment after May 1981 (SAL 2 )

Initiation of a protection study (SAL 1) Abolishment of the quota list in the 19x1 import regime and reduced licensing of imports (SAL 2) SAL 3 (1982)

Further reductions in licensing and simplified procedures; continual expansion of the liberalization list Commitment to rationalize the tariff system over the next five years Preparation of a list of prohibited items SAL 4 (1983)

Introduction of a rational tariff structure, adoption of a timetable by September 1983 for shifting from licensing system to one relying on tariffs SAL S ( I 984)

Reduce remaining licensing for imports to a negligible level during the fifth plan period (1985-89) drawing on the recommendations of the completed protection study Source: Yagci et al. (1985) and World Bank.

673

TurkeyIChapter 4

economy. We start our discussion here by looking more closely at SEE prices and agricultural support prices. Table 4.4 documents price changes for selected SEE and agricultural support items from 1979 to 1983. It also gives the official estimates for percentage annual changes in the wholesale price index (WPI), SEE revenue, and real GDP over the same period. The data in table 4.4 reveal that, relative to the WPI, SEE prices moved sharply upward and agricultural support prices moved generally downward. In interpreting the agricultural price movements, we should also note that prices of agricultural inputs (such as fertilizers and mafarin) shifted sharply upward in real terms, as subsidies were reduced. Thus, the impact of new policy measures was more pronounced on the net rather than gross prices of agricultural products. Moreover, delays in payments for agricultural support purchases also served as a mechanism to restrain farmers' real incomes. The SEE sales revenue showed a substantial rise in real terms in 1980-82, especially in the first year of the program. In the context of a falling real GDP in 1980, we may infer that the SEE sales revenue increased almost entirely due to price hikes, which were introduced in one major step in early 1980. Hence, the sharp rise in the inflation rate in 1980:I and 198031 (see table 4.5 below) reflected to a substantial extent the cumulative impact of the maxi-devaluation, SEE price corrections, and deregulation of private industrial prices. Table 4.4

Changes in Selected SEE and Agricultural Support Prices -

Percentage Annual Increase

A. SEE items Electricity Lignite Fertilizer (TSP) Cement sugar Motorin Paper Pig iron B . Agricultural suppport items Wheat Cotton Tobacco Tea Sugar beet Hazelnuts Memo items: WPI" SEE sales revenueb GDP (real)

1979

I980

38 72 0 62 62 72 21 89

153 131 824 177 171 225 23 1 121

57 82 22 21 58 74

64 69 -0.6

"Treasury wholesale price index. bExcluding budget transfers that cover duty losses.

1981

1982

49

46

61 18 10 30 I2 32 40 66

0 21 I9 28 14 21

103 I00 83 91 118 193

83 26 24 48 48 14

22 24 53 34 28 20

29 22 34 32 16 17

107 163 - 1.0

37 53 4.7

25 51 4.3

31 35 4. I

101

52 39 82 64 39

1983

10

40

674

Merih Cellsun and Dani Rodrik

Impressionistic evidence on SEE sales revenues and real GDP growth rates (as shown in table 4.4) suggests that the demand curves for most SEE products were largely inelastic in the short run. In the absence of sufficient competition from imports (before 1984), the emphasis on SEE price hikes to augment public revenue resulted in significant departures from competitive pricing in domestic product markets. Moreover, the available data on private industry suggest that the profit markups have not changed significantly in the post- 1980 period." These observations suggest that the deregulation of industrial prices has not produced by the mid- 1980s a highly competitive market structure in the Turkish economy.

4.7

Interest Rates, the Exchange Rate, and the Monetary Stance

Aside from SEE prices and agricultural support payments, two other key prices directly controlled by the government were the interest rate and the exchange rate. The sharp change in the policy attitude toward the latter two macroeconomic prices was perhaps the most noteworthy aspect of the post-1980 program. Table 4.5 provides quarterly data from 1979:I to 1985:IV on inflation rates, real interest rates, and real exchange rate changes. The inflation rates are measured as percentage changes in the WPI over the previous year; the real interest rates are after-tax returns on one-year bank deposits deflated by the WPI; and the exchange rate depreciations are the percentage changes over the previous year in an export-weighted index of the real exchange rate (increases correspond to depreciations). A scrutiny of the data shown in table 4.5, especially from 1980:II to 1981:11, points to the brisk realization of a 30 percent exchange rate depreciation and a nearly 45 percent rise in the real deposit rate at the outset of the program. These produced a ratchet effect in the overall price level, and in combination with the SEE price hikes led to an unprecedented price shock in the economy. In the absence of full monetary accommodation, the resulting squeeze on real money balances produced a temporary recession, but also served to cut the annual rate of inflation from 115 to 35 percent in a relatively short span of time. The effect of these relative-price shocks on the macroeconomic balance can be observed in the trend for real money balances. Figures 4.1 and 4.2 show the quarterly movements in the levels and annualized rates of change of the real monetary base (MB) and of real broad money (M2), from 1979:I to 1985:IV. As pointed out in chapter 1, M2 is a reasonable overall indicator for the scale of the Turkish financial sector, which is dominated by the banking system. From 1979:I to 1980:II, the cumulative real decline in M2 was nearly 40 percent, with most of the reduction taking place in the first two quarters of 1980. This gave rise to a heavy squeeze on the banking system and correspondingly to a rapid expansion of the unorganized credit market.

TurkeyIChapter 4

675

Table 4.5

InRation, Interest Rate, and the Exchange Rate % Change from Previous Year’s Same Quarter

1979:I I1 III IV 1980:l I1 111

IV 1981:l I1 111 IV 1982: 1

II 111

IV 1983:I I1 111

IV 19843

I1 111 IV 1985:I I1 111

IV

Inflation Rate (WPI)”

REERb Depreciation Rate

Real (Net) Interest Rate (I-yr deposits)

50.2 62.7 66.6 72. I 112.2 115.6 101.5 98.6 60.2 35. I 37.2 27.5 26.2 30.0 24.7 22.4 24.6 25.8 31.2 39.0 44.1 54.6 54.7 52.2 49.2 40.2 35.8 37.7

- 10.3 -20.2 -2.7 -8.6 27.2 38.6 25.1 30.5 - 0.6 - 2.4 -0.7 13.1 17.1 14.8 16.3 10.9 7.6 7.4 I .7 0.3 6.8 4.4 3. I I .o -6.4 0.3 4.6 4.3

-25.2 -29.6 -31.5 -35.9 -43.5 -42.0 -32.3 -32.5 -11.3 2.8 0.7 8.7 8.1 4.5 8.7 11.5 7.4 7.8 2.3 I .9 9.3 2.0 1.9 2.7 3.3 10.3 9.7 8.5

Source: Central bank of Turkey for the WPI and real (net) interest rates; own calculations for the real effective exchange rate.

*The WPI denotes the Treasury wholesale price index.

q h e REER is the real effective exchange rate (export weighted, using WPI)

In such a context, where the rate of decrease in real M2 had reached 30 percent (in annual terms) in 1980:II, the government deregulated deposit rates in July 1980 and freed nonpreferential lending rates. The deregulation of deposit rates was at first followed by a “gentlemen’s agreement” among major banks, which put a ceiling on the nominal rates. But eventually collusion collapsed under the impetus of the smaller, more aggressive banks, as well as the pressure from the unorganized money market. The real net interest rate on decontrolled deposits went from an average of - 32 percent in 1980 to 9 percent by 1981:IV, and to an average of 8 percent in 1982. Following the decontrol of deposit rates and the introduction of certificates of deposit (CDs), real balances for M2 increased at much faster rates than the real MB from 1980:II to 1982:IV, indicating a substitution of interest-bearing accounts for cash and an upward drift in the monetary

676

Merih Cellsun and Dani Rodrik

Fig. 4.1 Real money balances (in constant 1963 prices)

multiplier (M2/MB). Hence the high interest rates contributed to the monetization of the economy as it recovered from the recession. While broad money responded favorably to the rise in deposit rates, a consequence of financial reform was that the lending rates for nonpreferential credits soared. They reached unsustainable levels of 25-30 percent in real terms in 1981-82. A significant part (guesstimates running around 40 to 60 percent) of the nominal credit expansion in this subperiod was directed toward refinancing the interest payments connected with nonperforming loans. The illiquidity problems of the private corporate sector stimulated the activities of the so-called bankers (brokerage houses) that traded bank CDs and corporate bonds and provided loans at rates higher than nonpreferential bank lending rates. The unregulated activities of the unorganized credit market eventually produced a financial crisis in mid- 1982 after the collapse

677

TurkeyIChapter 4

Fig. 4.2 Percentage change in real money balances (from the previous year's same quarter)

of the largest broker (Banker Kastelli).'* The response to this crisis included, besides the replacement of key cabinet ministers, a relaxation of monetary policy, which had been quite tight at first, and the introduction of new guidelines for the financial system. At the end of 1982, the central bank was reauthorized to fix ceilings on deposit rates. As shown in table 4.5, real deposit rates in 1983-84 were lower than in 1981-82. However, they exhibited a large degree of variability during 1983-84. What was the cause of this variability, and why has it apparently lessened from 1985:II onward? The fluctuations in real deposit rates in 1983-84 reflected to a large extent the attempt of policymakers to realign the real rates according to the official inflation targets, which were, however, persistently exceeded by the realized

678

Merih Cellsun and Dani Rodrik

inflation rates. This caused an instability in expected rates, resulting in the stagnation of demand for broad money (see fig. 4.1). From 1985:II on, real deposit rates were stabilized at a higher level, partly as a response to the introduction of foreign exchange deposit accounts with Turkish commercial banks at competitive interest rates. The new interest rate policy began to take into account the arbitrage equilibrium conditions to contain currency substitution within reasonable limits. Hence, from early 1985 onward, domestic real deposit rates tended to match the world rates (about 4 to 5 percent) adjusted by a margin corresponding to the rate of real depreciation of the exchange rate plus a perceived risk premium. We may conclude that the prior experience with domestic financial liberalization was helpful in handling the new situation brought about by the partial decontrol of the capital account at a later stage. But the latter policy has also implied higher real interest rates domestically, under the joint influence of financial openness and sustained real exchange rate depreciations.

4.8 A Missing Element: Political Participation and Contestation By postponing the analysis of the actual outturn to chapter 5, in the present chapter we attempted to provide an overall review of the 1980-85 policy measures. Our discussion emphasized the supportive factors, sequencing patterns, and selected technical characteristics pertaining to the pricing aspects. Our policy review leads us to conclude that the policy mix in this period as a whole was one of the most comprehensive country adjustment programs applied in recent years with the full support of the multilateral lending agencies. Notwithstanding the social costs involved, the initial strength and sustained implementation of the program were quite impressive in technical terms, especially in the light of the ineffective policy trials of the earlier periods. In concluding this chapter, we may question whether the 1980-85 policy episode had any missing element in an important sense. Our answer is an affirmative one, and we suggest that broad political participation and contestation were crucial elements missing in this important national experience. The bulk of the 1980-85 program coincided with transitional military rule in Turkey, such rule having been instigated essentially on noneconomic grounds. The program did not sufficiently benefit from critical evaluations and possibly constructive proposals of the various groups of participants in the political and economic life of the country. The lack of political participation also undermined the medium-term policy planning process, which could have reduced the social costs of the program (to be reviewed at the end of the next chapter). On the other hand, the prevailing restrictions on political participation and contestation were clearly instrumental in providing the technocrats with the requisite autonomy to introduce a wide range of

679

TurkeyJChapter 5

radical reforms and the ability to withstand the distributional consequences. It is to be hoped that this missing element has not deeply hampered the foundations of a long-term social commitment to a more viable development strategy.

5

Performance and Adjustment Patterns in the 1980s

After the review of the policy measures in chapter 4, in the present chapter we aim to analyze the performance and adjustment patterns of the Turkish economy in the post-1980 period. Following a brief look at the actual outcome in section 5.1, in the remainder of the chapter we seek to explicate the major macrolevel mechanisms and linkages in Turkey’s recent adjustment experience. While leaving the quantitative treatment of selected topics to subsequent chapters, we focus here on the anatomy of the overall adjustment process, including the distributional aspects. The main argument in the present analysis is that changes in macroeconomic prices have played a determining role in Turkey’s overall adjustment effort. In this context, we consider the following as macroeconomic prices: the exchange rate, interest rates, SEE prices, real (urban) wages, and net prices (or domestic terms of trade) for major sectors. In the Turkish setting, real wages and sectoral net prices were determined by and large as residual variables, while the exchange rate, interest rate, and SEE prices served more directly as policy instruments. On the subject of relative prices and distortions, the economic literature has been mainly concerned with microlevel efficiency and welfare issues. As aptly analyzed by Balassa (1987) recently, and Krueger (1974b) and Bhagwati (1971) earlier, policy-induced market distortions tend to have adverse effects on resource allocation and employment. The permanent removal of these distortions would involve transitional costs, but could bring a continuous stream of future benefits, as emphasized by Fischer (1986). The post-1980 Turkish policies did make a genuine effort to remove a wide range of distortions that prevailed in the pre- 1980 period, but market imperfections and related inefficiencies continue to exist, as discussed in the context of SEE prices in section 4.6. The investigation of remaining microeconomic distortions and their allocational effects is an important item in the agenda of future empirical research on Turkey. But our emphasis in the present chapter is on the macroeconomic consequences of a sharply altered relative-price structure, as took place in the Turkish economy.

680

Merih Cellsun and Dani Rodrik

From the analytical standpoint, an integrated treatment of the various price and nonprice components of the overall adjustment process is a highly complex undertaking. It is preferably conducted within a multiperiod computable general equilibrium (CGE) framework, incorporating a macromonetary module that captures the inflationary phenomena. The CGE model described in the next chapter lacks explicit mechanisms for SEE pricing, inflation, and monetary details. Hence, we have opted in this chapter for an analytic review which is perforce more ad hoc, but has the advantage that it does not miss out on important aspects. We make extensive use of the nominal- and constant-price series on income, expenditure, and savings, which have not been closely examined in earlier studies on the Turkish economy.’ 5.1

Macroeconomic Performance and External Debt

Table 5.1 presents an overview of the macroeconomic performance from 1980 to 1985, including partially available data for 1986.3 The GDP growth figures for the industrial and Middle Eastern economies, shown as memo items in the table, bring out the inhospitable nature of the external environment, especially from 1980 to 1983. Besides the economic slowdown of its trade partners, Turkey, as an oil-importing country, was also faced with the second oil shock, which involved more than a doubling of imported oil prices in the wake of 1980. The latter shock, in combination with devaluation, resulted in a 23 percent fall in the terms of foreign trade in 1980. Because of continued exchange rate depreciations, the foreign terms of trade showed further declines in 1981-83. The policy initiatives in 1980 rapidly produced credible results at the macrolevel by lowering inflation and restoring growth in 1981-82. The resumption of growth reflected in large part export expansion, as can be observed from the changes in trade ratios. The year 1983 saw some setbacks in policy and performance with postponed adjustment in SEE prices. The liberalization process and price corrections gained renewed strength in 1984, as discussed in chapter 4. But inflation began to accelerate. In 1986 the sizable fall in the price of imported oil was not adequately exploited in the pursuit of price stabilization, thanks to the uncurbed rise in public expenditures. The figures for current account balances and their financing patterns are assembled in table 5.2 for the 1980-86 period. To bring out the contribution of the debt relief (granted by the OECD governments), current balances are given for before and after debt relief.4 In a complementary way, table 5.3 lists the main indicators of external debt for the 1981-86 period. In a cumulative way, debt relief accounted for nearly 44 percent of total financing for current deficits in 1980-85. Furthermore, Turkey benefited from large-scale emergency financial assistance from its major creditors. In

681

TurkeyKhapter 5

Table 5.1

Macroeconomic Performance, 1980-86 1980

1981

1982

1983

1984

1985

1986"

- 1.1

4. I 0. I 9.5 0.9 41.9

4.5 6.4 5.4 0.9 27.4

3.3 -0.1 8.7 0.7 28.0

5.9 3.5 10.2 1.3 49.8

5.1 2.4 5.5 43.6

7.8 7. I 10.2 2.1 32.8

-8.7 9.4

5.5 2.2

4.8 I .7

8.8 I .8

7.8 13.3

13.5 10.2

-4.9 8.4

0.6 0.9

4.2 2.0

4.7 1.3

5.5 3.0

3.6 3.2

7.8 7.4

5 .O 6.3 13.6 6.6 7.0

8 .0 10.2 15.2 6.6 8.6

11.0 14.8 16.5 7.0 9.5

11.6 15.6

14.7 19.4 21.4 7.3 14. I

15.6 21.5 21.9 6.8 15. I

12.9 18.4 19. I 3.9 15.2

1.9

-2.2

A . o/o Annual incrense

I . Real value added GNP Agriculture Manufacturing 2. Employment 3. GNP deflator 4. Fixed investment Private Public 5 . Consumption expenditure Private Public H . Trade ratios ( W of GNP,

I .7 -6.4 -0.1 103.9 - 17.3

-3.7

1.1

current prices)

I . Exports of goods (f.o.b.)b 2. Exports of goods and services 3. Imports of goods ( c . i . f . ) Oil Nonoil 4. Cument account balance (after debt relief)

18.1

7.2 10.9

-5.5

-3.5

-2.2

-4. I

2.0 3.5 -2.1 4.8 4.6

I .6 2.2 -1.8 0.7 4.7

0.5 1.6 -0.2 -0.4 5.0

2.6 1.3 0. I - 3.5 7.2

-2.8

-

Memo items:

Real GDP growth ('70) Industrial countries LDCs Middle East 15 heavily indebted Exporters of manufacturers

4.4 4. I 0.7 2.2 8.3

2.9 3.2 - 1.6 3.1 6.6

3.1 3.0 0.2 I .5 6. I

Soin-re: SPO and the central bank of Turkey for Turkish data; IMF World Economic Outlook 1986.

pp. 183-84, for memo items. "Provisional estimates (March 1987). blncludes transit trade.

1980, the first year of the program, the latter aid resulted in a net resource transfer from abroad (excluding relatively minor items connected with FDI) amounting to a positive 4.7 percent of GNP. The net resource transfer/GNP ratio was also quite significant in 1981, materializing around a positive 1 percent. In the post-1982 period ushered in by the Mexican financial crisis, no major LDC debtor benefited from such a positive and sizable resource transfer in launching its adjustment program. We will return to this aspect of the Turkish experience in chapter 9 when we look at external financial relations more closely. The data summarized in tables 5.2 and 5.3 also point to the reversal in debt-service trends starting in 1982. The noninterest current balances turned positive (implying negative resource transfers for the first time), and debt service as a percentage of GNP began to increase rapidly (especially after

Table 5.2

Financing the Current Account, 1980-86' (million $)

1980

1981

1982

-668 -2.270 -2,938

-1,184 -485 - 1,669

-1.465 630 -835

-3,408 -475 I8 -512

-1.919 -loo

1983

1984

1985

-1,441

- ,586

- 1,753

-387 - 1,828

179 - ,407

740 - 1.013

-1,898 - 62 46 -269 161

-

,407 150 1 I3 208 - 171 1.257 1,046 580 466 - 142 353

-

1,104

- I ,858

1986 (provisional)

1980-85 (billion $)

Parf A: Afrer debt re/ief

Interest payments Nonintereat current account Current account balanceb P a n B: Before debt relief Current account balance Nondebt financing Foreign direct investment Changes in reserves Counterpart to valuation changes Net foreign borrowing Long term (LT) Debt relieF Other LT (net) IMF (net use) Implied short term

95

3.883 2,194 1,450 744 422 1.267

-263 68 2,019 1.165 850 315 268 586

-586 1,254 2,745

-620 -1.804 577

19

-935 -229 55 -297 13 1,164

1 .960

1,030 750 280 129 5

303 I.000 - 697 117 I .540

-2.134 606 - 1,528

1,013 298 95 - 20 223 715 - 20

- 1.528

- 20

525 -241 1.413

- 103

838

- 169

125 - 545

25 1 1,697 525

-8.1 - 1.6

-9.7 10.6 -0.4 0.4 - 1.2 0.3

-

1 1 .o

5.7 4.6 1.1

0.7 4.6

-

100.0%)

( - 4.0%)

(4.0%) ( - 10.9%)

(3.0%) (104.0%) (54.1%) (43.8%) (10.3%) (6.5%) (43.4%)

Memo items:

Debt repayment (LT, after relief) Debt service (after relief) Net resource transfer

-

-952 -2,417 -401

-

1,066

-

- 2.507

- 2,690

139

- 329

I - 1.038 - 3,6l

- 2,145 -4,279 - 437

-6.2 - 14.3

1.7

Source: Central bank of Turkey, IMF, and OECD (1986). The Present data partly reflect the latest available estimates provided in the 1986 Annual Report of the central hank. NOIP: Numbers in parentheses are the percentage distribution of current account financing.

dBased on the revised presentation of the balance of payments. bCorresponds to foreign savings in the post-1983 national accounts. 'Interest plus principal.

683

TurkeyKhapter 5

Table 5.3

External Debt Indicators, 1981-86

A. Billion$ 1 . Debt,’ of which Short term ( I)h Short term (2)’ 2. Net international reserves Excluding gold Including goldd 3. Net indebtedness (incl. gold) B . B of GNP I . Debt 2. Interest payments 3. Debt service‘ 4. Net resource transfer‘ C. B of Exports of goods and services I . Debt 2. Debt service‘ D. B of Imports of goods (c.i.f.) International reserves (excl. gold)

1981

1982

1983

1984

1985

1986e

16.9 2.2 2.2

17.6 2.2 1.8

18.4 3.0 2.3

21.3 4.5 3.2

25.3 6.6 4.8

31.2 9.4 6.9

1.5

1.8

1.9

2.7 3.5 17.8

2.2 3.3 22.0

3.1 4.3 26.9

28.6 2.0 6.8

36.1 2.8 9.4 0.3

42.4 3.2

1 .o

32.8 2.7 7.9 -0.7

-0.7

47.8 3.3 15.9 -2.0

53. I 3.6 19.0 -0.7

280.2 66.6

222.2 53.5

231.4 60.3

218.2 60.7

223.3 74.3

278.8 99.8

16.8

20.6

20.9

24.9

19.3

27.8

11.8

Source: All figures are based on and/or derived from the central bank’s Annual Reports (1983-86) and measured in current prices and official exchange rates.

aExtemal debt outstanding and disbursed. blncludes all Dresdner scheme deposits. ‘Excludes Dresdner scheme deposits with one-year or longer maturity. %old is revalued at 216.7, 277.0, and 332.0 dollars per ounce at the end of 1984, 1985, and 1986, respectively.

‘Debt service refers to the sum of interest payments. amortization on medium- and long-term debt, and all short-term debt of less than one-year maturity. ‘A minus sign indicates net resource transfer (outflow) abroad. eF’rovisional estimates.

1984), generating additional strains on public finance, as will be discussed more extensively in chapter 8. Notice, however, that the rise in the external debt stock from 1984 to 1986 reflects not only the impact of new borrowing, but also the valuation effects of dollar depreciation in this subperiod. Only around half of Turkey’s external debt is denominated in U.S. dollars (see table A.18 in the stat. app.). Perhaps the most distinguishing feature of the post-1980 experience is the rapid rise in export/GNP ratios, as shown in table 5.1. Although a fraction of this export expansion is attributable to overinvoicing, as will be explored in some detail in chapter 7, the export success has been a remarkable achievement, given the limited familiarity with export markets in the pre-reform period. Turkish foreign policy has been a supportive factor in broadening export penetration in the Middle Eastern region, particularly in Iran and Iraq, on the basis of special trade arrangements involving oil import schemes (see Akder 1987). The setback in the 1986 export performance reflected, in part, the drop in the oil revenues of the Middle Eastern trading

684

Merih CelLun and Dani Rodrik

partners. This prompted the authorities to introduce additional export incentives in early 1987 (including differential premia on specific product categories) to redirect exports to the OECD economies. A disappointing aspect of the post-1980 recovery effort was the decline in private fixed investment in the initial two years of the program. The cumulative fall of about 40 percent in real terms in investment expenditures from 1977 to 1982 was finally halted in 1983, but a buoyant investment climate in export-oriented sectors had not yet arrived by the mid-1980s. As a consequence, the share of the private sector in total fixed investment declined to 41 percent in 1978-84 from 52 percent in 1973-77. Rapidly growing public investments have been concentrated in the energy and services sectors. There seems to have been a general expectation on the part of the authorities that the increase in the production of exportables would be generated by private sector projects. However, the share of agriculture, mining, manufacturing, and tourism-the main sectors for exportables-in total private investment decreased to 41 percent in 1978-84 from 51 percent in 1973-77. These investment trends point to the lag in the restructuring of the economy in a way that would sustain an export-led expansion in the medium-term future.

5.2 Movements in Macroeconomic Prices Since the burden of macroeconomic adjustment fell on key relative prices in the Turkish economy, we start by examining the changes in the structure of macroeconomic prices in the post- 1978 period. We will look subsequently at the resulting income and expenditure patterns. Table 5.4 shows the trends in key prices in two groups, separating those that exhibited increases (in real terms) from those with decreases. In this table, the net prices of some major sectors (such as construction, transportation, etc.) are not reported to avoid undue clutter in the presentation. We take 1978 as a base because from then on, the Turkish economy began to adjust to the debt-precipitated crisis of mid-1977. Once again, an upward movement in the exchange rate (TLiforeign currency) indicates depreciation. The pattern in 1979 is different from that after 1980 in one important way. In the absence of adequate policy measures, as we described in chapter 3, the inflationary process resulted in real exchange rate appreciation and a lower real interest rate (on time deposits). But it also eroded the real consumption wage and agricultural net price quite substantially, a process that would not be reversed later on. The net prices of manufacturing and trade (i.e., wholesale and retail commerce) moved up this year. The policy shock in 1980 achieved a large depreciation of the real exchange rate and additional increases in the net prices of energy, manufacturing, and trade sectors. In turn, it resulted in a further fall in the real consumption wage, while eroding government employee salaries. The net (after tax) interest rate remained negative in 1980.

685

TurkeyIChapter 5

With the establishment of transitional military rule at the end of 1980, these relative prices, and especially those that affected distribution directly, were largely consolidated. The implicit incomes policy treated the 1981 structure of relative prices more or less as a guideline in the determination of urban wages and agricultural support prices. Thus, the real income losses incurred by urban wage earners and agricultural producers during 1978-80 became an economically permanent and politically unchallenged component of the 1981-85 policies. It is obvious that this particular aspect of the post-1980 program may not be a transferable item in the design of adjustment plans elsewhere. In those manufacturing branches that had a narrow export orientation and limited access to concessional export credits, the fall in wage shares could not fully offset the rise in interest payments, causing a decline in net profit shares in 1981 and 1982.5 The resulting illiquidity, and in some cases insolvency, problems were partly eased by the provision of full tax deductibility of interest payments, which lowered the corporate tax burden. In the case of the agricultural sector, the fall in net prices reflected the effects of (a) restrained support prices, (b) reduced subsidies on inputs, the prices of which increased in real terms, (c) limited incentive benefits from export promotion which heavily relied on manufactures, (d) unfavorable world agricultural prices, and (e) domestic demand shifts originating from the substantially reduced purchasing power of urban workers. The sharp rise in nonagricultural prices and export incentives resulted in an increase in the profit margins in wholesale and retail trade, which is reflected in the trade sector net prices in table 5.4. Energy and mining prices soared after 1980 mainly to improve the SEE profit position. The real exchange rate was depreciated in response to additional trade liberalization, as well as to keep the export momentum going. In line with the partial liberalization of external financial flows, real deposit rates were also increased in 1985 (after a fall in 1984). After 1983 the government introduced tax rebates on wage and salary earners’ consumption expenditures. This mechanism served as an instrument to decrease (through a restrained stance on gross wage adjustments) real product wages at a faster rate than real consumption wages, as the relevant figures show in table 5.4. Finally, this review of the post-1978 price structure can be made more striking by comparing it with the pattern of pre-1978 prices. Figures 5.1 and 5.2 illustrate the behavior of selected macroeconomic prices for the entire 1973-85 period. Figure 5.1 shows the large blows delivered to the Turkish economy by two rounds of oil price increases in 1973-74 and 1979-80, which caused the foreign terms of trade to fall steeply. The pattern depicted in this figure is of interest in that it reveals clearly the sharply divergent attitude of policymakers toward urban workers before and after the 1977 crisis. Prior to 1977, the real consumption wage was on a sharp upward trend even though the Turkish economy was hit by substantial terms-of-trade

686

Merih CelAsun and Dani Rodrik Relative Price Movements, 1978-85 (1978 = loo)*

Table 5.4

A. Increases Real exchange rateb Real interest rate on I-yr deposits (9%) Real sectoral net pricesC Energy mining Manufacturing Wholesale & retail trade B. Decreases Terms of foreign trade ($) Real product waged Real consumption wage' Real sectoral net prices' Agriculture Government services' Memo irem: Financial sector net price'

+

1978

1979

1980

1981

1982

1983

1984

1985

100

89

116

1 I9

136

I42

148

149

- 25

-31

- 38

0

8

5

4

8

100 100 100

95 Ill

126 I20 I19

145

160

118

120

I20 121

161 123 I22

178 120 122

230 1 I6 122

100 100 100

100

YO 86

77 76 66

71 77 72

67 76 61

66 69 65

75 66 63

76 53 M)

100 100

88 99

84 74

85 64

79 67

77 63

79 48

76 46

100

80

76

95

88

88

128

136

109

Source: Central bank of Turkey for real interest rates (averages of quarterly highest returns); State Institute of Statistics for sectoral prices and terms of foreign trade; and SPO for nominal wage data (Tirkiye GeneIi Ifci Ucreti) deflated as noted below.

'For all prices: 1978

=

100, except for the real interest rate (annual).

bReal effective exchange rate (TLiforeign currency; export weighted). 'Ratio of sectoral value added (factor cost) deflator to GDP deflator. dGross wages (labor cost) deflated by the GNP deflator. 'Net wages (after tax and tax rebates) deflated by the implicit deflator for private consumption in the national accounts. 'Government services value added comprises gross employee salaries.

shocks. After this date, real consumption wages appear to have borne the full brunt of the deterioration in the external terms of trade. Figure 5.2 shows further details of the contrasting policies. During the mid- 1970s, the agricultural terms of trade improved and the manufacturing net price declined, while the exchange rate appreciated, lowering the real costs of imported inputs. These trends were reversed in a sharp manner after 1978-80. As the picture shows, the transformation in the structure of relative prices has been quite a radical one in comparison to previous experience.

5.3 Relative Price Changes In this section, the various balances of Turkey's macroeconomy are presented both in current prices and in constant 1983 prices. The purpose is to bring out the impact of relative price changes on basic macroeconomic relations and patterns in pre- and post-1980 periods. At the outset, a word of caution is in order as regards the use of the 1983 base as a benchmark for the post-reform price system. It may be plausibly

687

TurkeyKhapter 5

150-

I

I

I

I

I

1

I

I

I

I

I

I

I

-

140 130130.

120 120,

'\

\

TOT

'\

\J \

\

110110.

\

\

100100,

90 -

80 70 60 -

50- Id73 I

5:

I

I 77

I

I 79

I

I

81

I

I 83

I

I

I

85

Fig. 5.1 Terms of foreign trade (TOT) and real consumption wages (RCW)

Fig. 5.2.

Selected real prices (1978 = 100)

Note: EER is real effective exchange rate; AGR is agricultural net price; MAN is manufacturing net price; and RPW is real product wage.

688

Merih CelAsun and Dani Rodrik

argued that the flow equilibrium exchange rate in the mid-1970s was lower (in terms of TUforeign currency) than the actual 1983 real exchange rate, the latter reflecting the effects both of the second oil shock and lower world agricultural prices. DerviS and Robinson (1978) estimated that the equilibrium exchange rate exceeded the actual parity by about 25 percent in 1976 and 50 percent in 1977. As shown in figure 5.2, the 1983 value of the real effective exchange rate exceeds the 1976-77 values by about 40 percent. And as analyzed further in chapter 6, the 1983 exchange rate was in fact overvalued with respect to the objective of greater trade liberalization as borne out by the actual outcome in 1984 and 1985. Besides exchange rate considerations, the 1983 base also reflects both price corrections (e.g., the SEE price adjustments) that were needed by the Turkish economy for budget balancing and stabilization. On account of such adjustments, and with the proviso that the 1983 exchange rate be considered to be close enough to its “equilibrium” level, the use of the 1983 price benchmark for the review of income-expenditure patterns in real terms appears to be broadly acceptable. Largely for convenience, we draw upon SPO (1985) data in our subsequent discussion. 5.3.1 Disposable Income

A largely unexploited data set in the existing literature on the Turkish economy is the available SPO statistics on the public-private split of the nation’s disposable income, namely GNP. Public disposable income is the difference between the public sector gross revenue, including SEE profits and depreciation, and income transfers to the private sector and rest of the world, including interest payments. In turn, private disposable income is the residual between GNP and public disposable income. In Turkey’s macroeconomic setting, a familiarity with the public-private split of GNP is essential to analyze income-expenditure patterns and overall saving (or resource mobilization) behavior (see tables A . 1 -A.4 in the stat. app.). Figures 5.3 and 5.4 show the ratios of public and private disposable incomes to GNP both in current and constant 1983 prices during the 1973-83 period. These complementary figures point to the fact that the transfer of income to the public sector from the private sector, as a fraction of GNP, was larger in real terms than in nominal (current price) terms. In other words, the transformation in the structure of relative prices greatly increased the real purchasing power of the public sector, and its command over the economy’s resources was enhanced more than the nominal-price series shows. Hence, in the post-1980 period, the shift in relative prices definitely favored the public sector over the private sector. In turn, the rapid reduction in the share of private disposable income in GNP, in constant prices, was the major factor behind the restraint in private expenditures. This made export expansion possible from the viewpoint of demand management, as it moderated domestic absorption. The flip side of

689

TurkeyKhapter 5

100

PRDY 2 )

90

a Z

(3 rc

0

8 80

70

I I 1973

I I 75

I 1 77

I I 79

I I 81

1

83

Fig. 5.3 Private disposable income (PRDY) Nore: PRDY( I ) is PRDYGNP in current prices; and PRDY(2) is PRDYGNP in constant 1983 prices.

the coin was that the enhanced resources of the public sector, thanks to relative price changes, allowed public savings to rise, reducing the fiscal deficit. In general, an income transfer from the private sector to the public sector can reduce economywide expenditures only if the marginal propensity to save is higher for the latter than it is for the former. This was clearly the case in Turkey: the private sector matched its cut in income with cuts in expenditures, whereas the public sector used the transfer mostly to enhance savings rather than to expand expenditures.

5.3.2 Income-Expenditure and Savings-Investment Patterns Table 5.5 presents data that show more precisely the impact of relative price changes on Turkey's macroeconomic balances. Parts A and B in this

690

Merih Cellsun and Dani Rodrik

25

20

a 2

W

'& 15

/

8

/

10

5

I

1973

I

I

75

1

1 77

I

I I 79

I 81

I

I 83

Fig. 5.4 Public disposable income (PDY) Noret PDY( I ) is PDYiGNP in current prices; and PDY(2) is PDYlGNP in constant 1983 prices.

table provide estimates for the savings-investment balances in current and constant prices, respectively. These comparative data indicate that the improvement in these balances from 1980 to 1983 was more substantial in the public sector than in the private sector, and more sharply so in real than in current prices. Part C in table 5.5 displays data on the growth rates of real (domestic final) expenditures, which may be compared with the GNP growth rates given in the bottom row of the table. During 1978-79, the reduction in the current deficit and expenditures was engineered by import compression. From 1980 to 1982, the growth of expenditures at rates below the GNP growth was achieved through expenditure switching toward exports, which was induced by economywide price adjustments and supplementary export incentives. As is clear from the data, there was a setback in switching policy in 1983. After a reasonable attempt at expenditure restraint in 1984-85, the

Relative Price Effects on Investment-Savings Balances, and Growth of Real Expenditure, 1976-86

Table 5.5

1976-77 Average

1978-79 Average

1980

1981

1982

1983

1984

1985

1986‘

18.4 8.9 9.5 16.0 12.0 4.0 2.4

21.4 9.9 11.5 15.9 10.6 5.3 5.5

21.5 8.3 13.2 18.0 9.4 8.6 3.5

20.3 8.3 12.0 18.1 9.2 8.9 2.2

20.6 9.1 11.5 16.5 9.2 7.3 4.1

19.6 9.6 10.0 16.8 9.2 7.6 2.8

20.5 9.5 11.0 18.6 9.4 9.2 1.9

23.2 11.0 12.2 21.0 12.1 8.9 2.2

-0.6 -5.2 -5.8

3.1 -5.5 -2.4

0.7 -6.2 -5.5

-0.4 - 2.4 -2.8

-0.1 - 1.8 - 1.9

- 3.3

-9.1

-1.1 - 7.1 -8.8

- 0.4

.o

-3.9 -2.9

0.8 -2.7 - 1.9

I .7 -5.3 -3.6

-3.5 -0.5 -2.9

A. 70GNP, current prices

Investment Private Public Domestic savings Private Public Foreign savings” Sectoral savings-investment balances Private Public Total ( = - foreign savings) B. lo GNP, constant 1983 prices Sectoral savings-investment balancesb Private Public Total ( = - foreign savings) C. Growth of Real Expenditures (% per year)‘ Private Public Total Memo item (% growth per year): Real GNP Source: SPO

24.9 12.5 12.4 19.1 11.9 7.2 5.8

- 10.7 - 19.8

8.2 11.4 8.9 5.9

1.2

-4.6 -3.5

-3.1

-2.2

0. I -4.2 -4.1

-7.2 -7.6

0.4 -5.3 -4.9

0.4 -3.5 -3.1

0. I -4.2 -4.1

-6.3 2.0 -4.1

-0.3 5.2 0.8

4.3 2.1 3.8

4.7 I .5 4.0

5.9 2.3 5.2

4.0 8.9 5.0

8.4 9.0 8.6

- 1.1

4. I

4.6

3.2

5.9

5.1

7.8

1.1

0.9

1

1.1 - 2.2

(1985) and the central bank’s 1986 Annual Report for 1977-83 and 1984-86 data. respectively.

‘Current account deficit (after debt relief). The 1984-86 figures follow the revised presentation of the central bank for the balance of payments bSee table A.2 for other data in constant 1983 prices. ‘Domestic final expenditure excluding inventory changes. “Provisional estimates.

692

Merih Celasun and Dani Rodrik

control over demand management was considerably weakened in 1986, when there was also a slowdown in export expansion. In 1986 the current deficit/GNP ratio was higher in constant 1983 prices than in current prices, revealing a worsening trend in the external balance in real terms. The counterpart of this development was a rise in the external debt stock. 5.3.3 PSBR and Its Financing Our estimates for the overall public sector borrowing requirement (PSBR) and its sources and financing patterns are given, as a percentage of GNP, in table 5.6 for the 1980-85 period. To draw attention to the initial conditions, this table also contains average figures for the 1978-79 period. As will be discussed further in chapter 8, the PSBR figures differ in their basic orders of magnitude from the public savings gaps. The differences have been due mainly to the valuation adjustments for SEE inventories. The rise in public savings has contributed more to the reduction in public savings gaps than the restraint on public investment. The share of the nonfinancial SEEs in total public savings underwent a drastic change, going from a negative 25 percent in 1979 to a positive 14 percent in 1982 and 50 percent in 1985. SEE inventory management also began to improve in 1981. Hence the improved financial performance of SEEs, again mainly through wage-price adjustments, was a key factor in the adjustment process. After the initial fall in 1981, the PSBRGNP ratio remained high through 1985, averaging 5.8 percent in 1982-85. The share of the central bank in PSBR financing steadily declined, while the corresponding share of domestic borrowing increased rapidly from 1984 onward. The contribution of external borrowing to PSBR financing was substantial at the outset of the new

Table 5.6

The PSBR and its Financing, 1978-85 (current prices, as a percentage of GNP) 1978-79 Average ~~

1980 ~~~~

1981

1982

1983

19x4

1985

-3.3 -1.3

~~

Public savings-investment balance Public private capital transfer Inventory revaluation fund (SEES) Increase in accounts payable. net

-5.5 0.0 -3.5 0.6

-6.2 -0.1 -3.9 0.3

-4.6 1.2 -1.7 1.3

-3.0 -0.1 - 1.5 -0.3

-4.2 -0.2 -2.0 1.1

-3.0 -0.3

-1.8 -0.6 -2.0 -0.5

PSBR

-8.4

-9.9

-3.7

-5.0

-5.3

-7.9

-4.9

Financing: External borrowing, net Domestic borrowing Long term Short term (treasury bills) Central bank, net Othe?

2.2

3.2

1.1

1.1

1.4 -0.3 3.6 I .5

0.2 0.9 3.5 2.2

2.5 1.3 0.8 0.6 2.0 -2.2

1.0 1.5 06 0.8 0.3 2.2

1.4 0.9 1.8 -0.9 06 2.3

2.7 2.3 0.7 1.6 0.7 2.1

Source: SPO and the central bank of Turkey.

"Includes changes in holdings of deposits and currency. SEE arrears. and errors and omissions.

