Protectionism and World Welfare

This volume addresses highly topical issues at a crucial time in international economic relations. The world has never been closer to dismantling the liberal multilateral trading system which has been painstakingly established and successfully operated since the Second World War. In this volume many of the world's most distinguished economists examine the movement toward protectionism, bilateralism, and regionalism, and its causes, effects, and possible solutions. The contributors are theorists, researchers, and advisors to governments and international organizations who are at the forefront of trade theory, policy, and practice, and whose analyses have a real impact on international trade. By collecting together these analyses in a single volume, this book provides a unique survey for students and scholars of economics, and all those concerned with trade theory and policy in business and government.

Protectionism and world welfare

Protectionism and world welfare edited by

DOMINICK

SALVATORE

CAMBRIDGE

UNIVERSITY PRESS

Published by the Press Syndicate of the University of Cambridge The Pitt Building, Trumpington Street, Cambridge CB2 IRP 40 West 20th Street, New York, NY 10011-4211, USA 10 Stamford Road, Oakleigh, Melbourne 3166, Australia © Cambridge University Press 1993 First published 1993 A catalogue record for this book is available from the British

Library

Library of Congress cataloguing in publication data

Protectionism and world welfare / edited by Dominick Salvatore. p. cm. Includes indexes. ISBN 0 521 41455 5. - ISBN 0 521 42489 5 (pbk.)

1. Protectionism. HF1713.P77

382'.73-dc20

2. International trade.

1993

92-34450 CIP

ISBN 0 521 41455 5 hardback ISBN 0 521 42489 5 paperback

Transferred to digital printing 2004

I. Salvatore, Dominick.

To Luci/Ie

Contents

List of contributors Preface I

page xi xiii

Protectionism and world welfare: introduction

1

DOMINICK SALVATORE

1 The new protectionism: an overview 2

Fair trade, reciprocity, and harmonization: the novel challenge to the theory and policy of free trade

15 17

JAGDISH N. BHAGWATI

3

The revival of protectionism in developed countries

54

W. MAX CORDEN

4

Changes in the global trading system: a response to shifts in national economic power

80

ROBERT E. BALDWIN

5

"Fortress Europe" and retaliatory economic warfare

99

LAWRENCE R. KLEIN AND PINGFAN HONG

II 6

Trade theory, industrial policies, and protectionism

129

US response to foreign industrial policies

131

RICHARD N. COOPER

7

The current case for industrial policy PAUL R. KRUGMAN

IX

160

Contents 8

The case for bilateralism

180

RUDIGER W. DORNBUSCH

9

Restrictions to foreign investment: a new form of protectionism?

200

VITO TANZI AND ISAIAS COELHO

III

Exchange rates and protectionism

219

10

Floating exchange rates and the new interbloc protectionism: tariffs versus quotas

221

RONALD I. McKINNON AND K.C. FUNG

11 The theory of tariffs and monetary policies

244

ROBERT MUNDELL

12 The economics of content protection

266

MICHAEL MUSSA

13 The relationship between exchange-rate variability and protection

290

JOSEPH A S C H H E I M , GEORGE S. TAVLAS, AND MICHAEL ULAN

IV

The new protectionism in the world economy

309

14 Trade protectionism and welfare in the United States

311

DOMINICK SALVATORE

15 Protectionism and growth of Japanese competitiveness

336

RYU2O SATO, RAMA RAM AC H ANDRAN, AND S H U N I C H I TSUTSUI

16 Protectionism in Western Europe

371

NORMAN SCOTT

17 Protectionism and the developing countries

396

G.K. HELLEINER

18 Trade liberalization — the new Eastern Europe in the global economy

419

JOZEF M. VAN BRABANT

Author index

441

Subject index

445

Contributors

JOSEPH ASCHHEIM

George Washington University, Washington, DC ROBERT E. BALDWIN

University of Wisconsin, Madison, Wisconsin JAGDISH N. BHAGWATI

Columbia University, New York, New York ISAIAS COELHO

International Monetary Fund, Washington, DC RICHARD N. COOPER

Harvard University, Cambridge, Massachusetts W.

MAX CORDEN

Johns Hopkins University, Washington, DC RUDIGER W. DORNBUSCH

Massachusetts Institute of Technology, Cambridge, Massachusetts K.C.

FUNG

University of California at Santa Cru%, Santa Cru%, California G.K. HELLEINER

University of Toronto, Toronto, Canada PINGFAN HONG

United Nations, New York, New York LAWRENCE R. KLEIN

University of Pennsylvania, Philadelphia, Pennsylvania

XI

xii

Contributors

PAUL R. KRUGMAN

Massachusetts Institute of Technology, Cambridge, Massachusetts RONALD I. McKINNON

Stanford University, Stanford, California ROBERT MUNDELL

Columbia University, New York, New York MICHAEL MUSSA

International Monetary Fund, Washington, DC RAMA RAMACHANDRAN

New York University, New York, New York DOMINICK SALVATORE

Fordham University, New York, New York RYUZO SATO

New York University, New York, New York NORMAN SCOTT

Economic Commission for Europe, Geneva, Switzerland VITO TANZI

International Monetary Fund, Washington, DC GEORGE S. TAVLAS

International Monetary Fund, Washington, DC SHUNICHI TSUTSUI

Tulane University, New Orleans, Louisiana MICHAEL ULAN

US Department of State, Washington, DC JOZEF M. VAN BRABANT

United Nations, New York, New York

Preface

This volume comes at a very crucial time in international economic relations. Never before has the world been closer to dismantling the liberal multilateral trading system that has been so painstakingly put together since the end of the Second World War and served the world so well since then. Support for a liberal trading system in the United States has weakened considerably during the past decade and has given way to aggressive unilateralism. Japan's export successes and industrial targeting, in the face of serious Japanese institutional barriers against imports from the rest of the world, have resulted in serious frictions with the United States and Europe. European agricultural and industrial protectionism has put serious strains on the system. The breaking up of the world into three major trading blocs can potentially have ominous consequences for the world trading system and can lead straight down the path of managed trade. Even the time-honoured theory that a free-trading system maximizes world welfare and the welfare of each trading nation has recently come under attack by strategic trade theory and policy. All of these problems and trends are leading the world toward a trading system that is much less liberal, multilateral, and global than the one to which we have grown accustomed - to the detriment of all. The stalling of the Uruguay Round also means that a crucial opportunity to reverse or at least to contain these dangerous trends and set the world back on a liberalizing track has been missed. In this volume, many of the world's most distinguished economists address, with original articles, the issue of the recent deterioration of xm

xiv

Preface

international economic relations and movement toward protectionism, bilateralism, and regionalism, as to causes, effects, and possible solutions. These are the theorists, researchers, and advisors to governments and international organizations who are at the forefront of trade theory, policy, and practice and whose analyses, recommendations, and opinions can, and do, make a difference in matters of international economic relations in general and international trade in particular. Having all of these analyses, thoughts, and beliefs in a single volume can prove extremely useful to theorists, practitioners, students, and the enlightened public, especially now that trade theory and policy have left the ivory towers and specialized government offices and become part of general economics and reached the general public through newspapers and news programs. I take this opportunity to thank all participants in this volume and Patrick McCartan, the economics editor at Cambridge, without whose encouragement and support this volume would never have been undertaken or completed. Dominick Salvatore New York, June 1993 Fordham University

CHAPTER

Protectionism and world welfare: introduction DOMINICK SALVATORE

1

Introduction

Trade relations among the world's major industrial nations have taken a turn for the worse during the past two decades and are now threatened by new and more dangerous forms of trade restrictions, collectively known as the "new protectionism." This phrase, coined in the mid 1970s, refers to the revival of "mercantilism" whereby nations, particularly the industrial nations, attempt to solve or alleviate their problems of unemployment, lagging growth, and declining industries by imposing restrictions on imports and subsidizing exports. The instruments by which imports are restricted are also somewhat different from and less transparent than traditional import tariffs, and are called non-tariff barriers (NTBs). These refer to "voluntary" export restraints, orderly marketing arrangements, anti-dumping measures, countervailing duties, safeguard codes, and so on. Thus, at the time when tariffs were being reduced as part of the successive rounds of trade liberalization sponsored by the GATT (General Agreement on Tariff and Trade) and they are presently very low on most industrial goods, the number and importance of NTBs have grown rapidly since the mid 1970s and they have now become more important than tariffs as obstructions to international trade. As much as 50 percent of world trade is now affected by this new protectionism. This new protectionism now represents the greatest threat to the fairly liberal world trading system that has been so painstakingly put together over the past half a century and which has served the world so well since the end of the Second World War. As Corden points out in his chapter in this volume, this new protectionism arose as nations sought to protect industry

Z

DOMINICK SALVATORE

after industry from the adjustments required by international specialization and trade in a climate of slow domestic growth and rising unemployment. While not exactly "beggar-thy-neighbor" policies, the rise in NTBs leads to a shrinkage in the volume of world trade and to a reduction in the static and dynamic gains from trade that threatens the well-being of the entire world economy. This new protectionism is also closely related to other crucial policy controversies that the United States and other industrial countries are facing and are likely to continue to face in the foreseeable future. These include calls to adopt strategic trade and domestic industrial policies, to restrict excessive exchange volatility and misalignments and to promote greater international macroeconomic policy coordination; to form or to belong to a regional trading group and to deal with the problems that may arise from the resulting breaking up of the world into three major trading blocs; to reform the world trading system so as to stem the tide of protectionism and reverse the trend toward managed trade. All of these topics are examined in this volume and summarized in the rest of this introduction. 2

Post-war trade liberalization and the new protectionism

The 1950s and the 1960s saw unprecedented growth in the volume of international trade and in the economies of most countries of the world, particularly the industrial countries. The merchandise trade of the industrial countries grew at an average rate of 8 percent per year, and this fueled a growth in real gross domestic product of over 4 percent per year. This period can truly be regarded as a golden age of growth and stability. The growth of international trade was made possible by successive rounds of trade liberalization achieved under the Dillon Round (1960-1), the Kennedy Round (1964-7), and the Tokyo Round (1974-7). Tariffs were cut until today they are less than 5 percent on most industrial products. Effective tariffs (that is, tariffs on value added) are somewhat higher than nominal tariffs but are still very low both in absolute terms and by historical standards. Generally resisting the trend toward trade liberalization was trade in agricultural products and in services. As is well known, most nations have very elaborate domestic agricultural support programs and generally shield their agriculture and service sectors from outside competition by a powerful array of subsidies, tariffs, quotas, health regulations, and so on. The general trend toward trade liberalization was soon reversed, however, starting in 1975, when the world faced the deepest worldwide recession since the Great Depression of the 1930s. This gave rise to what has come to be known as the new protectionism, characterized by the imposition of many new non-tariff barriers. This fact did not go unnoticed

Protectionism and world welfare during the Tokyo Round of trade negotiations, and indeed codes were negotiated to restrict and regulate the use of NTBs. These codes, however, were unsuccessful in stemming the tide of new NTBs, to the point where today they represent the most serious threat to the post-war trading system and world welfare. From 1975 to 1979, world trade grew at about 5 percent per year (as opposed to an average of 8 percent during the previous two-and-a-half decades) and fell to a yearly average growth of 3 percent from 1980 to 1985. Clearly this trend of increased protectionism and declining growth in world trade had to be reversed. Thus, the Uruguay Round was initiated in 1986 and scheduled to be concluded at the end of 1990. The aim of the Uruguay Round was to establish rules for checking the proliferation of the new protectionism and reverse its trend; bring services, agriculture, and foreign investments into the negotiations; negotiate international rules for the protection of intellectual property rights; and improve the dispute settlement mechanism by ensuring more timely decisions and compliance with GATT rulings. The successful completion of the Uruguay Round would go a long way toward resolving the serious problems that face the present international trading system and restoring international confidence in the system. Negotiating rules of conduct to reverse the spread of the new protectionism proved extremely difficult, however, since issues of national sovereignty and desire on the part of most industrial nations to protect mature industries and stimulate high-tech ones are involved. In agriculture, the European Community and Japan objected to the US proposal to remove or phase out by the turn of the century all farm aid programs that interfere with international specialization and trade. In December 1990, the Uruguay Round stalled after the European Community (EC) rejected demands made by the United States and other large food exporters (among which Canada, Argentina, and Australia) for substantial cuts in EC agricultural subsidies. All attempts to revive and successfully conclude the negotiations had failed by December 1992. Thus, the world lost a crucial opportunity to reverse or even to check the slide toward a much less liberal and much more protectionistic world. 3

Strategic trade and industrial policies, and the new protectionism

The superiority of an international trading system characterized by greater multilateralism and international specialization over a trading system based on protectionism, bilateralism, and a division of the world into major trading blocs is by no means as clear cut today as it was a decade

3

4

DOMINICK SALVATORE

ago. There are two reasons for this. One is that many of the new non-tariff trade barriers are considered part of the arsenal of policies that a nation believes to be necessary in order to achieve some important domestic objectives. The second is that in recent years the very theoretical foundation of the modern theory of international trade, which for nearly two centuries has been consistently based on the alleged superiority of free trade over a system based on trade restrictions, is being questioned. Today all the leading industrial countries impose some restrictions on the importation of automobiles, steel, textiles, consumer electronics, and agricultural products. Practically all also provide direct and indirect subsidies to their computer and data processing, aircraft, and most other high-tech industries. Industrial nations regard these trade restrictions and subsidies to be crucial either to protect employment in their large and mature industries or to promote the growth of the high-tech industries deemed essential for the international competitiveness and future technological capability of the nation. These goals are promoted by tax benefits and subsidies for research, education and investments - and most nations regard these as purely internal matters. Japan is often given as the classic example of a nation using strategic trade and industrial policies. According to this, Japan provides protection from foreign competition and a maze of direct and indirect subsidies to an industry targeted for growth. After the industry has grown and is capable of meeting foreign competition, the industry, with the tacit approval and indirect support of the government, begins to dump the product (i.e., to sell the product at below domestic production cost) on the world market on a massive scale until it has driven foreign competitors out of business or rendered them impotent. Then the industry raises prices and proclaims full support for the principle of free trade, pointing to its then unprotected industry as a model of efficiency. According to this view, Japan has successfully and systematically applied this policy to steel, automobiles, computer memory chips, and is now doing this in computers and financial services. Understandably, Japan would not readily abandon an industrial strategy that proved so successful and instrumental in turning it into a first class economic power in just a few decades. While somewhat less aggressively and generally less successfully, the leading European countries have also used some of these same policies to create and stimulate the growth of Airbus and the Arianne space program, among others, and, to some extent, so did the United States (through the commercial applications of the technological discoveries arising from its military and space research programs) to promote its commercial aircraft industry. Thus, while the leaders of the major industrial countries pay lip service to the great benefits of and preference for a free multilateral trading

Protectionism and world welfare system, they have often violated those principles and have become much more protectionistic during the past two decades. Charging interference with national sovereignty, the leading industrial nations are even objecting to providing information to GATT on these new indirect forms of trade protection. As pointed out earlier, the attempt at the Tokyo Round (1974-7) to negotiate rules of behavior to limit the use of these new forms of protectionism and make them more transparent (for example, by replacing them with equivalent open tariffs) has, in general, not been successful - and the more recent Uruguay Round has stalled. Protectionism and bilateralism are also indirectly being encouraged by the recent questioning of the superiority of the time-honored free-trade model of international economics. Ricardo's theory of comparative advantage is attacked as being entirely static in nature and not very relevant to international trade in a world characterized by imperfect competition, technological breakthroughs, product cycles, intra-industry trade, multinational corporations, and integrated capital markets. Some of these criticisms are not new but they seem to have acquired new force and legitimacy from the fact that some leading theoreticians are joining in the criticism of traditional comparative advantage (see, for example, Krugman, 1986). It is now believed that most of today's international trade is based on comparative advantage that is created by industrial policies (which give rise to new technologies and new industries) rather than by traditional comparative advantage based on inherited international differences in factor endowments across nations. Strategic trade and industrial policies are also advocated to overcome market imperfections resulting from the existence of economies of scale and externalities and to combat foreign targeting. As discussed in the paper by Tanzi and Coelho, strategic trade and industrial policies also affect foreign direct investments. Opponents of strategic trade and industrial policies point out, however, that the theory of comparative advantage can be extended to incorporate dynamic changes in the form of new products and new technologies and that, in any event, it is often very difficult for the government to pick winners in the technological race. Furthermore, as Bhagwati (1971) clearly pointed out more than twenty years ago and repeats in his chapter in this volume, market imperfections should be corrected with taxes and subsidies in the markets where these imperfections occur rather than with trade restrictions. Most economists also believe that two wrongs simply do make a right in economics. There are, then, the many real-world examples of clear targeting failures, such as the abandoned Synfuel in the United States, the huge economic losses of the Anglo-French Concorde, and the still higher costs of electricity from atomic power than from conventional methods. To this, supporters of strategic trade and industrial policies reply that

5

6

DOMINICK SALVATORE

comparative advantage can indeed be created by targeting, as has occurred, for example, in the case of synthetic rubber, jetliners, and lunar landers and in the new advanced technology that sprang from them. They also point out that Lockheed, Chrysler, and Continental Illinois were all saved by large government loans. Indeed, Krugman, in his chapter included in this volume, states that industrial policy and targeting does have its practical uses and applications in the presence of substantial and clearly demonstrable externalities. Cooper holds similar views. Dornbusch's support for bilateralism is directed primarily at forcing Japan to open its markets much more widely to imports and his support for regionalism is aimed at speeding up the process of integration of Latin America and Eastern Europe into the world trading system. Bhagwati, on the other hand, remains the most uncompromising opponent to strategic trade and industrial policies and the stronger champion of a free multilateral trading system. 4

The new protectionism and fluctuating exchange rates

Excessive fluctuations in exchange rates in general and exchange rate misalignments in particular can also lead to protectionism. Indeed, McKinnon and Fung, in their chapter included in this volume, strongly support the view that exchange-rate volatility, or frequent and large fluctuations of exchange rates about equilibrium levels, by increasing uncertainty in international trade and payments, was an important reason for the spread of the new protectionism since the advent of flexible exchange rates in 1973. They also advance the view that currency instability encouraged the formation of trading blocs and this, by undermining the most favored nation principle, led to the spread of interbloc protectionism. The empirical results that Tavlas, Aschheim, and Ulan present in this volume, on the other hand, seem to show that the exchange fluctuations of the dollar did not appear to have reduced the volume of US trade in the period from 1975 to 1990 and, therefore, could not have provided grounds for protectionism. More important in giving rise to protectionist pressures are exchange rates that remain overvalued or undervalued for relatively long periods of time. The overvaluation of a nation's currency is equivalent to an import subsidy and an export tax by the nation. Many goods and services that would normally be exported with equilibrium exchange rates would be imported with an overvalued currency. Thus, potential exporters join import-competing producers in demanding protection. These demands are greatly strengthened by displaced workers and their labor unions and are difficult to resist, especially in periods of large unemployment and sluggish growth. Another result (discussed by Mussa) is content protection (i.e., the

Protectionism and world welfare requirement that final goods assembled for foreign-owned firms in the country use a minimum amount of domestic inputs). Ironically, demands for protection arise even in nations with undervalued currencies. An undervalued currency has an effect similar to an import tax, or tariff, and to an export subsidy. Thus, a persistent currency undervaluation leads to excessive expansion in the domestic production of import-competing and export industries. A subsequent realignment of exchange rates toward equilibrium levels and the elimination of the currency undervaluation then leads to increased competition and loss of production and jobs in import-competing as well as in export industries. Having become more or less entrenched during the period of currency undervaluation, these industries and their workers are likely to demand and frequently succeed in receiving protection when the currency appreciates toward the equilibrium level. The relationship between tariffs, money price levels, and exchange rates under a fixed and a flexible exchange rate regime is examined by Mundell in his chapter included in this volume. Bergsten and Cline (Cline, 1983, chapter 3) point out that the three periods of greatest protectionist pressures in the United States since the late 1960s coincided with periods of large dollar overvaluations. These were: (1) the period from the late 1960s to 1971, when the dollar overvaluation of about 20 percent (corrected when the Bretton Woods system collapsed and was replaced by flexible exchange rates) led to increased trade restrictions on imports of textiles and steel, and to widespread support for strongly protectionist measures. (2) The period 1975-6, when a dollar overvaluation of about 15 percent (corrected during 1977—8) coincided with the adoption of the trigger price mechanism (TPM) for steel and other protectionist measures. (3) The dollar overvaluation from 1981 to 1985, which led to new trade protection for automobiles, textiles, and steel. Conversely, periods of near exchange-equilibrium in the United States have led to trade liberalizations, such as the passage of the Trade Act of 1974 and the Trade Act of 1979. There is, however, the danger of a ratchet effect, whereby trade restrictions imposed during periods of exchange-rate misalignments are not removed when exchange rates return to near equilibrium levels. For example, the extension of trade restrictions to synthetic fibers in the early 1970s (at a time when the United States had a trade deficit in that account) were not removed afterwards when the United States achieved a trade surplus in synthetic fibers. 5

Protectionism around the world

During the past two decades, the new protectionism has spread to most industrial countries. In his chapter Salvatore points out that while the

7

O

DOMINICK SALVATORE

United States has generally adhered to the principle of free trade during the past thirty years, it has made an increasing number of compromises or exceptions to protect textiles, steel, automobiles, and other industries in exchange for the political support for the general principle of free trade, in the general context of increased global competition, unemployment, and lagging growth. Furthermore, concern over "fairness" and the desire to provide a "level playing field" in international trade in recent years has led to the revision of US trade laws and their interpretation away from non-discriminatory multilateralism toward aggressive bilateralism (i.e., requiring retaliation against countries that do not provide market access to US products equal to that which foreign countries and products enjoy in the United States). As Baldwin points out in his chapter, US support for a liberal multilateral trading system has wavered during the past two decades as a result of its loss of hegemonic power. This also moved the United States more and more toward regionalism, as the recent pursuit of a North American Free Trade Association indicates. In their chapter, Sato, Ramachandran, and Tsutsui conclude that the Japanese government did play a successful interventionist role through the Ministry of International Trade and Industry (MITI) in improving the competitiveness of home industries in international trade. The government (MITI) did this by selecting a small number of powerful companies to be initially protected from domestic and foreign competition. These companies received tax credits and subsidies to encourage them to develop technology through a cooperative research program. Once technology was developed, the government then strongly encouraged domestic competition in order to increase efficiency. A school system that emphasized science and high national savings and low interest rates made this possible. The authors also point out that much of the debate on Japan centers in the West on whether MITI has superior signaling abilities in encouraging the development of new technologies by firms and whether this requires a response from Western nations. The authors conclude, however, that the channels by which Japan and MITI oversee and direct the technology transfer and the methods of stimulating new technologies differ in degree from what takes place in other countries. As pointed out by Scott in his chapter, protectionism in the European Community (EC) has increased during the past two decades and generally paralleled the US increase in both strength and form. As in the case of the United States, the EC now protects its textile, steel, automobile, machine tools, and consumer electronic industries against the export of Japan, the newly industrializing countries (NICs), and other industrial countries. The tools that it uses are also similar; i.e., safeguard mechanisms, surveillance, anti-dumping duties, and voluntary export restraints. In deciding if

Protectionism and world welfare clumping has occurred, the EC sometimes adds as much as a 30 percent profit margin to the foreign cost of production. The foreign firm is then accused of dumping if it sells the product in the EC at below such a constructed price. Particularly troublesome is the EC's common agricultural policy (CAP). According to this, the EC first determines common farm prices and then imposes tariffs so as always to make the price of imported agricultural products equal to the high established EC prices. This not only severely restricts agricultural imports but also leads to huge agricultural surpluses within the EC and subsidized exports. The unwillingness on the part of the EC to reduce agricultural protectionism sharply was the primary cause for the stalling of the Uruguay Round in December 1990. A great deal of the new protectionism has been directed by the developed countries against the manufactured exports of the NICs. These nations (Brazil, Hong Kong, Korea, Mexico, Singapore, Taiwan) are characterized by rapid growth in gross domestic product (GDP), in industrial production, and in manufactured exports to industrial nations. The NICs have gained a comparative advantage vis-a-vis developed nations (including Japan) in textiles, shoes, television sets, consumer electronic products, steel, and shipbuilding. These are the very industries in which sharp cuts in employment have occurred, and are continuing to occur in developed countries. This, combined with the fact that these newly industrializing countries have little political power and are not in a position to threaten effective retaliation, has led developed countries to raise many new forms of protection against the manufactured exports of these newcomers. This has occurred in spite of the Generalized System of Preferences (GSP), negotiated by Western European countries and Japan in 1971—2 and by the United States in 1976, which grants preferential access for the exports of developing countries into developed countries' markets. Exception after exception to GSP has been "voluntarily" negotiated by the United States and other developed countries in many products, such as textiles, which are of great importance to developing countries. As Michaely (1987) pointed out, the NICs have simply been unlucky to be rapidly growing and to have increased their manufactured exports to developed countries at a time when the latter were facing large unemployment and slow growth. Had developed countries and international trade been growing in the second half of the 1970s and in the 1980s as rapidly as they did during the 1950s and 1960s, the growth in the manufactured exports of the NICs to developed countries would probably have been absorbed much more smoothly and without as much of a rise in protectionism on the part of developed countries. Having emerged in the wake of Japan's great industrial and export success, the NICs have been drawn inevitably into the trade disputes between industrialized nations and

9

10

DOMINICK SALVATORE

Japan and have been, to some extent, victimized by the ire of the former nations against the latter. It must be pointed out, however, that while the NICs were bitterly complaining about the new protectionism directed by developed countries against their manufactured exports, they were themselves heavily protecting their market against the exports from both developed and other developing countries. As pointed out by Helleiner in his chapter in this volume, it is essential to reassert and enforce the GATT principle of non-discrimination in trade matters and allow developing countries to increase their exports to developed countries, thereby helping to solve the former's foreign-debt problem and stimulate their development. If this does not happen very soon, we may witness a revival (and justification) of export pessimism and a return to inward-looking policies in developing countries. Trade problems in the West and developing countries have recently been compounded by the political and economic collapse in the East. Until a few years ago, trade among Eastern European countries and the Soviet Union was generally conducted on the basis of bilateral agreements and bulk purchasing, with market forces playing little if any role. Since the collapse of communism, Eastern European countries and the former Soviet Republics have been struggling to restructure their economies and foreign trade along market lines. This is a monumental task after many decades of central planning and gross inefficiencies. These countries are now facing increasing unemployment, high inflation, huge budget deficits, unsustainable international debts, and disrupted trade relations. Eastern European nations are having serious difficulties in expanding trade with the West sufficiently to make up for the collapse in trade among themselves and the former Soviet Republics because of the generally low quality of their manufactured products and protectionism in the EC and in other industrial countries against agricultural, steel, and textile products in which Eastern European countries seem to have a comparative advantage. These countries also need huge amounts of foreign capital and technology in order to restructure their economies and establish market economies. As van Brabant points out in his chapter, the ultimate goal of Eastern European countries is to become members of the EC, but this can only occur after these nations have accomplished the arduous task of restructuring their economies along market lines. 6

Protectionism and the future of the international trading system

The world has already and probably irreversibly moved into an international trade order characterized by three major trading blocs. The

Protectionism and world welfare

11

EC is already a political and economic reality, the United StatesCanada-Mexico free-trade area seems to be more or less already agreed upon and its implementation seems to proceeding rapidly. Least developed is the Asian trade bloc around Japan, but the dynamics of the situation is such that its formation is all but inevitable if the other two trading blocs continue to keep to their timetable for implementation. One could argue along the lines of customs union theory that such free-trade areas are second-best trade arrangements if a true worldwide free-trade system cannot be achieved under present conditions. The formation of these trade blocs, it is argued, will lead to increased specialization in production and raise world welfare if the net effect is to stimulate trade within each bloc without reducing trade among the blocs. The latter expectation, however, may not materialize and the formation of trading blocs may in fact have a net trade-diverting effect and lead to more generalized and, thus, more dangerous commercial disputes. As pointed out by Klein and Hong in the chapter included in this volume, this will impose efficiency and welfare costs on the world similar to those resulting from the oligopolization of previously near-perfect competitive markets. The best opportunity to prevent a drift toward greater protectionism and trade disputes among the major trading blocs into which the world is breaking up is to revive and successfully bring to conclusion the Uruguay Round. If this does not occur, commercial relations among the blocs would be seriously restricted and trade frictions would be the order of the day. The total volume of world trade may then fail to rise and may even decline, international specialization in production will be limited mostly by the extent of the market within each trade bloc, and the stimulating force that trade can play in the world economy will be seriously constrained. Even if the Uruguay Round is revived and brought to a successful conclusion, however, it is most unlikely that it would succeed in fully reversing the trend toward protectionism. What is possible is to negotiate for a reduction of traditional trade barriers and explicit non-tariff trade barriers and accept as inevitable the existence of some implicit non-tariff trade barriers (such as government aid to sunset and sunrise industries that the nation might be unwilling to give up or insists on having). Nations could counteract with appropriate domestic policies the most disturbing effects of implicit foreign non-tariff trade barriers in a manner that minimizes the resulting trade controversies. Success in these negotiations is made more difficult by the loss of the hegemonic position that the United States enjoyed during the 1950s and 1960s. The most that can reasonably be expected from a successfully completed Uruguay Round is that it would: (1) reassert the principle of an open multilateral trading system, (2) strengthen the dispute-settlement procedure

12

DOMINICK SALVATORE

of G ATT and raise its status to that of the International Monetary Fund and the World Bank, (3) set up the framework and establish the principle for subsequent liberalization in trade in services and agriculture. This would be no small accomplishment and would preserve a world trading system that served the world so well for nearly half a century. References Bhagwati, Jagdish N. (1971), "The Generalized Theory of Distortions and Welfare," in Jagdish N. Bhagwati (ed.), International Trade: Selected Readings,

Cambridge, Mass.: The MIT Press, pp. 171-89. (1990), Protectionism. Cambridge, Mass.: The MIT Press. (1991), The World Trading System at Risk, Princeton, New Jersey: Princeton University Press. Cline, William R. (1983), Trade Policies in the 1980s, Washington, DC: Institute for International Economics. Corden, Max W. (1987), "Protectionism and Liberalization: A Review of Analytical Issues." Occasional Paper 54, International Monetary Fund, Washington, DC. Council of Economic Advisors (1992), Economic Report of the President, Washington, DC: US Government Printing Office. Hathaway, Dale E. (1987), Agriculture andGATT: Rewriting the Rules, Washington,

DC: Institute for International Economics. International Monetary Fund (1984), Exchange Rate Volatility and World Trade,

Washington, DC: International Monetary Fund. (1992), World Economic Outlook, Washington, DC: International Monetary Fund, May. (1992), International Financial Statistics, Washington, DC: International Monetary Fund. Jones, Ronald W. and Peter B. Kenen (eds.) (1984 and 1985), Handbook of International Economics, Vols. 1, 2, Amsterdam: North-Holland. Krugman, Paul R. (1987), "Is Free Trade Passe?" The Journal of Economic Perspectives, 1(1), Fall, 131-44. Krugman, Paul R. (ed.) (1986), Strategic Trade Policy and the New International

Economics, Cambridge, Mass.: The MIT Press. McKinnon, Ronald R. (1984), An International Standard for Monetary Stabilisation.

Washington, DC: Institute for International Economics. Michaely, M. (1987), "The Demand for Protection Against Exports of Newly Industrializing Countries," in D. Salvatore (ed.), The New Protectionist Threat to World Welfare, New York and Amsterdam: North-Holland, pp. 471-81. Organization for Economic Co-operation and Development (1985), Costs and Benefits of Protection, Paris: O E C D .

Salvatore, Dominick (1988), "The New Protectionism with NontarifF Trade Instruments," in Christopher Saunders (ed.), Macroeconomic Management and the Enterprise in East and West, London: Macmillan, pp. 155—82. (1993), International Economics, 4th edn, New York: Macmillan.

Protectionism and world welfare

13

Salvatore, Dominick (ed.) (1987), The New Protectionist Threat to World Welfare, New

York and Amsterdam: North-Holland, (ed.) (1991), Handbook of National Economic Policies, Westport CT: Greenwood Press and Amsterdam: North-Holland, (ed.) (1992), Handbook of National Trade Policies, Westport CT: Greenwood Press and Amsterdam: North-Holland. World Bank (1992) World Development Report. Washington, DC: World Bank.

PART

i

The new protectionism: an overview

CHAPTER

Fair trade, reciprocity, and harmonization: the novel challenge to the theory and policy of free trade JAGDISH N. BHAGWATI

Writing in the new Journal of Economic Perspectives in 1987, my distinguished pupil and successor to my chair at M I T , Paul Krugman, declared: the case for free trade is currently more in doubt than at any time since the 1817 publication of Ricardo's Principles of Political Economy . . . In the last ten years the traditional constant returns, perfect competition models of international trade have been supplemented and to some extent supplanted by a new breed of models that emphasizes increasing returns and imperfect competition. These new models . . . open the possibility that government intervention in trade via import restrictions, export subsidies, and so on may under some circumstances be in the national interest after a l l . . . free trade is not passe, but it is an idea that has irretrievably lost its innocence. Its status has shifted from optimum to reasonable rule of thumb. There is still a case for free trade as a good policy, and as a useful target in the world of politics, but it can never again be asserted as the policy that economic theory tells us is always right.

Trade theorists of my generation, who spearheaded the modern theory of commercial policy from the 1950s through the 1970s, exploring several types of market failures and the appropriate nature of governmental interventions (including the use of trade tariffs, quotas, and subsidies) in each such instance, have found it puzzling that anyone should seriously suggest that "economic theory" was believed until the 1980s to imply that free trade was "always right." 1 The trade theorists who have used imperfectly competitive models recently and who then proceed to claim to be pioneers in thinking of possible theoretical arguments for departures from free trade, appear to us like young men who, on visiting a prostitute, boast of having robbed her of her virtue. 17

18

JAGDISH N. BHAGWATI

But there are two other objections, of equal significance, and they concern what Krugman says about the case for free trade being in its greatest crisis since 1817. The assertion is not valid historically. As I argue immediately below, a backward glance shows that earlier crises, some of equal and others of greater significance, have occurred with regard to the policy presumption in favor of free trade: all reflecting different blends of current economic circumstances and theoretical reflections. Moreover, and this is a theme that I develop principally in this chapter, the true and greater crisis that we face with regard to the theory and policy of free trade today comes, not from what Krugman has in mind (i.e., the theoretical modeling of imperfect competition in product markets during the 1980s), but from the growth of demand for "level playing fields," "harmonization," "fair trade," etc., all of which are variously undermining insidiously the legitimacy and feasibility of free trade since it is virtually impossible to harmonize everything so that playing fields are truly level in every way. There will always be something that an opponent of free trade will be able to find that is different in the country of one's successful rival and hence can be argued to make free trade unfair and therefore illegitimate and unacceptable. I intend here to explore this problem, which I consider to be the truly greatest threat since the early 1930s when (as reviewed below) Keynes was driven to abandon free trade in face of massive unemployment and many followed in his footsteps. But first, I consider the different ways in which the intellectual presumption that free trade was empirically the policy to follow has been shaken repeatedly over the period since 1817. 1

Free trade: national versus cosmopolitan formulations

Unfortunately, in the popular as well as in the academic discussion of free trade, the critical distinction between the "national" and the "cosmopolitan" formulations of free trade is blurred. They share common foundations, but they also have different implications for matters that have become contentious from time to time. The national formulation The national formulation of the case for free trade is the one that predominantly shapes the policy debate. It posits national welfare as the objective of policy and proceeds to demonstrate that free trade will maximize national welfare. In essence, the case depends of course on the set of assumptions that ensure that market prices reflect social costs, to use the old and apt Pigovian

Fair trade, reciprocity, and harmonization

19

way of saying things. While the case, if we must make it rigorously, requires the usual theoretical armor, the simplest and most intuitive way to state it is to say that, if there are any domestic or foreign market "distortions," free trade ceases to be optimal (and indeed even its superiority over autarky no longer follows as a necessity). Of course, these distortions imply market failures, including (in the cases where external prices of goods and services can be influenced by firms or governments, requiring us therefore to distinguish between average market prices and marginal revenues in trade) the failure of the market prices in trade to reflect the true social costs and benefits of trade. The post-war analytical developments in the theory of commercial policy, especially during the 1950s through the 1970s, have set out the basic principles, building an impressive edifice. They also constitute a substantial body of scientific analysis of several market failures, typically in the shape of factor market imperfections such as monopsony, generalized sticky wages, sector-specific sticky wages, and different varieties of wage differentials among sectors. The 1980s witnessed the extension of the theoretical analysis to product market imperfections, of both the large-group and the smallgroup varieties, completing the architecture of this theory. An important component of this theory, developed mainly during the 1960s, has also been the analysis of the implications for commercial policy when the conventional objective function that the economist maximizes becomes inappropriate. When "non-economic" objectives (such as the valuation in themselves of specific outputs such as manufactures or high-tech industry so that a dollar worth of output is valued at four, for instance) are admitted into the analysis, free trade will generally cease to be the optimal solution.2 Trade theorists have then analyzed, for such diverse non-economic objectives (which are abundantly in evidence and ceaselessly impose themselves on the political and policy scene), the optimal first-best and second-best ways of departing from free trade.

The cosmopolitan formulation While the study of commercial policy by economists has focused on national advantage and hence on the national formulation of the case for free trade, there is also the cosmopolitan formulation of the case for free trade. Premised again on the assumption that prices reflect social costs, the cosmopolitan argument simply elevates to the world level what the national argument did at the nation-state level. World efficiency follows free trade. But there is one overriding difference that follows for policy. If the case for national free trade holds as stated (i.e., the premises underlying it are

20

JAGDISH N. BHAGWATI

broadly reflected in the reality that the nation in question faces), then it really is irrelevant whether other nations follow free trade or not. "Fair trade" and "reciprocity," defined as the requirement that others follow free trade if one's own free trade is to make economic sense, are incompatible with this "essential" case for free trade. The prescription that emerges then is for unilateral free trade.3 Not so, however, under the cosmopolitan case. World efficiency requires that everyone follow free trade. It is simply not enough that one or more do so. The prescription now is: universal free trade. The question of reciprocity The contrast between the two cases for free trade, one implying that reciprocity does not matter and the other giving it central place, has however been less stark in practice. Unilateral free trade, as the ideal trade policy by one nation state, has been argued to be possibly deficient on theoretical grounds since Adam Smith. It has also been the subject of considerable debate among eminent economists, including at the end of the nineteenth century. Since free trade has not been widely practiced at most times, the demands for reciprocity have also implied threats to the pursuit of free trade for oneself, transmuting the issue therefore into one of free trade versus protection. At the same time, the appeal of reciprocity is such that it has spilled over in recent years into parallel demands for reciprocity and "level playing fields" in a variety of other ways, taking the form of objections to diversity among trading nations in their myriad governmental policies and national institutions, as preconditions for adopting free trade. It is necessary therefore to probe the issue of reciprocity further, prior to considering the old and the new challenges to free trade in sections 2 and 3 respectively. Reciprocity and the national case for free trade

Foreign barriers subject to change At the outset, it should be obvious that if the trade barriers of others can be regarded as changeable by the closure of one's markets, then a case can be made for departing from unilateral free trade in pursuit of national advantage.4 In fact, this was evident even to Adam Smith (1776, pp. 434-5): The case in which it may sometimes be a matter of deliberation how far it is proper to continue the free importation of certain foreign goods, is, when some foreign nation restrains by high duties or prohibitions the importation of some of our manufactures into their country. Revenge in

Fair trade, reciprocity, and harmonization

21

this case naturally dictates retaliation, and that we should impose the like duties and prohibitions upon the importation of some or all of their manufactures into ours. Nations accordingly seldom fail to retaliate in the manner. There may be good policy in retaliations of this kind, when there is a probability that they will procure the repeal of the high duties or prohibitions complained of. The recovery of a great foreign market will generally more than compensate the transitory inconveniency of paying dearer during a short time of some sorts of goods. To judge whether such retaliations are likely to produce such an effect, does not, perhaps belong so much to the science of a legislator, whose deliberations ought to be governed by general principles which are always the same, as to the skill of that insidious and crafty animal, vulgarly called a statesman or politician, whose councils are directed by the momentary fluctuations of affairs. When there is no probability that any such repeal can be procured, it seems a bad method of compensating the injury done to certain classes of our people, to do another injury ourselves, not only to those classes, but to almost all the other classes of them.

The case for the use of one's trade barriers to remove those of others raises, of course, several questions that have recurred in the debates on reciprocity through the last century: (i) Does one really have the power to effectively pry open others' markets; could this power not be overestimated, partly because of the self-interest-induced optimism of the lobbies that provide the political drive for such aggressive policies, leading to confrontational situations that create an ethos of mutual recriminations within which protectionist interests thrive? (ii) Will not the use of trade barriers for this purpose be "captured" by one's own protectionists for their purpose, leaving one saddled with tariffs? (iii) Could the power of example, rather than sanction, not suffice to induce others to follow free trade since one's own free trade should lead to prosperity at home whereas others' reliance on protection should undermine it abroad, making unilateral free trade a more efficacious and less dangerous path to reciprocity in free trade? As argued in section 2, the answers to these questions in end of nineteenth-century Britain and in end of twentieth-century United States have tended to be altogether different. In the former case, unilateralism survived the discord and debate. In the latter case, the use of muscle via threats to close markets in absence of concessions on unilateral demands made under the section 301 provisions of the 1988 Omnibus Trade and Competitiveness Act has emerged as US trade policy. 5

22

JAGDISH N. BHAGWATI

But while the questions posed above were predicated on the narrow question as to whether national advantage can be enhanced by the renunciation of unilateral free trade, the key questions (raised by the use of threats and actions to close one's markets to pry open others') take one inevitably into the cosmopolitan arena, i.e., one must ask questions such as: (i) Will not the use of power to extract unilateral concessions threaten the rule of law under which a trading regime should function, by legitimating the use of force and spreading such methods to other trading nations that can rise above impotence? (ii) Would it not also lead to concessions being granted to the powerful, diverting trade from the less powerful, replacing economic competitiveness by political clout as the guiding force behind trade, thus undermining world efficiency? (iii) Would not such use of power deteriorate also into unilateral (and therefore self-serving and biased) determination of the closedness of others' markets - a danger that is manifestly great when one is dealing with allegations of "invisible and inscrutable" trade barriers that can be flung freely at all and sundry but be forced to stick only for the benefit of the strong? (iv) Should we not regret the impact on world efficiency that would follow if the strong could fully indulge their own interest and browbeat the weak into removing protection or promotion (i.e., production subsidies) when in fact these could be correcting market failure rather than creating it?6 In fact, one could argue that if these adverse systemic (and hence cosmopolitan-efficiency-focused) implications of the pursuit of national advantage through the use of muscle to extract trade concessions are indeed likely to transpire, then the immediate gain in national welfare could well yield (in a reverse-J-curve outcome) to ultimate loss of welfare as the world trading system gets impaired. Foreign barriers unchangeable But what if the prospect of getting others to reduce their trade barriers is negligible? Then, the case for unilateral free trade stands intact, quite unambiguously. To recall Joan Robinson's famous observation, "you don't throw rocks into your own harbor just because others have done so in theirs." Of course, if there are any distortions to contend with, free trade ceases to be the optimal policy. And this applies equally to all market failures whether, for instance, in factor markets or in product markets. But there is an important implication for the question of reciprocity in trade which must be grasped since it is precisely the opposite of what is

23

Fair trade, reciprocity, and harmonization

Table 2.1. Reciprocity of free trade under alternative regimes: no-distortions and oligopoly Policy choices by home country / No distortions Unilateral FT (free trade) Reciprocity in FT

Foreign country

Utility of home country

Trade barriers given FT imposed or induced

II Oligopoly

Unilateral FT

Trade barriers given

Reciprocity in FT Appropriate intervention: departure from FT and/or domestic instruments*

FT imposed or induced

UFT{R) ^ UpT{U)

Trade barriers given

UINT > U

Source: *Eaton and Grossman (1986).

generally believed to be the theoretical consequence of market failures in the shape of imperfect competition and therefore provides ammunition to the interests that seek reciprocity. This implication is that when trade occurs with market failures (uncorrected by appropriate intervention), it is no longer possible to assert with Adam Smith that the successful removal of others' tariffs (and indeed other barriers and interventions) adds to the benefits from one's own free trade. In short, in the presence of market failure, it no longer follows that my free trade in the absence of yours necessarily makes me worse off than if we both had free trade (because, say, my insistence on reciprocity had led to the embrace of free trade by you as well). But reciprocity does offer added benefits to unilateral free trade when there are no distortions. But the popular belief, certainly in policy circles, is that the recent theoretical analyses of product-market imperfections imply that, because there are imperfect markets everywhere, we must insist on reciprocity in free trade: i.e., that the presence of oligopoly, for example, strengthens, instead of weakens, the case for reciprocity! Table 2.1 summarizes what we can say about this question, for the two contrasting cases: no distortions (and therefore also perfect competition)

24

JAGDISH N. BHAGWATI

and oligopoly (drawing on the recent analysis by the economists Brander, Spencer, Eaton and Grossman). It states that, in the former case, one can safely share Adam Smith's presumption that reciprocal free trade supplements the national gains from unilateral free trade. In the case of oliogopoly, however, this is no longer so. That, in some specific parametric cases one can show that reciprocal free trade is welfare-improving over unilateral free trade is, of course, compatible with the general proposition that it is not always or necessarily the case.7 Fairness and prudence The arguments so far presumed that the question of unilateral free trade versus reciprocity for national advantage could be assessed without reference to the role of emotions. But the policy question cannot be decided altogether in this fashion. Adam Smith, in the quote above, spoke of revenge as motivating the demand for reciprocity and compromising the ability to pursue a policy of unilateral free trade. But the sense that reciprocity is required by "fairness" is more pertinent today. Lack of reciprocity in free trade is generally considered, regardless of its impact on national welfare, as simply unfair. In consequence, there can follow a prudential argument for seeking reciprocity in free trade. The option of unilateral free trade may be impractical and attempts at keeping markets open unilaterally without attempting to extend free trade to one's trading partners and rivals may turn out in pluralistic countries to be a recipe for politically losing one's own free trade.8 Reciprocity and the cosmopolitan case for free trade

But the demands for reciprocity come perfectly naturally, as I have already indicated, if one approaches the question from the detached position of a "custodian" of world efficiency or, somewhat distinctly, from a "systemic" viewpoint. Cosmopolitan efficiency World efficient allocation of resources requires that no members of the world economy depart from free trade: the sin of one visits on all by compromising world efficiency (just as a monopolist is known to affect the Pareto-optimality of the economic equilibrium for all). It is necessary then to insist on all adhering to free trade: i.e., reciprocal embrace of free trade follows. To return to Joan Robinson's telling example, if any nation insists on throwing rocks in its harbor, we must throw rocks at it until it dredges them up. Systemic rationale But there is also the distinct argument that, if one is creating a trading regime, such as the GATT, to establish a rules-based

Fair trade, reciprocity, and harmonization

25

system, it is inevitable that it will impose symmetric or uniform rights and obligations on members. This is, of course, compatible with differences in obligations defined by accepted functional categories: e.g., developed and developing countries, the latter enjoying an asymmetry of obligations (i.e., Special and Differential Treatment in GATT language). But, outside of such functionally sanctioned discrimination, it is hard to imagine a regime that will tolerate asymmetries of free trade obligations. Reciprocity then belongs with the regime approach: it thus finds itself "naturally" embedded in the GATT. In fact, the GATT also generally embodies what I have called (Bhagwati, 1988) first difference reciprocity, i.e., matching reductions in trade barriers, as distinct from "zero option," or full reciprocity.9 2

The threats to free trade: from then to now

Now that I have distinguished among the alternative formulations, national and cosmopolitan, of the free trade doctrine, and analyzed the question of unilateralism versus reciprocity, it is possible to review with analytical clarity the manifold challenges to free trade since 1776 when Adam Smith invented both economics and international economics, grounding them in the principle of the division of labor. Several "episodes" of challenges to free trade can be distinguished over these two centuries and more. Of these, six (ending in the 1980s) fall broadly into one common pattern: they relate, in one way or another, to market failure. The seventh, now emerging, is radically different in nature and poses an altogether novel, and more complex, type of threat. The six market-failure related threats are now considered; the seventh, the principal subject*of this chapter, is addressed in Section 3. All are summarized and listed in table 2.2. EPISODE I Free trade with some theoretical exceptions: nineteenth century The repeal of the Corn Laws in 1846 ushered in free trade. It followed the shift from mercantilist thinking of Adam Smith and David Ricardo. The division of labor, and the gains from free trade, were in an essential way flip sides of the same coin. The relative importance of the roles played by the new economic doctrines and the industrial interests that would profit from cheap imports of corn have been the object of extensive debate; and the precise nature of the conversion by Prime Minister Sir Robert Peel has been analyzed with acute insight by Irwin (1987). For the rest of the nineteenth century, until near the end, unilateral free

Table 2.2. The threats to free trade since Adam Smith and David Ricardo" Episode

Economic circumstance

Economic theory

I 19th century

British dominance during prosperous century: no serious threat to British competitiveness since the repeal of the Corn Laws in 1846.

The case for free trade was qualified in three ways: 1 Infant industry argument in John Stuart Mill; 2 Monopoly power in trade argument in Robert Torrens and John Stuart Mill; 3 Possible use of tariffs to pry open others' tariffs; argument in Adam Smith.

Free trade with some theoretical exceptions

II End of 19th century Reciprocity and fair trade Infant industry protection

1 Britain's relative decline as No major theoretical developments transpired. Germany and the United States rose: the diminished giant syndrome (Bhagwati and Irwin, 1987) gripped Britain. "Fairness" and "Reciprocity" became fashionable demands. The belief in unilateral free trade was seriously challenged.

Threat level The conjunction of happy economic circumstance and the perceived inapplicability of the theoretical exceptions (to free trade) to British circumstance meant that these theoretical exceptions, of great potency in later periods, had very little impact on the embrace of unilateral free trade by Britain. •••

The threat to free trade was very real in the case of Britain, though it finally receded. The "latecomers," Germany and United States, practiced protection.

2 The flip side of the coin was that both Germany and United States were "coming from behind" and willing to use protection to industrialize, making the "infant industry" argument finally most potent. Ill

1930s

Macroeconomic failure

The onset of the Great Depression and the massive macroeconomic distress during and thereafter in the 1930s provided a compelling economic environment for protectionist demands to arise and flourish.

Keynes's change of mind away from free trade in the early 1930s foreshadowed the macroeconomic revolution that would effectively introduce the notion that tariffs could switch expenditure from foreign to domestic goods and thereby increase employment and national income. Later developments would include Joan Robinson's classic caveat about beggar-my-neighbor policies and, later still, the analysis of policies to maintain both external and internal balance under the adjustable peg system.

••••

The threat to free trade was not merely acute, reflecting both economic circumstance and new economic doctrine, but it also translated into reality through the 1930s.

Table 2.2. (cont.) Episode

Economic circumstance

Economic theory

IV 1930s and thereafter

The economic circumstance propelling protectionist forces was the Depression and its aftermath: it occupied center stage, to the exclusion of other forces. Nothing in the economic situation lent itself to demands for protection on grounds related to imperfect competition, in any event.

The pathbreaking developments in the theory of imperfect competition, by Edward Chamberlin and Joan Robinson, undermined the basic premise of the theory of free trade that prices would reflect social costs. However, unlike in the 1980s, this was not accompanied by significant developments in the trade-theoretic applications of imperfect competition. This reduced the overall forcefulness of this threat to free trade.

The demand for protection came, during this period, essentially from the developing countries. The presence of market imperfections, especially in their factor (rather than product) markets, was

Economic theory also developed, in tandem, to explore alternative factor market imperfections and to rank-order different trade and domestic policy interventions to improve welfare. Equally, the economic theory

Imperfect competition

V 1950s through 1970s Imperfect competition in factor markets and non-economic rtives

Threat level The lack of correspondence between economic circumstance and the new theory of imperfect competition, and the competing presence of the more powerful macroeconomic threat to free trade made the force of this threat to free trade pretty limited.

The developing countries did resort to protectionism generally. They successfully sought Special and Differential treatment at the GATT in order to do so. At the theoretical level,

considered a legitimate reason for protecting their nascent industrialization. Equally, "non-economic" objectives such as industrialization for modernization were considered to be valid reasons for departing from free trade. By contrast, the developed countries were committed to trade liberalization under GATT auspices. They enjoyed economic prosperity which reinforced, in a virtuous circle, the pro-trade policies.

of policy was extended to rank-order policy interventions, not to fix market failure, but to accommodate "non-economic" objectives. By granting legitimacy to such objectives, trade theorists developed new arguments for policy interventions, including departures from free trade. Both for factor market imperfections and for non-economic objectives, trade theory made gigantic strides in developing theoretical arguments for the first-best and second-best uses of tariffs, export duties and trade subsidies to improve on free trade.

however, the developments in the theory of commercial policy could be, and were, interpreted as narrowing down the scope for departures from free trade by showing that the first-best case for trade interventions was limited to cases where the market failures arose in the external markets whereas the best way to deal with other ("domestic") market failures ("distortions") was through domestic policy interventions instead. Equally, the use of trade instruments was optimal only when the non-economic objective was in the external sector.