0.5 2.7 1.9 09 I 3 0.4

693

TurkeyKhapter 5

program, facilitating the pursuit of financial stabilization in the Turkish economy. Further details on public finances will be covered in chapter 8. 5.3.4 Current Account Deficits and Imports The continuation and intensification of the economywide price distortions in the mid-1970s were enabled by heavy external borrowing. This observation is reinforced by figure 5.5, which shows the time patterns of the current (account) deficit as a fraction of GNP, both in current and constant price terms over the 1973-83 period. At the prevailing official exchange rates Turkey's GNP was unduly overstated in U.S. dollars in the mid-l970s, giving rise to a gross underestimation of the current deficit/GNP ratio, which at the time may not have looked alarming to most observers. At constant 1983 prices, however,

Fig. 5.5 Current account deficit (CA) Note; CA( 1) is CA/GNP in current prices; and CA(2) is CA/GNP in constant 1983 prices.

694

Merih Celasun and Dani Rodrik

this ratio was nearly 20 percent in 1976-77 and fell to about 9 percent in 1978-79. Accompanying this belated adjustment was a drop in real imports as shown in the tabulation of trade ratios in table 5.7. These official estimates may possibly contain some measurement errors. Nonetheless, they underline three simple, yet highly significant points. First, the scale of imports was unsustainable in 1976-77, but could not be evaluated realistically in current prices. The immediate problem in 1977 was not so much the adverse allocational effects of microlevel distortions produced by the trade regime, but rather the sheer size of merchandise imports to which the economy had become dependent in a structurally rigid way. This unsustainable pattern collapsed from the inherently destabilizing form of external borrowing described in chapter 2. Second, the import/GNP ratio in real terms was the same in 1980 as in 1979, which was an achievement in the face of the steep rise in the world oil price in 1980. This was made possible by the large-scale emergency financing extended to Turkey in the first year of the 1980-85 program. Third, the import/GNP ratio recovered only mildly in 1981-82 but, as the share of capital goods in total imports declined, there was increased scope for intermediate goods imports. This in turn made it possible to increase capacity-utilization rates and to embark on an export-led recovery in output. 5.4 Money and Credit

An integral part of the changes in the macroeconomic setting was the shifts in the money and credit system. Table 5.8 provides a summary of data on the major monetary variables and credit patterns. As discussed earlier in section 4.7 of chapter 4 in the context of quarterly data, the real demand for broad money (M2) responded favorably to the switch to positive real interest rates on time deposits. The new interest rate policy reversed the upward trend in the income velocity of M2 in the pre-1980 period and contributed to lowering the velocity from 5.0 in 1980 to about 3.5 in the 1982-85 period.6 A monetary modeling exercise by Fry (1986) suggests that the positive deposit rates contributed to the avoidance of protracted recession in the early 1980s through increased availability of

Table 5.7

Exports ( f .a. b. ) Imports (c.i.f.) Trade deficit Current deficit Source: SPO (1985)

Trade Ratios (as a percentage of GNP, constant 1983 prices)

1976-77

1978

1979

5.1 26.8 21.7 19.8

5.8 18.1 12.3 10.0

4.7 16.2 11.5 7.7

1980

1981

1982

1983

5.3 16.2 10.9 7.6

8.5 17.4 8.8 4.9

10. I 16.6 6.5 3. I

11.6 18.1 6.5 4. I

,

695

Table 5.8

TurkeyIChapter 5

Money and Credit Indicators, 1978-85"

Money multipliers M l/MB (monetary base) M2/MB Income velocity GNP/MB GNP/M I GNP/M2 Domestic credits ( D O b M2IDC GNP/DC Domestic credits (%Ib by: Deposit money banks Investment banks Central bank to: Public sector Treasury Public enterprises Private sector

1978

1979

1980

1981

1982

1983

1984

1985

I .3 I .5

I .4 I .6

1 .5

I .2 2.0

1.1

1.8

2.2

1.2 2. I

0.8 1.9

0.7 1.9

6.0 4.6 3.9

6.8 5.0 4.2

9.3 6.3 5.0

8.0 6.7 4.0

7.4 6.5 3.4

7.3 5.9 3.5

6.7 8.1 3.5

6.4 8.6 3.4

0.6 2.5

0.7 2.8 100 56 17 27 50

0.7 3.3

0.8 3.2

1 .o

3.3

1.0 3.3 loo 70 13 17 32

1.2 4.3

1.2 3.9 100 77 8

100

56 20 24 49

51

50

loo

loo

100

59 13 28 48 14 34 52

64 12 24 40 13 27 60

66 13 21 37 12 25 63

LO 22 68

loo

72 II 17 28 12 16 72

15

29 II 18

71

Source: Central bank of Turkey and OECD (1986). "Figures are rounded to the nearest unit. bNet of central bank advances to the banks.

credit (which is largely the counterpart of M2) to finance working capital, an important complementary input in the production process. With the reduction of the PSBWGNP ratio and decreased public sector reliance on the central bank, important shifts occurred in the structure of credits. The share of direct central bank credit (to the public sector) in total domestic credit could be reduced from 28 percent in 1980 to 17 and 15 percent in 1983 and 1985, respectively. The SEE share in direct central bank credit was reduced to 13 percent in 1985 from about 50 percent in 1980. Consequently, the reduced credit demand by the SEES enabled the rise in the private sector's share in total domestic credit from 52 percent in 1980 to about 72 percent in 1984. In this context, the rise in export credits was much faster than total credit expansion.

5.5 Anatomy of the Adjustment Process: A Summary The analysis of macroeconomic prices and balances in sections 5.2-5.4 complements the review of the actual performance in section 5.1. Our discussion has focused on the importance of preexisting conditions and on the heavily price-based policy mix. Prior to the 1980 program, Turkey's adjustment to the reduced capital inflows relied on import reduction and inflationary compression of domestic real expenditures, resulting in the loss of policy credibility at home and

696

Merih Cellsun and Dani Rodrik

abroad. Under such initial conditions, conventional wisdom suggested an export-oriented orthodox shock treatment, which had a chance to work in the absence of indexed price and wage adjustments. Having decided on the shock treatment option, policymakers relied heavily on sharp changes in the macroeconomic price structure, which in turn had vast consequences for unprotected income groups in the Turkish society. The exchange rate depreciation, SEE price hikes, and switch to positive real interest rates induced both income and substitution effects. The price corrections caused a transfer of real incomes to the public sector from the private sector, making a direct contribution to resource mobilization by the former. The reduction in private disposable income led to the contraction of private expenditures, with the brunt of adjustment falling on importintensive and interest-rate-sensitive components such as private fixed investments. The price-induced contraction of domestic demand (mainly for traded goods) provided room for export expansion, which was stimulated by a strong dose of price and nonprice incentives, as well as real wage cuts. The initial stage benefited from large-scale external financial assistance and debt relief, which eased the trade- and budget-correction problems, especially under the difficult circumstances of the 1979-80 oil shock. SEE price hikes resulted in larger operating surpluses, reducing dependence on central bank financing. The resources of the central bank could be channeled to the private sector on a larger scale through the banking system. After attaining a rise in exports, policy was increasingly turned to the liberalization of the Turkish economy. The liberalization reforms required a further depreciation of the exchange rate to compensate reductions in quantitative restrictions on imports. After contributing initially to price stabilization, the adjustment in macroeconomic prices remained as the key policy tool used to sustain the liberalization effort and strengthen savings generation in the Turkish economy (through SEE price hikes to boost public savings). In turn, a heavy and continual reliance on economywide price changes prevented smooth and steady reductions in the rates of domestic inflation. The counterparts of the exchange rate depreciation, SEE price hikes, and switch to positive deposit rates were lower real wages and salaries in the urban sector and worsened domestic terms of trade for the agricultural sector, which absorbs the bulk of employment in Turkey. The distributional consequences of these relative price shifts are reviewed in the next section.

5.6 Employment, Income Distribution, and Poverty As in most semi-industrial countries, employment and distributional statistics are generally less reliable and precise than the national accounts data. The predominance of the informal components of the rural and urban sectors gives rise to notorious measurement problems. The overriding

697

TurkeyKhapter 5

concern (of richer households) with tax liability results in underreporting of nonwage incomes in household surveys, which are not undertaken on a regular basis. Furthermore, the researchers often eschew consistency checks on the overall patterns suggested by population censuses and sectoral income aggregates. While being aware of these problems, we nevertheless believe that the available statistics permit a broad intertemporal evaluation of employment and distributional characteristics, provided we maintain our focus on proportional shifts over time. The presentation of this section draws on earlier detailed studies by Celisun (1986a, 1986b). Table 5.9 provides data on sectoral employment and productivity differentials for the benchmark years 1973, 1978, and 1983. The productivity differentials have been estimated in constant 1973 prices and in current prices. The differences in constant price and current price estimates reflect relative changes in the net prices (or alternatively, domestic terms of trade) of the major activity sectors of the Turkish economy. As regards the employment picture, table 5.9 shows the predominant position of agriculture in labor absorption and the limited employment generated by manufacturing, as noted in the cross-country comparisons in chapter 1. Compared with the 1973-78 period, the creation of new nonagricultural employment was much less in 1978-83, despite a massive cut in nonagricultural real wages. It is evident that factor substitution faces structural rigidities in the short and even medium run. Labor reallocation requires time and new capital formation in structurally rigid semi-industrial economies.

Table 5.9

Sectoral Employment Levels and Productivity Differentials, 1973-83 Productivity Differentials" Employment (thousand workers)

1. Agnculture

Constant 1973 Prices

Current Prices

1973

1978

1983

1973

1978

1983

1978

1983

9.580 -

9,537 -

9.45 I -

1.0

1.0

1.0

I .o

I .o

I60 1,419 456 544 682 1.417 -

217 1,610 562 646 1,083 1.594 -

226 1,685 586 696 1,204 1,720 -

5.9 4.2 4.0 8.0 6.1 4.8

7.3 3.9 3.5 1.7 3.9 5.0

4.7 4.4 3.3 7.9 3.9 4.3

6.0 4.6 3.6 8.1 3.9 5.0

10.2 7. I 3.5 12.9 3.3 6.3

4,678 -

5,712 -

6,126 -

5.1

4.7

4.6

4.9

6.5

14,258

15,249

15.577

2. Nonagriculture a. b c. d. e. f.

Mining and energy Manofacturing Construction Trade Public services Other services

g. Subtotal (a to f ) 3 Total Source: Celdsun (1986b)

"Productivity differentials are measured hy the ratios of per worker value added in agriculture to per worker value added in other sectors.

698

Merih Celisun and Dani Rodrik

If relative prices had remained constant between 1978 and 1983, sectoral productivity differentials would have exhibited the normally expected pattern, involving reductions in the ratio of per-worker-value-added in nonagriculture to per-worker-value-added in agriculture. Such a trend would have reduced income disparities in Turkey. However, measured in current prices, this ratio increased in a highly pronounced fashion after 1978. At the sectoral level, the observed changes in productivity differentials caused sizable income transfers from agriculture and public services to mining and energy, manufacturing, and trade sectors.’ In this context, it may be noted that the ratio of the value-added aggregate of the (wholesale and retail) trade sector to agricultural value added increased to 93 percent in 1983 from 55 percent in 1978. This was a significant distributional shift, considering the fact that trade sector employment was about 0.7 million workers as compared with 9.5 million workers absorbed in agriculture in 1983. After the review of productivity differentials and implied shifts in sectoral incomes, we may consider the proportional changes in factor shares, i.e., functional distribution of income as shown in table 5.10 for the benchmark years 1978, 1981, and 1983. Part A in this table gives the labor market aggregates, which point to the high levels of surplus labor in the Turkish economy.’ With the virtual termination of labor migration to Western Europe, Turkey’s rapid population growth-in combination with the remaining underemployment in agriculture-is likely to aggravate the unemployment problem in the future. Part B in table 5.10 provides estimates for economywide factor shares derived from the general equilibrium analysis of Celisun (1986a), which is calibrated to the official employment and wage data. Part C gives factor shares for the large manufacturing industry. For lack of relevant details, agricultural income in part B covers all income elements in this sector. The data point out the rapid reduction in the share of formal wage income (including civil servant salaries) in GDP from 33 percent in 1978 to about 21 percent in 1981, which reflects in a cumulative way the workings of the 1979-80 inflation and price adjustments.’ Part C data have a more narrow coverage, but largely confirm the falling wage shares in the post-1978 period. An earlier study by DerviS and Robinson (1980) examines the structure of income inequality in Turkey on the basis of household survey data for 1973. It concludes that the rural-urban household income differential is unusually large in Turkey and a major source of overall inequality in the size distribution of income. Taking this study as a point of departure, the analysis in Celhun (1986b) provides synthetic estimates for inequality and poverty measures for the benchmark years 1973, 1977, and 1983 on the basis of all available national income, employment, and distributional data in Turkey. The estimated Gini coefficients for 1973, 1978, and 1983 are 0.515, 0.509,

699

TurkeyIChapter5

Table 5.10

Labor Market and Functional Distribution of Income, 1978-81

A. Labor market (thousand workers)" I. Employment 2. Labor supply 3. Surplus labor B. Functional distribution of nominal GDP (f.c.), % 1 . Agricultural income 2. Nonagricultural income a. Formal wage laborb b. Nonwage labor c. Gross profits 3. Total ( = GDP) C. Factor shares in large manufacturing. % I . Private a. Nonprimary inputs b. Wages c. Gross profits d. Total ( = gross output) 2. Public a. Nonprimary inputs b. Wages c. Gross profits d. Total ( = gross output)

1978

1981

1983

15.250 16,640 I .390

15,370 17,620 2.250

15.580 17.775 2.195

25.3

22.5

19.7

33.2 6.1 34.8 100.0

21.4 4.6 51.5 100.0

20.8 4.3 55.2 100.0

61.7 12.7 25.6 100.0

66.0 9.6 24.4 100.0

66.9 8.7 24.4 100.0

58.9 19.9 22. I 100.0

68.7 12.9 18.4 100.0

64.5 8.7 26.8 100.0

Source: C e l h n (1986a) for parts A and B. and TUSIAD (1986) for part C Note: f.c.: factor cost.

"Rounded figures. hIncludes government employee salaries.

and 0.522, respectively (see tables A.27 and A.28 in the stat. app.). The share of households under the poverty line (about 12,000 TL in 1973 prices and $1,455 in 1983 prices) is estimated as 32, 25, and 30 percent in these three benchmark years. The estimated real mean income growth rates for top decile and bottom decile are 22.7 and 22.8 percent in 1973-78, and -8.4 and - 18.6 percent in 1978-82, respectively. This review of labor and distributional indicators brings out two main points. First, the heavy external borrowing of the mid-1970s temporarily allowed an improved income distribution. Second, the mechanisms used in extricating the economy from the crisis produced by the mismanagement of the macroeconomic balances and external debt mechanisms in 1974-77, namely large shifts in economywide prices, led to a dramatic reversal in distributional trends. 5.7

Recapitulation

How do we interpret the relatively successful adjustment of the Turkish economy in the post-1980 period? It seems clear that the radical changes in the structure of relative prices and the attendant shifts in patterns of income

700

Merih Cellsun and Dani Rodrik

distribution, were the key internal mechanism in attaining inflation reduction and initiating export-led recovery. Our review pinpointed the key role of economywide price changes, which produced demand restraint, enhanced savings performance in the public sector, greater financial intermediation, and expenditure switching toward exports. While the pre-reform ( 1978-79) period had already set into motion some of these changes, especially the decline in real wages and agriculture’s terms of trade, the policies of the 1980s consolidated and accentuated them. The structural rigidity inherited from the previous inward-oriented policies served to magnify the requisite changes in relative prices needed to bring about the desired consequences. The net effect was a substantial increase in the profitability of the traded manufactures sector and a sharp reduction in labor and farmers’ incomes. On the whole, these also implied an improvement in the terms of trade of the public sector vis-h-vis the private sector, which proved the key to the reduction of real private expenditures. Since the public sector is a high saver at the margin, these relative price changes not only resulted in expenditure switching, in the conventional manner, but were also instrumental in reducing absorption. This chapter has also shown the favorable macrolevel impact of external financial assistance in the earlier years of Turkey’s post-1980 program. The rise in debt-service ratios after 1983 poses new policy problems, however, as will be discussed in later chapters.

~

Selected Aspects of Debt and Adjustment

702

6

Merih Celisun and Dani Rodrik

External Borrowing, Real Wage Flexibility, and Equilibrium Exchange Rates: A General Equilibrium Analysis

Unlike the earlier stabilization episodes, the post- 1980 adjustment program featured a greater emphasis on export promotion, trade liberalization, and maintenance of realistic exchange rates. As discussed in chapters 4 and 5, policymakers attached a high priority to the attainment of an export-led output recovery in the earlier stages of the program. In this policy setting, a strong preference was also shown for large export subsidies to neutralize the preexisting anti-export bias in the trade regime, rather than for rapid import liberalization to achieve a broad neutrality in the protection and incentive system. Thus, after the removal of quotas in mid-1981, the import regime remained only partly liberalized, with a continued reliance on licensing and prohibited imports as restrictive tools until 1984 and 1985. Although import restrictions were not fully removed in the early 1980s and export subsidies were quite substantial, the overall adjustment policy displayed a strong commitment to the maintenance of an adequately depreciated real exchange rate in an effort to reduce domestic absorption and stimulate expenditure switching toward exports. In this vein, the authorities adopted a flexible exchange rate policy, which took the form of daily adjustments (from mid-1981 on) against a currency basket. As was shown previously (in tables 4.5 and 5.4) the real exchange rate depreciated considerably after 1979, despite the large differentials observed between domestic and world inflation rates. The Turkish experience with flexible exchange rates in the early 1980s is of general comparative interest. As reviewed and emphasized by Taylor (1979, 50-55), a devaluation from an initial position of deficit is likely to yield a contractionary effect on output, especially in structurally rigid economies. The Turkish setting of the 1980s featured a rigid economic structure (i.e., low substitution elasticities), but also contained three additional relevant elements, namely (a) sizable capital inflows, (b) downward flexibility in real wages, and (c) export demand shifts connected with the special conditions in the Middle East (as noted in chap. 4). An important question is the extent to which these elements contributed to the output recovery in a context where sharp depreciations of the real exchange rate threatened recessionary consequences. We concentrate in this chapter on the respective contributions of external financing and real wage flexibility, and leave to chapter 7 further discussion of the role of Middle Eastern demand. Our framework also allows us to estimate equilibrium exchange

703

Turkey/Chapter 6

rates and income levels that would have prevailed under a more rapid process of trade liberalization in the early 1980s. We present here the principal findings of a numerically based general equilibrium analysis of the Turkish economy in the early 1980s. The analysis summarized in this chapter has been conducted within the framework of a dynamic multisector general equilibrium model calibrated to the data observed from 1978 to 1983.' On the basis of a set of counterfactual experiments, we seek to examine what might have happened if Turkey had (a) pursued a more rapid trade liberalization, (b) received less capital inflow, and (c) experienced greater rigidity of real wages in the urban sector. The experiments suggest that a swifter trade liberalization under reduced borrowing would have produced an adverse effect on the level of economic activity, which would have been magnified if urban wages remained rigid in real terms. In the remainder of this chapter, section 6.1 presents the basic features of the model used. In section 6.2 the counterfactual experiments are defined and their results interpreted. This section also provides a crude estimate for the social productivity of external borrowing in the early and mid-1980s. In section 6.3 we recapitulate the main findings and policy lessons.

6.1 The SIMLOG-1 Model SIMLOG-1 is a dynamic, multisectoral, and computable general equilibrium (CGE) model of the Turkish economy, which has been calibrated for the 1978-83 period. The model closely reproduces the actual changes in relative prices, major sectoral variables, and macroeconomic balances observed in the 1978-81 and 1981-83 subperiods. The present model differs in two major ways from the earlier multiperiod CGE models built by DerviS and Robinson (1978) and Lewis and Urata (1983) in the context of the World Bank evaluations of the Turkish economy. First, the national accounts and public finances in the present model are structured around the system of accounts actually used by the SPO, which explicitly identifies private-public disposable incomes and savings as analyzed in chapters 5 and 8. Second, the nonlinear equation system of the new model is solved by the computational procedure designed by Yaprak (1982), which solves, in the Johansen (1960) tradition, for the log changes (or growth rates) of endogenous variables in a given time period, and updates their level values sequentially for new solutions in the subsequent periods of the simulation horizon. Hence the name SIMLOG: Simulation with an Interindustry Model based on Log-change variables. The full data base, functional forms, variables listings, and detailed Base Run results of SIMLOG-1 are presented in CelAsun (1986a), which also provides a discussion of its historical antecedents and relation to other applied general equilibrium models. Referring the reader to that study for

704

Merih Celasun and Dani Rodrik

technical details (and further sources and references), the distinguishing features of the model are briefly noted in the remainder of this section. 6.1.1

Input-Output Core, Trade Flows, and Primary Factor Supplies

SIMLOG-1 is built around a four-sector (agriculture, energy, manufacturing, and services) input-output (1-0)core based on an estimated 1-0table for the year 1978. The energy sector includes all commercial primary and secondary energy production, including coal, crude oil, petroleum refining, and electricity and gas production. The manufacturing sector excludes oil refining, but includes the relatively minor subsector of nonenergy mining. The combined employment and value-added figures for energy and mining correspond, therefore, to those of the industry sector defined in the national accounts. The model is designed in such a way as to analyze the import demands of the energy sector quite explicitly in the wake of the second oil price shock of 1979-80. It may be recalled that the ratio of oil imports to merchandise exports, in value terms, was 61, 82, and 64 percent in the benchmark years 1978, 1981, and 1983, respectively. In SIMLOG-1, imports are first grouped into oil imports and nonoil imports. The imported oil (the bulk of which is crude oil) is treated as a complementary input in aggregate energy production. In turn, nonoil imports are classified by sector of origin in their respective product categories as (in the conventional terminology of 1-0 literature) competitive imports. Merchandise exports are also classified by sector of origin in producers’ prices, with an explicit treatment of trade margins, which rose substantially after 1980. In the absence of precisely classified data on services trade, exports of (nonfactor) services are estimated net of imports in the services sector. The model distinguishes four types of labor: (1) agricultural, ( 2 ) nonagricultural wage, (3) nonagricultural nonwage, and (4) government employees with civil service status (appearing only in services). The nonagricultural (N-A) wage labor includes all workers in formal employee status (excluding government employees). N-A nonwage labor represents all informal workers, covering self-employed, small-enterprise, and family workers. In presenting the employment details of SIMLOG- 1, the rural and urban categories loosely correspond to agriculture and nonagriculture respectively, at the cost of ignoring the seasonally shifting status of marginal workers. The total labor supply is exogenously fixed, but its rural-urban composition is endogenously adjusted over time as a function of changing income differentials. For purposes of employment analysis, N-A nonwage laborers are treated as having imputed factor prices (wages) in the labor market, roughly reflecting minimum wage scales in the economy. The model defines a wage index (1978 = 1.O) for each N-A labor type, but allows variations in the movements of these wage indices over time for the analysis of

705

TurkeyIChapter 6

substitution between labor types. In the historical simulation, sectoral wage differentials for N-A labor types 2 and 3 are maintained at their base-year proportions. Besides labor, capital is also treated as a primary factor. Sectoral capital stocks are updated over time by taking into account the depreciation factor and the new capacity creation effects of fixed investments featuring one-year gestation lags. The capital composition matrix remains constant over time. 6.1.2 Production and Factor Demand In SIMLOG-1, producers are assumed to be price-takers in factor and products markets, and aim to equate factor prices with their marginal value products. In the model formulation, the treatment of production functions varies by sector, depending on the scope for input substitution. In agriculture, a constant elasticity of substitution (CES) aggregation of labor and capital is allowed for value-added formation, while fixed 1-0 coefficients are used to derive demands for intermediate inputs. The subsidies on manufacturing inputs into agriculture are explicitly incorporated. The (imputed) wage rate of agricultural labor adjusts to balance labor demand with endogenously determined rural labor supply. In the energy sector, which has only one type of labor, input substitution is treated through a three-level CES aggregation. First, capital and labor are aggregated to form a composite primary input. Second, the imported oil and intrasectoral energy use are combined to form a composite intermediate energy input. Third, a limited degree of substitution is allowed between these composite forms of intermediate energy and primary inputs to capture the workings of the policy emphasis on hydroelectricity investments, which aim to reduce the share of fuels in Turkey’s energy balances. The nonenergy intermediate inputs in this sector have fixed coefficients. The manufacturing and services sectors also feature a three-level CES analysis of input substitution in a slightly altered manner. In these two sectors, N-A labor types have a CES aggregation. At the remaining two levels, capital and labor are combined to form a composite primary input, which is then aggregated with intermediate inputs, all in a price-responsive fashion. For the analysis of urban employment conditions, the model can be operated under two variant modes, either with fixed or flexible real wages for N-A laborers (excluding government employees, who are always treated exogenously). The real (gross) wages are translated into nominal (gross) wages through an endogenously computed consumer price index. As regards technical progress in production, the model adopts the Hicks-neutral form of total factor productivity growth (TFPG) in the CES aggregation of capital and labor. The TFPG in manufacturing is partly related to changes in quantity rationing of nonoil imports. The observed changes in rates of capacity utilization are also captured by movements in TFPG.

706

Merih Cellsun and Dani Rodrik

6.1.3

Product Markets and Final Demand

The model treats the world prices of imported goods as exogenous data. On the export side, export volumes are sensitive to differences between Turkey’s own export prices and the world prices of other country’s exports, measured in dollars. The dollar prices of Turkey’s exports are endogenously determined on the basis of domestic production costs, trade margins, the exchange rate, and export subsidies which have fiscal and nonfiscal components. The export demand functions shift over time, reflecting real growth of world (or regional) trade in the relevant product categories. Nonoil imports (classified by sector of origin) and domestically produced goods are considered to be imperfect substitutes. For the base year (19781, the model specifies the levels of desired (nonoil) imports for each sector on the basis of the import restrictions observed in that period. From the base year onward, the ratio of desired imports to domestic output available for the home market is allowed to change in response to relative price movements within the framework of CES-type trade aggregation functions. The domestic demand flows are therefore valued in terms of composite good (or, simply, sectoral) prices. Under a restrictive trade regime, the total dollar value of desired imports may exceed foreign exchange supply available for their finance. As extensively discussed in DerviS, de Melo, and Robinson (1982, 288-316), quantity and/or premium rationing schemes may be specified to allocate available foreign exchange among competing sectors. Considering the heavy use of prohibited imports (mainly consumer goods) and user-specific licensing procedures in the simulation period, SIMLOG- 1 adopts the quantity rationing scheme and determines the quantity rationing factor, RIMP, as the ratio of available foreign exchange to total desired (nonoil) imports. For domestic final demand, the model identifies private consumption, public consumption, fixed investment, and inventory changes. The behavior of the average consumer is based on the linear expenditure system. The sectoral proportions of other final demand items are in the main exogenously determined. 6.1.4 Distribution and Disposition of Income The model identifies five groups of income: ( I ) agricultural, (2) N-A wage and salary, (3) N-A nonwage labor, (4) N-A profit, and (5) workers’ remittances from abroad. The gross incomes of the first four groups add to the GDP, and thus constitute the elements of the functional distribution of income. In turn, the model defines public and private sectors as two distinct institutional entities. In line with the official planning practice, public and private disposable incomes (in nominal prices) are determined after

707

TurkeyKhapter 6

distributing the tax burden and transfer payments over the relevant activities and income groups. The accounting system also tracks factor income flows from (and to) the rest of the world. For purposes of income flow analysis, Cellsun (1986a) provides a social accounting matrix for the base year 1978. For the public sector, public consumption is exogenous and public savings is determined residually. For the savings-generation process, the model introduces a forced-savings item corresponding to new money creation, the burden of which is distributed over the income groups proportionally. Forced savings is used to finance the public deficit and the increases in foreign exchange reserves, with the remaining part channeled to the private sector for investment financing. The voluntary savings of income groups are determined as functions of their disposable incomes and an (exogenous) real interest rate defined in the form suggested by Taylor (1979, 223-27). The direct taxes paid by various groups are also determined as functions of disposable incomes.

6.1.5 Price Normalization and Exchange Rate In the model, prices are normalized around an aggregate price index (GDP deflator). In multiperiod simulations, this index is exogenously fixed in two variant forms. In one variant, it is updated over time to reflect domestic inflation rates, taking also into account the absolute changes in all other exogenous prices. In the other variant, which is computationally more efficient, the aggregate price index is maintained constant from a particular benchmark year onward. In this second variant, all nominal variables and exogenous prices are measured in their price-level-deflated (PLD) forms. In this context, the dollar-valued balance-of-payments data (and world prices) also need to be deflated by a suitably chosen world price deflator (e.g., the world price of manufactured imports). In the applications, the nominal, real, and equilibrium concepts of the exchange rate need to be distinguished. The nominal exchange rate is the parity that converts U.S. dollars into domestic currency (TL). The real exchange rate refers to the purchasing-power-parity, price-level-deflated (PPP-PLD) version of the exchange rate, which is obtained by adjusting the nominal exchange rate for the differential between the domestic and world inflation rates. In turn, the equilibrium (or flexible) exchange rate is the endogenously determined parity given a set of policy instruments and (exogenous) balance-of-payments data.

6.1.6 Closure Rules for the External and Internal Balance To obtain solutions for the model, macrolevel closure rules need to be specified for the external and internal balance. The external balance corresponds to the current deficit, which is determined by adjusting the merchandise trade deficit for invisible flows in the current account of the balance of

708

Merih Cellsun and Dani Rodrik

payments. The current deficit is financed by external capital inflow and changes in reserves. In the simulations presented below, external capital inflows, reserve changes, and thus the current deficit are exogenously specified. Under a given system of import tariffs and export subsidies, the external balance is then achieved by determining one of the following two variables endogenously: the exchange rate or the RIMP for nonoil imports. For the savings-investment balance, we adopt the closure rule which treats private fixed investment as the main adjusting variable. In this particular version of the model, public fixed investment and foreign savings (current deficit) are exogenously specified. Inventory changes are determined on the basis of stock/output ratios. Private fixed investment therefore adjusts to the available savings in the economy. 6.1.7

Model Calibration and Validation, 1978-83

To establish a reliable numerical basis for its functional use, a model of the type outlined above should be simulated and tested over a given historical period in order to “validate” its capability to track the interdependent movements actually observed in the economy. As described in Celbun (1986a), the historical calibration and testing of the model over the turbulent period of 1978-83 was a cumbersome research effort, especially in view of the frequently revised ex post official data for key macro economic and public finance variables. Table 6.1 presents actual data and model estimates for selected macroeconomic variables for the benchmark years 1978, 1981, and 1983 in the Historical Base Run.* The actual and model-estimated labor market data are shown in table 6.2.3 The external closure rule for the Historical Base Run treats the RIMP for nonoil imports as the adjustment mechanism, while fixing the nominal exchange rate at its observed (official) annual average values over the 1978-83 period. A comparative review of the data shown in tables 6.1 and 6.2 indicates that the model closely replicates the actual observed values in this period. As shown in table 6.1, the model’s estimate for the RIMP is less than 100 percent in 1983, which implies that a portion (about 19.3 percent) of desired nonoil imports was repressed by quantitative restrictions (QRs). The Turkish economy witnessed a substantial but not complete import liberalization during 1980-83. After the elimination of quotas in 1981, the import regime continued to rely on QRs in the form of prohibited imports, licensing, and various approval mechanisms to limit actual imports to foreign exchange availability. The more comprehensive import liberalization measures taken in 1984 and 1985 also indicate that desired demand for nonoil imports was not fully (100 percent) met in 1983. The order of magnitude of the 1983 RIMP value appears therefore reasonable for purposes of model experiments, although it cannot be checked against a precisely observed figure.

TurkeyKhapter 6

709

Actual and Model Estimates for Value Added and Disposition of Income, 1978-83 (Historical Base Run)

Table 6.1

1978 Model A. Index of real value added (I978 = 100) I . Agriculture (f.c.1 2. Nonagriculture (f.c.) 3. ( I ) (2) = GDP (f.c.) B. Nominal value added (billion TL in current prices) I . Agriculture (f.c.) 2. Nonagriculture (f.c.) 3. ( I ) (2) = GDP (f.c.) 4. GNP (market prices) C. Disposable incomes (as % of nominal GNP) 1. Public 2. Private D. Savings-investment balance (as % of nominal GNP) 1. Public savings 2. Private savings 3. Current deficit' 4. (1) + (2) + (3) 5. Public fixed investment' 6. Private fixed investment 7. Public stock changes 8. Private stock changes 9. (5) + (6) + (7) + (8) E. Current accountb (million $ in current prices) 1 . Merchandise exports 2. Merchandise imports a. Oil b. Nonoil c. (a) + (b) 3. Exports of the service sector, net 4. Factor income from abroad, net" 5. Current deficit' F. Quantity rationing factor (RIMP) for nonoil imports G. Foreign terms of trade (I978 = 100)

+

+

1981 Model

103.0 103.1 103.1

100.0 100.0 100.0

30 1 778 1,188 1,285

1,348 4,655 6,033 6,483

Actual

Model

104.6 102.5 103.0

108.7 112.5 111.6

2,071 8,458 10,529 11,181

1,325 4,735 6,060 6,553

Actual

111.0 111.6 111.5

2,058 8,727 10,785 1 1.485

18.9 81.1

19.3 80.7

19.3 80.7

18.6 81.4

17.4 82.6

6.3 12.7 2.8 21.8 10.5 9.8

8.5 11.9 3.4 23.8 11.7 7.4 2.8 1.8 23.7

8.6 9.4 3.5 21.5 11.7 7.4 1.5

7.6 10.3 4.3 22.2 11.3 8.2 1.3 1.3 22. I

7.3 9.2 4.1 20.6 11.4 7.5 0. I 1.6 20.6

1

.o

0.5 21.8

1.1

21.5

2,288

4,709

4,703

- 1.396 - 3,203

- 3,724 -5,214 - 8,938

-3,878 -5,055 - 8,933

-4,599 402

1,066

49 1

1,195

- 1,418

- 1,968

0.697

0.603 100.0

68.4

Source: CelPsun (1986a). Note:

1983

f.c.: factor cost. ma.: precisely classified actual data not available

"Exogenously specified magnitudes in the Historical Base Run. bPresentation follows the old format of balance of payments.

~

5,732

5,728

3.525

-3,661 - 5,574 -9,235

- 5,705 - 9,230

n.a

1,274

n.a.

n.a.

I01 -2,123

n.a. -2,123

- 1,968 n.a 70.6

0.807 64.4

n.a. 66.4

710

Merih CelAsun and Dani Rodrik

Table 6.2

Actual and Model Estimates for Labor Market and Real Wages, 1978-83 (Historical Base Run) 1978

A. Labor market (thousand workers) I . Labor demand a. Agriculture b. Industry (energy manufacturing) c. Services d. (b c) Nonagriculture (N-A) e. (a + d) Total 2. Total labor supply 3. Surplus labor (N-A) 6 . N-A real wage indices (1978 = 100)" 1. N-A labor 1 2. N-A labor 2 3. N-A labor 3

+

+

1981

1983

Model

Model

Actual

Model

Actual

9,537 1,897 3,886 5,713 15.250 16,640 1,390

9,501 1.897 3.990 5,887 15,388 17,607 2,219

9,512 1,822 4,034 5,856 15,368 17,621 2,253

9,475 1,943 4,198 6,141 15.616 18.282 2,666

9,45 I 1,911 4,215 6.126 15,577 17,773 2,196

100.0 100.0 100.0

65.0 67.5 51.7

66.4 n.a. 50.9

63.4 65.9 51.7

65.0 n.a. 52. I

Source: Celisun (1986a) Note: n.a.: precisely classified actual data not avdikibk. N-A labor: 1 = wage labor; 2 = nonwage labor; 3 =

government employees. "The actual real wages for N-A labor 1 and 3 are obtained by deflating the Employers Federation nominal wages and net nominal salaries of civil servants (in 7th salary grade) by the Aggregate Price Index of the Historical Run. See sources cited in Cellsun (1986a).

6.2 Counterfactual Experiments 6.2.1 Preliminaries The counterfactual experiments are structured in such a way as to bring out the major lessons from the Turkish experience with wage, trade, exchange rate, and borrowing policies in the early 1980s. We distinguish between trade-liberalizing and deficit-reducing devaluations (following Krueger 1981), and carry out a number of counterfactual experiments designed to explore the general equilibrium effects of trade liberalization under varying targets for the payments d e f i ~ i t . ~ For purposes of counterfactual analysis, the Historical Base Run is resimulated in an altered form, which endogenously determines the exchange rate under exogenously fixed RIMP values. This restructured solution is simply referred to as the Base Run, which also treats real wage indices exogenously. With such adjustments in the model structure, the Base Run serves as a benchmark for all counterfactual experiments. For analytical convenience, the Base Run and counterfactual experiments are simulated from 1981 on under the constant 1981 level of the aggregate price index. As noted in section 6.1, this particular mode of price normalization requires the use of PLD domestic and external data. This

711

TurkeyKhapter 6

approach provides a basis on which to determine and compare equilibrium exchange rates in real (PPP-PLD) terms, excluding the effects of domesticworld inflation differentials. All counterfactuals are thus solved for the 1981-83 period, proceeding from the same 1981 solution obtained in the Base Run. 6.2.2 Description of Experiments To isolate and examine the effects of wage policy on equilibrium exchange rates, the counterfactual experiments are designed in two groups, A and B. In group A experiments, N-A real wages are exogenously specified, as in the Base Run, and N-A employment is endogenously determined. In group B, the model exogenously fixes the 1983 actual figure for total N-A employment as a target and flexibly determines the real N-A wages (for labor types 1 and 2, as described in sec. 6.1.1).5 To avoid further complications, the ratio of real wage indices for N-A laborers 1 and 2 is maintained constant. All experiments in groups A and B are simulated with equilibrium (or flexible) exchange rate specification. Group A includes the following experiments:

+

+

E-1A: Base Run fixed N-A wages additional trade liberalization involving a 10 percent increase in RIMP and 50 percent decrease in fiscal subsidy on manufactured exports.

+

E-2A: Base Run + fixed N-A wages 50 percent reduction in the 1983 value of the current deficit (which was about $2.1 billion in current prices) + 10 percent fall in foreign interest payments in 1983. E-3A: E-1A

+ E-2A

In group B, experiments E-lB, E-2B, and E-3B are the same as experiments E-lA, E-2A, and E-3A, respectively, except that they feature flexible N-A real wages as noted above. In each group, then, the first experiment investigates the consequences of enhanced trade liberalization, the second of reduced external borrowing, and the third of the two combined. Group A assumes fixed urban real wages, while B allows them to vary. The exchange rate equilibrates external accounts in all cases. 6.2.3 Economywide Results Table 6.3 presents a summary of the economywide results of the counterfactual experiments for the year 1983 in terms of ratios to the Base Run estimates, which closely replicate the actual 1983 data as shown previously in tables 6.1 and 6.2. Part A in table 6.3 lists the distinguishing characteristics of the experiments systematically. Under the fixed real wage regime, a trade-liberalizing devaluation (E- 1A) has a mild contractionary effect on N-A employment, with practically no

712

Merih Cellsun and Dani Rodrik

Table 6.3

Counterfactual Experiments: Basic Features and Some Economywide Estimates, 1983 (ratios to Base Run values for 1983)" Fixed N-A Real Wages

A. Counterfactual policy restrictions I . Current deficit (nominal $) 2. RIMP for nonoil imports 3. Subsidies on manufactured exports 4. N-A employment 5 . Real wage index (N-A labor I ) 6. Ratio of real wage indices for N-A labor I and 2 B. Counterfactual model estimates 1. Major prices a. Equilibrium exchange rate (real, 1981 prices, TW$) b. Real wage index (N-A labor I ) c. Foreign terms of trade 2. Real GDP (f.c.) 3. Real fixed investment 4. Real private consumption 5 . N-A employment 6. Foreign trade (nominal $) a. Merchandise exports b. Exports of the service sector, net c. Oil imports d. Nonoil imports

Fixed N-A Employment

E-IA

E-2A

E-3A

E-IB

1.0 1.1

0.5 I .o

0.5

I .o

1 .o

I .o

0.5 0.5 1.0 1.1 0.5 1.0 0.5 1.0 1.0 1.0 -----Endogenous-----

I .o

1 .o

I .o

I .o

1 .o

1 .o

1.252

1.417 I .m 0.567 0.989 0.939 0.947 0.973

1.172 0.979 0.835 1.002 1.023 0.988

1.275 0.934 0.699 0.995 0.905 0.971

1.447 0.913 0.536 0.997 0.929 0.958

l.m

l.m

1.m

1.064 0.944

1.097 1.100 0.916

0.951

1.010

1.035 1.038 0.973 1.060

1.069 1.071 0.949 0.955

1.108 0.923 1.015

1.1

0.5 1 .o 0.5 -----Endogenous---

1.165 l.m 0.843 1.ooO 1.026 0.984 0.994 1.034 1.036 0.972 1.059

l.m 0.723 0.989 0.914 0.962 0.979 1.063

E-2B

E-3B

1.0

1.1

1.104

Source: Celksun (1986a) "All counterfactual experiments take 1981 model solution as a point of departure and adopt the exogenous estimates of the Historical Base Run for 1981-83, with the exception of policy restrictions indicated in part A of the table

impact on the GDP level. The implication is that the contractionary impact of devaluation (in conjunction with reduced QRs) on import-competing sectors is offset by export-led output increases in the economy. In this experiment, the increase in the nonoil import bill (due to lowered QRs) is balanced by devaluation-induced reduction in oil imports and expansion in exports. Under flexible real wages in experiment E- l B , trade-liberalizing devaluation (at the same current deficit) is neutral in its effects on aggregate income and N-A employment, but requires a cut in real wages. A deficit-reducing devaluation in the context of an unchanged trade policy (E-2A) has a notable contractionary effect on GDP and N-A employment. To accommodate the reduced capital inflow, the devaluation requirement (in E-2Aj becomes quite large, leading to import contraction and export expansion, with particularly adverse consequences for real fixed investment and the foreign terms of trade. The flexibility of real wages in experiment E-2B reduces GDP losses, but results in lower real wages.