Table 2.2. (cont.) Episode

Economic circumstance

Economic theory

VI Renewed imperfect competition in product markets

The economic situation was precisely the opposite of that in the 1950s through 1970s. The developing countries turned increasingly to trade liberalization (having seen the success of the Far East); whereas the developed countries started flirting actively with protectionism; the diminished giant syndrome that characterized Britain in Episode II returned to center stage, now in the United States, driving again both protectionist demands and arguments for "fair trade" and for "reciprocity."

The demands for protection and for fair trade and reciprocity coincided with theoretical advances in the theoretical analysis of imperfect competition in product markets, and the resulting application thereof to international trade theory in significant ways. These developments underlined sharply the message of the 1930s, in Episode IV, which had dealt with imperfect competition (in both product and factor markets), by specifically working out the economics of welfare-improving intervention through trade instruments under imperfect competition (of both the large and the small group variety) in product markets.

Kenewed diminished giant syndrome Return of fair trade and reciprocity

Threat level The coincidence of the demand for protection and reciprocity (resulting from the diminished giant syndrome) with the theoretical developments that could produce renewed demonstrations of the possibly beneficial effects of indulging these demands (in specific parametric situations) led to a significant threat to free trade, greater than during Episode IV and closer to that under Episode II.

Increased competitiveness due to several structural changes Fair trade in the world economy has led Harmonization and level playing to enhanced sensitivity to any fields foreign governmental policies and institutions (whose continuation in existing forms is itself taken as implying policy tolerance and therefore by implication as an act of policy) that may confer competitive advantage on one's rivals. This leads to impossible demands for harmonization, failing which there follow demands for managed trade, protection from unfair trade and absence of level playing fields, etc. VII Late 1980s and 1990s

Economic analysis of these new threats to free trade is only just beginning; it is entirely novel, different from all earlier threats (which reflected market failures). It should dominate theoretical international economics during the 1990s. It will intersect, in turn, with the major new developments in the theory of political economy and international trade during the 1980s.

The threat to free trade from the rising tide of demands for "level playing fields" and the ensuing demands for managed trade has gathered strength. It has the potential to become far more serious than most of the earlier, market-failure-variety threats.

Note: Seven threats to free trade since Adam Smith and David Ricardo are tabulated here. They are tabulated chronologically, with stars awarded under the column, Threat level, to rank-order them in terms of their efficacy in affecting free trade as the preferred policy.

a

32

JAGDISH N. BHAGWATI

trade was Britain's policy. There was no serious threat to British dominance as a producer of manufactures during this period. The economic circumstance was therefore propitious for the continued embrace of free trade: strength makes it easier for politicians and industry to believe that the Darwinian process of survival of the fittest will serve one's interests well in competition. The economic theory of free trade, however, was not left unchanged and unqualified during this period. It is only a mild caricature to say that, ever since Adam Smith invented the case for the Invisible Hand (and hence also for free trade),10 economists have made their reputation by inventing reasons why the Invisible Hand fails. John Stuart Mill, in that respect no exception, produced the "infant industry" argument (1848, p. 922): n The only case in which, on mere principles of political economy, protecting duties can be defensible, is when they are imposed temporarily (especially in a young and rising nation) in hopes of naturalizing a foreign industry, in itself perfectly suitable to the circumstances of the country. The superiority of one country over another in a branch of production often arises only from having begun it sooner. There may be no inherent advantage on one part, or disadvantage on the other, but only a present superiority of acquired skill and experience. A country which has this skill and experience yet to acquire, may in other respects be better adapted to the production than those which were earlier in thefield. . . A protecting duty, continued for a reasonable time, will sometimes be the least inconvenient mode in which the nation can tax itself for the support of such an experiment. But the protection should be confined to cases in which there is good ground of assurance that the industry which it fosters will after a time be able to dispense with it; nor should the domestic producers ever be allowed to expect that it will be continued to them beyond the time necessary for a fair trial of what they were accomplishing. Mill started an argument whose theoretical formulation underwent considerable refinement during the last century. And while it proved to be of great influence in subsequent periods and in the latecoming nations, with the British free traders forecasting accurately its potency in practice - John Bright, Cobden's great ally in the movement to repeal the Corn Laws, remarked that Mill's one paragraph justifying infant industry protection "would cause hereafter more injury to the world than all his writings would do good" 12 - it did little to influence British policy, of course, and remained a theoretical curiosum there. So did the argument for the use of tariffs to extract greater gains from trade by improving the terms of trade, originally produced by Robert Torrens in 1844. Widely debated, by Mill, Bickerdike, Marshall, and

Fair tradey reciprocityy and harmonization

33

Edgeworth, it has now become the celebrated argument for an optimal tariff. But again, it found no place in British policy, a policy judgment supported by most of the eminent economists of the time but a folly according to the historian McCloskey (1980).13 I would also be remiss not to recall that the case for free trade was debated almost exclusively from the viewpoint of national advantage. And that it took a unilateral form, though it was seen from the time of Adam Smith himself that the unilateralism assumed that the trade barriers of others were immutable. All in all, therefore, the period until the rise of United States and Germany by the last quarter of the nineteenth century was characterized by the dominance of (unilateral) free trade in Britain, by the development of two remarkable theoretical exceptions to the case for free trade, and by lively academic debate over the policy merits of these exceptions and over the further issue raised by Britain's embrace of unilateral free trade and concerning the efficacy and the wisdom of British attempts at turning instead to reciprocity in free trade. Since Britain remained wedded to unilateral free trade throughout this period, and there was little sentiment among the leading economists of the time to change this policy despite the theoretical debates outlined here, the threat to free trade during this period can be dismissed as negligible. It was indeed the heyday of this doctrine. EPISODE II Reciprocity and fair trade: infant industry protection, last quarter of the nineteenth century By the end of the nineteenth century, however, the relative decline of Britain and the rise of the United States and Germany in the world economy had produced in Britain the "diminished giant" syndrome (that would recur in the United States a century later as it confronted the rise of Japan and the Far East).14 The consensus in Britain over her policy of unilateral free trade was now to be shaken. "Fair trade" and "reciprocity" became the common words of discourse. They also became the focus of organized lobbying: during the 1870s and 1880s, fair trade organizations such as the National Fair Trade League, the National Society for the Defense of British Industry, and the Reciprocity Free Trade Association arose on the British scene. The Conservative party was deeply split on the issue. Academic opinion, however, remained united behind the doctrine of unilateral free trade and the wisdom of continuing adherence to it by Britain despite manifest protection by her new and successful rivals.15 Indeed, if unilateral free trade survived in Britain into the twentieth century, protection was embraced by the new economic powers, Germany and the United States. The appeal of the doctrine of free trade was less

34

JAGDISH N. BHAGWATI

compelling for the latecomers. Instead, the argument for infant industry protection, propounded ably by John Stuart Mill, had a greater resonance in these countries, just as it would after the Second World War in the newly independent developing countries. Thus, this episode during the last quarter of the nineteenth century, was marked by a serious, if eventually unsuccessful, challenge to the doctrine of (unilateral) free trade in the country that had embraced it in 1846 in a pioneering departure from the protectionism of the past, and by the explicit practice of protection by the newly emerging industrial powers. Both in terms of threat, to itself, and actual breach of its tenets, free trade must then be regarded as having been subjected to a serious challenge during this period. EPISODE III

Macroeconomic failure, 1930s

The most dramatic episode in the life of the (national) free trade doctrine, however, was to come with the onset of the Great Depression. The massive macroeconomic distress during and thereafter in the 1930s provided a compelling economic environment for protectionist demands to arise and to flourish. Writing in 1951, Hicks recounted how the unemployment of these years had seriously undermined the belief in the doctrine of free trade:16 The main thing which caused so much liberal opinion in England to lose its faith in free trade was the helplessness of the older liberalism in the face of massive unemployment, and the possibility of using import restriction as an element in an active programme fighting unemployment. One is, of course, obliged to associate this line of thought with the name of Keynes. It was this, almost alone, which led Keynes to abandon his early belief in free trade. (1959, p. 48) Keynes's apostasy on free trade had been suggested in A Treatise on Money (1930) and in his evidence before the MacMillan Committee in February 1930 where he offered the view that tariffs, while unwise as a long-term policy, could immediately alleviate the slump.17 This viewpoint however becomes more pronounced in Keynes's thinking and writings through 1931, resulting in a celebrated controversy with Lionel Robbins and the Beveridge-led (1931) riposte by Robbins, Hicks et al. to Keynes in Tariffs: The Case Hxamined (Beveridge, 1931). Interestingly, Keynes seems to have anticipated the later objection that the superior intervention to achieve full employment was domestic reflation rather than expenditure-switching protection: If nations can learn to provide themselves with full employment by their domestic policy . . . there would no longer be a pressing motive why one

Fair trade, reciprocity, and harmonization

35

country need force its wares on another or repulse the offerings of its neighbours. International trade would cease to be what it is, namely, a desperate expedient to maintain employment at home by forcing sales on foreign markets and restricting purchases, which, if successful, will rarely shift the pattern of unemployment to the neighbour which is wasted in the struggle.18

Later theoretical analysis would then show how, under fixed exchange rates, the reflation would cause external imbalance and therefore the two policies, reflation and devaluation, would be generally necessary to attain the two targets of external and internal balance. Tariffs would then appear to be inferior to both devaluation and to the optimal combination of devaluation and reflation as the policy solution to unemployment. But these insights came later. During the 1930s, Keynes's renunciation of the doctrine of free trade remained a potent source of disbelief in the doctrine. Combined with the massive unemployment unleashed by the Great Depression and the lingering aftermath of its distress, this apostasy turned the 1930s into the most deadly episode among the challenges to the (national) doctrine of free trade. EPISODE IV

Imperfect competition, 1930s and thereafter

The 1930s also witnessed the emergence of a threat to free trade from an altogether different direction. It came, not from economic circumstance as the earlier threats had partly or wholly done, but entirely from theoretical progress — and that too in a curiously tangential way. In the early 1930s, Edward Chamberlin (1933) and Joan Robinson (1933) independently came up with important theoretical analyses of imperfect competition, opening up to systematic exploration the middle ground between perfect competition and pure monopoly.19 The result was to undermine seriously the notion that market prices reflected social costs, calling into question, more widely, the virtue of laisse^ faire, and, more narrowly with it, the merit of free trade as well. The economists of the Chicago School correctly saw this as a threat that would legitimate interventionism. Therefore, they proceeded to counter the threat by taking to econometrics to demonstrate that, although markets seemed imperfect to the naked eye, in reality there was "as i f or "working" competition and that the imperfections did not amount to a hill of beans.20 Today, we talk, not of "as if" competition but of "contestable markets": but, by and large, the key thought is the same. Despite Chicago's riposte, however, the skepticism about prices not reflecting social costs due to imperfect competition remained a potent source of erosion of the belief that free trade was a desirable policy. In his

36

JAGDISH N. BHAGWATI

earlier-cited essay of 1951, analyzing the different reasons why the doctrine of free trade had lost "much of its strength" and "been called into question," 21 John Hicks captured this reality well:22 the Monopoly-Competition argument . . . is of much less practical importance than the others, but it deserves at least a passing mention, because of the undoubted influence which it undoubtedly exercises - in a negative sort of way - upon the minds of economics students . . . If apparent costs only equal true costs under conditions of perfect competition, and competition hardly ever is perfect, the bottom seems to drop out of the Free Trade argument. This is in fact a fair description of the state of mind which quite a number of economics students seem to have reached. (1959, p. 46)

But the damage that the theory of imperfect competition did to the policy of free trade cannot be argued, as Hicks also suggests, to have been serious. There were two reasons. First, there was nothing in economic circumstance that led to the demand for protection; citing this kind of theory the attack on free trade was of the nihilistic variety and hard to tap by specific interest groups. Besides, free trade was already imperilled far more seriously by Keynes's desertion, by the economic circumstance of massive unemployment, and by the new macroeconomic ideas that I discussed as Episode III. When economists returned to imperfect competition in the 1980s, the threat would become more serious: economic circumstance would have changed, with more compelling craving for protection, and the idiom and substance of the new work on imperfect competition would be more readily accessible for capture by the protectionist interests. Episodes V and VI are more recent and familiar and require no elaboration. 3

Fair trade, reciprocity and harmonization, the 1990s

Until now, therefore, the challenges to the doctrine of free trade were concentrated on its national version. In turn, they reflected varying combinations of economic circumstance (fuelling the demand for protection) and economic theory (fuelling the supply of protection). 23 The latter again reflected different varieties of market failure that undermined the basic premise of the free-trade doctrine: that market prices should reflect social costs. But, even if there was no evident market failure, the question raised by Adam Smith had to be addressed: would one renounce free trade for oneself if others did not embrace it too? This issue of unilateralism versus reciprocity was never wholly dormant; but it turned to center stage during the two episodes (II and VI), a century apart, when Great Britain and the United States, both free traders in actual policy and

Fair trade, reciprocity, and harmonization

37

Table 2.3. Theoretical arrangements against free trade for oneself Episodes

Free trade: essential case

Unilateralism versus reciprocity

Episode I

Doctrine was accepted but with theoretical exceptions for two types (1) infant industry and (2) variable terms of trade^ by, for example, John Stuart Mill, Robert Torrens. Same as in Episode I

Unilateralism was accepted but a theoretical case was made for tariffs to pry open foreign markets, and hence for reciprocity, by Adam Smith. No new theoretical case of substance was developed for reciprocity, but several arrangements for unilateralism emerged in the great debate over fair trade and reciprocity that marked the British scene as Britain was afflicted by the diminished giant syndrome. Unilateralism was not an issue.

Episode II

Episode III

Episode IV

Episode V

Episode VI

Doctrine was frontally rejected by Keynes: macro unemployment was the source of market failure now. Doctrine was undermined because of the new theories of imperfect and monopolistic competition: (1) Imperfect competition in product markets: monopoly, oligopoly, and the large-group case; and (2) Imperfect competition in factor markets: monopsony (Joan Robinson). During this period, there was focus on two sets of market failure: (1) Imperfect competition in factor markets: monopsony, wage differentials between sectors, sticky or minimum wages, etc., and (2) Non-economic objectives-, the failure of markets to reflect non-economic objectives. The market failure focused now was that arising from imperfect competition in product markets: oligopoly; and the large-group case.

Unilateralism was not an issue.

Unilateralism again became an issue, reflecting both the renewal of the diminished giant syndrome (now in the US) and demonstration that lack of reciprocity may produce loss, rather than gains, vis-a-vis autarky.

38

JAGDISH N. BHAGWATI

self-perception, confronted the rise of new and successful trading rivals, either avowedly protectionist or accused of being deviously so, and were engulfed in demands for reciprocity in trade. "Free trade by oneself" would have to confront "free trade for all or free trade for none." Indeed the emphasis on reciprocity and its kin, fair trade, is a natural outgrowth of economic weakness or, more accurately, of perceivedeconomic weakness. The shifted focus on to what others are doing that may aid their producers vis-a-vis ours is an inevitable consequence of the "competitiveness" concerns that attend the rise of new and fiercely successful rivals abroad. In the earlier episode relating to the affliction of many in Britain by the diminished giant syndrome and for much of the 1980s in the parallel case of American affliction, the concerns were focused largely on whether others had trade barriers as well. Doubtless, tariffs by others reduced one's gains from trade: a matter that strong countries with commensurate faith in their strength would not elevate to a policy concern. But not so, when the strong, feeling weak, faced the weak. The giants, diminished relative to erstwhile Lilliputians, were still giants. So the academic (and policy) debate that ensued was mainly about the empirical relevance, and the wider systemic consequences, of Adam Smith's old "strategic" argument. In Randolph Churchill's classic phrase, the question became: should free trade be renounced in favor of protection as an instrument for opening foreign markets, the way one pried open oysters with a clasp knife?24 But, in the recent replay, the reciprocitarians and fair traders have gone considerably further than seeking a mutuality of trade barrier levels and reductions therein, raising an altogether different and more dangerous kind of threat to free trade. Let me explain. The fair trade notion now extends to any foreign governmental policy or institution that is different from one's own. Such differences, myriad in principle, are assumed to create an unfair advantage of one kind or another in favor of foreign producers, implying the absence of "level playing fields." This results in demands for getting these policies harmonized (in practice, to what the diminished giant prefers, reflecting of course its own political and ideological constraints and objectives), or (failing harmonization) for getting the alleged trade consequences of these different policies neutralized through countervailing duties, subsidies, etc. The latter set of policies, of course, amounts to protectionist intervention on behalf of one's industries. The proponents of managed trade, i.e., a fix-quantity as against a fix-rule regime, also draw comfort from the widespread notion that, if level playing fields are absent, free trade does not make sense. The delegitimation of free trade makes the case for managing trade through bureaucratically set

Fair trade, reciprocity, and harmonization

39

market shares and allocations of production and trade targets seem like the solution that is both natural and desirable when trading countries have different policies and institutions. Reasons for the rise of fair trade The demands for fair trade are not altogether novel. They have been more or less confined traditionally to foreign subsidies and predatory dumping, both of which have long been countervailed, both before the GATT arrived on the world stage and certainly under GATT auspices. But today the demands have grown out of hand.25 They have been extended, in the case of the Structural Impediments Initiative with Japan by the United States, to nearly 240 items including Japan's savings rate, her working hours, her retail distribution system, her keiretsus, etc. The 1988 Omnibus and Competitiveness Act now defines as unreasonable, and hence actionable, a variety of foreign policies and institutions such as labor standards and export targeting. Differences in environmental policies have also come under the fair trade rubric: Senators Baucus and Danforth have recently introduced legislation in Congress to countervail differentially smaller pollution abatement costs of production in foreign countries. Believe it or not, there have even been proposals to equalize similarly, through countervailing duties, the differentially higher legal costs imposed on domestic producers by the "extortionary" tort system that American lawyers have gifted to their nation 1 Many of my examples come from the United States where the diminished giant syndrome has prompted an acute search for unfair trade by others. The payments deficit and the special chord that fairness plays in the American political psyche have further fueled this search. But I suspect that there are also broader forces at work which will make this a continuing phenomenon that should further diffuse internationally: Since it is easier to get protection if one alleges unfair trade by others than if one simply pleads for assistance citing the difficulty of one's situation, once the unfair trade route is discovered it will be well traveled. Given the decline of transparent tariffs and the rise of nontariff barriers, of varying degrees of transparency, it is somewhat easier to charge others with unfair trade. Thus, for instance, the tendency has become manifest to "construct" what imports "should" be and are not, just as in anti-dumping actions it has become customary to "construct" fair value (and win through exploiting the scope for chicanery that this procedure opens up).

40

JAGDISH N. BHAGWATI

The slide to unfair trade allegations has also been facilitated by the fact that the main newly-emergent trading nation has been Japan with its manifest differences of culture and economic organization, and with its long history of being regarded since the 1930s by its fearful rivals as an unfair trader. The continuing globalization of the world economy, with its spider's web of criss-crossing investments and rising shares of trade to GNP, has increasingly made the force of international* competition more palpable to many, making producers ever more sensitive to whether, in one way or another, their rivals gain an unjustified, unreasonable, unfair advantage in the lethal struggle for markets and survival. The volatility of exchange rates, a fact of life despite "managed floats" since 1971, may have contributed to this sensitivity, though there is little econometric support for the thesis that trade volumes have been adversely affected in any significant fashion by the arrival of flexible exchange rates. For these powerful reasons, the concern over whether others are fairly trading, whether in our markets or in theirs or in third markets, has intensified. It will grow, rather than diminish. It can overwhelm the fix-rule, open trading system unless it is confronted; for, in principle and in revealed practice, the scope for unfair trade allegations is open-ended. A Pandora's box is open. Confronting fair trade How does one cope with the serious consequences of fair trade, ever looming larger, for the doctrine and regime of free trade? A pragmatic response would simply be to get nations to agree to a set of constraints beyond which the absence of level playing fields cannot be a ground for suspending free trade. This certainly should help; but it does not get to grips with the underlying theoretical problem that the issue of fair trade creates and which requires us to reconsider the way we think of free trade. This theoretical question arises because for over two centuries we have deduced free trade essentially as a mutual-gain trade arrangement among nations with private agents differentially endowed with tastes, technology, and resources, but with governments that exist only to implement our recommendations on trade policy. Free trade, and letting market-determined comparative advantage work, are thus two sides of the same coin: for, in our basic models of commercial policy where free trade and the gains

Fair trade, reciprocity, and harmonization

41

from trade are analyzed, governments simply play no independent role. But as soon as you let governments in, with their own myriad policies, they will wittingly or unwittingly affect comparative advantage: i.e., few policies are "neutral" in their impact on resource allocation. Thus, as soon as governmental policies are taken from the backburner and put on to the table, as the fair traders are now doing, there is simply no way that we can pretend that there is a (wholly) market-determined comparative advantage. Comparative advantage is inevitably "distorted," "created," in fact "shaped" by myriad governmental policies, wittingly or unwittingly. Therefore, if fair traders are to be confronted, we need a revolution in the way we think about the theory of commercial policy and gains from trade, divorcing it altogether from (untenable) notions of "market-determined" comparative advantage. In short, having discarded this baggage of marketdetermined comparative advantage, we need to ask questions such as: Can free trade produce mutual gains from trade even if two nations have different policies on, say, retail distribution or minimum wages or working hours or abatement of pollution entirely affecting one's own citizens? Or do such differences cause predation by one at the expense of the others under free trade (relative to autarky or other forms of intervention)? Would different types of harmonization in one's image, or the others', or in an altogether different fashion produce more overall gains from trade? If so, how would it affect the distribution of the gains from trade? When can one expect that non-harmonization will imply greater gains from trade for each nation, just as greater differences in private tastes between countries generally would increase trade gains? But before I consider these novel ways of getting at the way we must now address the question of free trade, putting governments and their actions squarely into our models, let me turn to the pragmatic agenda for containing the fallout from the growing flood of fair trade. Pragmatic solutions Slight reflection, if not familiarity with the way in which the traditional fair trade mechanisms in the shape of CVD and AD processes have evolved, shows at least four pragmatic forms of containment: Principle of proportionality For any specific policy where the absence of level playing fields becomes contentious, the principle of proportionality

42

JAGDISH N. BHAGWATI

can be applied: i.e., we can accept procedures which eliminate complaints that do not amount to a hill of beans. In essence, the application of an injury test is precisely a test of proportionality. Whereas it deals with proportionality of consequences, proportionality could equally be applied to the gradient on the level playing field: e.g., an anti-dumping margin of less than 20 percent may be declared inadmissible as a ground for action.26 Principle of intentionality Yet another way to contain the fair traders would be to ask whether any particular playing field is intentionallyfixedin favor of the home team or whether one is dealing with unwitting consequences. Thus, a nation subsidizing its higher education is shifting the endowment of skilled labor and therefore indirectly favoring those activities that used skilled labor. But the specific trade effect is unintended. So this range of governmental policies should be excluded from the fair traders' grasp. Principle of selectivity Again, one may agree to exclude policies that are not selectively designed in their impact. Thus, a general R&D credit, not confined to a specific industry or set of (say high-tech) industries, may be accepted as admissible even though it may have non-neutral effects on resource allocation. Equally, accelerated credit for investment amounts to an interest-free loan that can affect the choice of technique and also differentially impact on different sectors with different capital intensities. Yet, because it is applied non-selectively, it may be accepted as a policy that should be exempt from free trader's attention. Principle of proximity Finally, one may delimit fair trade concerns to include only policies that proximately, rather than indirectly, affect comparative advantage. Thus, agricultural export subsidies are offensive but not income support payments that keep farmers alive and may indirectly therefore lead to more production and exports. Principled solution: redoing the theory of free trade But these ways of containment of the fair traders, while helpful, do not get at the basic problem that fair traders find with free trade: that, in the absence of level playing fields, free trade is illegitimate since it distorts in one way or another the market-determined comparative advantage. Since governments will not self-destruct to oblige either Bakunin or Friedman, since politicians get elected or seize power to do myriad things, and since harmonizing everything is generally impossible, a theory of free trade and the gains from trade that does not squarely put governmental

Fair trade, reciprocity, and harmonization

43

policies (other than the trade policies which we seek to rank-order for benign governments to consider) into the specification of the model and its analysis will simply fail to address the concern that agitates the fair traders and undermines free trade. It is no longer enough, as in the classic welfare-theoretic analysis of Samuelson (1939, 1962), Kemp (1962), Kenen (1959), and others to argue that, in conventional economic models that postulate absent governments, we can prove well-known theorems such as that free trade dominates no trade (FT> NT) for national welfare and for world welfare, and that free trade is a mutual-gain policy for trading nations. We need to rewrite our models to allow for governments explicitly to have other policies, harmonized or otherwise, and to ask questions such as whether FT> NT in the presence of such policy differences or similarities. In particular, we need at minimum to investigate, for any particular such policy pursued by different governments in different ways so that level playing fields are absent, whether FT still dominates ATT and that F T leads to mutual gain for the trading nations. But this, and related questions which I suggest for theoretical analysis below, imply that governments must be explicitly modeled: i.e., we must build the theory of free trade on the foundations of the theory of political economy that has been developing through the 1980s, but with altogether different sets of questions. Equally, since these other policies may be driven (as in the case of environmental and labor-standards policies) by considerations which suggest that they define "non-economic objectives" (a la Corden, 1957; Johnson, 1965; Bhagwati and Srinivasan, 1969), it may be necessary in the case of some of these policies to augment the social objective function to include considerations such as: despite gains from trade, will FT not lead to more pollution or harm to child labor abroad? Thus, in the analysis required to address the concerns of fair traders and nonetheless argue for free trade, it would sometimes be necessary to combine two great traditions in the theory of commercial policy: the incorporation of non-economic objectives into the analysis, which is part of the early post-war development of the theory of commercial policy, and the explicit treatment of government, which is part of the recent (politicaleconomy) developments in the theory of commercial policy. Again, while the analysis so sketched would treat these other policies as exogenous to the choice of trade policy, so that comparisons such as F T versus NT are carried out subject to these policies pre-set in the analysis, we must also allow for the fact that some of these policies are endogenous to the choice of trade policy itself, e.g., if my labor standards are more stringent than yours, or my pollution abatement requirements are more drastic than yours, the competitive pressures that this builds on my industry in free trade may

44

JAGDISH N. BHAGWATI

lead to political lobbying to reduce environmental and labor standards at my end, in which case those of us who value environmental and labor standards would have good reason to object to free trade if harmonization of these standards to our own preference is not obtained. Thus, we need in the following analysis of fair-trade-prompted refashioning of the theory of commercial policy to distinguish among three classes of possibilities: (i) other policies exogenous; conventional objective function; (ii) other policies exogenous; non-economic objectives considered; and (iii) other policies endogenous; non-economic objectives considered. Let me indicate each type of analysis in turn, setting out the way we need to develop the theory rather than taking any particular policy and undertaking the theoretical analysis I argue to be necessary. Exogenous other policies; conventional objective function

This is the most "natural" case to consider, for a variety of policies where demands for harmonization (i.e., for level playing fields) have been raised. What questions should we ask here? Where we traditionally derive theorems that state that, for two countries I and II, F 7 > N T such that t/fT ^ U?7 and UFJ ^ 1/J7, we must now ask the following questions for the cases where there is harmonization (H) and non-harmonization (NH) and the countries then go from NT to FT. Where harmonization occurs with country Ts policy, denote it by H(I); at country IPs level, it is denoted by H(II). Question 1 The most important question which we must ask then, and which is the natural counterpart of the earlier free-trade doctrine in the absence of governments, is: will free trade, without harmonization, still lead to mutual gain* — i.e. can we still find that TTNH,FT^TTNH,NT

Uj

^Uj

,

TTNH,FT^

; and Un

TJNH,NT

^ Un

This is the key question since most fair traders seem to imply that predation, rather than mutual gains, will follow from free trade in the absence of harmonization. I would suspect that, in most cases, differential policies, left unharmonized, will be compatible with mutual gains from free trade; but this analysis is part of the agenda that awaits the theorists of commercial policy in the 1990s. Question 2 Harmonization: them like us - But the fair traders may well be interested in a different question as well, i.e., willfree trade with harmonization

Fair trade, reciprocity, and harmonization

45

imposed on the other country in one's image produce more gains than free trade without

harmonization? \ suspect that, in the case of the US's SII policy toward Japan where attempts are being made to remold Japan's retail distribution system, her keiretsusy etc. so that Japan gets closer on these policies to the US, the motivation is as much the fear that otherwise there will be predation (i.e., question 1 will be answered in the negative for the US) as the chagrin that the gains from trade are being diminished for the US by Japan's differential policies. This then leads to the comparison and question whether

Question 3 Harmonization: us like them - The question can then be turned around to ask if one's welfare would increase, relative to non-harmonization, if one changed one's policies in the image of the other. This has been suggested, for instance, by Japan-fixated writers such as Prestowitz, and by other admirers of the Japanese system as imagined by them, with regard to institutions and policies such as the keiretsus and industrial policy. In this case, the assumption to be examined, for any policy in question is, whether UH(II),FT>UNH,FT

An extremely important observation to make here is that there is no presumption at all that these harmonization versus non-harmonization comparisons will have the former win. If anything, policy differences between countries may act like private taste differences and lead to greater, rather than less, gains from trade. But this is precisely the issue that awaits formal analysis for different kinds of policies. Question 4 In case of both forms of harmonization, we should also ask the world-welfare or cosmopolitan-efficiency questions as to whether harmonization with one or other policy improves world efficiency over the case where non-harmonization prevails, and how each harmonization compares against the other, i.e., we should ask to compare27

{U?W'FT,

U%U)'FT} versus {U™-FT, U™-FT}

(2)

and

Exogenous other policies; non-economic objectives considered

As with the usual analysis of non-economic objectives, if the policies in question are treated as furthering desired social objectives, there

46

JAGDISH N. BHAGWATI

may be a problem and a necessity to incorporate these into the welfare analysis at hand. Take, for instance, labor standards. If child labor is prohibited in the EC, but not in South Korea or Mexico (whether de jure or de facto), then comparative advantage may indeed shift in favor of South Korea or Mexico for labor-intensive industries that rely on child labor. But this creates no particular social-objective problem within the US where child labor laws continue to be enforced. But suppose that non-governmental organizations claim jurisdiction over what happens in South Korea or Mexico, saying that, as planetary citizens, they have a stake in what happens to children abroad as well. Then, the expansion of child labor-using industries in South Korea or Mexico, thanks to free trade, creates a non-economic disutility which must be taken into account. The EC will have to incorporate into its utility function, if this non-economic objective is held to be significant within the EC, the fact that more child labor in employment abroad means less utility at home, i.e., as done in the formal analysis of non-economic objectives (Bhagwati and Srinivasan, 1969), the EC utility function must be rewritten as ,Cn;LFc)

where Cv . .. Cn are consumption levels of goods and services 1 .. . n in the EC as in conventional analysis, and Lc is the child-labor employment in foreign countries and is a source of disutility. While this is the formal way to rework the gains-from-trade analysis in this instance, I might add that the question of asserting one's preferences over what is conventionally regarded as a matter within the domestic jurisdiction of other countries by way of social policy, and then proceeding to suspend other nations' trading rights as defined by their GATT membership, has become a major issue today. Increasingly, of course, matters which were regarded as within purely domestic jurisdiction have been internationalized: human rights is the most striking example. The question has arisen now in an acute way with environmental issues, as I noted earlier, even with regard to issues such as differences in pollution abatement policies where no physical international spillovers on to other nations are involved and where instead the environmental impact is entirely domestic and therefore the environmental policy chosen by any nation should be a matter of its own choice, reflecting factors such as its intertemporal preferences over income and environment, resources, and technology (including the current and anticipated technology of pollution prevention now and pollution cleanup later). From the viewpoint of the world trading system, the tendency to so assert jurisdiction over what others do or do not do is dangerous because it

Fair trade, reciprocity, and harmonization

47

has prompted a tendency to enact national legislations and actions by governments to unilaterally suspend other nations' trading rights when they do not have policies that conform to one's own preferences. If each nation proceeds to act unilaterally in this fashion, we face a prospect where righteousness, instead of protectionism, will reproduce the trading chaos of the 1930s! Unilateralism in these matters is therefore a lethal development. But it is well in conformity with the short-sighted mentality that has produced the unilateralism embodied in the 301 policy in the US and which is applauded inside the beltway by those who want "results" and are unable, or do not bother, to see the systemic implications of such policies.28 Endogenous other policies: non-economic objectives considered

The theoretical analysis changes further when, in matters such as environment and labor standards, the policies designed to ensure a targeted fulfillment of such objectives become endogenous to the trade policy chosen. There is, in political-economy-theoretic modeling of these questions, an interesting interplay between trade policy and, say, environmental policy. The environmentalists fear free trade will, by putting competitive pressure on their industries that bear differentially higher abatement costs, trigger demands to reduce the environmental standards; so they ask for harmonization by others up to their own domestic standards or else they oppose free trade. The industry interests seek harmonization, given the stringent domestic standards, on competitiveness grounds. The two groups then unite in their demands for fair trade and harmonization by others to the more stringent standards at home, failing which free trade is to be rejected. Where such harmonization is rejected by others, either as incompatible with their optimal choice given their own constraints and objectives or as impossible to enforce effectively (as with, say, child labor laws), this becomes a surefire recipe for protectionism. By asking for harmonization that is impossible to get, one can then get the protection that is otherwise beyond reach. In fact, there is also the possibility then that protectionist lobbies may "strategically" set socially sub-optimal higher domestic environmental standards, joining with environmental groups equally to make demands for harmonization up to these higher standards which are then calculated to be impossible to meet by foreign rivals, and then ensuring the grant of protection because of lack of level playing fields. This observation also suggests that the theoretical analysis must also allow for the fact that these other policies, such as the environmental policy, may have been set at "distorted" levels and indeed in a sub-optimal fashion.

48

JAGDISH N. BHAGWATI

Thus, in the theoretical analysis I suggested earlier, comparing FT with NT subject to harmonized and non-harmonized such policies, the answers will vary, for my specific policy problem, depending on how these other policies are defined and set. 4

Concluding remarks

I have only been able to sketch here the kinds of questions that must now be explored by the theorists of commercial policy during the 1990s. Only if we rework the theory in these altogether new ways can we hope to contain the slide toward fair trade and its ill consequences: else, we will simply not be equipped to respond to the growing demands for level playing fields. I should add, in conclusion, that I have touched only tangentially on the unfair capture of the conventional fair trade mechanisms: the CVD and AD procedures. We are all in great debt to the economists such as Michael Finger, Brian Hindley, and Patrick Messerlin, and to the lawyers such as David Palmeter, John Jackson, and Michael Trebilcock, for illuminating analyses of the capture of the AD mechanism, in particular, by protectionist interests in the US and, even more so, in the EC. In turn, these findings raise yet another powerful agenda for complementary research to what I have suggested so far: how do we design these mechanisms so that we minimize the risk of such capture? We have to make sure that those who allege the lack of level playing fields are not really seeking the higher ground! The analysis of these questions takes us directly into the new microeconomics of information and incentives. It must take center stage as well in the agenda of research for the 1990s.29 In short, the agenda for fundamental thinking on the theory of commercial policy, or what James Meade beautifully entitled trade and welfare, in the 1990s will have to build squarely on foundations laid in the 1980s and earlier, not in the theory of market structure and related market failures, but in the theory of optimal intervention in the presence of non-economic objectives, in the theory of political economy in open societies that explicitly treats governments in the formal analysis, and in the theory of the design of institutions. Notes This paper is the culmination of a line of thinking which I had begun, in a somewhat primitive and preliminary fashion, in several writings from the mid-1980s, especially in the 1987 Bertil Ohlin Lectures (Bhagwati, 1988) and more notably in the 1989 Harry Johnson Lecture (Bhagwati, 1990). I have profited from conversations with Douglas Irwin who is currently engaged on a major historical

Fair trade, reciprocity, and harmonization

49

review of the vicissitudes of the free trade doctrine since 1776, tentatively titled Economic Thought and Free Trade. The views expressed here are strictly personal. A version of this paper will appear in a volume by R. Stern and A. DeordorfT forthcoming from the Michigan University Press. 1. Nearly one hundred and fifty years ago, even John Stuart Mill and Robert Torrens had developed the two celebrated theoretical exceptions to the case for free trade: the infant-industry and the monopoly-power-in-trade arguments, so that no student of international economics or informed policy-maker could possibly have believed that economic theory could not generate arguments for departure from free trade. Given however the susceptibility of many economists unfamiliar with the field of trade theory to such claims, and the eagerness of protectionists to embrace them, some of us have been moved to write against such claims even though they are trivially easy to refute and we have more interesting things to do (see, for instance, Bhagwati, 1989). 2. Strictly speaking, market failure will arise now in the sense that the true social costs, reflecting the costs and benefits as redefined by the inclusion of the non-economic objectives, will not equal the market prices. 3. I qualify this below to argue that this case for unilateral free trade has been modified, since Adam Smith himself, to allow theoretically for departures from free trade when free trade does not obtain elsewhere. 4. Strictly speaking, one would need to distinguish between the cases where the threat to close one's markets suffices to pry open others' markets and where the threat must be made credible by actually implementing it. In the latter case, one must at least set off the current or "first-period" losses from one's protection against the "later-periods" gains from reduced barriers abroad. The optimal policy choice should then reflect the usual considerations such as the intertemporal objective function of the nation. 5. For an extended analysis, see Bhagwati and Irwin (1987) and Bhagwati and Patrick (1990). The relevant Sections of the 1988 Act are 30O-310 but are popularly subsumed under the rubric of "301." 6. Alfred Marshall's objection to the use of British protection to remove the protection of latecoming nations was based in part on the notion that these foreign nations were justified in using "infant industry" tariffs (cf. Bhagwati and Irwin, 1987). 7. The effect of the demonstration of such a possibility of welfare-improvement in Paul Krugman's (1984) classic article on import restrictions leading to export promotion appears to have been precisely to mislead many into deducing erroneously that imperfect competition in product markets implies a strengthening, rather than weakening, of the case for reciprocal free trade. 8. Under imperfect competition, as noted in the table for the case of oligopoly, reciprocal free trade need not be a desirable policy (for national advantage) compared to imposing appropriate intervention in the presence of exogenously specified foreign trade barriers. 9. I have discussed several economic rationales tot first difference reciprocity in my 1990 Harry Johnson Lecture (Bhagwati, 1991). 10. One can correctly say that Economics and International Economics were born simultaneously in Adam Smith's The Wealth of Nations.

50

JAGDISH N. BHAGWATI

11. Quoted by Irwin (1991a). 12. Cf. Irwin (1991a, p. 203). 13. Also see Irwin (1987) who argues, based on econometric estimates of the relevant elasticities, that the optimal tariff could not have been large and that the maximal estimate of the loss from adopting free trade may have been of the order of 0.5 percent of GNP. 14. Bhagwati and Irwin (1987) compare the two historical phenomena, drawing both parallels and contrasts. 15. The academic defense of unilateral free trade by the best economists of the time, including the celebrated Alfred Marshall of Cambridge and F.Y. Edgeworth of Oxford, was richly textured and nuanced. It is reviewed in Bhagwati and Irwin (1987) and Bhagwati (1988). 16. Hicks' (1959) essay on free trade and modern economics was read to the Manchester Statistical Society in March 1951. 17. I am indebted to Barry Eichengreen (1984), Bernard Wolf and Nicholas Smook (1988), and Douglas Irwin (1991b), who offer a richly textured analysis of Keynes's views on free trade. 18. Cited in Irwin (1991b). Joan Robinson (1937) and Nicholas Kaldor (1950-1) analyzed further the argument that expenditure-switching policies to deflect expenditures on to oneself were "beggar-thy-neighbour" policies. 19. There were differences between the two pioneers. In particular, Chamberlin's analysis was deeper on product differentiation whereas Mrs Robinson's analysis of monopsony and price discrimination would prove seminal. Chamberlin insisted on differentiating his analysis of "imperfect" competition. Among those innocent of the resulting acrimony was D.H. Henderson (the Drummond Professor of Political Economy at Oxford at the time). Presiding over Chamberlin's lecture at All Souls' College, Henderson introduced him, with unintended double entendre, "as the father of the theory of imperfect competition." 20. I have discussed this response by Chicago at length in my obituary (1978) of Harry Johnson whose views, on migrating to Chicago from England, had progressively changed toward those of the Chicago school. An alternative response to the assertion that markets are imperfect and therefore, say, free trade must yield to welfare-improving intervention, would be to argue along with the Public Choice school that intervention will, in practice, make things worse rather than better. 21. Hicks writes: "Free Trade is no longer accepted by economists, even as an ideal, in the way it used to be. Economics has not lost authority, but the preponderance of economic opinion is no longer so certainly as it was on the Free Trade side." (1959, pp. 41-2). Doubtless, the reader will want to juxtapose Hicks' forthright statements on the challenges to free trade, as indeed his masterly analysis thereof, with Krugman's quote I started with in this chapter. 22. Interestingly, Hicks' does not explicitly cite the theory of imperfect competition or Mrs Robinson, preferring to talk of the Monopoly-Competition argument instead. Was this due to the lack of sentiment between the two great

Fair trade, reciprocity, and harmonization

23.

24.

25.

26.

27. 28.

29.

51

English economists? Or did Hicks harbor theoretical reservations about the foundation of the new theory, sensing the kinds of problems that later developments in the theory of industrial organization would begin to address meaningfully only in the latter 1970s and 1980s? In table 2.3,1 divide the theoretical arguments into two broad classes: those that pertain to free trade per se, reflecting market failures of one kind or another, and those that pertain to unilateralism versus reciprocity. This would not be the way the policy issue of reciprocity would arise for a weak country. Whereas the strong who feel weak can try to force others to reciprocate, the weak can only force reciprocity on themselves. I do not go into the theoretical and policy analysis of 301 actions of the US administration here, referring the reader instead to Bhagwati and Patrick (1990). A proper history would trace the beginning of this breakout to the early 1970s, in the US, in the wake of the dollar crisis that forced the US off gold convertibility, and on to the 1974 Trade Act. In practice, however, there are documented instances of apparently minuscule margins being countervailed. Proportionality of cause is distinctly less evident in practice as against proportionality of consequences. These would be "situation" comparisons of utility possibilities, of course, under each policy set. These issues are explored systematically, the GATT laws and rulings on the subject spelled out and their rationale explained, and alternatives to unilateralism are suggested in the forthcoming Annual Report of the GATT secretariat on environment and the trading system. The recent theoretical research of Robert Feenstra, John McMillan, Dan Coates, Tracy Lewis and others is already addressing this range of issues, while there is also a more institutional literature growing of the design of institutions that govern trade.

References Beveridge, William (ed.) (1931), Tariffs: The Case Examined, London, Parker & Co. Bhagwati, Jagdish (1977), "Harry G. Johnson," Journal of International Economics, 7(3), 221-30. (1988), Protectionism, Cambridge, Mass.: The MIT Press. (1989), "Is Free Trade Passe After All?" Weltwirtschaftliches Archiv, 125 (1), 17—44; reprinted in J. Bhagwati, Political Economy and International Economics, edited by Douglas Irwin, Cambridge, Mass.: The MIT Press (1991), chapter 1. (1991), The World Trading System at Risk, Princeton, New Jersey: Princeton University Press. Bhagwati, Jagdish and Douglas Irwin (1987), "The Return of the Reciprocitarians: US Trade Policy Today," The World Economy, 10(2), 109-30; reprinted in J. Bhagwati, Political Economy and International Economics, edited by Douglas Irwin, Cambridge, Mass.: The MIT Press (1991), chapter 3.

52

JAGDISH N. BHAGWATI

Bhagwati, Jagdish and Hugh Patrick (eds.) (1990), Aggressive Unilateralism: America's 301 Trade Policy and the World Trading System, Ann Arbor: Michigan University Press and Harvester Wheatsheaf. (1991), Aggressive Unilateralism, Ann Arbor: Michigan University Press. Bhagwati, Jagdish and T.N. Srinivasan (1969), "Optimal Intervention to Achieve Non-economic Objectives," Review of Economic Studies, 16, 27-38. Chamberlin, Edward (1933), The Theory of Monopolistic Competition, Cambridge, Mass.: Harvard University Press. Corden, W.M. (1957), "The Calculation of the Cost of Protection," Economic Record, 33, April. Eaton, Jonathan and Gene Grossman (1986), "Optimal Trade and Industrial Policy under Oligopoly," Quarterly Journal of Economics, 2, 383-406. Eichengreen, Barry, 1984, "Keynes and Protection," Journal of Economic Theory, 44, 363-73. Hicks, John R. (1959), "Free Trade and Modern Economics," in J.R. Hicks, Essays in World Economics, Oxford: Clarendon Press, reprinted in Hicks (1959). (1959), Essays in World Economics, Oxford: Clarendon Press. Irwin, Douglas (1987), "Welfare effects of British Free Trade: Debate and Evidence from the 1840s," presented to Mid-West International Economics Meetings, Ann Arbor. (1991a), "Challenges to Free Trade," Journal of Economic Perspectives, 5(2), 201-8. (1991b), "Keynes and the Macroeconomics of Protection," mimeo, Chicago University Business School. Johnson, H.G. (1965), "Optimal Trade Intervention in the Presence of Domestic Distortions," in R.E. Caves, P.B. Kenen, and H.G. Johnson (eds.), Trade, Growth and the Balance of Payments, North Holland, pp. 3-34. Kaldor, Nicholas (1950-1), "Employment Policies and the Problem of International Imbalance," Review of Economic Studies, 19, 42-9. Kemp, M.C. (1962), "The Gain from International Trade," Economic Journal, 72(288), December, 303-19. Krugman, Paul (1984), "Import Protection as Export Promotion," in H. Kierzkowski (ed.), Monopolistic Competition and International Trade, Oxford: Basil Blackwell. Krugman, Paul (ed.) (1986), Strategic Trade Policy and the New International Economics, Cambridge, Mass.: The MIT Press. McCloskey, Donald N. (1980), "Magnanimous Albion: Free Trade and British National Income, 1841—1881," Explorations in Economic History, 17, 303—20. Mill, John Stuart (1848), Principles of Political Economy, London, Parker & Co. Robinson, Joan (1931), The Economics of Imperfect Competition, London: MacMillan. (1937), "Beggar-my-Neighbour Remedies for Unemployment," in Essays in the Theory of Unemployment, New York: MacMillan. Samuelson, P. A. (1939), "The Gains from International Trade," Canadian Journal of Economics and Political Science, 5, 195—205.

(1962), "The Gains from International Trade Once Again," Economic Journal, 72, 820-9.

Fair tradey reciprocityy and harmonization

53

Smith, Adam (1776), The Wealth of Nations, New York: Cannan Edition, Modern Library, 1937. Wolf, Bernard and Nicholas Smook (1988), "Keynes and the Question of Tariffs/' in O.F. Hamouda and J.N. Smithin (eds.), Keynes and Public Policy After Fifty Years, Vol. II, New York: New York University Press.

CHAPTER

3

The revival of protectionism in developed countries W.

MAX

CORDEN

There is evidence of the revival of protectionist attitudes in developed countries, notably the United States and Western Europe. This chapter reviews some aspects of this protectionist renaissance, focusing particularly on the relationship between protection and macroeconomic events and policies. Protection and macroeconomic policies Recessions bring protection From 1980 to 1982 the developed world passed through a major recession created essentially by tight monetary policies designed to squeeze inflation out of the system. During this period protectionist pressures increased, and there were also some increases in actual protection. The issue arose again in the United States in 1991; as a result of the recession of that year, there was a powerful revival of protectionist attitudes directed particularly against Japan. This experience raises the important issue of the connection between two sets of government policies - protection policies and macroeconomic policies. One has to consider the case where macroeconomic policies may have induced a recession or may have failed to prevent a recession caused by other factors and where the recession in turn has induced pressures to increase protection. Policies of monetary tightness squeeze profitability and reduce employment, one aim - perhaps the primary one - being to moderate wage increases. If the moderation in wages anticipated the monetary squeeze, or 54

The revival of protectionism

55

at least followed it very closely, then profitability and employment would not need to fall, and the desired decline in the rate of inflation could be brought about without cost. But we know that profits do get squeezed and employment does fall. The declines in the rates of inflation in the United States, Japan, Germany, and Britain were brought about at considerable cost, especially in Britain. Additionally, if a country's monetary contraction is significantly greater than that of other countries (so that its real interest rate rises relatively), its real exchange rate is likely to appreciate and the decline in profitability and employment will be greater, the adverse effects being focused on the export and import-competing sectors of the economy. In 1981 and 1982, this was the experience both in the United States and in Britain. Pressures for protection, direct and indirect, are then inevitable. The protection may take the form not of tariffs, quotas, or voluntary export restraints (VERs), but of subsidization of private industries or the covering of losses of publicly owned industries. Recessions, whether policy-induced or not, always give rise to increased pressures for protection. This experience suggests that contradictions in government policy can arise here. The central bank and Ministry of Finance bring about a profitability squeeze which the Ministry of Industry - pushing for protection - seeks vainly to reverse. If the latter were successful, one would have to ask whether the original macroeconomic policy was justified: the profitability squeeze was meant to moderate wage increases, and these will not be moderated if the squeeze is avoided by protection. But in fact, in an overall sense, protection cannot undo what macroeconomic policy has created. It can only reshuffle the consequences. For example, with given fiscal and monetary policies, subsidies to particular industries must be balanced by fewer funds elsewhere. If taxes are raised, this may lead to increased wage demands, again offsetting the initial effects. Furthermore, protection to help some import-competing industries which have suffered from appreciation will lead to greater appreciation than otherwise, and so increase the adverse effects on other import-competing as well as export industries. These considerations do not rule out a role for microeconomic policy, including specific subsidies or even tariffs, but they do raise a question about protectionist policies designed to negate or soften the effects of macroeconomic policies. A recession - whether policy-induced or otherwise - may lead to increases in protection in spite of the implicit contradiction just discussed. The danger is then that protection will not be reduced once the economy recovers, since reducing protection may require a much more prolonged period of prosperity. Long-term costs are then imposed in the form of

56

W. MAX CORDEN

adverse effects on resource allocation and on the degree of competition. If protectionist pressures at a time of recession cannot be resisted, such adverse long-term effects must be taken into account when framing short-term macroeconomic policies. A temporary recession has, then, adverse long-term effects not only through the fall in investment that it brings about and through the loss of work experience of the temporarily unemployed but also through the recession-protection ratchet effect: a recession increases protection, but the following boom does not lower it to* the same extent. The association of increased protection with recessions or depression has also led some people to believe that protection can actually cause recessions. The association of the two in the 1930s helps to explain the free-trade movement in developed countries in the 1950s and 1960s. Others again think that protection can moderate a recession because of the favorable effects on employment in particular industries. In fact, it is not obvious that protection can either contribute to or moderate a recession. It can, of course, raise the inducement to invest in particular industries, but one must balance against this the indirect opposite effects in other industries. Protection and growth While the relationship between protection and recession goes only one way — recessions tending to increase protection but protection not necessarily moderating or intensifying recessions - the relationship between low long-term growth and protection is clearly two way. Widespread protection is likely to lower the long-term growth rate, especially if it is designed to protect losers from change rather than just consisting of a system of fixed tariffs or subsidies. If an industry that is having difficulty in coping with foreign competition can expect to obtain higher protection as a result, the incentive to improve its efficiency, or to transfer resources elsewhere, will be reduced. Protection reduces the flexibility of the economy, and hence the productivity of capital and labor. At the same time, low growth - and especially a shift of gear to a lower growth rate - means that adjustment between industries is difficult: relative losers may also have to be absolute losers, since fewer alternative job opportunities become available and since fewer resources are available to compensate losers. People become more security-minded, protection being part of an implicit "industry insurance system." Hence, it has been much easier to reduce protection at times of high growth. The relationship between the degree of protection and the rate of growth is not always clear-cut, one reason being that the degree of overall protection is difficult to measure satisfactorily. A very large economy which

The revival of protectionism

57

is in itself a free-trade area will lose less from some protection against imports from outside than a small economy. There is little doubt that the great economic success of the United States over a long period can be partially explained by the fact that it has been a very large area of free trade. The trend of its protection against trade with the outside world is probably less relevant in explaining its growth. For the same reason, some part of the high post-war growth in Europe can surely be explained by the freeing of intra-European trade and, especially, the establishment of the common market. The great economic success stories of the post-war period - Japan, Korea and Taiwan - have not been free-trade countries, and at some stage all practiced infant industry protection. But they have been outward-looking countries, developing during their crucial high-growth stages by export expansion. There has been little, if any, bias against exports in favor of import-competing industries, apart from agriculture. In this respect these countries have differed greatly from India and Latin American countries. It is also worth noting that Brazil's highest growth period, 1968-73, was the period when protection was reduced (though, by no means, eliminated), and when exports were fostered. In developing countries shortages of imports needed as inputs into domestic production have, at times of balance-of-payments difficulties, been an important explanation of low growth. Export expansion fostered growth by making it less necessary to restrict imports that were essential for domestic production. Protection and the exchange rate In the days of fixed exchange rates it was often said that a move to floating rates would obviate the need for tariffs and quotas to deal with balance-of-payments deficits. In the immediate aftermath of the first oil shock, it was usual to congratulate OECD countries for having avoided a revival of protectionism, the credit being given (at least to some extent) to exchange-rate flexibility. If one thinks of devaluation as a policy instrument that switches demand from foreign to home goods and, within the latter, from tradables to non-tradables, and switches output from supplying the domestic market to exports, then tariffs, tighter import quotas, export subsidies, and so on, are substitutes for devaluation. The standard argument is that they are inferior substitutes because they create distortions within the tradable goods (import-competing plus exporting) sector. Tariffs and quotas favor only import substitution relative to non-tradables, while devaluation or depreciation makes both import substitution and exporting more profitable relative to non-tradables, so that they are not anti-trade biased. Further-

58

W. MAX CORDEN

more, devaluation (or depreciation in a floating rate regime) has a uniform effect on the import side - being like a uniform ad valorem tariff - and so avoids the distortions between different import-substituting activities which result from differential tariffs or from any system of import quotas. It is thus an important argument in favor of exchange-rate flexibility that it is likely to reduce or eliminate the need to use tariffs, quotas, and similar devices for balance-of-payments reasons. Comparing two ways of improving the balance of payments (while simultaneously maintaining the level of demand for domestically produced goods and services in total) - the way of devaluation and the way of tariffs or quotas - the former has a more favorable resource allocation effect. The connection between protection, trade liberalization, and the exchange rate has been crucial for developing countries. In the 1960s and 1970s, many developing countries were very reluctant to devalue at a time of balance-of-payments crisis. The normal response was to tighten up imports restrictions - and these were not always relaxed when the crisis came to an end, so that, in many cases, a legacy of high import restrictions remained. Familiar distortions resulted. Not only did such policies introduce or strengthen a bias in favor of import-replacement against exporting, but they also distorted the pattern of imports and domestic import-competing production and, through the use of licensing as the main method of import restriction, led to rent-seeking and corruption. In the 1980s devaluations or frequent exchange-rate adjustments (crawling peg exchange rates) were more widely used, thus avoiding the tightening up of restrictions. This was partly the result of pressure from the International Monetary Fund and the World Bank. Even more important, several developing countries (notably Turkey, Morocco, and Mexico) liberalized imports, a process that had to be associated with devaluations of the exchange rates. It would certainly have been more difficult to engage in trade liberalization if the countries had been committed to fixed exchange rates or had been reluctant to devalue. It must be stressed that, for a given level of utilization of domestic labor and capacity, any improvement in the current account of the balance of payments requires a fall in aggregate expenditure, since the initial excess of expenditure over output has to be reduced if the current account is to improve. It is widely understood that this is so in the case of devaluation, but it is also true if tariffs or import quotas were, instead, to be used. Thus, trade restrictions are not painless ways of improving the balance of payments. We have considered here the case where a country has to improve its competitiveness because the balance of payments needs to improve. The argument also applies to the case where a change in the balance of payments

The revival of protectionism

59

is not needed but where a country is losing competitiveness owing to its inflation rate exceeding that of its trading partners. Exchange-rate depreciation can then restore competitiveness. If such exchange-rate adjustment were not possible, the resultant losses in profitability and employment in the export and import-competing industries would inevitably generate protectionist pressures. Adverse effects of exchange-rate fluctuations This standard approach views the exchange rate as a policy instrument. It implies that the exchange rate is neither fixed nor freely floating, but rather is managed so as to attain desired competitiveness or balance-of-payments outcomes. The rate could be pegged in the short run, with occasional or frequent policy decisions that alter the peg. A system of managed floating, with frequent interventions in the foreign exchange market by central banks, could also have the same result. But we must now consider a somewhat different situation, one where the exchange rate floats - possibly with some "leaning against the wind" intervention - and fluctuates over the medium term because of varying pressures originating in the capital market. There will then be changes, possibly very large ones, in competitiveness. This might be called the US problem - on which a number of economists, notably Bergsten and Williamson (1983), have focused. Whenever the United States loses competitiveness, pressures for US protectionism appear to intensify. The yen-dollar rate is particularly important in this respect. There have been three periods when the dollar was overvalued relative to the yen in terms of purchasing power parity (meaning some average longer-term real exchange rate). The overvaluation in the late 1960s and early 1970s gave rise to the Mills and Burke-Hartke bills, import controls on steel, and finally, in 1971, the import surcharge. The Burke-Hartke bill, if enacted, would have imposed strict quantitative limits on the levels and rates of growth of all imports into the United States. The overvaluation of the 1976-7 period led to major trade conflict between the United States and Japan. Finally, the overvaluation of 1981-5 led to voluntary export restraints on Japanese cars and numerous protectionist proposals in Congress. It is implied in this view that fluctuations in exchange rates originating in the capital market, especially in the yen-dollar rate, always generate pressures for increased protection in the United States when the dollar is in its real appreciation phase - when US competitiveness has declined - but that this is not offset by reductions in protection when the dollar is in its real depreciation phase. Thus, an asymmetry or ratchet effect is implied. When times are bad for US import-competing industries, they succeed in getting

60

W.