713

TurkeyIChapter 6

If, as in experiment E-3A, Turkey had pursued simultaneously a more rapid trade liberalization and a 50 percent smaller current deficit in 1983 (starting from the actual initial conditions in 1981 and other government policies remaining the same), this would have required a sharper real depreciation and depressed levels of private consumption and N-A employment. Again, downward flexibility in real wages (experiment E-3B) would have contributed to the maintenance of employment levels, but the required cuts in real wages would have been considerable (nearly 9 percent), especially in the aftermath of the 1979-80 erosion in wage earnings. 6.2.4 Social Marginal Productivity of External Borrowing In the medium term, trade liberalization aims at a neutral system of incentives for exports and import-substitutes, and to enhance efficiency in resource allocation. In the short run, however, a trade-liberalizing move (involving removal of nontariff barriers) in combination with a flexible exchange rate policy tends to entail some social costs in the form of reduced N-A employment or real wages, depending upon the wage policy adopted. In this context, the size of the capital inflow becomes a crucial variable affecting macroeconomic performance. It is often suggested that the social marginal productivity of external borrowing in the earlier stages of trade liberalization might be quite high “as long as the stabilization program appears to have a good chance of success,” as emphasized by Krueger (1981, 113). The counterfactual experiments reported in table 6.3 provide a quantitative basis to derive rough estimates for the social productivity of external borrowing under the observed conditions of the Turkish economy in the early and mid-1980s. It may be noted that a 50 percent reduction in the current deficit corresponds to about 2 percent of nominal GDP in 1983, which is approximately equivalent to 1.5 percent of real GDP measured in constant 1987 prices. A comparison of experiments E-1B and E-3B shows that the real GDP loss would have been about 0.5 percent if the additional trade liberalization were to be carried out in conjunction with a 50 percent lower deficit (net borrowing) target. These indicators broadly suggest that the social marginal productivity of external capital, in GDP terms, was about 33 percent (0.Y1.5). A comparison of the results of experiment E-2B with the Base Run data also reveals a similar order of magnitude. The welfare implications of external borrowing are much more complex and require a longer term horizon for an appropriate assessment. Nonetheless, the Turkish experience suggests that in the context of an adjustment program, external finance can be highly productive at the aggregate level. With real wages maintained constant, a $1.0 billion fall in external borrowing would have reduced N-A employment by 2 percent, implying a loss of the equivalent of 120,000 full-time jobs in the Turkish economy.

714

Merih Celisun and Dani Rodrik

6.2.5 Sectoral Responses to Exchange Rate Adjustments

In table 6.4 we summarize data on the sector-level responses of the Turkish economy under experiments E-2A, E-3A, and E-3B, which all involve deficit reduction under a flexible exchange rate policy. As revealed by the macroeconomic indicators of these experiments, deficit-reducing devaluations generate a deflationary impact on real GDP, which implies that depressed domestic expenditures are not fully offset by export expansion in real terms. A comparison of sectoral employment and output indicators for experiments E-2A and E-3A shows the resource-pulls originating from a Table 6.4

Counterfactual Experiments: Sector-level Adjustments to Reduced External Borrowing (ratios to Base Run values for 1983)O Experimentsb E-2A

A. Employment Agriculture Energy Manufacturing Services' B. Gross output Agriculture Energy Manufacturing Services C. Desired competitive imports Agriculture Energy (nonoil) Manufacturing D. Competitive imports (balanced) Agriculture Energy (nonoil) Manufacturing E. Complementary (oil) imports F. Exports Agriculture Energy (exogenous) Manufacturing Services (net of imports) G. Consumer demand Agriculture Energy Manufacturing Services

E-3A

E-3B

1 0.972 0.992 0.963

1.001 0.963 0.953 0.964

1.010 0.991 0.990 I .007

0.999 0.958 0.995 0.985

1 0.937 0.991 0.987

1.006 0.944 1.002 0.995

0.879 0.927 0.946

0.798 0.888 0.906

0.795 0.893 0.91 I

0.884 0.907 0.953 0.945

0.923 0.993 1.014 0.918

0.921 0.997 1.019 0.924

1.190 1.263 1.257

1.313 1 1.402 1.401

1.332 1 .Ooo 1.432 1.435

0.983 0.894 0.942 0.970

0.976 0.842 0.931 0.948

.ooo

1.CGU

.ooo

.ooo

0.979

0.858 0.944 0.963

Snurce: Celhun (1986a).

"See footnote a in table 6.3. All the variables indicated in this table are measured in real terms. %ese particular experiments feature 50 percent reduction in the current deficit (of $2,123 million) for the year 1983, which was about 4 percent of GNP in cument prices. 'Excluding government employees

715

TurkeyiChapter 6

trade-liberalizing adjustment in the exchange rate at the reduced level of current deficit. Under fixed urban real wages, agriculture and services tend to respond favorably in output and employment spheres to reduced QRs in the imports regime, while energy and manufacturing display downward trends. With flexible urban real wages, trade-liberalizing devaluation (E-3B) stimulates agriculture and services more vigorously, while also providing a mild output expansion in manufacturing (as contrasted with the results of E-3A). Consumer demand shifts in experiments E-2A and E-3A (both under fixed urban real wages) strongly underline the workings of relative price effects on the level and composition of real private consumption expenditure, which adjusts sharply to allow expenditure switching in favor of exports. It is evident that large depreciations concurrently serving deficit-reduction and trade-liberalization objectives have substantial short-term social costs in terms of reduced real consumption levels.

6.3 What Have We Learned? In the present chapter we have summarized the results of a general equilibrium analysis of the Turkish economy in the early 1980s. The study eschewed a formal consideration of money and inflationary dynamics. The focus of the analysis has been on the implications of alternative policies for trade liberalization, external borrowing, and urban wage settlement under a flexible exchange rate regime. Although the time frame of our counterfactual experiments extended over the relatively short period of 1981-83, the results nonetheless provide clues to the short- and medium-term impact of policy initiatives which were intensely debated in the wake of the 1980 adjustment program. The counterfactual analysis has shown that the Turkish recovery effort strongly benefited from external financial assistance extended in the post- 1980 period. In GDP terms, the social marginal productivity of external borrowing was around 33 percent in the early and mid-1980s. The counterfactual experiments have also revealed that a more rapid trade liberalization, under the actual levels of net foreign borrowing, would have generated a moderate fall either in the level of nonagricultural employment or in real wages, depending upon the wage policy adopted for the urban sector. Thus, a more speedy transition to an open trade regime would have entailed additional social costs in the early 1980s. The Turkish policymakers showed prudence in choosing a more gradualist approach in import liberalization and relying on export subsidies to neutralize the anti-export bias in the trade regime.

716

7

Merih Cellsun and Dani Rodrik

Trade Regime and an Anatomy of Export Performance

It would be possible to recast the Turkish experience with external debt in terms of a narrative involving trade flows exclusively. In this view, the rapid accumulation of debt in 1973-77 was the consequence of the rise in imports while exports stagnated. In 1978-79 the economy went into a tailspin as imports collapsed. The recovery after 1980 came alongside a phenomenal increase in exports, which allowed a revival in imports. To be sure, a perspective focusing on trade alone would be seriously misleading. The accumulation of external debt and its servicing are both clearly macroeconomic phenomena. These two are fundamentally linked to the relation between aggregate expenditures and national income. As such, the various microeconomic measures constituting a country’s trade regime play a somewhat secondary role. The trade regime can be of primary importance in determining the openness of an economy, as revealed for example by the ratios of exports and imports to national income. It can also clearly influence the efficiency with which resources are utilized. But in principle a particular net trade position can be achieved with any trade regime in place, no matter how restrictive that regime is. Consequently, explanations of debt crisis or successful adjustment in terms of the properties of a particular trade regime have to be held suspect at first sight. Still, important linkages exist between trade regimes and the quality of macroeconomic management. There are two important respects in which more open trade regimes tend to facilitate such management. First, the typical import-substitution regime shuts out all imports except for capital goods and intermediates. When macroeconomic adjustment calls for some import compression, the negative effects on domestic output and investment become hard to avoid. Since imports of consumption goods are a very small part of the total, they cannot provide a margin of safety. Secondly, for any given debt/GNP ratio, a higher share of exports in national income means a larger base of foreign exchange earnings with which to cushion external shocks and service debt. A country twice as open as another can generate twice as much foreign exchange through a proportionately identical increase in exports. Since commercial banks understand this, they watch the absolute level of exports closely. In Turkey’s case both factors have come into play at different junctures. During the 1970s, as in the preceding decade, the Turkish trade regime was characterized by typical import-substitution policies. As we have discussed in chapters 4 and 5, the 1980s have seen the dismantling of these policies as well as an export boom that has baffled all but the most optimistic observers. The present chapter provides an overview of these changes in the trade

717

TurkeyKhapter 7

regime and an analysis of their connection with the management of external debt. In the spirit of the remarks above, we will downplay the direct effects of the trade regime, and changes therein, in precipitating the debt crisis of 1977 and in extricating the economy from the crisis after 1980. Our main focus will be on the sources of the export boom, which has been the single most important success of the post-1980 policies as well as one of the key factors alleviating the burden of the debt crisis. We shall argue that the post-1980 export performance, remarkable as it has been, has had some disturbing features. First, a non-negligible share of the increase in exports after 1980 turns out to have been the result of a statistical fiction: to take advantage of generous export subsidies, domestic entrepreneurs appear to have changed their invoicing practices from mild underinvoicing to substantial overinvoicing. Secondly, an important stimulating role has been played by the Iran-Iraq war, which has created almost overnight a booming market for Turkish manufactured exports to both sides. The relative role of the most important policy variable, the exchange rate, appears to have been disappointingly small, despite vast amounts of real depreciation. Finally, the depressed state of private investment throughout the first.half of the 1980s suggests that very little export-oriented structural change has in fact taken place, with the bulk of exports coming from increased capacity utilization.

7.1 Import Regime

A complete description and evaluation of Turkey’s trade regime during the 1970s and 1980s are beyond the scope of the present work.’ Here, we will simply summarize some of the salient features and discuss briefly the linkages with the macroeconomy and external debt. Until the policy reforms of the 1980s, the predominant trade strategy in Turkey was that of import-substituting industrialization (ISI). The trade regime in place throughout the 1970s exhibited all the familiar characteristics of ISI: high rates of trade protection, biased against consumer durables; a wide variation of effective rates of protection across sectors; widespread use of QRs; selectivity and discretion in import-licensing policies; and reliance on administrative allocation of foreign exchange. These policies had the usual costs and distortions associated with them. Collectively, they biased production incentives away from exports and toward the domestic market. The domestic resource costs of the manufacturing industries thus sheltered were typically several times higher than the corresponding costs in export-oriented sectors.* Perhaps more serious than the allocative inefficiencies was the widespread rent-seeking encouraged by the import regime. Import licenses garnered healthy premia, and a sizable portion of Turkish GNP was devoted to capturing them. It is no coincidence that Krueger’s

718

Merih Celasun and Dani Rodrik

(1974b) classic article on rent-seeking drew heavily from her experience in Turkey. Given the predominant pattern of import rationing and foreign exchange licensing, the direct role of the official exchange rate in determining the level of imports was rather limited throughout the 1970s. Hence, econometric work with import demand equations typically discovers a low import demand elasticity with respect to the official exchange rate. Table 7.1 presents the results for an import equation estimated on quarterly data from 1970:II to 1983:IV. The explanatory variables are the trade-weighted real exchange rate, the black-market premium on the dollar (both lagged), the real domestic credit stock, a time trend, and seasonal dummies. The real exchange rate term has the expected negative sign, but the coefficient ( - 0.33) is low and falls short of statistical significance. The black-market premium, by contrast, is statistically significant at the 99 percent confidence level. The results suggest that a 10 percent increase in the black-market premium of the dollar reduces import demand by 7.8 percent. This provides strong evidence that the marginal cost of foreign exchange to importers was the black-market rate rather than the official rate. An important implication of this is that exchange rate devaluations which brought the official rate in line with the black-market one would not only fail to reduce imports, but could actually increase them.

Table 7.1

Determinants of Import Volume

From 197011 until 1983:IV 55 ,75627522 ,92401961 2.30905026

Observations R2

SSR Durbin-Watson

Degrees of Freedom R2

SEE

47 .71997579 .I4021409

No.

Label

Lag

Coefficient

Standard Error

t-Statistic

1

Constant LRXTWPIW BMXRPR LRCRED TREND ONE TWO FOUR RHO

0 1 1 0 0

4.9375060 - .3305008 -.7841130 ,3267 180 ,59827618-02 .8587800E-O1 ,1492214 ,6235854E-01 ,6523425

1,3271870 ,3433583 ,2607478 .3174550 ,576843OE-02 .4672353E-01 .4078458E-01 .4188279E-01 ,1307582

3.720278 - ,962553 -3.007 170 1.029179 1,037156 1.838003 3.658769 1.488882 4.988922

2 3 4

5 6 7 8 9

0 0

0 0

Note: Dependent variable is LMVOL. Corrected for first-order serial correlation. Key: LMVOL Volume of imports (in logs). LRXTWPIW: Import-weighted real exchange rate index (in logs). BMXRPR: Black-market premium of the dollar (percent). LRCRED Domestic credit deflated by wholesale price index (in logs). TREND: Time trend. ONE, TWO, FOUR: Quarterly dummies for first, second, and fourth quarters, respectively.

719

TurkeyKhapter 7

There can be little doubt than the IS1 policies of the period repressed exports-as a share of national output-below where they would have been in the absence of those measures. They also virtually eliminated imports of consumer goods, reducing such imports to less than 5 percent of all imports. Hence, in both respects they rendered a strategy of investment via foreign borrowing (as in the 1973-77 period) a risky one. To see this, it is sufficient to pose the following hypothetical question: How large would the increase in exports need to have been, assuming everything else was constant, in order to halve the current account deficit of 1977? Since the current account deficit then stood at 7 percent (of GNP) and exports around 5 percent (of GNP), the requisite increase in exports was no less than 70 percent. Had the export share been double its actual level, the corresponding export growth rate would have had to be only 35 percent. Therefore, the low levels of openness fostered by the IS1 strategy did complicate the adjustment process once foreign inflows stopped. Nonetheless, the direct links between IS1 and the onset of the debt crisis are extremely tenuous. It would be difficult to sustain any line of argument that gave a role to IS1 beyond the sort of calculations presented in the preceding paragraph. The IS1 regime per se cannot explain why the current account deficit, and hence the aggregate relationship between income and expenditures, got progressively out of line after 1973. Certainly there was no increase in the restrictiveness of the trade regime after 1973. If anything, there was some liberalization as the authorities made use of the freedom allowed by the ready availability of foreign funds. The problem in this period was not a particular level of imports or exports, but the worsening relation between the two, which in turn reflected the underlying macroeconomic balance (as discussed in chap. 2). Since 1980, as we discussed in chapter 4, a substantial amount of trade liberalization has been undertaken in line with the prevailing economic philosophy of the period. While small adjustments were made in the trade regime from January 1980 onward, the major break with the past came with the announcement of the 1984 import regime in December 1983. Under the new regime, commodities which had their importation banned or subject to license were explicitly listed, in sharp contrast to the previous system under which all commodities not specifically mentioned were effectively prohibited. This transition from the “positive list” to the “negative list” translated into a quantum jump in trade liberalization, as it became possible to import all products not specifically listed. In addition, large-scale elimination of QRs took place, affecting close to half of all imports (World Bank 1984, 26). While there was an overall downward movement in tariff rates as well, the government imposed specific duties on a wide range of commodities, especially consumer goods which were affected substantially by the elimination of QRs. The resulting import regime was one which was less protective overall, with an average tariff rate (including specific levies) of 30

720

Merih Cellsun and Dani Rodrik

percent (see table A.26 in the stat. app.), and in which price measures played a much greater role than quotas. Once again, it is fruitful to speculate about the effect of these trade reforms on the relatively successful macroeconomic adjustment after 1980. There can be little doubt that the rationalization of the import regime which took place after 1983 and the elimination of QRs in particular, have been salutary for the long-run health of the Turkish economy. But no fancy causality tests need be run to determine that trade liberalization could not have played an important role in the recovery after 1980. The adjustments in the trade regime prior to 1984 were by and large small. As pointed out above, the major reforms in the trade regime came into effect in 1984, well after the recovery. It is perhaps more appropriate to regard the improvement in the macroeconomic context as the enabling cause of trade liberalization, as opposed to the other way around. 7.2

Export Performance

In 1979 the share of merchandise exports in GNP stood at the meager level of 3.2 percent. By 1985 exports had risen to 14.9 percent of GNP. Figure 7.1 displays the phenomenal rise in Turkish exports after 1980. In dollar terms, exports nearly quadrupled, going up from $2.3 billion in 1979 to $8.0 billion in 1985. As figure 7.1 shows, a substantial portion of the increase has resulted from exports to non-OECD countries, principally the Middle Eastern markets. Exports to OECD countries have risen less

6000 w

-

'5

4000-

OECD /

I 0 1975 76

I

77

I 78

I 79

Fig. 7.1 Export performance, 1975-85

I

I

I

I

I

80

81

82

83

84

85

721

TurkeyKhapter 7

spectacularly than total exports, but have more than doubled in the same period. The latter is an important feat considering the generalized slowdown in industrial countries in the wake of the second oil shock. The significance of the export boom to the relative success of the Turkish adjustment experience after 1980 cannot be belittled. First and foremost, the expansion of exports has allowed a commensurate increase in imports and hence overall economic expansion, without creating undue strains on the current account. Imports have risen from 7.2 percent (of GNP) during the recession of 1979 to 21.7 percent in 1985. This is the critical factor distinguishing the Turkish adjustment experience from the typical pattern of import compression in other heavily indebted developing countries after 1982. Secondly, the export boom has allowed Turkey to reenter international private capital markets starting in 1982. The newfound confidence in the creditworthiness of the ’hrkish economy is based principally on the success with exports, and it has waxed and waned alongside export statistics. Since the Turkish authorities understand the connection all too well, the continuation of the export drive has become a paramount consideration in policymaking. Before going into the causes of the export boom, it is worthwhile to stress two transformations that have taken place in the structure of Turkish exports, as these provide some initial clues to the mystery. The first of these is in the geographical destination of exports, which has already been mentioned in our discussion of figure 7.1. As this figure shows, OECD countries were Turkey’s main trade partners prior to 1981. The increase in exports since then, however, has come predominantly from increases in sales to non-OECD markets, and to Middle Eastern (including North African) countries in particular. In 1980 the share of the Middle East in total Turkish exports stood at 17 percent; in 1981 this number rose sharply to 40 percent, and averaged 42 percent during 1981-85. Much of the increase was due to expanded trade with Iran and Iraq, two countries involved in a prolonged war with each other. The expanded demands of their war economy, coupled with Turkey’s transport cost advantages, provided an unprecedented opportunity for Turkish manufactured exports. These two countries alone took 23 percent of Turkish exports during 1981-85, compared to 8 percent in 1980. The second transformation has taken place in the product composition of exports. The export boom has been mainly in manufactured products, as agricultural exports have stagnated. As a consequence, the respective shares of the two sectors in aggregate exports have reversed between 1980 and 1983. Industrial products, which constituted 36 percent of exports in 1980, rose to 64 percent by 1983, while the share of agriculture fell from 57 to 33 p e r ~ e n t .As ~ tables A.23 and A.25 in the statistical appendix show, the expansion of manufactured exports took place across the board, and was not limited to specific subsectors, even though textiles, clothing, iron, and steel took the lead. The resulting structure of exports is at variance with the

722

Merih Celasun and Dani Rodrik

presumption that Turkey's comparative advantage lies in agriculture, processed agricultural products, and labor-intensive light manufactures. Exports of the first two categories have been clearly undistinguished. As for the last, certain capital goods sectors (iron and steel, metal products and machinery) have managed to score impressive successes, alongside the boom in textiles and clothing. But there is less of a puzzle than it would appear at first sight. This pattern of exports is the consequence of both the shift toward the less developed countries in the Middle East and the lavish export subsidies bestowed on manufactured goods (on which more later). What were the underlying determinants of the export boom? Explanations to date have relied on a long and varying list of causal factors. These can be categorized under three headings: (1) export-oriented policies of the governments since 1980, (2) fortuitous external circumstances, and (3) presence of key prerequisites which set the stage for the boom once the factors under (1) and (2) came into play. We shall take up each cluster in turn and attempt an overall evaluation. As discussed in chapter 4, the policies followed since January 1980 were explicitly oriented toward encouraging exports. Exchange rate policy in this period was geared toward achieving a massive increase in the relative profitability of supplying foreign markets. In a sharp break with the past, the exchange rate was actively used to offset the effects of higher domestic inflation and to provide a margin of real depreciation. The consequence was an unprecedented and largely continuous depreciation of the real exchange rate. Figure 7.2 displays the pattern after 1979. By the end of 1980, a 30

l

g

0

4

Fig. 7.2 Real effective exchange rate, 1979-85 (export-weighted, using WPI, 1979 = 100)

723

TurkeyIChapter7

percent real depreciation had already taken place (relative to the 1979 average). The daily adjustments of the nominal rate after May 1981 facilitated maintaining the real rate on track, and the cumulative real depreciation between then and the end of 1985 amounted to another 30 percent. The downward trend in the real rate and the reduced instability around trend both contributed to the increased attractiveness of exporting (see next section). The activist exchange rate policy of the period was supplemented by overt export subsidies geared almost exclusively to manufacturing industries. These subsidies took three forms: (1) export tax rebates, which were ostensibly designed to compensate exporters for indirect taxes, (2) subsidized export credits, and (3) preferential allotment of foreign exchange and duty-free imports. The relative importance of these have fluctuated since 1980, with a declining trend in the overall rate of subsidization since 1984. According to Branko Milanovic’s (1986) calculations, the ad valorem equivalent of these measures averaged around 20 percent during 1980-83, with a general tendency toward favoring metal products, machinery, and transport equipment (table 7.2). As far as exporters were concerned, the subsidy scheme was equivalent to a step-devaluation of the Turkish lira by the same magnitude. But a key difference was that these subsidies inserted a wedge between the profitability of manufactures exports and the profitability of other means of earning (or saving) foreign exchange. Turkish entrepreneurs, never too shy in exploiting arbitrage opportunities, used the wedge to their advantage. As we

Table 7.2

Ex Post Export Subsidy Rates in the Manufacturing Sector, 1980-84 (in percentages)

Sector

1980

1981

1982

1983

19X4

Food and beverage Textiles Leather and furs Chemicals Rubber & plastic Glass Cement Iron & steel Nonferrous metals Metal products Nonelectrical machinery Electrical machinery Transport equipment

10.8 20.3 9.4 16.6 6.4 36.5 24.5 16.3 16.7 62. I 71.1 54.0 96.2 51.6

11.7 19.3 14.5 21.6 16.9 24.9 21.6 15.6 15.3 63.4 70.2 21.8 69.8 47.0

13.0 18.9 20.6 28.1 16.4 29.6 23.9 20.9 23.9 52.3 101.5 25.4 43.8 31.7

10.6 21.7 25.1 20.9 25.4 21.2 18.3 28.2 29.2 41.2 159.5 29.2 68.7 28.7

8.2 13.6 16.8 9.9 15.7 20.0 16.9 18.0 21.0 25.0 69.7 12.9 29.7 25.3

Manufacturing average

22.1

20.5

20.6

23.4

15.1

Paper

Source: Milanovic (1986), table

VII. 4.

Note: These rates refer to the ratio of the value of combined subsidies to export values.

724

Merih Celasun and Dani Rodrik

will discuss below, a considerable amount of overinvoicing of exportsso-called fictitious exports-resulted. Besides exchange rate policy and export subsidies, two other aspects of the overall policy environment deserve mention. First, the relative restraint in domestic demand management may have had the natural consequence of forcing domestic entrepreneurs to look for export markets. This has no doubt played a role of some importance, which has been highlighted by the experience in 1986 when booming domestic demand led to a reduction in the dollar value of exports. But its importance ought not be exaggerated. The 1978-80 period, in which the cuts in real domestic demand were the most severe, experienced an export performance quite undistinguished relative to what was about to come. Second, the process of trade liberalization may have stimulated exports through the general equilibrium channels captured by the Lerner symmetry theorem. As explained above, however, the major steps in trade liberalization came in 1984 when the export boom was well under way. Among external circumstances, the Iran-Iraq war was of key importance. The flourishing export market created by the war between Turkey’s two southeastern neighbors has already been mentioned. While it is difficult to quantify precisely the effect of the war, it is clear that the increase in Turkish exports after 1980 was due in large part to exports to Middle Eastern markets. Table 7.3 shows that this increase coincided with the expansion of import demand in Iran and Iraq, which reached a combined peak in 1981 -83. A partially offsetting external factor was the generalized slowdown in economic growth in developed countries in the wake of the second oil shock. Finally, the export boom was facilitated by the existence of certain prerequisites. Foremost among these was the availability of large amounts of underutilized industrial capacity, which could be mobilized once imported

Table 7.3

Imports of Iran and Iraq (f.o.b., in billion $) Year

Iran

Iraq

Combined

1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

12.5 14.2 17.4 7.5 10.9 11.0 9.5 16.4 13.3 10. I

4.7 4.9 5.5 9.0 12.6 18.7 19.4 11.0 10.0 9.5

17.2 19.1 22.9 16.5 23.5 29.1 28.9 27.4 23.3 19.6

Source: IMF, International Financial Statistics, line 7 1yv, for each country, converted into dollars using period-average exchange rate.

725

TurkeyIChapter 7

inputs became available after 1980 and export incentives became sufficiently strong. As discussed in chapter 2 , the 1973-77 period had experienced a sustained investment boom, generating substantial industrial capacity. The foreign exchange bottlenecks after 1977 paralyzed the manufacturing sector, resulting in sharp declines in capacity utilization rates to around 45-50 p e r ~ e n t The . ~ availability of capacity is one important reason why Turkish industry responded so vigorously to the exchange rate and subsidization policies after 1980. While this allows us to put the investment drive of the previous decade in a somewhat more favorable light, the sectoral composition of the resulting exports, as discussed above, is prima facie evidence of the efficiency costs of having misallocated investment through the overvalued exchange rate and other pricing policies of the 1970s: the pattern of exports has been biased away from the sectors in which Turkey possesses a comparative advantage. It can be concluded that the export boom of the 1980s has had a high resource cost compared to a scenario in which the pricing policies of the 1970s had been more benign.

7.3 Anatomy of an Export Boom What were the relative contributions of the factors listed above to the export boom? The answer to this question will not only influence the choice of future policies in Turkey, it will also help determine the extent to which the Turkish adjustment experience can be “exported” to other countries. A complete answer would require an empirical general equilibrium model constructed specifically for that purpose. Since this is beyond the scope of the present work, we take a short cut instead. Our approach is to estimate a reduced-form regression equation for export volume and to simulate the consequences of alternative counterfactual scenarios for the 1980s. The marginal contribution of an exogenous variable can then be read as the difference between the “predicted” exports from two equations, one with all exogenous variables at their counterfactual values and the other with all variables at their counterfactual values except for the relevant variables. In principle, the appropriate regression model ought to include all of the variables discussed above, or at least a proxy for the unobservables. In practice, of course, this is not possible. The contribution of the investment drive during the 1970s cannot be simply read off from a regression. Neither is there any obvious way of capturing the role of the Iran-Iraq war except for a dummy variable. Subject to these limitations and if interpreted cautiously, however, the exercise is still of value. Table 7.4 contains the results of a quarterly regression for the 1970-84 period. Simple tests for the stability of the coefficients reveal no evidence of structural change after 1980. And since the longer time span provides better statistical results, we have preferred to work with data that go all the way back to 1970. Two additional variables were initially included in the model

726

Merih Celisun and Dani Rodrik Determinants of Export Volume

Table 7.4

From 1970 I until 1984: IV Observations

60 ,83458028 1.7155462 1.68483936

R2

SSR Durbin-Watson

Label

No.

Lag

Constant LRXTWPEW LIPDC LIVXRWP

D2 ONE TWO FOUR

-

Degrees of Freedom RZ

SEE

52 .8 1231224 . I 8 1635oO

Coefficient

Standard Error

r-Statistic

-2.0945750 ,4676637 1.0402740 -.7159398E-01 ,4138089 ,4278913 ,1089194 ,4641096

1.5953420 ,2316874 .2638576 .3479346E-01 .I017891 ,6696023E-0 1 .6688130E-01 .6717779E-01

-1.3 12932 2.01851 1 3.942559 -2.057685 4.065356 6.390230 1.628548 6.908676

Note: Dependent variable is LXTVOL.

Key: LXTVOL: Volume of exports (in logs). LRXTWPEW: Export-weighted real exchange rate index (in logs). LIPDC: Industrial production index for industrialized countries (in logs). LIVXRWP: Index of volatility of real exchange rate. D2: Dummy variable; 0 until 1981:I (inclusive), 1 thereafter. ONE, TWO, FOUR Quarterly dummies for first, second, and fourth quarters, respectively.

but have been dropped here because they proved to be statistically insignificant: real stock of domestic credit and a proxy for the export subsidy rate. The first was used to capture the state of domestic demand, and its lack of significance is perhaps not surprising in light of the discussion above. The unsatisfactory results with the proxy for the export subsidy are more disconcerting and probably reflect a large amount of “measurement” error.6 A time trend was originally included, but also turned out to be nonsignificant. Among the included variables, two are directly related to exchange rate policy. The leveE of the real exchange rate has the predicted positive effect on the volume of exports, with an elasticity slightly below 0.50, and is statistically significant at the 95 percent confidence level.7 The instability of the real exchange rate, as measured by an index of volatility, has a negative effect on export volume. The level of industrial production in the developed countries has the predicted positive effect, with an elasticity of just around unity. A dummy which takes on the value of 1 from 1981:II onward is also highly significant. Finally the quarterly dummies attest to the highly seasonal nature of Turkish exports. The overall fit of the regression equation is respectable, with an adjusted R 2 of 0.81. To calculate the contributions of the included variables on export growth we have to specify a counterfactual scenario regarding their time paths. In the following, we assume the following counterfactuals: (a) the real exchange rate remains unchanged from its 1979 level, (b) the volatility of the

727

TurkeyKhapter 7

real exchange rate remains unchanged from its 1978-79 average, (c) the developed countries continue to expand industrial production at the same rate as in 1970-79 (i.e., 3.36 percent per annum), and (d) the dummy variable for 1982:II remains zero throughout. The simulated levels of export volume over the 1979-84 period under the counterfactual scenario are listed in table 7.5. Notice that the simulated increase in exports is substantially less than the actual: by 1984, the actual volume of exports was almost 90 percent higher than the level that would have resulted under the counterfactual. Table 7.5 also includes a decomposition of the increase in exports between 1979 and 1984 due to the individual exogenous variables. These calculations are performed by simulating the trend of exports while holding all right-hand-side variables, except for one, on their counterfactual paths, and taking the difference between the level predicted in this fashion and the level in the counterfactual scenario. The results show that the bulk of the increase is due to the dummy variable for 1981:11, which alone “explains” 58 percent of the difference between the actual and counterfactual levels of exports in 1984. The real depreciation since 1979 explains 30 percent of the increase in exports, and the reduction in exchange rate volatility another 7 percent, bringing the total contribution of exchange rate policy to 37 percent. The slowdown in Sources of Export Performance, 197944

Table 7.5

Actual export volume Counterfactual” Simulationsb LRXTWPEW LIVXRWP LIPDC D2 Memo: Fitted values

1979

1980

1981

1982

1983

1984

100.8 103.8

100.0 108.8

161.8 112.6

194.7 116.7

202.2 120.7

235.2 125.1

103.8 103.8 103.8 103.8 103.8

123.2 101.7 104.2 108.8 110.2

128.8 113.5 105.0 170.3 183.3

142.3 122.1 101.3 176.5 195.6

150. I

158.3 132.7 112. I 206.6 227.6

128.0 103.6 195.6 206.6

Decomposition of 1984 Level of Export Volume (relative to 1979) Total increment to be explained Contribution of exchange rate policy: Level Volatility Total Contribution of slowdown in industrial-country growth Contribution of post-1981 dummy “Unexplained’ ’

235.2- 125.1 = 110.0 (100.0) 158.3-125.1 = 33.2 132.7-125.1 = 7.6 40.8 112.1 - 125.1 = -13.0 189.2-125.1 = 64.1 18.2

(30.2) (6.9) (37.1) (-11.8) (58.2) (16.5)

”The counterfactual scenario assumes the following: (i) an unchanged real exchange rate from the 1979 level; (ii) an unchanged level of real exchange rate volatility from the 1978-79 average; (iii) same level of industrial country growth in industrial production as in 1970-79, 3.36 percent p.a.; (iv) no dummy for the post-1981 period.

%e simulations are run by holding the values of all exogenous variables at their counterfactual level, except for the specified variable. For key to variable names, see table 7.4.

728

Merih Celiisun and Dani Rodrik

industrial countries, on the other hand, has made a negative contribution of - 12 percent. These results are striking in two respects. First, it is rather surprising that exchange rate policy has played such a comparatively small role in view of the vast real depreciations achieved since 1980. This, of course, is a reflection of the relatively small export supply elasticity estimated here. Secondly, the predominant role of the dummy variable points to a significant upward shift in export supply or export demand (or both) during 1981. It is tempting to label this effect as having been due to the Iran-Iraq war. Indeed, the expansion of exports to these two countries started during 1981. But in the absence of more direct evidence, this conclusion has to remain perforce tentative. The dummy variable may also be capturing the effect of export overinvoicing which appears to have begun in earnest in 1981. As explained above, the various export subsidies put into place, or strengthened, after 1980 greatly favored exports of manufactured commodities at the expense of other foreign-exchange-earning activities. The credit subsidies offered to exporters became particulary attractive as the banking sector’s real lending rates shot through the roof in 1981, due to the combined effect of financial liberalization and disinflation. This created an incentive to overinvoice such exports, or to simply declare exports where none had in fact taken place, in order to obtain subsidized credit. The fictitious component of the export earnings could then be made up by purchases from the black market or by reversed capital flight. This is one reason why a sustained premium for foreign exchange emerged in the black market after its virtual disappearance in the wake of the devaluation of January 1980 (see table A. 19 in the stat. app.). Since the traditional source of supply of foreign currency to the black market is workers’ remittances from abroad, the same phenomenon can also be observed in the declining trend of officially recorded remittances. How large was the magnitude of fictitious exports? A partial answer can be given by comparing partner-country import data with official Turkish export statistics. Two adjustments have to be made to partner-country data before they can be used for this purpose. First, one must adjust for the f.0.b.-c.i.f. difference in the recording of exports and imports. Here, we assume that the cost of insurance and freight adds 8 percent to the f.0.b. value of Turkish exports. This may be a bit too low, but it will allow us to form a conservative estimate of the extent of overinvoicing. Second, one must adjust for the presence of delivery lags.’ On this score we assume that goods spend an average of three months in transit, which allows us to match Turkish exports with partner-country imports moved forward by a quarter. Once these allowances are made, the difference between the two series can be interpreted as having been due to over- or underinvoicing on the part of Turkish exporters. Given the unreliability of trade statistics in most other countries, we carry out the analysis here for Turkey’s trade with other OECD countries only.

729

TurkeyKhapter 7

The results of the exercise are presented in table 7.6 for trade with the OECD as a whole and with West Germany. Look first at the implied patterns of overinvoicing in trade with the OECD. The period before 1981 seems to have been characterized by mild levels of underinvoicing, at an average rate of 4.2 percent. The reason presumably had to do with the existence of a black-market premium for foreign currency. Starting in 1981, there is an unmistakable transformation in the invoicing practices of exporters. The overinvoicing ratio is positive in every year during the 1981-85 period; it reaches its zenith in 1984 at a surprising level of 28 percent, before coming down to 8 percent in 1985 (presumably in line with the reduction of subsidies in that year). On average, the actual level of Turkish exports to the rest of the OECD appears to have been overstated by 13 percent during the 1981-85 period. Much of the overinvoicing appears to have taken place in trade with Germany, Turkey’s principal trade partner among OECD countries. If the numbers are to be believed, Turkish exports to Germany were overstated by 53 percent in 1984. However, note that overinvoicing appears to have been endemic to this particular bilateral trade relationship, even in the pre-1981 period, although not to the same extent. The combination of underinvoicing prior to 1981 with substantial overinvoicing since then makes for a less distinguished export performance in the 1980s than is revealed by official statistics. Once fictitious exports are eliminated, the average growth rate of Turkish exports to the OECD during

Overinvoicing of Turkish Exports, 197.5-85

Table 7.6

Exports by Destination (monthly ave., million $) Turkish Sources

Partner Sources”

Overinvoicing by Destination (%)

Year

OECD

West Germany

OECD

West Germany

OECD

West Germany

1975 1976 1977 1978 1979 1980

82.1 122.5 102.3 124.7 131.4 136.2

25.4 31.1 32.4 41 .9 45.6 48.8

102.2 104.9 105.9 130.3 137.0

26.2 28.2 30.4 34.1 40.1 48.8

-19.6 16.7 -3.4 -4.3 4.1 -8.6

-3.1 10.4 6.7 22.7 13.6 0.0

1975-80

116.5

37.5

121.6

34.6

4.2

8.3

1981 1982 1983 1984 1985

190.2 214.7 230.9 309.3 340.3

54.0 59 .O 70.6 105.8 115.3

186.6 194.0 205.1 241.9 314.4

47.2 47.9 58.4 69.4 83.0

1 .9

10.7 12.6 27.9 8.3

14.3 23.2 20.8 52.5 38.9

1981-85

257.1

80.9

228.4

61.2

12.6

32.2

~

149.1

~~

Source: OECD, Monrhly Bulletin of Foreign Trade Srarisrics, various issues. “Partner-countrytrade figures have been deflated by 1.08 to adjust for the f.0.b.-c.i.f.difference. In addition, the partner-source data have been moved forward a quarter to account for delivery lags (see text): the data for each year actually refer to the last three quarters of the relevant year and the first quarter of the next.

730

Merih CelAsun and Dani Rodrik

1980-85 is reduced to 16.1 percent per year (in dollar terms), which is not nearly as spectacular as the 20.1 percent calculated from official statistics, nor as impressively larger than the 7.8 percent for the earlier 1975-80 period. To make similar calculations with respect to Turkey’s global exports, we could assume alternatively that ( 1) there was no overinvoicing in exports to non-OECD areas, or (2) the extent of overinvoicing was the same in trade elsewhere as in trade with the OECD. The recorded and “actual” exports under the two scenarios then become as shown in table 7.7. Of the $5.0 billion increase in the level of total recorded Turkish exports between 1980 and 1985, the share of fictitious exports turns out to have been 9.2 percent ($0.5 million) under the first scenario, and 17.5 percent ($0.9 billion) under the second. It is not clear how alarming this phenomenon ought to be. To the extent that the export subsidies managed to increase the overall supply of foreign exchange to the economy, they would have to be deemed at least partly successful. Naturally, this success is diminished insofar as the foreign currency which came under the guise of export receipts would have come in the form of workers’ remittances anyhow. But more importantly, fictitious exports overstate the degree to which there was structural change in the economy in the direction of tradables in which a genuine comparative advantage exists. It also sheds some doubt on the permanence of the change of entrepreneurial attitudes in favor of exports. 7.4

Concluding Remarks

In a way, the export boom of the 1980s was an easy one to accomplish. The much-maligned investment-with-debt cycle of the 1970s had put into place a substantial industrial base. The economic collapse after 1977 in turn had led to large amounts of capacity underutilization. Once imported intermediate inputs became available after 1980-thanks to foreign official lending-and important export incentives were created through the exchange rate and overt subsidies, an export expansion of sorts was inevitable. Since export performance is the clearest success of the post-1980 program, whether the Turkish economy continues to be regarded as a case of successful adjustment depends to an important extent on the continuation of Table 7.7

Global Exports (in billion $) “Actual” Exports

1980 1985 Difference

Recorded Exports

(1)

(2)

2.910 7.958 5.048

3.065 7.641 4.582

3.186 1.352 4.166

731

TurkeyKhapter 8

the export boom. The boom, of greater concern to Turkish policymakers, is also needed to maintain credit worthiness. In some ways, Turkish exporters have shown remarkable flexibility in the face of adverse market development. As declining oil revenues choked Middle Eastern import demand, exporters have successfully reoriented their efforts toward the OECD. But export performance remains extremely sensitive to the domestic policy environment. A public-investment-led boom in domestic demand in 1986 resulted in an absolute reduction in export earnings. A renewed program of export subsidization since late 1986 appears to have revived exports, as well as overinvoicing, but the underlying fragility is still clearly there. A fundamental doubt regarding future export performance has to do with capacity constraints. The continuation of the export drive will henceforth require capital accumulation in export-oriented sectors, as output is reaching the limits of existing capacity. But, as discussed in previous chapters, private investment has remained soft since the late 1970s, and public investment continues to favor infrastructure projects with scant export potential. By late 1987, signs of a genuine structural transformation consistent with higher levels of trade orientation were still too few for comfort.