MAX

CORDEN

more protection, but when times become good, protection is not dismantled. There is possibly some tendency toward such a ratchet effect, though it has to be borne in mind that "protectionist pressures" do not always lead to actual increases in protection. Furthermore, one must also look at the Japanese side of the coin. In the 1980s there have been some reductions of protection in Japan (how much being hard to assess), and it seems very plausible that this has been caused not only by political pressure from the United States but also by the improved competitiveness of Japanese industries owing to yen depreciation. Medium-term fluctuations in exchange rates in response to forces originating in the capital market are not necessarily inappropriate. A country's exchange rate may appreciate because there is a transfer of long-term capital into the country based on correct expectations of favorable investment opportunities relative to other countries. Thus, for a time, the United States was seen by investors as a "safe haven." The exchange rate may also appreciate because of short-term capital inflows (or incipient inflows) that reflect particular expectations of prospective monetary and fiscal policies, expectations that may be perfectly rational, given available information. If the resultant loss in competitiveness leads to protectionist pressures, the question arises to what extent the exchange rate itself should be altered through exchange market intervention - which may be difficult, in any case, if market expectations are very firm - and to what extent effort should be put, rather, into resisting the protectionist pressures, while allowing the exchange rate to fluctuate. If the market expectations turn out to be justified, some reallocation of resources induced by the exchange-rate signals will also be justified. But this does not rule out an argument in favor of the monetary authorities forming a medium-term view about an exchange rate and, if this view differs from the market, sometimes cautiously acting on it. The US budget deficit and the decline in US competitiveness In the case of the United States there has been a connection between the budget deficit, the current-account deficit and the real exchange rate - and hence between the budget deficit and the apparent decline for some years in US competitiveness. The basic relationships are as follows. The budget deficit is financed by borrowing on the open capital market, and this tends to raise interest rates, draw capital into the United States and, thus, bring about appreciation of the exchange rate. The domestic expansionary effect of the budget would have an opposite effect on the exchange rate: it would lead to higher imports which, in itself, would tend to depreciate the exchange rate. But it is clear

The revival of protectionism

61

that in the US case the capital market effect on the exchange rate was stronger than the effect of the current-account deterioration. The net result was that indirectly the budget deficit was financed by foreign savings. The budget deficit led to the current-account deficit, which was financed by foreign capital inflow. Of course, the dollar also appreciated for other reasons, notably US monetary contraction which, for some time, was greater than that of the other major OECD countries. The private sector net financial surplus (excess of savings over investment) of the United States was insufficient to finance the large budget deficit. If there had been no international capital mobility, this surplus would have had to finance the budget deficit. This would have been brought about by real interest rates in the United States rising until sufficient investment was crowded out for the private surplus to reach the level of the budget deficit. There would have been no current-account deficit and hence no need for real appreciation. With an open capital market it was inevitable - and desirable from the US point of view - that the higher interest rates drew in capital from abroad. Inevitably the United States had to finance some of its budget deficit from foreign savings. If it had not done so and the structural budget deficit had attained the sorts of levels which it did, a significant decline in private investment relative to its normal recovery level in the United States would have been required. The United States had to run a current-account deficit, this being the way in which foreign savings became absorbed into the United States.1 Real appreciation is part of the mechanism by which a current-account deficit is brought about. The United States had to lose competitiveness if it was to generate the current-account deficit which was the counterpart of the capital inflow. There had to be some shift of resources, at least at the margin, away from export and import-competing industries toward non-tradables. Clearly there were gainers in the United States - above all, the industries stimulated by the extra defense spending and those benefiting from the extra consumer spending resulting from the tax cuts - but there were also losers, namely, employees and the owners of capital in the tradable goods industries, especially the more marginal ones. In addition, of course, there were gainers and losers from higher interest rates. Inevitably losers sought protection, irrespective of gainers elsewhere. It was never widely understood that the decline in US competitiveness was a by-product of expansionary fiscal policy - an inevitable one - and did not reflect any particular inadequacies of US industries. Such inadequacies surely existed, although they did not have to lead to a general loss of competitiveness, but only to a loss of competitiveness on the part of particular industries, offset - through the mechanism of exchange-rate

62

W. MAX CORDEN

adjustment - by a gain in competitiveness of other industries. The politicians whose taxation and spending policies generated the deficit should have gone to some trouble to explain that an overall loss of US competitiveness was necessary. In doing so, they might have been able to moderate some of the protectionist pressures. The following generalization might be added. If some increase in protection is a predictable by-product of real appreciation, while real appreciation in turn is a by-product of an increased fiscal deficit, then there is a reason to reduce the fiscal deficit additional to the usual reasons. It is usually argued that a large deficit would tend to crowd out private investment, might lead to the incurring of large interest commitments eventually payable by the US taxpayer, and would have adverse effects on developing-country debtors and new borrowers. To this is now added the resource misallocation cost resulting from extra protection. An increase in US protection could not alter the average effect on US import-competing and export industries of the fiscal deficit. It could only shelter particular industries or sectors at the expense of others. If an increase in particular imports is prevented by protection, other imports will have to increase even more, and so other import-competing industries will be even more adversely affected and the decline in exports will have to be greater. The key point is that the higher the level of protection for particular sectors, the greater the real appreciation of the dollar has to be to yield the required current-account deficit. The cost of protection to the United States results from the distortions set up within the US tradables sector and in distorting relative tradables prices facing US purchasers. Foreign suppliers of protected products will certainly lose, but foreign suppliers of other products will actually gain, benefiting from an even higher real appreciation. The continual real appreciation of the dollar actually came to an end in 1985. The dollar started depreciating relative to the major currencies, notably the yen and the DM, from 1985 on, and by 1990 it was (as measured by the International Monetary Fund) actually below 1980. Thus, one would expect the pressure for protection in the United States to have eased, and it did a little . . . though it revived again in the 1991 recession. The Federal budget deficit as a proportion of GNP remained high during 1984-6 (averaging 4.8 percent) but declined from 1987, and by 1989 was 3 percent. The current-account deficit clearly responded - though with a lag to the depreciation of the real exchange rate. In addition, the decline in the budget deficit played a part. The depreciation itself can be explained both by a shift in expectations and, later, by the decline in the budget deficit. Of course, relative monetary policies of the United States and the other major economies also played a role. The current-account deficit as a proportion of GNP reached its peak in

The revival of protectionism

63

1987, at 3.5 percent, and by 1989 was down to 2 percent. What is particularly striking is the growth in the volume of US exports that had taken place. In 1987 it was 14.5 percent and in 1988 22.5 percent, and over the whole period 1987—90, average annual export volume growth was 14 percent (compared with 4 percent for Japan). Thus, US competitiveness has certainly improved, whether measured by the movement in the real exchange rate or the actual trend in export volume. OECD protection and the developing countries Protection and the debt problem The effects of OECD protection on the developing countries are crucial. To start with, the relationship between OECD protection and the developing countries' debt will be discussed. It seems obvious that if the indebted developing countries are to meet their interest obligations and eventually to repay at least some of their debt, they must be allowed to increase their exports. It also seems obvious that it is in the interests of the developed countries, and especially their financial system, that the debt issue be resolved without open or implicit default. It might be argued that improvements in a current account can be brought about as much by reductions in imports as in increases in exports. Clearly the indebted developing countries must operate on both fronts. But if OECD countries were unwilling to accept substantial extra imports of manufactures from developing countries, then the latter would be forced to bring about the necessary balance-of-payments improvements mainly by reducing their own imports from OECD countries. This would mean that the necessary improvements would be brought about at greater cost to the developing countries. They would be deprived of the potential benefits of further exploiting their comparative advantage in labor-intensive products. More specifically, there tends to be some minimum requirement of imported components and raw materials for domestic manufacturing production (at least in the short run), so cutting imports beyond a point is likely to lead to increased unemployment. Furthermore, reduced imports would raise the cost of living, so tend to raise nominal wages to compensate, and hence lead to reduced employment in the manufacturing industry owing to higher wage costs. It follows that the indebted countries have to increase their exports - and to be allowed to do so by OECD countries - if they are to bring about the necessary balance-of-payments improvements at tolerable cost. How important has this issue been for the principal debtors? In the case of a number of products exported by them, OECD protection must have

64

W. MAX CORDEN

reduced their income substantially. The Community's Common Agricultural Policy has had an adverse effect on Argentinean agricultural exports. Some of the debtor countries are substantial exporters of textiles and clothing. The Republic of Korea is a debtor country (which has been repaying its debt) and obstacles have been placed by the United States on its exports of textiles, clothing, and color television sets. In addition to these products, Community countries have restricted imports from Korea of steel products and (in the case of France at least) many other goods. Its footwear exports generally meet trade obstacles in the Community on a whole range of primary product exports. Japan has quotas on footwear imports. These are just examples. Perhaps more important in the case of several of the countries are the actual and expected obstacles against products which would form the basis of a feasible export expansion program. The potential for expanding exports of labor-intensive goods of various kinds would seem to be very large. I shall return to this point below. The effects of OECD protection on the developing countries Until the 1980s one could say clearly that OECD protection in the manufacturing sector grossly discriminated against developing countries, especially the more successful exporters among them. This was in spite of the Generalized System of Preferences, which had a modest effect compared with the various and complex restrictions imposed under the umbrella of the Multi-Fibre Arrangement (MFA). It really seems outrageous that over a long period of time severe limits have been placed on imports of textiles and clothing from developing countries. It appears that these restrictions, if anything, have been strengthened since the renewal of the MFA in 1981. This arrangement provides a framework for bilateral agreements (utterly contrary to the most-favored-nation GATT principle) and is the basis for numerous VERs. At the end of 1982 the United States had bilateral agreements with twenty-two developing countries, mostly embracing all textiles and textile products (clothing), and with the Community, twenty-nine. The 1981 MFA renewal endorsed continued quantitative restrictions arranged bilaterally, especially against the most successful suppliers. Furthermore, the complexities of the arrangements are likely to inhibit exporters, especially the less experienced ones. Textiles and clothing are by no means the only products from developing countries that have suffered from restrictions, but it is here that the comparative advantage of developing countries (as well as some of the ex-socialist countries of Eastern Europe) is particularly strong. The Common

The revival of protectionism

65

Agricultural Policy also has an adverse effect on many developing countries which are exporters of wheat, beef, rice, and sugar; Poland's exports are held back by this policy. It is worth noting that some developing countries which are importers of products the Europeans export at heavily subsidized prices are gainers. This would include particularly major grain importers. It cannot be said that the Common Agricultural Policy clearly discriminates against developing countries since it has a significant adverse effect on temperatezone exporters such as Australia, Canada, and New Zealand (though Argentina is also in that group). In addition, because of restrictions imposed primarily on Japanese exports in recent years, one cannot say for sure that overall protection by the United States and the Community discriminates against developing countries. But it does discriminate against particular labor-intensive products, and, in any case, in some fields at least, it puts a severe limit on export expansion by developing countries. The importance of these protectionist policies directed against exports from developing countries can perhaps be overstated.2 After all, manufactured exports from developing countries to OECD countries have continued to increase and rose steadily, even relative to world trade, over the whole post-war period. It is worth noting that in 1970 the share of developing-country imports in the total (apparent) consumption of manufactured goods in industrial countries was 1.7 percent, and by 1980 it had risen to 3.4 percent. In the United States it rose from 1.3 percent to 2.9 percent. Over the period 1983 to 1989, the volume of exports from all those developing countries that were predominantly exporters of manufactures rose by 8.5 percent per annum, a very impressive result. The growth rates of exports from Korea, Turkey, Malaysia, Thailand, and the People's Republic of China were particularly high. So protection may have slowed the growth of exports somewhat but did not halt it. Even exports of textiles and clothing have steadily increased. Taking a broad view, the severest restrictions have clearly been applied in the field of clothing and textiles. Even here some suppliers have managed to cope with the restrictions by upgrading their products: when restrictions are on a quantitative basis it pays to export higher-value products. This, of course, is not necessarily an economically sound adjustment for a country that may have a comparative advantage in low-value, low-quality products. In addition many developing countries have widely diversified their exports, moving into areas where there have been rather fewer or looser restrictions: machinery and non-metallic mineral products (including china and glassware), and miscellaneous goods such as sports equipment, toys, and musical instruments. Furthermore, extensive import controls have been imposed on products from the Far Eastern exporters; nevertheless, they have increased their shares of the OECD market. Restrictions on

66

W. MAX CORDEN

footwear imports into most OECD countries have not been as tight as for clothing and textiles, although Japan has strict quotas, the Community has negotiated VERs with Korea and Taiwan, and tariffs are still moderately high (averaging 13 percent in the major developed countries). On the other hand, the possibilities of export expansion are very great. There are widespread restrictions on imports of consumer electronics from developing countries, notably Korea, and yet these must be regarded as having a major potential for growth. Developing countries clearly have a comparative advantage in textiles, clothing, and footwear and yet still have only a small part of the market of OECD countries. (In 1976 the developing countries had about 2 percent of the total US market for textiles and 10 percent of the market for clothing, though, of course, they had a much larger share of imports. In 1980 the share in the market of a group of eleven industrial countries as a whole - including the main ones - was 5.4 percent for textiles and just over 16 percent both for clothing and for footwear.) Not only are there possibilities of further expansion from the major existing exporters but there are also many potential exporters around the world who could, after an initial infant-industry period, become suppliers of such goods to the industrial countries without subsidization. But there is always the danger that they will be discouraged by the prospects of restrictions as soon as they manage to break into a market. Protection by developing countries Perhaps the worst aspect of the continuance of such protection and the revival of protectionist sentiments in the developed countries is that it could lead to a revival of export pessimism in developing countries, leading again to inward-looking policies. In the 1950s export pessimism was fashionable, especially in India and Latin America, and, arguably, led to severely growth-inhibiting importsubstitution policies. For a long time it was widely believed that there was little hope of breaking into the markets for manufactured goods of the developed countries so that, if developing countries were to build up their manufacturing industries, they would have to do so by replacing imports. On the basis of such beliefs some highly uneconomic industries were built up in many countries, for example India and Brazil. Partly as a result of the successes of the new industrial countries, as well as of academic and World Bank research and writing, there has been a gradual shift of opinion, leading to significant moves to trade liberalization in many developing countries, the outstanding example being Mexico. Through hard experience the disadvantages of import-substitution policies have been realized. The experiences of Korea and Taiwan seemed to justify export optimism. In fact, in major

The revival of protectionism

67

countries, notably India and Brazil, the shift of policies has not been so great. But there is always a danger that this tendency will be reversed. There are really two separate issues here. One is whether protection in developed countries justifies protection by developing countries. It might be noted here that most developing countries have much more all-embracing systems of protection — and, on the average, at much higher rates — than OECD countries have. If the OECD rates of protection were given, there would be no case for developing countries keeping their protection just because of OECD protection. This is a well-known proposition of trade theory: one country's protection does not justify another's. Protection by a trading partner lowers the real income both of the partner and at home, and protection at home would add to the income loss both at home and in the partner country. The matter would be different if OECD protection consisted of fixing the quantity of (rather than rates of protection on) imports from developing countries. It would then pay the latter to restrict their exports so as to obtain the highest possible prices, given the fixed quantity they can export. An indirect form of export restriction is through quotas and tariffs on imports (with the exchange rate adjusting to bring exports down to the reduced level of imports). But, of course, such an extreme situation does not really exist. Thus there is still a case for these countries in general to reduce their trade restrictions. The second issue is crucial. In practice, it has been quite difficult to persuade policy-makers and the general public in developing countries to liberalize their trading arrangements. The protectionist arguments used have been prevalent in developed countries as well, and are very similar. They are the types of arguments that have been analyzed by economists for over two hundred years and generally have been found wanting. If there is a revival of protectionist attitudes in developed countries, if it seems that OECD protection has greatly increased (though it may not actually have done so), and if it is possible to obtain ready-made arguments from intellectuals and policy-makers in developed countries in favor of protection, then it is quite possible that the liberalization trend in the developing world would be reversed. Increased protection by developing countries themselves - or even a failure to continue liberalization trends - may do them more harm than have the possibly modest increases in protection in Europe and the United States, but the latter may encourage the former. Some arguments for protection It is worth looking briefly at some arguments for protection currently popular in the United States and Europe. 3 The main argument

68

W. MAX CORDEN

from the point of view of politicians is presumably that protection may raise the incomes of particular groups in their countries or - more important may avoid falls in incomes that might otherwise take place, and that these groups have political influence. But a variety of arguments is used to show that either fairness or the national interest justifies such protection. With regard to the national interest, the implication is that there is some national gain - either that there will be no significant losers from protection but only gainers or that over time losers could be compensated and a net gain would remain. In fact, no new arguments for protection have been developed in recent years. Indeed, some of the currently popular arguments were advanced in the nineteenth century in the United States, notably the "pauper labor" argument. One of the more acceptable arguments in current conditions is for the imposition of temporary restrictions to modify a sudden import surge that might impose serious and unexpected injury on some domestic industry. This will be referred to later in connection with a possible safeguard code and GATT Article XIX. Here let us look briefly at the employment argument, the pauper labor argument, the fairness argument, and the dumping argument. Subsequently I shall consider the supposed "problem" of Japan. The employment argument Protection of an industry may contribute to preserving employment in that industry. At the same time there is evidence that the actual declines in employment in particular industries (such as textiles, clothing, automobiles, and steel) that have taken place in the United States and Western Europe have not been primarily caused by increases in import shares. In the United States in particular, declines in rates of change in overall demand for various industries' products as well as labor productivity growth have been quantitatively much larger in their impact on employment (as shown in Krueger, 1980). Nevertheless, if import competition is significant and the volume of imports is large, common sense suggests sufficient protection could offset the adverse employment effects of these other factors. The weakness of the employment argument is that it is narrowly partial equilibrium, focusing only on particular industries. One should note the adverse effects, for example, of protection of steel on steel-using industries. Higher steel costs reduce effective protection (protection related to value added) for the automobile industry and will tend to reduce employment there, and this could more than offset the gains in employment in the steel industry. More important are the general equilibrium effects operating

The revival of protectionism

69

through effects on wage levels and the exchange rate. Here let us note the oft neglected exchange-rate aspect. If protection of some industries is increased and yet the balance of payments is not to change, then - given the average level of nominal wages as well as foreign costs - the exchange rate has to appreciate. And this will have an adverse effect on employment in those tradable-goods industries where protection has not increased. To consider the case where the general level of protection for import-competing industries is reduced, such reduction would have to be associated with depreciation of the exchange rate and thus would have beneficial effects on employment in export industries. The possibility of transitional unemployment effects cannot be denied: the demand for labor in one set of industries may fall while that in others increases, but labor may not readily move, and in the short run the former industries may shed labor before the latter absorb extra labor. But such a problem suggests simply the need for temporary measures to foster adjustment, not protection designed to prevent change. In addition there may be some general real wage rigidity. If initially there is unemployment because real wages are too high, then, if protected industries are labor-intensive relative to non-protected industries, an increase in protection would raise overall employment. The gain in employment in the protected industries would more than offset the losses in the non-protected industries. But here it must be noted that United States exports, not import-competing industries, tend to be relatively laborintensive (though there are measurement problems which require this to be stated with caution) so that, given the real-wage rigidity assumption, a general rise in tariffs or imposition of quotas, as proposed for example in the famous Burke-Hartke bill, would actually lower US employment overall. In any case, the real-wage rigidity model is too simple. If the nation as a whole gains from a particular measure (that is, non-labor and labor incomes combined rise), there is scope through the tax system for maintaining after-tax real wages while lowering pre-tax real wages to the levels required for full employment. The pauper labor argument The pauper labor argument is that protection should be imposed on goods originating from countries where wages are low. In fact, in many developing countries labor costs per unit of output are not relatively low, since the benefits of low wage costs are offset by low labor productivity. But let us suppose that labor costs per unit are low for particular products, presumably labor-intensive ones. This simply means that these countries have a comparative advantage in such products. There must be other

70

W.

MAX

CORDEN

products where costs are relatively high, for otherwise they would be only exporting, not importing. If they were only exporting, then presumably they would need to alter their exchange rates unless they were primarily interested in importing bonds - that is, lending money - rather than importing goods and services. Another version of this argument is that it is "immoral" to buy goods produced by cheap or "sweated" labor. On the basis of such an argument some labor unions have advocated restricting imports from Hong Kong, for example. Yet if the demand by the United States for Hong Kong products fell, Hong Kong producers would either have to lower their prices - and hence in due course the wages they paid - in order to unload more exports on other markets or unemployment in Hong Kong would increase. Fairness argument To industrialists faced with competition from imports produced in more favorable conditions than at home, such competition seems "unfair." Thus one gets the view that conditions - wages and other factors affecting the cost of labor - should be equalized around the world. In the extreme this could be interpreted to mean that protection should offset all comparative advantage differences, so that all trade would cease. More plausibly it may be meant that the flow of trade should depend purely on differences in managerial efficiency and entrepreneurial skill, all other factors being offset by protection. In any case, the appropriate analysis is the same as in the case of the pauper labor argument. But there is one complication. Suppose foreign governments subsidize certain industries or the exports of certain products. They may provide indirect assistance, as Britain used to do when it covered the losses of publicly owned industries, or the United States does through its defense spending or its space program. One country may pump funds into research and development, another into its educational system, a third into its agricultural sector, and a fourth into reviving its steel industry. Does this mean that it is in other countries' interests to engage in countervailing protection? If a country is concerned only with maximizing its national income and neither with sectional interest nor with fairness, such protection is not to its advantage. The various interventions by country A should be taken as given by country B when the latter formulates its optimal policy, at least unless it can induce country A to change its policies in more favorable directions. But some intervention policies by country A may actually be favorable for A's trading partners, for example, policies that lower the prices of A's exports relative to its imports. In the case of unfavorable

The revival of protectionism

71

policies, such as tariffs that restrict the flow of trade, if B responds by imposing tariffs of its own, the cost of protection will be increased. All this must be qualified for the case where the foreign interventions are expected to be temporary, in which case they can either be ignored or some temporary offsetting measure might be applied to avoid short-term dislocations. Dumping In the 1980s, anti-dumping measures have become very common instruments of protection. The principal users are the United States, the European Community, Canada, and Australia. These measures are subject to the GATT Anti-Dumping Code, which was amended in 1980 and which is actually quite permissive about such selective, discriminatory measures (Stegemann, 1991). Dumping generally means that a country exports its products at prices lower than it sells them at home. Naturally the competing industries abroad will seek to get the country to raise its export prices or will try to persuade their own governments to impose countervailing duties. Yet, there is no logical argument for such reactions. It should not really be of interest to either the import-competing producers of a country or its purchasers of imports at what prices these goods are sold to purchasers in the foreign country. In general a nation benefits when its imports become cheaper relative to its exports, even though import-competing producers may lose. If there is a sudden fall in the price or a surge of imports, there may be some argument for temporary protection, but this has little or nothing to do with whether the price charged to consumers in the supplying country is above or below the import price. The "Japan problem" There has been great concern in Europe and the United States about the growth of Japanese exports and, even more, there has been fear even paranoia - about their prospective growth. A good deal of the revival of protectionism has been directed against Japan, notably in the case of motorcars, steel, and consumer electronics. Partly this appears to be connected with trade and current-account balances. Japan has had large trade surpluses with the United States. This is reduced but not eliminated when services are allowed for; that is, when the focus is on the current rather than the trade account. But world trade is not meant to be balanced bilaterally, so clearly one must at the minimum look at Japan's overall current account.

72

W. MAX CORDEN

In the twelve years from 1970 to 1981 Japan ran current-account deficits in four years, balance in one, and surpluses in the other seven. The two big surplus periods were 1971-2 and 1977-8. But the surpluses were low as proportions of Japanese GNP - the maximum was 2.5 percent in 1971, and in 1977 and 1978 they were only 1.6 percent and 1.7 percent respectively. But in the five years 1984 to 1988, the surplus averaged 3.4 percent of GNP. In general, changes in the Japanese overall current account have been closely correlated with bilateral current-account movements with the United States, and the explanations can be largely found in the relationships between Japanese and US macroeconomic policies. Essentially there have been two explanations: divergences in overall macroeconomic demand levels or pressures (in 1977-8, the United States followed expansionary and Japan relatively contractionary policies) and divergences in the monetaryfiscal policy mix, as recently. Apart from these cyclical movements, there does appear to be some general tendency for the Japanese current account to be in surplus, a surplus substantially larger than required to maintain constant the real value of Japan's foreign financial assets. These current-account surpluses are not necessarily matters for concern. The Japanese household sector has a very large savings ratio by world standards, and even though most of these savings are absorbed by Japanese private investment and by the Japanese budget deficits, there is on average still something left over for the rest of the world, so that Japan as a nation tends to be a net buyer of financial assets in exchange for goods. This is not necessarily a bad thing. It tends to lower world interest rates. It is the obverse of the situation of most developing countries and of the United States situation. In fact, as the United States became a large net borrower on the world capital market as the result of its budget deficit, there had to be net lenders - countries that ran current-account surpluses - and Japan was the premier candidate for this role, a role in which it was practiced and which, for some time, its cautious citizens saving massively for their old age have been eager to fulfill. There are surely gains from international trade in financial assets against goods, as there are in goods-goods trade. There are, furthermore, plenty of similar examples from the history of countries that ran surpluses over long periods, notably Britain in the nineteenth century. There seem to be two problems connected with Japan's impact on the world. The first is that a large, high-growth, high-productivity country, by the very fact of its importance and of the changes it generates in the rest of the world, provokes fears. These fears were provoked by the United States in the 1950s. For those who do not understand the law of comparative advantage, there is the fear that this country is getting better at everything -

The revival of protectionism

73

and presumably will end up only exporting and not importing. There is also the feeling that, just because the country happens to have the world's highest productivity growth rate now, this must go on forever - even though Japan's growth rate has actually slackened dramatically since 1974. The second problem is the true problem. There are losers as well as gainers both from longer-term Japanese economic expansion (with a given current-account balance) and from periods (like 1971-2, 1977-8, and 1981-4) when there is a substantial Japanese current-account surplus. In the former, long-term growth case, the gainers are the consumers of Japan's export products and the suppliers of food and raw materials; while in the latter episodes the gainers include those consumers as well as borrowers who can draw indirectly on Japanese savings. But the focus, inevitably, is on the losers. They are the competitors with Japanese exports, above all. The problem is particularly acute when these exports expand suddenly. In Japan, unlike other countries, a deflation of domestic demand leads very quickly to a significant increase in exports, so that variations in Japanese domestic macroeconomic policy manifest themselves in variations in exports. From 1975 to 1981 the volume of Japanese exports increased 71.5 percent while import volume rose only 21.4 percent. Just to show how quickly exports can change, in 1976 export volume was 43 percent above 1973. It is no wonder that competitors in other countries were unhappy. But it is not difficult to explain these developments. Japan suffered a sharp deterioration in its terms of trade owing to the two oil price shocks, and compensated by pushing exports. A big current-account deficit in the first half of 1974 had by the second half of 1974 been turned into a surplus. Taking the whole period from 1975 to 1981, when export volume increased 71.5 percent and import volume rose only 21.4 percent, the country's terms of trade deteriorated 52 percent. The reaction to these developments has been increased protectionism outside Japan. In particular, the two oil shocks created major adjustment problems, some of which manifested themselves indirectly through increased competition from Japan in certain products, as Japan sought to avoid large current-account deficits. It has to be accepted that rapid expansion of Japanese export volume however justified by the rise in its import prices - inevitably generates protectionist pressures in other countries. So it may be advisable in such situations for Japan to moderate rapid export expansion and accept temporarily larger budget deficits (so as to reduce its current-account surpluses) even though such restraint is in the interests neither of consumers in other countries nor of borrowers. One should look rather carefully at the full implications of proposals that Japan open its own markets further to imports from the United States and

74

W. MAX CORDEN

Europe. First it must be stressed that this would have an adverse effect on third countries, such as Australia and many developing countries, if it led to discrimination in favor of the United States (and there are some signs that this is happening). Apart from that, reductions in Japanese protection would have favorable efficiency effects on the standard grounds that more of the benefits of comparative advantage differences would be reaped. But the source of current complaints has been the surge in Japanese exports. If Japan accelerated its growth rate or opened up its markets more, its imports would increase, which would inevitably lead to more exports. There is no reason why the current account would disappear or even turn into a deficit, since this depends on the aggregate savings-investment balance, which is only very indirectly affected by changes in Japanese protection levels, if at all. The tendency would be for the yen to depreciate in real terms even more, thus improving Japanese competitiveness and, no doubt, generating increased pressures from US and European industries for protection. It is doubtful that these pressures would be moderated by the knowledge that the Japanese have reduced their own protection. The industries in the United States that benefit from the extra sales to Japan need not be the same ones that would lose from a new Japanese export expansion. The current pressures in the United States designed to reduce the bilateral trade imbalance between the two countries are misguided for five reasons. First of all, the focus on bilateral balances is grossly misguided. World trade balances - insofar as it does - multilaterally. It is the inevitable nature of comparative advantage that bilateral "imbalances" will emerge. Thus, Australia has a surplus with Japan and a deficit with the United States. In a triangular relationship there is some canceling out of imbalances. Surely Australia should not impose restrictions on imports from the United States, with the aim of favoring Japan, because of its bilateral deficit with the United States, nor should Japan restrict imports from Australia for the same reason. Second, the focus on the trade balance - which does not include trade in services - is misguided (and even more, when it is focused on trade in particular categories of goods, such as automobiles). The United States actually has a surplus in trade in services with Japan. Third, even the concern with overall current-account balances is misguided. In a world of capital mobility, it can be strongly argued that the current account, in itself, does not really matter at all (see Corden, 1992). As noted earlier, it simply reflects the exchange of financial assets for goods. If capital flows into a country in order to finance a private investment boom, and a current-account deficit naturally comes about as part of the transfer process, this may well be a desirable flow. It all depends on the nature of the investment, whether private investment decisions are distorted by public

The revival of protectionism

75

policies, or whether they are based on unwise private investment decisions. If there is a problem, it will not be the current account as such but rather the particular investment decisions. When the current-account deficit results essentially from a budget deficit, as in the case of the United States (where, on the whole, there has not been a very significant change in private savings and investment ratios), the problem is the budget deficit - which imposes a burden on future taxpayers - and not the current account. The United States is lucky to have been able to draw on Japanese and German savers to finance its deficit, and so avoid crowding out of US borrowers. If the United States had continued running the budget deficit but, instead of increasing its foreign debt, had reduced its domestic investment, it would probably have imposed an even greater burden on future generations. Fourth, the pressure on Japan to take the initiative to reduce its surplus and hence the US deficit - was misguided because such action would be harmful to the United States. By what mechanism would an increase in the Japanese budget deficit - which would appreciate the yen and raise world interest rates - reduce the US deficit? In the absence of a change in the US budget deficit, it could only reduce the US deficit if US savings were raised or investment reduced. The rise in world interest rates - and hence US interest rates - would indeed reduce investment. Since it is doubtful that savings respond much to interest rates, it is doubtful that the savings ratio would be raised. Assuming that US monetary policy would, in any case, have assured a level of demand in the United States that yielded the "natural rate of unemployment," there would not be any scope for savings rising owing to an increase in US real income induced by the real depreciation of the dollar which would be the mirror image of the real appreciation of the yen. Thus, success in the US pressure, while benefiting tradable goods producers in the United States and increasing US competitiveness, would bring about a decline in US investment. Finally, there is the most misguided aspect of the pressure for increasing US protection and reducing Japanese protection in order to alter current accounts. There is no reason to believe that overall current accounts would be affected by changes in protection. This point has already been made above. The current-account deficit is definitionally equal to the excess of private investment over private savings plus the budget deficit. Why would US protection lower investment, raise savings, or reduce the budget deficit? Of course, an increase in tariffs might do so insofar as it brought in revenue, but in this respect it is no different from any other tax increase. An increase in US protection might well reduce US imports, but it would lead to a real appreciation of the dollar that would stimulate exports. Protection affects

76

W. MAX CORDEN

the pattern of US tradable goods industries, the protected industries expanding and the others contracting. The current-account balance depends on the net capital flows which, in turn, depend on savings and investment, private and public. At this point a qualification should be noted. If the United States discriminated against imports from Japan, it might succeed in reducing its bilateral deficit with Japan. But, given savings and investment in the United States and Japan, the dollar would appreciate and the US bilateral surplus with Australia, for example, would decline, and its deficit with Taiwan would increase further. As for Japan, when its market in the United States declines owing to the discriminatory measures directed against it, its exchange rate would have to depreciate, and this would reduce its imports from Australia, as well as from the United States. On balance, there would be a reshuffling of bilateral balances, with Japan selling less to the United States and more to other countries, and the United States buying less from Japan and more from other countries. Changes in rules and changes in attitudes Much of the current policy discussion on international trade issues is concerned with institutions and rules. Among the developed countries the post-war movement toward the freeing of trade was built around GATT, and GATT itself was built around the most-favored-nation principle - the principle that tariffs and quantitative import restrictions should not discriminate among sources of supply. Only agriculture and textiles were left out of this process. But GATT has been bypassed more and more, and a feature of the "new protectionism" is that it is discriminatory, leading to bilateral arrangements, and that it involves devices that are not subject to GATT rules and procedures. Various proposals for changes have been made. In general Americans tend to favor extending and tightening rules, ensuring that "codes" are obeyed, and so on, while the Europeans are inclined toward more pragmatism, which at the moment means letting every country be as protectionist as it wants, subject only to some limitations that can be rather easily bypassed. The GATT ministerial meeting of November 1982 came up against this conflict and no progress was made, whether to slow up the existing movement toward more protectionism, to tidy it up and make it more transparent (the latter needed especially with regard to VERs), to enforce and extend the new subsidies code agreed upon at the Tokyo Round, or to extend regulations to include trade in services (as the United States wished). The most urgent need seems to be to bring the various non-tariff

The revival of protectionism

11

restrictions that now exist - the VERs, the quotas, and especially the bilateral arrangements under the auspices of the MFA - within the ambit of GATT rules. This, of course, should be an outcome of the Uruguay Round negotiations. Some allowance must be made for the desire of countries to impose at least temporary restrictions on particular imports when there is a sudden inflow likely to impose substantial damage on particular domestic industries (even though such an import surge would benefit consumers or industries that use these imports as inputs). Article XIX of GATT is a famous article that provides for quotas on these grounds but sets certain conditions. Restrictions must be nondiscriminatory, must be temporary, and must be justified to GATT. Countries imposing restrictions under that article must notify the affected parties in advance and must consult with them, are expected to compensate countries against which affected action is taken, and could be subject to retaliatory action in the form of the withdrawal of equivalent concessions by adversely affected exporters. Countries have bypassed Article XIX by inducing suppliers to accept VERs or what are called "orderly marketing arrangements," of which the MFA is the main case. Suppliers have agreed reluctantly because the alternatives would probably not be Article XIX restrictions but unilaterally imposed import quotas which would be even less satisfactory from the exporters' point of view. Attempts at the Tokyo Round negotiations or later to negotiate a new "safeguards code," involving a revision of Article XIX to make it more acceptable and to bring much of the new protectionism within its ambit, failed. The main obstacle was the determination of the European Community to preserve the principle of selectivity - discrimination among countries — something that was not acceptable to most developing countries and probably also not to the United States. It is well known that the United States was the original protagonist of the principle of non-discrimination, having viewed the British preferential system of the prewar years with great disfavor. But perhaps there is a deeper reason why no agreement had been reached by January 1992 to bring the various existing unilateral or bilateral safeguard arrangements within GATT rules. The major nations simply see no reason to subject their policies to international scrutiny. But, in any case, this matter will be resolved by the outcome of the Uruguay Round negotiations. It could be argued that the central problem is one of attitudes. If attitudes became less protectionist, agreements would be readily reached and it would be found easy to strengthen GATT. If they harden - if protection becomes the new conventional wisdom - then surely in time the attitudes will be reflected not just in a maintenance of the new protectionism but in increased actual protection and thus in a substantial movement away from

78

W. MAX CORDEN

the open trading system. The greatest concern must be about protection that is designed to prevent adjustment in OECD countries to export expansion by developing countries. Essentially, protection as currently practised in developed countries involves protecting one sector of an economy at the expense of others, with the nation as a whole losing, at least in the long term. The gains from protection are usually very visible, perhaps concentrated on one industry, while the losses are more indirect and thinly spread. The understandable argument for protection is that it helps a particular sector, this being a sufficient argument for the relevant lobby. But, as noted above, it is often argued incorrectly and yet persuasively that there are national gains - in particular that, in the absence of protection, employment lost at one end of the economy as a result of an inflow of imports will not be compensated for by employment gains elsewhere. As a result, it is usually thought by non-economists that protection yields national gains at the expense of foreign suppliers, thus providing the basis for trade negotiations. The central message that economists regularly preach, but fail to convey, is that protection is likely to inflict a loss on the protecting nation itself. Yet it is a message that - if understood - should carry much weight. The question thus arises why attitudes have become more protectionist and how a shift in attitudes back to a belief in the gains from trade can be fostered. While this issue seems to arise particularly in the European Community, none of the developed countries seems immune from the revival of protectionist sentiment. It is not difficult to find protectionists in the United States Congress or in the US labor movement. If a shift back to a belief in the advantages of relatively free trade does not occur, it is at least possible that the fairly open world trading system in manufactures that we had until the early 1970s will be gradually eroded with long-term adverse effects for the world economy, and especially for the growth of many developing countries.

Notes This is a revised version of W.M. Corden, The Revival of Protectionismy Occasional Paper No. 14, The Group of Thirty, New York, 1984. 1 An excellent exposition of these issues, making many of the key points stressed in the present paper, appeared in Council of Economic Advisors (1983), chapter 3. 2 The following discussion draws on Hughes and Waelbroek (1981) and Hughes and Krueger (1983). 3 One argument not discussed here is the infant-industry argument. This is used currently by advocates of protection for high-technology or "sunrise" industries. This argument has slightly more to be said for it than some of the other

The revival of protectionism

79

arguments discussed here but is not as widely applicable as often believed. It is analyzed in Corden (1974), chapter 9. References Bergsten, C.F. and W.R. Cline (eds.) (1983), Trade Policy in the 1980s, Washington, DC: Institute for International Economics. Bergsten, C.F. and J. Williamson (1983), "Exchange Rates and Trade Policy," in C.F. Bergsten and W.R. Cline (eds.), Trade Volicy in the 1980s, Washington, DC: Institute for International Economics. Corden, W.M. (1974), Trade Volicy and Economic Welfare, Oxford: Clarendon Press. (1980), "Relationships Between Macroeconomic and Industrial Policies," The World Economy, 3, 167-84. (1992), "Does the Current Account Matter? The Old View and the New," in Jacob A. Frenkel and Morris Goldstein (eds.), International Financial Policy: Essays in Honor of Jacques J. Polak, Washington, DC: International Monetary Fund. Council of Economic Advisers (1983), Annual Report. Washington, DC: US Government Printing Office, chapter 3. Hughes, H. and A.O. Krueger (1983), "Effects of Protection in Developed Countries on Developing Countries' Exports of Manufactures," in R. Baldwin (ed.), The Structure and Evolution of Recent US Trade Policies, Chicago: University of Chicago Press. Hughes, H. and J. Waelbroek (1981), "Can Developing Country Exports Keep Growing in the 1980s?" The World Economy, 4, 127-48. Keesing, D.B. and M. Wolf (1980), "Textile Quotas Against Developing Countries," Thames Essays No. 23, London: Trade Policy Research Centre. Kelly, M., Naheed Kirmani, Miranda Xafa, Clemens Boonekamp, and Peter Winglee (1988), "Developments in International Trade Policy," Occasional Paper No. 16, Washington, DC: International Monetary Fund. Krueger, Anne (1980), "Trade Policy as an Input to Development," American Economic Review, May, 288-92. Lai, D. (1981), "Resurrection of the Pauper-Labour Argument," Thames Essays No. 28, London: Trade Policy Research Centre. Stegemann, K. (1991), "The International Regulation of Dumping: Protection Made Too Easy," The World Economy, 14, 375-405.

CHAPTER

4

Changes in the global trading system: a response to shifts in national economic power ROBERT E. BALDWIN

Introduction Significant changes are taking place in the global trading system. The growing importance of regional trading arrangements is one such development, as indicated by the negotiation of the North America Free Trade Agreement, the enlargement of the European Community, and the revival of regional trading arrangements in Latin American and Africa. Another is the increased use of non-tariff measures to restrict trade, e.g., the voluntary export-restraint agreements in steel, automobiles, and machine tools, the price fixing agreement in computer chips, and the anti-dumping and countervailing duties imposed on a wide range of products, coupled with the greater difficulty of securing trade liberalization through multilateral negotiations, such as the Uruguay Round. Using the threat of closing one's own markets as a means of opening foreign markets is still another major change, as demonstrated by the more frequent use of Section 301 of the 1974 Trade Act and the inclusion of a Super 301 provision in the 1988 Trade and Competitiveness Act. This chapter relates these changes in trade policies to significant structural changes in world industrial production that have brought about a decline in the dominant economic position of the United States, a concomitant rise to international economic prominence of the European Economic Community and Japan, and the emergence of a group of newly industrializing developing countries (NICs). The first two sections describe the rise of the United States to a dominant position in international economic affairs in the immediate post-war period and indicate the types of hegemonic actions followed by this country. "Shifts in International 80

Changes in the global trading system Economic Power" explains how changes in trade, finance, and energy conditions, have led to modifications in national trade policy behavior, particularly on the part of the United States, while the following section discusses the trade policy response of countries to these shifts in economic power. We then speculate about the nature of the international trading regime that is evolving under the present distribution of economic power among nations. The chapter's final section is a summary and conclusion.

The rise in US hegemony The role of the United States in the evolution of the modern trading system has been central. Although this country became an important trader on the world scene after the First World War, it gave little indication at the time of a willingness to assume a major international leadership role. The American share of the exports of the industrial countries rose from 22.1 percent in 1913 to 27.8 percent by 1928 (Baldwin, 1958), but during this period the United States chose political and economic isolation by rejecting membership in the League of Nations and by erecting in 1930 the highest set of tariff barriers in its peacetime history. The failure of the London Economic Conference of 1933 due to the inward-looking economic position of the United States marks the low point of US internationalism in the interwar period. A major policy reorientation toward participation in international affairs began in the United States during the late 1930s and especially in the Second World War. More and more political leaders and the electorate generally began to accept the view of key policy officials in the Roosevelt Administration that continued international isolationism would bring renewed economic stagnation and unemployment to the American economy and the likely prospect of disastrous new worldwide military conflicts. Consequently, active participation in the United Nations was accepted by the American public as were the proposals to establish international economic agencies to provide for an orderly balance-of-payments adjustment mechanism for individual nations and to promote reconstruction and development. International trade had long been a much more politicized subject, however, and all that was salvaged (and then only by executive action) from the proposal for a comprehensive international trade organization was the General Agreement on Tariffs and Trade (GATT). The economic proposals initiated by the United States were not, it should be emphasized, aimed at giving this country a hegemonic role. Rather, they envisioned the United States being one of a small group of nations that would cooperate to provide the leadership necessary to avoid the disastrous nationalistic policies of the 1930s. The envisioned leadership group

81

82

ROBERT E. BALDWIN

included the United Kingdom, France, China, and hopefully, the Soviet Union. Hegemony was thrust upon the United States by a set of unexpected circumstances. The failure of the United Kingdom to return to anything like her prewar position as a world economic power was unforeseen. US officials thought, for example, that the US loan of $3.75 billion to the United Kingdom in 1946 would enable the country to restore sterling convertibility and return to its earlier prominent international role. But the funds were quickly exhausted, and it was necessary to restore exchange control. The 1949 devaluation of the pound was equally disappointing in its failure to revitalize the country. Economic reconstruction in Europe also proved much more costly than envisioned. The resources of the International Bank for Reconstruction and Development (the World Bank) proved to be much too small to handle this task and massive foreign aid from the United States became necessary. The US economy also grew vigorously after the war rather than, as many expected, returning to stagnant conditions. The failure of either China or the USSR to participate in the marketoriented international economy placed an added leadership burden on the United States. But perhaps the most important factor leading to US hegemony was the effort by the Soviet Union to expand its political influence into Western Europe and elsewhere. American officials believed they had little choice from a national viewpoint but to assume an active political, economic, and military leadership role to counter this expansionist policy, an action that most non-communist countries welcomed. The significant expansion of productive facilities in the United States during the war coupled with the widespread destruction of industrial capacity in Germany and Japan gave American producers an enormous advantage in meeting the worldwide pent-up demand of the 1940s and 1950s. The US share of industrial country exports rose from 25.6 percent in 1938 to 35.2 percent in 1952 (Baldwin, 1958). (The combined share of Germany and Japan fell from 24.0 percent to 11.4 percent between these years.) Even in a traditional net import category like textiles, the United States maintained a net export position until 1958. Static trade theory suggests that a hegemonic power will take advantage of its monopolistic position by imposing trade restrictions to raise domestic welfare through an improvement in its terms of trade. However, like the United Kingdom when it was a hegemonic nation in the nineteenth century, the United States reacted by promoting trade liberalization rather than trade restrictionism. A restrictionist reaction might have been possible for a highly controlled, authoritarian economy that could redistribute income fairly readily and did not need to rely on the traded-goods sector as a major source of employment generation or growth, but the growth goals of

Changes in the global trading system free-market firms together with the nature of the political decision-making process rule out such a response in modern industrial democracies. Industrial organization theory emphasizes that firms in oligopolistically organized industries take a long-run view of profitability and strive to increase their market share. In doing so, they try to prevent both new competitors from entering the market and possibly causing losses to existing firms and old competitors from increasing their shares to the point where others might suffer progressive and irreversible market losses. US oligopolistic firms seized the post-war competitive opportunities associated with American dominance to expand overseas market shares both through increased exports and direct foreign investment. The desire of US political leaders to strengthen non-communist nations by opening up American markets and providing foreign aid complemented these goals of US business, and business leaders actively supported the government's foreign policy aims. Even most producers in more competitively organized and less high-technology sectors such as agriculture, textiles, and miscellaneous manufactures favored an outward-oriented hegemonic policy at this time, since they too were able to export abroad and were not faced with any significant import competition. The United States behaved in a hegemonic manner on many occasions in the 1950s and early 1960s. As Keohane (1984, chapter 8) emphasizes, in doing so, it did not coerce other states into accepting policies of little benefit to them. Instead, the United States usually proposed joint policy efforts in areas of mutual economic interest and provided strong incentives for hegemonic cooperation. In the trade field, for example, US officials regularly pressed for trade-liberalizing multilateral negotiations and six such negotiations were initiated between 1947 and 1962. The United States traded short-term concessions for possible long-run gains, since the concessions by most other countries were not very meaningful in trade terms due to the exchange controls they maintained until the late 1950s. The US goal was to penetrate the markets of Europe and Japan as their controls were eased and finally eliminated. One instance in which the United States did put considerable pressure on its trading partners to accept the American viewpoint was in the Kennedy Round of multilateral trade negotiations. At the initial ministerial meeting in 1963, US trade officials - with President Kennedy's approval - threatened to call off the negotiations unless the European Community accepted the American proposal for a substantial, across-the-board tariff-cutting rule. Members of the Community had regained much of their economic vitality and the United States wanted economic payment for its earlier nonreciprocated concessions and its willingness to support a customs union arrangement that discriminated against the United States.