8

The Public Sector: Fiscal Adjustment and Resource Mobilization

In combination with price, incomes, and external borrowing policies, the government’s fiscal policy has had a close bearing on the conditions of macroeconomic stability, trade balance, resource mobilization, and growth in the Turkish economy in the post-1973 period. Fiscal policy has affected macroeconomic performance through the workings of public sector deficits and their financing mechanisms, and the mix of public revenues and spending. The 1973-77 period saw a surge in public spending and widening deficits, which were financed mainly through domestic credit expansion. Under reserve decumulation and heavy external borrowing, the expansion of credit to the public sector was largely sterilized by falling net foreign assets of the central bank, producing only a moderate monetary expansion and inflation. The unprecedented rise in imports also served to dampen the inflationary pressures. In turn, the reduced capital inflows in 1978-79 could no longer sterilize deficit financing through central bank credits, leading to a sharp

732

Merih Cellsun and Dani Rodrik

acceleration in monetary growth and inflation. Against the backdrop of rapidly worsening fiscal performance, the post- 1980 adjustment program required policy actions to lower public deficits and to restore more sustainable fiscal conditions as part of an effort to pursue an outwardoriented approach in the growth process. As pointed out in chapters 4 and 5, the fiscal correction in the post-1980 era included revenue mobilization in the SEE sector, real wage cuts for government workers, expenditure restraint, and a tax reform effort. To offset the contractionary effects of price shocks and falling private investment, public investment was maintained on a steadily rising path. The post-1983 period under the Ozal administration exhibits a new set of trends in fiscal strategy and adjustment, including fiscal decentralization, introduction of the value-added tax system, and increased reliance on domestic borrowing (at high interest rates) in financing fiscal deficits. We conclude that fiscal retrenchment has not been adequate in Turkey, portending serious policy difficulties for the late 1980s in coping with inflation, public debt, and deficits in an increasingly competitive political context. The present chapter provides an overview of fiscal adjustment and resource mobilization in Turkey’s public sector in the post- 1980 period. We will attempt to document and interpret the major fiscal trends within the framework of public sector accounts arranged on the basis of national income accounting concepts. The chapter ends with an assessment of issues in and prospects for fiscal policy.

8.1 Public Finance: Scope, Size, and Structure Turkey lacks a unified system of public sector accounts that is strictly adhered to in the presentation of subsector budgets. The classification and treatment of transactions have not been uniform across government units and have showed variation over time. Intrasectoral transfer payments and capital flows are not reported in sufficient detail on a regular basis. In the present Turkish setting, the highly aggregated data base of the SPO serves as the most consistent source of information on overall public finance. The SPO data on public finance include figures on public disposable income, savings, and investment, from which estimates may be derived (through the use of other relevant data) for public sector borrowing requirements (PSBRs). 8.1.1

Institutional Components

In the SPO data system, the Turkish public sector comprises six major components: ( 1) central government, (2) local government, (3) nonfinancial SEEs, (4) financial SEEs, (5) revolving fund agencies, and (6) extrabudgetary funds.

733

TurkeyKhapter 8

Central government covers all the usual public service departments. Its so-called consolidated budget (including budgets of several annexed agencies) serves as the central vehicle to mobilize public revenue and appropriate expenditures under the general scrutiny and approval of the Parliament. Local government comprises municipalities, special provincial administrations, and villages, which have had limited revenue-generating capacity until recently. The nonfinunciul (or operational) SEEs are directly engaged in the production of marketable goods and (nonfinancial) services, often requiring subsidies for their current operations and budgetary transfers for their investment programs. The financial SEEs include state-owned banks and, until recently, social security institutions. From 1984 on, the social security institutions have become revolving fund agencies, which exercise, as adjunct governmental entities, a considerable autonomy in their financial management. In the post-1983 period, the Ozal administration has introduced a large number of extrabudgetary funds, presumably to increase flexibility in revenue mobilization and expenditure allocation. The largest such funds in operation before 1983 were the Petroleum Price Stabilization Fund and the Support and Price Stabilization Fund (for fertilizer subsidies). As off-budget parastatals, these funds face less strict budgetary control and receive protection from general budget cuts. The main financial sources of extrabudgetary funds are: (a) various earmarked taxes and surcharges on foreign trade, bank credits, and other transactions, (b) income-sharing certificates of public utilities and enterprises, (c) interest income on the funds' financial assets, (d) foreign credits, and (e) donations and transfers from other funds. Among the new funds, the particularly sizable ones are the Mass Housing Fund (for residential construction), the Public Participation Fund (for public infrastructure investment), the Resource Utilization Support Fund (for export and investment incentives), the Development and Support Fund (for animal feed stock), the Mutual Assistance and Support Fund (for income transfers to the poor), the Petroleum Consumption Fund (for highways and municipal investments), and the Defense Industry Support Fund. The share of public services value added (comprising the gross salaries of government employees) in GNP was 9 percent in 1980, and declined to 6 percent in 1985. In turn, the ratio of the SEE value added to GNP was 11 percent in 1980, and increased to 17 percent in 1985. The share of SEEs in industrial value added was about 24 and 27 percent in 1980 and 1985, respectively. In terms of generating new employment, the public sector has played a restrained role in the post-1980 stabilization period. In the mid-l980s, the shares of public services and SEEs in nonagricultural employment were

'

734

Merih Celssun and Dani Rodrik

nevertheless sizable, about 20 and 11.5 percent, respectively. As discussed in chapter 5, the real wage cuts from 1978 on have caused significant changes in the functional and size distribution of income. 8.1.2

Public Revenue, Disposable Income, and Expenditure

Table 8.1 shows public sector revenue, current transfers, disposable income, and final expenditure as a percentage of GNP (in current prices) over the 1978-86 period. The flows in this table exclude capital transfers to and from the private sector and abroad, and thus may be viewed as current account items. It should be emphasized that the receipts from off-budget funds are not included in the revenue figures of the pre-1984 period.' The SPO coverage of extrabudgetary funds in public revenue is partial in the post-1984 period, but progressively increases from 1984 onward. Current transfers as an item includes subsidies, interest payments, tax rebates, and other income transfers to (and from) the private sector and rest of the world, making up the difference between public revenue and disposable income. Final expenditure is the sum of public consumption and investment, including inventory changes. In this table, the revenueexpenditure balance is equivalent to the savings-investment balance (or

Table 8.1

Public Sector Current Account, 1978-86 (as a percentage of GNP, current prices)

Public Sector I . Revenueb 2. Current transfers' 3. Disposable income ( I minus 2) a. Consumption b. Savings 4. Final expenditure (consumption + investment) 5. Total expenditure (2 4Id 6. Savings-investment balance (3 minus 4) Memo items: Wealth tax Public disposable income (1983 prices) of which Consumption Savings

+

Source:

1978-79 Average

1980

1981

1982

1983

1984

1985

1986"

20.3 2.9

19.8 2.3

22.2 2.9

22.9 3.2

22.4 5.0

21.8 5.2

23.8 6. I

25.6 8.1

17.4 13.4 4.0

17.5 12.2 5.3

19.3 10.7 8.6

19.7 10.9 8.9

17.4 10.1 7.3

16.6 9.0 7.6

17.7 8.5 9.2

17.5 8.6 8.9

22.9 25.8

23.7 26.0

23.9 26.8

22.7 26.0

21.6 26.6

19.0 24.2

19.5 25.6

20.8 28.9

-5.5

-6.2

-4.6

-3.1

-4.2

-2.4

- 1.8

-3.3

0.2

0.1

0.1

0.1

0. I

0.2

0.2

0.4

12.4 9.8 2.6

15.5 10.9 4.6

18.6 10.5 8. I

18.8 10.2 8.6

17.4 10.1 7.3

16.2 9.8 6.4

17.4 9.6 7.8

16.0 9.4 6.6

SPO (1985), Central Bank (1987), and various SPO Annual Programs

"Provisional estimates. bExcludes wealth and capital Rows. 'Includes subsidies and interest payments; excludes capital transfers. dExcludes capital transfers and debt (principal) repayment.

735

TurkeyKhapter 8

savings gap) of the public sector. Conceptually, wealth taxes (on property and motor vehicles) are not considered as an income flow, but as a capital transfer item. They are thus excluded from public revenue figures, but shown as a memo item in table 8.1. Leaving the review of fiscal adjustment to section 8.2, we may now point to a number of basic trends emerging from the data shown in table 8.1. Public expenditure (excluding capital transfers) has been consistently around 26 percent of GNP during the 1978-85 period. Public revenue has remained several percentage points below public expenditures in this period. Notice that there has been a sharp rise in current transfers since 1983. This reflects mainly increased interest payments, which have required a downward adjustment in final expenditures in relation to domestic product. The constant-price share of public disposable income in GNP (shown as another memo item in table 8.1) may be compared with the corresponding share in current prices to note the strong impact of relative price changes on the fiscal position in 1980 and 1981. Table 8.2 gives the breakdown of public revenue (by sources) and disposable income (by institutional components) for selected benchmark years from 1979 to 1986. The data in the table show that the post-1980 economic program markedly altered the pattern of revenue mobilization in the public sector. The increased revenue contribution of factor income from

Table 8.2

Structural Change in Public Revenue and Disposable Income, 1979-86 I979

A. Public revenue (I) I. Taxes a. Direct taxes b. Indirect taxes c. Subtotal 2. Nontax budget revenue 3. Factor income from propertyb 4. Total B. Public disposable income (PDY.%)c 1. Central government budget 2. Local government 3. SEEs (nonfinancial) 4. SEEs (financial) 5. Revolving fund agencies 6. Extrabudgetary funds 7. Total

1982

1985

I 986a

56.6

46.3

25.7

26.3

43.7 -

100.3 -

32.2 78.5 -

38.1 63.8 -

41.4 -

-3.8 -

14.4 -

3.5

100.0 92.2 7.7 -4.1 3.3 1 .o

l00.od

7. I

100.0 81.2 7.4 6.3 4.2 0.9 -

100.0

Source: SPO. ’Provisional estimates. bIncludingnet surplus of social security institutions. ‘The PDY figures for subsectors are derived after taxes and intrasectoral transfers. dColumn does not add up exactly due to rounding errors.

9.2 27.0 100.0

51.0 12.4 26.0 0.6 3.6 6.4 100.0

-

67.7 7.5 24.8 100.0 _ .

52.6 14.7 19.4 0.7 3.2 9.4 100.0

-

736

Merih Celasun and Dani Rodrik

property (including SEE profits and depreciation allowances) provided room for tax reform initiatives, which tended to lower the tax/GNP ratio in 1980-84. Only after the introduction of the VAT (in 1985) and extrabudgetary funds did the relative share of indirect taxes begin to shift upward. Correspondingly, the structure of public disposable income shifted away from the central government in favor of SEEs, local government, and extrabudgetary funds, with sharp implications for the savings structure as discussed in section 8.4. 8.1.3 PSBR In the absence of official indicators of the overall PSBR, we have derived two sets of estimates under two variant procedures. In variant A, the PSBR(A) is estimated by adjusting the public savings-investment balance for the following three factors: ( 1) nondebt capital transfers (including wealth tax, grants, acquisition or sale of property, and capital flows connected with state participations); (2) valuation differences for the year-end SEE inventories; and (3) increase in accounts payable (or arrears) in the central government budget. In variant B, the PSBR(B) is estimated as the total cash deficit of the central government and nonfinancial SEEs, excluding the cash needs and/or surpluses of other public sector entities. It may be noted that the cash deficit of the central government (included in PSBR(B)) is the budget deficit adjusted for arrears. Table 8.3 shows the estimates PSBR(A) and PSBR(B) for the 1980-86 period. Despite differences for particular years, the trends displayed by the two series are not too dissimilar. The estimation of PSBR(A) is conceptually more satisfactory, even though the quality of underlying data on nondebt capital transfers is somewhat q u e ~ t i o n a b l e .This ~ qualification as regards data notwithstanding, the PSBR(A) estimates appear to be closer to the actual cash deficits of the overall public sector in the post-1980 adjustment period, and for this reason we have chosen to rely on this version here and in other chapters. Table 8.3 also gives the financing items for PSBR(A), which point to the tendency toward reduced central bank financing and increased domestic borrowing, especially after 1983. The particular fiscal characteristics of the intervening years in 1980-86 are reviewed in the next section.

8.2 Fiscal Adjustment, 1980-86 8.2.1

Overall Fiscal Policy

In main, four sets of factors have shaped the overall fiscal policy in the 1980-86 period. First, public sector deficits prior to 1980 required immediate corrective actions, both in the budget and SEEs, to regain control over monetary growth and inflation. Second, the policymakers attached

737

TurkeyKhapter 8 PSBR Estimates, 1980-86 (as a percentage of GNP)

Table 8.3

1. Variant A: PSBR(A)

Financing (net) External borrowing Domestic borrowing (Treasury) Budget, long term Budget, short term Central bank Other 2. Variant B: PSBR(B) Cash deficit a. Central government budget b. SEEsb b) = PSBR(B) c. Total (a Memo items: Interest rate on government bonds (%) Increase in WPI (%)

+

1980

1981

1982

1983

1984

1985

1986'

9.9

3.7

5.0

5.3

1.9

4.9

4.1

3.2 1.1 0.2 0.9 3.5 2.2

2.5 1.4 0.8 0.6 2.0 -2.2

1.0 1.5 0.6 0.8 0.3 2.2

1.4 0.9 1.8 -0.9 0.6 2.3

2.7 2.3 0.7 1.6 0.7 2.1

0.5 2.1 I .9 0.9 1.3 0.4

0.5 2.8 1.2 1.6 0.6 0.8

5.0 6.6 11.6

1.3 4.1 6.0

2.1 3.9 6.0

2.1 2.9 5.0

5.2 2.8 8.0

2.8 2.8 5.6

3.2 2.4 5.6

107

31

25

45 31

59 52

56

40

50 21

Source: Variant A: Estimates based on SPO data, see tables A.7 and A.8 in statistical appendix. Variant B: OECD (1986) for 1980-84 and Central Bank (1987) for 1985-86 estimates. "Provisional estimates bCash deficit of nonfinancial SEEs after budgetary and parabudgetary transfers, and before arrears and State Investment Bank Credits.

importance to output recovery and growth objectives, and chose to reduce the public savings gap through increased savings rather than through a major reduction in public investment, especially in the context of falling private fixed investment. Third, the tax system needed qualitative changes and restructuring to halt bracket creep (or fiscal drag) in income taxation, reduce evasion, and revitalize indirect tax revenues, which had fallen rapidly in proportion to GNP in the latter part of the 1970s. Moreover, financial liberalization and various supply-side concerns required adjustments in tax burdens and incentives. Hence, a complex tax reform package had to be introduced with rather uncertain prospects for the tax/GNP ratio in the medium run. Fourth, a fiscal decentralization away from central government was perceived to be essential by the authorities to increase allocational flexibility in general and improve local government finances in particular. Thus, the overall fiscal policy evolved under a number of cyclical and structural constraints in such a manner as to support the stabilization, recovery, and liberalization objectives in the post- 1980 adjustment program. For adjustment mechanisms, the fiscal policy relied mainly on flexible pricing in the SEEs, real wage reductions, tax restructuring, and lately domestic borrowing in an effort to sustain an acceptable growth of developmental expenditures in the economy. The multiplicity of policy

738

Merih Cellsun and Dani Rodrik

objectives and instruments produced conflicts, however, in the implementation process. The fiscal adjustment in the intervening years is reviewed in the rest of this ~ e c t i o n . ~ 8.2.2

1980-82

This subperiod saw a determined price-stabilization effort, which resulted in a sizable fiscal retrenchment. The PSBR was nearly halved in this policy phase, mainly due to the rise in public revenue, disposable income, and savings. The external debt relief was helpful in avoiding a rapid increase in current transfers (including interest payments), and therefore indirectly contributed to the attainment of deficit reduction. Central bank financing of the PSBR declined sharply from 3.5 percent of GNP in 1980 to 0 .3 percent in 1982. Revenue mobilization was boosted in 1980-82 primarily by the rise in SEE factor incomes. With the maintenance of government employment at reduced real wages, public consumption continued to grow in real terms, providing social services to the economy at sharply lowered relative prices. The tax reform initiated in late 1980 showed positive results in 1981, yielding an increase in tax elasticity (i.e., the proportional response of tax revenues to increases in income) from 0.9 to 1.2 in that year. However, the tax elasticity dropped to 0.8 in 1982 because of altered income brackets and lowered tax rates. 8.2.3

1983

The public savings gap widened in 1983 because of the decline in the ratio of public disposable income to GNP, which was in turn due to the sizable increase in external interest payments of the central government budget. That year saw a legislative initiative to simplify the settlement of tax arrears and to assess taxable income in relation to observed expenditures. The further fall in tax/GNP ratio could not be adequately offset by SEE revenue increases. The PSBWGNP ratio could be held around 5 percent in 1983, mainly through a large increase in the arrears of the central government. 8.2.4

1984

After the general elections of November 1983, the Ozal administration could not reverse in 1984 the deteriorating revenue performance and the rising current transfers of the public sector. Foreign interest payments were grossly underestimated in the central government budget, but had to be fully serviced. The increase in the nominal value of SEE inventories (including agricultural support purchases) was unexpectedly large. In terms of public finance shares in GNP, the burden of adjustment was mainly on final expenditure categories. A restraint on nontransfer expenditures could not, however, halt the widening of the overall cash deficit, which increased above 7 percent of GNP, requiring an expansion in central bank financing and a significant rise in borrowing by the Treasury.

739

TurkeyKhapter 8

In the first half of 1984, the public sector financing pressures on money supply were augmented by pressures from the balance of payments as export earnings were stimulated by trade liberalization and new incentives. Despite the large rise in interest rates, the liquidity expansion could not be contained, and the inflation rate surged to 52 percent (as measured by the WPI) in 1984 from 31 percent in 1983. The reduction in real demand for money base after the introduction of foreign exchange deposits might also have contributed to excess money supply in 1984. In order to attract private financial savings, the interest rate on Treasury borrowing was raised to nearly 60 percent with vast consequences for the interest burden on the central government budget. 8.2.5

1985-86

The worsening fiscal trends were partially reversed in 1985. Despite the continuing rise in current transfers, the shares of public disposal income and savings in GNP could be raised that year mainly through a rapid recovery in public revenue. The sources of revenue expansion were SEE factor incomes, the VAT, nontax revenues, and various levy collections for the extrabudgetary funds. The reversed fiscal trends more or less continued in 1986. The PSBWGNP ratio was lowered from above 7 percent in 1984 to less than 5 percent in 1985-86, together with the reduction in the inflation rate to 27 percent in 1986 from 52 percent in 1984. In the context of falling dollar prices of oil and nonoil imports in 1986, a larger reduction in inflation could have been attained through a moderate rise in public expenditures and smaller cash deficits.

8.3 Public Debt As argued in chapters 4 and 5, the post-1980 Turkish economic recovery benefited from the external debt relief extended in 1980-84. With the termination of debt relief in 1984, Turkey began to face an increase in external debt service, which may extend, if the debt is not partly rescheduled, over the 1985-90 period. This debt overhang gives rise to two familiar problems: the need for noninterest surpluses in the current account of the balance of payments, and additional domestic resource mobilization by the public sector for debt servicing. As in other heavily indebted middle-income countries, the bulk of Turkey’s external debt is held by the public sector. The rise in external debt service raises current transfers (through larger than usual interest payments), lowers public disposable income, and puts pressure on the budget balance. Furthermore, the repayment of principal reduces the volume of net foreign borrowing as a financing item for the PSBR, requiring the expansion of either central bank financing and/or domestic borrowing through the issue of government securities. Also, as part of the policy actions taken to increase

740

Merih Celisun and Dani Rodrik

international competitiveness and generate noninterest current account surpluses, the exchange rate depreciation tends to increase the relative size of debt service in relation to budget revenue and expenditures. In Turkey in 1984-86, policymakers faced all these problems and chose to use domestic borrowing as the main mechanism to cope with the remaining fiscal disequilibrium. However, a heavy reliance on domestic borrowing accelerated the rise in public debt from 1984 on, creating a larger debt claim on limited public sector resources. Table 8.4 shows the principal indicators of the Turkish public debt from 1981 to 1986. In the presentation of figures for debt stock in part A of this table, following the official practice, the public sector includes the SEES, while excluding the central bank. As a proportion of GNP, the public debt was about 35 percent in 1981-82, nearly 45 percent in 1983-84, and 49 percent in 1985-86.6 From 1983 on, the ratio of public debt to public revenue has fluctuated around 200 percent. Within the public sector, the central government budget carries the major burden of servicing public external debt. With an increasing recourse to Table 8.4

Indicators of Public Debt, 1981-86

A. Public debt stock 1. Public external debt a. billion $ b. trillion TL 2. Public domestic debt (trillion TL) a. Consolidated debtb b. Nonconsolidated debt c. Subtotal 3. Total public debt ( l b 2c) a. trillion TL b. % G N P c. % public revenue B. Debt service in central government budget (% GNP)' 1. External 2. Domestic 3. Total Memo items: Central government budget Revenue (% GNP) External debt (billion $) Central bank Private sector 'hrkey

+

1981

1982

1983

1984

1985

1986a

10.3 1.1

11.6 1.9

1.6 2.6

13.0 4.7

14.9 7.7

18.7 12.5

0.5 0.5 1.o

0.7 0.7 1.4

1.7 0.8 2.5

2.5 1.o 3.5

3.9 1.8 5.7

4.1 2.9 7.0

2.1 32.0 144.1

3.3 37.7 164.6

5.1 44.4 198.2

8.2 44.8 205.5

13.4 48.5 203.8

19.5 49.1 191.8

0.9 0.9 1.8

1.5 0.8 2.3

2.4 0.7 3.1

2.7 0.9 3.6

3.1 1.2 4.3

3.7 2.8 6.5

20.3

19.6

18.7

15.4

17.0

17.4

4.3 2.2 16.8

4.1 2.0 17.7

5.3 1.5 18.4

5.6 2.6 21.2

6.7 3.7 25.3

7.6 4.9 31.2

Source: Central bank of Turkey, SPO, and OECD (1986) "Provisional estimates. %e bulk of consolidated debt covers devaluation-induced valuation changes in external debt held by the central bank. 'Including principal and interest payments

741

TurkeyXhapter 8

Treasury borrowing in financing the cash deficits, the domestic debt service burden also rose sharply in the post-1984 period as shown in part B of table 8.4. The ratio of total debt service to revenue in the central government budget increased from about 15 percent in 1981-83 to nearly 40 percent in 1985-86. The latter has required a tight stance on social outlays and personnel expenditures, lowering the general quality of social services (e.g., health and education) in Turkey. In the review of public debt, it should also be noted that the so-called consolidated debt (of the central government) constitutes a significant portion of domestic public debt. Besides including obligations previously contracted, this debt stock primarily covers devaluation-induced differences in the Turkish lira value of external debt held by the central bank, which are treated as Treasury liabilities in the monetary authorities’ accounting system. The nominal interest rates on consolidated debt are very low, and therefore average interest rates on total domestic debt have been much lower than interest rates on the newly issued government securities. From 1984 on, the domestic borrowing for PSBR financing has relied basically on two financial instruments: short-term Treasury bills and longer term government bonds, with small nominal differences in their respective interest rates (see memo items in table 8.3). Since May 1985, the interest rates of these securities have been determined through weekly auctions. These default-free, tax-exempt, and high-yield money-market instruments have become quite popular for commercial banks, which are allowed to hold them against their liquidity requirement^.^ Table 8.5 provides data on security issues in Turkey during 1982-86, with the volumes issued by the public and private sectors expressed as percentages of GNP. The indicators in the table crystallize the predominance of the public sector in the issue of securities in the Turkish financial system after 1983. While contributing to the promotion of financial intermediation, the disproportionately high share of the public sector in new security issues

Table 8.5

Securities Issued, 1982-86 (as a percentage of GNP) ~~

I . Public sector a. Government bonds (long term) b. Treasury bills (short term) c . Income-sharing certificates d. Total 2. Private sector a. Bonds b. Equities c. Total

1982

1983

1984

1985

1986

0.9

2.2

-

2.0 6.9

3.2 4.5 0.6 8.3 0.3 0.3 0.6

-

-

0.9

2.2

8.9

4.3 5.1 0.5 9.9

0.2 I .2 1.4

0.1 0.8 0.9

0. I 0.6 0.7

0.1 0.3 0.4

-

~

Source Central Bank (1987) Note Em-dashes indicate that percentages were negligible or zero

_

_

_

_

_

742

Merih Cellsun and Dani Rodrik

has tended to crowd out suitable domestic financing for private investment, especially in the manufacturing sector, as further discussed at the end of the following section.

8.4 Role of the Public Sector in Savings and Investment 8.4.1 Basic Trends

A review of Turkey’s public finances also requires a broad discussion of the share and role of the public sector in economywide savings and investment. The public sector influences capital formation directly through its own savings effort and investment programs. It affects private savings and investment behavior indirectly through demand management, external borrowing, and structural policies. Keeping in mind the longer term growth requirements of the Turkish economy, policymakers have placed a high premium on public investment in the 1980s (with some restructuring of its contents) at the cost of a continued fiscal imbalance. Figure 8.1 illustrates the changes in investment rates (as percentages of GNP in current prices) from 1978 through 1986. The investments portrayed in this figure cover inventory changes (excluding valuation adjustments for year-end inventory stocks). Besides pointing to the relatively larger share of the public sector in total investment, figure 8.1 also brings out the active demand management role of the public sector in offsetting the large slack in private investment in the wake of the post- 1980 adjustment program. In turn, the changes in the composition of economywide savings over the same period are illustrated in figure 8.2. The time paths for various categories of savings (also measured as percentages of GNP in current prices) bring out four major points. First, the public sector saving effort rapidly improved under the post-1980 adjustment program, with some slippage in 1983-84. Second, notwithstanding the interest rate reform, the private savings/GNP ratio declined in 1980-83. Third, foreign savings were sizable in 1980-81, lending support to domestic measures aiming at macroeconomic stabilization. Fourth, from 1984 on, the share of private savings in GNP began to show some recovery from the lowest point reached in 1983. To crystallize the savings-investment patterns further, table 8.6 gives (in part A) figures for the relevant ratios (to GNP), and provides (in part B) disaggregated data for the public sector. In effect, this table complements table 8.4, which presented an institutional breakdown of public disposable income for the same benchmark years between 1979 and 1986. While part A in table 8.6 quantifies the improvement in public savings from 1979 on, it also invites attention to the sharp rise in private savings and investment in 1986. In part B of table 8.6, the breakdown of public savings and investment by institutional components over time clearly shows the decentralization trends

TurkeyXhapter 8

743

30

I

I

I

I

I

I

I

I

I

25 IP + IG

a z 20rc O

8

G

1510-

/-. IP

---d

*------/

/

/.

0.

\,--

50

I

I

I

1

I

I

I

I

I

in the public sector. The share of central government in public savings sharply fell to about 17 percent in 1985-86 from more than 100 percent in 1979. Under flexible pricing policies, the SEES managed to increase their capacity for resource mobilization quite significantly, albeit at the cost of preventing deeper inroads into price stabilization. The investment programs of local government and extrabudgetary funds also became the beneficiaries of the reduced role of central government in the overall public finance. 8.4.2

Domestic Saving Behavior

Table 8.7 lists the major indicators of domestic savings from 1978 to 1986, which are based on current price data.8 As these indicators show, the private saving performance was sluggish in 1980-85 despite the switch to positive real interest rates (on time deposits), which played a crucial role in monetary adjustment as discussed in chapters 4 and 5 . In turn, the saving drive of the public sector contributed more effectively to the adjustments in

744

Merih CelIsun and Dani Rodrik

I

I

I

I

I I I I 197879 80 81

I

1

I

I

I

I I 1 I 1 82 83 84 85 86

Fig. 8.2 Savings (% GNP, in current prices) Note: SP is private savings; SG is public savings; and F is foreign savings

the real side of the economy. The share of public sector in domestic savings increased to 50 percent in 1985 from 25 percent in 1978-79. How can we then explain the disappointing saving performance in the private sector in the earlier part of the 1980s, especially in the aftermath of a major reform of the interest rate policy?' The relevant income base for private savings (and consumption) is not national income, but private disposable income, which dropped 5 percent in real terms in 1980 after stagnating in 1978-79. It also failed to increase in 1981. In per capita terms, the cumulative real fall in private disposable income was about 10 percent from 1978 to 1981. At the aggregate level, one possible hypothesis is that private agents strived to protect their consumption levels by lowering their short-run propensities to save. The interpretation of private saving behavior in the early 1980s is confounded by another factor, namely the observed distributional change in favor of nonagricultural capital income, which is expected to have a priori a higher savings ratio. The distributional analysis summarized at the end of

Table 8.6

Structural Change in Savings and Investment, 1979-86

~

1979

1982

1985

I986*

Savings Investment

Savings lnvestment

Savings Investment

Savings Investment

~_______

A. Sector totals (% GNP, current prices) 1. Public 2. Private 3. Rest of the world 4. Total B. Public sector (%) I . Central government budget 2. Local government 3. SEES (nonfinancial) 4. SEES (financial) 5. Revolving fund agencies 6. Extrabudgetary funds 7. Total Source: SPO

Note: N.A. means not applicable.

aProvisional estimates.

2.7 13.5 2.1 18.3

9.5 8.8 N.A. 18.3

107.4 -8.1 -24.7 19.6 5.7 N.A. 100.0

33.6 3.8 60.0 0.2 2.4 N.A. 100.0

-

-

8.9 9.2 2.2 20.3

12.0 8.3 N.A. 20.3

9.2 9.2 1.2 20.5

11.0 9.5 N.A. 20.5

8.9 12.1 2.2 23.2

68.7 6.2 13.9 9.2 2.1 N.A. 100.0

39.3 4.7 54.1 0.3 1.6 N.A. 100.0

16.1 14.2 49.6 1.2 6.8 _12.2_ 100.0

32.3 9.6 50.8 0.4 1.4 - 5.5_ 100.0

18.7 17.1 38.1 1.3 6.2 _ 18.6 100.0

- -

- -

11.0 12.2 N.A. 23.2

35.0 15.4 37.9 0.3 2.3 9.1 100.0

746

Merih CelPsun and Dani Rodrik

Table 8.7

Savings Ratios, 1979-86* (in percentages) 1978-79 Average

1980

1981

1982

1983

1984

1985

1986b

1. Ratios

a. Private savings to private disposable income b. Public savings to public disposable income c. Domestic savings to GNP d. Foreign savings to GNP e. Total savings to GNP 2. Composition of domestic savings a. Private b. Public C. Totdl Memo items: Growth of private disposable income Growth of private fixed investment Real interest rate (I-yr deposits, after tax)

14.6

12.9

11.6

11.5

11.2

11.0

11.4

14.6

21 .0 16.0 2.4 18.4

29.8 15.9 5.5 21.4

43.4 18.0 3.5 21.5

45.2 18.1 2.2 20.3

42.1 16.5 4.1 20.6

46.3 16.8 2.8 19.6

52.3 18.6 1.9 20.5

50.9 21.0 2.2 23.2

50.8

55.8

54.8

50.5

75.0

66.7

52.2

25.0 -

33.3 -

47.8 -

0.9

-5.1

0.3

-8.9

-17.3

-27.1

-37.6

100.0

100.0

100.0

-8.7 0.2

44.2 49.2 -

49.5 45.2 -

57.6 42.4 -

100.0

100.0

5.1

7.6

3.4

9.8

5.5

4.8

1.8

13.3

13.5

8.2

4.9

4.0

7.9

15.9

100.0

100.0

4.3

100.0

Source: SPO (1985) for 1978-83, and Central Bank (1987) for 1984-86 data. "Parts 1 and 2 are based on current price data. bProvisional estimates.

chapter 5 points to the large income gains of the wholesale and retail trade sector from relative price changes observed in the post-1978 period. The saving propensity of income recipients in the trade sector might have been somewhat lower than the saving rates in other productive sectors, which contributed more substantially to savings in the earlier periods. A further observation in this context relates to the composition of the private savings aggregate defined in the official national accounts, which we have used. Besides covering household savings, this aggregate also includes depreciation allowances and undistributed enterprise profits, which declined in some manufacturing firms confronted with illiquidity problems in the early 1980s. Moreover, the private savings aggregate in the national accounts is not voluntary savings in the pure sense, as it implicitly includes the forced transfers (to the public sector) effected via new money creation. The contraction of real money base from 1979 on (recall fig. 4.1) definitely played a role in lowering the seignorage portion of private savings in the early 1980s. Finally, it may be noted that the nature of adjustment mechanisms (or closure rules) for the savings-investment balance also affects the determination of private savings in a general equilibrium context." In the face of low capacity utilization rates, reduced aggregate demand, and high user cost of

747

TurkeyEhapter 8

capital, investors’ enthusiasm was definitely at a low level prior to 1985, possibly dampening real demand for private savings. The actual outturn for 1985-86 suggests that private savings adjusted quite rapidly to the rising investment demand in the private sector in the wake of generous new incentives provided by the extrabudgetary funds, especially for mass housing projects. In concluding our brief discussion on saving behavior, we would like to draw attention to the fact that domestic saving rates in Turkey have not been high by cross-country standards, as was shown in table 1.2. In relation to the size of its disposable income base, the public sector’s direct contribution to resource mobilization for investment has been substantial in the 1980s, requiring a prolonged tight stance on public consumption. In turn, private saving rates have been rather low, despite the policymakers’ preference for a market-directed growth process in the future. The present discussion suggests that private saving may respond favorably to new incentives under more viable and stable macroeconomic conditions for investment and growth. 8.4.3 Investment Allocation To complement the analysis of saving and investment patterns at the macrolevel, table 8.8 provides sectoral data on the distribution of fixed investment for the periods 1973-78, 1979-83, and 1986. The sector-level data on investment allocation reveal four notable trends. First, the share of manufacturing in total investment exhibits a continuing decline from 1978 on. Bearing in mind the heavy reliance on manufactured goods in the Turkish export drive, this trend gives rise to the question of the sustainability of export growth in the early 1980s. Second, the share of

Table 8.8

Fixed Investment by Major Activity Sectors, 1913-86 (as a percentage of total) 1986b

Agriculture Energy & mining Manufacturing Transportation Tourism Housing Education & health Other services

1973-78 Total”

1979-83 Totala

Private

Public

Total

11.0 11.1 30.2 19.0 0.7 17.1 3.8 7. I

10.0 20.6 25.6 18.0 0.6 15.8 3.4 6.0

6.6 1.6 31.4 17.0 2.2 36.0

8.0 28.5 8.3 33.2 1.1 1.8 5.6 13.5

7.4 16.9 18.2 26.2 1.6 16.5 3.6 9.6

Source: SPO.

“In constant 1983 prices. bProvisional estimates reported in Central Bank (1987).

1 .o

4.2

748

Merih CelAsun and Dani Rodrik

agriculture in total investment has been steadily low, despite the predominant position of this sector in total employment. From 1978 on, the depressed levels of investment in this sector may be attributed to the fall in farmers’ real incomes and generally weakened external demand conditions for agricultural products in the mid-1980s. Third, the rising share of energy and mining (mainly hydroelectric power and coal) is a positive element in the pursuit of reduced dependence on energy imports in the growth process. Fourth, transportation and housing investments have definitely been boosted from 1983 on, following relatively low levels of investment activity in these sectors in 1978-83. The 1986 data also point to an encouraging revival in tourism investments. To arrive at sound judgments on investment allocation, a wider analysis is needed of Turkey’s comparative advantage, factor proportions, and sectoral interdependence in the longer term growth process. The investment trends emerging from data shown in table 8.8 broadly suggest that an allocational shift is needed from services to manufacturing to avoid capacity bottlenecks in outward-oriented growth in the medium-term future. In the context of investment allocation, a critical problem is the efficient deployment of various categories of private savings to high-priority investment programs. In the past, Turkey’s financial markets have played a highly limited role in coping with this problem. In the late 1980s, vigorous recovery of manufacturing investment requires, besides the sector’s own resources, additional savings from the financial system at moderate costs. As the data presented earlier in table 8.5 show, the share of private bonds and equities in total security issues was less than 10 percent during 1984-86, a period which saw an unprecedented rise in the Treasury borrowing at high real rates of interest. Hence, the size and pattern of private investments will be closely affected by the fiscal adjustment in the public sector.

8.5 Fiscal Policy: Achievements, Issues, and Prospects We have attempted in the present chapter to broadly document and review the patterns of fiscal adjustment and public resource mobilization in the 1980-86 period. The examination of fiscal trends suggests that SEES, extrabudgetary funds, and local government should be integrated into public accounts for a wider assessment of fiscal policy links with macroeconomic performance. For revenue mobilization, fiscal policy has relied heavily on SEE price adjustments. SEE prices were clearly out of line before 1980 and required immediate and large corrections at the outset. However, in the absence of deeper rationalization of the SEE system, a continuing and excessive reliance on SEE price hikes-in conjunction with markup pricing practices in the private industrial sector-has produced conflicts in the price stabilization process and contributed to the volatility of inflation rates.

749

TurkeyKhapter 8

Turkey’s tax reform effort resulted in tax revenue losses toward the mid-1980s (after a short-lived tax rise in 1981) and contributed to the weakening of fiscal retrenchment. Again, the initial conditions were important in the choice of policy direction. The tax reform focused on tax restructuring to overcome bracket creep in income taxation, reduce the tax burden on financial intermediation, and broaden the base for indirect tax revenue. The illiquidity problems of part of the private sector also required a relief in corporate taxes. The tax reform initiatives began to pay off after the introduction of the VAT system (in 1985) and the switch to tariff protection in the trade regime. Despite their unduly complicated nature, the taxes, surcharges, and nontax revenues collected by local government and new extrabudgetary funds also contributed to the rebound in public revenue in 1985-86. The Turkish tax reform experience shows that tax restructuring is a lengthy process and should not be exclusively relied upon as a vehicle for additional revenue mobilization in the early phases of adjustment programs. In this connection, the important lesson is that the tax system requires continual modernizing improvements in a rapidly changing economic structure. In the future, the Turkish tax system should have a broader base to encompass taxable incomes in the services sector. A rapid output recovery after the 1978-80 episode required the maintenance of real growth of public investment with some alterations in its allocational pattern. Because of rising current transfers, the brunt of adjustment in the disposition of available net income fell on public consumption, mainly through real reductions in the salaries of government employees. The latter mechanism seems to have produced a general worsening in the quality of social services. Turkey faces a substantial increase in its external debt service in the 1985-90 period. In 1985-86 the burden of external interest payments prevented a notable recovery in public disposable income (in relation to GNP) despite a marked rise in public revenue. With the reluctance to restrain public investment, the overall fiscal deficit (PSBR) could not be lowered adequately after reaching a peak in 1984. In the face of reduced net foreign borrowing, policymakers chose to rely on domestic borrowing at high real rates of interest, which crowded out investment financing in the private sector, especially in manufacturing, whose dependence on the financial system has increased in the post-1980 era. From 1984 on, the heavy reliance on Treasury bond issues in coping with fiscal disequilibrium has given rise to a complex set of policy issues. The bond-financing of deficits adds a rising domestic debt burden to the public sector’s external debt service. At the macro level, the internal adjustment shaped by fiscal policy becomes compatible with the balance-of-payments constraint at the cost of very high real rates of interest, to which private savings did not sharply respond in the 1980-85 period. On the other hand,

750

Merih Cellsun and Dani Rodrik

high interest rates aggravate the illiquidity problems and tend to discourage a vigorous investment recovery in the manufacturing sector, which provides the major productive base, besides tourism and some services, to sustain export expansion. Favorable supply conditions for export and GNP growth are essential to sustain the trade-liberalization process and to maintain an adequate capacity to service external debt. As we argued in section 8.4, macroeconomic stability is particularly important for a harmonized savings-investment balance in the Turkish economy. The large fiscal deficits pose a threat to macroeconomic stability and growth because their monetization remains a tempting policy option, given the historical precedents in the Turkish context. With limited possibilities to reduce public consumption further, a noninflationary fiscal adjustment will have to be based on a socially acceptable mix of public revenue increase and restrained public investment for a brief transitional period. This process may require complementary actions for a partial refinancing and/or rescheduling of the public sector’s external debt in such a way as to avoid a deterioration in Turkey’s hard-earned creditworthiness in the international financial markets. Finally, toward the end of the 1980s, Turkey’s public finance system requires further qualitative changes and improvements aiming at noninflationary methods of revenue mobilization, deeper SEE rationalization (including gradual privatization), and streamlining of extrabudgetary funds and local government finances. While seeking a greater macrolevel flexibility in fiscal policy, efforts may also be usefully directed to the redesign of budgetary methods for a more efficient allocation of public resources.

9

External Financial Relations and Debt Management

Since the early 1970s, Turkey’s relations with the international financial community have been consistently out of synch with those of most other highly indebted countries. Turkey entered its debt crisis in 1977, at a time when a general crisis was still far off in the horizon. Its recovery and export boom in 198 1 coincided with increasing difficulties experienced by debtors in Latin America and elsewhere. In 1982, just as the rest of the developing world became engulfed in a debt crisis and new flows from commercial banks dried up, Turkey reentered private international capital markets. Since

751

TurkeyKhapter 9

then, Turkey has emerged as the international financial community’s only major example of a successful recovery from debt crisis-“a Baker Plan country before the Baker Plan,” as the former secretary of the treasury is reported to have told Prime Minister Ozal. Previous chapters have stressed the role of the domestic policies behind the debacle of 1977 and the adjustment of the 1980s. In the present chapter we turn to some specific aspects of Turkey’s external financial relations which have played a critical role. Foremost among these is Turkey’s important geopolitical role in the Middle East and as a NATO member bordering on the Soviet Union. We will emphasize that after 1979 this has been critical in mobilizing Western support-in terms of both debt relief and new flows-of a magnitude not experienced in any other recent case. The enthusiasm of official lenders has also affected the policies of the IMF and the World Bank, which showered Turkey with generous amounts of program lending. This chapter provides some detail on these flows and discusses the various rounds of debt rescheduling that took place between 1978 and 1982. The narrative here throws into sharper relief the fact that Turkey was spared the necessity, thanks to the combined effects of debt relief and new official inflows, of going through as intense a squeeze on the current account as most other heavily indebted countries have experienced since their debt crisis. This, in turn, greatly facilitated the implementation of a wide range of economic reforms as discussed in chapter 4,as well as setting the stage for a recovery. We conclude the chapter with a discussion of some debt-management issues which have plagued the Turkish macroeconomy. We will stress here the recumng penchant of policymakers to rely on rather exotic borrowing arrangements with relatively short-term maturities and high premia over international rates. These have not served Turkey well in the past and there are some questions as to whether they will do so in the future.

9.1 Official Assistance since 1977 It is instructive to start with a look at the overall volume of external financial flows in the aftermath of the 1977 crisis. Between mid-1977 and 1982, Turkey was effectively cut off from private capital markets.’ As reserves had already been substantially depleted, current account deficits had to be financed exclusively by official inflows (and initially to some extent by payments arrears, which show up as short-term credits in official statistics). An interesting exercise is to compare the net inflows to Turkey in its post-crisis period with the analogous figures for other highly indebted developing countries after 1982. Table 9.1 displays the current account and noninterest current account for Turkey and a group of seventeen other highly indebted developing countries. The years after 1977 and after 1982, respectively, provide the relevant post-crisis periods for purposes of comparison. It

752

Merih Celasun and Dani Rodrik

Table 9.1

Post-Crisis Comparisons Turkey (% of GNP)

17 Highly Indebted LDcs (% of GDP)

Yea1

CA deficit

Noninterest CA deficit

CA deficit

Noninterest CA deficit

1911 1978 1979 1980 1981 1982 1983 1984 1985

7.1 2.6 2.1 5.5 3.5 2.2

6.4 1.9 1.3 4.4 1.5 -0.5 0.7 -0.4 1.4

5.9 I .9 0. I 0. I

0.8 -2.8 -4.8 -4.2

3.5 2.8 I .9

~

Sources: For Turkey, OECD and SPO; for others, Choksi (1986)

is immediately clear from the figures that the net resource transfers to Turkey were substantially larger during the first few years after its debt crisis than were the corresponding transfers to the other countries. During the first four years after the crisis of 1977 (1978-81), external flows allowed the Turkish economy to run an average current account deficit of 3.4 percent (of GNP) and an average noninterest current account deficit of 2.3 percent. In fact, the Turkish noninterest current account turned positive for the first time only in 1982, a comfortable five years after the original crisis. The contrast with the post-1982 experience of other countries could not be starker. During 1983-85, these countries experienced a sharp reduction in their current deficits to an average of 0.7 percent (of GDP) and a drastic turnaround in their noninterest current account balance to an average surplus of 3.9 percent. It is hard to avoid the conclusion that Turkey was treated more favorably by the international financial community than were the post-1982 cases of near-default. Table 9.2 provides a closer look at the kinds of financial assistance that were made available to Turkey in the immediate aftermath of the 1977 crisis. For the table we calculated a hypothetical figure for each year, corresponding to the total amount of external financing needed. This is taken to be the sum of the current account balance (before debt relief) and of principal amortization on long-term debt, and adds up to $12 billion over 1978-81.* The gap was closed by debt relief granted through the reschedulings undertaken by the OECD consortium for Turkey (29 percent); bilateral official assistance (26 percent); and lending from multilateral institutions, mainly the World Bank and the IMF (22 percent). In addition, there were hidden private inflows-which do not show up in table 9.2 on either side of the ledger-insofar as maturing short-term debt owed to private creditors (e.g., CTLDs) was rolled over and then rescheduled.

753

TurkeyKhapter 9

Table 9.2

Financing the Balance of Payments, 1978-81 (in million $) 1978

Noninterest current account Interest payments (before relieP) Current account balance (before reliefa) Debt repayment (before relieP)

- 676

- 403

- 489

- 1,010

- 1,165

- 1,413

-451

- 945

1.616

-

Total financing a. Debt relieQ lnterest Principal b. Multilateral lending (excl. IMF) of which: IBRD c. IMF (net) d. Bilateral lending of which: special assistance under OECD auspices e. Othe?‘ of which: short-term credits change in reserves

1979

1980

1981

- 2,270

?h Financed

- 476

-1,138

- 1,443

-3,408

- 1,919

- 1,556

- 1,185

2,358

4.964

295 100 195 236 173 213 356

924 464 460 390 280 11 75 1

1,450 470 980 507 313 422 1,078

850 250 600 655 454 268 894

0 516

225 282

996 1,507

315 437

42 I

194 76

216 -512

- 263

- 187

1978-81

3,104 -

12,042 -

100.0

3,519

29.2

1.788

14.8

914 3.079

7.6 25.6

-212

Sources: Central bank, World Bank, and OECD.

’Refers to debt relief granted by OECD consortium. bIncludes short-term credits, change in reserves, FDI, and errors and omissions.

These were the outcome of a complicated series of negotiations and initiatives on the part of Turkey and its creditors after 1977. The negotiations involved a wide circle of concerned parties: commercial banks, creditor governments, foreign export suppliers, the IMF, and the World Bank. They soon turned into “a tangled web of condition and precondition with the conclusion of an agreement with one creditor group wholly dependent upon implementation of the measures by the ~ t h e r . ”Commercial ~ banks, in particular, insisted that Turkey submit to IMF tutelage before they would reschedule the CTLDs, and dragged their feet until the Turkish government gave in. The next two sections will discuss more fully some of the key components of this exceptional experience and the underlying political environment that influenced it. We will first review the string of debt renegotiations that took place during 1978-82. Next, we will turn to the official and multilateral flows that have taken place since 1977. 9.2 Debt Renegotiations

At the end of 1977, more than half of Turkey’s foreign debt consisted of short-term liabilities, most of which had been incurred under the CTLD scheme and, increasingly after 1976, as a result of payments arrears to

754

Merih Celisun and Dani Rodrik

commercial suppliers (see table 2.10). Long-term liabilities to public and publicly guaranteed sources amounted to only 38 percent of the total. Between 1978 and 1982, Turkey experienced an extensive restructuring of its debt, both long- and short-term, in a series of debt renegotiations which were the largest undertaken to date by the international financial community. By March 1982, practically all short-term debt had been consolidated and transformed into long-term debt. In addition, all of the principal payments on OECD-guaranteed debt falling due between May 1978 and June 1983 had been rescheduled. Such was the scope of these renegotiations that by year-end 1981 no less than 70 percent of Turkey’s total debt (exclusive of liabilities to multilateral organizations) had been rescheduled. Table 9.3 summarizes these renegotiations and the terms agreed upon. With respect to official creditors, there were three consecutive arrangements with OECD governments involving bilateral loans and guaranteed export credits amounting to a total of $ 5 . 5 b i l l i ~ n The . ~ terms of the successive agreements reveal a general relaxation over time with respect to grace and maturity periods, a consequence of the deepening economic crisis in Turkey as well as of a fundamental change in OECD attitudes toward Turkey after 1979 (on which more below). The last of the three agreements was the widest ranging in scope. Signed in July 1980, it entailed a multiyear rescheduling (unusual in official debt renegotiations) and consolidated payments falling due during the next three years. With regard to private creditors, two arrangements stand out. The more important of the two was the consolidation and rescheduling of $2.3 billion of CTLDs, constituting more than 90 percent of outstanding CTLD liabilities to foreigners. These CTLDs were converted into long-term debt and tied to a schedule of repayment on terms indicated in table 9.3. With this arrangement, the Turkish central bank shouldered the CTLDs as its own liability, effectively completing a socialization process which had begun de facto with the exchange g ~ a r a n t e e .The ~ negotiations with the group of eight banks, representing some 220 individual creditor institutions, proved to be long and arduous.6 In the end, the Turkish side was unsuccessful in obtaining the banks’ agreement on two key demands: (1) no IMF supervision, and (2) longer grace and maturity periods than the three and seven years, respectively, eventually settled on. On account of the first sticking point, the final agreement with the banks was delayed for a year while Turkey’s existing standby with the Fund went awry. The banks signed the agreement only when a new standby arrangement came into effect in July 1979. Meanwhile, the banks came up with a new syndicated loan of $407 million. Three years later (in March 1982) they also agreed to modify the original arrangement to somewhat improve the maturity profile (see table 9.3). The $1.2 billion of “suppliers’ arrears” settled in April 1980 was an altogether different story. Ostensibly these were arrears on nonguaranteed trade credits extended by foreign exporters. They resulted from the inability

755

TurkeyIChapter 9

Table 9.3

Date Official Creditors May 1978

Summary of ’hrkish Debt Renegotiations, 1978 to Present

Creditor

Type of Debt

Amount Renegotiated (million $)

Terms”

OECD governments

bilateral loans, M&ST insured export credits

1,300

3 yrs grace, 5 y n maturity (for 80% of M< debt); 2 yrs grace, 4 yrs maturity (for ST debt); 4.7%-7.5%

May 1978 July 1979

Iraq OECD governments

oil debt bilateral loans, M&ST insured export credits

312 1,200

July 1980

OECD governments

bilateral loans, M&ST insured export credits

3 .OOo

interest free 4 yrs grace, 5 yrs maturity (for M< debt); 3 yrs grace, 4 yrs maturity (for ST debt); 5.2’+8.7% 5 yrs grace, 10 yrs maturity; (4 yrs grace, 8 yrs maturity for previously rescheduled loans)

commercial banks

banker’s credits

commercial banks suppliers

convertible TL deposits nonguaranteed suppliers’ arrears

August 1981

commercial banks

March 1982

commercial banks

third-party reimbursement claims convertible TL deposits

Private Creditors June 1979

August 1979 April 1980

TOTAL RENEGOTIATED

429

3 yrs grace, 7 yrs maturity; LIBOR + 1.75

2,269

same as above

1,200

4% yrs grace, 10 yrs maturity, interest S 8%; option of payment in TL for specified uses no grace, 3 yrs maturity; LIBOR 1.50

100

(2.269)

+

renegotiation of the August 1979 agreement: 2-yr extension on grace & 3-yr extension on maturity

9,810

Sources: World Bank (1983a). vol. 2 (3 June 1983). table A2.7; and IMF (1983), tables 7 and 12

’These are the key features of the agreed terms. Some of the agreements (notably those with the OECD) contain additional terms which apply to certain small portions of the consolidated amounts.

of the central bank to undertake the necessary foreign exchange transfers to foreign exporters whose goods had already been received, and for which the local importers had already deposited with the central bank the requisite amounts in domestic currency. In reality, few foreign exporters were willing to extend nonguaranteed credits to Turkish importers at a time when Turkey was in deep financial trouble; they naturally demanded payment up front. And, by all accounts, most foreign suppliers had indeed been prepaid by domestic importers, who had obtained the foreign exchange from the black market. Given the widespread shortages of the period (see chap. 3),

756

Merih Celasun and Dani Rodrik

importing was profitable even if it meant paying for it twice: once to make the requisite deposit with the central bank, and the second time around to purchase the foreign exchange on the black market. The authorities were not unaware of what was going But as long as it helped imports flow in, it became a convenient fiction for the authorities to suppose that the central bank’s growing foreign-currency arrears were to foreign suppliers rather than to domestic importers. The realization that these “suppliers’ arrears” were predominantly liabilities to domestic residents encouraged the government to propose an imaginative approach to settling them. Under a plan put forth in early 1980, the creditors were given two options. They could elect to receive payments in foreign currency, in which case they would be reimbursed over a ten-year period with a 4%-year grace period at interest rates depending on the currency but not exceeding 8 percent. Or, they could choose immediate payment in Turkish currency for purposes of specified investments and other activities within Turkey, at varying rates of discount depending on the activity selected. This early form of debt-equity swap turned out to be relatively successful. Roughly half of the outstanding liabilities were eventually redeemed in domestic currency.

9.3 Official Flows and Changing Western Attitudes While these complicated debt restructurings were taking place, the attitude of Western governments toward Turkey was also undergoing a transformation. Initially, the American and European governments had showed little interest in the economic crisis brewing in Turkey. For reasons discussed below, the attention devoted to the Turkish crisis grew over time, culminating in a major rescue operation launched in 1979. Until late 1978, relations between Ankara and the Western capitals were dominated by military considerations. Foremost on Prime Minister Ecevit’s agenda was the lifting of an arms embargo imposed by the U.S. Congress in the aftermath of the Turkish landing on Cyprus in 1974. To underscore his seriousness, Ecevit initiated a policy of rapprochement with the Soviet Union. Alarmed by the implications for the southern flank of NATO, President Jimmy Carter made the repeal of the embargo his highest foreign policy priority by mid-1978, and his efforts were successful in October of that year. Insofar as foreign relations were concerned, the arms embargo and the importance attached to it by both sides had overshadowed the economic crisis in Turkey. But with the arms issue out of the way, the economic dimension began to attract increasing attention in Western capitals. Further, events elsewhere in the Middle East fortified the perceived importance of Turkey. The Iranian revolution and the eventual fall of the Shah, in particular, served to concentrate the collective mind of the Western alliance

757

TurkeyIChapter 9

on Turkey as it had never been before. By the beginning of 1979, it became a commonplace assessment that “[tlhe strategic importance of Turkey . . . is too great for Ankara’s fate to be left to the [International] Monetary Fund and commercial banks abroad.”9 As a New York Times editorial succinctly put it, “Turkey is now the only clearly pro-Western state between the Soviet Union and the Middle East; it guards the straits between the Soviet Black Sea fleet and the Mediterranean and offers the main remaining land site from which electronic intelligence bases can monitor Soviet missile-test launchings” (22 January 1979). The Soviet invasion of Afghanistan in early 1980 added further urgency to Western efforts. In early January 1979, a rescue operation was started by the Big Four (U.S., West Germany, France, and Great Britain) in their Guadeloupe summit. After much hard work by the Americans and Germans behind the scenes, the OECD countries agreed in May to pledge close to $1 billion in bilateral assistance. This included fast-disbursing emergency loans (at low interest rates and with long repayment periods) as well as special trade credits to finance Turkish imports. The catch was that the loans were conditional on Turkey’s acceptance of a new IMF standby arrangement to replace the earlier one (of April 1978) which had proved unsuccessful. With the OECD arrangement and the CTLD restructuring with commercial banks both hinging on an IMF program,’o Ecevit finally succumbed in June and undertdok the major policy change advocated by the IMF: a devaluation of the currency. This cleared the way for a standby arrangement in July 1979 and for the OECD funds. Further OECD assistance was pledged in subsequent years. As table 9.2 showed, the impact of the OECD program was felt most heavily in 1980 when $1 billion flowed in, providing essential support in the wake of the second oil crisis. The OECD program also played a role in triggering additional flows to Turkey. As is shown in table 9.4 summarizing all medium- and long-term commitments received in 1979-81, Turkey became the recipient of flows from such diverse sources as the European Investment Bank (EIB), OPEC (principally Saudi Arabia), and the centrally planned economies (CPEs). The total medium- and long-term commitments received in 1979-81 were on average twice as large as in the earlier 1975-78 period, and public commitments were three times as large. As the title of a contemporary news account in the Stuttgarter Zeitung put it bluntly, “Ecevit turn[ed] Turkish geography into dollars.”” The generous mood of the OECD governments was also reflected in the policies of the World Bank and IMF. From 1980 on, Turkey became the recipient of exceptional flows from these two institutions. Alongside its regular project credits, the World Bank extended five consecutive SALs totaling $1.6 billion, the largest number of such loans ever made to a single country. These loans, containing mild levels of conditionality, were made in support of the economic reforms undertaken since January 1980

758

Merih Celasun and Dani Rodrik

Table 9.4

Medium- and Long-Term Commitments, 1975-81 (million $) 1975-78 Annual Average

Official sources Bilateral OECD OPEC CPEs Multilateral World Bank EIB ERF Other Private sources of which: syndicated loans Total

623 360 226 11 124 263 224 10 25 4 494 131 1,117

1979

1980

1981

1,131 659 596 54 9 472 306 112 39

2,683 1,671 1,138 288 245 1,012 616 27 1 104 21 299 0 2,982

1,799 988 49 I 56

15

634 407 1,765

441

811 570 55 110

76 249 0 2,048

Source: World Bank (1983a), vol. 2 (3 June 1983), table A2.5 Note: CPEs are centrally planned economies; EIB is the European Investment Bank; and ERF is the European

Reconstruction Fund.

(see chap. 4). The IMF entered into a three-year standby arrangement in June 1980 for a total of SDR 1.25 billion (table 9.5). This amounted to 625 percent of Turkey's IMF quota at the time, and together with previous purchases brought total IMF commitments to Turkey to 870 percent of quota, the largest multiple awarded by the IMF until then. Indeed, many executive directors of the Fund felt uneasy about the special flexibility shown to Turkey and the speed with which such resources were made available, at a time when Turkey was one of many countries experiencing economic difficulties. It is hard to judge the extent to which these multilateral flows were directly influenced by the overall politicaVstrategic importance placed on Turkey by the OECD countries. Whatever that influence might have been, it Table 9.5

Date

Chronology of Standby Arrangements with the IMF, 1978-85

Duration Envisaged (yrs)

Amount (million SDR)

% of Quota

Comments

300 250

150

not drawn fully SDR 230 million drawn; replaced by next arrangement fully drawn cancelled at request of new government after elections of November 1983 replaces previous standby; last purchase (one-fourth of total) not made

April 1978 July 1979

2

June 1980 June 1983

3 1

1,250 225

625 75

April 1984

1

225

52

1

125

759

TurkeyKhapter 9

is clear that the World Bank and the IMF became especially enthusiastic as a result of the signals sent by Turgut Ozal, who was appointed in December 1979 by the incoming prime minister Demirel as his deputy in charge of economic affairs. A former staff member of the World Bank, Ozal appears to have been intent on showing his seriousness about economic reform. Hence the devaluation and price hikes announced on 24 January 1980 were rumored to have been in excess of what the IMF was willing to settle for. The Fund is reported to have advocated a depreciation of the lira to somewhere in the range of TL 60-70 to the dollar (from a rate of TL 47.10 for most transactions), with some up-side padding presumably added to leave room for negotiation. Ozal’s strategy was to distinguish himself clearly from previous Turkish negotiators by picking the top of the range (TL 70 = $1). The visiting IMF chief of mission is reported to have been ecstatic upon hearing of Ozal’s intentions (Colagan 1983, 61- 139). Since the January 1980 reform package, the enthusiasm and support of the Bretton-Woods institutions for Turkey has been unfailing. This despite an awkward episode in 1983 in which the Turkish authorities were caught doctoring monetary statistics in order to fulfill IMF ceilings. In essence, the central bank was found to have instructed the state-owned Agricultural Bank to make cash payments to it at the end of each week, with the transaction reversed at the beginning of the next. Since the IMF calculated the domestic credit extended by the central bank by averaging figures for domestic assets each Friday, this enabled the authorities to exceed the IMF’s ceilings and give the economy a boost on the eve of the general elections of November 1983. The IMF staff eventually became suspicious as the calculated money multiplier started taking odd turns. The episode took place after Ozal had resigned from the government, so he was not directly implicated. In fact, since it was Ozal’s party that won the elections and he became prime minister, the IMF decided to keep quiet about the transgression and to give the new government the benefit of the doubt.” It is scarcely in the interest of the Fund to make waves: after years of strong support and having hailed Turkey as a success story, the IMF and the World Bank both have a substantial interest at stake in seeing that the Turkish economy indeed sails smoothly. 9.4 Debt Management Turkey’s historical experience with private international capital markets has not been a happy one. A borrowing binge during the nineteenth century had left the public finances of the disintegrating Ottoman empire almost completely under the control of foreigners. The memory of this event partly accounts for why Turkish authorities invited practically no foreign lending by private sources until 1975. But as we have discussed in chapter 2, the resort to private capital markets came with a vengeance after this date. The CTLDs

760

Merih Cellsun and Dani Rodrik

were the worst possible kind of borrowing: they were short term, high cost, and encouraged overborrowing. By the very nature of the scheme, the authorities were unable to control the level of borrowing that took place. The irony is that at the time Turkey could well have relied on more commonplace (and safer) means of borrowing, such as syndicated bank loans. l3 That it did not is explained partly by the lack of recent experience with private lenders and partly by the false sense of security generated by the pace at which CTLD funds were flowing in. Since the debt debacle of 1977, Turkish authorities have paid considerably more attention to issues of debt management. The oversight agency for public sector debt is the Treasury, which has reported to the prime minister’s office (rather than the finance ministry) since December 1983. The Treasury is empowered to set targets for overall flows in light of balance-of-payments and investment requirements. All direct foreign borrowing by the public sector is subject to the approval of the Treasury, as is all borrowing by the private sector exceeding two years’ maturity. l4 Responsibility for monitoring debt flows rests with the Treasury (for medium- and long-term debt) and the central bank (for short-term debt). An ongoing computerization project promises to make aggregate data and information available to policymakers in a more timely fashion than has been possible so far. Yet in many ways the debt-management issues of the 1980s are little changed from those of the 1970s. Two such issues deserve special emphasis. First, there is a recurring tendency to rely on special lending arrangements which are both short term in nature and relatively costly. The period since the debt restructurings of 1978-82 has seen a renewed rise in the share of short-term debt. Part of this rise is due to the liberalization of capital account flows and the greater reliance of private banks and enterprises on short-term foreign credits. But an important part is due to the increasing importance of the Dresdner Bank scheme, which accounted for 40 percent of all short-term debt by the end of 1985.15 This scheme is an arrangement whereby the Dresdner Bank makes available to the central bank deposits made by Turkish workers abroad. To attract such deposits, the central bank has been paying the Dresdner Bank-and ultimately the Turkish workers-fairly high interest rates. As of the summer of 1986, the rates on two-year deposits were 12 percent on dollar accounts and 11 percent on deutsche mark accounts. (Prior to January 1985, the rates had stood at 14 percent for all currencies.) This implies a spread over LIBOR of 4-5 percentage points, well above what most other developing countries have to pay for syndicated loans. Hence, preexisting long-term debt gets serviced by being transformed into substantially more expensive short-term debt. Interestingly, the highest spread witnessed during the recent spate of reschedulings has been 2.25 percentage points. In principle, then, Turkey could be better off simulating a debt crisis than continuing servicing debt in the present fashion. However, short of another round of debt reschedulings, the Dresdner scheme provides

761

TurkeyKhapter 10

a useful financing function as its cost is still modest compared to domestic borrowing. A second and related issue has to do with the important role played by the central bank, and of the banking sector in general, in debt accumulation. The CTLDs of the 1970s and the Dresdner Bank accounts (as well as various balance-of-payments loans) of the 1980s have ultimately been the liability of the central bank. The advantage of these kinds of borrowing is that they provide a degree of latitude in their use which project credits do not allow. But this may also be a disadvantage to the extent that they allow a disjuncture between decisions on debt accumulation on the one hand, and decisions on resource allocation and investment patterns on the other. Whether this is dangerous or not depends on how finely tuned the central bank is to the investment possibilities in the public and private sectors. These arrangements have an additional consequence: they tend to bias the debt-servicing process toward money creation rather than public sector budget adjustments.

9.5

Concluding Remarks

This chapter has focused on Turkey’s external financial relations in the aftermath of the crisis of 1977. The importance of the support provided during this period by the international financial community (mostly OECD governments and eventually the IMF and World Bank) cannot be underestimated. No other country has been the beneficiary of comparable amounts of financial assistance. We have argued here that the West’s concern with the Turkish economy was at heart strategic; as one foreign banker colorfully put it, “supranational agencies such as the IMF, as well as Western governments, showed little interest in Ankara’s financial difficulties until Turkish real estate suddenly became more valuable to NAT0.”l7 In this key respect, Turkey’s adjustment experience is likely to prove nontransferable. Of course, this qualification does not reduce the importance of the domestic policies undertaken since January 1980, nor does it diminish their relative success. But it puts the experience into a proper perspective.

10

Conclusions and Prospects

In many ways, the Turkish encounter with foreign debt has combined the best and worst in the debt-management experience of the developing world. During the 1970s, Turkish policymakers got the country into a debt crisis by relying on an intrinsically destabilizing form of foreign borrowing, and

762

Merih Celasun and Dani Rodrik

ensured that the adjustment to the crisis would be extremely painful by following policies that consistently discouraged exports. The tail end of the decade witnessed a typical pattern of import compression and inflation, spurred by the authorities’ unwillingness to undertake serious adjustment measures. The 1980s,on the other hand, have been a period of recovery and regained creditworthiness. At the policymaking level, the gains made since 1980 can be credited in large part to Turgut Ozal, who as deputy prime minister pushed for the radical adjustment package of 1980 and who as prime minister consolidated the outward orientation of the economy after 1983. In this, he was assisted both by the special freedom provided to the technocrats under the military rule of 1980-83 and by the exceptionally generous official capital inflows stimulated by Turkey’s important geopolitical role. Without these two enabling circumstances, we doubt that Ozal’s program could have been carried out to its fruition. This is an important cautionary note for those who believe that Turkey’s experience can be easily transplanted in other contexts. A convenient way to summarize some of our arguments in previous chapters is via the perspective provided by key debt indicators. Table 10.1 displays the trends in debt/GNP, debuexports, and debt-service ratios over the 1973-86 period. A quick glance at the numbers for the mid-1970sshows what was wrong with the debt strategy at the time. By 1977 the debVGNP ratio had almost doubled to 27 percent from 15 percent in 1973. But this in itself was a rather undramatic rise, given the small initial base. What proved

Table 10.1

Principal Debt Indicators, 1973-86 (in percentages)

Year

DebVGNP Ratio”

DebVExports Ratiob

Debt-service ratio‘

1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 I986

14.9 11.7 13.9 19.0 27.4 32.7 31.2 27.8 28.6 32.8 36.1 43.4 47.8 53.1

207.2 178.4 234.0 286.5 491.9 507.2 479.1 441.5 280.2 222.2 231.4 218.2 223.3 278.8

26.0 23.1 69.1 133.3 289.0 275.3 154.5 101.7 66.0 52.7 60.4 60.2 73.4 103.3

Sourcrs: Central bank. SPO

=Debt is converted into national currency by using period-average exchange rates: black market rates for 1974-79, and official rates for 1973 and 1980-86. bExports refer to exports of goods and services. ‘Ratio of interest payments, short-term debt (with less than one-year maturity), and amortization payments on medium- and long-term debt to exports of goods and services.

763

TurkeyKhapter 10

disastrous in the end was a sharp deterioration between 1975 and 1977 in the ability to service the new liabilities: the debt-service ratio quadrupled in two years from 70 percent to an incredible 290 percent. This was partly the consequence of the anti-export bias of the growth policies of the period. But more importantly, it was the result of a debt strategy which, by providing a blanket guarantee against exchange losses, encouraged the private sector to incur as many short-term liabilities as possible, as quickly as possible. During the two-and-a-half years of muddling through which followed the crisis, the economy came crashing down, inflation accelerated, and income distribution worsened considerably as unprotected sectors became the casualty of the rise in prices. Yet as table 10.1 shows, the debt-service ratio started to descend quite rapidly from its peak in 1977-78. Before the January 1980 reform package was announced, this ratio had already come down to 155 percent, and fell further to 102 percent in 1980 before the export boom had gotten under way. This drastic improvement in the economy’s ability to service its debt was of course not the consequence of adjustment policies, which were quite lacking prior to 1980. The trick was performed by a series of debt reschedulings, which reduced amortization payments substantially, and by the conversion of short-term liabilities into long-term debt. The export boom starting in 1981 took another 40-50 percent off the debt-service ratio, but this reduction looks rather unimpressive in comparison with the one accomplished by debt renegotiations. To be sure, these debt renegotiations served only to postpone the servicing of the existing debt, Together with the new borrowing of the early 1980s, the reschedulings have now come back to haunt the Turkish economy, requiring ever-improving export performance just to maintain the debt-service ratio level. In fact, as table 10.1 shows, the principal debt ratios have witnessed a marked worsening since 1982. The debt/GDP ratio now stands at an all-time high, and the debt-service ratio has inched its way up to around 100 percent. Both the debVGDP and debvexports ratios are currently higher in Turkey than they were in the heavily indebted Latin American countries just prior to their debt crisis (the Turkish debt-service ratio looks better, however). This renders the balancing of exports, creditworthiness, and sustained growth a very delicate high-wire act. In a way, the statistics overstate Turkey’s external debt. A considerable share-amounting to 16 percent at the end of 1986-of the “foreign” debt actually constitutes a liability to Turkish workers who reside abroad.2 But, accounting conventions aside, we do not think that this fact makes much of a difference in practice. It would be a mistake to consider these liabilities as “safer” and more reliable than conventional forms of external indebtedness. As remittance behavior over the last two decades has shown, Turkish migrant workers are quite sensitive to overall macroeconomic conditions in Turkey. Hence, they are unlikely to keep rolling over their deposits if confidence wanes. And any difficulty in servicing these foreign currency liabilities would likely send a disastrous signal to Turkey’s genuinely foreign creditors.

764

Merih Cellsun and Dani Rodrik

How will Turkey manage the debt-service hump? Its debt burden could be eased substantially by another round of reschedulings or by refinancing existing liabilities and converting short-term debt into longer term debt. However, the current global environment is not particularly encouraging about the prospects for this. Few other countries have traveled as far down the adjustment path as has Turkey, and hence are as deserving of favorable treatment. Ironically, this very fact makes it more difficult for Turkey to openly seek relief hard-earned creditworthiness is at stake. Nonetheless, it is to the advantage of all concerned-including foreign creditors and multilateral institutions-that Turkey’s model prove a viable one in the longer run. For this reason, common interest may well dictate that some workable arrangements be devised for reducing Turkey’s debt-service burden. It is scarcely fair that a “successful” country should be servicing its rescheduled debt at several points above LIBOR while problem debtor countries are rescheduling theirs at a margin less than 1 percent over LIBOR. Irrespective of any rescheduling or refinancing, how Turkey comes out in the short to medium run will depend overwhelmingly on two aspects of economic performance: exports and fiscal balance. The continued servicing of the debt will require both generating sufficient foreign exchange and improving resource mobilization in the public sector, whose liability foreign debt primarily is. The dilemma is that many of the current policies appear to be working at cross purposes with respect to these two goals.

10.1 Export Performance So far, Turkey’s export performance has confounded the export pessimists. The question that remains is the extent to which the export boom represents a genuine structural transformation and a permanent increase in the economy’s capacity to generate foreign exchange. Questioning the permanence of a boom which has now been going on for more than six years (with a temporary-it appears-setback in 1986) may seem ungracious, yet we have pointed out at several junctures some unsettling aspects of the export performance to date. We reiterate two of these here. First, maintaining the export boom has required a continuous process of real depreciation of the Turkish currency and alongside it an explicit and generous program of export subsidization. Neither of these two, continuous, real depreciations and subsidies is a policy option that can continue to be exercised without damaging consequences elsewhere. The subsidies themselves, even leaving aside the rather large overinvoicing to which they have given rise, are in clear contradiction of Turkey’s obligations under the GATT and will quite evidently attract restrictions on market access by the leading importers as Turkey’s exports become more important. Subsidies are also costly to the budget; they therefore render public sector resource mobilization, the second desideratum alongside export performance, more problematic.

765

TurkeyKhapter 10

As to the policy of real exchange rate depreciations, it is an impractical solution for the longer run. By building a real interest rate premium on Turkish assets, a continuous and expected real depreciation raises the cost of capital and ultimately defeats the purpose of export expansion by choking private investment in tradables. In addition, just like export subsidies, it increases the burden of debt servicing on the public sector budget: everything else being equal, a decrease in the real value of the Turkish lira increases the debt-service/GNP ratio and requires a correspondingly smaller public sector deficit (as a share of GNP) to finance it. Finally, real depreciations are ultimately deleterious to distributional goals. While in theory the link between the real exchange rate and real (consumption) wages is ambiguous, the Turkish experience suggests that in practice it is likely that currency depreciations have to be validated by real wage cuts in order to yield the desired effect on competitiveness. From the social viewpoint, this is an undesirable policy given the magnitude of real wage cuts that have already taken place. The second aspect of export performance we want to highlight is the evident absence of private investment in tradables that underlies it. So far, the export boom has come from existing capacity; there was considerable room for output expansion given the low rates of capacity utilization in 1980. The continuation of the export drive clearly requires new investment in tradables. The government has pinned its hopes on the private sector’s ability and willingness to provide the necessary capital accumulation. While private investment in manufactures has recovered somewhat from its trough in 1980-82, it is still well below its level in 1976-77. Disconcertingly (for exports), much of the recent increase in private investment has been geared toward housing. The overall sluggishness in investment performances is at least partly related to the excessively high real rates of interest currently prevailing. The latter in turn is the consequence of the fact that the fiscal balance is still out of control (see be lo^).^ In sum, policy geared toward export promotion will have to start relying less on exchange rate and subsidy policies, and more on increases in private capital formation. A critical prerequisite, then, is to reduce the real cost of credit to the private sector, an outcome that can be achieved only if the demands of the public sector on the economy’s resources can be moderated. 10.2 Fiscal Balance As our argument above indicates, maintaining an appropriate fiscal stance is not only important for generating sufficient resources with which to service the public sector’s debt, but is also crucial for avoiding the crowding out of private investment in export-oriented sectors. So far, the Turkish adjustment experience has been characterized by an undistinguished performance in terms of public sector resource mobilization. While some of

766

Merih Cellsun and Dani Rodrik

the reforms undertaken since 1980 have been long needed-e.g., the rationalization of public sector prices, adjustment of tax brackets against inflation, introduction of the VAT-the overall retrenchment in the public sector deficit has not been large, a fact which finds its counterpart in the continuing need for moderate amounts of foreign borrowing. In many ways, the public sector balance remains the Achilles’ heel of the Turkish economy. A given public sector deficit can be financed in a noninflationary manner in only one of two ways: domestic or foreign borrowing. As we have pointed out in previous chapters, both of these forms of borrowing are now extremely costly. Despite its newfound creditworthiness, the amount of voluntary lending flowing into Turkey is not substantial, and the Dresdner Bank accounts used to draw in Turkish workers’ savings from abroad pay substantial premia over Euromarket rates. Domestic borrowing, in which the government liberally indulges, is even more costly, at real rates of interest far exceeding the growth rate of the public sector’s revenue base. This last aspect, in particular, raises serious questions about the sustainability of the current strategy. By all indications, then, reducing the public sector deficit is going to be the main challenge to policymakers in the years ahead. Not all of the adjustment here need be borne by public expenditures. A successful outcome would involve both decreased expenditures and expanded tax revenues. The experience in the early 1980s has shown the difficulty of resource mobilization via taxation, but we suspect that a broader based tax effort aiming at, among other things, the enlarged profit margins in many services sectors has considerable promise.

10.3 Income Distribution Improving the fiscal balance is likely to clash head on with an issue that we expect to become increasingly important over the next few years. As we have highlighted throughout our account, the Turkish experience with income distribution has been singularly disappointing ever since the onset of the debt crisis in mid-1977. What makes this experience even more striking is that it took place after a period during the 1970s in which real wages and rural incomes had more than amply shared in the spoils of economic growth. As the political system opens up, a process which by and large was consolidated with the referendum of September 1987, we expect that distributional issues are going to become increasingly important in the political agenda. An unavoidable question is the nature of the link between the adjustment policies of the 1980s and the adverse trends in distribution. The first point to be made in this connection is that, despite a widespread impression to the contrary, the deterioration in income distribution started in 1978 and had already become seriously entrenched by the time the 1980 reforms were

767

TurkeyiChapter 10

taking effect. As we analyzed above, this initial sharp deterioration was a consequence of the lack of adjustment policies. Inflation triggered by the external constraint and the shortages wiped out the real incomes of the least-protected sectors of the economy. The sad irony was that the half-hearted nature of the pre-1980 reforms was the result of a concern that doing more might have jeopardized the distributional gains of previous years. It is true nonetheless that the post-1980 policies did not exactly have salutary effects on distribution either. Some of the distributional trends at that time can be linked to the military’s role in freezing the factor shares inherited in late 1980. It is also clear that distributional issues were of secondary importance to Ozal compared to economic recovery and regaining creditworthiness. In general, theory makes no predictions about the distributional consequences of adjustment policies. But as we have stressed, the very nature of the policies followed in this period, relying on sharp changes in economywide relative prices, ensured that the outcome would not be distributionally neutral. Take, for example, real wages. The reduction in real wages served a number of important purposes in the adjustment process. First, it allowed an increase in competitiveness of the traded sector. Secondly, it eased the cost pressures brought on by the high cost of credit in an already highly indebted private manufacturing sector. Third, it contributed to the improvement in public sector finances by reducing the wage bill of state enterprises. The deterioration in agriculture’s term of trade served many of the same purposes. The reduction in farm price supports and the phasing out of input subsidies enhanced public sector savings. The emphasis on the subsidization of manufactured exports, on the other hand, denied agriculture most of the gains that conventional analysis had posited would follow from outward orientation. Was there an alternative? We suspect that policies geared directly toward improving the distribution of income would have complicated tremendously the recovery effort in the chaotic conditions of 1980. Yet some of the post-1980 trends could perhaps have been avoided if the emphasis in the program had been less on relative prices and more on policies that reduced expenditures directly. In practice, what this means of course is that the scale of public sector expenditures, both current and investment, ought to have been more tightly controlled. As we argued in chapter 8, this would have been somewhat costly in terms of growth, as public investment played an important stimulating role in the absence of private investment. In addition, we know very little about the distributional implications of direct cuts in government expenditures. Nonetheless, it is quite likely that such a change in the overall thrust of the program would have eased the requirement of turning the terms of trade of workers and farmers sharply against them in order to generate savings for the public sector.

768

Merih Celisun and Dani Rodrik

In addition, it is possible that the real interest rate consequences of the post- 1980 reforms could have been moderated had policymakers been less dogmatically attached to financial and capital account liberalization. The latter reforms have not only necessitated a reorganization of factor shares at the level of firms, but they have also complicated macroeconomic management by engendering currency substitution. The relative openness of the capital account has rendered speculative attacks and capital flight a dangerous possibility, emphasizing all the more the need for a careful balance on the fiscal front.

Appendixes A. Political Chronology, 1970-87 1970 March 1971

April 1971 April 1972 May 1972

March 1973 April 1973

Suleyman Demirel’s Justice Party (JP) government under increasing strain as political violence grows. The military present the president with a memorandum threatening a takeover. Demirel’s cabinet resigns. Nihat Erim (Republican People’s Party, RPP) forms new government. Martial law proclaimed in eleven provinces. Erim resigns. Ferit Melen forms a coalition government. Biilent Ecevit takes over from Ismet Inonu as leader of the RPP. President Cevdet Sunay’s term expires. Fahri Koriiturk is elected president by the Grand National Assembly, after repeated ballots fail to generate enough support for the military’s favored candidate, Faruk Gurler. Melen’s cabinet is succeeded by one formed by Naim Talu, an independent senator.

September 1973

Martial law comes to an end, as the military greatly reduce their interventionism of the past two years.

October 1973

General elections fail to produce a majority government, but Ecevit’s RPP emerges with a plurality of seats.

January 1974

RPP forms a coalition government with the Islamist National Salvation Party (NSP) led by Necmettin Erbakan. Ecevit becomes prime minister.

769

TurkeyIAppendix A

July 1974 September 1974

February 1975 March 1975 July 1975 June 1977 August 1977

January 1978 April 1978 October 1978 December 1978

June 1979 October 1979

A coup in Cyprus led by the Cypriot National Guard prompts the Turks to land on the island. Ecevit resigns in the hope of capitalizing on his enhanced popularity in early elections. Other parties block early elections and Sadi Irmak leads a caretaker government . U.S. imposes arms embargo on Turkey. Demirel returns to power, leading a right-wing coalition. Turkey takes over U.S. bases in retaliation for the arms embargo. General elections fail to produce a majority. After an unsuccessful try by Ecevit, Demirel forms a second right-wing,coalition government. Key ministerial posts are given to Erbakan’s NSP and the ultraright-wing Alpaslan Turkes’s Nationalist Action Party (NAP). Erbakan opposes vehemently any deal with the IMF. Political violence continues to grow. Ecevit forms minority government with the support of independents and defectors from JP. Ecevit adopts a stabilization program and signs a standby arrangement with the IMF. U.S. arms embargo is lifted by Congress. Political violence culminates in the most serious outbreak of sectarian fighting in decades in Kahramanmara? (southeastern Turkey). Martial law is imposed in thirteen provinces, all in the east except for Istanbul and Ankara. New Letter of Intent is signed with the IMF. Ecevit’s RPP loses ground to JP in by-elections. Ecevit resigns, and Demirel forms a new government with the backing of NSP and NAP.

January 1980

A package of economic measures is announced to deal with the ongoing economic crisis.

April 1980

Koriitiirk’s term as president expires; Parliament is deadlocked and unable to elect a new president. IMF approves a three-year standby arrangement for SDR 1.25 billion. Controls over interest rates on bank credits and deposits are lifted.

June 1980 July 1980

770

Merih Celasun and Dani Rodrik

September 1980

The armed forces, led by General Kenan Evren, chief of the general staff, seize power in a bloodless coup. Evren becomes head of state. A new government is formed under a retired naval commander, Biilend Ulusu, but Turgut Ozal keeps his post as deputy prime minister for economic affairs.