83

84

ROBERT E. BALDWIN

In the financial area the $3.75 billion loan to the United Kingdom in 1946, the large grants of foreign aid after 1948 under the Marshall Plan, and the provision of funds to help establish the European Payments Union in 1948 are examples of hegemonic leadership by the United States. American leaders envisioned a post-war international monetary regime withfixedand convertible exchange rates in which orderly adjustments of balanceof-payments problems would take place. When the International Monetary Fund (IMF) proved inadequate to cope with the magnitude of post-war payments problems, the United States provided financial aid until the affected countries were strong enough economically for the IMF to assume its intended role. A US hegemonic role was also exercised in the energy field, as American companies, with the assistance of the US government, gained control over Arab oil during the 1940s and 1950s. Shifts in international economic power Trade competitiveness

The hegemonic actions of the United States aimed at maintaining the liberal international economic framework established largely through its efforts and at turning back the expansion of the Soviet Union succeeded very well. By 1960 the export market shares of France, Germany, Italy, and Japan had either exceeded or come close to their prewar levels. Among the industrial countries only the United Kingdom failed to regain its prewar position by this time. The restoration of peacetime productive capabilities in these countries meant that the exceptionally high market shares of the United States in the early post-war years declined correspondingly. The 35.2 percent US export share of 1952 had dropped to 29.9 percent by 1960, a figure that was, however, still higher than its 1938 share of 25.6 percent (Baldwin, 1962). For manufactured products alone, the picture is much the same. The US world export share decreased sharply from 29.4 percent in 1953 to 18.7 percent in 1959, while shares of Western Europe and Japan rose from 49.0 percent to 53.7 percent and from 2.8 percent to 4.2 percent, respectively (Branson, 1980). The export market share of Western Europe remained unchanged in the 1960s, but the Japanese share continued to rise and reached 10.0 percent in 1971. At the same time the US share of world exports of manufactures fell to 13.4 percent by 1971. While aid from the US government played an important part in restoring the trade competitiveness of the European countries and Japan, the governments of these nations themselves were the prime driving force for revitalization. The French government, for example, formulated an industrial modernization plan after the war and two-thirds of all new

Changes in the global trading system

85

investments between 1947 and 1950 were financed from public funds. Similarly, the British government under the Labour Party created an Economic Planning Board and exercised close control over the direction of post-war investment, while even the relatively free market-oriented German government channeled capital into key industries in the 1950s. Government investment aid to the steel, shipbuilding, and aircraft industries and the use of preferential government policies to promote the computer sector are other examples of the use of trade-oriented industrial policies in Europe during this period. Japan is perhaps the best-known example of the use of government policies to improve international competitiveness. During the 1950s and 1960s the Japanese government guided the country's industrial expansion by providing tax incentives and investment funds to favored industries. Funding for research and development in high technology areas also became an important part of the government's trade policy in the 1970s. Governments of newly industrializing developing countries used industryspecific investment and production subsidies to an even greater extent than any of the developed nations in their import substitution and export promotion activities. Not only had the prewar export position of the United States been restored by the late 1960s, but the absence of significant import pressures in major industries ended. Stiff competition from the Japanese in the cotton textiles industry was evident by the late 1950s, and the United States initiated the formation of a trade-restricting international cotton textile agreement in 1962. A broad group of other industries also began to face significant import competition in the late 1960s. The products affected included footwear, radios and television sets, motor vehicles and trucks, tires and inner tubes, semi-conductors, hand tools, earthenware table and kitchen articles, jewelry, and some steel items. Trade pattern changes in the 1970s and early 1980s were greatly influenced by the price-increasing actions of the Organization of Petroleum-Exporting Countries (OPEC). This group's share of world exports rose from 5.4 percent in 1970 to 13.7 percent in 1981 {Economic Report of the President, 1986). By 1985 OPEC's share had, however, fallen to 7.1 percent as the power of the cartel declined. The 1970s saw a continued decline in the US share of world exports, although at a greatly decreased rate. The US share of world exports fell from 13.8 percent in 1970 to 11.7 percent by 1981, while that of the European Community dropped from 36.2 percent to 30.7 percent. Japan managed to increase its share from 6.2 percent to 7.6 percent. The non-OPEC developing countries also performed well, increasing their share from 10.9 percent in 1970 to 12.9 percent in 1981.

86

ROBERT E. BALDWIN

The dominant factor influencing US trade in the 1980s was the dramatic appreciation of the dollar during the first half of the decade and its depreciation during the second half. The real trade-weighted value of the US dollar increased from an index of 84.8 in 1980 (1973 = 100) to 131.9 in 1985 and then back to 86.4 by 1990. The appreciation brought about an enormous increase in the merchandise trade deficit from $25 billion in 1980 to $160 billion by 1987. Due to the lag in the adjustment process the deficit was still at $102 billion in 1990. The US share of total world exports actually increased somewhat during this period, rising from 11.1 percent in 1980 to 11.7 percent in 1989 (see International Trade, 89-90, General Agreement on Tariffs and Trade for these and the following figures). However, the US share of world exports of manufactured goods declined from 13.3 percent in 1980 to 11.4 percent in 1988. Japan continued its growth in market share with total exports rising from 6.4 percent in 1980 to 8.9 percent by 1989. For manufacturing the rise was from 11.2 percent in 1980 to 12.7 percent in 1988. However, the most impressive performers in manufacturing were the developing countries. Their share of world exports of manufactured goods rose from 9.5 percent in 1980 to 15.1 percent in 1988. An important feature of the trade-pattern shifts in industrial countries during the 1970s and 1980s was the severe import competition faced not only by labor-intensive sectors like textiles, apparel, and footwear but also by large-scale, oligopolistically organized industries such as steel, automobiles, and shipbuilding. Machine tools and consumer electronic goods also came under increasing import pressure. International financial and other economic changes As a decline in the dominant trade-competitive position of the United States became increasingly evident in the 1960s, both the United States and many other countries became dissatisfied with the US role in the international monetary system offixedexchange rates. Since the supply of gold in the world increases only slowly, the demand for additional international liquidity that accompanied the rapid growth in world trade had to be met by greater holdings of dollars, the other official form of international reserves. However, as these holdings grew, a number of countries became concerned about the freedom from monetary and fiscal discipline that such an arrangement gave the United States and resented the seigniorage privileges it granted. The US also became increasingly dissatisfied with its inability to change the exchange rate of the dollar as a means of adjusting its balance of payments. Another indication of the decline in US hegemony was the creation in 1969 of a new form of

Changes in the global trading system international liquidity in the International Monetary Fund (IMF), namely, Special Drawing Rights (SDRs), designed to reduce the dependence of the international economy on the dollar. The shift to a flexible exchange-rate system in 1971 was the clearest manifestation of the decline in US dominance in the monetary field, however. Although the results of this action have not given countries the expected degree of freedom from US financial influence, the role of the dollar as a reserve and vehicle currency has declined. Another institutional change directed at reducing the monetary influence of the United States was the formation of the European Monetary System in 1979. The severe difficulties faced by the industrial nations in the energy field as a consequence of the success of OPEC have already been mentioned. This development was an especially devastating blow to the international economic prestige of the United States. Trade policy responses to the redistribution of national economic power The non-hegemonic members of the international trading regime, i.e., countries other than the United States, responded to the inevitable industry disruptions caused by the shifts in comparative cost patterns in a manner consistent with their earlier reconstruction and development policies. With the greater post-war emphasis on the role of the state in maintaining full employment and providing basic social welfare needs, these governments intervened to prevent increased imports and export market losses from causing what they considered to be undue injury to domestic industries. Assistance to industries such as steel and shipbuilding injured by foreign competition in third markets took the form of subsidies, including loans at below-market rates, accelerated depreciation allowances and other special tax benefits, purchases of equity capital, wage subsidies, and the payment of worker social benefits. Such activities had been an integral part of the reconstruction and development efforts of the 1940s and 1950s, and the provisions of the GATT dealing with subsidies other than direct export subsidies did not rule out such measures. Because of the difficulties of modifying the tariff-reducing commitments made in the various earlier multilateral trade negotiations, import-protecting measures generally did not take the form of higher tariffs. By requiring compensating duty cuts in other products or the acceptance of retaliatory increases in foreign tariffs, increases in tariffs could have led to bitter disputes and the unraveling of the results of the previous negotiations. Therefore, to avoid such a possibility, governments negotiated discriminatory quantitative agreements outside of the GATT framework with

87

88

ROBERT E. BALDWIN

suppliers who were the main source of the market disruption. For example, quantitative import restrictions were introduced by France, Italy, the United Kingdom, and West Germany on Japanese automobiles and on radios, television sets, and communications equipment from Japan, South Korea, and Taiwan (Balassa and Balassa, 1984). Flatware, motorcycles, and video-tape recorders from Japan and the NICs of Asia were also covered by such import restrictions on the part of various European countries. In the agricultural area, which had been excluded from most of the rules of the GATT, governments did not hesitate to tighten quantitative import restrictions (or restrictions like those under the European Community's Common Agricultural Policy that have the same effect) or provide export subsidies to handle surpluses produced by high domestic price-support programs. In the United States the disrupting effects of the post-war industry shifts in competitiveness throughout the world produced basic policy disputes that continue today. Except for two politically powerful industries, oil and textiles, up until the latter part of the 1960s import-injured industries were forced to follow the administrative route provided for import relief under the escape clause provisions of the GATT. Moreover, many of the industry determinations by the International Trade Commission were rejected at the presidential level on foreign policy grounds, i.e., the need for the hegemonic power to maintain an open trade policy. Industry subsidies provided by foreign governments, though subject to US countervailing duty laws, were also largely ignored by the executive branch for the same reason. The official position of the United States began to change under the strong import pressures of the late 1960s. As their constituents described the competitive problems they faced, fewer members of Congress accepted the standard argument that a liberal US trade policy was essential to strengthen the free world against communism. The intensity of congressional views on trade issues is indicated by the rejection by that body of President Lyndon Johnson's 1968 request for new trade authority and by the near approval in 1970 of protectionist legislation covering a wide variety of products. The growing unwillingness of the allies of the United States in the Cold War to accept the unquestioned US leadership in international political, military, and economic affairs also caused officials in the executive branch to question the traditional American position on trade policies. The view that gradually gained the support of the major public and private interests concerned with trade matters was that a significant part of the increased competitive pressure on the United States was due to unfair trade policies of foreign countries such as government subsidization, dumping by private and public firms, preferential government purchasing

Changes in the global trading system procedures, and discriminatory foreign administrative rules and practices relating to importation. This argument had appeal for several reasons. No new legislation was required to provide import relief; a stricter enforcement of long-existing domestic legislation seemed to be all that was necessary. After a material injury clause was introduced into the US countervailing duty law in 1979, these laws also were consistent with the provisions of the GATT dealing with unfair trade practices. Consequently, stricter enforcement of US unfair trade laws was unlikely to lead to bitter trade disputes with other countries. By placing the blame for their decline in competitiveness on unfair foreign actions, US managers and workers could avoid the implication that this decline might be due to a lack of efficiency on their part. Finally, government officials could maintain that the United States was still supporting the rules of the liberal international regime that the country had done so much to fashion. The emphasis on the greater need for fair trade was clearly evident in the Trade Act of 1974. The best-known provision of this legislation, Section 301, authorizes the president to impose new duties or other import restrictions as well as withdraw previous trade concessions unless foreign countries remove a wide variety of perceived unfair trade practices. As amended by the 1979 Trade Act, these practices include the maintenance of unjustifiable, unreasonable, or discriminatory policies that burden or restrict US commerce or deny benefits to the United States under any trade agreement. Some 70 cases have been brought under Section 301 since 1975 with a noticeable increase in the frequency of its use since 1985. The 1974 Act also authorized US participation in the Tokyo Round of multilateral negotiations. In reshaping the president's proposal, the Congress stressed that the president should seek "to harmonize, reduce, or eliminate" non-tariff trade barriers and tighten GATT rules with respect to fair trading practices. Officials in the executive branch supported these directives not only on their merits but because they deflected attention from more patently protectionist policies. The Tokyo Round codes, which were enacted into US trade law in the Trade Act of 1979, by no means fully satisfied those who strongly stressed the need for fairer trade, but their provisions and the attention that the subject received established the framework for many of US trade-policy actions that followed. There has, for example, been a marked increase in the number of anti-dumping and countervailing duty cases since the 1979 Trade Act. The number of anti-dumping investigations totaled 494 between 1980 and 1990, while the number of countervailing duty investigations amounted to 300 over the same time period. Trade legislation in the 1980s continued to broaden the scope for taking actions against import practices alleged to be unfair and against foreign

89

90

ROBERT E. BALDWIN

countries that restrict access by US exporters to their domestic markets. For example, the Trade and Tariff Act of 1984 and the Trade and Competitiveness Act of 1988 make it easier to obtain anti-dumping and countervailing duties when foreign producers try to avoid such duties either by shipping components of the product subject to the duties rather than the product itself or by shipping another good containing the dumped or countervailed product as an intermediate input. However, so-called Super 301 is the most important provision of the 1988 Act dealing with unfair trade. Under this provision, particular countries targeted for their unfair trade-distorting practices must reach an acceptable agreement with the United States over the practices within a specified time or risk retaliatory action by the United States. One of the most important protectionist actions taken by the United States in the 1970s and 1980s, namely, the gradual tightening of controls over steel imports (over twenty-five countries are now covered by such controls), has also been justified mainly on the grounds of unfair trade practices by foreign producers. For example, the trigger price mechanism (TPM) introduced by the president in 1978 that in effect established minimum import prices for steel was designed to offset foreign dumping. When a series of voluntary export restraint agreements with leading steel exporting nations were concluded in late 1984, a spokesperson for the US Trade Representative stated: "We are responding to unfair trade in the US; defending yourself against unfair trade is not, in our opinion, protectionism" (New York Times, December 20, 1984). The unfair trade argument has been used in support of most other trade-restricting or trade-promoting actions taken by the United States in recent years. The textile and apparel sectors have been described by government officials as "beleaguered" by disruptive import surges, thus justifying more restrictive import controls. Similarly, when temporary orderly marketing agreements (OMAs) were negotiated in the 1970s with selected east and southeast Asian countries, the implication conveyed was that these were responses to unfair export activities of these nations. Even the Japanese voluntary export restraints on automobiles introduced in 1981 are often justified by American industry and government officials on the grounds that the industry's competitive problem was in part due to the unfair targeting practices of the Japanese government. On the exportpromoting side, it is routinely claimed that subsidized export credits through the Export-Import Bank are necessary to counter unfair foreign practices in these areas. In short, fair trade arguments using such phrases as the need for "a level playing field" or "to make foreign markets as open as US markets" have become the basic justification for the greater use of trade-distorting measures by the United States.

Changes in the global trading system Although the United States became more aggressive in the 1970s and 1980s in unilaterally restricting access to its own markets when it believed foreigners were adopting either export-promoting or import-restricting unfair trading policies, US officials continued to press for multilateral negotiations as a means of dealing with such practices. For example, the United States proposed a new multilateral negotiation round at a meeting of trade ministers in 1982. However, both the European Community and the developing countries rejected the US initiative, although they eventually agreed to new negotiations in 1986 (the Uruguay Round). This rejection had a major effect in causing the United States to negotiate a series of regional agreements as a means of securing some of its trading objectives. These include the Caribbean Basin Initiative, the US-Israel Free Trade Agreement, the US-Canada Free Trade Agreement, and the North American Free Trade Agreement. The future of the international trading regime The US fared well economically in its hegemonic role: American exporters and investors established substantial foreign market positions from which they are still benefiting greatly. The open trade policy that US officials were able to maintain for so long also promoted growth and resource-use efficiency and thus extended the period of US economic dominance. But the post-war recovery of Europe and Japan and the emergence of a number of developing countries as industrial economies brought an inevitable relative decline in US economic and political power. The comparative economic position of Western Europe also receded from its post-war recovery level as Japan and the industrializing developing countries grew more rapidly. The outcome in the area of commercial relations has been a relative decline in the multilateral approach to trade policy and increased emphasis on such approaches as non-tariff protection, regionalism, and market opening measures. No country or country group is likely to assume a dominant role in the world economy during the rest of the century and first part of the next. Japan would seem to be the most likely candidate for this leadership role with its highly competitive industrial sector, but this country appears to be too small economically to be a hegemonic power. Moreover, like the United States in the 1920s, Japan is still quite isolationist. Government and business leaders are conditioned by the disastrous outcome of the country's expansionist efforts in the 1930s and 1940s and by its past history of inwardness. Furthermore, when a potential hegemonic nation first demonstrates its competitive strengths over a wide range of products, certain traditional sectors, for instance, agriculture, that are faced with difficult adjustment problems tend to be able to prevent the national commitment to

91

92

ROBERT E. BALDWIN

trade openness required by a dominant economic power. This occurred in the early stages of both the British and American rise to economic dominance and is now hampering Japan from assuming a commitment to openness commensurate with its competitive abilities. In addition, Japanese consumers have not yet developed the taste for product variety needed to make Japan an important market for foreign manufactured goods. The European Community possesses the size and resources to be the dominant economic power, but the very diverse economic nature of its members and the severe structural adjustment problems faced by almost all of them preclude a hegemonic role for this economic bloc. The developing countries of East Asia are becoming major players in world markets for manufactured goods, but they are not sufficiently united politically to act as a dominant power. The United States remains the country most able to identify its trading interests with the collective interests of all. However, a number of the industries that were among the most competitive internationally during the rise of US hegemony have become victims of their success. The relatively high profits these oligopolistically organized industries were able to maintain provided the investment funds needed to take advantage of the expanding market opportunities at home and abroad. But their economic structures were also favorable to the development of powerful labor unions that wished to share these profits through higher wages. The outcome was wage increases in these industries that far exceeded wage increases in manufacturing in general. Consequently, as other countries developed their productive capabilities, these American industries found themselves penalized by above average labor costs and an institutional framework that made it very difficult to adjust to the new realities of international competition. Management in some of these industries also failed to keep up with the most advanced practices. Another very important feature of these industries is their ability to obtain protection by exerting political pressure at the congressional and presidential levels, if they fail to gain it through administrative routes involving the import-injury, anti-dumping, and countervailing duty laws. The consequence of these developments is that protectionism has gradually spread in the United States as such industries as steel, automobiles, computer chip, and machine tools have come under severe international competitive pressure. European governments are faced with even stronger protectionist pressures for similar reasons and have also moved toward more restrictive import policies. As Mancur Olson (1983) has argued, organized common interest groups such as these industries tend to delay innovations and the reallocation of resources needed for rapid growth.

Changes in the global trading system There seems to be no reason why the recent trend toward increased non-tariff protectionism, more regionalism and greater unilateral efforts to open export markets will not continue in the world economy. But one should not conclude from this that the present international trading regime will turn into one where protectionism is rampant and multilateralism is abandoned. There are - and will continue to be - dynamic, export-oriented industries in the older industrial countries that will seek access to foreign markets and see the relation between this goal and open markets in their own country. Moreover, such industries will have considerable political influence, as US high technology and export-oriented service industries have demonstrated. These sectors will continue to provide the United States, Western Europe, and Japan with the economic power that makes international openness a desirable trade-policy objective. Moreover, the GATT is likely to continue to prove to be a mutually beneficial institution in which to resolve trade disputes and reach agreements on matters affecting the trade of most countries. Consequently, none of these trading blocs is likely to adopt a policy of general protection or cease to support the multilateral trading system. But will not creeping protection at the industry level eventually bring a de facto state of general protection? And will not increased unilateralism and regionalism, in effect, destroy the multilateral system? These are, of course, real possibilities. However, one reason this conclusion need not follow is that protection usually does not completely stop decreases in employment in declining industries. Even politically powerful industries usually only have sufficient political clout to slow down the absolute fall in employment. Furthermore, while employment tends to increase due to the fall in imports from the countries against which the controls are directed, offsetting forces are also set in motion. These include a decrease in expenditures on the product as its domestic price tends to rise; a shift in expenditures to non-controlled varieties of the product, to either less or more processed forms of the good, and to substitute products; a redirection of exports by foreign suppliers to more expensive forms of the item; and, if the import controls are country-specific, an increase in exports by non-controlled suppliers. The larger industry profits associated with the increased protection are also likely to be used to introduce labor-saving equipment at a more rapid pace than previously. The continued decline in employment after increased protection is well documented from histories of protection in particular industries (e.g., United States International Trade Commission, 1982). In the European Community and the United States even such politically powerful industries as textiles and apparel and steel have been unable to prevent employment from falling despite increased import protection.

93

94

ROBERT E. BALDWIN

While there are many factors that determine an industry's effectiveness in protection seeking, its size in employment terms is one important factor. With declining employment, an industry is likely to face diminishing political power not only because of a fall in its voting strength but because of a decrease in its ability to raise funds for lobbying purposes. The decline in the political power of the US agricultural sector as the farm population has declined is an example that supports this hypothesis. It seems likely, therefore, that highly protected industries such as textiles and apparel will gradually lose their ability to maintain a high degree of import protection. Consequently, in older industrial nations the spread of protection to sectors in which newly industrializing countries gradually acquire international competitiveness may be offset by a decrease in protection in currently protected sectors. Counter-protectionist pressures also build up as industry-specific protection spreads. The stagnating effect of this policy becomes more obvious as do the budgetary and economic efficiency costs. A state of affairs may thus be reached in which protectionism will not increase on balance in the current group of industrial countries or only at a very slow rate. Meanwhile, export-oriented high technology and service sectors will encourage continued international cooperation to maintain an open trading regime. The formation of regional groupings also in itself lowers trade barriers, although on a discriminatory basis. The enlargement of the European Community (EC) and the development of special free-trade relationships between the EC and other countries in Europe represent a major liberalization of trade as does the North American Free Trade Agreement. Furthermore, after such regional agreements are negotiated, strong pressures are usually exerted on these groups by non-member countries to undertake protection-reducing multilateral trade negotiations so that the discriminatory impact on non-members will be reduced. One can make a case that regional groupings may lead to a faster pace of overall liberalization. Even if this sanguine scenario takes place, the international trading regime is likely to operate quite differently than it did in the years of US dominance. Industrial countries will seek short-run economic reciprocity in their dealings with each other. In particular, the United States seems no longer willing to trade access to the American market for acquiescence to US international political goals and the prospects of long-term penetration of foreign economic markets. The developing countries and nations with special political relationships with particular major trading powers will probably continue to be waived from the full reciprocity requirement but their trade benefits from this waiver will be closely controlled. Greater emphasis will be placed on bilateral negotiations in reducing non-tariff trade distortions, though these negotiations may still take place at general meetings of GATT members. The articles and codes of the GATT will

Changes in the global trading system provide the broad framework for such negotiations, but the variety and discriminatory nature of non-tariff measures make true multilateral negotiations too cumbersome. Bilateral negotiations will also be used to a greater extent in handling trade disputes. The GATT dispute-resolution mechanism will be utilized by smaller countries in their dealings with the larger trading nations and by the larger nations to call attention to actions by one of their members that are outside of generally accepted standards of good behavior. These means of settling disputes will not differ essentially from the practices followed throughout the history of the GATT. Greater discrimination among countries in the granting of trade benefits and the imposition of restrictions is another feature of the emerging international trading regime. This is an inevitable outcome of the increased number of regional trading arrangements. In addition, the increased use by countries of countervailing and anti-dumping duties and the greater willingness to adopt 301-type actions reinforce this trend. More state assistance for the development and maintenance of high-technology and basic industries is likely to be another characteristic of the international trading order. The governments of industrial countries and developing nations will continue to insist on the use of subsidies to develop a certain minimum set of high-technology industries and to maintain a number of basic industries domestically on the grounds that these are needed for a country to become or remain a significant economic power. The international trading regime described above is quite different than the ideal global trading system favored by economists. It will not yield the degree of static economic efficiency and economic growth that economists believe are achievable in an open, non-discriminatory trading order. But this is an essay on the most likely nature of the future international trading order and not on the regime economists would most like to see evolve. Free trade is not a politically stable policy in an economic world of continuing significant structural shifts involving severe adjustment problems for some politically important sectors and demands by infant industries for special treatment. But neither is general import protection a politically stable state of affairs in modern industrial democracies with dynamic export sectors. Politically stable conditions in this type of world economy involve a mix of openness, discrimination, and protection/subsidization across countries and industries. The particular mix will vary depending on such factors as the country distribution of national economic power and the pace of structural change. Summary and conclusions The changing global trade system is related to significant structural changes in world production that have brought about a decline in the

95

96

ROBERT E. BALDWIN

dominant economic position of the United States, a concomitant rise of the European Community and Japan to international economic prominence, and the emergence of a highly competitive group of newly industrializing countries. The trading regime expected to develop after the Second World War involved a shared responsibility on the part of the major economic powers for maintaining open and stable trading conditions. However, the unexpected magnitude of the immediate post-war economic and political problems thrust the United States into a hegemonic role. US economic dominance manifested itself in the trade, finance, and energy fields and enabled American producers to establish strong export and investment positions abroad. Yet, by facilitating the reconstruction and development of Western Europe and Japan as well as the industrialization of certain developing countries, US hegemonic activities led eventually to a marked decline in the American share of world exports and a significant rise of import competition in both labor-intensive sectors and certain oligopolistically organized industries. These developments also significantly diminished the leadership authority of the United States. Most industrial countries responded to the inevitable market disruptions associated with these shifts in comparative advantage by providing extensive government assistance to injured industries in the form of subsidies and higher import barriers. Such behavior was consistent with the extensive role the governments of these countries played in promoting reconstruction and development. For the hegemonic power, the United States, the policy adjustment has been more difficult. Government and business leaders gradually adopted the view that unfair foreign trading practices were a major cause of the country's competitive problems. Therefore, the United States has more vigorously enforced US laws and GATT rules covering alleged unfair trade practices and has also enacted new legislation making it easier to take unilateral actions designed to open foreign markets. Furthermore, in response to their perception that US trade-policy objectives could not be successfully attained through multilateral trade negotiations, US trade officials have attempted to gain some of these goals by negotiating special regional agreements with certain countries. No trading bloc seems able or prepared to become a hegemonic power. However, free trade is not a politically stable policy in a dynamic economic world in the absence of such leadership. Without the foreign policy concerns of the dominant power, domestic sectors injured by import competition and the loss of export markets are able to secure protection or other forms of government assistance through the political processes of industrial democracies. Nevertheless, these industries are unlikely to be able

Changes in the global trading system

97

to prevent market forces from halting a decline in employment and thus an erosion of their political influence. General protectionism is also not a politically stable policy in a rapidly changing economic environment. Politically important export industries that are able to compete successfully abroad will press for the opening of foreign markets and realize the need to open domestic markets to achieve this result. While it is possible that the spread of trade-distorting measures for particular industries, the increased numbers of regional agreements, and the greater emphasis on aggressive bilateral policies will essentially destroy the multilateral trading system that has operated quite successfully for the last fifty years, a more sanguine outcome seems more likely. It involves establishing a reasonably stable trading regime characterized by more trade-distorting government interventions and greater regionalism than at the height of American hegemonic influence. However, while new industries will be added to the list of protected or subsidized sectors over time, others will be withdrawn as they lose political influence so that, on balance, the list does not change much over time. The rate of increase in regional arrangements is also likely to slow down over time and their discriminatory effects on other nations reduced through multilateral pressures. Furthermore, the growing importance of trading problems shared by all nations, as economic and environmental interdependence increases, will continue to require the existence of a multilateral organization such as the GATT for solving these problems. Such a regime may not yield the growth and efficiency benefits of an open trading system, but at least it will not lead to the disastrous economic and political consequences brought about by the type of trading order prevailing in the 1930s. References Baldwin, R.E. (1958), "The Commodity Composition of Trade: Selected Industrial Countries, 1900-1954," The Review of Economics and Statistics, 40, 50-68. (1962), "Implications of Structural Changes in Commodity Trade," in Factors Affecting the United States Balance of Payments, Part 1, Washington, DC: Joint Economic Committee, 87th Congress, 2nd Session. Balassa, B. and C. Balassa (1984), "Industrial Protection in the Developed Countries," The World Economy, 7, 179-96. Branson, W. (1979), "Trends in US International Trade and Investment Since WW II," Princeton University, Princeton, NJ: Department of Economics. Economic Report of the President (1986), Washington, DC: US Government Printing Office. General Agreement on Tariffs and Trade (1990), International Trade 89-90, Vol. II. Geneva, Switzerland. Keohane, R.O. (1984), After Hegemony: Cooperation and Discord in the World Political Economy, Princeton, NJ: Princeton University Press.

98

ROBERT E. BALDWIN

New York Times, December 20, 1984. Olson, M. (1983), "The Political Economy of Comparative Growth Rates," in D.C. Mueller (ed.), The Political Economy of Comparative Growth Rates, New Haven: Yale University Press. United States Department of Commerce (1985), United States Trade: Performance in 1984 and Outlook, Washington, DC: US Department of Commerce, International Trade Administration. United States International Trade Commission (1982), The Effectiveness of Escape Clause Relief in Promoting Adjustment to Import Competition, USITC Publication 1229, Washington, DC: United States International Trade Commission.

CHAPTER

^

"Fortress Europe" and retaliatory economic warfare L.R. KLEIN a n d PINGFAN HONG

Ever since protectionist tendencies began to appear on an increasing scale during the 1970s, Project LINK has tried to estimate the effects on the world economy. At first, there was general consideration of the effects of the increased tariffs (see Klein and Su, 1979). Not surprisingly, the system validated the arguments in favor of free trade, in the sense that protectionism generated overall economic loss in the form of reduced global production and trade volume. The microeconomic effects of trade barriers are argued to be welfare losses and departures from (Pareto) optimality. It was, however, the objective of LINK to examine the macroeconomic effects. Trade barriers extend far beyond the imposition of tariffs, but, for understandable reasons of quantification and also ease of computation, tariff changes served as the instrument of protectionism. In principle, non-tariff barriers to international trade can be approximated by a corresponding tariff rate, but we did not undertake the extensive analysis that would be necessary to pair non-tariff and tariff barriers. For the present exercise, tariffs are imposed on total imports of given groups of countries, specifically, the EEC (Common Market) and the ROW (Rest of the World). On occasion, within the context of LINK simulations we have further examined tariff influences on classes of imports, manufactures, e.g., crude oil, or combinations excluding some primary products (see Klein, Pauly, and Petersen, 1987). The impacts of more specific forms of protectionism are localized, but, generally speaking, LINK results at the macrolevel are consistently in support of free trade. 99

100

L.R. KLEIN AND PINGFAN HONG

External fears of the single market in Europe The European Economic Community (EEC) has been a practical success. It has expanded over the years and become an accepted institution. While the economist's ideal might be one of unrestricted multilateral free trade, in the framework of welfare economics, most people admire the achievements of the EEC in giving a sense of economic unity to Europe and in providing a framework into which peripheral and emerging countries in Europe will fit. Some or all the countries in the European Free Trade Association (EFT A) and the countries of the former Conference on Mutual Economic Assistance (CMEA) seem to want to join the enhanced EEC, after 1992. It is easy to understand how the EEC might improve trading conditions within the European territory that is directly involved. It can, however, fall short of optimality in the strict economic welfare sense if the gains from liberalization accrue only to member countries and divert trade or other possible economic benefit from countries outside the system. In particular, countries of North America and the Asia-Pacific community, admire, from the outside, what the European countries are doing for themselves, but they are apprehensive about the new step toward more complete economic unification in Europe. They fear that a "Fortress Europe" could be erected to promote as much internal trade within the EEC as possible, to the exclusion of trade with outsiders. Experience with the attempts to remove barriers to agricultural trade under the Uruguay Round has left outsiders pessimistic that European countries will liberalize trade with outsiders and raises the possibility of the imposition of new barriers against trade with non-Europeans. We have accordingly made new trade scenario calculations with the Project LINK system to examine the effects of tariffs imposed on imports generally by EEC countries against ROW, followed by retaliation on the part of ROW against EEC countries in a virtual trade war. Of course, these calculations are only expository. The barriers, as in the case of agricultural trade, would not be entirely, if at all, in the form of tariffs. Also the entire ROW or the entire merchandise trade volume will not be treated uniformly. Nevertheless, the calculations do look interesting and reveal the folly of economic warfare. Some conceptual and computational issues The import flow from country j to / can be specified as

"Fortress Europe"

101

Xji = real merchandise imports of / fromy, GDPt = real GDP of/, Pf = domestic price index for /, PXjt = price of imports of / fromy (countryy's export price converted to /'s currency units including CIFO adjustment), Tji = tariff rate imposed by / on goods fromy, PX.t = weighted average of price of imports of / from all countries except y. The first argument has a positive effect on xjh while the next two have negative effects. If country / should increase its tariff rate on goods fromy, the cost of imports fromy relative to its domestic goods will rise, and imports fromy will be expected to fall. Also the cost of imports fromy relative to the price from other countries will rise, and this, too, will cause imports fromy to fall. Country j's exports will show a fall in the "mirror" statistics. Other countries will gain an advantage, relative toy in exporting to /. Depending on the strength of the relative price effects, the reduced demand for imports in / could lead to an improved current-account balance. Domestic spending might fall in country /, but /'s treasury collections of import duties will tend to improve the budget balance and there could be some secondary effects in financial markets. If countryy retaliates by raising its duty on goods imported from /, then the same chain of reasoning applies to variables associated withy's imports. The final outcome for both countries, worked out by the LINK system, depends on the strengths of several elasticities or effects. If the EEC, as a whole, raises their duties on goods from ROW (but not against EEC trade partners) and if ROW does likewise against EEC, then trade between the two regions will fall, but trade within each of the two regions might increase. There are some practical difficulties in implementing these changes in the LINK system in precisely the way that was outlined above because bilateral import functions in LINK are not all specified as above. Import functions are not directly estimated from bilateral flows between / andy, but as a country's total imports, which are then allocated among trading partners according to calculations from a trade-share matrix. The overall import equation for country / is

m^ = real merchandise imports of /, GDPi = real GDP of/,

102

L.R. KLEIN AND PINGFAN HONG

PMi = average import price of / (calculated as a weighted average of partners' export prices), Tt = weighted average of all tariff rates for /, Pi = domestic price index for /. When country / increases the tariff rate on imports fromy, the average rate, Tf, will increase just enough to reflect the contribution of j to the average rate. This will lead to a decrease in m{ because the second argument in the function above will increase, and its effect on m{ is negative. The decline in mi would be allocated to all trade partners of/ and not toy, as in the previous analysis. We must, therefore, modify the LINK calculation based on a given trade-share matrix. We have, in the protectionist scenario, increased tariff rates by 10 percent for all EEC countries and in a preliminary LINK solution obtained an increase in exports by the EEC countries, as a whole (by summation). In the LINK algorithm, a country's exports are expressed as a weighted sum of partner's imports,

VXL(= £ oiijVMj+CAi, ieEEC VXLt = value of/'s exports computed by linkage, OLij = ij element of trade-share matrix, VM( = value of/'s imports, CAt = adjustment term to /'s exports (for world as a whole

t

i=l

We then reallocate VXLh

/'GEEC,

A jeROW

£

ieEEC

to ROW by the formula

VXL

<

By adding the vector of ACAj to the LINK equations for exports that were generated by a given trade-share matrix, we will get export values that are approximately the same as those generated by a variable trade-share matrix. The resulting trade pattern among countries will reduce ROW export shares to EEC countries by the amount of changes in EEC countries' import demands from outside. Where there is no tariff change (within EEC and within ROW) the export shares will remain unchanged. By following these steps, we formed a non-retaliatory trade simulation in which EEC tariffs are raised against all members of ROW. We then

"Fortress Europe"

103

superimposed on this solution, a retaliatory solution, following the same procedures for an increase in ROW tariffs against EEC countries. Some estimates of protection and retaliation If the EEC countries raise their tariffs against ROW, without any retaliatory action by the latter, we estimate a decrease in world trade volume by about 2.5 percent annually. The scenario begins in 1992, with just a 2.3 percent drop, but it builds up quickly to a 2.9 percent drop, settling soon at 2.5 percent. The EEC countries realize a very small gain in nominal exports of merchandise, but a slightly larger drop in import value. All the percentage changes here are presented as deviations from the time path of the base-case simulation. North American exports drop by as much as 3.6 percent in nominal value, while Japanese exports drop by as much as 4.3 percent. Industrialized countries outside the EEC, North America, and the Pacific Basin lose about 4-5 percent of their export value. Within the EEC, the import reductions are not uniform, in percentage terms. UK imports are cut back by only 1.0 to 3.0 percent. Developing countries lose exports too, by 3 to 3.5 percent. World production, measured as the sum of GDP over all countries (gross world product) is estimated to fall by just under 1.0 percent, building up steadily in size for the period to 1996. Japan is estimated to lose more production, in percentage terms (more than 1.0 percent), than many of the others. Some countries in the EEC gain, especially France, but others lose, especially the UK and Germany. The developing countries also lose GDP. By and large, the whole world is worse off, in terms of production, as a result of protectionism. This is an expected macroeconomic result and confirms earlier studies of protectionism. With production down, the unemployment rate ought to rise. It does rise in this scenario but not by very much, either by 0.1 or 0.2 in most cases but not by more than 0.7 at the extreme. Similarly, inflation for the OECD countries rises and then falls, by about 0.3 percentage points. It falls, however, in North America and the industrial countries of the Far East, where output is set back by the EEC protectionism but also in Western Europe. Summary table 5.1 presents more numerical results for the case without retaliation. Now, let us look at table 5.2, where both EEC and ROW raise tariff rates by 10 percent. The changes in an environment of economic warfare are even larger, and, of course, perverse. The percentage decline in world trade is estimated to be more than double the previous results. In this scenario, the United States, Germany and France appear to have cut back imports by more than 10 percent, more than other industrial countries appear to have

Table 5.1 A. World exports, imports and trade balances (f.o.b.): Project LINK - United Nations/DIES A (Exchange rates endogenous) EC raise their tariff rate by 10% (a) Spring 1992 1992 World exports World export price World exports real

-58.96 0.03 -19.24

Developed market economies Exports -29.87 Imports -57.91 28.04 Balance North-America Exports -16.00 Imports -1.96 -14.04 Balance United States -12.48 Exports Imports -2.16 -10.32 Balance Developed East Exports -12.60 Imports -0.27 Balance -12.33

%chg

1993

%chg

1994

-1.6 0.7 -2.3

-61.76 0.06 -25.89

-1.5 1.4 -2.9

-1.1 -2.2

-23.62 -59.75 36.12

-2.7 -0.3

%chg

1995

%chg

1996

%chg

Billions of US$ -50.93 - 1 . 2 0.08 1.7 -26.42 - 2 . 8

-49.41 0.07 -25.73

-1.0 1.6 -2.5

-54.70 0.07 -26.38

-1.0 1.5 -2.5

-0.8 -2.1

-10.80 -48.91 38.12

-0.3 -1.6

-8.36 -47.21 38.85

-0.2 -1.4

-9.67 -52.05 42.37

-0.3 -1.4

-22.86 -5.57 -17.29

-3.6 -0.8

-23.54 -4.53 -19.01

-3.4 -0.6

-23.12 -4.09 -19.03

-3.0 -0.5

-25.36 -6.45 -18.92

-3.0 -0.7

-2.7 -0.4

-17.57 -5.80 -11.76

-3.5 -1.1

-18.26 -4.86 -13.41

-3.3 -0.8

-18.04 -4.43 -13.61

-3.0 -0.7

-19.70 -6.62 -13.08

-2.9 -0.9

-3.3 -0.1

-17.13 -0.38 -16.75

-4.2 -0.1

-17.23 -0.49 -16.74

-3.9 -0.1

-16.68 -0.71 -15.97

-3.4 -0.2

-18.09 -1.03 -17.05

-3.5 -0.3

Japan Exports Imports Balance EC Exports Imports Balance Germany Exports Imports Balance France Exports Imports Balance United Kingdom Exports Imports Balance Rest of industrialized Exports Imports Balance Developing countries Exports Imports Balance

-11.25 -0.40 -10.85

-3.4 -0.2

-15.21 -0.70 -14.50

-4.3 -0.3

-15.07 -0.94 -14.13

-3.9 -0.3

-14.38 -1.14 -13.24

-3.5 -0.4

-15.51 -1.41 -14.10

-3.5 -0.4

11.35 -57.81 69.16

0.8 -3.9

31.17 -58.84 90.01

2.0 -3.7

44.43 -50.87 95.30

2.7 -2.9

45.66 -49.16 94.82

2.6 -2.7

48.67 -51.46 100.13

2.5 -2.6

2.38 -37.63 40.01

0.5 -9.1

7.27 -27.86 35.14

1.5 -6.3

10.89 -21.60 32.49

2.1 -4.5

11.00 -19.99 30.99

2.0 -4.0

10.99 -18.95 29.93

1.8 -3.4

1.51 -4.49 6.01

0.7 -2.0

4.23 -12.25 16.48

1.8 -5.1

6.13 -12.46 18.59

2.4 -4.7

6.48 -12.46 18.94

2.3 -4.3

7.27 -14.96 22.23

2.4 -4.7

1.30 -1.07 2.37

0.7 -0.5

3.43 -5.58 9.01

1.7 -2.4

4.84 -4.15 8.99

2.2 -1.6

4.98 -2.52 7.50

2.1 -0.9

5.29 -0.85 6.14

2.0 -0.3

-12.62 2.12 -14.75

-4.7 0.7

-14.80 5.04 -19.84

-5.1 1.6

-14.46 6.97 -21.42

-4.5 2.0

-14.23 6.75 -20.97

-4.0 1.8

-14.90 6.89 -21.78

-3.9 1.7

-25.30 -1.77 -23.53

-2.8 -0.2

-34.39 -2.78 -31.61

-3.4 -0.3

-36.70 -2.62 -34.07

-3.2 -0.2

-37.56 -2.32 -35.25

-2.9 -0.2

-41.47 -2.71 -38.76

-2.9 -0.2

Table 5.1 A. (cont.) 1992

%chg

1993

%chg

1994

%chg

1995

%chg

1996

%chg

Billions of US$ Latin America, Caribbean Exports Imports Balance Africa Exports Imports Balance North Africa Exports Imports Balance Sub-Sahara Africa Exports Imports Balance Nigeria Exports Imports Balance South, East Asia Exports Imports Balance

-3.77 0.06 -3.84

-2.9 0.0

-5.06 0.05 -5.10

-3.6 0.0

-5.32 0.04 -5.36

-3.5 0.0

-5.44 0.14 -5.59

-3.3 0.1

-6.13 0.11 -6.24

-3.4 0.1

-2.94 0.00 -2.94

-3.7 0.0

-3.42 0.14 -3.56

-4.0 0.1

-3.52 0.02 -3.54

-3.9 0.0

-3.61 -0.18 -3.44

-3.7 -0.2

-3.88 -0.36 -3.52

-3.8 -0.3

-1.70 -0.18 -1.52

-4.6 -0.4

-1.91 -0.12 -1.79

-4.5 -0.3

-1.95 -0.25 -1.71

-4.3 -0.5

-1.99 -0.37 -1.62

-4.1 -0.6

-2.09 -0.49 -1.60

-4.0 -0.8

-1.24 0.18 -1.42

-3.0 0.4

-1.51 0.26 -1.77

-3.5 0.5

-1.57 0.26 -1.83

-3.5 0.5

-1.62 0.19 -1.81

-3.4 0.3

-1.78 0.13 -1.91

-3.5 0.2

-0.33 0.07 -0.40

-3.1 1.2

-0.32 0.14 -0.46

-3.1 2.2

-0.30 0.18 -0.48

-2.8 2.5

-0.28 0.18 -0.46

-2.6 2.3

-0.29 0.18 -0.47

-2.5 2.2

-13.42 -2.19 -11.23

-2.7 -0.4

-19.68 -2.56 -17.12

-3.4 -0.4

-21.33 -2.27 -19.05

-3.2 -0.3

-21.82 -1.83 -19.99

-2.9 -0.2

-24.36 -2.41 -21.95

2.8 -0.3

China Exports Imports Balance West Asia Exports Imports Balance Oil Export Ctrs. Exports Imports Balance Oil Import Ctrs. Exports Imports Balance Mediterranean Exports Imports Balance Eastern Eur. and USSR. Exports Imports Balance USSR Exports Imports Balance

1.51 0.00 1.52

-2.3 0.0

-1.98 0.01 -1.99

-2.8 0.0

-2.03 0.08 -2.11

-2.6 0.1

-1.99 0.07 -2.06

-2.3 0.1

-2.13 0.09 -2.21

-2.2 0.1

1.67 0.24 1.91

-2.4 0.4

-1.96 0.56 -2.52

-2.5 0.8

-2.20 0.72 -2.92

-2.4 0.9

-2.41 0.59 -3.00

-2.4 0.7

-2.50 0.53 -3.03

-2.3 0.6

1.52 0.18 1.70

-2.3 0.4

-1.78 0.45 -2.22

-2.5 0.8

-2.02 0.58 -2.60

-2.4 0.8

-2.23 0.45 -2.69

-2.4 0.6

-2.32 0.39 -2.71

-2.3 0.5

0.15 0.06 0.21

-2.7 0.7

-0.18 0.11 -0.29

-3.0 1.3

-0.18 0.14 -0.32

-2.7 1.5

-0.18 0.14 -0.31

-2.5 1.4

-0.19 0.14 -0.32

-2.5 1.3

1.50 0.23 1.72

-4.7 0.6

-1.71 -1.02 -0.70

-4.9 -2.3

-1.68 -1.33 -0.35

-4.4 -2.7

-1.69 -1.25 -0.44

-4.1 -2.3

-1.82 -0.76 -1.06

-4.0 -1.3

3.80 0.73 4.53

-3.2 0.6

-3.74 0.93 -4.67

-3.1 0.8

-3.44 0.81 -4.25

-2.7 0.6

-3.47 0.26 -3.73

-2.6 0.2

-3.56 0.32 -3.88

-2.4 0.2

2.16 0.73 2.89

-2.9 1.0

-2.12 1.07 -3.19

-2.9 1.4

-1.90 0.92 -2.82

-2.5 1.2

-1.79 0.49 -2.28

-2.2 0.6

-1.83 0.61 -2.43

-2.0 0.6

Table 5.1A. (cont.) 1992

%chg

1993

%chg

1994

%chg

1995

%chg

1996

%chg

Billions of US$ World exports World export price World exports real FRG Exports Imports Balance Eastern Europe 5 Exports Imports Balance Dev. Ctrs (Excl. China) Exports Imports Balance Least dev. countries Exports Imports Balance

-58.41 0.03 -19.35

-1.5 0.7 -2.2

-60.21 0.07 -26.38

-1.4 1.5 -2.9

-48.66 0.08 -27.14

-1.1 1.8 -2.8

-47.11 0.08 -26.43

-1.0 1.6 -2.6

-52.33 0.08 -27.14

-1.0 1.6 -2.5

2.61 -36.03 38.64

0.5 -8.2

8.08 -28.50 36.58

1.5 -6.1

12.15 -23.08 35,23

2.2 -4.6

12.25 -21.69 33.94

2.1 -4.1

12.17 -20.66 32.83

1.9 -3.5

-1.64 0.00 -1.64

-3.8 0.0

-1.62 -0.14 -1.48

-3.6 -0.3

-1.54 -0.11 -1.42

-3.1 -0.2

-1.69 -0.23 -1.46

-3.2 -0.4

-1.73 -0.28 -1.45

-3.0 -0.5

-23.79 -1.78 -22.01

-2.8 -0.2

-32.42 -2.79 -29.63

-3.4 -0.3

-34.67 -2.70 -31.97

-3.2 -0.2

-35.58 -2.39 -33.19

-3.0 -0.2

-39.34 -2.79 -36.55

-3.0 -0.2

-0.25 0.00 -0.24

-3.5 0.0

-0.31 -0.04 -0.27

-4.0 -0.2

-0.33 -0.05 -0.28

-3.8 -0.3

-0.34 -0.05 -0.29

-3.6 -0.3

-0.38 -0.05 -0.33

-3.7 -0.2

Net energy exp. ctrs Exports Imports Balance Net energy imp. ctrs Exports Imports Balance

-8.35 -0.30 -8.05

-2.5 -0.1

-10.40 0.56 -10.96

-2.8 0.2

-10.98 0.60 -11.58

-2.7 0.2

-11.20 0.48 -11.68

-2.5 0.1

-12.08 0.14 -12.22

-2.5 0.0

-16.95 -1.47 -15.48

-3.0 -0.2

-24.00 -3.34 -20.65

-3.7 -0.5

-25.71 -3.22 -22.49

-3.5 -0.4

-26.37 -2.80 -23.57

-3.1 -0.3

-29.39 -2.85 -26.54

-3.1 -0.3

Notef Absolute and percentage deviations.

Table 5. IB. World gross national product (1970 US$) Project LINK - United Nations/DIES A EC raise their tariff rate by 10% Spring 1992 1992

%chg

1993

%chg

6520.01 6499.33 -20.68

1.3 1.0 -0.32

6709.21 6660.86 -48.35

Developed market economies 4603.19 4584.66 -18.53 North-America 1933.63 1924.33 -9.30 United States 1758.03 1748.74 -9.30 607.07 Developed East 603.96 -3.11 528.85 Japan 526.14 -2.71 EC 1871.34 1866.10 -5.24

1.9 1.4 -0.40 1.6 1.1 -0.48 1.4 0.9 -0.53 2.6 2.0 -0.51 2.5 1.9 -0.51 1.9 1.6 -0.28

4742.33 4706.57 -35.77 1994.38 1985.98 -8.40 1810.25 1801.85 -8.40 628.95 624.04 -4.91 548.76 544.40 -4.37 1923.14 1902.12 -21.02

World total

%chg

1996

%chg

Baseline/Alternative/Difference 2.9 6918.87 3.1 7150.24 2.5 6869.87 3.1 7092.06 -0.72 -49.00 - 0 . 7 1 -58.18

3.3 3.2 -0.81

7377.66 7311.64 -66.01

3.2 3.1 -0.89

3.0 2.7 -0.75 3.1 3.2 -0.42 3.0 3.0 -0.46 3.6 3.3 -0.78 3.8 3.5 -0.80 2.8 1.9 -1.09

2.9 2.8 -1.00 3.3 3.1 -0.15 3.1 3.0 -0.17 3.1 3.0 -0.99 3.3 3.2 -1.02 2.5 2.3 -1.93

5149.43 5092.17 -57.26 2192.67 2188.20 -4.47 1985.70 1981.22 -4.47 689.04 681.37 -7.66 602.63 595.64 -6.99 2055.04 2011.62 -43.42

2.5 2.4 -1.11 3.1 3.0 -0.20 3.1 3.0 -0.23 2.8 2.7 -1.11 2.9 2.7 -1.16 1.8 1.6 -2.11

1994

4881.46 4839.83 -41.63 2059.88 2059.98 0.10 1867.84 1867.95 0.10 649.81 643.92 -5.89 567.25 561.95 -5.29 1969.76 1935.51 -34.25

%chg

2.9 2.8 -0.85 3.3 3.7 0.01 3.2 3.7 0.01 3.3 3.2 -0.91 3.4 3.2 -0.93 2.4 1.8 -1.74

1995

5023.40 4972.93 -50.46 2126.86 2123.64 -3.22 1926.65 1923.43 -3.22 670.08 663.46 -6.62 585.83 579.83 -6.00 2018.54 1979.59 -38.95

Germany France

United Kingdom

Rest of industrialized

Developing countries

Latin America, Caribbean

Africa

North Africa

Sub-Sahara Africa

928.62 926.67 -1.95 224.36 224.89 0.53 145.97 145.06 -0.91 191.16 190.27 -0.89 1285.62 1281.84 -3.78 330.54 329.78 -0.76 136.53 135.89 -0.63 74.51 74.01 -0.50 62.02 61.89 -0.13

1.7 1.5 -0.21 1.3 1.6 0.23 2.0 1.4 -0.62 1.5 1.0 -0.46 5.5 5.2 -0.29 3.6 3.4 -0.23 3.6 3.1 -0.46 4.1 3.4 -0.67 3.1 2.8 -0.22

953.68 937.65 -16.03 230.16 232.64 2.48 150.05 147.95 -2.10 195.87 194.44 -1.43 1360.72 1354.80 -5.92 344.40 343.32 -1.08 141.40 140.64 -0.76 77.49 76.92 -0.57 63.91 63.72 -0.19

2.7 1.2 -1.68 2.6 3.4 1.08 2.8 2.0 -1.40 2.5 2.2 -0.73 5.8 5.7 -0.43 4.2 4.1 -0.31 3.6 3.5 -0.54 4.0 3.9 -0.73 3.0 3.0 -0.30

969.78 940.96 -28.82 236.65 240.99 4.34 155.25 152.58 -2.67 202.02 200.43 -1.60 1442.67 1436.06 -6.60 361.01 359.66 -1.35 147.07 146.14 -0.92 81.09 80.38 -0.71 65.98 65.76 -0.21

1.7 0.4 -2.97 2.8 3.6 1.83 3.5 3.1 -1.72 3.1 3.1 -0.79 6.0 6.0 -0.46 4.8 4.8 -0.37 4.0 3.9 -0.63 4.6 4.5 -0.88 3.2 3.2 -0.32

987.11 955.56 -31.55 243.55 248.50 4.95 160.13 157.10 -3.03 207.93 206.25 -1.68 1528.31 1521.45 -6.86 377.82 376.22 -1.61 152.79 151.84 -0.95 84.69 83.97 -0.72 68.10 67.87 -0.23

1.8 1.6 -3.20 2.9 3.1 2.03 3.1 3.0 -1.89 2.9 2.9 -0.81 5.9 5.9 -0.45 4.7 4.6 -0.43 3.9 3.9 -0.62 4.4 4.5 -0.85 3.2 3.2 -0.34

998.32 964.39 -33.93 249.76 254.60 4.84 164.51 161.32 -3.20 212.69 210.98 -1.71 1616.75 1609.25 -7.50 395.02 393.11 -1.91 159.42 158.43 -0.98 88.99 88.26 -0.74 70.42 70.17 -0.25

1.1 0.9 -3.40 2.6 2.5 1.94 2.7 2.7 -1.94 2.3 2.3 -0.80 5.8 5.8 -0.46 4.6 4.5 -0.48 4.3 4.3 -0.62 5.1 5.1 -0.83 3.4 3.4 -0.35

Table 5.IB. (cont.)