February 1981

Tax reform law is enacted, changing brackets and rates so as to eliminate the distortions due to inflation.

May 1981 October 1981

Daily adjustments of exchange rates are instituted. The National Security Council (NSC) orders dissolution of all political parties and confiscates their assets. Turkey’s largest money broker and securities house, Banker Kastelli, collapses. Ozal resigns subsequently.

June 1982 November 1982 June 1983 November 1983 December 1983

April 1984

January 1985 February 1985

December 1985 September 1986

April 1987 July 1987

A new constitution is adopted in a referendum, and Evren is voted president. IMF approves one-year standby agreement for SDR 225 million. Ozal’s Motherland Party wins majority of the seats in the first parliamentary elections since the coup. The new government liberalizes the trade regime further and eliminates many QRs. Residents and nonresidents are allowed to open foreign exchange deposit accounts. Existing IMF standby agreement for SDR 225 million is replaced by new one for the same amount. Ozal’s party emerges as winner in local elections. Subsidized export credit is eliminated. The VAT comes into effect. Turkey signs the GATT subsidy code and agrees to phase out all subsidies on exports by the end of 1989. Martial law is lifted in Istanbul. Party supported by ex-premier Demirel (True Path Party) shows strength in by-elections by taking a strong second place behind Ozal’s party. Turkey applies to join the European Community. Ozal vetoes the military’s choice for chief of general staff.

771

TurkeyIAppendix B

September 1987

Referendum enables, by a very slim margin, pre1980 party leaders to return to political life.

B . Statistical Appendix (Tables A. 1 -A.28 on pages 772-96.)

Table A . l

Macro Balances (current prices, billions of TL) ~~

GNP Foreign savings Total resources Invesrmenr Fixed capital formation a. Private b. Public Changes in stocks a. Private b. Public Consumption a. Private b. Public Source: SPO.

~

~~

~

~~

~~

~

1972

1973

1974

1975

1976

1977

1978

1979

I980

1981

1982

1983

1984

1985

240.8 .I 240.9 48.5 46.9 26.9 20 1.6 1.7 -.I 192.4 164.4 28

309.8 -6.8 303 56.2 59.3 34.2 25.1 -3.1 .3 -3.4 246.8 210 36.8

427.1 9.9 437 88.5 76.1 41.1 35 12.4 1.4 II 348.5 301.5 47

535.8 26.9 562.7 120.3 107.9 54.1 53.8 12.4 .8 11.6 442.4 378.5 63.9

675 36.5 711.5 166.9 153.7 79 74.7 13.2 9.6 3.6 544.6 460 84.6

872.9 60.4 933.3 217.8 210.8 103 107.8 7 .7 6.3 715.5 599.2 116.3

1,290.7 34.1 1,324.8 239 279.6 143.3 136.3 -40.6 -26.5 - 14.1 1,085.8 913.2 172.6

2,199.5 46.5 2,246 402 449.3 213.8 235.5 -47.3 - 21 -26.3 1.844 1,550 294

4,433.2 244. 1 4,679.3 947.9 863.6 378.7 484.9 84.3 60.8 23.5 3,731.4 3,187.3 544.1

6,553.6 230.6 6.784.2 1,408.6 1,241.4 474.5 766.9 167.2 69.9 97.3 5,375.6 4,675.5 700.1

8,735. I 187.7 8.922.8 1,774.6 1,646.9 641.4 1,005.5 127.7 88 6 39. I 7,148.2 6.208.8 939 4

11,551 410 11,961 2,260 2,131 1,183 948 I29 - 46 175 9701 1,175 8,526

18,375 513 18,888 3,595 3,331 1,831 1.501 263 2 26 1 15,294 1,622 13,672

27,715 525 28,240 5,682 5,442 3,107 2,333 240 - 4R 288 22,559 2,332 20,227

Table A.2

GNP Foreign savings Total resources investment Fixed capital formation a. Private b. Public Changes in stocks a. Private b. Public Consumption

a. Private b. Public Source: SPO

Macro Balances (1983 prices, billions of TL) 1972

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

7,357 740.2 8,097.2 1,394.8 1,351.6 780.6 571 43.2 45.5 -2.3 6,702.4 6,122.6 579.8

7,752.4 513.3 8,263.7 1,453.2 1,527.4 881.4 646 -74.2 6.4 - 80.6 6,812.5 6,173.3 639

8325 947.1 9,272.1 1,872.7 1,65 1.7 895.8 755.9 22 1 25.2 195.8 7,399.4 6,697.5 701.9

8,987.5 1.360.6 10.348. I 2,252.6 2,05 I . I 1,038.3 1,012.8 201.5 13 188.5 8.095.5 7,297.2 798.3

9,699.4 1,888. I 11.587.5 2,650.4 2,464.6 1,266.9 1,197.7 185.8 135.8 50 8,937.1 805 I 886.1

10,076.3 2,027.6 12, 103.9 2,657.3 2,577.4 1,248.4 1,329 79.9 8.3 71.6 9.446.6 8,527.6 919

10,364.9 1,024.4 11,389.3 2,020.3 2,320.6 1,172.7 1,147.9 -300.3 - 196.5 - 103.8 9.369 8.364

10,323.3 796.8 11.120. I 2,023.4 2,236.8 1,036.2 1.200.6 -213.4 -94.9 - 118.5 9.096.7 8,073.3 1,023.4

10.213.2 771.8 10.985 2.197.3 2.01 3.4 857.2 1,156.2 183.9 132.8 51.1 8,787.7 7,677.9 1,109.8

10,636.9 520.9 I I , 157.8 2,314.3 2,047.6 782.8 1,264.8 266.7 111.5 155.2 8,843.5 7,723.6 1,119.9

11,126 343.9 10,985 2,280.9 2,118.2 825.5 1,292.7 162.7 112.9 49.8 9,189 8,046.9 1.142.1

11,485.2 475.6 11,960.8 2,375.7 2,179.4 865 1,314.4 196.3 187.5 8.8 9,585. I 8,428.1 1,157

1,005

774

Merih Cellsun and Dani Rodrik

Table A.3

Macro Balances, 1983-85 (1986 prices, billions of TL)

GNP Foreign savings Total resources Investment Fixed capital formation a. Private b. Public Changes in stocks a. Private b. Public Consumption a. Private b. Public

Source: SPO.

1983

1984

1985

33,082 733 33,815 7,113 6,744 2,818 3,926 369

35,048 622 35.670 7,563 7,062 3,066 3,997 501 497 4 28.107 25,042 3,065

36,823 435 37,258 8.155 7,836 3,306 4,530 319 383 -64 29,103 25,939 3,164

500 - 131

26,702 23,724 2,915

Table A.4

Structure of Public-Private Disposable Income and Savings, 1973-83

A IGNP, current pnces 1 Public disposable income 2 Pnvate disposable income B %GNP, 1983 constant pnces I Public sector a Disposable income b Consumption c Savings d Investment e Savings investment gap 2 Pnvate sector a Disposable income b Consumption c Savings d Investment e iiavings mvestment gap 3 Current dehca

Source: SPO

1973

1974

1975

I976

20.7 79.3

18.4 81.6

20.9 79.1

20.6 79.4

13.5 8.2 5.2 7.3 -2.1

12.9 8.4 4.5 11.4 -7.0

14.0 8.9 5.1 13.4 -8.3

86.5 79.6 7.0 11.5 -4.6 6.6

87.1 80.5 6.7 11.1 -4.4 11.4

86.0 81.2 4.8 11.7 -6.9 15.4

1977

1978

1979

1980

1981

1982

1983

19.7 80.3

18.6 81.4

16.1 83.9

17.5

82.5

19.3 80.7

19.7 80.3

17.4 82.6

12.4 9.1 3.3 12.9 -9.6

11.3 9.1 2.2 14.0 -11.7

12.9 9.7 3.2 10.1 -7.0

11.9 9.9 2.0 10.5 -8.5

15.5 10.9 4.6 11.8 -7.2

18.6 10.5 8.0 13.4 -5.3

18.8 10.3 8.6 12.1 -3.5

17.4 10.1 7.3 11.5 -4.2

87.6 83.0 4.6 14.5 -9.9 19.5

88.7 84.6 4 1 12.5 -8.4 20.1

87.1 80.7 6.2 9.4 -3.0 9.9

88.1 78.2 9.9 9.1 0.8 7.7

84.5 75.2 9.3 9.7 -0.4 7.6

81.4 72.6 8.8 8.4 0.4 4.9

81.2 72.3 8.8 8.4 0.4 3.1

82.6 73.4 9.2 9.2 0.1 4.1

Table A S

Supply and Use of Resources (in percentage volume change over previous year)

Gross value added: Agriculture, forestry. and fishing Industry Mining Manufacturing Energy Construction Wholesale and retail trade Transport and communications Financial institutions Ownership of dwellings Private professions and services Government, health, and education Gross domestic product at factor cost Gross national product at market prices Foreign balance' Total domestic demand Fixed capital investment Public Private Stock changesa Consumption Public Private

1976

1977

1978

1979

1980

1981

1982

1983

1984

198Sb

7.7 10.0 5.0 9.8 18.4 8.3 9.6 9.6 7.9 13.5 6.9 7.0 8.9

- 1.3 10.2 38.2 7.3 10.6 5.5 4.9 6.7 9.8 3.8 4.4 6.0 4.9

2.6 6.6 26.7 3.6 12.4 4.1 3.9 2.5 4.8 4.0 3.2 6.2 4.3

2.8 -5.6 - 16.3 -5.3 8.0 4.2 -2.3 -4.4 3.0 3.9 -0.9 4.2 -0.6

1.7 -5.9 -11.1 -5.4 -4.5 0.8 -4.1 -3.6 1.8 4.1 - 1.0 -4.8

6.4 4.6 -5.8 5.1 11.6 0.5 4.6 0.6 1.9 2.9 4.0 5.4 4.3

-0.1 8.2 1.9 9.0 2.2 0.6 1.6 3.4 0.5 2.8 3.5 4.2 4.1

3.5 9.3 -0.2 10.3 9.0 1.9 8.4 6.7 4.5 2.8 6.1 2.6 5.8

2.8 5.5 7.8 4.8 10.3 2.9 5.5 4.9 3.5 2.6 4.9 3.3 4.3

7.9

3.9

2.9

- 0.4

- 1.0 - 1.1

0.1 7.6 0.2 8.7 5.9 0.4 12.6 4.3 2.0 3.0 5.3 4.0 4.7

(2.0) -2.2 -3.6 4.6 -11.6 (0.6) -2.5 1.7 -3.1

(0.2) - 1.2 - 10.0 -3.7 - 17.3 (3.9) -3.4 8.8 -5.2

(-5.3) 11.6 17.7 18.3 17.2 (0.3) 10.2 10.8 10.1

(-1.4) 4.5 3.9 11.0 -2.1 (-0.1) 5.0 3.2 5.2

(9.2) -5.4 - 10.0 - 13.7 -6.0 (-1.7) - 2.4 9.9 -3.9

Source: OECD (1986). Tontribution to GNP growth. bProvisional. Expenditure figures are not available on the basis of the latest GNP estimate (March 1986)

4.1 (2.5) 1.6 I .7 9.4 -8.7 (0.8) 0.6 0.9 0.6

4.6 (1.7) 2.8 3.5 2.2 5.5 (-1.0) 3.9 2.0 4.2

3.3 (-1.3) 4.7 3.0 1.9 4.7 (0.2) 4.7 1.7 5.0

5.9

5.1

(0.8) 4.9 3.8 I .4 7.3 (0.4) 4.8 3.2 4.9

(0.8) 4.2 12.9 17.1 7.0 ( - 1.0)

3.1 3.3 3.7

Table A.6

Prices (annual percentage change) 1975

1976

I977

1978

1979

1980

1981

1982

1983

1984

1985

10.3 17.3 -0.1

15.7 15.0 16.6

23.9 22.8 26.3

52.6 45.0 65.7

63.9 48.9 87.5

107.2 100.3 115.7

36.8 41.6 31.1

25.2 21.2 30.2

30.6 26.4 35.4

52.0 61.3 41.9

40.0 36.7 44.2

19.0 21.2 16.2

16.6 17.4 16.7

22.5 25.8 24.5

53.2 61.9 43.7

61.8 63.5 71.1

101.4 94.2 105.7

33.9 37.6 41.9

28.3 32.7 27.2

30.8 28.8 28.0

47.3 45.6 49.9

44.9 45.0 43.6

-2.3 15.7

13.8 12.4

25.2 23.4

43.1 54.9

59.6 67.7

172.8 231.0

38.9 46.1

43.8 44.6

21.8 31.0

64.8 63.9

44.3 44.1

Whoksnle prices (1963 = 100)

General index Food and feeding stuff Raw materials and semi-finished goods Consumer price index (1963 = 100) Ankara Istanbul GNP dejlator Foreign rrade prices (TL) Export prices Import prices Source: OECD (1986)

PSBR, Determining Factors, 1973-85 (current prices, billions of TL)

Table A.7

1973

1974

1975

1976

1977

1978

1979

1980

1981

I982

1983

1984

1985

309.8 64.0 27.2 21.7

427.1 78.5 31.5 46.0

535.8 111.9 48.0 65.4

675.0 139.1 54.5 78.3

872.9 171.9 55.6 114.1

1,291.0 240.5 67.8 122.2

2.200.0 353.2 59.2 209.2

4,435.0 776.9 232.8 508.4

6,554.0 1.264.2 564.1 864.2

8,735.0 1.720.6 781.2 1,044.6

11,485.0 1,997.9 840.9 1,323.2

18,317.0 3.048.0 1.426.0 2,023.0

27,715.0 4,895.0 2,563.0 3,060.0

5.5

- 14.5

-17.4

-23.8

-58.5

- 54.4

- 150.0

-275.6

-300.1

-482.3

-597.0

-497.0

-3.3

-6.7

-9.7

-22.1

4.0

-12.4

-26.3

-244.0

-162.0

-24.1

-33.5

- 80.6

- 50.4

- 155.2

-280.5

-219.0

-275.8

-508.6

-841.0

-659.0

- 83.4

- 174.3

- 110.3

- 131.9

-235.4

-544.0

-551.0

42.0

15.0 -439.8

87.0 -242.3

-30.0 -437.7

131.0 -613.0

- 1,441.0

A. National accounts

I. 2. 3. 4. 5.

GNP Public disposable income Public savings Public investment Public savings-investment gap (PSIG) B. PSBR 6. Public-private capital transfers, net 7. Public sector capital balance (5 6) 8. Inventory revaluation fund (SEES) 9. Increase in accounts payable, neta 10. PSBR ( 7 + 8 + 9 ) C. Ratios (%) 11. PSIGGNP 12. PSBRiGNP

+

-7.7 -2.2

-

17.8

-4.0

-4.1

-3.1

-9.8

-21.4

-42.0

n.n.

-6.2

n.a. -21.9

-5.4 -32.6

-1.3 -44.6

9.7 -92.3

- 101.9

1.8 -2.0

-3.4 -5.1

-3.2 -6.1

-3.5 -6.6

-6.7 10.6

-4.2 -7.9

Source: SPO, Central bank of Turkey, and OECD. Note: n.a. = not available.

Budget, deferred payments; a positive number indicates an increase.

-

-9.5

-5.2

- 196.6

-6.8

- 8.9

-4.9

-6.2 -9.9

81.1

-4.6 -3.7

263.4

-3.0 -5.0

-4.2 -5.3

-56.0

-3.3 -7.9

-150.0 - 1,360.0

1.8 -4.9

-

Table A.8

PSBR, Sources of Financing, 1973-85 (current prices, billions of TL) ~

PSBR Financed by: 1. Foreign borrowing a. Budget, net Receipts from loans Repayment on loans b. SEEs, net c. Subtotal (a b) 2. Domestic borrowing (budget, long term) a. Receipts b. Repayments c. Net (a-b) 3. Short-term borrowing a. Central bank, net (budget + SEEs other public) b. Treasury bills, net 4. Other (changes in holdings of deposits and currency, SEE arrears, and errors and omissions) Memo items: Changes in public deposits at central bank (negative number means an increase)

+

+

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

-6.2

-1.9

-32.6

-44.6

-92.3

-101.9

-183.6

-440.0

-242.0

-438.0

-613.0

15.2

1.8

-0.7

I .5

5.5

4.3

14.1 29.3

42.4 47.3

74.0 81.0 -7.0 67.0 141.0

41.0 63.0 -22.0 122.0 163.0

77.0 -27.0 50.0

- 34.0

4.9

- 18.0

1984 -1,441.0

1985 -1,360.0

26.0 171.0 - 151.0 138.0 164.0

319.0 555.0 -236.0 177.0 496.0

-250.0 170.0 -420.0 389.0 139.0

56.0

237.0 -300 207.0

195.0 -58 0 137.0

686.0 -1720 514.0

36.0 - 54.0 104.0 86.0

2.9

3.6

7.9

10.0

12.5

16.9

31.0

38.0 -30.0 8.0

0.5 -0.8

15.3 1.3

10.0 -0.4

30.2 0.4

57.7 6.3

36.0 -3.1

96.0 -6.5

155.0 40.0

134.0 38.0

30.0 74.0

67.0 -98.0

134.0 284.0

347.0 244.0

1.8

2.4

14.1

1.5

11.5

22.8

15.8

96.0

- 143.0

192.0

273.0

390.0

116.0

0.3

-4.9

44.1

-221.0

94.0

140.8

-0.1

0.2

-0.5

-0.6

-0.8

90.0

-31.1

-

70.2

Table A.9

Consolidated Budget (billions of TL)’

Revenues Tax revenues Nontax revenues Expenditures Personnel expenditures Other current expenditures Interest payments Foreign borrowing Domestic borrowing Investment Transfers to SEES Other transfers Budget balance Change in accounts payable, net Cash balance Financing Foreign borrowing Receipts from loans Payments on loans

1980

1981

838 750 88 1,073 321 176 28 9 19 I70 165 213 -235 15 - 220

1,329 1,191 I38 1,503 390 255 67 34 33 310 267 214 -174 87 -87

74 81 -7

41 63 - 22

1982

1983

1984

1985

1.424 1,305 119 1.575 440 280 77 53 24 344 233 201 -151 - 30 -181

2,156 1.933 223 2,533 667 390 180 I30 50 463 292 54 I -377 131 -246

2,831 2.369 462 3.731 877 595 315 264

4,691 3,857 834 5,313 1,277 815 595 427 I68

683 275 926 -900 -56 -956

181 ,341 -622 -150 -772

- 18

26 177 -151

319 555 -236

-250 170 -420

36 - 54

111

,104

1986 Current Estimate 7,000 6,050 950 7,595 1,725 1,200 1,135 635 500 1.400 150 1,985 - 595 - 595 - 180

500 - 680

Percentage Change over Previous Year 1985/84

1986185

65.7 62.8 80.5 42.4 45.6 37.0 58.7 61.7 51.4 61.6 --34.2 44.8

49.2 56.9 13.9 42.9 35. I 47.2 90.7 48.7 197.6 26.8 - 17.1 48.0

Domestic borrowing Receipts from loans Payments on loans Central bank Treasury bills, net Other Memo items (% GDP): Revenues Tax revenues Expenditures Budget balance Cash balance Debt service (principal Of which: Foreign

8 38 - 30 103 40 -5

18.9 16.9 24.2 -5.3

+ interest)

-5.0 1.5

0.4

50

56

77 - 27 39 38 -81

90

20.3 18.2 22.9 -2.7 -1.3 I .8 0.9

- 34

32 74 31 19.6b 17.9b 21.6b -2.lb -2.5b 2.3b ISb

207 237 - 30 72 - 98 39 18.7 16.7 21.9 -3.3 -2.1 3.I 2.4

I37 195 -58 190 284 26 15.4 12.9 20.3 -4.9 -5.1

3.6 2.7

5 I4 686 -172 266 244 -2

17.0 14.0 19.3 -2.3 -2.8 4.3 3.I

775 2,075 - 1,300

18.9 18.9 20.5 -1.6 - 1.6 8.4 3.5

Source: OECD (1986).

1980 and 1981 fiscal year: March to February; 1982:March to December. From 1983 on, the fiscal year and the calendar year coincide bFor comparison, the ten months’ data for 1982 have been multiplied by 1.2.

782

Merih Cellsun and Dani Rodrik

Table A.10

Financial Account of Nonfinancial SEEs (billions of TL) -~

Sales revenues Operating expenses Operating surplus Direct taxes Income after taxes Depreciation Subsidies Transfers from budget Cash flow Fixed investment Changes in stocks External financing requirement Financing Foreign borrowing, net Borrowing Repayment Domestic borrowing, net Central hank, net Borrowing Repayment State investment bank, net Other domestic borrowing, net

1980

1981

1982

1983

1984

1985"

1986 Estimate

1,146 1,169 - 23

1,767 1,764 3 41 -38 28 74 241 305 406 210 -311

2.650 2,583 67 57

3.596 3,630 - 34 I26 - 160 I55 108 292 395 585 145 - 33s

6.310 5,845 465 I I7 348 240 173 239 963 545 - 508

9,279 8,433 846 357 489 374 248 181 1,292 1,706 517 -931

12,863 11,593 1,270 699 57 I 488 250 150 1,459 2,024 397 -962

I38 222 - 84 197 -5

177 520 - 343 331 - 56 132 - 188 49 338

389 728 -339 542 81 279 - 198 30 43 1

32 1 672 -351 641 73 I07 - 34 14 554

15

-38 23 30 149 164

28 I 178 -295 68 102 -34 227

so 100 -50 16 161

122 180 -58 189 32 72 -40 16 141

10

48 76 20s 339 533 151 - 345 104 192

88 24 1 31 33 -2 59 151

-

-

-5 28 I74

1 ,O00

Source: OECD (1986). "Provisional.

Table A.11

Employment and Wages in Nonfinancial SEEs Gross Wages Billion TL

% lncrease

Billion TL

535 548 547 522 529 554

92.2 142.9 236.2 310.9 370.2 461.3

71.1 55.1 65.3 31.6 19. I 24.6

117.4 305.1 699.6 1,147.3 1,629.3 2,142.3

9c Increase

~

1978 1979 1980 1981 1982 1983

Source: SPO and SIS.

'Nominal wages are deflated by GNP deflator

Wage Rate (1978 = 100)

Gross Output

Employment (thousands)

~

~

Wage Share (%)

Nominal

Real"

52.0 46.8 33.8 27.1 22.7 21.5

100.0 151.2 250.4 345.6 406.3 482.7

100.0 88. I 70.9 69.0 63.7 59.2

~~~~

30.7 72.0 129.3 64.0 42.0 31.5

Table A.12

Assets and Liabilities of the Banking System: Monetary Survey, 1973-85 (billions of TL, end-year)

A. Assets 1. Foreign assets 2. Domestic credit Of which: to government to public enterprises to private sector to other financial institutions" 3. Bonds and participations 4. All other assets Total assets 8 . Liabilities I. Foreign liabilities 2. Money (MI) 3. Quasi money (time deposits) 4. Other deposits (M4-M2)b 5. All other liabilities total liabilities

1973

1974

1975

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

31.8 91.5

24.9 129.7

22.2 184.4

26.4 261.0

18.9 365.2

47.7 460.9

68.2 707.4

319.6 1,204.8

601.9 1,867.3

985.0 2,394.8

1,674.1 3,082.1

2,571.6 3,750.9

3,370.0 6,534.6

11.3 16.2 63.3 0.7 8.9 33.8 166. I

16.4 24.5 82.0 6.7 9.5 39.6 203.7

21.8 30.4 118.3 13.9 11.6 58.8 276.9

27.4 53.3 157.7 22.6 18.8 74.9 381.3

52.6 76.6 200.7 35.3 26.3 115.1 525.4

66.3 101.3 252.9 40.4 35.9 174.2 718.7

102.7 187.8 370.1 46.7 47.7 289.8 1,113.1

199.4 316.3 640.8 48.3 85.8 687.4 2,297.6

279.4 376.0 1,157.9 54.1 144.5 I , 139.9 3,753.6

283.0 406.7 1,632.5 72.6 269.2 1,910.5 5,559.5

376.6 428.7 2,192.2 84.5 325.5 2,909.8 7,991.4

582.6 185.8 2,928.0 54.6 836.1 5,805.2 12,963.8

1,070.5 537.7 4,860.1 66.4 1,712.7 9,089.0 20,706.3

18.9 69.8 20.5 13.1 43.7 166.1

21.6 88.7 24.6 14.6 54.2 203.7

34.9 117.6 29.0 19.9 75.5 276.9

73.2 150.4 30.8 25.9 101.1 381.3

82.6 209.1 34.4 32.2 167.2 525.4

131.9 283.6 44.4 55.6 203.2 718.7

198.8

646.1 704.1 177.9 192.0 577.5 2,297.6

829.3 972.0 665.1 504.9 782.3 3,753.6

1.193.5 1,341.9 1,212.2 622.2 1,189.7 5,559.5

2,129.4 1,941.0 1,347.5 693.3 1,880.3 7,991.4

4,061.8 2,252.7 2,926.3 759.6 2,963.4 12,963.8

6,929.9 3,208.7 4,936.8 1,053.0 4,577.8 20,706.3

444.5 83.3 82.1 304.4 1,113.1

Source: Central bank of Turkey.

"Excluding credits classified under "other financial institutions," which also include investment and development banks. central bank classification, M4 equals M2

+ public deposits + other deposits with central bank + deposits of investment banks.

Table A.13

Monetary Base (MB) and Money Supply, 1973-85 (billions of TL, end-year)'

A. MB 1 . Net foreign assets 2. Import deposits 3. Domestic assets

B. C. D.

E.

Public sector Deposit money banks Other financial institutions" 4. Net other items Total MB Narrow money (MI) Broad money (M2) Money multipliers I . MliMB 2. MZlMB Income velocity 1 . GNP/MI 2. GNP/M2

I973

1974

1975

I976

1977

13.3 -1.2

3.9 -1.0

16.3

-47.7 -11.5

- 46.5

21.4 17.6

29.5 25.2

48.4 40.7

77.7 70.1

1.4 -3.8 48.7 69.8 90.3

7.5 -4.6 60.6 88.7 113.3

14.7 -4.6 80.9 117.6 146.6

23.5 -9.5 102.6 150.4 181.2

-

- 1.9

-65.2

138.6 96.3 37.1

- 8.7 151.6 209. I 243.5

1978

-88.1 -70.8

1979

- 136.9 -99.5

1980

-368.5 -99.9

220.8 125.2

384.2 162.9

706.3 274.7

42.3 13.2 216.1 283.6 328.0

48.3 -35.3 323.6 444.5 527.8

49.8 -83.9 478.5 704.0 881.9

-

1981

1982

1983

-327.3 -77.5

-321.7 -52.0

-609.7 -55.5

894.5 403.4

1,277.3 355.8

1,684.2 622.7

54.3 132.6 814.8 972.0 1637.2

-

67.5 137.7 1,189.2 1341.9 2554. I

-

1984

- 1,660.7 - 84.0

1985

-3,263.8 -59.2

4,487. I 300.5

7,179.0 356.4 49.5 20.8 4,279.7 3,208.7 8,145.5

76.0

32.2

- 147.6

- 356.2

1,570.2 1.941 .O 3,288.4

2,718.9 2,252.7 5,179.0

1.43 1.86

1.46 1.87

1.45 1.81

1.47 1.77

1.38 1.61

1.31 I .52

1.37 I .63

I .47 1.84

1.19 2.01

1.13 2.15

1.24 2.09

0.83 I .90

0.75 1.90

4.44 3.43

4.82 3.77

4.55 3.65

4.49 3.73

4.17 3.58

4.55 3.94

4.95 4. I7

6.30 5.03

6.74 4.00

6.51 3.42

5.92 3.49

8.13 3.54

8.60 3.39

Source: Central bank of Turkey.

"Figures may not add up exactly due to rounding. bIncluding State Investment Bank.

Table A.14

The Current Accounl and its Financing (million $) Nondebt Financing Implied Net Foreign Borrowing:

Year

Current Account Balance” (1)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 ‘1984 1985 1986

Foreign Direct Investment (2)

- 55

58 45 43 79 33 I I4 10 27 34 75 18 95 55 46 113 99 125

-40 47 534 - 662 - 1,648 2,029 -3,140 - 1,165 - 949 - 2,938 - 1.669 - 835 - 1,828 - 1,407 - 1,013 1,528 ~

~

Change in Reserves” (3) - 186 - 320 - 639 - 704

424 429 54 349 - I87 16 -512 - 263 - 297 269 - 1,715‘ 395 - 494 ~

Counterpart to Valuation Changes‘ (4)

Net Use of 1MF Resources (5)

Excluding I M P (6)

Including I M F (7)

n.a. n.a. n.a. n.a. n.a. -40 30 -1 -4

37 -7 - 67 0 0 248 143 18 213 II 422 268 I29 I17 - 142 - 103 - 241

146 322 616 91 205 897 1,792 2,747 1,109 747 2,991 1,501 935 1,773 2,473 399 1,887

183 315 549 91 205 1,145 1,935 2,765 1,322 758 3,413 1,769 1.064 1,890 2,331 296 1,646

40 19 68 13 161 678‘ 223 25 1

Sources: Central bank of Turkey’s Annual Reports, and OECD Economic Surveys

“To render this series compatible with the central bank’s revised presentation of the balance of payments, “imports with waiver” and “agricultural surplus” items have been moved from the capital account to the current account to adjust the O E D figures prior to 1975. The central bank’s (revised) series has been used for years since 1975. bExcludes use of IMF resources. Includes changes in holdings of reserves (gold and foreign exchange) of the central bank and of the deposit money banks, changes in holdings of SDR. and changes in the reserve position with the IMF.

‘For official reserves only. Valculated as

- (1)

‘Calculated as (5)

-

(2) - (3)

- (4)

+ (6)

‘In 1984 the central bank’s statistics contain a fictional entry of $849 million under “other claims” of the CB, offset by a counterpart entry. This is done to account for the transfer of the foreign exchange holdings of the deposit money banks, previously shown under official reserves. Since the present table aggregates CB and DMB reserves, we have subtracted this 849 million from the reserve changes and added it to the counterpart item. This gives a more accurate picture of the actual change in reserve holdings.

786

Merih Cellsun and Dani Rodrik ~~

Table A.15

~~~~~

Balance of Payments (new format, million $)

Trade balance" Exports (f.0.b.) Imports (f.0.b.) Invisibles, net Services Tourism Investment income Interest paymentsb Other Other services Transfers Official' Private Workers' remittances Otherd Current balanceb Long-term capital, net Direct investment Credits received Project credits Other official credits' Private credits Debt relief Principal Interest Debt repaymentsb Official Private Basic balance Short-term capital Errors and omissions Counterpart items Overall balance Change in official reserves Net use of IMF Other

1980

1981

1982

1983

1984

1985

-4,603 2,910 7,513 1,195 - 976 222 - 1.163 -1,138 - 25 - 35 2,171 18 2,153 2,071 82 -3,408 2,212 I8 2,300 547 1,588 165 1,450 980 470 - 1,556 n.a. n.a. - 1,196 92 1,082 19 -3 3 422 -419

-3,864 4,703 8,567 1,945 - 630 277 - 1,434 - 1,442 8 527 2,575 16 2,559 2,490 69 - 1,919 1,263 95 1,538 642 840 56 850

-2,628 5,890 8,518 1,692 - 602 224 - 1,473 - 1,565 92 647 2,294 105 2,189 2,140 49 - 936 1,085

-2,990 5,905 8,895 1,092 - 693 292 -1.449 -1,511 62 464 1,785 236 1,549 1,513 36 - 1,898 349 46 1,299 508 535 256

-2.942 7,389 10,331 1,535 - 579 271 -1,387 -1.586 199 537 2,114 229 1,885 1,807 78 1,407 1,159 113 2,150 733 873 544 580 580 n.a. - 1,684 n.a. n.a. -248 36 317 - 171 - 66 66 - 141 207

- 2,975

Mx)

250 - 1,220 n.a. n.a. - 656 - 307 478 68 -417 417 268 149

55 1,882 754 982 146 750 650 100 - 1,602 n.a. n.a. 149 - 83 - 75 27 18 - 18 133 - 151

1.OOo

930 70 - 1,996 n.a. n.a. - 1,549 958 507 180 96 - 96 78 - 174

-

Source: OECD (1986).

%chiding transit trade. bBefore debt relief. 'Including grants. dIncluding workers' imports. eIncluding European Resettlement Fund loans, World Bank SALs and bilateral program loans.

8,255 11,230 1,962 - 29 770 - 1.320 - 1,753 433 52 I 1,991 229 1,762 1,714 48 - 1,013 75 95 1,838 926 283 629 n.a. n.a. n.a. - 1,858 1,711 - 147 - 938 1,656 -810 370 278 - 278 - 255 - 23

787

TurkeyiAppendix B

Table A.16

External Debt, 1973-81 (million $) Medium and Long Term

Year

Public”

Private

Short Term

1973 1974 1975 1976 1971 1978 1979 1980 1981

2.869 3.126 3.176 3,590 4,326 6,467 11,113 12,000 12,789

I I5 146 160 248 479 557 630 668 556

279 216 1,155 3,050 6,191 7,176 3.556 2,500 2,194

IMF

Total

243 39 1

3,263 3,488 4,734 7.279

409

1 1.405

622 633 1,054 1,322

14,822 15,932 16,222 16.861

-

Source: World Bank and central bank. %blic or publicly guaranteed.

Table A.17

External Debt, 1982-86 (million $)

By maruriry: Total outstanding disbursed Medium and long term Short term By borrower: Medium and long term Public sector (incl. SEEs) Central bank (Dresdner Bank Scheme)b Private sector Short term Public sector (incl. SEEs) Central bank Private sector By lender: Medium and long term Multilateral agencies IMF IBRD, IDA, IFC European Investment Bank European Resettlement Fund Islamic Development Bank OPEC Fund International Fund for Agricultural Development Bilateral Lenders OECD countries OPEC countries Other countries

1982

1983

1984

1985

1986‘

17,619 -

18,385 -

21.258 -

25.349 -

-

31,228 -

15,855 -

16,104 -

18,078 -

15,855 1.764

I 1,497 3,453 (400) 905 1,764 73 63 I

1,060

16,104 2,281

11,439 3,825 (758) 840 2,281 I94 1,432 655

15,104 -

-

-

2 6,560 5,607 535 418

4,531 1,455 2.1 I5 420 384 1 I7 40

7,115 6,146 587 382

16,352 4,916 1,512 2,488 393 399 22 40

18,078 3,180

12.805 4,453 (1.326) 820 3,180 215 1,183 1,782

18,078 5,494 1.426 3,044 39 1 554 35 40

4 7.204 5,987 603 614

20,590 4.759

20,590 -

24,317 6.91 I

24,317 -

14,1688 5,037 (1,858) 865 4,759 230 1,685 2,844

-

18,091 5.269 (2.480) 957 6.911 597 2,335 3,979

20,590 -

24,317 -

7 7,955 6,528 640 787

8 10.187 8,270 1,027 890

6.157 1,326 3,490 449 815 35 35

-

6,588 1,085 3,643 573 1,197 53 29

788

Merih Cellsun and Dani Rodrik

Table A.17

(continued) 1982

Commercial banks Private lenders (Dresdner Bank scheme) Short term Islamic Development Bank Bilateral lenders Commercial banks Private lenders (Dresdner Bank scheme) (Foreign exchange deposit accounts) By type of credit: Medium and long term Project and program credits Eurocurrency loans Rescheduled debt

CTLDs Bankers' credits TPRCs NGTA Private credits (Dresdner Bank scheme) Short term Public sector Bankers' credits Overdrafts Dresdner Bank program Petroleum credits Other Private credits CTLDs Acceptance credits Pre-export financing Foreign exchange deposit accounts Other

1983

1984

3,229 980

3,262 1,366

I985

3,704 1,676 (L,326) 3,180 65

(400)

(758)

1.764 -

2,281 -

-

-

-

73 68 I46 1,477 (417) 0

15,855 11,321 720 2,859 1,996 429 84 350 955

(400) 1,764 704

-

48 417 68 171 1,060 585 276 199

-

94

392 1,795 (493) (83)

-

16,104 -

11,244 902 2,710 1,886 429 45 350 1,248 (758) 2,281 979 65 164 493

-

257 1,302 647 318 254 83 ~

1986"

4,351 2.127 (1,858) 4,759 30 -

4,833 2,709

(2,480) 6,911 33

-

94 1 2,174 (452) (544)

1,465 3,264 (820) (724)

2,624 4,254 (1,308) (1.250)

-

20.590 -

24,317 -

-

-

18,078 12,311 1,653 2,401 1.65 1 400

-

350 1,713 (1,326) 3,180 1,337 195 417 452

13,616 2,616 2,004 1,420 315

269 2,354 (1,858) 4,759 1,897 432 376 820

16,211 3,438 1,624 1,166 229 229 3,044

(2,480)

-

-

6.91 1 -

-

-

-

273 1,843 61 703 414 544 121

269 2,862 18 1,093 609 724 418

Sources: Central bank of Turkey and the Undersecretariat of Treasury and Foreign Trade.

"Provisional. bDeposits with more than one-year maturity are classified under medium- and long-term debt.

2,926 944 77 1,308

597 3,985 6 1,061 629 1,250 1,039

Table A.18

Currency Composition of External Debt (percentage distribution at year-end 1985)

Commercial Banks Pre-export fiancing Acceptance credits FX deposits FX credits Central bank Overdrafts Bankers’ credits CTLD Dresdner scheme Nonfinancial public sector ID9 TUPRAS Total short-term debt Total medium & long-term debt Total external debt

ATS

BF

DM

FF

HFL

SFR

LIT

YEN

STG

Others(US$)

57.30 59.29 25.76 76.56

0.00 0.89 1.95 0.00

0.18 0.27 1.11 0.00

12.58 13.28 51.02 10.46

0.57 1.31 4.15 0.14

0.13 0.73 8.23 0.00

27.48 8.40 3.56 12.15

0.67 1.55 0.03 0.00

0.00 9.19 2.34 0.00

1.08 5.08 I .02 0.61

0.00 0.00 0.82 0 09

608.87 1,093.70 724.00 418.00

83.71 97.64 16.76 2.84

0.00 0.00 0.00 0.67

0.00 0.00 0.00 0.08

14.43 2.36 61.45 88.70

0.00 0.00 0.00 0.69

0.45 0.00 0.00 4.00

0.35 0.00 5.03 2.63

0.00 0.00 10.61 0.00

0.21 0.00 0.00 0.00

0.00 0.00 0.00 0.09

0.85 0.00 6.15 0.28

375.70 432.20 17.90 2,677.60

0.00

0.00

0.00

30.00 200.00 6.6 16.97 17,064.24 23.68 1.2I

100.00

0.00

100.00 -

E

0.00

0.63

0.98

0.22

57.94 -

39.12

52.68

0.88

0.55

0.00

0.00

0.00 -

0.00 -

o.00

I .O1

2.69

15.59 -

3.09 -

0.96 -

46.63

24.26

2.51

Source: Central hank. Key: US$ ATS BF DM

FF

= = = = =

U.S. dollar Austrian shilling Belgian franc Deutschemark French franc

Total Value (million $)

US$

HFL SFR LIT YEN STG

= = = = =

Dutch guilder Swiss franc Italian lire Japanese yen Pound sterling

0.00

1.44

0.00

0.00 0.00 -

o.00

6.68 0.21 -

6.23 -

6.19

6.54

0.32 0.24

I .79

4.99

0.00

0.00

0.00 -

0.00 -

I .78 -

5.87 -

1.13

I .59

0.27 4.31

790

Merih Cellsun and Dani Rodrik Black-Market Premium of the U.S. Dollar, 19691-1983IV

Table A.19

I 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983

0.544 0.541 0.061 0.031 0.022 0.035 0.092 0. I33 0.106 0.249 0.736 0.182 0.049 0.112 0.179

I1 0.547 0.512 0.044

0.016 0.003 -0.012 0.048 0.077 0.109 0.122 0.660 0.083 0.105 0.175 0.194

111

IV

0.490 0. I72 0.031 0.020 -0.049 0.050 0.104 0.082 0.133 0.087 0.429 0.009 0.151 0.133 0.114

0.528 0.076 0.063 0.035 0.019 0.049 0.125 0.106 0.346 0.356 0.463 0.175 0. I47 0.077 0.119

Note: Calculated as the ratio of the black-market rate to the official rate minus one. The hlack-market rate for each quarter is taken to he the geometric average of end-of-month rates from Pick’s Currenrv Yearbook.

Table A 2 0

Real Effective Exchange Rate (export-weighted), 1969:1-1985:1V (1969:I= 100)

1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

100.0 101.1 156.0 134.9 130.0 118.5 119.4 117.8 119.1 117.0 105.0 133.6 132.8 155.5 167.3 178.7 167.2

100.4 102.9 154.3 134.1 134.3 116.3 121.3 118.2 119.5 126.3 100.8 139.8 136.5 156.6 168.3 175.7 176.2

102.1 141.6 150.6 131.8 138.3 117.5 124.2 120.6 116.8 119.1 115.9 145.0 144.0 167.5 170.4 175.7 183.8

103.2 164.7 142.9 128.8 125.8 121.9 123.5 120.2 114.I 114.0 104.2 136.0 153.8 170.7 171.2 173.0 180.4

Nore: Calculated as the geometric average of bilateral real exchange rates, using wholesale price indices. The weights correspond to the export shares of the six largest OECD trade partners; the balance is split between the US$ (two-thirds) and the DM (one-third). The resulting weights are: U.S. (0.43). Germany (0.36). Italy (0.07). Switzerland (0.05). France (0.051, U.K. (0.04).

791

TurkeyiAppendix B

Table A.21

Real Effective Exchange Rate (import-weighted), 1969:1-1985:IV (1969:I= 100)

1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

100.0 101.1 158.7 136.1 130.7 118.4 119.2 118.0 119.2 116.8 104.3 134.0 134.2 158.6 171.1 184.0 172.3

100.6 102.9 156.9 135.3 135.2 115.9 121.2 118.4 119.7 126.6 100.I 140.6 138.4 159.8 172.4 180.7 181.8

102.3 143.6 152.9 132.8 138.9 117.4 124.5 120.8 116.8 119.0 115.6 146.1 146.5 171.7 175.0 181.0 189.7

103.3 168.0 144.7 129.6 125.9 122.0 123.9 120.2 113.9 113.6 103.4 137. I 156.6 175.0 175.9 178.3 186.3

Nuze: Calculated as the geometric average of bilateral real exchange rates, using wholesale price indices. The weights correspond to the import shares of the six largest OECD trade partners; the balance is split between the US$ (two-thirds) and the DM (one-third). The resulting weights are: U.S. (0.48), Germany (0.32), Italy (0.03, Switzerland (0.05). France (0.05). U.K. (0.05).

Index of Real Exchange Rate Volatility, 1970:1-1985:IV (19709 = 100)

Table A.22

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985

I

II

111

IV

100.0 885. I 142.6 86.4 403.4 165. I 198.3 76.7 143. I 434.4 932.2 319.2 146.9 217.2 107.1 85.6

110.7 9 19.7 155.0 157.2 336.9

836.6 534.2 162.4 154.8 295.4 149.3 180.2 71.5 357.2 586. I 829.8 322.8 196.6 103.4 152.3 257.3

814.7 135.4 142.3 380.2 256.9 122.5 180.6 80.3 382.2 665.3 725.6 239.6 164.7 87.0 168.8 261.8

144.0 186.5 76.8 285.4 126.1 862.9 289.9 176.5 218.1 149.4 223.8

Note: Calculated as the standard deviation of the quarterly (percentage) change in the export-weighted real exchange rate index over the last four quarters (inclusive of present quarter).

Table A.23

Exports by Commodities (million $)"

A. Agricultural products

1. Cereals 2. Fruits & vegetables a. Hazelnuts b. Dried fruit c. Citrus fruit d. Other 3. Industrial crops & forestry products a. Cotton b. Tobacco c. Other 4. Live animals & sea products B. Mining & quarrying products C. Processed & manufactured products 1. Processed agricultural products 2. Manufactured products a. Textiles & clothing b. Hides & leather c. Forestry d. Chemicals e. Rubber & plastics f. Petroleum products g. Glass & ceramics h. Cement i. Iron & steel j. Nonferrous metals k. Metal products & machinery 1. Electrical equipment & products m. Other Total

Source: OECD (1986). "Excluding transit trade

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

1,254 70 375 203 68 87 17 734 438 25 1 45 75 I10 596 98 498 263 50 7 47 3 16 20 16 22 17 16 2

1,041 I20 440 25 1

1,542 262 561 33 1 I45 44 41 617 348 225 44 102 124 622 I10 512 309 40

1,344 I67 647 353 166 53 75 446 227 176 43 84 I32 785 151 634 378 44 2 23 3

1,672 181 754 395 187 86 86 606 323 234 49 131 191 1,047 209 838 424 50 4 76 16 39 36 40 34 18 30 II 60 2,910

2,219 326 795 302 208 125 160 813 348 395 70 285 194 2,290 412 1,878 803 82 20 94 72 I07 102 198 100 30 85 26 I59 4,703

2,141 337 649 24 I 168 77 163 74 1 297 348 96 414 I75 3,430 569 2,861 1.056 Ill 33 148 60 343 I04 207 362 45 I43 75 I74 5,746

1,881 376 59 I 246 120 72 153 531 I97 238 96 382 I89 3,658 670 2,988 1,299 192 15 120 77 232

1.749 267 646 305 I I9 62

1.719 234 561 255 73 58 I75 659 I70 330 I59 265 244 5,995 647 5.348 1.790 484 106 266

19

1.960

100

77 12 432 210 I76

46 49 126 586 137 449 260 52 1

1

34 3

24 2

-

27 9 14 20 14 3 12 1,753

-

30 41 21 12 18

4 10 2.288

-

37 45 31 15 18 4 3-1 2,261

108

81 407 79 I22 69 I87 5,728

160

492 I68 216 I08 343 240 5,144 808 4,336 1,875 40 I 24 173 97 409 I46 56 576 86 134 100

259 7,133

108

372 190 44 969 I16 450 I19 334 7,958

Table A 2 4

Imports by Commodities (million $4"

A. Agricultural products & livestock B. Mining & quarrying I . Oil a. Crude oil h. Oil products 2. Other C. Industrial products 1. Agriculture-based processed products 2. Industrial products a. Chemicals h. Fertilizer c. Rubber & plastics d. Textiles e. Glass & ceramics f. Iron & steel g. Nonferrous metals h. Metal products i. Machinery j. Electrical appliances k. Motor vehicles 1. Other industrial products D. Imports with waiver Total

Source: OECD (1986)

"Excluding transit trade. "Including $269 million gold imports.

1976

1977

1978

1979

1980

1981

1982

1983

1984

1985

77 1, I91

112 1,546 1,436 1,152 284 I10 4,037 58 3,978 555 214 266 51 25 690 97 15 1,060 29 I 572 141 102 5,797

50 1,486 1.396 1.044 352

36 1,818 1,712 962 750

50 4,006 3,862 2,952 910

I25 4,098 3,878 3,258 620 220 4.641 228 4,412 919 280 240 78 40 605 141 23 1.223 336 356 171 69 8,933

I76 3.961 3.749 3,528 22 I 212 4.657 I76 4,482 839 51 237 103 34 59 I 122 37 1,309 374 594 191 49 8,843

138 3,864 3,665 3,242 423 199 5,177 203 4,974 1,032 I19 25 I 98 57 675 I95 30 1,432 398 478 209 56 9,235

417 3,908 3,637 3,373 264 271 6,432 434 5,998 1,212 I28 359 1 I7 63 862 220 34 1,618 573 517 295

375 4.186 3.612 3,321 29 1 574" 7,052 48 1 6,565 1,294

1,106

1,002 104

85 3,725 I47 3,578 554 98 183 58 25 546 89 27 1,070 274 518 134 I36 5,129

90

106

144

2.943 50 2,893 476 283 154 50 18 408 42 20 76 1 218 378 83 120 4,599

3,092 115 2,977 524 356 145 46 28 345 55 14 903 25 I 22 I 88 123 5.069

3,759 30 I 3,458 727 395 181

79 35 462 87 23 843 270 223 133 94 7,909

343 I46 63 1,060 224 38 1,551 664 812 370

-

-

10,757

11,613

794

Merih Cellsun and Dani Rodrik

Table A25

Export Shares in Manufacturing Gross Output (in producers’ prices) ~~

Gross Output (1983 billion TL)

Food & beverages Textiles & clothing Hides & leather Forestry Paper & printing Chemicals Rubber & plastics Petroleum products Glass & ceramics Cement Iron & steel Nonferrous metals Metal products & machinery Electrical equipment & products Agricultural equipment Transportation vehicles Other Total

Average Export Share in Gross Output (9%)

1978

1983

1978

1983

Marginal Export Share in Gross Output (%)

!,155.0 752.3 159.7 476.6 121.7 432.2 120.2 910.9 175.5 137.0 301.9 113.5 327.8

2,818.0 851.2 219.6 535.3 150.4 693.9 192.0 212.7 121.7 467.5 157.4 454.4

6.19 23.04 5.71 0.08 0.20 1.72 0.53 0.40 3.95 7.01 1.47 3.11 0.98

7.70 39.35 20.36 2.39 3.53 7.90 5.47 4.59 11.49 14.84 17.08 9.10 7.38

12.59 163.34 59.38 21.22 17.67 18.11 13.73 24.74 47.08 -54.95 45.52 24.57 23.96

190.0 48.4 283.4 6.3 6,712.2

251.5 96.4 337.1 7.9 8,667.2

1.29 0.36 0.82 0.02 5.38

6.65 21.81 4.45 4.19 11.01

23.23 43.43 23.46 21.05 30.33

1.100.2

Source: SPO

Table A.26

Import Protection, 1985 Nominal Protection Rates

Sector Agriculture Industry Mining Manufacturing Consumer goods Intermediate goods Capital goods

Import Duties

Import Duties plus EBF Levies”

0.22 0.25 0.16 0.29 0.5 1 0.25 0.31

0.25 0.30 0.18 0.36 0.58 0.30 0.39

Source: Ali Turhan, “lthalatta Sektorel Vergi Tahsilat Oranlari, Agirlikli Vergi Oranlari ile Koruma Oranlari.” PIunlumu Dergisi, no. 20, State Planning Organization (Ankara, 1986). aEBF refers to the extrabudgetary funds.

Table A.27

Distributional Statistics, Agricultural and Nonagricultural Households, 1973 1973 Estimates Ab

1973 Survey Data"

Relative distribution (7%) Quintile: I' L

3 4

Gini coefficient Log variance Mean incomes (1973 TL) Overall Top decile Bottom decile Ratio: Top decile-bottom decile Ratio: Mean-Turkey mean Share (Sof): Poor (<12,000) Wealthy (>72,000)

Turkey

Agricultural

NOndgricU~tUral

3.45 7.90 12.83 20.49 55.33 100.00

2.53 6.20 11.06 19.86 60.34 100.00

4.97 9.28 13.73 20.47 51.55 100.00

Turkey

2.75 7.15 12.83 22.06 55.21 100.00

2.48 5.97 10.79 19.69 61.07 100.00

-

4.97 9.50 14.45 22.39 48.59 100.00

-

0.50 0.97

0.56 1.26

0.45 0.65

0.5146 1.1714

0.5726 1.26

0.432 0.65

24,570 96,090 2,726 35.25 I .oo

22,516 97,009 1,799 53.92 0.92

26,344 95,282 4,839 19.69 1.07

31,660 118,120 2,690 43.91 I .OO

22,610 98,780 1,830 53.98 0.71

41,340 131, I 80 7,610 17.24 1.31

38.41 5.11

49.10 5.31

29. I8 4.93

32.01 9.52

49.88 5.56

12.90 13.76

"Source: DerviS and Robinson (1980, 112) estimates aggregated from the 1973 survey data. bSource: The authors' estimates (excluding workers' remittances). 'Lowest (poorest) 20 percent.

Nonagricultura!

-

Source: Celisun (1986b).

dHighest (richest) 20 percent.