South, East Asia

China

West Asia

Oil Export Ctr. Oil Import Ctr.

Mediterranean

Eastern Europe and USSR

USSR

1992

%chg

1993

%chg

1994

%chg

387.30 385.50 -1.80 355.44 355.17 -0.27 21.35 21.23 -0.12 16.83 16.71 -0.12 4.52 4.52 0.00 54.47 54.27 -0.20 631.20 632.83 1.63 494.93 497.01 2.08

5.9 5.4 -0.46 8.2 8.1 -0.08 7.8 7.2 -0.56 9.5 8.7 -0.71 1.7 1.7 0.00 0.7 0.3 -0.36 -9.5 -9.3 0.26 -11.0 -10.6 0.42

412.61 410.11 -2.50 380.04 379.69 -0.35 23.10 22.91 -0.19 18.48 18.29 -0.19 4.62 4.62 0.00 59.17 58.13 -1.04 606.17 599.50 -6.67 468.04 461.67 -6.37

6.5 6.4 -0.61 6.9 6.9 -0.09 8.2 7.9 -0.82 9.8 9.5 -1.02 2.2 2.2 0.00 8.6 7.1 -1.76 -4.0 -5.3 -1.10 -5.4 -7.1 -1.36

Billions of US$ 439.40 6.5 436.82 6.5 -2.58 -0.59 406.68 7.0 406.36 7.0 -0.32 -0.08 25.01 8.3 24.84 8.4 -0.18 -0.72 20.29 9.8 20.11 9.9 -0.18 -0.88 4.73 2.4 4.73 2.4 0.00 0.00 63.50 7.3 62.25 7.1 -1.25 -1.97 594.76 - 1 . 9 593.99 - 0 . 9 -0.76 -0.13 452.57 - 3 . 3 452.19 - 2 . 1 -0.38 -0.08

1995

%chg

1996

%chg

468.27 465.77 -2.50 435.18 434.90 -0.29 26.56 26.38 -0.19 21.72 21.54 -0.19 4.84 4.84 0.00 67.68 66.36 -1.32 598.53 597.68 -0.85 450.36 450.08 -0.27

6.6 6.6 -0.53 7.0 7.0 -0.07 6.2 6.2 -0.71 7.1 7.1 -0.86 2.4 2.4 0.00 6.6 6.6 -1.95 0.6 0.6 -0.14 -0.5 -0.5 -0.06

497.44 494.56 -2.88 465.52 465.24 -0.28 27.75 27.57 -0.19 22.77 22.59 -0.19 4.98 4.98 0.00 71.60 70.34 -1.26 611.49 610.23 -1.26 456.05 455.45 -0.59

6.2 6.2 -0.58 7.0 7.0 -0.06 4.5 4.5 -0.67 4.8 4.9 -0.82 2.8 2.8 0.00 5.8 6.0 -1.76 2.2 2.1 -0.21 1.3 1.2 -0.13

FRG Eastern Europe 5

Dev. Ctr. (excl. China)

Least Dev. Ctr.

Net energy exp. ctrs

Net energy imp. ctrs

885.17 882.21 -2.97 136.27 135.82 -0.45 930.18 926.67 -3.51 25.10 25.04 -0.06 623.38 622.01 -1.37 662.24 659.83 -2.41

1.4 1.1 -033 -3.6 -3.9 -0.33 4.5 4.1 -0.38 3.6 3.4 -0.25 6.8 6.6 -0.22 4.3 3.9 -0.36

906.49 889.95 -16.54 138.13 137.83 -0.29 980.68 975.11 -5.57 25.97 25.86 -0.11 661.58 660.06 -1.53 699.13 694.75 -4.39

2.4 0.9 -1.82 1.4 1.5 -0.21 5.4 5.2 -0.57 3.4 3.3 -0.42 6.1 6.1 -0.23 5.6 5.3 -0.63

919.04 889.31 -29.73 142.18 141.80 -0.38 1035.99 1029.70 -6.29 26.83 26.71 -0.12 703.68 702.00 -1.68 738.99 734.07 -4.92

1.4 -0.1 -3.23 2.9 2.9 -0.27 5.6 5.6 -0.61 3.3 3.3 -0.46 6.4 6.4 -0.24 5.7 5.7 -0.67

931.73 899.36 -42.37 148.17 147.60 -0.58 1093.13 1086.56 -6.57 27.73 27.61 -0.12 747.47 745.84 -1.63 780.84 775.62 -5.22

1.4 1.1 -3.47 4.2 4.1 -0.39 5.5 5.5 -0.60 3.4 3.4 -0.45 6.2 6.2 -0.22 5.7 5.7 -0.67

938.73 903.97 -34.76 155.44 154.78 -0.66 1151.23 1144.01 -7.22 28.66 28.54 -0.13 795.16 793.48 -1.68 821.59 815.77 -5.82

0.8 0.5 -3.70 4.9 4.9 -0.43 5.3 5.3 -0.63 3.4 3.4 -0.45 6.4 6.4 -0.21 5.2 5.2 -0.71

114

L.R. KLEIN AND PINGFAN HONG

Table 5.1C. OECD unemployment rate: Project LINK - United Nations/ DIESA EC raise their tariff rate by 10% (°) Spring 1992. Mean0

1992

1993

1994

7.6 7.7 7.8 8.0 3.5 3.7 9.4 9.4

7.4 7.7 7.5 7.7 3.4 3.6 9.3 9.6

7.2 7.4 6.9 7.0 3.5 3.5 9.2 9,8

6.9 7.1 6.3 6.4 3.8 3.7 9.0 9.7

7.0 7.2 6.2 6.4 4.2 4.1 9.3 9.9

7.2 7.4 7.0 7.1 3.7 3.7 9.3 9.7

12.5 12.6

12.4 12.5

11.5 11.7

10.8 11.0

10.3 10.4

11.5 11.6

1995

1996

Percent OECD° b

North-America" North-America6 Developed East0 Developed East6 EEC° EEC6 Rest of OECD0 Rest of OECD6

Notes: Unemployment rate in Baseline. 6 Unemployment rate in Alternative. c Average unemployment rate for period. Excl. Greece, Iceland, and Switzerland.

a

Table 5.ID. OECD private consumption deflator (Inflation rate in local currency weighted with GNP in current US$) Project LINK - United Nations/DIESA EC raise their tariff rate by 10% (fl) Spring 1992

OECDfl b

North-America" North-America6 Developed East0 Developed East6 EEC

EEC6 Rest of OECD° Rest of OECD6

1992

1993

3.1 3.2 2.5 2.5 2.5 2.4 3.3 3.5 4.0 4.2

3.9 3.8 3.3 3.2 2.2 2.1 4.2 4.2 3.6 3.8

1994

1995

1996

Meanc

Percentage change

Notes: ° Inflation rate in Baseline. 6 Inflation rate in Alternative. c Average inflation rate for period.

3.7 3.5 3.7 3.7 2.5 2.4 3.8 3.5 4.4 4.5

3.5 3.2 3.9 3.9 2.4 2.3 3.6 3.2 4.2 4.2

3.4 3.1 4.0 3.9 2.3 2.3 3.4 3.0 3.9 3.9

3.5 3.4 3.5 3.4 2.4 2.3 3.6 3.5 4.0 4.1

"Fortress Europe" Notes to Tables 5.1 and 5.2

115

Definition of aggregates: North America: Canada, USA Developed East: Japan, Australia, New Zealand. EC: Belgium/Luxemburg, Denmark, France, Germany, Greece, Ireland, Italy, Netherlands, Portugal, Spain, UK. Rest of Industrialized: Austria, Iceland, Finland, Norway, Sweden, Switzerland, South Africa, Israel. Latin America and Caribbean: Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Mexico Paraguay, Peru, Uruguay, Venezuela, Caribbean. Africa: Algeria, Egypt, Ethiopia, Gabon, Ghana, Kenya, Libya, Morocco, Nigeria, Sudan, Tunisia, Other African Countries, African Least Developed Countries. North Africa: Algeria, Egypt, Libya, Morocco, Tunisia Sub-Sahara Africa: Ethiopia, Nigeria, Sudan, Gabon, Ghana, Kenya. Other African Countries, African Least Developed Countries. South and East Asia: Hong Kong, Indonesia, Korea, Malaysia, Philippines, Singapore, Taiwan province, Thailand, India, Pakistan, South East Least Developed, Other South Asian Countries. West Asia: Iran, Iraq, Kuwait, Saudi Arabia, West Asia Oil Importers. Mediterranean: Cyprus, Malta, Turkey, Yugoslavia. Eastern Europe and USSR: Bulgaria, Czechoslovakia, Hungary, Poland, Romania, USSR Energy export countries and regions Bolivia, Ecuador, Mexico, Peru, Venezuela, Algeria, Gabon, Libya, Nigeria, Egypt, Tunisia, Iran, Iraq, Kuwait, Saudi Arabia, OWAO, Indonesia, Malaysia, China Energy import countries and regions Argentina, Brazil, Chile, Paraguay, Uruguay, Caribbean, Colombia, Ethiopia, Ghana, Kenya, Morocco, Sudan Other African Countries, African Least Developed Countries WAOI, Hong Kong, Korea, Philippines, Singapore, Taiwan province, Thailand, India, Pakistan, South East Least Developed, Other South Asian Countries. Cyprus, Malta, Turkey, Yugoslavia. Note": Regions are defined in the respective country tables. Note*: In the trade aggregation table, the SITC 3 exports of WAOI has been moved from the Energy import group to the Energy export group.

Table 5.2A. World exports, imports and trade balances (f.o.b.): Project LINK endogenous)

- United Nations/DIESA

(Exchange rates

EC and ROW raise their tariff rate by 10%" Spring 1992

World exports World export price World exports real

1992

%chg

1993

%chg

1994

-119.26 0.07 -39.16

-3.2 1.5 -4.6

-183.13 0.10 -59.00

-4.5 2.2 -6.5

Billions of US$ -227.78 - 5 . 2 0.09 2.1 -67.48 - 7 . 1

-3.3 -4.0

-143.10 -162.15 19.05

-4.9 -5.6

-181.08 -202.77 21.69

-2.7 -4.6

-22.91 -50.76 .27.85

-3.6 -7.5

-2.7 -5.9

-18.17 -53.05 34.88

-2.9 -4.6

-15.59 -14.88 -0.71

Developed market economies Exports -89.86 Imports -108.89 Balance 19.03 North-America Exports -15.71 Imports -28.68 12.97 Balance United States Exports -12.48 Imports -30.00 17.52 Balance Developed East Exports -11.33 Imports -12.75 Balance 1.42

%chg

1995

%chg

1996

%chg

-262.26 0.07 -69.46

-5.5 1.5 -6.9

-291.60 0.06 -69.68

-5.5 1.1 -6.6

-5.8 -6.5

-210.83 -235.38 24.56

-6.2 -7.0

-231.35 -261.98 30.64

-6.2 -7.1

-25.51 -66.07 40.56

-3.7 -9.1

-26.81 -72.72 45.90

-3.5 -9.2

-30.53 -74.48 43.95

-3.6 -8.5

-3.6 -9.6

-20.33 -68.59 48.26

-3.7 -11.6

-21.54 -75.51 53.97

-3.6 -11.7

-24.60 -77.65 53.05

-3.7 -10.8

-3.8 -4.9

-17.36 -17.01 -0.35

-3.9 -5.1

-18.42 -19.04 0.62

-3.8 -5.3

-21.09 -21.08 0.00

-4.1 -5.4

Japan Exports Imports Balance EC Exports Imports Balance Germany Exports Imports Balance France Exports Imports Balance United Kingdom Exports Imports Balance Rest of industrialized Exports Imports Balance Developing countries Exports Imports Balance

-10.09 -13.08 2.99

-3.0 -5.7

-13.82 -15.34 1.52

-3.9 -6.2

-15.24 -17.34 2.10

-4.0 -6.4

-16.03 -19.08 3.04

-3.9 -6.4

-18.27 -20.86 2.60

-4.1 -6.5

-48.72 -68.37 19.65

-3.3 -4.6

-85.71 -95.90 10.20

-5.5 -6.0

-116.48 -115.92 -0.56

-7.0 -6.7

-141.37 -135.22 -6.15

-7.9 -7.4

-152.41 -154.25 1.84

-7.8 -7.7

-15.76 -40.99 25.23

-3.5 -9.9

-27.42 -40.38 12.95

-5.7 -9.1

-37.22 -45.17 7.95

-7.3 -9.4

-45.16 -53.48 8.32

-8.4 -10.7

-49.34 -62.48 13.14

-8.3 -11.3

-7.13 -8.70 1.57

-3.3 -3.9

-12.55 -25.24 12.69

-5.4 -10.4

-16.85 -32.83 15.97

-6.7 -12.4

-20.99 -35.93 14.94

-7.6 -12.5

-22.85 -38.46 15.61

-7.5 -12.1

-6.86 -1.88 -4.99

-3.5 -0.8

-11.80 -10.17 -1.63

-5.7 -4.3

-15.60 -12.41 -3.20

-7.1 -4.9

-18.13 -13.70 -4.42

-7.8 -5.1

-18.95 -14.84 -4.11

-7.3 -5.2

-14.10 0.91 -15.01

-5.3 0.3

-18.89 -0.60 -18.29

-6.5 -0.2

-21.73 -3.77 -17.96

-6.8 -1.1

-24.23 -8.41 -15.82

-6.9 -2.3

-27.32 -12.17 -15.15

-7.1 -3.0

-25.17 -11.06 -14.11

-2.8 -1.2

-35.16 -21.56 -13.60

-3.4 -2.1

-40.88 -26.11 -14.77

-3.6 -2.2

-44.54 -29.48 -15.06

-3.5 -2.3

-51.60 -33.59 -18.01

-3.6 -2.3

Table 5.2A. (cont.) 1992

%chg

1993

%chg

1994

%chg

1995

%chg

1996

%chg

Billions of US$ Latin America, Caribbean Exports Imports Balance Africa Exports Imports Balance North Africa Exports Imports Balance Sub-Sahara Africa Exports Imports Balance South, East Asia Exports Imports Balance China Exports Imports Balance

-2.8 -3.1

-4.89 -4.97 0.07

-3.4 -3.5

-5.66 -5.99 0.34

-3.7 -3.8

-6.38 -5.10 -1.28

-3.8 -3.0

-7.65 -5.08 -2.56

-4.2 -2.7

-3.13 -1.22 -1.91

-4.0 -1.3

-3.94 -2.54 -1.40

-4.6 -2.6

-4.52 -3.15 -1.37

-5.0 -3.0

-5.02 -3.87 -1.14

-5.2 -3.4

-5.72 -4.99 -0.73

-5.5 -4.1

-1.83 -0.96 -0.88

-4.9 -2.1

-2.24 -2.21 -0.03

-5.3 -4.5

-2.54 -2.57 0.03

-5.5 -4.8

-2.75 -2.90 0.15

-5.6 -5.0

-3.05 -3.62 0.57

-5.9 -5.8

-1.29 -0.26 -1.03

-3.1 -0.6

-1.69 -0.33 -1.37

-3.9 -0.7

-1.98 -0.58 -1.41

-4.4 -1.1

-2.27 -0.97 -1.30

-4.7 -1.8

-2.67 -1.38 -1.29

-5.2 -2.3

-13.24 -6.42 -6.82

-2.6 -1.2

-19.84 -10.19 -9.65

-3.4 -1.7

-23.05 -11.69 -11.36

-3.5 -1.7

-24.89 -13.83 -11.07

-3.3 -1.8

-28.56 -16.01 -12.55

-3.3 -1.8

-1.55 -0.16 -1.39

-2.4 -0.3

-2.12 -0.81 -1.31

-2.9 -1.2

-2.40 -0.95 -1.45

-3.0 -1.2

-2.63 -1.19 -1.43

-3.0 -1.4

-3.08 -1.39 -1.69

-3.2 -1.5

-3.61 -4.04

+ 0.43

West Asia Exports Imports Balance Oil Export Ctrs. Exports Imports Balance Oil Import Ctrs. Exports Imports Balance Mediterranean Exports Imports Balance Eastern Eur. and USSR. Exports Imports Balance USSR Exports Imports Balance

-1.62 -0.03 -1.59

-2.3 -0.1

-1.84 -0.04 -1.80

-2.4 -0.1

-2.37 -0.52 -1.85

-2.6 -0.7

-2.39 -1.40 -0.99

-2.4 -1.6

-2.82 -2.32 -0.50

-2.6 -2.5

-1.47 -0.11 -1.35

-2.2 -0.2

-1.64 -0.18 -1.46

-2.3 -0.3

-2.15 -0.65 -1.49

-2.6 -0.9

-2.15 -1.49 -0.66

-2.3 -2.0

-2.55 -2.38 -0.17

-2.5 -2.9

-0.16 0.08 -0.24

-2.8 1.0

-0.20 0.14 -0.34

-3.2 1.6

-0.22 0.14 -0.36

-3.4 1.5

-0.24 0.09 -0.33

-3.4 1.0

-0.27 0.05 -0.33

-3.7 0.5

-1.63 -0.11 -1.52

-5.1 -0.3

-2.11 -4.14 2.03

-6.0 -9.5

-2.43 -4.93 2.50

-6.3 -10.0

-2.79 -5.08 2.29

-6.7 -9.4

-3.26 -4.67 1.41

-7.2 -7.9

-4.23 0.60 -4.83

-3.6 0.5

-4.87 0.41 -5.28

-4.1 0.3

-5.84 0.92 -6.76

-4.6 0.7

-6.88 1.81 -8.69

-5.1 1.3

-8.66 2.39 -11.05

-5.8 1.5

-2.34 0.48 -2.82

-3.2 0.6

-2.65 1.39 -4.04

-3.6 1.8

-2.91 2.50 -5.42

-3.8 3.2

-3.40 4.12 -7.52

-4.1 4.8

-4.05 5.51 -9.56

-4.4 5.8

Table 5.2A. (cont.) 1992

%chg

1993

%chg

1994

%chg

1995

%chg

1996

%chg

Billions of US$ World (Two-Germany) World exports World export price World exports real FRG Exports Imports Balance Eastern Europe 5 Exports Imports Balance Dev. Ctrs (Excl. China) Exports Imports Balance Least dev. countries Exports Imports Balance

-122.30 0.07 -39.95

-3.2 1.5 -4.6

-188.71 0.10 -60.53

-4.5 2.2 -6.6

-235.55 0.10 -69.25

-5.2 2.0 -7.1

-271.83 0.07 -71.00

-5.5 1.5 -6.9

-302.09 0.05 -70.92

-5.6 1.0 -6.5

-17.83 -42.99 25.16

-3.5 -9.8

-30.73 -47.49 16.77

-5.7 -10.1

-41.49 -54,46 12.97

-7.3 -10.8

-50.19 -62.88 12.69

-8.5 -11.8

-54.80 -71.40 16.60

-8.4 -12.0

-1.89 0.12 -2.01

-4.4 0.3

-2.22 -0.98 -1.24

-4.9 -2.1

-2.92 -1.58 -1.34

-5.9 -3.2

-3.48 -2.31 -1.17

-6.5 -4.3

-4.61 -3.12 -1.49

-8.0 -5.3

-23.62 -10.90 -12.72

-2.8 -1.3

-33.04 -20.75 -12.29

-3.5 -2.1

-38.48 -25.16 -13.32

-3.6 -2.3

-41.92 -28.28 -13.63

-3.5 -2.3

-48.52 -32.20 -16.32

-3.6 -2.4

-0.25 -0.79 0.54

-3.5 -5.0

-0.34 -1.17 0.83

-4.3 -6.9

-0.40 -1.39 0.98

-4.7 -7.7

-0.47 -1.58 1.11

-5.0 -8.2

-0.57 -1.75 1.18

-5.5 -8.5

Net energy exp. ctrs Exports Imports Balance Net energy imp. ctrs Exports Imports Balance

-8.21 -1.50 -6.71

-2.4 -0.5

-10.34 -6.79 -3.56

-2.8 -2.0

-11.96 -8.19 -3.77

-2.9 -2.2

-12.67 -8.58 -4.09

-2.8 -2.1

-14.63 -10.28 -4.35

-3.0 -2.3

-16.95 -9.55 -7.40

-3.0 -1.5

-24.82 -14.77 -10.04

-3.8 -2.1

-28.92 -17.93 -10.99

-3.9 -2.3

-31.87 -20.90 -10.97

-3.8 -2.3

-36.97 -23.30 -13.66

-3.9 -2.3

Note? Absolute and percentage deviations.

Table 5.2B. World gross national product (1970 US$) Project LINK

- United

NationsjDIESA

fl

EC and ROW raise their tariff rate by 10% ( ) Spring 1992 1992

%chg

1993

%chg

6520.01 6488.21 -31.80

1.3 0.8 -0.49

6709.21 6615.67 -93.55

Developed market economies

4603.19 4572.25 -30.94 1933.63 1935.59 1.96 1758.03 1759.99 1.96 607.07 607.28 0.21 528.85 529.41 0.57 1871.34 1839.51 -31.82

1.9 1.2 -0.67 1.6 1.7 0.10 1.4 1.5 0.11 2.6 2.6 0.03 2.5 2.6 0.11 1.9 0.2 -1.70

4742.33 3.0 4663.80 2.0 -78.54 - 1 . 6 6 1994.38 3.1 1994.97 3.1 0.03 0.59 3.0 1810.25 2.9 1810.85 0.03 0.59 3.6 628.95 629.02 3.6 0.01 0.07 3.8 548.76 3.8 549.42 0.12 0.66 2.8 1923.14 0.5 1848.10 -75.04 - 3 . 9 0

United States

Developed East

Japan

EC

%chg

1995

Baseline/Alternative/Difference 2.9 7150.24 6918.87 3.1 7041.92 6800.19 2.8 2.0 - 1 . 3 9 -118.69 - 1 . 7 2 -108.32

World total

North-America

1994

4881.46 4774.52 -106.94 2059.88 2054.25 -5.63 1867.84 1862.21 -5.63 649.81 650.23 0.43 567.25 568.43 1.19 1969.76 1873.22 -96.54

2.9 2.4 -2.19 3.3 3.0 -0.27 3.2 2.8 -0.30 3.3 3.4 0.07 3.4 3.5 0.21 2.4 1.4 -4.90

5023.40 4929.96 -93.44 2126.86 2131.12 4.27 1926.65 1930.92 4.27 670.08 671.02 0.94 585.83 587.66 1.82 2018.54 1925.11 -93.43

%chg

1996

%chg

3.3 3.6 -1.51

7377.66 7270.38 -107.27

3.2 3.2 -1.45

2.9 3.3 -1.86 3.3 3.7 0.20 3.1 3.7 0.22 3.1 3.2 0.14 3.3 3.4 0.31 2.5 2.8 -4.63

5149.43 2.5 5060.26 2.6 -89.17 - 1 . 7 3 3.1 2192.67 2.7 2187.91 -4.75 -0.22 3.1 1985.70 1980.94 2.6 -4.75 -0.24 689.04 2.8 690.44 2.9 0.20 1.41 2.9 602.63 605.06 3.0 0.40 2.43 2055.04 1.8 1974.63 2.6 -80.41 - 3 . 9 1

Germany France

United Kingdom

Rest of industrialized

Developing countries

Latin, American, Caribbean

Africa

North Africa

Sub-Sahara Africa

South, East Asia

928.62 1.7 908.95 - 0 . 4 -19.67 - 2 . 1 2 1.3 224.36 0.4 222.17 -2.19 -0.98 2.0 145.97 0.7 144.07 -1.89 -1.30 1.5 191.16 0.8 189.88 -1.28 -0.67 1285.62 5.5 5.2 1282.43 -3.19 -0.25 330.54 3.6 330.49 3.6 -0.05 -0.01 3.6 136.53 3.0 135.75 -0.77 -0.57 4.1 74.51 3.2 73.86 -0.65 -0.87 62.02 3.1 2.8 61.90 -0.12 -0.20 5.9 387.30 5.5 385.83 -1.47 -0.38

2.7 953.68 902.72 - 0 . 7 -50.96 - 5 . 3 4 2.6 230.16 2.0 226.58 -3.58 -1.56 2.8 150.05 146.20 1.5 -3.85 -2.57 2.5 195.87 1.0 191.71 -4.16 -2.12 5.8 1360.72 5.6 1354.34 -6.38 -0.47 4.2 344.40 4.0 343.81 -0.59 -0.17 3.6 141.40 3.4 140.30 -1.10 -0.78 4.0 77.49 3.7 76.62 -0.87 -1.12 3.0 63.91 2.9 63.68 -0.23 -0.36 6.5 412.61 6.4 410.57 -2.04 -0.49

969.78 904.19 -65.59 236.65 233.03 -3.62 155.25 151.00 -4.25 202.02 196.83 -5.20 1442.67 1445.20 -7.47 361.01 359.15 -1.86 147.07 145.78 -1.29 81.09 80.09 -0.99 65.98 65.68 -0.29 439.40 437.20 -2.21

1.7 0.2 -6.76 2.8 2.8 -1.53 3.5 3.3 -2.74 3.1 2.7 -2.57 6.0 6.0 -0.52 4.8 4.5 -0.51 4.0 3.9 -0.87 4.6 4.5 -1.22 3.2 3.1 -0.44 6.5 6.5 -0.50

987.11 928.56 -58.56 243.55 240.60 -2.95 160.13 156.40 -3.73 207.93 202.71 -5.22 1528.31 1520.21 -8.10 377.82 374.66 -3.16 152.79 151.58 -1.21 84.69 83.84 -0.85 68.10 67.74 -0.36 468.27 465.82 -2.46

1.8 2.7 -5.93 2.9 3.2 -1.21 3.1 3.6 -2.33 2.9 3.0 -2.51 5.9 5.9 -0.53 4.7 4.3 -0.84 3.9 4.0 -0.79 4.4 4.7 -1.00 3.2 3.1 -0.53 6.6 6.5 -0.52

1.1 998.32 2.6 952.88 -45.44 - 4 . 5 5 2.6 249.76 247.78 3.0 -1.98 -0.79 2.7 164.51 3.4 161.67 -2.84 -1.73 2.3 212.69 2.3 207.27 -5.41 -2.54 5.8 1616.75 5.8 1607.83 -8.91 -0.55 4.6 395.02 4.3 390.64 -4.38 -1.11 159.42 4.3 4.2 157.95 -1.47 -0.92 5.1 88.99 4.9 87.94 -1.06 -1.19 3.4 70.42 3.4 70.01 -0.41 -0.59 6.2 497.44 6.2 494.57 -2.87 -0.58

Table 5.2B. (cont.) 1992 China

West Asia

Oil Export Ctr.

Oil Import Ctr.

Mediterranean

Eastern Europe and USSR

USSR

%chg

8.2 355.44 8.1 355.20 -0.24 -0.07 7.8 21.35 7.1 21.22 -0.13 -0.59 9.5 16.83 8.7 16.70 -0.13 -0.74 1.7 4.52 1.7 4.52 0.00 0.00 0.7 54.47 53.94 - 0 . 3 -0.52 -0.96 631.20 - 9 . 5 633.53 - 9 . 2 0.37 2.33 494.93 - 1 1 . 0 497.83 - 1 0 . 5 0.59 2.90

1993

%chg

6.9 380.04 7.1 380.40 0.10 0.36 8.2 23.10 8.0 22.92 -0.19 -0.80 9.8 18.48 9.5 18.30 -0.19 -1.00 2.2 4.62 2.2 4.62 0.00 0.00 8.6 59.17 4.5 56.35 -2.82 -4.77 606.17 - 4 . 0 597.53 - 5 . 7 -8.64 -1.42 468.04 - 5 . 4 460.71 - 7 . 5 -7.34 -1.57

1994 Billions 406.68 407.86 1.18 25.01 24.83 -0.19 20.29 20.10 -0.19 4.73 4.73 0.00 63.50 60.39 -3.11 594.75 590.47 -4.28 452.57 450.44 -2.13

%chg of US$ 7.0 7.2 0.29 8.3 8.3 -0.76 9.8 9.8 -0.93 2.4 2.4 0.00 7.3 7.2 -4.90 -1.9 -1.2 -0.72 -3.3 -2.2 -0.47

1995

%chg

7.0 435.18 7.2 437.26 0.48 2.08 6.2 26.56 6.3 26.39 -0.17 -0.64 7.1 21.72 7.2 21.55 -0.17 -0.79 2.4 4.84 2.4 4.84 0.00 0.00 6.6 67.68 6.8 64.49 -3.19 -4.71 0.6 598.53 0.2 591.76 -6.77 -1.13 450.36 - 0 . 5 446.31 - 0 . 9 -4.05 -0.90

1996 465.52 468.66 3.15 27.75 27.57 -0.19 22.77 22.59 -0.19 4.98 4.98 0.00 71.60 68.45 -3.15 611.49 602.30 -9.19 456.06 450.66 -5.38

%chg 7.0 7.2 0.68 4.5 4.4 -0.67 4.8 4.8 -0.82 2.8 2.8 0.00 5.8 6.1 -4.40 2.2 1.8 -1.50 1.3 1.0 -1.18

FRG Eastern Europe 5

Dev. Ctr. (excl. China)

Least Dev. Ctr.

Net energy exp. ctrs

Net energy imp. ctrs

885.17 864.82 -20.35 136.27 135.70 -0.57 930.18 927.24 -2.94 25.10 24.97 -0.13 623.38 622.87 -0.51 662.24 659.57 -2.68

1.4 -0.9 -2.30 -3.6 -4.0 -0.42 4.5 4.1 -0.32 3.6 3.1 -0.54 6.8 6.7 -0.08 4.3 3.8 -0.40

Note" Absolute and percentage deviations.

906.49 855.83 -50.66 138.13 136.83 -1.30 980.68 973.94 -6.74 25.97 25.70 -0.27 661.58 661.48 -0.11 699.13 692.87 -6.27

2.4 -1.0 -5.59 1.4 0.8 -0.94 5.4 5.0 -0.69 3.4 2.9 -1.02 6.1 6.2 -0.02 5.6 5.0 -0.90

919.04 853.50 -65.54 142.18 140.03 -2.15 1035.99 1027.34 -8.65 26.83 26.51 -0.32 703.68 704.54 0.86 738.99 730.66 -8.33

1.4 -0.3 -7.13 2.9 2.3 -1.51 5.6 5.5 -0.83 3.3 3.1 -1.21 6.4 6.5 0.12 5.7 5.5 -1.13

1.4 931.73 2.3 873.48 -58.25 - 6 . 2 5 4.2 148.17 3.9 145.45 -2.73 -1.84 5.5 1093.13 5.4 1082.94 -10.18 - 0 . 9 3 3.4 27.73 27.37 3.3 -0.36 -1.31 6.2 747.47 749.27 6.3 0.24 1.80 5.7 780.84 5.5 770.94 -9.90 -1.27

938.73 0.8 893.69 2.3 -45.04 - 4 . 8 0 4.9 155.44 4.3 151.63 -3.81 -2.45 5.3 1151.23 5.2 1139.17 -12.06 - 1 . 0 5 3.4 28.66 28.27 3.3 -0.40 -1.38 6.4 795.16 797.62 6.5 0.31 2.46 5.2 821.59 810.22 5.1 -11.37 - 1 . 3 8

126

L.R. KLEIN AND PINGFAN HONG

Table 5.2C. OECD unemployment rate: Project LINK- United Nations/ DIESA EC and ROW raise their tariff rate by 10% 0 Spring 1992. 1992

1993

1994

7.6 7.7 7.8 7.8 3.5 3.6 9.4 9.7

7.4 7.8 7.5 7.5 3.4 3.5 9.3

7.2 7.8 6.9 7.1 3.5 3.7 9.2

10.4 12.4 12.4

10.8 11.5 11.7

1996

Meanc

6.9 7.4 6.3 6.3 3.8 3.9 9.0

7.0 7.4 6.2 6.2 4.2 4.2 9.3

7.2 7.6 7.0 7.0 3.7 3.8 9.3

10.7 10.8 11.1

10.5 10.3 10.6

10.4 11.5 11.7

1995 Percent

OECD° b

North-America0 North-America6 Developed East0 Developed East6 EEC

EEC" Rest of OECD° Rest of OECD*

12.5 12.6

Notes: ° Unemployment rate in Baseline. 6 Unemployment rate in Alternative. c Average unemployment rate for period. Excl. Greece, Iceland, and Switzerland. Table 5.2D. OECD private consumption deflator (Inflation rate in local currency weighted with GNP in current US$) Project LINK - United Nations/DIESA EC and ROW raise their tariff rate by 10% (°) Spring 1992

OECDfl b

North-America" North-America* Developed East0 Developed East6 EEC

EEC6 Rest of OECD0 Rest of OECD6

1992

1993

3.1 3.3 2.5 2.8 2.5 2.7 3.3

3.9 3.9 3.3 3.7 2.2 2.6 4.2 4.1 3.6 4.5

1994

1995

1996

Meanc

Percentage change

3.5 4.0 5.6

Notes: a Inflation rate in Baseline. 6 Inflation rate in Alternative. c Average inflation rate for period.

3.7 3.3 3.7 3.8 2.5 2.6 3.8 3.3 4.4 4.6

3.5 3.1 3.9 3.9 2.4 2.3 3.6 3.0 4.2 4.1

3.4 3.1 4.0 4.1 2.3 2.2 3.4 3.0 3.9 3.8

3.5 3.4 3.5 3.7 2.4 2.5 3.6 3.4 4.0 4.5

"Fortress Europe"

127

done. The EEC no longer shows a small gain for exports; everyone seems to lose on both the export and import side if retaliation takes place. Developing country imports are particularly hurt in the retaliatory case. They lose imports, which are vital for their growth process. World production falls, eventually, by much more than 1 percent below the base-case values (where no new protectionism takes place). The feedback effect on the EEC (fortress Europe followed by ROW retaliation) brings about an estimated decline in production, while Japan appears to come out on top, by gaining a bit in production. North American production is set back by only a small amount. Developing countries, however, would suffer. Quite large increases in unemployment, are estimated to follow the decline in EEC output, but North America and Japan have very small increments. As for inflation, it might be only a bit higher at the beginning of the policy changes, but recedes as output gains are reduced. References Klein, L.R. and V. Su (1979), "Protectionism: An Analysis from Project LINK," Journal of Policy Modeling, 1 (January), 5-35. Klein, L.R., P. Pauly, and C.E. Petersen (1987), "Empirical Aspects of Protectionism: Results from Project Link," in D. Salvatore (ed.), The New Protectionist Threat to World Welfare, New York: North-Holland, pp. 69-94.

PART

ii

Trade theory, industrial policies, and protectionism

CHAPTER

6

US response to foreign industrial policies RICHARD N. COOPER

US industry in the 1980s felt under siege from foreign competitors, and pointed to rising shares of imports in their home market and falling shares of US exports in world markets. Attention was drawn to a host of foreign practices that apparently help to explain the increased competition from foreigners, ranging from specific export promotional tactics and specific import prohibitions to broadly drawn industrial policies and industrial targeting that allegedly provided impetus to foreign exports and simultaneously discouraged imports from the US and elsewhere. These practices in one variant or another were discovered to be widespread, being used not only by other industrial countries, but by less developed countries as well, particularly the newly industrialized countries such as Korea and Brazil. But Japan was held up as the main culprit, not so much because its practices were more extensive than those in other countries, as we will see, but because they were somewhat mysterious and lost behind Japanese reticence and linguistic ambiguity, and above all because Japan emerged as the most successful competitor in a number of industries in which Americans had hitherto considered themselves unrivaled. It is foreign success rather than the practices themselves - which in many cases existed for many years and in the case of Japan actually diminished in importance - which gave rise to such widespread concern, and led to calls for US action ranging from retaliation to emulation. This essay addresses these American concerns about what have come to be called "unfair trade practices" from the perspective of public policy. What response should the US government make to these foreign practices 131

132

RICHARD N. COOPER

and their alleged impact on the foreign trade - and even the industrial viability - of the United States? With respect to the future viability of American industry, the view has been expressed that the theory of comparative advantage which provides the intellectual underpinnings for a liberal trade policy is fundamentally misleading, since it takes comparative advantage as given, whereas in fact it is determined by government policy. In particular, it is claimed that the country that gets the head start in a period of rapid technical change is likely to develop a "comparative advantage" in the product in question (see, e.g., Borrus etal.y 1986). Government actions that lead to early development thus determine comparative advantage. This chapter will take up the objectives and instruments of industrial policy in several other countries, focusing on Japan. It will then inquire into US governmental influences on the American economy and its structure of trade, which on close inspection turn out to be more pervasive than most Americans think, and with some unexpected twists. The chapter then addresses the "making" of comparative advantage by government policy, and finally turns to various alternative approaches to US policy for dealing with the pervasive influence of governments on the composition of output and trade, both in a cooperative framework and by acting on its own. Macroeconomic context Before we turn to these matters, it is important to address a fundamental analytical error that underlies much of even informed discussion with respect to so-called unfair trade practices and other policy distortions to international trade. That is the supposition that the overall trade balance, and lying behind that economic growth and employment, are determined or strongly influenced by the foreign practices that are typically singled out for admiration or condemnation, i.e., by policies or practices that limit particular imports or promote particular exports. This error is an example of the "fallacy of composition," whereby summing up a number of separate actions each with well-defined effect - e.g., enlarging the Japanese import quota on citrus will surely lead to larger Japanese imports of citrus products - will not in fact have the expected overall effect. Concretely, the Japanese could remove every policy measure and alter every practice foreigners point to as restricting imports, and the consequences would not involve a reduction in the large Japanese trade surplus, and might not even result in an overall increase in Japanese imports. The reason is that a country's balance of trade in goods and services must always equal the difference between national saving and national investment, and it is only

US response to foreign industrial policies

133

insofar as actions influence savings and investment that they can influence the trade balance: X-M=(S+(TG))-I Here the difference between exports (X) and imports (M) equals net foreign investment (unilateral transfers are neglected), and net foreign investment in turn must equal the difference between national saving ( = private-sector saving (S) plus the government operating surplus (T—G)) and national investment (I). This relationship is an accounting identity, from the definition of the concepts in the national accounts; decisions that affect one of these aggregates automatically must affect one or more others in a consistent way. This relationship is of fundamental importance in assessing overall trade performance. While lifting the Japanese citrus quota will almost certainly raise Japan's imports of citrus, and will probably raise overall Japanese imports (but by less than the increase in citrus imports), it would not reduce the large Japanese trade surplus except insofar as lifting the quota affected national savings and/or investment, i.e., probably not at all, or only negligibly. Given a large savings-investment imbalance, a rise in one particular import will be offset by a decline in other imports and/or a rise in exports. Paradoxically, it is even conceivable that a stripping away of all Japan's protective practices1 would lead to a rise in the trade surplus rather than a decline, if the sudden import competition reduced Japanese investment more than it reduced saving; but since "saving" includes retained corporate earnings, it could equally plausibly reduce the trade surplus by reducing corporate profits but at the same time stimulating corporate investment, in order to survive in the new, more competitive environment. The point is that to assess the overall impact one must look beyond the commodities immediately affected by the change in policy, whether it be import restriction or export promotion. The significance of this observation in the present context is that from 1981 to 1987 the United States had a rapidly growing trade deficit, whereas Japan and several European countries developed large surpluses. The development of these overall imbalances had a strong influence on the rise in import shares in the US market and the decline in US export shares mentioned at the outset, and correspondingly on the competitive pressures felt by American firms. But they arose from overall macroeconomic developments, not from "unfair trade practices," however many of those there might be. It is of course unrealistic to expect industries affected by severe foreign competition at home or abroad to sort out the macroeconomic influences from the specific foreign (or domestic) policies

134

RICHARD N. COOPER

affecting the industry or commodity in question. But for purposes of addressing policy it is essential to make this distinction; failure to do so is bound to lead to frustration, since even a vigorous use of the wrong instruments will not achieve the desired objective.2 In 1984 the United States ran a current-account deficit of $95 billion, or 2.6 percent of GNP, a record high up to that point (it reached 3.4 percent in 1987, the peak). Gross private savings were 18.4 percent of GNP, also a record high since 1945. Gross investment was 17.4 percent of GNP, typical for a boom year. But the deficit on governmental account (including state and local governments) was 3.4 percent of GNP, down from 1983 but nearly an historic high, excluding the Second World War. Under these circumstances, the United States was bound to have a large trade deficit. Paradoxically, additional investment stimulated by additional tax concessions to investment, frequently urged by the business community as a means to become more competitive internationally, would actually worsen the trade balance during the period of investment, both by increasing investment and by increasing the budget deficit. An analogous development occurred in Japan. Private savings and investment rates had been considerably higher in Japan than in the United States over the past three decades, and were on average roughly equal to one another until 1975. Investment fell sharply then, and has remained low by Japanese standards since then. Japan greatly extended its social welfare programs in the early 1970s, and that combined with an economic slowdown produced large government deficits, reaching over 4 percent of GNP in 1979. Japanese policy was to reduce the budget deficit in the early 1980s; but private savings remained in excess of investment, so a large current-account surplus emerged in 1981; by 1986 it had reached 4.4 percent of GNP, receding thereafter. How can these movements in savings and investment, inclusive of government, get translated into trade surpluses or deficits? A simple version of the answer is that the excess private savings of Japan, to the extent it could not be satisfied by claims on government (resulting from the budget deficit), sought investment abroad. This in turn put downward pressure on the yen, which depreciated until 1985 and made Japanese goods more competitive in world markets. In the United States, in contrast, a combination of government deficit and private investment in excess of private savings strengthened interest rates and thereby attracted capital from abroad, thus leading to appreciation of the dollar and worsening the competitive position of American products both at home and abroad. So the main mechanism of equilibration was the exchange rate. Under a system of fixed exchange rates this pressure of a desired increase in net private savings cannot weaken the currency, so it reduces income (causes a

US response to foreign industrial policies

135

recession) to the point at which the reduced savings, the increased budget deficit (due to lower tax revenue), and an improvement in the currentaccount balance (due to lower imports) preserve the equality between net national savings and net foreign investment. The recession may also put downward pressure on wages and prices, and improve international competitiveness in that way. Either way, the link between the overall trade position and overall savings-investment behavior is inescapable, and no fiddling around with particular trade policies or practices will get around it. For instance, the occasional call for a US import surcharge would improve the trade balance mainly through the increased revenue to the government (reducing the dissaving by the government), and the increased investment that some proponents want it to stimulate would undermine even that improvement if the investment were not accompanied by higher private savings. Of course, from the perspective of an individual industry these macroconomic developments seem unimportant relative to the industryspecific policies of countries; and it is true that such policies can strongly influence the commodity structure and geographic direction of trade, if not the overall balance. But the industry perspective is myopic; in assessing the impact of industrial policy on trade as a whole, it is crucial to take into account the macroeconomic environment. Industrial policy in Japan and elsewhere "Industrial policy" has no well-defined boundaries, and at its broadest we can take it to mean any government policies that affect the structure of output, or, slightly more narrowly, any government policies whose intended purpose is to affect the structure of output. But perhaps the best way to characterize industrial policy of the type that concerns many Americans is to discuss briefly but concretely the policies of the country that is seen to be most threatening, i.e., Japan. This task is somewhat more difficult than it might seem, since Japanese policies have changed substantially over the past forty years, and some of the measures most frequently cited in fact belong largely to the past. So Japanese industrial policy will be described here in the three phases frequently cited by Japanese analysts: end of post-war price control (1952) to the early 1960s, aimed at establishing the base for a modern industrial economy; 1960—72, following elimination of foreign exchange controls, aimed at excelling in exports; and 1972 to the present, aimed at shifting the structure of the Japanese economy from capital-intensive to knowledgeintensive activities.3 The dividing lines between these periods are of course somewhat artificial, and some continuity of both policies and problems can

136

RICHARD N. COOPER

be found during the past forty years, but over the period substantial changes in both policy orientation and instruments of policy occurred. The fifties In the early 1950s, the Japanese desired to build a modern industrial society, drawing insofar as possible on the strengths they already had, which included a skilled workforce in steel-making, shipbuilding, and optics (for binoculars) built up during the Second World War. Strong government guidance in modernization had been part of Japanese history during the previous eighty years, and it was turned to again. Crucial industries were identified, forecasts of demand were made, and the industries were encouraged to invest accordingly. A rationalization program was begun for steel in 1951. Special industry promotion laws were passed for synthetic fibers (1953), petrochemicals (1955), machinery (1956), synthetic rubber (1957), electronics (1957), and aircraft (1958). Assembly of automobiles was made a priority in Japan in 1952 (Warnecke, 1978, p. 127). The Occupation-imposed Anti-monopoly Act was amended in 1953 to permit the formation of "recession" and "rationalization" cartels with government approval in order to prevent cut-throat price competition and to induce orderly reduction in capacity during periods of slack demand (such as immediately following the Korean War, when the revisions were passed). The Export and Import Trading Act was also passed in 1953, permitting cartels to fix prices and to limit imports. By 1971 there were 195 legally recognized cartels under the Export-Import Trading Act, thirteen under the Anti-monopoly Act, and twenty-three under separate legislation pertaining to the machinery, electronics, and fertilizer industries (Yamamura, 1982, p. 82), a decline in number by about 20 percent from the mid 1960s (Caves and Uekusa, 1976, p. 148). It was during this period too that the Fiscal Investment and Loan Program (FILP) was created, whereby postal savings accounts (historically an important depository for household savings) and public pension reserve funds were channeled into a series of trust funds (about forty in the mid 1980s) for the promotion of public policy. These trust funds include the Japan Development Bank (JDB) and the Export-Import Bank of Japan. Investment by the favored industries was encouraged by special tax incentives and by loans from JDB. Japanese industry drew 13 percent of its external financing from JDB in 1952-5, and another 15 percent from other FILP programs (these figures had declined to 4 and 10 percent, respectively, by 1971-5). Four key selected industries - electric power, shipping, coal, and steel - got 24 percent of their external financing from JDB, and another 13 percent from other FILP programs in 1952—5 (Noguchi in Yamamura, 1982, p. 131). So this government finance was a key instrument of policy; moreover, a JDB loan often provided a signal for lending (at commercial

US response to foreign industrial policies

137

terms) by the quantitatively more important private banks, on the grounds that the favored firms were likely to involve lower risk than other business loans. Finally, foreign exchange control existed throughout this first period, and import licenses were used to further the industrial policy. For instance, foreign cars were more durable and commodious than domestic cars, and as the taxi industry revived and thrived its demand rose for foreign cars (including used cars from American forces resident in Japan). Purchase of such cars was limited under foreign exchange regulations in 1951, imports were liberalized in 1952, but sharply tightened again in early 1954 and a "buy domestic" campaign was started with the taxi companies. Japanese production with improved quality required foreign technology. MITI in 1952 promulgated guidelines for auto assembly licensing agreements which inter alia stipulated that after a specified period of time eleven key auto parts had to be produced in Japan (Yakushiji in Aoki, 1984, pp. 278-81). Foreign exchange regulations were used to limit imports in many other industries as well, and to shape the development of each favored industry. The Sixties Although Japan recognized the importance of exports from as early as 1949, when export promotion was adopted as MITFs main goal, the 1950s can be characterized principally as a period of import substitution, as Japan began to produce more sophisticated products that permitted a reduction in imports of those products. In 1960 there was a major trade and foreign exchange liberalization, associated with pressure from the United States and Europe to move toward currency convertibility and to accept fully the obligations of the General Agreement on Tariffs and Trade (GATT), which Japan had joined in 1955, preparatory to Japan's admission in 1964 to the Organization of Economic Cooperation and Development (OECD), the club of industrialized countries. This move eliminated exchange controls and general use of overt import restrictions as a major instrument of industrial policy, although some approved cartels continued to limit imports, presumably with MITI knowledge. The relative importance of the JDB also declined sharply, as the private banks became both stronger and more assertive. A consequence of the liberalization of foreign exchange regulations and of the growing strength of sources of domestic credit in Japan is that MITI had to rely more on moral suasion, less on directives. It did not always work. An effort to pass a new law supporting selected industries unexpectedly failed due to domestic opposition in the early 1960s, and in 1965-6 the Sumitomo steel company flouted MITFs administrative guidance to cut back steel production. The growth of independent banks inhibited the development of "national champions" in Japan, since each

138

RICHARD N. COOPER

bank wanted to have within its "family" of firms a representative from each major industry. The degree of competitiveness among major Japanese banks and firms, and the difficulty it sometimes poses for government, was already encountered in 1955 when a MITI plan to create a single small, inexpensive People's car (along the line of Volkswagen) to open the mass market was leaked to the press (Japan's press is more competitive and more aggressive than that in the United States) and created a storm of protest from the actual and would-be auto makers. The plan was abandoned and vigorous competition developed among Japan's auto makers. A general policy can continue in the face of even major exceptions and derogations. But it is also likely that such exceptions remind the officials, if they need reminding, that there are distinct limits to their authority, and they therefore influence what MITI calls for, and what industry calls on MITI to call for if the industry is not unanimous. It is not true, as foreigners are sometimes led to believe, that Japanese business leaders are unwilling to take risks on their own and to stand out from the crowd. By the 1960s steel and shipbuilding were commercial successes, and relative emphasis shifted to encouragement of the machine tool industry, which, like steel, and on the basis of close observation of the American economy, was seen to be a prerequisite for a modern industrial economy. It continued to get tax breaks, modest subsidies, and favored procurement. The Seventies In the early 1970s there was a marked shift in Japanese policy in a number of respects. The Japanese public had become restless about growing pollution, about the inadequate welfare system, and even about the fact that they had to pay much higher prices for some Japanese products than Japan's overseas customers did (television sets, especially, became a local cause celebre). Moreover, MITI and other officials became concerned about the rapid growth of Japan's imports of oil, even before the oil shock of late 1973. In response to general public pressure, Japan adopted in 1972 a much more generous social security system and introduced stiff pollution standards. In response to the dollar crisis of 1971-2, Japan took a number of steps to liberalize imports, to liberalize direct investment inflows, and to monitor exports with a view to restraining too rapid growth. Japan also eliminated its "buy only Japanese" policy with respect to government procurement, and loans under FILP were made subject to Diet approval, something that had not been required before 1972, and which greatly reduced their flexibility as an instrument of industrial policy. Looking to the future, MITI emphasized the growing importance of "knowledge-intensive" industries and encouraged Japanese industry to move in that direction. The Agency for Industrial Science and Technology (AIST) was created within MITI to finance research projects in ceramics,

US response to foreign industrial policies

139

computers, seabed mining, and flexible computer-aided manufacturing systems. The criteria for AIST support are that the item in question is not yet on the commercial market and that the research effort would be too large or too risky for private firms to undertake alone. An early success was a desalination process that was later commercialized by private firms and sold in the Middle East. MITI in 1983 had a total R & D Budget of about $250 million, which was divided between MITI's own fifteen laboratories and support for R & D by private firms, compared with about $10 billion in total spent annually by Japanese firms on research and development. In 1989 total Japanese government expenditures on all R & D amounted to $13.5 billion about one-fifth the comparable figure for the United States, but the Japanese government financed only 1.5 percent of business enterprise expenditures on R & D, compared with one-third in the United States (OECD, p. 53). The most heavily publicized cases of government R & D support concern the VLSI (very large-scale integrated circuit) project started in 1976, on which some $120 million in conditional loans (repayable only if there was commercial success) were made by the government between 1976 and 19794; and government support of a so-called fifth generation computer starting in 1982. But there are other sources of support of high-technology industries than research grants and conditional loans. After 1978 the National Aeronautics and Space Development Agency gave preference in its procurement to satellites with high local content, which came to exceed 60 percent. However, Japan's semiconductor market "can be said today [1984] to be completely open to American-owned companies," (US-Japan Trade Study Group, 1984, p. 54), a contention that was challenged by the US Semi-Conductor Industry Association, which managed to persuade the US Government to extract from Japan an objective of 20 percent foreign sales (see Tyson, 1992, chapter 3). The Eighties Japan does not only help the products of the future. Since 1978 it has had a program to "restructure" Japanese industries that are depressed, for whatever reason. Firms accounting for two-thirds of a depressed industry's output can petition the government for a restructuring plan, which involves an agreed reduction in capacity, with loan guarantees and tax benefits accruing to firms that scrap capacity under the plan. In mid 1984 there were twenty-two officially designated industries, of which five (paper, ethylene, compound fertilizer, polyolefins, and PVC resin) had formed legal cartels to restrict output and price competition (US-Japan Trade Study Group, 1984, p. 64). This represented a substantial decline in the use of cartels from twenty years earlier. In addition, FILP support for Japanese industry became very much less (proportionately) than it used to

140

RICHARD N. COOPER

be, and the government loan rate differential below market rates dropped from around 3 percent in the early 1960s to about 1 percent in the early 1980s (Noguchi in Yamamura, 1982, p. 137). Nonetheless, the practice of making concessional loans to distressed firms, especially small business, continued into the late 1980s. In general, however, Japanese government involvement in determining industrial structure is far less than it once was in part because the two instruments of foreign exchange, licensing and credit control are no longer available for disciplining large firms. But it continues to provide hortatory guidance in MITI's "Visions" and other government pronouncements, and to back these up with direct or indirect funding on a modest scale, and with directed government procurement (see Okimoto, 1989). Europe Other countries have industrial policies which are similar to those in Japan, and generally more extensive. For example, Britain, France, and Italy all have had organizations analogous to JDB, that channel publicly raised funds to private enterprises. All provide tax breaks to encourage investment in general and favored investment (by industry and by region) in particular. All had extensive state-owned industrial enterprises, something that is rare in Japan, that receive periodic infusions of new "equity" capital, which is difficult to distinguish from subsidies when the firms are running operating losses. (Britain privatized many of these state-owned enterprises during the 1980s.) All have given extensive support to their steel and textile industries, to consolidate operations and to scrap obsolete capital. Britain, France, and Germany have also provided government support to cushion declining demand in their shipbuilding industries. In the high-technology area, all have provided extensive government funding and preferred government procurement in aerospace, computers, and telecommunications (for a survey of actions, see Carmoy in Warnecke, 1978). For instance, national Post, Telephone, and Telegraph (PTT) organizations in these countries rarely procure foreign-made equipment; the same is true of national power companies with respect to heavy electrical generating equipment. During the 1980s, European policies shifted increasingly toward the high-technology sectors, became more generic in nature, and more and more were encouraged on a Community-wide level, as opposed to uncoordinated national actions. On the one hand, the European Commission took an aggressive stance in the late 1980s against state subsidies to particular industries, as providing unfair competition to other European firms in a single European market. On the other hand, inter-European cooperative efforts were encouraged (with public financial support) in semi-conductors, high definition television, biotechnology, and automatic car navigation systems. Public efforts were also made to foster university-

US response to foreign industrial policies

141

business collaboration, and to disseminate new ideas more rapidly (see Pavitt and Sharp, 1992, and Tyson, 1992, chapter 6). Newly industrializing countries also have adopted strong industrial policies, apart from the traditional technique of restricting imports that compete with the production of favored industries. Korea at least in one area adopted a novel incentive: Koreans who worked on contract engineering and construction projects in the Middle East in the 1970s allegedly were exempt from the draft and given priority in public housing when they returned home (Warnecke, 1978, p. 133) Observations on industrial policies The purpose of this sketchy survey is to illustrate how widespread are the uses of "industrial policies" and how diverse are the instruments of support that governments may use. They raise the question of how can such extensive interference with market forces be reconciled with a liberal trading system predicated on the mutual gains that flow from reliance on comparative advantage to determine each country's structure of output and trade, with certain acknowledged exceptions having mainly to do with national security. In particular, the discovery that foreign governments intervene extensively in national economic development even at the sectoral level, and that these interventions may impinge on the market for American products both at home and abroad, led in the 1980s to a number of stated or implied recommendations for US policy (see, e.g., LICIT, 1983 and Magaziner and Reich, 1983). The US government, it was suggested, should: 1 2 3 4

insist that other countries give up the practices which allegedly represent "unfair competition" for American firms; adopt measures similar to foreign actions that are deemed to have been successful, e.g., create a revived Reconstruction Finance Corporation modeled on the Japan Development Fund; raise import barriers to goods enjoying "unfair competition," sometimes only after (1) has failed, or as a threat to encourage foreign compliance with (1); and match the foreign competitive measures in third country markets (e.g., through generous Exim Bank financing) and/or restrict unrelated imports into the United States from countries engaging in unfair competition in third markets.

Before we turn to policy issues for the United States, however, several remarks should be made about the overall economic impact of these foreign industrial policies.

142

RICHARD N. COOPER

First, most of them are not new, and have been around for a long time, so whatever real stimulus they provide should have been adjusted to by adaptation elsewhere in the economic system. France, Italy, and Japan among major countries began extensive industrial policies in the early 1950s. Britain launched on this course in earnest in the mid 1960s. Indeed, if anything, the impact of industrial policies has declined sharply in Japan in recent years, and the Thatcher government in Britain sharply reduced that country's sectoral policies in the 1980s. Second, in many cases a costly quid pro quo is exacted for government support of a firm or industry, especially in terms of loss of freedom or flexibility over decisions regarding employment, plant closings, new investment, and diversification. This loss of freedom and flexibility is directly or indirectly cost increasing. Such factors are typically not allowed in the reckoning as "negative" unfair competition, a partial offset to the aids that are received. Third, it is often forgotten that the United States, while not having an industrial policy as such, nonetheless has many measures which directly or indirectly assist American business. These are sufficiently extensive and variegated to warrant extensive treatment in a moment. Fourth, it should not be forgotten that measures which discriminate in favor of certain firms or industries by that very fact discriminate against other firms or industries. In economics there are only rarely opportunities for a "free lunch," and someone has to pay for the special aids that are granted to others. This is an important analytical point and will be discussed further below. The United States government has had and continues to have extensive sectoral involvement in the US economy. For example, in the nineteenth century it gave land grants to the railroads and more recently it has constructed a vast highway system to open up areas of the country and provide cheap inland transport. It provides cheap, underpriced water to irrigate Arizonan and Californian farms, producing the citrus which growers complain about being unable to sell "fairly" to the Japanese and Europeans. It has extensive research and development programs by the Department of Defense and by the National Aeronautics and Space Administration which have generated commercially valuable spin-offs such as jet engines, helicopters, and the Boeing 747 (see Nelson and Winter, 1982). It is true that Europeans and Japanese exaggerate the quantitative importance of the commercial impact of defense R & D; but most American analysis of foreign government policies is also qualitative, leaving the impression of greater quantitative importance than is generally warranted. Extensive charitable deductions under the US tax system permit the United States to provide higher education to a much larger percentage of the labor

US response to foreign industrial policies

143

force than the Europeans could afford with their tax supported systems and in "unfair competition" with the private universities of Japan. It is perhaps useful to address more systematically government subsidies and other policies that influence American exports, since the US government engages in a host of actions that influence the competitiveness of American exports, ranging from direct actions to encourage exports through activities to stimulate production and general support for business activity to actions which by discouraging certain industries lead indirectly to encouragement of others. The following list proceeds from the most direct form of export encouragement to less direct forms (this list is adapted from Cooper in Warnecke, 1978). 1 Economic and military assistance to less developed countries, tied to the procurement of American goods. In this case the US government in effect buys the American goods and gives them away, or lends them on very easy terms. Foreign aid represents an extreme form of export subsidization, but it is accepted as contributing to economic development or national security, and the importing countries in this case would be unlikely to hold the United States accountable for unfair import competition or to impose countervailing duties. But third world countries may lose export orders because of foreign aid shipments tied to US procurement. The subsidization of American exports would cease if foreign aid grants and loans were freely usable for the purchase of goods and services anywhere, as is the case with loans from the World Bank. US foreign assistance and military credit sales amounted to about AVi percent of total US exports of goods and services in the mid 1980s. 2 Under US tax laws until 1984, corporations that derived at least 95 percent of their gross receipts from exports could qualify as domestic international sales corporations (DISC) and could defer payment of corporate profits tax until dividends were remitted to the parent corporation. This provision, which cost in excess of $1 billion in annual revenue foregone in the mid 1970s, amounted to an interest-free loan from the government for expenditures involved in the promotion of exports. The subsidy element - about $60 million a year - was thought to be much less than the foregone revenues, since the taxes would have eventually to be paid. In fact, in the 1984 Tax Act which eliminated the DISC and permitted in its place the Foreign Trade Corporation, many of the unpaid taxes were waived, so the interest-free loan turned out ex post to be a direct subsidy. 3 The government subsidized both the construction and operation of merchant vessels under US registry. Construction subsidies do not increase

144

RICHARD N. COOPER

exports since such subsidies are available only to purchases by US flag companies, but of course operating subsidies to shipping make it easier for Americans to export shipping services. Both programs are of longstanding, and in the mid 1970s the operating subsidy amounted to around $200 million a year. Ship construction and shipping services generally involve heavy government involvement throughout the world. 4 The Export-Import Bank provides medium-term credit for American exports. For a number of years the interest rates were below market rates, so a direct subsidy was involved. In the 1980s, the Bank has tried to keep its lending rates above its borrowing rates by enough to cover its operating costs, except when necessary to meet foreign competition. The subsidy to American exports is thus the more subtle (and smaller) one which arises from the use of US government credit in borrowing in the capital market plus the absence of a requirement to pay dividends on the Bank's capital. 5 Until 1973 the Commodity Credit Corporation (CCC) gave substantial subsidies to US exports of many agricultural products, the counterpart of a system of high domestic price supports combined with the view that in the absence of agricultural policy the United States would be a substantial exporter of agricultural products, especially grains, cotton, tobacco, etc. The high price supports stimulated output, so the program also involved limitations on acreage. This system was reinstituted on a lesser scale in the early 1980s. It is difficult to say whether agricultural exports would be larger or smaller than they would have been in the absence of the government support program, since the support prices and the acreage controls could be expected to have opposite effects on farm production. 6 For a number of years investment in plant and equipment in the United States enjoyed a 10 to 25 percent investment tax credit. The credit in effect lowered the cost of domestic investment by that amount, and thus stimulated the productive capacity of the economy. The credit operated for all investment, however, so it is not obvious whether on balance exports or imports were stimulated more by the tax credit. The first round effects of increased production and income could go either way. The major long-run effect of the investment tax credit was to make American industry somewhat more capital intensive than it would be without the credit, both in each industry taken separately and in its overall industrial structure. In addition, from 1981 to 1988 there was extremely generous write-off provisions for the depreciation of new investment. These provisions had a similar effect to the investment tax credit. The investment tax credit was eliminated in 1987, but a variant is likely to be reinstituted in 1993.

US response to foreign industrial policies

145

Depletion allowances for oil and other minerals have the effect of stimulating domestic production of such products and thus serve to reduce imports or to increase exports of these products. Until 1975 this tax privilege was available to American-owned mineral investment anywhere in the world, but thereafter it was limited to production in the United States, and it was further restricted in the 1986 Tax Act. 7 Direct government expenditure also often supports business enterprise. Examples are federal spending, net of user charges, for airports and air traffic control, for dredging rivers and harbors, and for providing postal service. Government funds by the billions have been devoted to the development of water resources, which provide both cheap hydroelectric power in the areas covered by them, and cheap water for irrigation in the southwestern part of the country, resulting in a great stimulation of agricultural output there. In addition, the Rural Electrification Administration subsidized the electrification of the rural parts of the American economy for nearly fifty years, at low interest rates, thus making farming somewhat less costly than it would otherwise be. 8 Price controls on domestically produced natural gas cheapened energy for Americans with access to the price-controlled gas, and hence provided "subsidies" (but not revenue-reducing ones) for American exports as well as for domestic sales of products that required gas to be used in their production. By 1993 gas price controls will be completely removed, so this "subsidy" will have disappeared (decontrol was largely completed in 1987). For nearly a decade before 1981 US oil prices were also held below world market prices, with similar effect. 9 Government expenditures on research and development help to cover the initial cost of new economic activities, which often later lead to exports. The classic example is agricultural research, which has been financed by government for over a century and which has led to vast improvements in the productivity of American agriculture and to improvements in the quality of agricultural products. Sometimes too, large export sales are a distant bi-product of military research development and development expenditures, as was the case with the jet engine. The government has also spent substantial sums on research and development in the energy sector, both on nuclear power and on such possibilities as liquefaction of coal. To the extent that the last proves to be economically feasible, it may augment future exports of American coal. 10 Extensive government purchases sometimes lead to the development of products which are highly competitive in world markets, by helping private firms to spread their own research and development costs as well as

146

RICHARD N. COOPER

other overhead expenditures over a larger number of sales. The list of products here in principle is a long one, but the point is quantitatively important in relatively few industries, such as military equipment, avionics, some kinds of telecommunications equipment, and ground tracking stations for satellites. 11 The most pervasive influences, and quantitatively probably the most important stimulus to exports of particular goods, but also the least obvious, arise from the host of government regulations on production which have been introduced to improve the working environment or the natural environment. Such items as effluent controls, safety regulations, minimum wage legislation, and restrictions on child labor can have a profound effect on the competitiveness of particular industries, and hence on the relative competitiveness of other industries less directly affected. Since most observers would not mention such government actions in a list of export "subsidies" - and indeed they do not normally give rise to a loss of government revenue, except where the government occasionally incurs some of the costs, for example, of anti-pollution actions - it is worthwhile to trace through the influence on exports of one of these regulations, the minimum wage. The key assumptions in this analysis are that over time balance is maintained in international payments, for example by movements in the exchange rate of the dollar, and that the government takes whatever steps are necessary to assure full employment of the labor force. So we are looking here for the sectoral effects, the relative stimulation or retardation of production in particular sectors of the economy which arise from the regulations in question. The minimum wage, if it is set high enough to exceed the wages that would otherwise be paid in some industries, reduces the international competitiveness of those industries by raising their cost. A directly affected industry will find it is more difficult to compete with products from abroad. Imports will rise, and restoration of equilibrium in the balance of payments will require some depreciation of the dollar relative to what it would otherwise be. The depreciation, in turn, will increase the competitiveness of all sectors where wages are not directly influenced by the minimum wage. Put more concretely, it is likely in the United States that the minimum wage discourages the production of apparel (which are displaced to some extent by imports) and encourages the production and export of machinery. Thus in an indirect fashion, via adjustment of the exchange rate, the export of machinery is "subsidized" (but again not in a fashion that reduces revenue to the government). A similar argument holds, mutatis mutandis^ for other government regulations. For example, meeting required safety standards will raise costs more in some industries than in others, and via adjustments

US response to foreign industrial policies

147

in the exchange rate will increase the competitiveness of industries or firms whose costs for safety have increased least. (See Kalt [1988] for a finding that environmental regulations have had a discernable effect on the composition of US trade.) So the influence of government actions on international competitiveness is pervasive, but it may also be so indirect as to be difficult to trace in detail with any confidence. It may nonetheless be substantial, even when it flows from actions aimed at objectives quite different from a desire to stimulate exports. The United States has a special characteristic which it shares with a few other countries such as Canada, Australia, and Germany, namely the federal structure of its government. The Federal government accounts for only about one-third of total government expenditures in the United States and for about one-fifth of total civilian government employment. For the most part, the influence of state and local governments on the structure of production and costs falls into categories (7) and (11) in the foregoing list: expenditures which support business enterprise in a general way, and regulation on the conditions of production or marketing. In addition, local governments sometimes support particular firms in the form of cheap land, low utility rates, or cheap credit. Starting in the 1950s a number of localities used their privilege of floating tax exempt securities to provide cheap credit to new firms through the issue of so-called industrial development bonds, a practice which continued until the late 1980s. The business promoting activities of state and local governments are not of course aimed at encouraging exports from the United States, but rather from the particular state or locality to the rest of the United States, and only incidentally to other countries. We saw above in the case of the minimum wage that a measure which has a negative impact on one industry indirectly affects others positively. The same point works in the opposite direction as well. Discrimination in favor of certain firms or industries automatically involves discrimination against firms and industries that are not specially favored. If subsidies or tax breaks or below market credits are given to favored firms, others have to make up the difference by paying more taxes or more for credit than would be true in the absence of the discriminatory measures. And the currency will appreciate to the extent that exports are stimulated, thus putting others at a disadvantage with respect to foreign competition. When Japan allows the formation of cartels in periods of recession or for industrial restructuring, the impact of which is to hold domestic prices above what they would otherwise be, this represents a negative action for those that buy from the cartel. Where consumers are made to bear the additional cost, as in the notorious color TV case, exports are not adversely affected. But where the

148

RICHARD N. COOPER

cartel raises the price of steel or industrial chemicals or machinery or a host of other products, this represents increased costs for the downstream purchasers. (Of course, for domestic prices to remain above world market prices some form of import restriction must exist.) It may also be true, as Yamamura points out, that the possibility of creating recession cartels has generated higher capacity than otherwise would occur, by removing from investing firms some of the risk of a major downturn in the market, and that this excess capacity in turn encourages exports at something below average cost. In that event, the yen will be stronger, and cartelization will put other firms at a disadvantage for that reason too. These points are generally neglected by those who criticize the unfair trade practices of Japan and other countries. A similar argument applies to required procurement of domestic products. An enduring and justified complaint of the American heavy electrical generating equipment industry is that potential purchasers in other major industrial countries buy exclusively domestic products, regardless of cost. This is true especially in Europe, where electricity supply is typically provided by public authorities, but it is even true in Japan, which like the United States has extensive private utilities. In contrast, the American market for heavy generating and transmission equipment is relatively open, and perhaps one-fifth of US purchases are from abroad, mainly Japan. However, the requirement by foreign utilities to buy domestic in general will raise the costs of generating electricity there, and those costs must be either subsidized by the government or passed on to consumers, including industrial consumers. So, while the US heavy electrical equipment industry suffers from this practice, the rest of American industry in general benefits, by virtue of paying lower electricity charges than its foreign competitors must do. Similar arguments apply to the requirement in some countries that computers should be purchased locally. Over time, unless the local computers are competitive in quality and price, the (mostly state-owned) enterprises that are burdened by this obligation will surfer in competitiveness. We can now return to a point raised at the outset: Can comparative advantage be made? Of course it can. In a tradition going back at least to Frank Taussig, it has been recognized that comparative advantage is determined not only by a country's natural endowments, but also by its social and political and educational systems.5 It is determined by the quality as well as the quantity of its labor force, by the motivation of its workers, and by their willingness to work diligently. Insofar as government provides for inland transport, education, efficient banking and other financial transactions, these too can influence comparative advantage, as can the long list of items discussed above for the United States, especially the regulatory

US response to foreign industrial policies

149

environment. It would be absurd to pretend otherwise and to treat each geographic area as a tabula rasa with natural endowments but with no social or political system. That is perhaps not what is meant when people talk about "making" comparative advantage rather than "finding" it. Rather, they may have in mind permanent cost advantages that are generated by a headstart or by extensive production of a particular product. If important and durable economies arise from "learning by doing," then "doing" can give an advantage, however it is brought about, as analyzed by Krugman. However, two points should be kept in mind before rushing headlong to do many things to reap the learning by doing cost advantages that they may engender. First, undertaking production which is not at once profitable in order to reap the advantages of learning by doing represents an investment, and to undertake any investment on the supposition of future positive rates of return is not sufficient. It will be a good investment only if the yield on the investment is at least as high as that on alternative investments. If on analysis that seems to be the case, then as with any new investment opportunity many parties may simultaneously want to undertake it, and that would not necessarily be a bad thing. Second, "decisive" cost advantages for a given product or product group achieved by learning by doing turn out to be remarkably transient in many instances, given the large number of products which have had their commercial introduction in the United States, but the production of which has subsequently been relocated abroad. Evidently there are cost factors which eventually overwhelm the cost advantages achieved even by a relatively long headstart in production and the learning by doing gains that are thus achieved. There remains the possibility, however, that other countries do not recognize these two points, and that they therefore undertake government support for activities which turn out to be bad investments, but which in the meantime create competition for otherwise successful American firms. What if anything should be done about it?

Possible policy directions The problems posed above have been around for a long time. What we now call industrial policy goes back, in the strictly commercial arena, at least to the time of the early eighteenth-century, when France tried (successfully) to create a high-quality porcelain industry in competition with the Saxon industry in Dresden. The effort in textiles is even older, as when in the early seventeenth-century England attempted (unsuccessfully) to start a finished textile industry in competition with the Flemish cities, by prohibiting the export of wool.

150

RICHARD N. COOPER

Existing arrangements The General Agreement on Tariffs and Trade (GATT) of 1947 dealt with the problem of subsidies and international trade by suggesting (in Article 16) that export subsidies on manufactured goods — note the exclusion of primary products - should generally be eschewed, and permitted contracting parties to impose countervailing duties when such subsidies cause material injury. (In addition, Article 19 provides a general escape clause from GATT commitments, permitting countries to reimpose tariffs, but only on a non-discriminatory basis, whenever a domestic industry is subject to substantial injury by imports.) A GATT working party in 1960 attempted to define more precisely exactly what export subsidies were. In the mid 1970s the United States introduced its "traffic light" proposal, distinguishing between subsidies that were prohibited (red), subsidies that were clearly permitted (green), and subsidies which were potentially troublesome and which were subject to consultation and possible countervailing action when they caused injury to some other contracting party to the GATT (yellow). This proposal was not adopted in the form in which it was presented, but it provided a framework for the Code on Subsidies and Countervailing Duties that was adopted in 1979 as part of the Tokyo Round of multilateral trade negotiations. The 1979 Code prohibits export subsidies except on certain primary products and except by developing countries. Code signatories undertake not to use such subsidies (Article 9) and their use creates a (rebuttable) presumption of adverse effects, and are thus subject to countervailing duties without an injury test, but subject to international approval. The Code also provides, in its Annex, an illustrative list of export subsidies, which goes a long way toward defining them. The Code acknowledges the widespread use of, and permits, many other subsidies, but signatories recognize that these can hurt the trade interests of other countries and they therefore "seek to avoid causing such effects through the use of subsidies." In evaluating the use of subsidies in pursuit of domestic economic objectives, signatory countries are to "weigh . . . possible adverse effects on trade" (Article 11). If other countries are injured by such subsidies, they can countervail the subsidies in question (see Cooper in Spence and Hazard, 1988). This general language leaves a large area for interpretation, not to mention the cracks and overlaps in the illustrative list of prohibited subsidies. US practice has gravitated toward an interpretation that requires a domestic subsidy to provide a "special favor" to a firm or industry before it is countervailable under US law. That is to say, the United States tries to rule out as countervailable subsidies such broad legislative favors as accelerated depreciation, investment tax credits, research and development tax credits, and so on. Thus a domestic subsidy is countervailable if it causes material injury to an American industry through stimulated exports and if

US response to foreign industrial policies

151

the subsidy is selective rather than general in its impact on the foreign firm or industry. On the whole, the approach embodied in the GATT Code and in evolving US practice seems very sensible. Export subsidies are prohibited, and other subsidies are permitted but are countervailable if they are selective and if there is injury to another party. But several practical problems remain. First, on this track we must await sharper definition of the distinction between selective and general subsidies on a case-by-case basis. Second, during the period in which this sharper definition is occurring, the possibility exists for harassing imports by bringing test cases, and creating uncertainties whenever there is a preliminary finding of subsidy, through "suspension of liquidation" under which an importer is put on notice that he may have to pay a higher duty at some subsequent date, but does not know what the duty will be. Third, there is no assurance that other countries will move in the same direction as the United States in their interpretation of the distinction between selective and general subsidies. The United States tends to be the most active country in developing the case law, perhaps because other countries have less formal mechanisms for restricting imports if they choose to. The Uruguay Round of multilateral trade negotiations provides an occasion to resolve especially the last difficulty, but in early 1992 no clear resolution was in prospect. In their detailed study of treatment of subsidies to international trade, Hufbauer and Erb suggest that the list of prohibited export subsidies needs to be tightened and that the distinction between general and specific subsidies needs to be substantially clarified. But they basically accept the existing framework. Their most novel suggestion concerns remedies to subsidies, where they propose the countervailing subsidy, to be financed by an (internationally approved) import duty on goods coming from the country providing the offending subsidy (Hufbauer and Erb, 1984, p. 129). The advantage of a countervailing subsidy as opposed to a countervailing duty is that it permits the injured country to continue to sell in third markets and even in the home market of the subsidizing country. Its obvious disadvantage is that it would affect fourth country competitors, who now would have to compete with two subsidizing countries, and that might engender a race toward subsidies. It would perhaps ultimately lead to a "disarmament" negotiation to remove the subsidies, as happened during the late 1970s and early 1980s with respect to official export credits, thereby reducing the distortions to international trade. But once subsidies are introduced, eliminating them is a prolonged, difficult, and not wholly successful process as the effort to get consensus on official export credit restraint also illustrates.

152

RICHARD N. COOPER

Three alternative criteria A major difficulty with public debate in this area is that it mixes three quite different criteria in assessing subsidies: distortions to resource allocation, fairness or equity, and injury. It is reasonable to conjecture that if there were no injury the issues of fairness and distortion would not arise in the practical world of policy, although academic economists would be concerned with the misallocative effects of distorting subsidies. However, when injury does arise, the three different criteria tend to be co-mingled, yet each points to a rather different solution. Subsidies, domestic or export, induce distortions in the allocation of resources and hence should be eliminated or offset for the sake of economic efficiency, except in those cases where the subsidy is itself designed appropriately to offset some other distortion, e.g., an externality of some kind. This is the principle that underlies the prohibition on export subsidies, and the permissible imposition of a countervailing duty. As pointed out above, however, the countervailing duty cannot undo the distortion with respect to competition in third markets. A plan to eliminate all of the distortions introduced by government action would be a counsel of perfection given the extensive government intervention in modern economies, enumerated above. It would even be difficult against all selective subsidies, which would invariably be justified on grounds of correcting a distortion in capital or labor markets, or as exploiting an externality, or on national security grounds, or in extreme cases as necessary to insure domestic peace and political harmony. Governments are likely to abandon their extensive array of domestic subsidies only when they come to deem the costs, defined broadly, to be greater than the rewards. Emphasis on equity or fairness leads to a second approach to policy, which would be to harmonize domestic measures among countries. If an activity has positive externalities or national security value, that attribute is presumably not limited to a particular country. Most countries are concerned about having substantial local production of the staple food, perhaps of steel, perhaps of small arms, and increasingly of electronics, on grounds of national security. Nations can negotiate broad ground rules on what is and is not acceptable. That would offer one concrete meaning to the otherwise obscure phrase, providing "a level playing field." The rules would be permissive rather than obligatory, but they would also be limiting, in that countries could not subsidize beyond what was agreed. The United States tried in the Tokyo Round to negotiate a more detailed elaboration of acceptable but possibly troublesome subsidies, but the result was the less specific Code described above. The United States tried again in the Uruguay Round to get greater international agreement on domestic subsidies that are permissible and those that are sufficiently potentially troublesome that they should be avoided, with modest success, although it

US response to foreign industrial policies

153

was neglected for higher priority issues. Inevitably such a negotiation would have to be on a conditional MFN basis, as the 1979 subsidies Code was, since the interests of over 100 GATT members are too diverse to permit any meaningful harmonization among all of them. By general consensus, developing countries are held to a lesser standard than are the fully industrialized countries, although the extent to which the standard should be relaxed is still a matter of considerable and continuing dispute. Negotiation on harmonization of subsidies would be a cooperative approach. If other countries decline to cooperate willingly, the United States could approach the matter unilaterally. One way to do this is through the notion of "reciprocity" as embodied in Senator Danforth's 1985 telecommunications bill. This bill, subsequently watered down and incorporated in the International Trade and Competitiveness Act of 1988, was motivated by the fact that Japanese and major European markets in telecommunications equipment were effectively closed to American (and other foreign) sellers, since most of this procurement is undertaken by government-owned (PTT) monopolies. Danforth's bill would have required other countries to open their individual markets to imports of US telecommunications equipment within three years under the threat of a sharp increase in US tariffs on telecommunications products coming from them, from roughly 8 percent today to about 35 percent. The underlying principle was that the terms of access should be the same in each product field, telecommunications in this case, and that the United States should persuade others to adopt US practices - which, with the 1984 breakup of AT&T, means relatively open procurement by private telecommunications companies in the United States. In the absence of foreign adoption of US practice, the United States would greatly reduce foreign access to the US market, country by country. There are two problems with this as a general approach. First, it does not address the question of exports of third countries. (That is perhaps not a major problem in telecommunications; indeed the United States continued to hold 38 percent of the world market in telecommunications products in 1982, the latest data available when the Danforth bill was framed, compared with only 11 percent for Japan despite its alleged export prowess.) Second, success flowing from the threat embodied in the Danforth bill is likely to vary from country to country. Japan has a large stake in the US market. France, whose practices were if anything even more exclusive than Japan's, did not. Japan might comply in its negotiations with the United States, Britain and Germany might comply partially, and France might not comply at all. The result at the end of three years would be that the United States would have to have different levels of restriction on products from each of these countries and value-added criteria to prevent geographic arbitrage

154

RICHARD N. COOPER

among them. The issue is even more complicated when one allows for the difference between national ownership and location of production, through foreign subsidiaries. Thus new distortions would be introduced into international trade, the discrimination implied would represent a sharp break with the most favored nation principle embodied in the GATT, and the consequence would be much closer surveillance over imports, with correspondingly greater opportunities for threats and harassment. Such developments would not be in the overall US interest. A second unilateral approach toward a "level playing field," or more accurately toward reducing the protectionist mischief that can be done under that label, would be to introduce a uniform tariff of, say, 10 percent on all US imports of manufactures, against a presumed entitlement of zero, combined with a stipulation that any industry wanting protection above that level would bear an exceptional burden of proof to show that it was substantially injured by a substantial foreign subsidy, one that was well above 10 percent, except in the case of clearly prohibited subsidies. By imposing a levy on all imports of manufactures, the uncertainty that foreign exporters face with respect to US administrative law in the area of subsidies and countervailing duties could be greatly reduced. The disadvantage of this kind of measure, of course, is that it would impose a burden on all consumers of imports, and also directly and indirectly reduce the competitiveness of US exports, mitigated only in part by giving drawbacks of the duty on their imported inputs. Also, over time it might be difficult to sustain the principle that the 10 percent duty covered most foreign subsidies, against an asserted entitlement of no duty. The third major approach shifts the emphasis to injury, and away from the subsidies as such, except for the clearly prohibited subsidies. It recognizes that government influence is pervasive and of long standing, and that in many ways such influence has been absorbed into the existing prices and economic structure. It focuses instead on large changes in government policy and the injury that such changes may cause. This approach would call for prior international notification of all major changes in government policy, and gradualism in the implementation of these changes. When gradualism is not followed, it would permit digressive (that is, gradually phased out) relief to the injured parties, with a view to encouraging ultimate adjustment to the new situation. Procedures would be necessary for discussing new policies and for considering measures that would achieve the same objective with smaller external impact. These procedures would cover any major change in policy with external repercussions on particular industries, and would have the objective of modifying the proposed action so as to reduce the imposition of costs on other countries. A gradual introduction of the new measures, like the multi-year staging of tariff

US response to foreign industrial policies

155

reductions, is one way to reduce the costs of adjustment. The principles involved here cover subsidization that stimulates exports. Instead of calling for harmonization of policies, an exacting requirement in a world of independent nation states, this approach calls for extensive prior consultation on any industrial policy changes that are likely to cause injury, with a view to avoiding those changes but also with a view toward mitigating any injury that does occur. Variants of the harmonization approach and the approach emphasizing injury are not incompatible with one another. It would be possible to emphasize injury as the main strategy, while still negotiating with other countries to reduce the discrepancies among their industrial policies. But extreme harmonization — attainment of a fully "level playing field" — would not be necessary. A.dvice to the United States In framing US policy, it is necessary to consider what is in the best interest of the United States, given its capabilities and limitations. The United States cannot favor particular industries secretly; that is not consistent with the way American government operates. Should it do so openly? My answer would be that open support for particular industries should be given only if a strong public policy case can be made for such support on its merits: either on grounds of national security, or if there are substantial and demonstrable externalities which are engendered by such support. This approach, while sounding blandly obvious, in fact has strong implications with respect to some of the proposals that have been made. For example, the Labor Industry Coalition for International Trade has complained of closed markets in other industrial countries for electric power generating equipment, in contrast to imports by the United States that on occasion have reached 20 percent of US purchases. This, they contend, is unfair (see LICIT, 1983, pp. 79-82). But the United States did not lose any of the economies of scale that it might otherwise obtain. In fact the US share of world exports actually rose between 1965 and 1980, from 5 to 10 percent. In addition, Americans presumably get lower cost power by permitting import competition. Power is an important input into industry. It is possible, for instance, that part of the difficulties of aluminum refiners in Japan was due to preferred Japanese procurement of domestic electric generating equipment. While it would certainly be desirable to be able to sell US equipment in France and Japan, it is not obviously in the US interest to stop purchases from those countries even if they decline to open their markets to US equipment, and indeed the prevailing situation may on balance even favor US industry as a whole. We simply do not know about the general equilibrium effects. Furthermore, US actions should not be motivated by arguments that rest

156

RICHARD N. COOPER

on the fallacy of composition, that is, that the action would generate employment or improve the overall trade balance. Except in the short run, employment and the overall trade balance are determined by the macroeconomic conditions of each economy, not by particular trade policies. Where beneficial practices in other countries may also have beneficial effects in the United States, making allowance for differences in institutions, training, and temperament, Americans should emulate the foreign practices. That is, Americans should not fall into the habit of rejecting ideas that "are not invented here." Fortunately there is little risk of that, since Americans have traditionally been open to ideas whatever their source, and American business has often been quick to commercialize inventions made elsewhere, especially in Europe. Two experiments drawing from Japanese experience of government-industry cooperation were inaugurated in the second half of the 1980s: the creation of Sematech, a business research coalition with Federal government financial support to work in the area of semi-conductors and machinery for making high-quality integrated circuits, and the creation of the National Center for Manufacturing Sciences, a similar effort to improve and especially to disseminate new manufacturing technologies. In early 1992 it is too early to tell how successful these efforts will be, although neither had yet had any striking results after several years of existence. In short, the United States should look at other countries (1) for practices which are illegal under GATT and its codes, and (2) for policy measures that may have useful application in the United States. But in the end the United States should adopt measures that are best for the United States in the institutional setting of the United States, not just because they seem to have worked well abroad. Many actions by foreign governments have been costly and largely unsuccessful; those that have been successful would not necessarily or generally be successful in the American context. The United States could adopt many of the practices of Europe or Japan; but it would be unwise to do so. In particular, Americans should have to make a specially strong case for discriminatory treatment in favor of any particular industry. It is much easier for foreign countries to "pick winners" because they observe successful new industries in the United States. How would the US government pick the industries of the future? To what experience could Americans turn? Instead, the United States relies on private rewards to thousands of firms and individuals, each making a guess as to which activities will be winners in the future. Most will be wrong. Some will be right. But what kind of position does this leave the United States in with respect to bargaining with other countries on their practices? Might it not be desirable for the United States to introduce Danforth-style reciprocity or in

US response to foreign industrial policies

157

other ways threaten to close US markets with a view to getting foreign countries to open theirs? Here we must recognize the limitations of the United States. The American government cannot bargain subtly in the economic arena, and it has always found it difficult to back away from a publicly stated and argued position introduced for bargaining reasons, but not fully acknowledged as such. Many Americans will believe the arguments that have been advanced. In short, a pluralistic, open society cannot bluff. It simply cannot carry it off. Foreigners will be skeptical of any non-serious threat, and once domestic political support has been built for the threat, the threat becomes serious but it ceases to be a bluff. For this reason the United States should threaten actions only when it is clearly willing to undertake them. And in general it should be willing to undertake them only if they are in its best interests, not simply as a bargaining tactic. It is true that the ambiguity surrounding decision-making within the United States, and particularly the relationship between the President and the Congress when it comes to trade matters, is sometimes helpful in bargaining with foreigners, since foreigners can never be certain that Congress will not take the upper hand and move in a way which puts them at a disadvantage. This uncertainty is a reality of American politics, and it can sometimes be used successfully in the bargaining context. But it is quite different from the bluff of a poker player. If other countries continue to be recalcitrant in pursuing practices that the United States considers unacceptable, should the United States retaliate as a demonstration that it is willing to take action hurtful to the other country, even if it is also hurtful to the United States? As suggested above, the general answer to this question is negative. However, it cannot absolutely be ruled out, if the action has a reasonable prospect of inducing a change in behavior by the other country or by third countries, that is beneficial to the United States. But this prospect is so heavily conditioned on the exact manner, timing, and context of the threatened retaliatory action that general rules are not likely to be helpful. It is a case where "playing it by ear" is preferable to playing by score or, to mix metaphors, by recipe. That is where the art of diplomacy comes into the economic arena. Notes 1 I use the term "practices" since by 1990 most difficulty in import penetration in Japan is due, in my judgment, less to governmental policy than to deeply ingrained purchasing practices by Japanese firms and mid-level ministry officials. 2 President Bush's trip to Tokyo in January 1992, accompanied by CEO's of US automobile and other companies, was unfortunately premised on this fallacy of composition: US businessmen repeatedly linked low Japanese auto imports to the large Japanese surplus.

158

RICHARD N. COOPER

3 The following description is a blend from several sources; Namiki in Warnecke, 1978, US-Japan Trade Study Group, 1984, Yakushiji in Aoki, 1984, Yamamura, 1982, and interviews with MITI officials. 4 Japanese government support was terminated in 1979 under pressure from the United States. The private sector continued the program without government help. See Tyson, 1992, chapter 4. 5 Taussig (1927, pp. 57-8) pointed to Germany's well-trained chemists and lab assistants as the basis for that country's comparative advantage in chemical dyes and other chemical products.

References Aoki, Masahiko (ed.) (1984), The Economic Analysis of the Japanese Firm, Amsterdam: Elsevier. Borrus, Michael, Laura Tyson, and John Zysman (1986), "Creating Advantage . . .," in Paul Krugman (ed.), Strategic Trade Policy and the New International Economics, Cambridge, Mass.: The MIT Press. Caves, Richard E. and Masu Uekusa (1976), Industrial Organisation in Japan, Washington, DC: The Brookings Institution. Cline, William R. (1982), Reciprocity: A New Approach to World Trade Policy?, Washington, DC: Institute for International Economics. Cline, William R. (ed.) (1983), Trade Policy for the 1980s, Washington, DC: Institute for International Economics, 1983. Hufbauer, Gary Clyde, and Joanne Erb (1984), Subsidies in International Trade, Washington, DC: Institute for International Economics. Jacquemin, Alexis (ed.) (1984), European Industry: Public Policy and Corporate Strategy, Oxford: Clarendon Press. Kalt, Joseph P. (1988), "The Impact of Domestic Regulatory Policies on International Competitiveness," in Michael A. Spence and Heather A. Hazard (eds.), International Competitiveness, Cambridge, Mass.: Ballinger. Krugman, Paul (1990), "Notes on Trade in the Presence of Dynamic Scale Economies," in Krugman, Rethinking International Trade, Cambridge, Mass.: The MIT Press. Labor-Industry Coalition for International Trade (LICIT) (1983), International Trade, Industrial Policies, and the Future of American Industry, Washington. Magaziner, Ira C. and Robert B. Reich (1983), Minding America's Business, New York: Vintage Books. Ministry of International Trade and Industry (MITI) (1980), The Vision of MITI Policies in the 1980s, Tokyo. Nelson, Richard R. (ed.) (1982), Government and Technical Progress: A Cross-Industry Analysis, New York: Pergamon Press. Nelson, Richard R. and Sidney G. Winter (1982), An Evolutionary Theory of Economic Change, Cambridge, Mass.: Belknap of Harvard University Press. O E C D (1991), Strategic Industries in a Global Economy: Policy Issues for the 1990s,

OECD International Futures Programme.

US response to foreign industrial policies

159

Okimoto, Daniel I. (1989), Between MITI and the Market: Japanese Industrial Policy for High Technology, Stanford University Press. Patrick, Hugh and Henry Rosovsky (eds.) (1976), Asia's New Giant, Washington, DC: Brookings Institution. Pavitt, Keith and Margaret Sharp (1992), "Key Technologies and New Industrial Policies," prepared for Europe and Global Economic Interdependence, College of Europe, Bruges, January. Taussig, Frank W. (1927), International Trade, New York: Macmillan. Tyson, Laura (1992), Who's Bashing Whom: Trade Conflict in High-Technology Industries, Washington, DC: Institute for International Economics. US-Japan Trade Study Group (1984), "Progress Report: 1984," Tokyo. Warnecke, Steven J. (ed.) (1978), International Trade and Industrial Policies, London: Macmillan. Yamamura, Kozo (ed.) (1982), Policy and Trade Issues of the Japanese Economy, Seattle: University of Washington Press.

CHAPTER

7

The current case for industrial policy PAUL R. KRUGMAN

On the whole, economists have been highly critical of government efforts to target particular industries for protection or promotion. The actual policies of governments have, of course, been criticized for their clumsiness, their responsiveness to special interest groups, and often their sheer corruption. Beyond this critique of practice, however, economists have been extremely negative about the idea of industrial policy even in principle. The general presumption of most economic theory is that the best industrial policy is to let the market work - that the decentralized incentives of the market place will push resources to the places with the highest expected return, and that no second-guessing of market decisions is necessary or desirable. The purpose of this chapter is to argue that whatever the failings of industrial policy in practice, there is in fact a pretty good case for such a policy in principle - and that recent developments in economics give us a reasonable set of criteria for thinking about which industries to promote. This represents a change in position on my part. Like other economists, I have written a number of critiques of proposals for industrial policy (Krugman, 1983, 1984). These critiques reacted (correctly) against the crude misconceptions that seemed to underlie most proposed criteria for industrial targeting. They went on, however, to argue that, while there might be potential reasons for an active industrial policy based on more sophisticated criteria, it was unlikely that such a policy could actually do much good. I have now changed my mind, and have gone, at least slightly, soft on industrial policy. The reasons for my initial skepticism can be briefly summarized as follows. It has been clear for some time that an intellectually respectable case 160

The current case for industrial policy

161

for industrial policy must rest on external economies. But are external economies sufficiently important to worry about, and can they be empirically identified with enough reliability to be a guide to useful policy? Until recently the answer to both questions seemed to be "No." Standard theory said that the only external economies relevant for policy are technological externalities; while these surely exist, it seemed that they would be limited to a few industries (and would also, a point discussed later, often be international rather than national issues). Thus they were probably not very important. Furthermore, technological external economies are by their very nature extremely difficult to identify - they leave no "paper trail" of market transactions. Thus it seemed unlikely that one could give any useful advice about industrial policy even where an omniscient economist might have advocated government activism. What has now become clear to me, however, is that this skepticism went too far, for two reasons. First, it is now clear that the standard view that there is a sharp line between technological externalities, which matter for efficiency, and pecuniary externalities which do not, is wrong. Or to be more accurate, it would be correct only in a constant-returns/perfect competition economy. In the real world, where increasing returns are pervasive and most competition is oligopolistic, pecuniary external economies often matter a great deal. In particular, Hirschmann (1958)-type forward and backward linkages make perfectly good intellectual sense, and are probably quite important in practice. Second, there is a growing awareness that most external economies (broadly defined) probably apply not at an international level but at a regional or metropolitan level. If one accepts this, then it becomes possible to look for external economies by examining the geography of economic activity within countries - and the task of identifying likely sectors for industrial targeting becomes much less hopeless. It is important to emphasize that this argument, which will be brought out at much greater length, is not a blanket endorsement of any and all industrial policy. Much, probably most, actual industrial policy continues to be based on economically irrational criteria. In addition, the political economy of industrial policy remains very problematic, with the risks of capture by special interest groups very high. The point of this chapter is therefore not to sound a clarion call for massively increased government intervention; it is simply to call attention to the point that intellectually, at least, industrial policy should be more respectable than it currently is. This chapter is in four sections. The first section discusses the general issue of criteria for industrial policy. The second section presents some illustrative models of the modern view of external economies, which is crucial to the argument. The third section discusses some of the evidence on

162

PAUL R. KRUGMAN

external economies provided by examining economic geography. Finally, the fourth section discusses the possible shape of a sophisticated industrial policy. 1

Criteria for industrial policy

Why should a government encourage some industries at the expense of others? One of the reasons that industrial policy has gotten a bad name among serious analysts is that most advocates of industrial targeting offer badly thought-out answers. Probably the most common argument is that the government should push resources into "high-value" activities. The normal working of a market system, however, already gives factors of production the incentive to seek out the highest possible rate of return. If the purpose of government intervention is simply to be smarter than the market, to "pick winners," it is unlikely to do more good than harm. In fact, many popular proposals for industrial policy (and much industrial policy in practice, especially in developing countries) amount to pushing industries that have high value added per worker, but which correspondingly have low value-added per unit of physical or human capital. Since capital is a scarce resource, such schemes do not raise overall output, and can easily lower it. One may turn the question around. Since the market already leads factors of production to seek their maximum return, how can government targeting increase that return? At a basic logical level, there are only two ways for industrial targeting to be successful. First, targeting could generate excess returns by creating market power. If a government policy succeeds in winning a new or expanded monopoly position for domestic factors of production that would not otherwise have had it - and if the monopoly rents are collected from foreigners rather than domestic residents - then a targeted industrial policy can raise a country's income at foreign expense. Thus one possible rationale for industrial policy is rent creation.

Second, targeting can raise overall national income if there are some industries in which the social rate of return exceeds the private rate - that is, in which the resources committed by individual firms indirectly raise the earning of other firms' resources. This is, of course, the case of external economies.

Traditional economic analysis of the case for intervention, because it assumed perfect competition, focused entirely on the external economy issue. While the case was understood in principle, however, it was generally concluded (erroneously) that external economies were probably not an important issue or were at any rate too elusive to be a useful concern of policy. Thus when during the 1980s the possibility was advanced of

The current case for industrial policy

163

devising an industrial policy to take advantage of rent creation instead, this seemed like a major advance. The theory of "strategic trade policy" advanced by Brander and Spencer attracted extraordinary attention. What I want to argue now is that strategic trade policy in the Brander-Spencer sense turns out to be intellectually interesting but probably not very important in practice, while external economies are much more important than we have realized. The rise and decline of strategic trade policy In the early 1980s James Brander and Barbara Spencer (1983,1985) suggested a particular approach to trade policy analysis under the new theory that created a massive stir. The Brander-Spencer analysis did three things. First, it offered a particularly clever way of setting up the case for activist trade policy, one which simplified the issue enormously and thereby revealed its core. Second, it seemed to suggest that the new trade theory provided at least limited support for a kind of neo-mercantilism, for the assertion that governments could in fact raise national income at other countries' expense by supporting national firms in international competition. Third, and not without importance, the Brander-Spencer approach could be succinctly described with a term that, while accurate, seemed to promise a larger prize than Brander and Spencer themselves ever suggested: "strategic trade policy." What the Brander-Spencer approach actually consisted of was the following: we imagine two firms, from each of two countries, competing for some export market. Domestic consumers in this sector are ignored or assumed away, so that the approach is inherently biased toward a view of trade as competition rather than mutual gain. The firms compete by choosing the level of some strategic variable: perhaps output, perhaps capacity, perhaps R&D. In this kind of competitive situation, firms would like to convince each other of their aggressiveness. That is, each would like the other to believe that it will invest or produce massively, thereby inducing the other to produce or invest less, perhaps even to avoid entering the market at all. The problem is to find a way to make the threat of aggressive competition credible. The answer suggested by industrial organization theorists is that firms will make "strategic" moves - that is, take actions that do not directly raise profits, but that are intended to make aggressive behavior more credible and therefore have a deterrent effect on potential rivals. The quintessential strategic move is construction of excess capacity, which a firm does not expect to use, but which it builds in order to deter entry of potential competitors.