Agricultural

-

Table A.28

Distributional Statistics, Agricultural and Nonagricultural Households, 1978-83 ~

~~~

1978 Estimates A" Turkey Relative distribution (70) Quintile: I

2 3 4 5

Gini coefficient Log variance Mean incomes (1973TL) Overall Top decile Bottom decile Ratio: Top decile-bottom decile Ratio: Mean-Turkey mean Share (%) of Poor (<12,ooO) Wealthy p72.000)

Agncultural

1983 Estimates Ad

Nonagncultural

Turkey

Agricultural

Nonagncultural

2.84 7.33 12.99 22.13 54.71 100.00

2.48 5.97 10.79 19.69 61.07 100.00

4.97 9.50 14.55 22.39 48.59 100.00

2.63 6.93 12.59 21.39 55.93 -

2.48 5.97 10.79 19.69 61.07 -

4.58 9.01 14.09 22.15 50.I7 -

0.5089 1.1424

0.5726 1.26

0.432 0.65

0.5224 1.2212

0.5726 1.26

0.4504 0.515

39,300 144,940 3,450 42.01 I .00

27.900 121.890 2,260 53.93 0.71

50.000 158,660 9,200 17.25 1.27

35,020 132,830 2,810 47.27 1 .OO

21,750 95,020 1,760 53.99 0.62

46,030 152.570 7,690 19.84 1.31

24.98 13.99

42.45 7.99

8.58 19.62

29.89 11.67

51.25 5.18

12.16 17.06

Source: Celisun (1986b).

aThe authors' estimates (excluding workers' remittances)

100.00

100.00

100.00

797

Turkey/Notes

Notes Acknowledgments It is impossible to name and thank every individual who contributed to this project in the two years it took to complete. We include here a partial list only. At the data collection stage, we were assisted generously by the Central Bank, the State Planning Organization (SPO), the Undersecretariat of Treasury and Foreign Trade (UTFT), and the Directorate of Mass Housing and Public Participation. Riiadii Saracoglu and Zekeriya Yildirim of the Turkish central bank took time from their busy schedules to discuss with us some general issues regarding Turkey’s debt strategy. We were also assisted at the central bank by Nezihi Alpturk who helped answer our questions and directed us to other sources of information. At the SPO, Necati Ozfirat was particularly helpful and gave generously of his time. We are also grateful to Yavuz Arinsoy, Zafer Yukseler, Sevki Emin Kahyagil, and Ibrahim QinakGi (all of the SPO), Omer Altay and Ernur Demir Abaan (both of the central bank), Tandogan GuGbilmez (Directorate of Mass Housing and Public Participation), and Kadir Gunay and Ayae Oktem (both of UTIT) for providing us with data and information, as well as for clarifying some puzzles. Many others contributed by giving us their reactions to earlier drafts of individual chapters or in discussions. We would like to thank in particular Tosun Aricanli, Osman AtaG, Marta Castello Branco, Pat Conway, Kemal Dervia, Avinash Dixit, Nazim Ekinci, Peter Kenen, and Jeffrey Sachs. We are grateful to Tevfik Yaprak for his most efficient and enthusiastic all-around research support at every stage of the study. Excellent research assistance was also provided by Diana Edge and Isabel Marshall. Meral Sariasma and Jean McKeown graciously undertook much of the typing under severe time pressures. Finally, we want to thank the National Bureau of Economic Research for the excellent research environment it provided during the project. We are extremely grateful to Kirsten Foss Davis in particular for her expert handling of many administrative details along the way.

Chapter 1 1. See Krueger (1974a) for a trade-focused and exceptionally well-documented analysis of the Turkish economy from 1950 to 1971. 2. The initial conditions and origins of etatism are analyzed by Okyar (1965). 3. For an analysis of policy shifts in 1946-50, see Tekeli and Ilkin (1974). 4. The basic reference for national accounts from 1923 to 1948 is Bulutay et al. ( 1974). 5. Singer (1977) provides an interpretive study of the political context, economic policy, and performance in the 1940s and 1950s. 6. For empirical assessments of Turkey’s import-substitution experience in the 1960s, see Krueger (1974a), Krueger and Tuncer (1980), and Celkun (1983). 7. It should be noted that capital inflow figures in table 1.2 measure the net imports of goods and nonfactor services, and thus exclude interest payments and workers’ remittances. In terms of current account deficits, capital inflows were moderate by cross-country standards and averaged about 1.5 to 2 percent of GNP

798

Merih Celasun and Dani Rodrik

during 1963-73. The 1953 and 1963 actual data refer to 1952-54 and 1962-64 averages, respectively. 8. See Hatiboglu (1978) and Celksun (1983) for price-distortion effects on sectoral income differentials. 9. Aksoy (1982) examines the deficit-financing requirements of agriculture-based public enterprises as part of a wider analysis of structural aspects of inflation in Turkey. 10. See Akyuz (1984) for a disaggregated analysis of Turkey’s financial system and flow of funds for these benchmark years. In reviewing table 1.4, it may be noted that the share of the monetary system in the total assets of the financial system in 1963 stood at around 39 percent in developed economies and at around 69 percent in developing countries. For the U.S., Akyuz reports that the share of equities and bonds in total issues of the domestic real sector was 37 percent in 1900 and 46 percent in 1960. The latter share for Turkey averaged around 18 percent in 1970-81. 11. Walstedt (1980) provides a detailed evaluation of the economic and financial performance of SEES from 1960 to 1974. For a review of SEE reform proposals and legislation, see Karata? (1986).

Chapter 2 1. For an analytical account that stresses the lack of exchange rate and fiscal adjustment to the first oil crisis, see Lewis (1986). This study neglects the role of the borrowing strategy which we will emphasize here. 2. While the terms-of-trade effect has direct welfare consequences, the reduction in exports per se does not, unless the domestic economy has unemployment exacerbated by lower foreign demand or the exportables sector makes oligopolistic profits. See Dornbusch (1985) for a discussion of this point. 3. This is a conservative procedure because the trend rate for 1972-74 was considerably lower than for a time span stretching further back. Also, we are assuming a constant dollar value of remittances despite the depreciation of the dollar against the deutsche mark from 1974 to 1976. 4. The ratios of foreign savings to GNP displayed in table 2.7 are very close, but not identical, to the current account ratios of table 2.6 due to conceptual differences in the measurement of the current account in the national accounts and the balance of payments, respectively. 5. These numbers do not add up exactly due to rounding. 6. The discrepancy between the public savings-investment gap displayed in the national accounts and the PSBR in table 2.8 arises from several sources. For example, the national accounts treat the joint ventures of state enterprises with private firms as private sector activities. The item “public-private capital transfers” in table 2.8 adjusts for that. 7. These numbers are higher than those reported in table 2.5 because they include public enterprises and local governments alongside the consolidated central government. 8. It should be kept in mind that the term “public sector” here excludes the central bank. 9. Prior to May 1975, a limited program of convertible deposits for Turkish residents and workers abroad already existed. This explains the presence of CTLD items in table 2.10 prior to 1975.

799

Turkey/Notes

10. There is in reality an additional step. The exchange guarantee implies that the central bank, acting on behalf of the Treasury, takes over the foreign exchange liability of the commercial bank, leaving it with a liability denominated in liras. 1 1 . A World Bank (1980, 209) report suggests the average gestation lag to be around two-and-a-half years. 12. One U.S. banker is quoted as having commented: “I can’t understand why the American banks aren’t rushing into this market. We can net 6 percent with no difficulty at all” (Brennan 1976, 84). 13. Due to the shortage of foreign exchange, payments on maturing CTLDs, but not on interest, were apparently stopped by the central bank in July 1977 (Bleakley 1978, 50). 14. Notice that the progressive increase in these front-end fees is indicative of a positively sloped supply curve on the part of foreign lenders. In view of country-risk considerations, banks must have been willing to increase their exposure to Turkey only by being compensated for doing so. 15. Stability is also possible with signs reversed for both expressions, i.e., A < q*/p and (1 - apA) < 0. But in this case any negative shock to the current account will lead to decumulation of foreign debt (i.e., current account surpluses), so we ignore it. 16. This possibility depends on the precise configuration of the parameters involved and, in particular, on the foreign interest rate, q*. Remember that q* was actually being driven up by Turkish borrowers willing to pay increasingly higher front-end fees. This aspect of the process, not captured in the simple model discussed here, would naturally make stability more problematic. 17. See Cellsun (1980), table 1. 18. The exchange guarantee was finally lifted, but not for existing obligations, in February 1978.

Chapter 3 1. A good example of the overambitiousness of the authorities is provided by the targets of the fourth five-year plan. Announced in September 1978 in the midst of a foreign exchange crisis, the plan foresaw an 8 percent annual average rate of growth. See Celbun (1980, tables 2 and 3) for a comparison of planned with actual macro aggregates for 1978 and 1979. 2. Data on black-market rates are from Pick’s Currency Yearbook. 3. The figures here for net foreign assets and domestic assets have been adjusted to eliminate the effects of valuation changes. Hence, the changes in foreign assets are net of valuation effects, as are the changes in domestic assets. Since these adjustments completely offset each other, the figures on the money base remain unaltered. 4. Since import deposits and some other items are not included under either domestic or foreign assets, the growth rate of base money is not constrained to lie in between the growth rates of these two. 5. We are grateful to Marta Castello Branco for drawing this passage to our attention. 6. The agricultural sector is a net exporter, and therefore benefits from real depreciations.

800

Merih Cellsun and Dani Rodrik

Chapter 4 1. See, e.g., Balassa (1983) and Kopits (1986) for the Turkish experience, and Michalopoulos (1987) for the World Bank’s perceptions of adjustment and growth in developing countries in the mid- 1980s. 2. Besides Balassa (1983) and Kopits (1986), see also the following studies and evaluations of Turkey’s post-1980 economic experience: Boratav ( 1986), SenSes (1983), Okyar (1983), Onis (1986), World Bank (1982, 1983a), Euromoney (1982), Yagci et al. (1985), and various issues of OECD Economic Surveys on Turkey. 3. See Akder (1987) for an analysis of Turkey’s export expansion to the Middle East. Akder also provides estimates for similarity indexes of exports to the Middle East and the European Community. 4. See Syrquin (1986) and Cellsun (1983) for analyses of structural transformation in Latin America and Turkey, respectively. 5. See IMF (1986). 6. See a candid interview with Turgut Ozal on the introduction of the 1980 policy package in Euromoney (1982). 7. See Okyar (1983) and Kopits (1986) on the suspension of labor union activities and wage negotiations under collective bargaining in the post-September 1980 military period. Okyar and Kopits also discuss the wage settlement arrangements under the High Arbitration Council, which was established by the military government. 8. For a review of FDI activities, see Erdilek (1986). 9. See, e.g., McKinnon (1982). 10. Evidence on this will be provided in chapter 5. 11. See TUSIAD (1986). 12. These brokers offered very high yields on CDs which they bought wholesale from commercial banks and then invested the receipts in doubtful ventures. The scheme was in effect a Ponzi game and collapsed as soon as the inflow of new deposits fell short of interest payments coming due. Chapter 5 1. For price distortions before the mid-I980s, see the study by Yagci (1984), which provides quantitative measures on the incidence of the protection-subsidy system in Turkish manufacturing in 1981. 2. See Conway (1987) for a quantitative analysis built around the incomeexpenditure framework. 3. See also the supplementary tables in the statistical appendix. 4. For an alternative presentation for the first two years of the program, see also table 9.2. Notice that the debt relief in question includes only the reschedulings undertaken with respect to liabilities to ofJiciul creditors. The effect of the CTLD reschedulings does not show up here. Their inclusion would naturally magnify the role of debt relief. See chapter 9 for more information. 5. This is brought out in a World Bank (1983b) survey of 127 companies. 6. In the estimation of income velocities and multipliers in table 5.8, year-end values of monetary variables are used. 7. See Cellsun (1986b) for the estimated values of intersectoral income shifts induced by changes in domestic terms of trade during 1973-83. It may furthermore

801

Turkey/Notes

be noted that the value added of the public services sector comprises government employee salaries. 8. In Turkey’s official planning data for the labor market, agricultural employment includes underemployed labor (or disguised unemployment) in this sector. Hence, the surplus labor shown in table 5.10 corresponds to urban unemployment. The urban labor supply is the difference between total labor supply and agricultural labor. 9. The share of formal wage laborers (including government employees) in total economywide employment was 24.8 percent in 1978. For disaggregated labor data, see Celdsun (1986a).

Chapter 6 1. We gratefully acknowledge the computer programming support of Tevfik Yaprak in undertaking the study summarized in this chapter. 2. The actual data in table 6.1 may show minor deviations from actual data reported elsewhere in this monograph, mainly due to differences in data sources. The current deficit figures in this table come from the official format for the balance of payments that was used in the pre-1985 period. 3. The indicators for the functional distribution of income estimated by using this model have been previously reported in table 5.10. 4. For an econometric analysis of Turkish trade deficits in the pre-1980 period, see Conway (1986). 5. Salaries of government employees are maintained at their Base Run values in all experiments. Chapter 7 1. For useful accounts of the trade regime in Turkey, see Krueger (1974a) and Baysan and Blitzer (forthcoming). 2. For a detailed study of the structure of domestic resource costs and effective rates of protection in Turkish industry in 198 1, see Yagci (1984). 3. Krueger estimated that the rents on import licenses amounted to 15 percent of Turkish GNP in 1968. 4. The balance is exports of the mining sector. 5. See World Bank (1983a), vol. 2, table 8.10. 6. The proxy was constructed by using a constant value of 10 percent for 1970-79, and using Milanovic’s (1986) estimates for the period thereafter. Part of the measurement error could be due to the fact that Milanovic’s estimates are for ex post subsidies, whereas export behavior is determined by anticipated subsidies. 7. Kopits (1987, fn. 49) reports a much higher long-run export supply elasticity of 2.1 for the shorter period 1977-84, but does not present the regression he estimated. In our regressions, the real exchange rate is no longer statistically significant when lagged more than one quarter (which is consistent with a speedy response arising from excess capacity). We also find that the inclusion of dummy variables (for seasonal effects and for 1981%) reduces substantially the estimated export supply elasticity. 8. When exports are growing rapidly, the presence of delivery lags could lead to substantial divergences between partner-country data. See McDonald (1985, fn. 6). McDonald finds Turkish primary exports to have been susceptible to the incentive to smuggle during the earlier 1962-79 period.

802

Merih Cellsun and Dani Rodrik

Chapter 8 1. See High Control Board (1987) for the shares of public services and SEEs in value added and employment. 2. In pre-1984 data, off-budget subsidies to SEEs are included as a negative item in current transfers. Thus, the estimates for public disposable income are conceptually consistent indicators of the public sector’s net spendable income. 3. Tables A.7 and A.8 in the statistical appendix show the details of the PSBR estimates under variant procedure A for the 1973-85 period. For lack of data, the SEE arrears are included in the “other” category of financing items. 4. In particular, the sources of the very large figure for nondebt capital transfers in 1981, as reported in SPO (1985), could not be verified. 5. See OECD (1984) for a disaggregated review of post-1980 trends in Turkish public finance. 6. In expressing the public debt stock as a percentage of GNP, the year-end dollar values of external debt have been converted to domestic currency by using annual average exchange rates. The figures for domestic debt are year-end data. 7. See Central bank (1987) for a review of capital and money markets in 1985-86. 8. SPO (1985) provides private disposable income, savings, and consumption series both in current and in constant 1983 prices. These series imply, however, somewhat spurious price deflators for private savings. Hence, we have chosen to use current price data in the computation of annual average saving propensities. 9. See Gazioglu (1986) and Maktavli (1986) for regression studies on private consumption and savings for the earlier period. The ex ante and ex post savings ratios in the SPO Annual Programs (from 1973 to 1979) are reviewed in Cellsun (1980). 10. See Ekinci (1987) for a recent macroeconometric modeling study on Turkey, which treats private savings as a variable adjusting to a specified investment behavior. Chapter 9 1. The only significant exception is a $407 million syndicated loan negotiated in 1979 jointly with a rescheduling agreement for the CTLDs. 2. Debt relief here, as in chapter 5, refers to the reschedulings undertaken with official creditors only. There were additional renegotiations over short-term debt owed to private sources (e.g., CTLDs and suppliers’ arrears), but these are not included here. See section 9.2 below for an account of the reschedulings. 3. World Bank (1983a), vol. 2, p. 32. 4. As a member of the OECD, Turkey objected to these negotiations taking place under the auspices of the Paris Club. They were carried out instead in the OECD Consortium for Turkey, even though the general principles followed were the same. 5. Apparently, the central bank still collects from domestic commercial banks the Turkish lira equivalent of the principal repayments on these CTLDs, but at the original exchange rate! Given the thirtyfold depreciation of the lira against the U.S. dollar in the intervening period, these amount to token payments only. 6. The eight banks were Morgan Guaranty, Citibank, Deutsche Bank, Dresdner Bank, Chase Manhattan, Barclays Bank, Swiss Bank Corp., and Union Bank of Switzerland. For an entertaining account of the negotiations, see Bleakley (1978). 7. Stories about the situation circulated at the time both in the domestic and foreign press. For one account, see Business Week, 1 April 1978, p. 92.

803

Turkey/Notes

8. See New York Times, 2 June 1978, p. A l . 9. The quote is from a New York Times editorial of 3 January 1979. 10. The recurring need to obtain the IMF’s “green light” gave rise to the following joke in Turkey, related to Libyan president Qaddafi’s reported promise to send a vast amount of assistance to Turkey were Ecevit to break ties with the U.S. Ekevit was said to have asked Qaddafi when the Libyan money would start flowing in. His answer was: “As soon as you reach agreement with the IMF.” This is reported in Euromoney, June 1979, p. 43. 11. The original is titled “Ecevit schlagt Dollar aus der turkischen Geographie,” Sturtgarter Zeitung, 28 June 1979, p. 4. While wholesale assistance started when Ecevit was prime minister, it is difficult to give him credit for the later flows, as he left office in October 1979. 12. A later interview with Adnan Bager Kafaoglu, the finance minister in power at the time, sheds some interesting light on the attitude of the authorities toward IMF conditionality. Kafaoglu claims that the central bank’s tricks originated during the government of Ecevit in 1978-79. Upon taking office and discovering the cash transactions between the central bank and the Agricultural Bank, he reports being amused by the fact that the transactions involved the physical transportation of cash from one place to the other. Whereupon, he appears to have given instructions that the transactions be carried out simply by writing out and canceling receipts. This way, he reasoned, there was no risk that the money would get stolen along the way! See interview with Adnan Bager Kafaoglu in Colagan (1985). 13. One banker is quoted in the aftermath of the CTLD episode as saying: “With $2.1 billion in foreign exchange reserves at the end of 1974, Turkey could have gone for medium-term loans beginning in early 1975-and still kept to its industrialization plan” (Bleakley 1978, 49). 14. The administrative background is described more fully in Ayse Oktem (1985). 15. Since part of the Dresdner accounts have maturities exceeding one year, they are not, technically speaking, short term. The central bank has recently reclassified Dresdner accounts with longer than a year’s maturity as medium-term debt. 16. An interesting question is whether the Dresdner Bank liabilities should be properly considered as foreign debt, as they are ultimately the central bank’s liabilities to Turkish nationals. Presumably, the intermediation of the Dresdner Bank was supposed to provide an element of confidence to the Turkish workers who may otherwise have been less inclined to repatriate their savings home. It appears, however, that the Dresdner Bank is not liable for the deposits in case the central bank stops servicing them. 17. Business Week, 15 December 1980. Quoted in Sevket Pamuk (1981, 27).

Chapter 10 1. The weighted average debt/GDP, debt/exports, and debt-service ratios for the heavily indebted Latin American countries stood at 31.3, 271.5, and 153.8 percent, respectively, in 1981 (Sachs 1985). 2. This refers to the Dresdner Bank scheme and the foreign exchange deposit accounts in domestic banks of Turkish workers abroad. See table A.17 in the statistical appendix. 3. Two explanations of the high real interest rates are the continuous depreciation of the Turkish lira and the large PSBR. These two explanations are complementary if

804

Merih Celasun and Dani Rodrik

we view foreign and Turkish assets as imperfect substitutes for each other. The Turkish real rate of interest will then exceed the world real interest rate by a margin that equals the expected real depreciation of the domestic currency plus a risk premium, the latter being an increasing function of the outstanding stock of the government’s domestic debt. 4. See, for example, the analysis in de Melo and Robinson (1982).

References Akder, H. 1987. Turkey’s export expansion in the Middle East. The Middle East Journal 41 (4):553-67. Aksoy, A. M. 1982. Structural aspects of Turkish inflation, 1950-79. World Bank Staff Working Paper no. 540. Washington, D.C.: World Bank. Akyiiz, Y. 1984. Turk ekonomisinde mcili yapi ve iligkiler (Financial structure and relations in the Turkish economy). Istanbul: Turkish Industrial Development Bank. Akyuz, Y., and H. Ersel. 1984. Turk ekonomisinde rnrili yapi ve iligkiler: Istarisrik ekler (Financial structure and relations in the Turkish economy: Statistical annexes). Istanbul: Turkish Industrial Development Bank. Balassa, B. 1983. Outward orientation and exchange rate policy in developing countries: The Turkish experience. The Middle East Journal (December):429-47. _ _ . 1984. Adjustment policies in developing countries: A reassessment. World Development 12(9):955-72. -. 1987. The interaction of domestic distortions with development strategies. Report no. DRD249. Washington, D.C.: World Bank. Baysan, T., and C. Blitzer. Forthcoming. The timing and sequencing of trade liberalization: The case of Turkey. Washington, D.C.: World Bank. Bhagwati, J. N. 1971. The generalized theory of distortions and welfare. In Trade balance of payments and growth, ed. J. N. Bhagwati et al. Amsterdam: North-Holland. Bleakley, F. R. 1978. The rush to rescue Turkey. Institutional Investor [international edition] (November):47-64. Boratav, K. 1986. Distribution, external linkages and growth under orthodox policies: The Turkish economy in the early 1980s. Mimeo, University of Zimbabwe, June. Brennan, P. J. 1976. Will Turkey become a big-league borrower? Institutional Investor [international edition] (November):8 1-88, Bulutay, T., N. Yildirim, and Y. Tezel. 1974. Tiirkiye’nin milli geliri: 192348 (National income of Turkey: 1923-48). Publication no. 375. Ankara: Faculty of Political Sciences, Ankara University. Celgsun, M. 1980. Real and monetary aspects in Turkish economic planning. METU Studies in Development 7(112):1-36. -. 1983. Sources of industrial growth and structural change: The case of Turkey. World Bank Staff Working Paper no. 614. Washington, D.C.: World Bank.

805

Turkey/References

_ _ . 1986a. A general equilibrium model of the Turkish economy, SIMLOG-I.

METU Studies in Development 13(1/2):29-94. _ _ . 1986b. Income distribution and domestic terms of trade in Turkey,

1978-83. METU Studies in Development 13(1/2):193-216. Central Bank. 1987. Annual Report, 1986. Ankara: Central Bank. Chenery, H. B., and M. Syrquin. 1975. Patterns of development: 1950-70. London: Oxford University Press. Choksi, A. 1986. Adjustment with growth in the highly indebted middle-income countries. CPD Discussion Paper no. 1986-36. Washington, D.C.: World Bank, October. ColaSan, E. 1983. 24 Ocak: Bir donemin perde arkasi. Istanbul: Milliyet Yayinlari. . 1985. Biz kirk kisiyiz birbirimizi biliriz. Istanbul: Milliyet Yayinlari. Conway, P. 1986. Decomposing the determinants of trade deficits: Turkey in the 1970s. Journal of Development Economics 21(2):235-58. -. 1987. Economic shocks and structural adjustments: Turkey after 1973. Amsterdam: North-Holland. Dervig, K., J. de Melo, and S. Robinson. 1982. General equilibrium models f o r development policy. Cambridge: Cambridge University Press. Dervig, K., and S. Robinson. 1978. The foreign exchange gap, growth and industrial strategy in Turkey: 1973-83. World Bank Staff Working Paper no. 306. Washington, D.C.: World Bank, November. . 1980. The structure of income inequality in Turkey (1950-73). In The political economy of income distribution in Turkey, ed. E. Ozbudun and A. Ulusan. London: Holmes and Meier Publishers, Inc. Dornbusch, R. 1985. Policy and performance links between LDC debtors and industrial nations. Brookings Papers on Economic Activity 2:303-68. . 1986. Dollars, debts and dejicits. Cambridge, Mass.: MIT Press. Ekinci, N. 1987. M%li baski ve Turkiye iqin bir ekonometrik model (Financial repression and an econometric model for Turkey). Ph.D. diss., Institute of Social Sciences, Ankara University, Ankara. Erdilek, A . 1986. Turkey’s new open-door policy for direct foreign investment: A critical analysis of problems and prospects. METU Studies in Development 13(1/2):171 -9 1. Euromoney. 1982. Supplement to Euromoney: Turkey. London, February. Fallen-Bailey, D. G., and T. A. Byer. 1979. Energy options and policy issues in developing countries. World Bank Staff Working Paper no. 350, table VI-9. Washington, D.C.: World Bank, August. Fischer, S. 1986. Issues in medium-term macroeconomic adjustment. The World Bank Research Observer l(2): 163-82. Fry, M. 1986. Turkey’s great inflation. METU Studies in Development 13(1/2):95116. Gazioglu, S. 1986. Government deficits, consumption and inflation in Turkey. METU Studies in Developmenf 13( U2): 117-34. Hatiboglu, Z. 1978. An unconventional analysis of the Turkish economy: An essay on economic development. Istanbul: Aktif Biiro Basim Organization. High Control Board. 1987. Kamu iktisadi teSebbiisleri gene1 raporu, 1985 (General report on state economic enterprises, 1985). Ankara: Prime Ministry.

806

Merih Celasun and Dani Rodrik

Hirschman, A. 0. 1985. Reflections on the Latin American experience. In The politics of injarion and economic stagnation, ed. L. L. Lindberg and C. S. Maier. Washington, D.C.: Brookings Institution. IMF. 1983. Recent multilateral debt restructuring with official and bank creditors. IMF occasional Paper no. 25. Washington, D.C.: IMF, December. -. 1986. IMF Survey. Washington, D.C.: IMF, December. Johansen, L. 1960. A multi-sectoral study of economic growth. Amsterdam: North-Holland. Kafaoglu, A. B. 1986. DCM dosyasi. Istanbul: Alan Yayincilik. Karata?, C. 1986. Public economic enterprises in Turkey-Reform proposals, pricing and investment policies. METU Studies in Development 13(U2): 135-69. Kopits, G. 1986. Structural reform and economic stabilization: A case study of Turkey. Paper presented at a seminar on the Role of the Fund in the International Monetary System, IMF Institute, Washington, D.C. -. 1987. Structural reform, stabilization, and growth in Turkey. IMF Occasional Paper no. 52. Washington, D.C.: IMF, May. Krueger, A. 0. 1974a. Foreign trade regimes and economic development: Turkey. New York: Columbia University Press for the NBER. _ _ . 1974b. The political economy of the rent-seeking society. American Economic Review 64(3):29 1- 303. _ _ . 1981. Interaction between inflation and trade regime objectives in stabilization programs. In Economic stabilization in developing countries, ed. W. R. Cline and S. Weintraub. Washington, D.C.: Brookings Institution. Krueger, A. O., and B. Tuncer. 1980. Microeconomic aspects of productivity growth under import substitution: Turkey. NBER Working Paper no. 532. Cambridge, Mass.: National Bureau of Economic Research. Kubo, Y., and S. Robinson. 1979. Sources of industrial growth and structural change: A comparative analysis of eight countries. Paper presented at the Seventh International Conference on Input-Output Techniques, Innsbruck, Austria. Lewis, J. D. 1986. Coping with adjustment: Turkey, 1973-81. Mimeo, Harvard University, April. Lewis, J. D., and S. Urata. 1983. Turkey: Recent economic performance and medium term prospects, 1978-1990. World Bank Staff Working Paper no. 602. Washington, D.C.: World Bank. McDonald, D. C. 1985. Trade data discrepancies and the incentive to smuggle: An empirical analysis. IMF Staff Papers 32, no. 4 (December):668-92. McKinnon, R. 1982. The order of liberalization: Lessons from Chile and Argentina. In Economic policy in a world of change, ed. K. Brunner and A. H. Meltzer. Amsterdam: North-Holland. Maktavli, S. 1986. Private consumption and savings trends in Turkey, 1963-1983. M.Sc. thesis, Institute of Social Sciences, Middle East Technical University, Ankara. de Melo, J., and S. Robinson. 1982. Trade adjustment policies and income distribution in three archetype developing economies. Journal of Development Economics 1O(February):67-92. Michalopoulos, C. 1987. World Bank Programs for Adjustment and Growth. Report no. VEPERS 11. Washington, D.C.: World Bank.

807

Turkey/References

Milanovic, B. 1986. Export incentives and Turkish manufactured exports, 19801984. World Bank Staff Working Paper no. 768. Washington, D.C.: World Bank. OECD. 1984. Turkey. OECD Economic Surveys, Paris. ___ . 1986. Turkey. OECD Economic Surveys, Paris. Oktem, A. 1985. Turkey. In External debt management, ed. H. Mehram. Washington, D.C.: International Monetary Fund. Okyar, 0. 1965. The concept of etatism. Economic Journal 75, no. 297 (March):98- 11 1. -. 1983. Turkey and IMF: A review of relations. In 1MF conditionality, ed. J . Williamson. Washington, D.C.: Institute for International Economics. Onis, Z. 1986. Stabilization and growth in a semi-industrial economy: An evaluation of the recent Turkish experiment, 1977-84. METU Studies in Development l3( 1/2):7-28. Pamuk, S . 1981. Political economy of industrialization in Turkey. MERlP Reports (January):26 -32. Sachs, J. 1985. External debt and economic performance in Latin America and East Asia. Brookings Papers on Economic ActiviQ 2523-64. __ . 1986. Managing the LDC debt crisis. Brookings Papers on Economic Activity 2:397-431. $eyes, F. 1983. An assessment of Turkey’s liberalization attempts since 1980 against the background of her stabilization program. METU Studies in Development 10(3):271-321. Singer, M. 1977. Economic development in the context of short-term public policies: The economic advance of Turkey, 1938-60. Ankara: Turkish Economic Society Publications, Ayylidiz Matbaasi. SPO. 1985. V c i Plan destek qaliymalari (Fifth plan supplementary studies). Vol. 1. Ankara: State Planning Organization. Stiglitz, J. E., and A. Weiss. 1981. Credit rationing in markets with imperfect information. American Economic Review 71, no. 3 (June):393-410. Syrquin, M. 1986. Growth and structural change in Latin America since 1960: A comparative analysis. Economic Development and Cultural Change 34(3):434-53. Taylor, L. 1979. Macro models for developing countries. New York: McGraw-Hill. Tekeli, I., and S. Ilkin. 1974. Savaj sonrasi ortaminda 1947 Tiirkiye iktisadi kalkinma plani (The 1947 economic development plan of Turkey in the environment of the postwar period). Publication no. 24. Ankara: Faculty of Administrative Sciences, Middle East Technical University. TUSIAD. 1986. Turkiye’de enflusyon ve enjlasyonla savajta bayari kopllari (Inflation in Turkey and conditions for success in the fight against inflation). Report T/86.7.90. Istanbul: Turkish Industrialists and Businessmen’s Association. Walstedt, B. 1980. State manufacturing enterprise in mixed economy: The Turkish case. Baltimore: The John Hopkins University Press. World Bank. 1980. Turkey: Policies and prospects for growth. Washington, D.C.: World Bank, March. -. 1982. Turkey: Industrialization and trade strategy. Washington, D.C.: World Bank. . 1983a. Turkey: Country economic memorandum-Recent economic development and medium-term prospects. Washington, D.C.: World Bank.

808

-.

Merih Celasun and Dani Rodrik

1983b. Turkey: Special economic report, policies for the financial sector. Washington, D.C.: World Bank. ___ . 1984. Turkey: Trade policy issues in the structural adjustment process. Washington, D.C.: World Bank. ___ . 1985. Turkey-The fifih five year plan in the context of structural adjustment. A review. Washington, D.C.: World Bank. Yagci, F. 1984. Protection and incentives in Turkish manufacturing. World Bank Staff Working Paper no. 660. Washington, D.C.: World Bank. Yagci, F., S. Kamin, and V. Rosenbaum. 1985. Structural adjustment lending: An evaluation of program design. World Bank Staff Working Paper no. 735. Washington, D.C.: World Bank. Yaprak, T. 1982. Log-linear method for solution of general equilibrium models: Design of a computer program and numerical application. M.Sc. thesis, Department of Operations Research, Middle East Technical University, Ankara.

Biographies

Merih Celbun is a professor of economics at the Middle East Technical University in Ankara, Turkey. Susan M. Collins is an associate professor of economics at Harvard University and a faculty research fellow of the National Bureau of Economic Research.

Robert S. Dohner is a visiting professor at the Graduate School of Business, Stanford University. Ponciano Intal, Jr. is an assistant professor and chairman, department of economics, University of the Philippines at Los Banos. Anwar Nasution is lecturer in monetary economics, Faculty of Economics, at the University of Indonesia. Won-Am Park is a fellow of the Korea Development Institute. Dani Rodrik is an associate professor of public policy at the John F. Kennedy School of Government, Harvard University and a faculty research fellow at the National Bureau of Economic Research.

Jeffrey D. Sachs is the Galen L. Stone Professor of International Trade at Harvard University and a research associate at the National Bureau of Economic Research. Wing Thye Woo is an assistant professor of economics and head of the Pacific Rim Studies Program at the University of California, Davis.

809

Contributors

Merih Celisun Department of Economics Middle East Technical University Ankara, Turkey

Won-Am Park Korea Development Institute P.O. Box 113, Chungryang Seoul, Korea

Susan M. Collins Department of Economics Harvard University Littauer Center M-7 Cambridge, MA 02138

Dani Rodrik John F. Kennedy School of Government Harvard University 79 Kennedy Street Cambridge, MA 02138

Robert S. Dohner Graduate School of Business Stanford University Stanford, CA 94305-5015 Ponciano Intal, Jr. Department of Economics University of the Philippines at Los Banos College, Laguna, The Philippines Anwar Nasution Institute for Social and Economic Studies Faculty of Economics University of Indonesia Jalan Ray Salemba 4 Jakarta 10430, Indonesia

810

Jeffrey D. Sachs Department of Economics Harvard University Littauer Center M-14 Cambridge, MA 02138 Wing Thye Woo Department of Economics University of California Davis, CA 95616

Name Index

Abueva, Jose V., 377, 385, 594n1.8 Adelman, I., 300 Agcaoili, Mercedita, 463 Aghevli, B., 154, 219, 358n3.2, 359n4.8, 3621113.1 Akder, H., 683, 800114.3 Aksoy, A. M., 798111.9 Akyiiz, Y., 789111.10 Alburo, Florian, 442, 568, 598nn3.1,3.5, 3.6 Amatong, Juanita, 597n2.31 Amsden, A., 281, 359n4.8, 3611110.9 Aquino, Benigno, 374, 559 Aquino, Corazon, 407, 473, 561-62 Arafin, Bustanil, 63 Arndt, H. W., 6, 91, 142nn3.4.3.6, 144n6.4 Asian Development Bank, 600n5.2 Atatiirk, M. Kemal, 619 Bacha, E., 359115.2 Balassa, Bela, 2, 632-33, 679, 800nn4.1,4.2 Baldwin, Robert, 439, 442, 594111.6, 598nn3.1,4.7 Barker, R., 473, 599114.6 Bautista, Romeo, 598nn3.1,3.1 I ,3.18 Baysan, T.,801n7. I Benedicto, Roberto, 463-64, 470-74, 491 Bhagwati, J. N., 679 Bleakley, F. R., 644,799n2.13, 802n9.6, 803n9.13 Blitzer, C., 801117. I Booth, Anne, 68

Boratav, K., 654, 800114.2 Brennan, P. J., 641, 653, 799n2.12 Briones, Leonor, 421 Brown, J., 599n4.2 Bucheit, Lee, 604117.15 Buenaventura, Corazon, 423 Bulutay, T.,797111.4 Callison, C. Stuart, 410 Campos, Jose, 470 Canlas, Dante, 464, 472, 599nn4.4,4.10, 601115.14, 606118.1 1 Canoy, Reuben, 599n4.15 Carter, Jimmy, 756 Cellsun, Merih, 5-6, 623, 697-98, 703, 707-8, 797111.6, 798n1.8, 799nn2.17,3.1, 800nn4.4,5.7, 801n5.9, 802118.9 Central Bank, 802118.7 Chang Myon, 155, 184 CheneIy, H., 167, 623 Cho, Y. J., 290-91, 297, 300, 362111 1.10 Choo, H., 300-301, 314 Clarete, Ramon, 468 Cline, William, 146n8.3 Clunies-Ross, James, 598113.18 Cojuangco, Eduardo, 417, 450, 467-68, 470, 472, 474, 491 Colaaan, E., 759, 803n9.12 Cole, D., 236, 290-92, 295, 297, 300, 358nn3.4,3.6,4.2, 359114.5. 361nl1. I , 362nn11.9, 11.10 Collins, Susan M., 3, 4, 364 Conway, P., 800115.2, 801116.4

811

812

Name Index

Corbo, V., 289-90, 3611111.6 Corden, Max, 100-101, 144n6.1, 145n6.21 Coronel, S., 599n4.12 Crouch, Harold, 142n3.13 Cuddington, John, 146118.4, 603n6.9 Cuenca, Rodolfo, 470-72, 477, 500 Cumby, Robert, 603n6.9 Dake, Antonie, 60 David, Cristina, 473, 599n4.6 de Borja, E., 594111.7 de Dios, Loreli, 597n2.26, 598n3.9 Dee, Dewey, 471, 477, 491, 496, 502 de Melo, J., 706, 804n10.4 Demirel, Siileyman, 621, 635, 663, 759 DerviS, Kemal, 645-46, 688, 698, 703, 706 de Vries, Barend, 434 Diaz Alejandro, C., 270 Diokno, Benjamin, 600114.19 Disini, Heminio, 407, 470-73, 477-78, 497 Doherty, John F . , 599nn4.15.4.16, 601115.11 Dohner, Robert S., 603nn6.9,6.10,6.11, 605118.7, 607n8.30 Dooley, Michael, 146118.4 Dornbusch, Rudiger, 146n8.3, 176, 231, 270, 358n 1.2, 359nn4.8,7.1, 360118.3, 663, 665, 798112.2 Ecevit, Bulent, 635, 663, 756-57 Eichengreen, B., 593111.3, 597n2.14 Ekinci, N., 802118.10 Emery, Robert, 601115.16 Enrile, Juan Ponce, 385, 467 Erdilek, A,, 800114.8 Esguerra, Emmanuel, 605117.21 Espinosa, Roberto, 473, 599n4.13 Euromoney, 667, 800nn4.2.4.6 Fischer, S., 679 Floirendo, Antonio, 470, 473, 478 Frank, C., 167, 185, 190, 360n9.1 Fly, Maxwell, 602nn5.30,5.31 Gablis, V., 90 Gazioglu, S., 802n8.9 Geertz, Hildred, 60 Gillis, Malcolm, 144n6.7, 145116.14 Giovannini, A., 186, 241, 602115.31 Glassburner, Bruce, 142n3.2, 143n3.20, 144116.2, 146nn7.2,7.3 Golay, Frank H., 379, 593nnl.2,I .3, 594nnl.4,1.5, 597n2. I5

Government of Korea, 205, 360n8.4 Gratz, W., 593111.3, 597n2.14 Gray, Clive, 91 Gregorio, Rosario. 598113.11 Gurley, J . , 358114.2 Haggard, S., 359n4.8, 361n10.9, 606118.22 Haque, Nadeem, 146118.4 Haryono, Subaliono, 127, 146n7.6 Hatiboglu, Z . , 798n1.8 Hawes, Gary, 468 Hicks, George L., 598113.7 High Control Board, 802118. I Hirschman, Albert, 607118.38 Hong, W. T., 184-85, 187. 293, 358nn3.4,3.6, 359n4.6, 360n9.1, 3621111.7 Hooky, Richard, 459, 598n3.3, 599n3.25 Hoon Yu, 308 Huhne, Christopher, 146118.3 Hunter, Alex, 120 Hutton, C. Peabody, 604117.15 Ibon Facts and Figures, 478, 600nn4.22.4.23 Ilkin, S., 797111.3 inonii, Ismet, 620 Intal, Ponciano, Jr., 464, 467-68, 593 International Labor Office (ILO), 412, 441, 606n8.12 International Monetary Fund (IMF), 66, 72, 110, 410, 481, 485, 493, 495, 524, 539, 553, 556, 567, 597nn2.19,2.22,2.23,2.27,2.30, 600n5.1, 601nn5.9,5.11,5.17, 602n6.5, 603n6.7, 605117.20. 800114.5 Jacinto, Fernando, 473 Jackson, Karl, 64,142n3.2 Johansen, L., 703 Johnson, Bryan, 605n8.4 Jones, L. P., 198, 361n10.7 Jung, Y. D., 300 Kafaoglu, Adnan Bager, 803n9.12 Kanesa-Thasan, S . , 144n6.I KarataS, C., 798nl.11 Kessler, Richard J . , 580 Khan, Mohsin, 146118.4 Kim, D. M.. 308 Kim, K. S., 167, 185, 190, 226, 228, 359n7.2, 360119.1 Kim Sookon, 278-79, 360n10.4, 361nn10.5,10.6,10.9

813

Name Index

Kincaid, G. Russell, 144n6.6, 145116.13 Kintanar, Agustin, Jr., 597n2.16 Koo, B. Y., 360n9.2 Kopits, G . , 800nn4.1,4.2,4.7, 801117.7 Koppel, Bruce, 606118.25 Korea Productivity Center (KPC), 358114.1 Krueger, Anne O., 178, 358nn3.4,3.6, 679, 710, 713, 717- 18, 797nn1.1,1.6, 801nn7.1,7.3 Kuznets, S., 316 Lal, Deepak, 594111.7 Lamberte, Mario, 427, 484, 489, 500, 597n2.32, 600nn4.24.4.25, 601n5.18 Lande, Carl, 567 Laya, Jaime C., 500 Lever, Harold, 146118.3 Levich, Richard, 603n6.9 Lewis, J . D., 703, 798112.1 Licaros, Gregorio, 494 Licuanan. Victoria S., 485-86 Liddle, William, 142n3.2, 143n3.23 Lindauer, D. L., 278-81, 3601110.2 Lipsky, Seth, 146n7.2 Lyman, P., 358nn3.4.3.6 Macapagal, Diosdado, 38 I McCawley, Peter, 65, 68, 145116.16, 146n7.2 McDonald, Hamish, 123, 801n7.8 McDougald, C., 472 Mackie, J. A. C., 144n6. I McKinnon, Ronald, 90, 358114.2, 800114.9 Magsaysay, Ramon, 376 Maktavli, S., 802n8.9 Manasan, Rosario, 423, 597n2.14 Marcos, Ferdinand, 376, 381-84, 462, 470 Marcos, Imelda, 410, 470, 472 Marquez-Ruarte, J., 154, 194, 358113.2, 359n4.8, 3621113.1 Mason, E., 283, 300, 358nn1.2,2.1, 361nn10.7.11.1, 362nn12.2,12.3 Medalla, Erlinda, 598nn3.1,3.15 Mijares, Primitivo, 473, 599n4.14 Milanovic, Branko, 723, 801117.6 Montelibano, Teodoro, 478, 498 Montes, Manuel M., 556-57, 597n2.26, 605nn7.23,7.27 Moon, C., 359n4.8, 361n10.9 Moreno, Honorata, 49 1, 493, 601nn5.11,5.14 Morgan Guaranty, 146118.4, 359n4.13 Mortimer, Rex, 142n3.2

Nam, S. W., 197, 200, 289-90, 359n4.8, 360nn8.3,9.1, 361nl1.6 Nasution, Anwar, 90, 144nn5.1,5.2,5.6 National Economic and Development Authority (NEDA), Philippines, 468, 599n4.8, 606n8.14 Nelson, Gerald, 463 Ocampo, Sheilah, 468 OECD, 802118.5 Oktem, A,, 803n9.14 Okyar, O., 797n1.2, 800nn4.2,4.7 Ongpin, Roberto, 409, 525-26 Onis, Z., 800114.2 Oshima, Hany, 594nl.7, 596n2.11 Ozal, Turgut, 664, 666-67, 759 Palacpac, A,, 473, 599n4.6 Palmer, Ingrid, 142n3.2, 143n3.22 Pamuk, 5.. 803n9.17 Pangestu, Mari EIka, 145nn6. 11.6.13 Panglaykim, Jusuf, 143n3.22 Park, J. K.,226, 228, 359n7.2 Park, Won-Am, 3, 4, 207, 215-18 Park, Y. C., 187, 193, 207, 212-14, 219, 231, 236, 265, 290-92, 295, 300, 358nn1.2,4.2, 359nn4.5,7.1, 360nn8.3,9.1,10.1, 361nll.1, 362n11.9, 812 Park Chung Hee, 155, 157, 184, 199, 201 Pascual, Alfredo, 600115.1 Patrick, Hugh, 358114.2, 491, 493, 601nn5.11,5.14 Paz, W., 594n1.7 Penny, D. H., 59 Perkins, Dwight, 144116.7 Philippines, 569-70, 605nn7.25,8.5,8.8, 606n8.15 Pitt, Mark, 144116.2, 145116.22 Posthumus, G. A., 115, 143n3.24 Power, John H., 377, 442-43, 451, 464, 467-68, 598nn3.1,3.11,3.15 Presidential Commission on Government Reorganization, 596n2.12, 597n2.25 Quiambao, D., 471 Race, Jeffrey, 594111.10 Rao, D. C., 300 Renaud, B . , 300 Reyes, L., 604117.15 Rivera-Batiz, Francisco, 146nn8.1,8.2 Rivera-Batiz, Luis, 146nn8.1,8.2

814

Name Index

Robinson, S., 167, 300, 645-46, 688, 698, 703, 706, 804n10.4 Rodrik, Dani, 5-6 Roemer, Michael, 144116.7 Roh Tae Woo, 157 Romualdez, Alfredo, 470 Romualdez, Benjamin, 470, 472-73 Roumasset, James, 468

Taylor, L., 702, 707 Tekeli, I., 797n1.3 Tengco, Jose, 427, 471 Thakur, S., 360118.6 Thalib, Dahlan, 59 Tidalgo, Rosa Linda, 605117.21 Timmer, Peter, 142n3.12 Tuncer, B., 797111.6 TUSIAD, 800n4.11

Sacerdoti, Guy, 472 Sachs, Jeffrey, 142n3.I , 598113.2, 663, 803n10.1 Sakong, 1. L., 198, 308, 361n10.7 Senga, K., 442 SenSes, F., 800n4.2 Shaplen, Robert, 384 Shaw, E., 358114.2 Shepherd, Geoffrey, 442, 598nn3.1,3.5,3.6,3.21 Shin Hyon Whak, 200 Sicat, Gerardo, 377, 442-43, 451, 472, 595111.4, 598nn3.1,3.6, 599n4.11 Silverio, Ricardo, 470-72, 497, 500 Singer, M., 797n1.5 Snodgrass, Donald, 144116.7 Soeharto, President of Indonesia, 20 Soekamo, President of Indonesia, 21, 32-33 SPO, 8, 802nn8.4,8.8 Stiglitz, Joseph, 647 Suh, S. M., 310, 362n12.4 Sundararajan, V., 360118.6 Sundhaussen, Ulf, 57 Sutowo, Ibnu, 120-22, 127 Suwidjana, Njoman, 144115.1 Syrquin, M., 167, 623, 800114.4 Szall, R. J . , 300, 309

Walstedt, B., 798111.11 Warr, Peter, 145116.8 Watson, Maxwell, 605n7.18 Weiss, A., 647 Westphal, L., 167, 185, 190, 198, 360n9.1 Wideman, Bernard, 464, 471 Woo, Wing Thye, 6, 142n3.3 World Bank, 59, 110, 115, 128, 144n4.5, 145116.22, 146n7.1, 153, 330, 358111.2, 360nn7.6,8.1,9.1, 361n10.13, 362111 1.10, 387, 411-12, 419, 445, 481, 485, 493, 495, 556, 596n2.10, 597n2.18, 598nn3.13,3.14,3.19,3.20,3.22, 600115.1, 601nn5.9,5.11,5.17, 605nn7.23,7.24,8.1, 606nn8.23,8.24, 607nn8.32,8.34,8.35,8.36, 719, 799n2.11, 800nn4.2S.5, 801117.5, 802n9.3

Tait, Alan M., 593n1.3, 597n2.14 Tan, Edita, 442, 499-501, 598113.18, 601n5.9

Yagci, F., 800114.2, 801n7.2 Yaprak, T., 703 Yusuf, S., 360118.5

Urata, S . , 703 van Wijnbergen, S., 186, 360118.6 Veloso, R., 471 Villaruel, Teresita, 604117.15 Villegas, Bernardo, 599n4.16 Virata, Cesar, 468, 598n3.12

Subject Index

Agricultural sector: Soeharto regime, 60-61; Turkey, 622, 673-74 Aquino administration, 425, 473, 524, 561-66, 589-93 Army: Indonesia, 62; Philippines, 385, 388, 425 Asian Development Bank, 505 Association Agreement (1963) and Additional Protocol, 622 Balanced budget rule, Indonesia, 21, 56, 58-59, 84-86, 107, 138 Balance of payments: Indonesia, 84, 98-99, 107; Philippines, 447 Bank Indonesia, 24, 25, 37 Bank Indonesia Act of 1953, 36 Banking system: Indonesia, 87; Korea. 187, 291-93, 295-97; Philippines, 482, 484, 490-91; Turkey, 670 Bank Negara Indonesia, 37, 42 Bank of Korea (BOK), 180 BAPPENAS. See National Planning Body Barriers to entry, nontariff Philippines, 44243 BASECO. See Bataan Shipyard and Engineering Company Bataan nuclear power project, 407 Bataan Shipyard and Engineering Company (BASECO), 473 Bell Act. See Trade Agreement Act of 1946 Big Push, Korea (1973-79): effect of, 192, 197-98, 251, 268-69; exchange rate policy and, 256; government intervention

and, 295-97. See also Heavy and chemical industries Board of Investments (BOI), Philippines, 440, 444-46 BOK. See Bank of Korea Borrowing, domestic: Korea, 187; Philippines, 382 Borrowing, foreign: Korea, 184-87, 22122; Philippines, 382, 393-95, 398, 401, 493, 516; Turkey, 631-32, 637 Budget, government: Indonesia, 38-40; Philippines, 430-33 BULOG, Indonesia, 79, 139 Business firms, Korea, 172 Capital flight: Indonesia, 20, 94; Philippines, 393, 400, 516-17 Capital inflows, foreign: Korea, 187, 19697, 293 Capital markets: Philippines, 481 -82; Turkey, 626 Cargill (firm),476 CCCN. See Customs Cooperation Council Nomenclature CCSF. See Coconut Consumer Stabilization Fund CDCP. See Construction and Development Corporation of the Philippines Ceilings on crediudebt: Indonesia, 24, 8485, 90-91; the Philippines, 522-23 Central hank: Indonesia, 24; Philippines, 427-29, 483-84, 519; Turkey, 638 Chang Myon administration, Korea, 250

815

816

Subject Index

Civil Relief in Korea (CRIK), 163 COCOFED. See Coconut Producers Federation Coconut Consumer Stabilization Fund (CCSF), 467 Coconut industry, Philippines, 466-69 Coconut Industry Development Fund, 417, 456, 467 Coconut Investment Company, 466 Coconut Investment Fund, 466 Coconut Oil Mills (UNICOM), 467-68, 476 Coconut Producers Federation (COCOFED), 466-68 Colonial period, Korea, 159-60 Commodity markets: Indonesia, 117; Philippines, 389-90, 456-57, 462. See also Exports; Oil export revenues; Oil imports Communist Party of the Philippines (CPP), 385 Competitiveness, international: Korea, 191 Comprehensive Stabilization Plan (CSP), Korea, 199-200, 288, 297 Construction and Development Corporation of the Philippines (CDCP), 470 Consultative Group for the Philippines, 383 Contribution to Regional Development (IPEDA), Indonesia, 68, 71 Convertible Turkish lira deposits (CTLDs), 640-55, 753-54, 760-61 Counterfactual experiments, Turkish economy, 710-15 CPP. See Communist Party of the Philippines Credit markets, international, 28, 85, 116 Credit system: Indonesia, 24, 84; Philippines, 489-91. See also Ceilings CRIK. See Civil Relief in Korea Crony capitalism, Philippines: the economy and, 399, 459, 460; fiscal activity and, 415, 477-79; institutionalization of, 46981 CSP. See Comprehensive Stabilization Plan CTLDs. See Convertible Turkish lira deposits Curb market, Korea. See Financial institutions Currency devaluation: Indonesia, 26, 29, 56-57, 59-60, 71-72, 99- 113; Philippines. 447-49. See also Dutch disease Current account: Indonesia, 33-34, 98-99, 110; Korea, 196, 207- 15, 229-34; Philippines, 382, 391, 398, 447, 514, 55051; Turkey, 638

Customs Cooperation Council Nomenclature (CCCN), 261 DBP. See Development Bank of the Philippines Debt, external: Aquino government plan for renegotiation, 575-78; Indonesia, 30, 114- 15; Korea, 170-72; Philippines, 383, 391, 395-96, 400, 503, 511, 51719, 522-24, 541 -46; Turkey, 620, 631, 638-40, 665-66, 739-42. 753-56. See also OECD Aid Consortium Debt accumulation: Korea, 186-87, 19198; Turkey, 628, 716 Debt Certificates (SBIs), Indonesia, 25, 9293 Debt crisis: Indonesia, 19, 29; Philippines, 433, 459, 501-3, 524-58; Turkey, 63031, 664-65, 750-51. See also Trade policy DebUexport ratio, Indonesia, 118- 19 Debt-service ratio: Indonesia, 29-30, 110, 119; Philippines, 391-93, 512, 515; Turkey, 68 1, 683 Democratic Justice Party (DJP), Korea, 157 Democratic Party (DP), Turkey, 620 Development assistance, Philippines, 387 Development Bank of the Philippines (DBP), 427, 484 DJP. See Democratic Justice Party Dutch disease, Indonesia, 25-26, 100- 102, 105-6 Dwifungsi, Indonesia, 62-63, 127, 142 ECA. See Economic Cooperation Administration Economic advisors. See Technocrats Economic Cooperation Administration (ECA), U . S . , 163 Economic growth: Indonesia, 28, 33-38; Korea, 166, 186-87, 189-91, 219-20, 226-29, 250; Philippines, 390, 433; Turkey, 623-24, 630, 680 Economic nationalism: Indonesia, 22, 6263; Philippines, 378, 440 Economic planning, Korea, 184-86 Economic Planning Board (EPB), Korea, 180-81 Economic policy: Indonesia, 31 -32, 56; Philippines, 396, 566-69; Turkey, 619, 666-71 Educational expansion, Korea, 164-65

817

Subject Index

EEC. See European Economic Community Egalitarianism, Korea, 164-65 Employment policies, Korea, 279 Energy projects, Philippines, 405-8 EPB. See Economic Planning Board Etatism, Turkey, 619, 626 Ethnic groups, Indonesia, 60-61 European Economic Cummunity (EEC), 622-23 Exchange rate: Indonesia, 19, 20, 26, 56, 100-101; Korea, 185-86, 255; Philippines, 438, 440, 448-50; Turkey, 674 Exchange rate devaluation, Korea, 188, 194-95, 202 Exchange rate, multiple, Indonesia, 20, 25, 36, 40, 97 Exchange rate policy: Indonesia, 20-22, 25-28, 31, 67, 97-98; Korea, 256; Philippines, 447-51, 525-27; Turkey, 718 Export certificate plans, Indonesia, 45, 48 Export growth: Korea, 155; Philippines, 434, 440, 453-57; Turkey, 720-31. See also Export policy; Investment incentives Export Import Bank (U.S.), 407 Export Incentives Act of 1970, Philippines, 443-44,455 Export market: Indonesia, 30, 43-48, 77, 79; Korea, 155-56, 255, 259-61; Philippines, 383-84, 391-93, 434-35, 440, 515 Export policy: Philippines, 398, 408, 433, 455-56, 585-88; Turkey, 683-84, 716, 720-25 Exports, nonoil, Indonesia, 23, 56, 101-2, 105-7, 109 Export taxation, Philippines, 412, 462 FCDUs. See Foreign currency deposit units Financial crisis: Korea, 189, 294; Philippines, 393, 399, 411, 498 Financial deepening: Indonesia, 25; Philippines, 491 Financial institutions: Indonesia, 87; Korea, 291-94, 297-98; Philippines, 481-85, 487-91, 493-96, 570-71. See also Banking system Financial liberalization, Korea, 292-95, 298 - 300 Financial system: Indonesia, 24-25, 56, 8991; Korea, 283-84, 290-300; Philippines, 399-400, 498-99, 588-89; Turkey, 626

Fiscal policy: Indonesia, 23-24, 109; Korea, 189, 202-3, 287-89; Philippines, 41218, 443, 447, 569-70; Turkey, 731-32, 736-39, 748-50 Fixed capital formation, Korea, 242-43 Foreign aid: Indonesia, 28; Korea, 155-56, 160-66; Turkey, 619-20, 680-81, 75153. See also International Monetary Fund; OECD Aid Consortium; World Bank Foreign borrowing. See Borrowing, foreign Foreign Capital Inducement and Promotion law (1983), Korea, 185, 187, 269 Foreign currency borrowing, Philippines, 493 Foreign Currency Deposit System, Philippines, 504, 519 Foreign currency deposit units (FCDUs), 504 Foreign exchange. See Exchange rate Foreign Exchange Control Act, Korea. 172 Foreign exchange markets, Philippines, 48687 Foreign investment, Philippines, 379-80, 390. 41 1

GARIOA. See Government Appropriations for Relief in Occupied Areas General Electric (firm),407 GOLKAR party, Indonesia, 57 Government Appropriations for Relief in Occupied Areas (GARIOA), U.S., 163 Government intervention: in domestic market (Philippines), 389; in financial markets (Korea), 295-97; with firms during financial crisis (Korea), 294-95; in sugar trading (Philippines), 462 Government Service Insurance System (GSIS), Philippines, 425, 470, 484 Government spending: after second oil shock (Philippines), 393, 408- 12; energy projects (Philippines), 405-8; financing by foreign sources (Indonesia), 115, 117; under martial law (Philippines), 401-5; Philippines, 388-90, 401, 411, 430, 433; stimulation in Indonesia by oil export revenue, 84; Turkey, 731-32. See also Gross domestic product, regional; State enterprises Grain Management Fund, Korea, 189, 197, 285 Gross domestic product, regional (RGDP), Indonesia, 79

818

Subject Index

Gross national product (GNP), Korea, 22124 GSIS. See Government Service Insurance System Heavy and chemical (HC) industries, Korea, 191, 193-94, 197-98. See also Big Push Homogeneity of Korean society, 302 Human Settlements Development Corporation, Philippines, 410 Hyperinflation, Korea, 164 ICOR. See Incremental capital output ratio IGGI. See Inter-GovernmentalGroup on Indonesia IMF. See International Monetary Fund Import controls: Indonesia, 27. 56-57, 6263, 111-13; Korea, 261-63; Philippines, 438-40.451, 461; Turkey, 717-18 Imports: Korea, 225-26; Turkey, 716 Import stimulation, Indonesia. 85 Import substitution: Philippines, 406,4089, 433, 451-52; Turkey, 619, 717 Incentives, industrial. See Industrial policy; Investment incentives Income distribution: Korea, 300; Turkey, 659-62 Incremental capital output ratio (ICOR): Philippines, 397; Korea, 221 -22 Indonesia: economic policy influenced by political considerations, 21; fiscal system in, 23-24; independence (1949). 32; influence of political constituency in, 20; legacy of economic policies (1950-69, 20; political conflict with Singapore and Malaysia, 36; political system in, 21 Indonesian Communist Party (PKI), 21, 60, 67, 77 Industrial Incentives Act of 1967, Philippines, 379-80,440. 443-44, 455 Industrial policy: Korea, 265-69; Philip pines, 408-9, 433-37, 443-46, 458, 572-73; Turkey, 619-20, 717. See also Export growth; Heavy and chemical industries; Import substitution Industrial sector, Indonesia, 61-65 Industries, intermediate, Philippines, 458-59 Inflation: Indonesia, 20-21, 25, 36-38, 4243, 58-59; international, 42; Korea, 164, 168-69, 186-87, 189, 191-92, 195-97, 202-3; Philippines, 440, 450, 488, 531, 546, 551-56; Turkey, 620, 658-62, 674, 680. See also Hyperinflation, Korea

INPRES program allocations, Indonesia, 77 Institutional memory, effect in Indonesia of, 31, 56, 67, 96 Integration policy, Indonesia, 60-61 Interest rates: Korea, 186; Philippines, 48789, 501-3; Turkey, 626 Inter-GovernmentalGroup on Indonesia (IGGI), 28, 30, 115- 19, 135-36 International Monetary Fund (IMF): Korea, 202, 294; Philippines, 383, 441, 523-25. 532-40, 557-58; Turkey, 620-21, 656,

672- ?2

Investment: Korea, 242-48; Philippines, 382, 390-91, 396-97, 443, 493; Turkey, 636-37, 684, 747-48 Investment incentives, Philippines, 443-47, 455. See also Industrial policy Investment Incentives Acts (1967, 1983), Philippines, 443, 446 Investment Priorities Plan, Philippines, 44445 Investment program, K m a , 156-57. See also Big Push IPEDA. See Contribution to Regional Development Islamist National Salvation Party, Turkey, 635 Japan: and development of Korean financial system, 282-83; effect of occupation of Korea, 302-3; influence on Korean economy, 160, as source of financial aid to the Philippines, 505 Javanese peasantry, Indonesia, 20, 21, 67, 77 KDB. See Korea Development Bank KDI Quarterly Macroeconomic model, 2012, 206, 216-18 KEB. See Korea Exchange Bank KEXIM. See Korea Export Import Bank Korea Development Bank (KDB), 180 Korea Exchange Bank (KEB), 180 Korea Export Import Bank (KEXIM),180 Korean War:and Indonesia, 33, 34, 42; and South Korea, 155, 161, 304; and lbrkey, 620 Korea Reconstruction Bank, 185

Labor force,Korea, 280-81, 305-6 Labor Standards law, Korea, 278 Land reform: Korea, 165-66, 303-4; Philippines, 386, 388, 390, 418, 573-75

819

Subject Index

Laurel-Langley Agreement of 1955, US.Philippines, 379-80, 439 Law for Fostering Capital Markets, Korea, 187 Lending, domestic. See Borrowing, domestic Lending foreign. See Borrowing, foreign Lever Brothers (firm), 476 Loan guarantee system, Korea, 185-87, 293-94 Loans, concessional, Indonesia, 30 Macroeconomic policy, Philippines, 383 Major Industrial Projects (MIPS), Philippines, 408-10, 445 Malaysia, 409 Management of External Debts and Investment Accounts Department (MEDIAD), Philippines, 503-4, 522 Manufacturing sector: Philippines, 435-37, 439-40, 445-46, 451, 458-59; Turkey, 717 Marcos administration, 381-83, 401; banks associated with, 491; and crony capitalism, 469-76; and energy projects, 405; final years, 558-61; opposition to and effect of, 514-15; re-election and martial law, 384-90, 396, 401, 417-18, 434, 470. See also Crony capitalism Marinduque Mining Corporation, Philippines, 47 1, 500 Marshall Plan assistance, 619 Martial law, Philippines, 385-89, 395, 401, 417-18, 434, 470 MEDIAD. See Management of External Debts and Investment Accounts Department Military personnel, 62. See also Army Ministry of Finance: Indonesia, 65-66; Korea, 172 MIPS. See Major Industrial Projects Monetary policy: Korea, 189, 202-3; Indonesia, 24, 41-43, 48, 56, 109; Philippines, 382, 447, 487-89, 529-31, 55156; Turkey, 674-78 Money market, Philippines, 484-86 Money Market Instruments (SBPUs), Indonesia, 25, 92-94 Money supply, Indonesia, 84 NASUTRA. See National Sugar Trading Corporation National Development corporation (NDC), Philippines, 409, 41 1

National Economic and Development Authority (NEDA), Philippines, 386 National Electrification Administration (NEA), 407 National Investment Fund (NIF), Korea, 194, 267 National Investment Trust (F'T Danareksa), Indonesia, 88 Nationalization, Indonesia, 36 National Planning Body (BAPPENAS), Indonesia, 65, 122 National Power Corporation (NPC), Philippines, 406-8. See also Bataan nuclear power project National Sugar Trading Corporation (NASUTRA), Philippines, 463-65 NDC. See National Development Corporation NEA. See National Electrification Administration NEDA. See National Economic and Development Authority New Order government, Indonesia, 20 New People's Army, Philippines, 385, 425 NIF. See National Investment Fund Nivico (firm), Philippines, 417, 472-73 NPC. See National Power Corporation ODA. See OECD Development Assistance Committee OECD Aid Consortium, 664, 666, 754, 757 OECD Development Assistance Committee (ODA), 387 Offshore banking units, Philippines, 487 Oil export revenues, Indonesia: decline (1982), 29, 70-71, 110; effect (197178), 25, 84; importance (1970s), 23, 70 Oil price shocks: Philippines, 390, 393, 396, 405- 12; Turkey, 680 Open market instruments, Indonesia, 25, 92-96 Outer Islands, Indonesia, 20, 21 Overseas Private Investment Corporation (OPIC), U.S., 476 Ozal administration, 732-33 Park administration, 155-56, 184, 199, 201, 250, 283-84 Partition, Korea, 161 Patronage. See Crony capitalism, Philippines Pertamina (firm), Indonesia, 28-29, 64-65, 120-27, 135-36 PHILEX. See Philippine Exchange Company

820

Subject Index

Philguarantee. See Philippine Export and Foreign Loan Guaranty Corporation Philippine Coconut Authority, 467 Philippine Exchange Company (PHILEX), 462-63, 465-66 Philippine Export and Foreign Loan Guaranty Corporation (PhilGuarantee), 427 Philippine National Bank (PNB), 427, 463, 47 1,482 Philippine National Oil Company (PNOC), 407,471 Philippine Sugar Commission (Philsucom), 463 Philippine Veterans Investment and Development Company (PHIVIDEC), 416 Philsucom. See Philippine Sugar Commission PHIVIDEC. See Philippine Veterans Investment and Development Company PKI (Parrai Kommunis Indonesia). See Indonesian Communist Party PNB . See Philippine National Bank PNOC. See Philippine National Oil Company Political constituency, Indonesia, 20-22, 67, 96 Political parties, Indonesia, 57 Political system: Philippines, 376, 383, 38788; Turkey, 626 Pribumi, Indonesia, 24, 89, 122 Price control: Indonesia, 43; Korea, 197, 203; Philippines, 53 1 Price ratio, Indonesia, 103-4, 109 Prices, macroeconomic, Turkey, 684-95 Privatization: Indonesia, 63; Philippines, 571-72 F’rocter and Gamble (firm), 476 Productivity differentials, Turkey, 698 Productivity growth, Philippines, 457-58 Progressive Manufacturing Program (PMP), Philippines, 445 Protectionism: Indonesia, 56-57, 100, 109; Korea, 251; Philippines, 437-47, 45758; Turkey, 717. See also Exchange rate; Import controls; Quantitative restrictions; Quotas PSBR. See Public sector borrowing requirement PT Danareksa. See National Investment Trust Public corporations/enterprises.See State enterprises Public sector, overlap with private in Indonesia, 63. See also Dwifungsi

Public sector borrowing requirement (PSBR), Turkey, 637, 692, 695, 736-39 QRs. See Quantitative restrictions Quantitative restrictions (QRs), Indonesia, 27, 111- 13. See also Import controls; Quotas Quotas, Indonesia, 27, I1 I Rebellions, Indonesia (1950-65), 20-21, 36 Recession, Philippines, 408, 41 1 Recession, world (1982), 71, 91, 107, 110, 41 1 Reconstruction, postwar: Korea, 163, 166; Philippines, 438 Regional Development Banks (RDB), Indonesia, 87 Republic Act 6142 (1970), Philippines, 5034, 512 Republican People’s Party, Turkey, 620, 635 Republic of Korea, 161 Republic v. Quash.379 Reserve money: Indonesia, 25, 29, 84; Philippines, 529-31 Revenue, government: Indonesia, 39-40; Philippines, 41 1- 18; Turkey, 734-36. See also Tariffs; Tax policy Rice intensification program, Indonesia, 24, 89 Rural development policy, Indonesia, 60-61, 79 SAL (structural adjustment loan). See World Bank Savings, domestic: Korea, 186-88, 236-42, 248-49; Turkey, 742-47 SBIs. See Debt Certificates SBPUs. See Money Market Instruments Singapore, 409 SITC. See Standard International Trade Classitication Small and Medium Industry Bank, Korea, 283 Smuggling, Indonesia, 40,98 Social Security System (SSS), Philippines, 425,484 Soeharto regine: 28-29, 57-59, 97-98, 115 Soekamo regime, 32-48, 56 Spending, private, Indonesia, 45 SSS. See Social Security System Stabilization plan, U.S.-Korean (1960s), 284

821

Subject Index

Standard International Trade Classification (SITC), 261 State economic enterprises (SEEs), Turkey, 627-28, 673-74 State enterprises: Indonesia, 40, 88; Korea, 284-85; Philippines, 398, 405, 408, 411, 418-25, 459; Turkey, 626. See also State economic enterprises (SEEs), Turkey State Planning Organization (SPO),Turkey, 620 Stock exchange: Indonesia, 38, 87-88; Philippines, 481; Turkey, 670 Subsidies, trade, 259-61 Sugar industry, Philippines, 461 -66 Sugar lobby, Philippines, 376-77, 439, 450 Sulawesi, 36, 60 Sumatra, 36 Syngman Rhee regime, 155, 304 Taiwan, 409 Tariffs: Korea, 264; Philippines, 412, 415, 439, 441-43, 446-47. See also Barriers to entry, nontariff Tax, value-added (VAT), Indonesia, 72-73 Tax policy: Indonesia, 68-71; Philippines, 412-18, 456 Tax reform: Indonesia, 23, 56, 70-74; Korea, 203; Philippines, 415-16 Technicians. See Technocrats Technocrats: Philippines, 386, 390; Indonesia, 22, 57, 65-67, 82-83 Tailand, 409 Tradable goods industry, effect of Dutch disease on, Indonesia, 100- 102 Trade Agreement Act of 1946 (Bell Act), U.S.-Philippines, 379, 437-38, 441 Trade liberalization: Korea, 187, 258-65; Philippines, 446; Turkey, 670-71, 71920. See also Export market; Import controls; Protectionism; Subsidies, trade; Tariffs

Trade policy: Philippines, 433, 435-43, 457, 525-21, 572-73; Turkey, 626-27, 716-17. See also Baniers to entry; Exchange rate; Protectionism; Tariff policy Trade restrictions: in Indonesia, 2. See also Baniers to entry, nontariff; Import controls; Protectionism; Quantitative restrictions: Quotas UCPB. See United Coconut Planters Bank UMM. See Unofficial money market UNICOM. See Coconut Oil Mills Union Oil Company, 476 Unions, Korea, 279-80 United Coconut Planters Bank (UCPB), Philippines, 467-68 United Nations Korean Reconstruction Program (UNKRA), 163 United States, foreign aid programs: South Korea, 163; Philippines, 505 United States Army Military Government in Korea (USAMGIK): establishment of public education and land reform systems, 160-65 UNKRA. See United Nations Korean Reconstruction Program Unofficial money market (UMM), Korea, 236, 283, 297 USAMGIK. See United States Army Military Government in Korea Usury Law of 1916, Philippines, 487 Wages: Korea, 195, 197, 274-81; Philippines, 383-84, 455 Westinghouse (firm), 407 Worker Council law, Korea, 279 World Bank, as source of financial aid: to the Philippines, 415, 446, 505; to Turkey, 67 1-72 World War ll: and the Philippines, 438; and Turkish economy, 619