164

PAUL R. KRUGMAN

What Brander and Spencer pointed out was that trade policies could serve the same strategic purpose. Suppose that one of the two firms is backed by a government, which commits itself to subsidize the firm's sales. Then the other firm will know that an aggressive policy by the subsidized firm is rational, and curtail its own plans. The result can be to raise the firm's profits by much more than the actual subsidy outlay. And as a result, such a "strategic" trade policy can raise the aggressive nation's income at other countries' expense. It is a lovely analysis in the way that it cuts through the complexities. Yet it is also subject to abuse: Brander and Spencer's work has enabled crude advocates of aggressive trade policies to give their views a new intellectual gloss. Thus the theory of strategic trade policy has been subject to an unusually detailed academic critique. And the upshot of this critique has been to show that what Brander and Spencer offered was an example, not a general result. Eaton and Grossman (1986) showed that the case for strategic aggressiveness was sensitive to the assumed form of competition; Horstmann and Markusen (1986) showed that the benefits of strategic trade policy might be dissipated by entry of new firms and the resulting excess capacity; Dixit and Grossman (1986) showed that competition for scarce resources among industries complicates greatly the task of devising a welfare-improving policy; and Dixit and Kyle (1985) argued that strategic trade policies should be seen as part of a larger game in which it would often be better for governments to rule out their possibility. What this academic critique showed was not that the strategic trade policy concept was wrong, but that it was not necessarily right. Or to put it more accurately, the case for strategic trade policies was not like the traditional case for free trade, which (in the old trade theory) could be made a priori without consideration of the specific details of industries. Strategic trade policies could be recommended, if at all, only on the basis of detailed quantitative knowledge of the relevant industries. So what the new trade theory gave rise to was not a prescription for policy, but a program of research. Making models of imperfect competition quantitatively operational is not an easy task. Indeed, it is sufficiently difficult that to a considerable extent the field of industrial organization proper seems to have punted: since 1970 a remarkable and fascinating body of industrial organization theory has arisen, with an equally remarkable lack of operational content or empirical confirmation. In international economics, however, such a state of affairs has been viewed as unacceptable by all concerned. Perhaps because free trade is such a powerful symbol, perhaps because international economists are more likely than industrial organization theorists to commute to Washington or

The current case for industrial policy

165

Brussels, there was almost immediately a demand to put up or shut up: to determine when, if ever, the new arguments against free trade were relevant. But how was this to be done? Ideally one would estimate models of imperfectly competitive industries econometrically. In practice this is extremely hard to do, because of the difficulty of identifying firm behavior. In fact, it is actually very hard to estimate models even of perfectly competitive industries; adding the potential complexities of oligopoly is beyond what anyone has managed to do. The only method that has seemed to offer any hope of progress has been a "calibration" technique similar to that used in computable general equilibrium (CGE) models. In this technique, parameter estimates are drawn from econometric and engineering estimates wherever possible; the number of remaining parameters of the model is then narrowed down by a priori assumptions until the model can be fully identified by requiring that it match data for some base period. This is by no means an ideal method - most papers in this area contain some kind of disclaimer, an acknowledgement that the results should not be taken too seriously - but it at least allows research to go beyond purely theoretical speculation. There is now a reasonably large selection of calibrated new trade models, including Dixit (1988), Baldwin and Krugman (1988a, 1988b), Smith and Venables (1988), Venables and Smith (1986), Baldwin and Flamm (1989), and others. What do these models tell us? Three main points seem to have emerged. First, the models generally suggest that the positive economics of trade policy - its consequences for output and trade flows - are quite different from the predictions of conventional trade theory. In particular, protection, by encouraging entry of domestic firms, often promotes exports. In some cases, as in Baldwin and Krugman (1988a), this result alone is of some importance for policy disputes. Second, the models have for the most part supported the view that modest tariffs and/or subsidies, if imposed unilaterally, do improve on free trade. Dixit's initial model suggested that tariff rates in the low double-digit range were optimal; similar results have recurred in a number of other papers. Third, however, the calibrated models universally suggest that the gains from even an optimal strategic trade policy are extremely modest. Indeed, the gains are so small that even a mild cynicism about the ability of governments to act intelligently and/or honestly would be enough to persuade one that it would be better not to get into the strategic trade policy business at all. Thus while the effort at quantification has provided some support for the

166

PAUL R. KRUGMAN

strategic trade argument in principle, it has also suggested that it is a bad idea in practice. But if rent capture is not a good justification for industrial policy, we are left with external economies. And until recently, it was conventional to argue that external economies were not a very plausible justification for targeting either. The conventional case against external economies Some years ago, in a policy-oriented paper (Krugman, 1984), I provided a very clear and quite incorrect case against taking external economies seriously as an industrial policy issue. The argument involved three steps. First, I argued - conventionally - that only technological, not pecuniary, external economies are a valid source of public concern. This is the standard view in welfare economics. A firm that buys more of an input, such as iron ore, will by its action raise the incomes of iron ore producers; but its failure to take that effect into account does not, in standard models, lead to any misallocation of resources. Only when there is a spillover directly into some other firm's production function - for example, when a firm makes an innovation that other firms can imitate — is there a market failure in the first firm's lack of concern. Second, I argued that since the only relevant external economies were purely technological, they were likely to be both limited in extent and difficult to observe. Knowledge spillovers seemed to me at that time likely to be important only in a few very innovative high technology sectors. Furthermore, by definition a purely technological spillover leaves no record in market transactions, and thus will be very elusive as an empirical matter. Finally, to the extent that technological spillovers are important, they seemed likely to be international rather than national in scope. A US-made computer can be reverse-engineered in Japan, and vice versa. If external economies are international rather than national, however, industrial policy is not an international competitive strategy; if a country supports an industry that generates international spillovers, the main benefits will accrue to other countries. These three points, taken together, seemed to make a strong case against placing much weight on external economies as a basis for industrial policy. I now believe, however, that all three were wrong. First, the conventional sharp distinction between pecuniary and external economies is incorrect. It is valid only when there are constant returns and perfect competition; in a world of increasing returns and imperfect competition, the range of significant external economies is much larger. In

The current case for industrial policy

167

particular, there are true external economies associated with a variety of market-size effects. Second, once one recognizes the importance of market-size effects, one becomes aware that such market-size effects are in fact pervasive in modern economies. Finally, many of the important market-size effects apply not at the level of the international or even the national economy but at a regional or local level. The argument that the gains from support of industries that generate external economies will be dissipated abroad is therefore mostly wrong. The key point in this new argument is the insight that pecuniary external economies matter. 2

The new economics of external economies

The easiest way to explain the new view of external economies is with some illustrative examples. Once we have these examples under our belt, it will be possible to sketch out a broader version. Model 1

The Big Push

In the 1940s, Rosenstein-Rodan (1944) suggested that poor countries might be caught in a low-level equilibrium trap in which firms were unwilling to invest because of an inadequate market, and the market was small because of the lack of investment. Unfortunately, neither Rosenstein-Rodan nor other theorists of the great age of development economics - Hirschmann and Myrdal in particular - succeeded in making the key assumptions of the Big Push idea clear. Over time, the concept became blurrier instead of sharper; by the 1970s, it seems fair to say that most economists regarded it not simply as wrong in practice but as not even making sense in theory. Only recently has it been shown (by Murphy, Shleifer and Vishny, 1989) that the Big Push concept can be given a simple formalization that is very close to Rosenstein-Rodan's original vision. I want to offer a quick exposition of this formalization, then point out what it shows about the nature of external economies. Consider, then, an economy closed to international trade, in which labor is the only factor of production. The economy produces a large number of goods, each of which receives the same share of expenditure. We imagine that each good can be produced using two different technologies. First, there is a "traditional" technology subject to constant returns; we choose units so that labor productivity under this technology is unity. Second, there is a "modern" technology subject to increasing returns. Specifically,

168

PAUL R. KRUGMAN

suppose that the labor required to produce Q units of each good under the modern technology is

We also suppose - an assumption that as we will see plays a crucial role in this particular model - that for some reason workers must be paid a wage premium to use the modern technology. For the purposes of this example, we let the wage premium be 60 percent - i.e., the wage rate in the modern sector is 1.6 times that in the traditional sector. If no modern technology were available, all goods would be produced using the traditional technology. Assume for the sake of the example that in this case ten units of each good would be sold (and produced with ten units of labor). Now suppose that someone considers producing a good with the modern technology. Given the increasing returns, such a producer would be a monopolist; her price could not, however, be higher than that charged by traditional producers, so if this limit is binding the price of a good produced by modern technology would be the same as that of one produced traditionally. (This is one simple solution to the problem of modeling imperfect competition. I will discuss others below.) If an individual producer shifts over to modern technology, then, she will sell the same ten units that traditional producers would have sold. This will require only seven units of labor. But since the wage rate paid by modern producrs is 1.6 times that paid by traditional, the wage cost of this production would be 11.2, versus only 10 for traditional production. So from the point of view of any individual producer, a shift to modern techniques is not worth making if everyone else stays traditional. But suppose instead that all producers go over to modern techniques. This increases the size of the market, so that each firm will sell more. A shortcut lets us tell just how much more: because of symmetry, the same ten units of labor will now be used to produce each good, which under modern techniques allows production of twenty units of output. The wage cost of twenty units of output under modern techniques, however, is sixteen, versus twenty using the traditional technique. So if everyone uses modern techniques, it turns out to be cheaper to produce using the modern technique - and each firm not only manages to pay higher wages, but earns a profit of four besides. What does this example tell us? There are indeed, just as RosensteinRodan suggested, two equilibria. The low-level trap exists because of a kind of external economy: it is not worth investing in modern technology unless you expect enough other firms to do so as well. But there is, by assumption, no technological external economy; there is no interdependence between

The current case for industrial policy

169

firms' production functions. The externality is entirely pecuniary, and yet it has real welfare significance. On reflection, the nature of this externality is clear. Because there are increasing returns in production, the size of the market matters. The assumption of a wage differential between traditional and modern sectors means that the market is larger, the larger the size of the modern sector; and so there is a strategic complementarity among firms without the need to postulate any external economies in the traditional sense. This model may at first seem a long way from saying anything about industrial policy in modern countries. In particular, the wage differential story, which seems to link the model with Arthur Lewis-type surplus labor concepts, seems at first to restrict the analysis to very underdeveloped nations. It makes the important point, however, that once there are increasing returns external economies need not take a purely disembodied form. And in fact the wage differential is just one of a number of ways in which the size of the relevant market may depend on the decision of firms, giving rise to market-size externalities. We can see this immediately by looking at another, related model, this time of regional inequality. Model 2

Core and periphery

Recently I have argued for the crucial role of increasing returns and market-size externalities in shaping regional economics (Krugman, 1991). I offer in particular a basic story of geographic concentration that relies on the interaction of increasing returns, transportation costs, and demand. Given sufficiently strong economies of scale, each manufacturer wants to serve the national market from a single location. To minimize transportation costs, she chooses a location with large local demand. But local demand will be large precisely where the majority of manufacturers choose to locate. Thus there is a circularity that tends to produce endogenous regional inequality.1 Doing this story right involves taking some care about market structure. (The trick used in the Big Push model doesn't work, for technical reasons; it turns out to work best to imagine monopolistic competition instead.) But the basic idea can be conveyed with a simple numerical example if one is willing to be a little sloppy. Imagine a country in which there are only two possible locations of production, east and west, and two kinds of production. Agricultural goods are produced using a location-specific factor (land), and as a result the agricultural population is exogenously divided between the locations; for the moment we assume that the division is 50-50. Manufactured goods (of which there are many symmetric varieties) can

170

PAUL R. KRUGMAN

Table 7.1. A manufacturing location story

Distribution of manufacturing employment

~ . , - .. , Costs ofc typical firm ifc it produces in East

Both

West

East only

Fixed Transportation Total

4 3 7

8 0 8

4 7 11

50-50 split

Fixed Transportation Total

4 5 9

8 0 8

4 5 9

West only

Fixed Transportation Total

4 7 11

8 0 8

4 3 7

be produced in either or both locations. If a given manufactured good is produced in only one location, transportation costs must be incurred to service the other market. On the other hand, if the good is to be produced in both locations, an additional fixed set-up cost is incurred. The manufacturing labor force in each location is proportional to manufacturing production in that location. Finally, assume that the demand for each manufactured good in each location is strictly proportional to that location's population. The basic idea can then be illustrated with a simple numerical example. Suppose that 60 percent of a country's labor force are farmers, divided equally between East and West. Suppose also that the total demand for a typical manufactured good is ten units. Then if all manufacturing is concentrated in one location, that location will demand seven units (three demanded by the local farmers, four by the manufacturing workers), while the other demands three; if manufacturing is evenly divided between the locations, each location will offer a local demand of five. To figure out what happens, we need to specify the fixed costs and transportation costs; suppose that the fixed cost of opening a plant is four, and that the transportation cost per unit is one. Then we have the situation shown in table 7.1. The table shows the costs to a typical firm of three locational strategies, contingent on the locational strategies of all other firms. Thus suppose that all other manufacturing is concentrated in east. Then our firm will have a local demand in east of seven units, a local demand in west of only three units. If it serves the national market from a single plant in east, it will incur a fixed cost of four and a transport cost of three. This is obviously less than serving the national market from a plant in west,

The current case for industrial policy

171

which will have the same fixed cost plus a transport cost of seven; it is also less than building a plant to serve each local market, which saves the transport cost but incurs a double fixed cost of eight. In this case, then, the typical firm will choose to produce in east for a national market. If each firm concentrates its production in east, however, then manufacturing production as a whole will be concentrated in east - which is what is assumed. So concentration of production in east is an equilibrium. But it is not the only equilibrium. As the rest of the table shows, if manufacturing is concentrated in west, each firm will similarly also want to concentrate its production in west. And if production is split between east and west, each firm will want to split its production, too. So in fact all three distributions of production - all in east, all in west, and a 50-50 split - are equilibria in this example. Clearly, this example is very similar in its basic logic to the Big Push story. What is crucial is the combination of increasing returns and a channel through which the decisions of firms can affect the sizes of each others' markets; in this case factor mobility between regions rather than extraction of surplus labor from a traditional sector plays the key role. Model 3

Labor markets and industrial specialization

In this third model I want to show why a particular industry might want to concentrate in a particular location because of a somewhat different kind of market-size effect: the advantages of a large market for specialized labor.2 This is, of course, a traditional argument, going back to Marshall; yet it is also an argument that had until recently fallen into disuse because economists were no longer sure that it made sense. Again, a simple example shows that it makes perfectly good sense. Imagine for a moment that there is some industry that consists of just two firms, each of which can produce in either of only two locations. These firms both use the same distinctive kind of skilled labor. For whatever reason, however, the firms' demands for labor are not perfectly correlated. For example, they may produce differentiated products that face uncertain demand; or they may be subject to firm-specific production shocks. Whatever the reason, the labor demand of the firms is both uncertain and imperfectly correlated. To make matters more concrete, suppose that each firm may experience either "good times," in which it would like to hire 125 specialized workers at the going wage, or "bad times," in which it would like to hire only seventy-five. We also suppose that there are 200 of these workers in total, so that average demand for labor equals supply. (In this example I take the wage rate for the specialized labor as given, so that there may be

172

PAUL R. KRUGMAN

either excess demand or excess supply for labor. If you like, imagine a wage bargaining process that sets the wage at an expected market clearing level before the shocks to labor demand are revealed - not too unrealistic an assumption. This assumption is not, however, essential; see Krugman, 1991b.) Now we may ask: will firms and workers be better off if the two firms choose different locations - each thus forming a company town with a local labor force of 100 - or if the two firms choose the same location, with a pooled labor force of 200 that can work at either firm? First, consider the situation from the firms' point of view. If each has its own town, with a labor force of 100, then it will be unable to take advantage of its good fortune when labor demand is high: during good times, there will be an unfillable excess demand of twenty-five workers. If the two firms are in the same place, however, then at least occasionally one firm's good times will coincide with the other firm's bad times, and there will be additional workers available. Next consider the situation from the workers' point of view. If they live in a company town, then the firm's bad times are their bad times too: whenever the firm has low labor demand, twenty-five workers will be laid off. If the firms are in the same place, then at least sometimes one firm's bad times will be offset by the other firm's good times, and the average rate of unemployment will correspondingly be lower. This is a pretty trivial example. Yet it shows yet another context in which market-size effects generate what amounts to an external economy. Note in particular that the example shows that uncertainty alone won't generate localization. You need increasing returns as well. The key point is that in order to make a pooled labor market advantageous I needed to assume that each firm had to choose one location or the other, not both. If each firm could produce in both locations, or for that matter if each firm could be split into two identical firms, one in each place, then the full "portfolio" of firms and workers could be replicated in each location, and the motivation for localization would be gone. But the most natural justification for the assumption that firms do not locate in both places is that there are sufficient economies of scale to militate for a single production site. So it is the interaction of increasing returns and uncertainty that makes sense of Marshall's labor pooling argument for localization. What have we learned? I have just offered three models (out of many possible models) that show how market-size effects can create true social external economies even if there are no pure technological spillovers. The examples all look quite a

The current case for industrial policy

173

lot like old-fashioned development economics, which is not an accident: in my view, the development theorists of the 1940s and 1950s were groping toward precisely this insight, without quite finding the technical tools they needed to express this insight clearly. What the models show is that external economies need not be an abstract and ethereal concept: they can be very straightforward and visible. And two out of the three models, the two that seem most relevant to developed countries, suggest that these external economies are most likely to occur at the regional or local rather than national or international level. In the next section of the chapter I take this last observation as a lead and present some suggestive discussion of the importance of external economies as illustrated by the existence of local and regional concentrations of production. 3

Geographic evidence for external economies

The best evidence for the practical importance of external economies is so obvious that it tends to be overlooked. It is the strong tendency both of economic activity in general and of particular industries or clusters of industries to concentrate in space. In this section I briefly review some of the evidence on geographic concentration at three levels - regional, urban, and industrial - and suggest what it implies about the nature and importance of external economies. Regional concentration Both the United States and Europe exhibit a strong core-periphery pattern in economic activity, albeit in different forms. In the United States, what we observe is that even though the country as a whole is not very densely populated, most of the population is concentrated either in the traditional manufacturing belt of the northeast, or in a few more recent sunbelt areas. In Europe the population is less unevenly distributed, but there is instead a strong core-periphery pattern in per capita income. The traditional US manufacturing belt essentially includes the northern areas that were settled before the building of the railroads. For nearly a century - from about 1870 to about 1960 - this area dominated US manufacturing, in spite of a steadily declining resource base. The distinctive role of the manufacturing belt emerged between about 1870 and 1910. In 1870 the northeast and East North Central regions - within which the emerging manufacturing belt lay - accounted for 44 percent of US "resource extraction" employment (agriculture, mining, forestry, fisheries). By 1910 this share had already fallen to 27 percent; yet these regions still

174

PAUL R. KRUGMAN

accounted for 70 percent of manufacturing employment. And this dominance persisted for a very long time. PerlofF et al. (1960), using a slightly more precise definition, estimated that as late as 1957 the manufacturing belt still contained 64 percent of US manufacturing employment - only slightly reduced from its 74 percent share at the turn of the century. Why did the manufacturing belt play such a dominant role for so long? It was clearly not a case of an enduring advantage in natural resources: the manufacturing belt persisted even as the center of gravity of agricultural and mineral production shifted far to the west. The answer obviously rests in the importance of external economies: each individual manufacturing facility stayed within the manufacturing belt because of the advantages of being near other manufacturers. And the apparent incentive for manufacturers to cluster together explains the persistence of the manufacturing belt even after the bulk of US primary production had shifted to other regions. Once the belt had been established, it was not in the interest of any individual producer to move out of it. The population distribution of Europe does not at present exhibit anything like the unevenness of the American distribution. Within countries there are some core-periphery patterns: the continuing pull of greater London or of the Ile-de-France are obvious. But in spite of considerable migration from south to north in the 1960s and early 1970s, there has been no wholesale concentration of population and employment in the areas of early industrialization. The reason is obvious: Europe has historically been far less integrated, both in terms of factor mobility and in terms of trade, than the US. On the other hand, Europe is characterized by a very strong centerperiphery pattern when one considers, not population, but purchasing power. Interregional income differentials within Europe are much larger than within the US, and they are closely associated with geographical position. There is a strong correlation between measures of centrality (i.e., access to markets) and per capita income. It is not hard to develop a variant of the core-periphery story that does not need factor mobility to function. Suppose that through forward and backward linkages, a region that has accumulated a lot of physical and human capital tends to have a higher, rather than a lower, rate of return on investment than a region where these factors are scarce. And suppose that the rate of capital accumulation itself depends on the rate of return. Then one can imagine an unequalizing spiral in which an economy becomes endogenously differentiated into rich -and poor nations. Admittedly, in Europe it is likely that the main causation runs the other way. That is, northwestern Europe is relatively rich for reasons that have to do more with culture than with geography. And as a result, the richer

The current case for industrial policy

175

regions are also relatively close to the large markets, which are themselves. Nonetheless, it is hard to reject the view that external economies have played at least some role in the process of differentiation.

Urbanization A colleague of mine once asked what evidence would make me think that increasing returns are actually of any economic importance. When I replied with one word - "cities" - he was offended (and remains offended to this day). It seems ridiculous to suggest that a major area of controversy could be settled with an appeal to such an obvious fact of life. Yet the importance of increasing returns in general and external economies in particular is demonstrated more clearly by the concentration of population in cities than by any other evidence. Urban external economies have actually been the subject of a large empirical literature; see in particular Henderson (1988). The strength of such external economies is attested by the fact that the largest, most expensive, and most congested city in the industrial world - Tokyo - is actually growing at the expense of its smaller and more liveable rival, Osaka.

Industrial concentration Porter (1990) has recently emphasized the point that international competitiveness is often the product of successful geographical clusters within countries. He uses the classic example of Italian ceramic tiles to make his point, but in fact examples of geographic clustering are pervasive. Famous examples like Silicon Valley, Milan's fashion industry, or London's financial service industry are outnumbered by less glamorous concentrations like the Belgian carpet industry or the group of property insurance companies in Connecticut. What is particularly noticeable is that, on careful examination, industrial clusters that are hidden by conventional statistics often become visible. Some are concealed by excessive aggregation. For example, the software industry in the United States appears at first sight to be dispersed among a number of sites. The specialized lines within that industry, however, are remarkably concentrated: operating systems in Washington, word processing and networking in Utah, spreadsheets in Massachusetts. Others are concealed by misclassification. For example, the New York city of Rochester, home of Kodak, Xerox, and Bausch & Lomb, appears at first to have a diversified industrial structure; but in fact it is essentially focused on industrial applications of optics.

176

PAUL R. KRUGMAN

What the evidence shows Geographic concentration is the most convincing evidence of the importance of external economies in real economies. In fact, the importance of such concentration at various levels is in effect a decisive refutation of the competitive model of economic equilibrium. Equally important, by looking at patterns of geographic concentration, especially at industrial clusters, one has a reasonable hope of identifying sectors subject to particularly important external economies. In a remark that he will long regret, the current chairman of the US Council of Economic Advisers is reported to have rejected industrial targeting by saying that silicon chips were no more important to the economy than potato chips. The clear response is that there is a Silicon Valley, but not a Potato Valley; the concentration of the semiconductor industry in a narrow area, and the willingness of firms to pay high costs in order to remain in that area, amount to clear evidence of important externalities. Similarly, the optics concentration in Rochester offers clear evidence that instruments, certain business machines, and photographic equipment constitute a valid industrial cluster with significant linkages in the Hirschmann sense. One can certainly go beyond this level of observation. By studying the motives of firms that cluster together, the nature of the external economies can often be pinned down. For example, surveys of Silicon Valley firms suggest that the pooled market for specialized labor is a key motivation for firms to remain there. While solid, reliable quantitative estimates of external economies are still a long way off, we are by no means as much in the dark as many economists, myself included, have supposed. But how can this evidence be used to devise policy? That is where we must now turn. 4

Criteria for industrial targeting

The analysis above suggests a criterion for industrial policy, and a possible way to make that criterion operational. The criterion is the following: an industry is a likely candidate for special attention - call it a "strategic" industry - if it seems that the industry's success is largely a social construction - that individual firms do not stand or fall on their own efforts, but depend on the mutually reinforcing effects of each other's success. This is just a non-technical way of saying that positive external economies are the key, but putting it this way may help to make the idea seem less abstract. In most cases, the best evidence for the importance of external economies will come from geographic clustering. And such geographical clusters will in fact help us to define what is an industry. Simply observing a cluster is

The current case for industrial policy

177

not, however, enough: one must then ask why the industry is clustered, and make a judgment about whether the external economies, technological or market-size in nature, are sufficiently important to warrant government support. It is immediately apparent that some familiar sectors will fit the definition of strategic industries. Semiconductors (at least the sophisticated ones that are still designed and to at least some extent manufactured in Silicon Valley) fit the bill. So do software, minicomputers, scientific instruments. Less familiarly, financial services, high-end furniture, and some agricultural products also would qualify. Of course in order to be useful a criterion must exclude some industries as well. Autos, whose production is increasingly being dispersed, do not seem to be an industry with many external economies. The same is true of tires, and of textiles. And perhaps surprisingly, passenger aircraft - produced in only a few locations, but by only a few giant firms - also show little evidence of external economies. (In this sense an external-economy approach to industrial policy yields different recommendations from the BranderSpencer approach.) To readers familiar with the recent work of Porter (1990) this analysis will seem familiar. Indeed, Porter's analysis of international competition is largely a discussion of the importance of geographically restricted external economies. Porter does not follow through and assert that he has offered a justification for industrial policy, but in effect he has. The point is that it is in fact easy to offer a justification for industrial policy, and not even very hard to specify plausible targets for such a policy. The difficult questions are how to implement such a policy in practice, and how to manage the political economy of such a policy in such a way as to avoid the usual mistakes. Fortunately, this chapter is a conceptual essay rather than a blueprint. Thus at this point I break off, and leave the hard issues for another time. Notes 1 The story as just sketched out here focuses entirely on "backward linkages" from production to demand (which are indeed the only effects in the Murphy et al. [1989], Big Push). One could also introduce "forward linkages" from production to supply. Indeed, in the formal model introduced by Krugman (\99\a) such forward linkages emerge as a necessary part of the story. The important point is that Hirschmann (1958) was basically right: forward and backward linkages are useful and indeed inevitable concepts in an economy subject to increasing returns. 2 One can also explain localization via market-size external economies involving the production of non-traded intermediate goods. This is a familiar point in

178

PAUL R. KRUGMAN

urban economies — see, for example, Fujita (1989) - which has not been fully appreciated in the rest of economics.

References Baldwin, R. and H. Flamm (1989), "Strategic Trade Policy in the Market for 30-40 Seat Aircraft/' Weltwirtschaftliches Archiv, 125 (3), 484-500. Baldwin, R. and P. Krugman (1988a), "Market Access and International Competition: A Simulation Study of 16K Random Access Memories," in R. Feenstra (ed.), Empirical Methods for International Trade, Cambridge, Mass.: The MIT Press. (1988b), "Industrial Policy and International Competition in Wide-Bodied Jet Aircraft," in R. Baldwin, (ed.), Trade Policy Issues and Empirical Analysis, University of Chicago Press. Brander, J. and B. Spencer (1983), "International R&D Rivalry and Industrial Strategy," Review of Economic Studies, 50, 707-22. (1985), "Export Subsidies and Market Share Rivalry," Journal of International Economics, 18, 83-100. Dixit, A. (1988), "Optimal Trade and Industrial Policies for the US Automobile Industry," in R. Feenstra (ed.), Empirical Research in International Trade, Cambridge, Mass.: The MIT Press. Dixit, A. and G. Grossman (1986), "Targeted Export Promotion with Several Oligopolistic Industries," Journal of International Economics 21, 233—50. Dixit, A. and A. Kyle (1985), "The Use of Protection and Subsidies for Entry Promotion and Deterrence," American Economic Review, 75, 139-52. Eaton, J. and G. Grossman (1986), "Optimal Trade and Industrial Policy under Oligopoly," Quarterly Journal of Economics, 101, 383-406. Fujita, Masahisa (1989), Urban Economic Theory, Cambridge University Press. Henderson, J.V. (1988), Urban Development: Theory', Fact, and Illusion, New York: Oxford. Hirschmann, A. (1958), The Strategy of Economic Development, New Haven: Yale University Press. Horstmann, I. and J. Markusen (1986), "Up your Average Cost Curve: Inefficient Entry and the New Protectionism," Journal of International Economics, 20, 255-49. Krugman, P. (1983), "Targeted Industrial Policies: Theory and Evidence," in Federal Reserve Bank of Kansas City, Structural Change and Public Policy, Kansas City: FRBKC. (1984), "Foreign Industrial Targeting and the US Economy," Brookings Paper on Economic Activity, 77-121. (1991), "Increasing Returns and Economic Geography," Journal of Political Economy, 99(3), June, 483-99. Murphy, K., A. Shleifer, and R. Vishny (1989), "Industrialization and the Big Push," Journal of Political Economy, 97, 1003-26. Myrdal, G. (1957), Economic Theory and Underdeveloped Regions, London: Duckworth.

The current case for industrial policy

179

Perloff, H. (1960), Regions, Resources, and Economic Growth, Baltimore: Johns Hopkins. Perloff, H., E.S. Dunn, E.E. Lampard, and R.F. Muth (1960), Regions and Economic Growth, Lincoln, Nebraska: University of Nebraska Press. Porter, Michael (1990), The Competitive Advantage of Nations, New York: Free Press. Rosenstein-Rodan, P. (1944), "Problems of Industrialization in Eastern and South-Eastern Europe," Economic journal, 53, June—September, 202-11. Smith, A. and A. Venables (1988), "Completing the Internal Market in the European Community: Some Industry Simulations," European Economic Review, 32, 1501-25. Venables, A. and A. Smith, (1986), "Trade and Industrial Policy under Imperfect Competition," Economic Policy, 1, 622—72.

CHAPTER

8

The case for bilateralism RUDIGER W. DORNBUSCH

Until recently, US commercial policy had as its central tenet GATT, multilateralism and ////conditional most favored nation treatment. That policy stance emerged in response to the massive and chaotic trade disruption of the 1930s which was an important part of the Great Depression.1 The objectives of the new post-war world are well summarized by Jacob Viner (1944, p. 53): The American objectives with respect to' the pattern of post-war international economic relations are in their general outlines clear. These objectives are a post-war world in which: trade barriers will be as moderate and as nondiscriminatory as governments and peoples, including our own, can be persuaded to make them; exchange rates between national currencies will have assumed stability; currencies will be freely convertible into each other, at least with respect to transactions on ordinary current account; longterm capital will move from capital-rich to capital-poor countries in quantities and on terms mutually satisfactory to creditor and debtor; ways of international collaboration aiming at maintenance of high-level employment will be devised; abundant scope will be preserved for the exercise in the international economic field of competitive enterprise, subject to standards and rules intended to prevent abuse but not to stifle and suppress private initiative. Multilateralism set the system apart from Imperial Preferences of the 1930s and from the bilateralism practiced in the Schachtian system. GATT has been the chief vehicle for executing the multilateral trade liberalization that underlay the US conception of an open world economy. And GATT has served us well! In progressive rounds of tariff cutting, tariff barriers 180

The case for bilateralism

181

among industrial countries have come down to negligible levels. And trade expansion has been vigorous, running far ahead of the growth in real GNP. Why then shift to another approach, bilateralism, putting quite possibly at risk not only further progress on the tried and proven course but perhaps even some of the accomplishments of the past five decades? The case made here for bilateralism in no way questions the usefulness, past, present, and future of the GATT system. Rather it advocates that there is room for, and indeed a good reason for a parallel track move on to a bilateral basis. The case for this extra approach arises in two directions. First, relative to Japan the industrialized countries' trade situation is thoroughly unsatisfactory: Japan continues to be substantially closed de facto^ decades of trade liberalization notwithstanding. Hence there is a need for a new approach that takes on Japan firmly and quietly to offer the choice of integration like every other country or else exclusion. Second, on a regional basis there is important room to push ahead trade liberalization on a wider and deeper basis than appears immediately possible in the current GATT round. Such regional moves can and must be conducted in a GATT compatible way. Their attraction to the system is that they may well motivate further progress in the framework of GATT in that they stimulate competitive liberalization. The regional opportunities deserve special attention today because of the opportunities and needs of Eastern Europe and Latin America. In these regions the pendulum, has swung to favor free markets and open economies. At the same time the pressure of migration forces a recognition that either there is a bold move to open trade or else increasing migration pressure becomes unavoidable. A move to unrestricted free trade suddenly emerges as a plausible response to the circumstances. The notion of regionalism involving developed and developing regions is radically new (preferences, of course, are a one-way precedent) and offers a special problem but so far the political reception has been surprisingly favorable. This chapter sets out the case for bilateralism and regionalism - the case is made in terms of two issues: a US—Latin American trade strategy and the question of what policies to use to deal with the "Japan problem." Before going into the details of these issues it remains to emphasize the basic motivation for defending these policy approaches: the underlying preoccupation is how to achieve freer trade; it is definitely not some obsession with managed trade, a fondness for protectionism, or dissatisfaction with the accomplishments of multilateral diplomacy and GATT. Freer trade should be pursued on all fronts and with all means. Some experimenting with new approaches can be seen as a threat only by the most conservative defenders of the status quo, i.e., professional multilateralists. Fortunately they do not have a monopoly on either common sense or commercial policy.

182

RUDIGER W. DORNBUSCH

Trade theory and trade policy The professional prejudice of economists is that wide open trade is best and that it is unwise to start compromising on this position. Fortunately that position, while often challenged, commands such broad support in the profession that politicians have been wary of challenging it frontally. Theory gives us relatively little guidance on how best to negotiate further trade openings, starting from where we are. The advice is clearly not the same for each country, large or small, and it is not independent of strategic interactions among the players. Patently, unilateral trade liberalization is preferred, particularly in small countries, to the status quo of high protective barriers. But, as Wonnacott and Wonnacott (1981) have shown, bilateral trade liberalization is even better because it yields market access as one of its payoffs. For a large country it is clearly the case that on strictly economic grounds there is an optimal tariff - free trade is demonstrably not the best strategy. But, caught in a tariff-ridden equilibrium, even a large country should want to go some way toward trade liberalization, though not necessarily all the way. Keeping one market restricted, not counting any dynamic benefits from learning, research and development etc., still yields terms of trade improvement and hence has a positive aspect. Of course, on the other side is an excluded seller who, if that country practices free trade, must suffer a worsening of its real income. Far more complicated is the situation of trade blocs. The two effects at work are trade creation and trade diversion. The only guidance we have comes from theorems of Kemp (1976) and Kemp and Wan (1972). Specifically, regional liberalization that keeps external trade unchanged (by suitable tariff adjustment) can be Pareto improving; it improves world welfare by raising welfare among the members without reducing it on the outside. That is a powerful result. With a presumption that there are gains from freer trade to be reached, the economists' presumption is to push trade liberalization at every margin. There is no theorem around to argue that pushing regional ahead of the multilateral trade liberalization is welfare reducing and no result in game theory to argue that the mechanism is inevitably unstable. In the area of trade policy a good dose of common sense must fill the gap left by an absence of hard theory that might otherwise set the guideposts. Bilateralism, regional trade arrangements and GATT Concerns about pursuing a regional trade liberalization stem to some extent from fears about the dynamics of the world trade game. If the

The case for bilateralism

183

United States were to pursue a bilateral route and set up preferential trade blocs, would not there be a risk of the formation of other, competing blocs? And if that were to happen, could one be certain that there would not be a 1930s style decline in world trade? The opposition to multilateral approaches is thus more than anything a reaction to the untried. But it also seems that the principle of bilateralism is far more offensive and readily challenged than any particular implementation, and more so when it is suggested for the case of the United States than anywhere else. The principle of multilateralism and the liberal trading system are so sacred because the United States sees itself as the custodian of a policy tradition that it helped create in the 1930s. The system has served us well and still does, but it certainly does not represent a unique track strategy toward freer trade. The policy debate in the United States has been until quite recently hostile to regional trade arrangements or bilateralism. The status quo position of those who favor an open trading system is the GATT-based multilateral approach. Few among the protagonists of the status quo ask of the GATT process where the gains will be and when they come, if at all negotiations take a decade or more. The beneficial effects of the status quo are taken for granted and the only counterfactual is a world without trade. Violations of the open trading system by Japan are played down by arguing that Japan is "already" opening up. Bilateralism practiced abroad, as in Europe's post-war experience or the joining of EFTA and the EC, is welcomed, even if the same policy used by the United States is seen as a threat to the system and the process. Any bilateral proposition is presumed subversive and argued to death on grounds of impracticality, political unfeasibility, here or abroad, or insignificance of effects or on principle. After free trade with Mexico came on the agenda, the special aspects of the relation were played up to justify the exception. Free trade with all of Latin America is certainly not an accepted strategy. Regionalism (and even more so, bilateralism) has an unnecessarily bad name. The gains from the multilateral approach in the past have been significant and are not in question; but the pace at which the process delivers extra gains is slowing down and the liberal trading system is eroding. Moreover, trade policy initiatives in Europe and in Asia, are working quite possibly to the detriment of US-located production. In such an environment a search for a more effective US trade policy assumes special importance. To reap the real income gains that freer trade can offer, the United States must not be blocked by limiting negotiations to a potentially unproductive GATT process. Moreover, if the system is open in the sense of allowing conditional MFN access, a bilateral initiative can become a vehicle for freer trade on a regional

184

RUDIGER W. DORNBUSCH

Table 8.1. Post-Tokyo Round tariff rates (percent) Semi-finished Raw materials manufactures 0.2 0.5 Canada 0.5 0.2 EEC US

Japan

4.5 4.6 8.3 4.2

Finished manufactures 8.0 6.0 8.3 6.9

Source: Economic Report of the President, 1989, p. 156.

basis. Regional initiatives are a policy option that pushes the world economy toward freer trade, complementing the GATT process where it heads in that direction or filling a vacuum in the quest for freer trade in those areas where GATT has tacitly accepted the status quo or even a slide into protectionism. Supporters of a multilateral approach - GATT, the Uruguay Round, unconditional MFN - argue two points. One is that the approach has served us well in the past. They can rightly point to the dramatic reduction in tariff barriers. Remaining tariff barriers on most goods are negligible even if the effective rates of protection are, of course, higher than what might appear from table 8.1. The reason, of course, is the tariff structure which taxes imported intermediate goods at rates lower than final goods. This consideration is particularly relevant for developing countries whose exports are more likely to be in the industries with relatively high rates of effective protection.2 Moreover, hard-core protection in the form of nontariff barriers and quotas add to the negative side of an otherwise very open trading system. Second, proponents of multilateralism argue that alternative approaches, including a dual track strategy, implicitly weaken if not outright undermine the multilateral approach. Faced with this choice they would throw their undivided support to the multilateral approach. Crediting on an exclusive basis the multilateral trading system with reconstruction of world trade after the breakdown of the 1930s, and with the growth of world trade far higher than world real income, would certainly be overdoing it. Several qualifications to that perception are necessary. To start with, many if not most LDCs never opened up in the first place. Trade liberalization in developing countries is a fashion of the 1980s (see table 8.2). Moreover, the reconstruction of world trade started on a bilateral basis. In Europe a dual track strategy has been followed all along. In the past thirty years, GATT notwithstanding, Europe has used regional approaches

185

The case for bilateralism Table 8.2. Trade restrictions in

LDCs Per capita income

$500-$1000

$1001-$1500

$1501-$5000

Average tariff rate (%)

41

54

34

Tariff positions subject to NTBs (%)

77

83

36

Source: I M F Issues and Developments in International Trade Policy Occasional Paper

No. 63 Tables A22 and A23.

over and over again, from the European Payments Union in the 1950s discrimination against the United States explicitly condoned as a strategy of rebuilding industrial Europe - to the Common Market and EFTA, and to the Europe 92 initiative. Few questions have been raised about the wisdom of that strategy, whether it amounted to deepening the extent of integration or widening the scope to include Greece and Portugal, North Africa, and Turkey. Developments in Eastern Europe offer the prospect that this region will soon enjoy a preferred trade status with the Common Market just as all of EFTA already does. The major regional effort in which Europe is involved removes any conceivable argument that a US free-trade bloc policy would undermine an otherwise intact multilateral system. The Europe 92 project so clearly foreshadows trade discrimination that the EFTA partners are scrambling to get inside the deal for fear of being left out in the cold. A US policy of building a trade bloc is certainly not the first or even a decisive trespass on a systenrof more open trade. On the contrary, it is giving energetic support to the idea of trade liberalization. In fact, it is the best idea in free-trade thinking for a long time. Regional integration represents a parallel track strategy to the GATT process. GATT has served us well in the post-war period, especially in the North Atlantic trade relations. It can yield extra benefits in the ambitious targets set in the Uruguay Round liberalization for agriculture, and trade in services and dispute settlement. GATT explicitly recognizes regional free-trade agreements as an exception to the MFN rule; Europe has taken advantage of that exception throughout the post-war period and has derived peace and prosperity from that strategy. The United States can only gain from emulating the European precedent. Finally, as we will see below, Japan never opened up in the first place. If there is any threat to the multilateral, open system it is posed by the continuing closedness of the Japanese market, not by the pursuit of freer

186

RUDIGER W. DORNBUSCH

trade on the part of the United States. New initiatives in commercial policy should be seen in the context of these three issues: they are necessary to achieve significant progress in the direction of further opening of trade. US free trade with Latin America? Free trade with Latin America and the Caribbean is in the national interest. Trade policy should create good jobs rather than focusing primarily on the protection of bad jobs. Free trade with Mexico represents a step in the direction of enlarging US export markets. Because the US is already wide open to world competition, and especially to Mexico, further bilateral trade liberalization cannot have dramatic dislocation effects but will on balance create more and better jobs in the US. Several important trends in the world make pursuit of free-trade agreements with Latin America an important, productive trade strategy for the United States. Most impressively, all of Latin America is reforming and modernizing in a way we have not seen since early in this century. The momentum, direction, and success of these reforms must be strengthened by a partnership that makes it difficult for Latin America to move back, and delivers tangible benefits to offset the immediately visible costs of opening up. Having encouraged the modernization, it would be unwise for the United States to walk away from participation and partnership in the process. Economic modernization demands a wider scope for economic activity and the regional level offers the most concrete setting in which to visualize the benefits and hence find the willingness to make the concessions. Moreover, Latin America is our fastest growing market. Of course, free trade with Latin America cannot be a panacea for economic US problems; but even if Latin America is not most of the solution of our problems, it still is an important market and our trade posture has important implications for their prosperity and hence our security. It would be a mistake to write off Latin America as an important partner. One must also explore a favorable longer-run perspective of Latin American markets. Latin America's population is almost twice that of the United States and economic growth in the years to come will be substantial (see table 8.3). If Latin America recovers economically, and the United States can certainly invest in that prospect, there is ultimately a very significant market for US exports. A clear demonstration is given by the (unilateral) Mexican trade liberalization in 1988. Within a single year Mexican imports from the United States increased by $6 billion. That represents a larger increase than the entire prospective gain from the US-Canadian free-trade agreement.

The case for bilateralism

187

Table 8.3. Comparative data, 1989 Population (Mill.) US Canada W. Hemisphere Brazil Mexico

248.8 26.2 436.0 147.3 84.6

GDP (Bill. $US)

GDP Per Capita (JUS)

5,202.4 498.6 837.0 374.1 170.0

20,910 19,031 1,918 2,540 2,010

Source: World Bank.

The Mexican trade liberalization is of interest because it highlights two features. First LDCs' markets are extraordinarily closed and hence, in a situation of trade liberalization, offer a potential for very major increases in US exports. The second point is that the first round in Mexican liberalization was multilateral and hence, even though it benefited overwhelmingly the United States, also allowed other countries to participate. A free-trade agreement with Latin America would yield for the United States a more privileged status, at least for a while until a move toward multilateralism comes about. It might be argued that the Mexican gains are already in place and that Mexico is one of the larger countries so that there is little left. Of course, there is Brazil which is far larger and offers a very striking opportunity in the perspective of the next twenty years. Brazil does have a strong interest in unimpeded access to the US market and it offers an important market for US exports in return. The current level of trade with Brazil is far below the potential of the country as a market. This point is quite apparent from table 8.4. Brazil ought to be opened up. Another argument might be that Mexico was getting ready for an agreement after years of successful reform (much as Chile), but that the same cannot be said of most of the region. The response surely must be that we are interested in export markets and we are interested in regional stability. We are now trading with these countries and nobody can possibly argue that existing trade restrictions, here or there, help make their economies or their societies function better. It can be argued, however, that more intensive trade and investment links will. Hence, particularly for economies where much is to be accomplished - notably Brazil - the prospect of an FT A raises the ante and will enhance and speed up the need for reform. Nobody is thinking of political union or EC-style deep integration for which a very high level of community is essential. What is at stake is the removal of impediments to trade and investment.

188

RUDIGER W. DORNBUSCH

Table 8.4. US Trade: 1990 (Bill. $US)

World Canada

Japan EC W. Hemisphere Mexico Brazil Chile

Exports

Imports

393.6 83.0 48.6 98.0 54.0 28.4 5.1 1.7

517.0 93.8 93.1 95.5 67.2 30.8 8.6 1.6

Source: IMF.

Trade and employment effects Much of the controversy surrounding the proposed FTA with Mexico, and surely an FTA with all of Latin America, is misplaced. Free trade with Mexico will not bring about "an economic and social disaster for US workers and their communities" as argued by the AFL-CIO. 3 Any job losses are bad news at a time where real wages are depressed and employment at best stagnant. But these issues must not become an argument for stopping a good move in trade policy, which creates good jobs at home. Even if trade liberalization inevitably causes some dislocation, that must not mesmerize us into maintaining the status quo for poor jobs. It is bad trade policy, to keep workers and their children in poor jobs and even pervert protection to the point where we attract immigrants to perform this work. We should not let go of competition. But, equally important, displaced workers should get adjustment programs, skill building, and education to help them get into good jobs. We do not perform at all an adequate job in this area. The scope for worker training and adjustment assistance should be enhanced and broadened to include both trade-andproductivity-caused losses in jobs. Concerns about the effect of free trade on US jobs focus on the low level of Mexican labor cost. True, Mexican labor costs far less than US labor but that is true with and without free trade. Short of closing our economy we will be unable to escape from the increasing ability of developing countries to produce manufactures at highly competitive prices. But we can turn to our advantage the situation by gaining access to their markets with our goods. The dramatic effect of opening markets in Mexico to our exports is already amply demonstrated by the experience of the past four years (see table 8.5).

The case for bilateralism

189

Table 8.5. US Nonoil Trade With Mexico and US Job Creation (Billion SUS) US Exports 1986 1990

US Nonoil Imports

Net nonoil Trade Balance

12,391 14,040 -1.649 28,375 25,206 3,169 Net US Job Creation: 1986-90 / 30 Jobs per $1 Mill. 25 Jobs per $1 Million 144,531 120,450

Note: 30 jobs per $1 million exports is the number used by the Economic Policy Institute.

Several factors support the assertion that an FTA with Mexico cannot bring major harm and is very likely to be beneficial. First, Mexico is very small relative to the United States. Any significant increase in Mexican exports (measured on the US scale) would increase labor requirements and wages in Mexico dramatically and thereby squash competitiveness. Second, although Mexican labor costs are low relative to those in the United States, these labor costs also reflect in many cases a low level of productivity and in some areas such as textiles the very low quality of output. The quality factor especially is a major obstacle to a dramatic development of Mexican exports. Third, the United States is already a very open economy. Competition from abroad is not a threat but a complete reality. Protection continues only in a few sectors, not across the board in all lines of activity. Moreover, Mexico enjoys already a privileged position both as a result of the GSP and more importantly as a consequence of the maquila program which exempts reimports from US duties except for the Mexican value-added component. The combination of factors reduces the extra impact of US trade liberalization to a few sectors and to a total effect that have simply no chance of amounting to much in terms of aggregate employment or output. And what goes for Mexico also goes for other countries in Latin America. In the 1980s a vastly overvalued dollar brought about abnormal import competition and hence job losses and a reduction in real wages in the US labor market. No doubt, some of the competition came from Latin America and hence helps explain the 0.5 to 1 percent decline in real wages attributable to import competition.4 But the remedy is not closing down, but rather to seek new markets and thus create extra opportunities for manufacturing which have a demonstrable positive effect on earnings. With

190

RUDIGER W. DORNBUSCH

Mexico that has now already happened and more is to come as Mexico prospers in the 1990s; much more is to be gained from access to Brazil and the other protected markets of Latin America. Aggressive pursuit of market access is the only sensible strategy for a country like the United States which is already open and does not wish to close. Negotiations A useful approach to Latin America would be to target Mexico and the Southern Cone group (Brazil, Argentina, Uruguay) for a free trade area, including if possible Canada. Once the principle is accepted and the outline of an agreement is in place other Latin American countries could join. The dynamics of concluding an FT A are important because of the trade diversion aspect.5 It would be a mistake to have a long drawn out process with much uncertainty about who can or cannot join in the end. Trade diversion in the interim and a halt to investment pending clarification could create unnecessary problems for countries who ultimately may be members but are not in the early rounds at the negotiating table. This argument enhances the case for a blueprint negotiation with Mexico and for an early and short window of time during which other countries can adhere. Anything more than a two-year period would be unnecessarily and unproductively long. The fast track approach is clearly the right way of proceeding. Negotiations in any other setting are an invitation to cannibalize the agreements with exemptions and loopholes so that what is left in the end may not be worth having, or worse. The Caribbean with its exception of sugar, textiles, and more is a case in point. In a setting other than fast track it is virtually a foregone conclusion that politicians will have to seek "relief" for their customers and special advantages; they will be judged not by the total result but by the deviation from the norm that they can secure. The result, unfailingly is a worthless agreement. Even as an FTA with Mexico is being implemented, it must be made clear that Latin America should practice opening and clearing the tradefieldof the myriad of obstacles in place today. Countries like Brazil have a lot of work to do in removing the extensive controls on trade and the tariffs. Other countries like Chile or even Argentina are already far ahead. But it would be a mistake to rush into an agreement with say Chile because the country is already open but proceed slowly with a country like Brazil. Chile has boycotted the Mercosur on the grounds that it was in an express lane for an FTA with the United States and that its neighbors were too unstable. The United States should maintain an environment where only good neighbor practice collects rewards.

The case for bilateralism

191

The focus on some countries being "ready" for an FTA and others not yet is largely misdirected. The United States trades today with many countries who have not achieved financial stability or significant modernization. There is no plausible reason why we should want to maintain tariffs and quotas on their trade, or should want them to restrict our access to their market just because they have not yet reached an advanced level of stability. The guiding principle should be much more our interest in market access and the real possibility that the prospect of a free trade agreement can help exert important pressure to stabilize. Broader regional objectives Narrow economic benefits from freer trade - a better resource allocation - are not the only goal to be derived from a regional strategy. The aims go further to political objectives that may well be far more important. When the United States sponsored French-German integration in the immediate post-war period it initiated a strategy that has made totally inconceivable the previous pattern of recurrent wars. Likewise, in USLatin American relations, political objectives must play a central role.6 Latin America represents a national security issue and as such deserves attention. If freer access to the US market, and less restriction on trade, at least from the United States, can help reduce the economic problems of Latin America then a useful purpose is immediately served. One only needs to contemplate the risks of Latin America turning in the direction of poverty and destitution experienced in Haiti to recognize that there would be serious risks, certainly in terms of migration, for the rich North. With the spreading of democracy in Latin America, US concerns have focused on deepening democracy and broadening the focus of trade policy to economic and special objectives including in particular labor standards and the environment. Opening and modernization of Latin America through trade and investment links is a certain way to encourage the process and help shape the agenda.7 Limiting trade opportunities would not only set back modernization, but it would certainly force lower standards for workers and for the environment. The political objectives are primarily three: one to create prosperity on one's borders because prosperity fosters stability. Second, to spread active and deep democracy. Third, to transplant broader social objectives ranging from workers' rights to industrial safety and environmental standards are obvious issues on the political agenda of modern, open economies. A free-trade agreement supports modernization in the region and thus nurtures these objectives. It will also help raise wage levels, back to their 1980 level and beyond. By contrast, trade restriction here must mean even

192

RUDIGER W. DORNBUSCH

more poverty there; and with more poverty there will be fewer rights and the risk of political radicalism, neither of which is in the American interest. Today the United States faces an unusual opportunity to implement an outward looking trade policy with Latin America. Looking for modernization as the way out of a difficult economic situation, Latin America today is open to far-reaching trade reform. If we miss this opportunity, we are bound to fail in building an important western hemisphere trade and investment bloc in the 1990s. Turning our back on trade opening means inviting a slowdown if not failure of the reform movement and a resurgence of protectionism throughout Latin America and beyond. In concluding, nothing makes the case for a regional pervasive opening more strongly than the challenge of Cuba. The United States needs to prepare a strategy for handling Cuba. It will not take long before the communist regime falls. At that point, just as in the case of East Germany, from one day to the next there will be a need for concrete answers: if the United States does not offer a program of stabilization and open trade, mass migration is bound to be the result. A far-reaching trade opening, including sugar, will be necessary to avoid Cuba suffering the extent of economic decline experienced in Haiti or Nicaragua. Trade diversion An Americas FTA will involve the possibility and even the likelihood of some trade diversion. Trade partners in Europe, Africa, and Asia will not enjoy the same preferential access that insiders share. The resulting trade diversion is the down side of regional liberalization. But it hardly can be a reason for not going ahead. Moreover since the United States is already a very open economy, actual trade diversion will be minor. Increased prosperity and stability of the entire region and increased opening of Latin America creates prosperity and hence opportunities for outside countries that might not otherwise come about. Regional developments are not a threat to multilateralism; rather, they, are a means to break down barriers on a limited regional scale which create precedents, blueprints, and competitive forces that help drive a broader liberalization effort. Trade diversion can be a virtue. There is little doubt that the European opening to the East, ultimately including Russia, Belarus, and the Ukraine, and free trade in the Americas will create an important impetus for a new round of world trade liberalization. The pressure will be on Japan, immediately, to provide the market access that Europe and the United States are preparing to grant. Moreover, the liberalization of trade in hard core protection areas that is on the agenda in the regional context eliminates these sectors as a central obstacle to further

The case for bilateralism

193

trade liberalization. In this optimistic perspective ten years of successful regional liberalization may well be the prelude to a major round of world trade liberalization of the kind that today seems difficult to reach. The Japan problem Four decades of trade liberalization notwithstanding, the Japanese markets for manufactures, for services, and for agriculture remain virtually closed.8 Manufactures cannot make their way into Japan, except when produced by Japanese firms abroad. The difficulty of access has nothing to do with special economic conditions of the Japanese economy nor with the shoddiness of foreign goods. It reflects a culture determinedly opposed to trade as a two-way street. That attitude is increasingly challenged by Europe and the United States. It is becoming more and more likely that it will lead to a major trade conflict. The western perception In merchandise trade and in services it is exceptionally hard for European, US, or Asian firms to break into the Japanese market. Firms that have cultivated the market for a long time do sell and, because margins are extremely high, enjoy a profitable business. But successful firms are outnumbered 100:1 by those who have not been able to make it even if they try hard by international standards. The episodes are telling. Security firms that ultimately manage to cut through bureaucracy and get registered in Tokyo, find their bond issues boycotted. Manufactures find it impossible to cut through the complexities of trade access. Service firms find it impossible to compete in public tender. The game is rigged and it takes an age to make headway. By then Japanese competitors have in place the technology, innovation, or financial product that a foreign firm was trying to introduce. Of course, these are only episodes and there are plenty of messages to the contrary. Successful foreign firms operating in Japan sing an "all-is-well" chorus which is not really persuasive. We are told that US firms are making profits in Japan. What is surprising about this? US firms make profits everywhere; what is suspect is the need to even assert it. In fact, the very profitability of these firms has probably more to do with the closed Japanese markets where margins are phenomenal than with the achievements of the chosen few who have gained access. The United States has responded to the perception of a closed Japanese market by challenging Japanese business practices in the context of the Strategic Impediments Initiative (SII). This involves negotiating an

194

RUDIGER W. DORNBUSCH

Table 8.6. Import penetration in manufacturing Consumption Share

Canada Germany UK US Japan

GNP Share

1975

1985

1980

1989

19.5 22.9 14.2 5.5 4.7

38.7 31.7 33.2 12.9 5.3

14.0 12.0 13.2 4.5 2.4

18.8 15.6 17.9 7.1 3.1

Note: The data refer to manufacturing. Source: OECD "The OECD Compatible Trade and Production Data Base: 1970-85," Paris, mimeo, 1988. GATT and IMF.

opening piece-by-piece. The procedure is noisy and not very effective and one of its regrettable side effects is already apparent: a long queue offirmsin Washington lobbying to get on the waiting list. The facts Japan tells us that most problems lie abroad: US budget deficits and the poor quality of American goods. Or else, that Japan is not closed but must export manufactured goods to compensate for its lack of natural resources. Finally, that Japan may have been closed in the past but liberalization is underway by leaps and bounds. These arguments either miss the point or run counter to the facts. The mechanisms that close Japan to outsiders are not clear - neither tariffs nor quotas play a role - but by any definition Japan is closed. Manufacturing import penetration

By this measure Japan is far out

compared to other industrial countries. In Europe the ratio is more than 10 percent of GNP. Even the United States, a far larger country, has still twice the Japanese import penetration ratio (see table 8.6). Korea's import penetration in manufacturing is 18.7 percent - six times that of Japan. Of course, Korea is a developing country and hence the ratio might be high for that reason. But this would suggest that Japan might have had a high penetration ratio in the past. In Japan manufacturing imports have moved between 1.5 and 3 percent of GNP, without much change for a quarter of a century (seefigure8.1).

195

The case for bilateralism

1

1965

1968

1971

1974

1977

1980

1983

1986

1989

1992

Figure 8.1 Japan: import/GNP ratio (percent of GNP) GATT rounds of trade liberalization and major swings in exchange rates have done almost nothing to change Japan's openness. This evidence supports the view that Japanese protection is like an onion; it has multiple layers and the inner most are cultural, not the conventional restrictions in the form of quotas or tariffs. Considering next the entire range of non-oil imports, figure 8.2 represents a dramatic portrait of Japan's situation. While Germany's import penetration increased steadily, that of Japan actually declined over the past three decades. Figure 8.2 shows that Japanese import penetration in manufacturing is not much different today than it was thirty years ago trade liberalization unlike anywhere else had no effect. Intr'a-industry trade In open, developed economies consumers have the advantage of choosing from a broad range of product qualities and varieties produced throughout the world. Given the diversity of consumer tastes and the specialization of firms, any country would both import and export consumer goods or capital goods in many categories. Intra-industry or two-way trade is the common experience of advanced countries. The extent to which countries do pursue two-way trade is readily measured by an index which assumes a value of 1 when trade is completely two-way, i.e., in a particular commodity group imports equal exports. The index reported in table 8.7 assumes a value of zero when trade is a one-way street.

196

RUDIGER W. DORNBUSCH

Table 8.7. Intra-industry trade in selected manufactures (Index: one way = two-way = 1) Category

Germany

S. Korea

Japan

Finished Manufactures Machinery & Transp. Equip't. Elect. Mach. & Apparatus Automotive Products Textiles

0.72 0.66 0.89 0.71 0.90

0.72 0.94 0.68 0.91 0.33

0.33 0.25 0.27 0.17 0.36

Source: GATT International Trade.

1958 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 Figure 8.2 Non-oil imports as percent of GNP

All three countries are resource poor. Korea and Japan have their geography and transport costs to the West in common; Korea is poor while Japan and Germany are rich. Whichever way we look at these data, Japan is severely closed to intra-industry trade. Any story of resource endowments, geographic location, or the state of development simply fails. In Japan protection is at work, by invisible hand. There is no other explanation for a value of intra-industry trade of finished manufactures in Japan of 0.33 versus 0.72 for each, Korea and Germany. A good specific example is automotive products. Germany does have superior products and even so has intra-industry trade with a two-way index at 0.71. In Japan the corresponding number is entirely absurd at 0.17.

The case for bilateralism

197

Japan does not practice two-way trade. Manufactured goods are produced and exported, they are rarely imported. To some extent resource endowments influence the index: clearly, a country without natural resources will be a net exporter of manufactures to pay for oil imports. Yet, the puzzle is this: why does Japan look so different from Germany? Germany does not have natural resources any more than Japan. Do Japanese consumers, unlike consumers everywhere else in the world, not like imports? Or are they still taught to save foreign exchange and favor home industry as might have been plausible in the immediate post-war period? Or are there mechanisms we cannot see that plainly keep imports out? The skyrocketing of imports when liberalization does occur lends weight to this last hypothesis. Policy options Radical, rapid, and complete opening of the Japanese market is now a must. But it is not enough for Japan just to start catching up with other countries. Moving too late and too little means a path straight to trade conflict and beyond. Dissatisfaction in America with its own performance and rising populism in response to the middle-class squeeze will make America increasingly antagonistic. Therefore the state of limbo cannot last; Japan must act. Japan must make a real and determined effort to become a leading, active part of the world system. Japan must recognize that there are two ways to go. Either the country participates in a system of common goals and common responsibilities where the largest countries drive the initiatives and bear the burdens, or else Japan divides the world and builds its new Asian empire, based on confrontation and hostility to the West. It is tempting for America to set out an agenda that Japan must accomplish; full market opening over the next decade or else face retaliation. But it may already be too late for such an approach. US firms are fearful that in a confrontation they might not have access to Japanese components, US business is reluctant to jeopardize access to technology, and our military even more so. The United States has already become too dependent to make sanctions easy or even plausible. So the responsibility for a farsighted approach now rests with Japan. A resolution of the Japan problem is the best thing that could happen to GATT and the open international trading system. Those who question the openness of the world economy and the benefits of a liberal trade policy have all too easy a time today to point to Japan and argue that trade is not taking place on a level playing field. In time they will use the argument to do real damage to an open, non-discriminating system.

198

RUDIGER W. DORNBUSCH

Of course, Japan's opening would be no panacea for what in the United States is understood as the "Japan Problem" which in truth must be called mostly the US problem - poor economic performance. Notes 1 Some would argue that the Second World War, at least in Asia, had trade conflicts among its leading origins. In this perspective Cooper (1987) argues that the Smoot-Hawley tariff undercut the liberal faction in Japan, gave the military the upper hand, and led straight to Pearl Harbor. 2 This point has been known for a long time. See Johnson (1965). 3 Thomas R. Donahue, Secretary-Treasurer, AFL-CIO, before the Senate Finance Committee on February 6, 1991: "The enactment of a free trade agreement with Mexico, as proposed by President Bush, would be an economic and social disaster for US workers and their communities, and do little to help the vast majority of Mexican workers." 4 See Brauer (1991) for evidence of the effect of import competition on wages. 5 See McMillan (1986). 6 The great success of Europe invites the thought that the same ought to be tried elsewhere in regions where conflict is the rule. On this argument one ought to welcome trade initiatives in Africa involving South Africa, in the Middle East involving Israel and the Arab countries and some means of bringing together India and Pakistan. 7 On this view see Rubio (1992). 8 See, too, Balassa and Noland (1988) and Lawrence (1987). References Balassa, B. and M. Noland (1988), Japan in the World Economy, Washington DC: Institute for International Economics. Bauer, D. (1991), "The Effects of Imports on US Manufacturing Wages," Quarterly Review, Federal Reserve Bank of New York, Spring, 14-26. Cooper, R. (1987), "Trade Policy as Foreign Policy," in R. Stern (ed.) US Trade Policies in a Changing World Economy, Cambridge, Mass.: The MIT Press. Johnson, H.G. (1965), "Optimal Trade Intervention in the Presence of Domestic Distortions," in R.E. Caves, P.B. Kenen, and H.G. Johnson (eds.), Trade Growth and the Balance of Payments, North Holland, pp. 3—34. Kemp, M. (1976), "An Elementary Proposition Concerning the Formation of Customs Unions," Journal of International Economics, 2, 95-7. Kemp, M. and H.Y. Wan (1972), "The Gains from Free Trade," International Economic Review, October, 509-21. Lawrence, R. (1987), "Imports in Japan: Closed Markets or Minds," Brookings Papers on Economic Activity, 2, 517—54. McMillan, J. (1986), Game Theory in International Economics, New York: Harwood Publishers.

The case for bilateralism

199

Rubio, L. (1992), "Democracy and Economic Reform in Mexico," in Democratic Institutions, 1, 13-24. Viner, J. (1944), "The Views of Jacob Viner," in M. Shields (ed.), International Financial Stabilisation, New York: Irving Trust Company. Wonnacott, P. and R. Wonnacott (1981), "Is Unilateral Tariff Protection Preferable to a Customs Union? The Curious Case of the Missing Foreign Tariff," American Economic Review, September, 704—14. World Bank (1987), World Development Report, Washington DC.

CHAPTER

9

Restrictions to foreign investment: a new form of protectionism? VITO TANZI AND ISAIAS COELHO

I

Background

The traditional theory of international investment is based on the economic opportunities offered by potential markets which, for a variety of reasons, cannot be exploited through exports. It is also based on the abundance of location-specific resources, such as mineral deposits, which can be exploited more easily or more cheaply with the help of foreign technology and foreign capital. Moreover, the existence of specific skills in firms, that successfully combine advanced technology, superior managerial ability, and ample financial resources, explains why "direct" investment, often undertaken by multinational enterprises, cannot be easily replaced by "portfolio" investment or by the direct export of goods.1 If markets are allowed to perform their roles, actual or expected differences in relative prices across countries will point to profitable opportunities to trade, and differences in rates of return, assuming equality of risk, will signal where capital is scarcer, thus pointing toward higher payoffs from investing in those areas. Ceterisparibus, as capital movements take place, the return to capital in the various locations (and in the various investments) will tend toward equalization. Indeed, the theory predicts that all capital movements will cease once marginal returns are equalized at a given level of risk. If neither capital-exporting countries nor capitalimporting countries introduce obstacles or incentives to capital movements and, if the prices of output and input are the correct ones, capital movement will maximize potential benefit for the world as a whole and for each country.2 That foreign capital has historically been seen as a positive factor in the development of the host country, increasing production and employment and thus bringing prosperity, is manifest, for example, in the history of the United States. For two centuries that country steadily pursued policies that created a receptive climate for foreign investors. Its policy-makers were 200

Restrictions to foreign investment

201

aware that its great development potential could be better exploited by increasing the labor supply, through immigration, and the capital stock, through foreign investment and foreign borrowing. Similar behavior has been observed in many other countries where capital and know-how were scarce, and governments often offered incentives, such as tax holidays, and even guarantees of a minimum return to foreign investors. At the turn of this century, however, a literature developed, especially in England, on the "imperialistic" nature of foreign investment, a theme that was quickly taken over by Lenin and other Marxist writers.3 The view that foreign capital could be an instrument of political domination, or a tool that could be used by foreign interests to buy domestic politicians and important economic or even political concessions, reappeared with renewed strength in the period after the Second World War, especially in connection with the process of decolonization of Africa and Asia, a process which was followed, in many instances, by socialist and nationalistic regimes. In fact, decolonization was often accompanied by the nationalization of the existing enterprises, which had often been foreign owned. Nationalization was also accompanied by prohibition against foreign investors in order to protect the interests of the newly nationalized firms. The perception of foreign capital as a tool for exploitation found great intellectual support in Latin America, where a whole generation of students of economics came to believe in a "secular deterioration of terms of trade," which would drive commodities-exporting countries to poverty.4 This attitude resulted in the implementation of inward-looking policies and especially in a strategy of industrial development based on import substitution. The concern that foreign investors would have undue political power within the country, implied that import substitution ought to be carried out by local investors with substantial incentives and behind walls erected against imports. When foreigners were allowed to invest, they were often' limited to natural resources, or their voting power within an enterprise was kept below 50 percent. In the 1970s particular events forced a major change in policy orientation regarding foreign investment.5 The sharp rise in oil prices in 1973-4 caused great income losses for oil-importing countries and great gains for oil-exporting countries. These losses or gains were compounded by the second oil shock in 1979. The oil-importing countries experienced not only sizeable real income losses but also major disruptions in their production because of changes in relative prices. These changes made some energyintensive industries obsolete almost overnight, leading to sharp rises in unemployment. If the hardships caused by the higher cost of energy were, in the short run, more visible in some industrial countries, it was because they adjusted faster to the new reality. For example, Japan was able to reduce

202

VITO TANZI AND ISAIAS COELHO

dramatically and quickly its use of petroleum. However, during the adjustment process, it experienced major inflation and sizeable deficits in its current account. Many developing countries, however, tried to maintain the levels of consumption by additional foreign borrowing. In many cases, they also increased the subsidies to the domestic enterprises. At this point, the view that the state enterprises were important for the economic development of the countries was yet not challenged. These policies required foreign capital which often came as debt to the government and the state enterprises. This borrowing was facilitated by the aggressive lending behavior of international bankers who were attempting to recycle the large surpluses created by the "petrodollars." Many developing countries experienced large increases in foreign debt during this period. The low, or even negative, real interest rates that characterized international lending in the second half of the 1970s allowed the developing countries to postpone adjustment. The situation became precarious when interest rates in international capital markets rose sharply after 1979; it became unsustainable when the banks lost their desire to lend as a consequence of Mexico's default in the summer of 1982. Debt financing ceased to be an option, forcing many debtor countries to default on their foreign debt. The debt crisis had arrived with a vengeance. Once debt financing became unavailable, developing countries could only hope that (in addition to official lending) some capital would continue to flow in as private investment. However, this form of capital would follow a different dynamic and would respond to different incentives. It would require a fundamental change in attitude and in economic policy on the part of the developing countries. 2

Recent tendencies

In the 1980s, governments became more favorably disposed toward direct investment, in part because of the foreign exchange squeeze, in part because they had become progressively disillusioned with the past development strategy, and in part because they became fully aware of the rigidities that the servicing of foreign debt imposes on the balance of payments. Equity capital appeared a far more attractive option than in the past. The change in attitude came to be reflected in a gradual change of foreign investment laws and regulations. Within a few years these laws were reformed from being very restrictive to being inviting to foreign investors. Foreign investors started finding a red carpet in places where they had not been welcome. Other factors contributed to this change in attitude. One important

203

Restrictions to foreign investment Table 9.1. Major sources of external financing for developing countries (in billions of dollars)

Year

Foreign direct investment (A)

Net external borrowing from private creditors (B)

(A) as percentage of (AB)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989

2.79 3.11 2.25 4.36 -1.50 6.11 1.92 4.54 6.95 8.26 3.71 17.64 19.81 13.50 13.50 10.50 10.20 14.40 16.70 16.70

11.04 9.52 13.34 16.32 31.64 24.08 31.24 33.25 50.27 38.32 60.22 83.54 53.06 55.50 34.50 25.90 50.60 36.20 28.50 43.80

20.2 24.6 14.4 21.1 -5.0 20.2 5.8 12.0 12.1 17.7 5.8 17.4 27.2 19.6 28.1 28.8 16.8 28.5 36.9 27.6

Note: Column B Includes short-term official debt flows (primarily interest arrears). Source: International Monetary Fund. factor was the realization, by countries such as those which formed the Andean G r o u p , that foreign investors would not invest in their countries when alternative and safer outlets for their resources and expertise were available. During the 1970s the flow of direct investment into developing countries had been very limited (see table 9.1) and highly concentrated in a few countries. N o t surprisingly, countries which had been hostile to foreign investment were avoided by foreign investors. As mentioned earlier, in this period, most developing countries were getting foreign resources in the form of inexpensive debt. Economic views underwent a major revolution at the beginning of the 1980s. Of direct relevance to investment, the decade started with Mrs. Thatcher's privatization policy, which, in a process that is still reverberating

204

VITO TANZI AND ISAIAS COELHO

around the world, influenced one country after another. In the United States, which did not have state enterprises, President Reagan continued the process of deregulation started by the Carter administration, and launched a major tax reform aimed at increasing efficiency, which inspired other reforms around the world. Reagan also attempted to reduce in many other ways the role of government in the economy, promoting the view that the government was often the problem rather than the solution. As the decade approached its end, the disintegration of the Soviet Union and the liberation of Eastern Europe represented a mortal blow to central planning and state-controlled investment. The trends mentioned above influenced the developing countries. In these countries, the 1980s were a decade of progressive and accelerating economic reorientation toward more openness in economic relations with the rest of the world and toward a lesser role for the government in the economy. Many countries replaced existing restrictions on foreign investment with new and more receptive attitudes, which came to be reflected in revised investment codes, that simplified the investment process and provided guarantees to foreign investors. These changes, however, could not compensate for problems associated with poor economic performance and continuing mismanagement of the economy. For example, in spite of these changes and the fact that investment in many host countries came to be protected to some extent by the Multilateral Investment Guarantee Agency (MIGA), 6 foreign investment in Africa remained negligible. The economic decline of that continent and the persistent effects of nationalization during the 1965-75 period continue to be determining factors. Asia was, perhaps, slower in opening its doors to foreign investors than other regions. For example, Korea has only recently allowed foreigners to invest in retail trading. It was only in 1992 that the government of India started to show a more liberal attitude toward foreign investment. Until 1992, investment in India was held back by very restrictive laws and a heavy bureaucracy. Other countries restricted investment in various ways. Despite this, during the 1980s Asia's share of foreign investment was by far the largest among all continents (see table 9.2). This reflected, first, the fact that the region went almost unscathed by the debt crisis and, second, the remarkable economic performance of the major recipients of foreign investment, namely, Singapore, Hong Kong, Taiwan, China, Malaysia, and Thailand. Latin America has recently improved its attractiveness to foreign investment through a combination of deregulation, privatization of state enterprises, access to foreign investors of sales of government equity in state enterprises, improved foreign exchange regimes, and new financing mechanisms, especially the debt-for-equity swaps. The improvement in the

Restrictions to foreign investment

205

Table 9.2. Directions of foreign direct investment (in billions of US dollars) 1984

1985

1986

1987

1988

1989

1990

49.1 48.3 0.8 0.2 0.3 — 0.1 0.1

57.1 55.8 1.3 0.1 1.0 — 0.1 0.1

92.9 91.0 1.9 0.1 0.8 — 0.4 0.7

140.3 137.7 2.6 0.1 1.9 — 0.3 0.2

170.7 164.6 6.1 0.1 5.3 — 0.4 0.3

215.7 205.9 9.8 0.1 8.4 — 0.6 0.7

229.2 220.7 8.5 0.1 7.5 — 0.4 0.5

0.1

0.1

0.7

0.3

0.4

0.7

0.6

0.7

1.2

1.2

2.3

5.7

9.1

7.9

53.8 38.5 15.3 1.1 4.6 0.2 6.1 3.2

48.2 36.2 12.0 0.7 4.9 0.2 2.2 4.0

75.7 63.8 11.9 0.6 5.7 0.2 2.3 3.1

121.8 149.8 107.6 127.9 14.2 21.9 1.2 1.4 12.7 8.5 0.2 0.5 (0.1) 1.4 4.2 6.1

192.4 164.3 28.1 2.7 15.8 1.1 1.9 6.7

179.5 150.9 28.6 1.2 18.6 1.2 0.4 7.3

5.6

1.8

1.5

(0.1)

1.1

2.9

1.5

9.7

10.3

10.4

14.3

20.8

25.2

27.2

28.4

24.9

15.7

11.7

14.6

14.6

15.9

From:

Total Industrial countries Developing countries Africa Asia Europe Middle East Western Hemisphere Including: Oil exporting countries Non-oil developing countries To: Total Industrial countries Developing countries Africa Asia Europe Middle East Western Hemisphere Including: Oil exporting countries Non-oil developing countries Developing countries in % of total

Note: Differences between totals are due to statistical discrepancy. Source: International Monetary Fund.

macroeconomic environment in several Latin American countries since 1991 has undoubtedly also helped. Up to 1990, however, the results were still modest (see table 9.2), owing to the pervasive economic instability and the foreign-debt overhang that, up to that time, affected much of the region. The debt crisis had kept away many potential investors. The figures shown in table 9.2 for Latin America are much influenced by debt-for-equity swaps (see table 9.3).7 They do not represent genuine movements of financial resources. When the data for the period after 1990 become available, they

206

VITO TANZI AND ISAIAS COELHO

Table 9.3. Debt-equity swaps in selected countries" 1985 Argentina Brazil Chile Costa Rica Ecuador Honduras Jamaica Mexico Nigeria Philippines Uruguay Venezuela Total Argentina Brazil Chile Costa Rica Ecuador Honduras Jamaica Mexico Nigeria Philippines Uruguay Venezuela

469 537 323 1,329

1986 (in millions 176 974 7 413 81 1,651

1987

1988

1989

of US cdollars) 764 336 2,095 1,997 2,927 44 89 127 261 14 9 4 5 1,680 1,056 70 451 931 104 49 45 8,320 4,738

1,180 942 2,767 124 31 47 16 532 304 630 53 544 7,170

1990 6,670 292* 1,100 17 58 32 9C 652 157 630 716 10,333

(in percent of net inflow of foreign investment) 51.0 0.0 0.0 66.6 114.8 42.4 99.4 30.9 75.0 521.0 1,708.8 2,058.8 2,685.3 1,068.3 12.2 117.4 36.2 107.9 326.3 38.8 169.3 23.2 29.9 125.7 (41.7) 56.9 7.5 27.1 40.7 23.7 51.8 18.6 14.6 63.8 146.9 99.5 130.7 233.7 (281.3) 233.3 706.5

"Face value of debt converted under official ongoing schemes. b J anuary-J une. 'J anuary-September. Source: International Monetary Fund.

are likely to reflect an increasing role of foreign direct investment largely promoted by privatization policies and improved economic conditions in Argentina, Chile, Mexico, and a few other countries. Some of the direct investment may represent the return of capital that fled during the debt crisis. The growing popularity of foreign investment during the 1980s was not limited to the developing world. Indeed, most European countries have tried to attract foreign investment, advertising their political stability and

Restrictions to foreign investment

207

their relative steadiness of policies. Some countries, as for example Ireland, have indeed gone a long way, through tax incentives and other measures, to make themselves attractive to foreign capital. Helped by the economic difficulties in the developing countries, these policies resulted in a significant shift of foreign investment toward the industrial countries, as indicated in the bottom line of table 9.2.8 This receptive approach, and the anticipation that trade barriers between the European Community and the rest of the world would make exports to the Community more difficult, resulted in a surge of foreign direct investment in Europe in the early 1990s. Another factor expected to have an impact on the location of international direct investment is the joining of the Eastern European countries, including the new Soviet republics, in the map of the market economies. The policy-makers of these countries are aware that the modernization of their economies can only be accomplished by the massive imports of Western technology and investment. As a consequence, foreign direct investment is now seen as not just a source of financing but also as a vehicle for modernization. In future years the limited and falling supply of world saving will have to finance foreign investment in Europe, the developing countries, and the economies in transition of Eastern Europe. Economic and political conditions will determine where this investment will finally go. 3

Prospects for foreign investment in developing countries

The new attitude on the part of developing countries toward foreign investment is very much influenced by the realization that it could make an appreciable contribution to development by providing badly needed financing, by introducing new technologies and management skills, and by opening foreign markets for their exports. Not long ago, the profits made by the branches or the subsidiaries of foreign concerns were seen as the result of exploitation or monopoly power. However, the anxiety that government officers felt toward foreign investment started fading as they became more knowledgable about markets and economic conditions and more skeptical that the earlier strategy based on import substitution would work. They became more confident in their ability to draft legislation and negotiate contracts and foreign participation in specific projects in terms advantageous for both their countries and the foreign investors. This process has undoubtedly been facilitated by the fact that many of the current policy-makers have been trained abroad and are thus familiar with foreign institutions. In spite of these favorable developments, over the next few years prospects may still not be very bright for large inflows of foreign direct investment to specific developing countries. The reasons are several.

208

VITO TANZI AND ISAIAS COELHO

First, the change in attitude toward foreign investment - from one of prohibition or strict regulation to one of encouragement - has taken place at about the same time all over the world. As a result, the change in policies in one country or one region does not significantly affect its relative attractiveness to foreign investors, as would have been the case a few years earlier. Second, apart from investment in natural resources, foreign investment has often taken place in order to sell in local markets behind the protection of tariff walls and other import restrictions.9 With the replacement of import-substitution policies by policies aimed at integrating a country's economy with the rest of the world, this reason for investing in a specific country disappeared. Now direct investment must be induced by other motives. The removal or lowering of tariff walls may induce multinational corporations to increase exports to developing countries rather than to attempt to produce in those countries in order to sell there. Third, the growing economic integration within the European Community, and the fear that the Community might raise its barriers against imports from the rest of the world is likely to divert some investment from developing countries toward the huge European market where all major multinational enterprises would want to be established. Once inside the European market, an enterprise would be able to sell to all the member countries. More generally, "cross-hauling" investment (reciprocal investment flows) among developed countries has replaced to a large extent investment flows toward developing countries. Fourth, debt-for-equity swaps, which have been an important component of the foreign flows toward developing countries as recorded in the statistics, are not genuinely foreign investment in the common sense of a new and actual transfer of resources. Another component of the investment statistics consists of reinvested profits, which in some cases, especially in the 1980s, were not repatriated because of shortages of foreign exchange in the host countries. Therefore, in recent years, the statistics on foreign investment to developing countries contained an element of overestimation which will disappear in the future. Fifth, privatization certainly offers new opportunities for foreign investment. However, when many countries - including Eastern Europe and the former Soviet republics - are all trying to privatize at the same time, it is likely that either the privatizing assets will be acquired at low prices or the amount of investment attracted by such schemes in a specific country will be relatively small. Sixth, the fall in the savings rate over the past decade in most industrial countries and the persistence of sizeable fiscal deficits have inevitably reduced the amount of saving available to finance investment in foreign

Restrictions to foreign investment

209

countries. The fall in the savings rate is expected to continue in part because of demographic factors (the aging of the population in industrial countries). Finally, the reduction in the corporate income tax rates in several industrial countries (especially the United States and the United Kingdom) has reduced the relative attractiveness, for tax purposes, of investing in developing countries. 4

Investment into common economic areas

The anxiety vis-a-vis foreign investment has declined in developing economies only to reappear in the industrial countries. In an earlier paper,10 we discussed the appearance of this phenomenon in the United States and the United Kingdom, during the oil crisis, when Arab capital was acquiring fixed assets and equity in those countries. More recently it has been stimulated by the increasing inflow of investment capital from Japan generated by the large current-account surpluses experienced by this country.11 The acquisition of American landmarks by the Japanese has led to an increasingly strong reaction on the part of a sector of the American public. Complaints that Japan is buying America and fears that soon many Americans will be working for the Japanese have been heard with growing insistence and stridency.12 Over the years European governments developed complicated procedures for screening foreign investment. These procedures are being used against Japanese investment, which is also generating anxiety in Europe. These procedures have succeeded (a) in preventing the entrance of capital aimed at taking over major national enterprises, be these privately or publicly owned, (b) in welcoming green-field investment which does not significantly compete with local firms in the domestic markets, and (c) in enthusiastically attracting non-polluting investment especially if aimed at generating exports. The current anxiety regarding foreign investment observed in industrial countries is in part the result of considerations that come into play when a country is a member of a common market, or even of a free-trade area. Over the years, costs of transportation have become much less significant to trade flows than barriers imposed by trade policies since traded products have generally become less bulky and more highly priced. Therefore, locating production facilities inside a large free-trade area has become a major factor for capital-exporting countries in determining where to invest. For the capital-exporting country, ceteris paribusy almost any country within the free-trade area is as good as another. However, for the countries belonging to the free-trade area or the common market, the location of the foreign

210

VITO TANZI AND ISAIAS COELHO

investor inside the free-trade area is not a matter of indifference since there are particular benefits that accrue only to the specific country where the investment is located. The generation of employment, the development of skills and technology through training and R&D, and the payment of taxes are examples of important benefits that will accompany new production facilities. As a consequence there will be competition for foreign investment among the countries, the regions, or the states within the common economic area. In the United States, for example, states compete for foreign investment by offering a number of incentives. Within the European Community the competition is among the member countries. Countries join a common economic area primarily to take advantage of economies of scale. However, they do not want to lose technologically advanced industries because of foreign competition. Consider, for example, the case of automobiles. For a number of reasons, several European countries want to preserve their automotive industry. Therefore, they raise visible and invisible barriers to the importation of cars, which result in higher consumer prices and allow even less efficient car manufacturers to continue operating. Because of these barriers, car makers from outside the Economic Community have a strong incentive to establish themselves as producers within the Community. They will be welcome to a country that already had an automotive industry provided that the country believes that they will produce mainly for export. In other words, the host country wants the employment and the taxes from the new facility but not the competition that the new car production will provide to its own car industry: it wants the benefits but not the perceived costs of the foreign investment. The (other) importing countries in the Community, however, see this process as a back-door penetration of their domestic markets. They see only the costs and not the benefits of the foreign investment. The question here is not one of moral hazard, whereby, for example, the United Kingdom, by attracting Japanese investment for production aimed predominantly at clients living across the Channel, is exploiting its associates. Rather, it is a problem typically associated with goods that generate large externalities. The question is how the benefits generated by the investment can be shared by each member of the Economic Community. The question does not arise when the choice of locations for foreign investment is between two different points in the same country. In this case, the host country policies would, ceterisparibus, be indifferent as to location.13 The difficulty resides in understanding what is the national interest. To be indifferent as to the location of investment within the Community would imply that the well-being of other Europeans is given the same weight as the well-being of nationals. For the near future this is clearly unrealistic so that each country of the Community will continue assigning a greater weight to the benefits derived by its nationals.

Restrictions to foreign investment

211

A solution to this dilemma will not come from the spiritual feelings associated with the concept of the Nation, but from the more rational, operational, and more economically significant concept of the state. At least in principle, the state should be capable of accommodating conflicting national interests on the basis of mutual economic interest and of creating redistributive mechanisms that would make every region or member state better off than if it were pursuing its interests in isolation. Such redistributive mechanisms are now not available to any significant extent within the Community and it is unlikely that they will be available for some time. As a consequence, the conflicts outlined above will continue to play a major role in determining the individual governments' attitude vis-a-vis foreign investment. The establishment of free-trade areas and, more generally, the raising or removal of barriers to trade in large markets have significant implications for investment decisions and for foreign direct investment, because the larger the market and the greater the actual or potential barrier against the outside world, the greater will be the incentive for multinational enterprises to establish facilities within it. Two examples illustrate the point. The start of negotiations for the creation of a North American Free Trade Area (NAFTA), which would unite the economies of Canada, the United States, and Mexico in a common trade area, has prompted interest groups to call for protection against "abusive" trade practices. These trade practices may be closely linked to investment decisions. According to these groups, Asian exporters, eager to penetrate the American market, would use Mexico as an export platform. Mexico might have an incentive to invite the new production facilities onto its territory since much of the production would be exported to the United States while the employment and tax benefits would stay in Mexico. To deal with this possibility, some argue that the United States should demand that only imported goods containing a high proportion - say at least 75 percent - of value added in Mexico would qualify as intra-NAFTA exports. The danger and the irony is that the argument could easily be pursued further, and the US authorities could be asked to suggest restrictions to foreign investment in Mexico, a move that would contradict the long-standing US policy of open door to foreign investment. A second example is provided by the South American Common Market (MERCOSUR), which is being created by Argentina, Brazil, Paraguay, and Uruguay.14 When MERCOSUR was announced, multinational enterprises operating in the region started revising their plans for the location of future production facilities, since, once the Common Market started operating, the larger Brazilian market would be equally accessible from Sao Paulo (Brazil) or Montevideo (Uruguay).15 In such a case the foreign enterprise might, for example, choose to establish its production in the country with the lowest

212

VITO TANZI AND ISAIAS COELHO

taxes, and countries might be pressured to compete in giving tax and other incentives to foreign investors. The important conclusion is that trade policies and policies regarding foreign investment cannot be separated. The distinction between trade and investment becomes even more blurred by such arrangements as joint ventures and franchising. It is becoming progressively more difficult for countries belonging to free-trade agreements to design consistent and transparent policies for foreign investment. The solution involves coordination of trade and investment policies in each country and across the countries of the common economic space. However, as the Andean Group experience has shown, agreement on a uniform treatment for investment from outside the area is difficult to reach, and almost impossible to adhere to. In reality, a common external treatment is not enough to ensure coherence in the group, because it does not provide (a) an acceptable distribution of the benefits from foreign investments, which are hard to predict and assess, and (b) does not remove the desire to preserve the control by nationals over technologically advanced sectors of the economy. 5

Investment protection

Let us define as investment protection all the distinct policies utilized by a country to prevent or limit unwanted direct foreign investment. Investment protection is much more difficult to identify, measure, and overcome than trade protection. The latter is an intellectual pariah,16 which can be understood, if not justified, only in terms of gains for one sector of the economy (an industry, a region) at the expense of another.17 Investment protectionism, however, has a much stronger intellectual claim on legitimacy. Not on purely economic grounds, where its chances might not be better than those of trade protectionism, but on grounds of power and control, if better words cannot be found to describe the fear of losing the privileged position, in the inner circle of the affluent nations, that comes with knowledge and technology and with technologically advanced industries. The history of investment protectionism has still to be written. So far much of the attention has been directed toward trade protectionism. In periods when trade protectionism was stronger, for example, during the mercantilist period, foreigners who could produce trade-competing goods were especially welcome.18 In the second half of this century, capital started moving in every direction, thereby making it difficult to distinguish between capital-importing and capital-exporting countries. Countries with a long history of economic nationalism, where industrialization had to some extent resulted from industrial policies of the government (and where

Restrictions to foreign investment

213

frequent wars had built strong links between industry and defense), were simply not prepared to let important sectors of the economy fall under foreign control. In no other sector is this appeal for self-reliance sharper than in the defense industry. However, other sectors can easily be added to the list of the industries considered essential for the defense of the country, such as coastal and maritime transportation. But then it becomes easy to use the argument of "economic security" to limit many sectors of the economy to investment by nationals. The banking sector is also highly regulated in most countries, and more so when the banks are foreign owned. Several countries forbid foreign investment in the banking sector. Abstracting from the small countries that constitute tax havens/banking centers, industrial countries provide a relatively unfavorable climate for foreign-owned banks. In the United States, the states are accorded regulatory powers on the banking industry and often use them to discriminate against foreign banks.19 As a consequence of the Bank of Credit and Commerce International's scandal, in 1991 the US Treasury sought to restrict the operations of foreign banks in the United States. This move met with only partial success owing to the opposition by the Federal Reserve Board, which was concerned about a possible reduction in lending in the United States by foreign banks. In another instance, when privatizing its financial system in the late 1980s, Portugal placed limits on the shares that could be acquired by foreign investors to ensure that Portuguese nationals kept control of the more important institutions. The air passenger transportation industry is another sector where investment protection is notable. Virtually every country has its "national airline," owned and subsidized by the government. In the United States the airlines are private, but foreign investment in an American air carrier cannot exceed 25 percent of its capital. In addition, the US government uses - as every other country - the access to its aviation market as a tool to obtain entry overseas for US carriers. Financial difficulties, however, are softening investment protection in the airline industry. In 1990, when every US airline recorded excess of operating expenses over revenues, the US Department of Transportation announced its intention to allow an increase in foreign equity to 49 percent. Argentina's airline has been privatized, and its control has been acquired by foreign investors. Progress in this area has been slow, however. Aeroindustry is another sector where investment protection runs high, certainly because of its connection with the defense industry. A consortium of four European countries supports the Airbus project, which is the main competitor to large airplanes made in the United States. The United States has challenged the European Community on the grounds that the Airbus is

214

VITO TANZI AND ISAIAS COELHO

subsidized in violation of the GATT Subsidies Code. EC officials, in their turn, have alleged that US aviation firms benefit from military contracts, which act as indirect subsidies.20 On April 1, 1992, a breakthrough agreement on limiting subsidies on civil aircraft was reached after six years of negotiations, although details must be worked out. 21 In many countries some services are also protected against foreign participation. Japan has been recently challenged by the United States for its restrictions on the exercise of law practice by US lawyers, and also for not giving opportunity to US firms to bid for construction projects. In the United States, the official position is that the government should restrict only foreign investment which can pose a threat to national security.22 National security is, however, an elusive concept, which can be easily extended to cover a number of technologically advanced sectors where foreign investment may take place. It, thus, provides a pretext for lobbying by representatives of various industries against foreign penetration of their domestic markets. Rules concerning foreign investment are still more obscure in other industrial countries. In reality, restrictions to foreign investment are fundamentally intended to insure that the country remains "competitive" through a technological edge held by nationals. 23 For this reason governments subsidize high-technology sectors directly or indirectly (through equity participation, procurement practices, provision of soft credit lines, and so forth). To liberalize investment activities by foreigners in some sectors of a host country is a complicated matter because it is difficult to disentangle the activities of certain enterprises from the government. It is even more difficult to distinguish clearly the purely commercial motivation of the foreign investor in whose country the investment originates. This explains why airlines, banking, insurance, and telecommunication industries are so often highly regulated. In addition, much high-technology research is closely linked with, or is a spinoff of, defense projects. It can be predicted that multilateral and bilateral negotiations regarding investment liberalization will be difficult and complex. Clear and non-discriminatory rules for foreign investment go a long way toward creating a climate advantageous for foreign direct investment. However, such rules are at times, insufficient to remove the worry of capital-importing countries that they may not be receiving a fair share from the investment. This worry seems to be particularly pronounced with respect to the taxation of the profits of multinationals. It is well known that these enterprises have the ability, and often use that ability, to allocate profits among the countries in which they operate by manipulating transfer prices. The objective is to minimize the total, worldwide tax liability for the

Restrictions to foreign investment

215

multinational. Transfer prices may be manipulated (a) with respect to the manufactured goods that the enterprises buy from related enterprises (branches or subsidiaries) in other countries; (b) with respect to the payments that they make abroad for the use of intangibles, such as patents, copyrights, and trademarks; and finally, (c) with respect to payments made for technical, professional, financial, and other services. At present there are about 45,000 foreign-controlled corporations in the United States and the number is growing fast. A recent study by the Internal Revenue Service24 has cited evidence that, over the 1984-9 period, US companies 50 percent or more owned by foreign shareholders, consistently reported higher rates of growth in assets and in receipts than US companies. At the same time, they reported rates of return and net incomes consistently and significantly below those of non-foreign owned US companies. This study confirmed the widespread view, strongly held by some Congressional members, that the foreign owned companies are not paying their fair share. The Internal Revenue Service has been increasing the resources it allocates to ensuring that foreign inbound investment contributes its fair share. At the same time, an army of lawyers, accountants, and tax economists is assisting the foreign-owned enterprises in arguing their cases before the Internal Revenue Service. These costs must clearly be subtracted from the benefits that countries receive from foreign investments. Congress has been taking a progressively more aggressive stance. Legislation introduced on May 27, 1992 by Congressmen Rostenkowski and Gradison (H.R. 5270, the "Foreign Income Tax Rationalization and Simplification Act of 1992") reflects this concern.25 This kind of legislation conflicts with established agreements between countries, especially as reflected in tax treaties, and brings about frictions with other countries.26 These frictions once again represent a cost connected with foreign investment that economists must take into account. 6

Concluding remarks

Economists have paid a lot of attention over the years to trade restrictions. Only now they are waking up to a new phenomenon with important implications for policy and for the allocation of resources; this is the phenomenon of restrictions to foreign investment as a form of protectionism. This chapter has discussed some aspects of this phenomenon. However, its full implications still remain to be analyzed. The analysis of investment restrictions is likely to prove very difficult. The chapter has provided a taste of these difficulties, especially within the context of a free trade area.

216

VITO TANZI AND ISAIAS COELHO

Notes 1 The standard reference is still the excellent survey by Caves (1982). 2 Foreign investment will reduce the rate of return to capital in the capitalimporting country, thus hurting holders of the guest country pre-existent capital. This loss, however, can be shown to be smaller than the gain accruing to other factors of production. Therefore, an income redistribution that would make everyone better off is theoretically possible. The same applies, mutatis mutandis, to the capital-exporting country. On this, see Jones, Coelho, and* Easton (1986). 3 The seminal work is Hobson (1902). 4 To this attitude that implies that the interests of countries are contrasting belong the concepts of dependent development, center-periphery, neocolonialism, North-South, and Third World. 5 Policies of industriali2ed countries regarding investment outflows toward developing countries are covered in Bachmann (1991). 6 MIGA, an affiliated agency of the World Bank, is also a creation of the 1980s. 7 Data in tables 9.2 and 9.3 are not fully comparable because they originate from different statistical sources. The reader should be aware of the difficulties involved in the measurement of international capital movements. 8 A detailed analysis of investment capital flows in the period is available in International Monetary Fund (1991a, 1991b, 1991c, 1991d). 9 This is less so in Southeast Asian countries where foreign investment was predominantly oriented toward the development of export platforms. 10 See Tanzi and Coelho (1991). 11 In a recent interesting paper McCulloch (1992) has discussed the changing and still ambiguous attitude of US negotiators regarding international investment flows. 12 A recent best-selling fictional work, Rising Sun, has capitalized on this fear. 13 We are abstracting, of course, from development policies targeted to the less developed parts of the country, like the programs for industrial development in Italy which discriminate in favor of the Mezzogiorno. We are also abstracting from environmental considerations which may make one location preferable to another. 14 Bolivia may join MERCOSUR at a later date. 15 Other factors, of course, affect the decision to invest, among them the tax structure. On this, see International Monetary Fund (1990). 16 The main defense of (temporary) protection rests on dynamic comparative advantage in the infant-industry argument. Practice, however, shows that temporary incentives become permanent privileges. A case in point is the auto industry in Brazil, which was built thirty years ago behind a prohibitive tariff wall. As of today, the industry operates with a productivity level much lower than that of Europe and the United States. Any attempt to introduce some competition in the market — which undoubtedly would benefit consumers - is resisted by the industry citing the threats of unemployment. 17 Smith (1965) describes how, in the late 1980s, an ill-advised protectionist policy

Restrictions to foreign investment

18

19

20

21 22

23

24 25 26

217

brought disaster to the South of Italy, but profited the more industrialized North. In Calonne's France, ". . . economic relationship with Britain was governed by what was called state opportunism. Engineers had been brought to Northern France to install spinning jennies and Crompton's mechanical mule, and at the end of 1786 hopes were raised of spiriting away the famous Matthew Boulton and James Watt from the British Midlands" (see, Schama, 1989). During the Great Depression, legislation was passed to restore confidence in the banking system, which resulted in segregation among financial institutions. For example, banks cannot deal in the securities market. Also, according to Treasury Secretary Nicholas Brady, "We [the United States] are the only modern country that does not permit national banking" {Washington Post, December 1, 1990). That is, American banks are not allowed to branch over state lines. "Where would we be today in our balance of payments if we did not have the technology of our jet commercial aircraft industry, which was underwritten partly by the Department of Defense?" (Senator Frank Murkowski's intervention at a hearing before the Subcommittee on Foreign Commerce and Tourism, on the US Senate, on July 19, 1990). Financial Times, April 2, 1992. An analysis of US policies regarding foreign investment, and the economic impact of foreign-owned enterprises in the United States, is provided by Graham and Krugman (1991). "In the Intelligence Committee we are searching for the appropriate role to gather intelligence to ensure the national security interest of our international competitiveness. We are not set up to do it. We do not know how to do it. We just feel if we do not do it, and the French are doing it and the Japanese are doing it, how are we going to survive in this world? It is a competitive world out there" (Senator Frank Murkowski, see note 20.). Department of Treasury, Internal Revenue Service: Report on the Application and Administration of Section 482 (Washington, DC: April 1992). See Department of Treasury, Treasury Notes, July 21, 1992. A letter from the Chairman of the Committee on Fiscal Affairs of the OECD to Mr. Brady, the US Treasury Secretary, in June 1992 outlined the concern of other countries with the proposed legislation. References

Bachmann, H. (1991), "Industrialized Countries' Policies Affecting Foreign Direct Investment in Developing Countries", PAS Research Paper Series, Vol. 1: Main Report. Washington, DC: MIGA/PAS. Caves, R.E. (1982), Multinational Enterprise and Economic Analysis, Cambridge: Cambridge University Press. Department of Treasury (1992a), Internal Revenue Service, Report on the Application and Administration of Section 482, Washington, DC, April. (1992b), Treasury News, July 21. Graham, E. and P. Krugman (1991), Foreign Direct Investment in the United States.

218

VITO TAN2I AND ISAIAS COELHO

Washington, DC: Institute for International Economics. Hobson, J.A. (1965), Imperialism, a Study, Ann Arbor: Michigan University Press. International Monetary Fund (1991a), Balance of Payments Statistics, Vol. 42, Parts 1 and 2, Washington, DC: International Monetary Fund. (1991b), "Determinants and Systematic Consequences of International Capital Flows," Occasional Paper No. 77, Washington, DC: International Monetary Fund, March. (1991c), Private Market Financing for Developing Countries, Washington, DC: International Monetary Fund, December. (1991d), World Economic Outlook, Washington, DC: International Monetary Fund, October. International Monetary Fund, Fiscal Affairs Department, Tax Policy Division (1990), "Tax Policy and Reform for Foreign Investment in Developing Countries," in Taxation and International Capital Flows, Paris: Organization for Economic Cooperation and Development, pp. 163-235. Jones, R., I. Coelho, and S. Easton (1986), "The Theory of International Factor Flows," Journal of International Economics, 20, May, 313-27. McCulloch, R. (1992), "The Impact of Foreign Ownership on U.S. Business," paper presented at the International Monetary Fund/Research Department seminar held in Washington, DC, March 31. Schama, S. (1989), Citizens: A Chronicle of the French devolution, New York: Knopf. Smith, D. (1965), Storia d'Italia dal 1861 al 1958, Bari: Laterza. Tanzi, V. and I. Coelho (1991), "Barriers to Foreign Investment in the United States and Other Nations," Annals of the American Academy of Political and Social Science, 516, July, 154-68.

PART

in

Exchange rates and protectionism

CHAPTER

io

Floating exchange rates and the new interbloc protectionism: tariffs versus quotas RONALD I. MCKINNON AND K. C. FUNG

1.

Introduction I regard as a key advantage of free exchange rates the likelihood that they will lead to freer world trade, will promote a dismantling of exchange controls and import quotas and a reduction of tariffs. (Milton Friedman, 1967, p. 71)

The removal of the balance-of-payments (objective) is an important positive contribution that the adoption of flexible exchange rates could make to the achievement of the liberal objective of an integrated international economy. (Harry Johnson, 1971, p. 210)

Earlier arguments in defense of floating exchange rates often contained high hopes that by allowing the exchange rate to adjust naturally in response to the ebb and flow of international payments, there would be no need for the government to intervene to correct balance-of-payment disequilibria. International payments adjustment would become virtually automatic. This was taken to imply, among other things, that there would be less likelihood that protectionist policies such as tariffs and quotas would be used. In other words, floating exchange rates should lead to free trade, or at a minimum, freer trade within the world economy. Behind the earlier textbook arguments for the linking of free trade to floating exchange rates were two closely related assumptions. First, it was 221

222

R. I. MCKINNON AND K. C. FUNG

presumed that exchange rates would adjust toward market equilibrium in a smooth and orderly fashion. Milton Friedman argued that destabilizing speculators would on average incur losses while stabilizing speculators made profits - thus ensuring that stabilizing speculation would predominate. The second presumption was that the "equilibrium" exchange rate for which stabilizing speculators would aim is governed by conditions in the current account of the balance of payments. Taking the spot markets for international goods and services as the reference point, the traditional advocates of floating had it that the exchange rate would adjust smoothly toward an equilibrium that balanced "international competitiveness" in importing and exporting. Economists were somewhat divided as to whether this equilibrium exchange rate would (1) roughly balanceflowsof imports and exports as per the elasticities approach to the balance of trade or (2) roughly align national price levels for tradable goods as per the principle of purchasing power parity. Both groups, however, looked to current commodity flows, rather than to the capital account of the balance of payments, to determine equilibrium. Floating exchange rates, it was imagined, would limit trade deficits with but modest departures from purchasing power parity. Real exchange rates would remain fairly constant. Therefore, continuously smooth adjustment in nominal exchange-rates to balance international competitiveness was seen - from a macroeconomic perspective - as a way of eliminating the balance-of-trade motive for protection. A government need no longer worry about untoward deficits or surpluses in its balance of trade. At the microeconomic level, floating exchange rates were seen to insulate domestic producers from abrupt changes in their ability to compete on international markets - and thus make them less likely to petition the government to relieve their distress. The demand for protection would diminish. With our experience of the floating exchange-rate system, we now know that many, if not all, of the elements of the textbook pre-1973 arguments for floating exchange rates are empirically and/or logically incorrect. Floating rates do not move smoothly, steadily, or orderly. They move more sharply and more frequently than most economists had expected - as likely away from purchasing power parity as toward it in the short run. But domestic goods prices invoiced in home currencies evolve more slowly, i.e., they remain "sticky." The correspondingly large changes in real exchange rates have led, in the 1970s and 1980s, to abrupt changes in industrial competitiveness and increased demands for protection. (But, as we shall see, protection more in the form of quantitative restrictions on trade flows rather than tariffs.)

Floating exchange rates

223

Where did the traditional arguments for floating go off the rails? By concentrating on commodity flows and the trade balance, floaters misjudged what would determine a sustainable equilibrium in the foreign exchanges in the short and intermediate runs. Instead, it is now wellestablished (Frenkel and Mussa, 1980) that, under floating, the spot exchange rate behaves like a forward-looking asset price which is dominated by financial institutions holding portfolios of interest-bearing securities and money denominated in foreign and domestic currencies. But such portfolio holders are more concerned with the future evolution of the exchange rate than they are with any level that balances current commodity flows. Such an asset price is highly sensitive to "news" such as, how much the future money supply in the US might be expanded relative to that in Germany or Japan, or how the US future capital gains tax may change or what the prospects for changes in a country's terms of trade may be. Without a common monetary standard and a par value system for exchange rates, how national monetary policies will evolve vis-a-vis each other is essentially random. Thus, exchange rates fluctuate as "virtual" random walks in the short run. They cannot be predicted satisfactorily using past information of money growth, GNP growth, trade deficits and so on (Meese and RogofT, 1983). The sharp volatility of the exchange rates is vividly illustrated by episodes of dollar gyrations in the 1980s. After being undervalued in early 1980, by early 1985 the dollar had overshot any equilibrium defined with reference to the current state of the commodity markets, say, by the principle of purchasing power parity. Similarly, by the end of 1987, it had fallen far below purchasing power parity (McKinnon and Ohno, 1989). Since then, the dollar has continued to gyrate - rising or falling 15-25 percent in the course of a year against other major currencies. The arguments set forth below will show that quantitative restrictions in the form of quotas, voluntary export restraints, and market-sharing schemes are increasingly relied upon to reduce exchange-rate risk as relative currency values become more volatile. One possible caveat to our analysis is that it is static. From a dynamic standpoint, one may be tempted to argue that a country can engage in intertemporal optimization under the floating exchange-rate system by hedging efficiently in the forward exchange market. If exchange risks are fully hedged, there will then be no need for quota protection to stabilize domestic prices. However, the crucial assumption in this line of thinking is that exchange risks can indeed be effectively hedged. As we will argue below, under many circumstances, economic agents cannot fully offset currency risks by utilizing existing markets for forward exchange. First, however, let us consider the problem in the absence of forward hedging against exchange risk.

224

R. I . M C K I N N O N AND K. C. FUNG Domestic

Net excess demand for imports

DD

Imports

Qa

Figure 10.1 Exchange-rate and domestic price fluctuations under free trade

2

The non-equivalence of tariffs and quotas in the face of exchange fluctuations

In the face of severe exchange-rate volatility, countries will attempt to utilize policies that can minimize domestic price exposure to such fluctuations. We argue that traditional fixed-rate tariffs are ineffective in reducing exchange risk. Instead, import quotas and other market sharing schemes will reduce the domestic price instability induced by exchange-rate changes. Furthermore, the severity of the quantitative restrictions required increases as the degree of exchange variability rises. As an alternative, sliding-scale tariffs and export subsidies, e.g., as utilized by the European Common Agricultural Policy program, can act as substitutes for import and export quotas for preserving domestic price stability. To focus our analysis, figure 10.1 shows an auction market under exchange uncertainty, e.g., a market for a homogenous agricultural good where exchange-rate changes are fully passed through into the domestic prices. DD is the net import demand curve (domestic consumption less domestic production) and fi is the mean domestic price with the corresponding mean of the exchange rate E (domestic currency/foreign

225

Floating exchange rates Domestic price

Net excess demand for imports

Imports

Qa

Figure 10.2 Exchange-rate and domestic price fluctuations under an ad valorem tariff

currency) set initially at one. As the exchange rate depreciates, the domestic price rises above \i while an exchange appreciation causes the domestic price to fall below \i. For illustrative purposes, we assume that under free trade, the domestic price varies from fi + 2atofi — 2
226

R. I. MCKINNON AND K. C. FUNG

Import quota

DD

Qa Exports

Figure 10.3 Exchange-rate and domestic price fluctuations under either import or export quotas

In order to stabilize prices over the range of Ee[l +2(7,1 — 2cr], the import quota must be more restrictive and set no higher thanj2. Essentially the quota truncates the import demand curve and the corresponding price acts like a price floor. From figure 10.3, it is easily seen that the restrictiveness of the quota necessary to stabilize domestic prices depends on the variability of the exchange rate. In other words, as the volatility of the exchange rate increases, the import quota must be reduced to keep the import price from fluctuating. Trade then becomes more and more restricted and the average domestic price and the (variable) license rent get higher. By focusing on the third quadrant of figure 10.3, we can further illustrate the case where the country is exporting instead of importing. This will be the situation if the average price is at fix with P e[fix — 2a,fix + 2a] under free trade. By similar argument as before, a restrictive export quota at jg* will stabilize the domestic price at fix — 2a or slightly below.

Floating exchange rates 3

227

The hedging problem

In response to the sharp volatility of the exchange rate, financial centers all over the world have developed a variety of instruments for hedging currency and interest-rate risks. With the rapid development of a rich menu of financial instruments that allow participants in the foreign exchange market to take forward positions, most economists are convinced that international currency risk associated with trade and investment can be effectively hedged. Nonetheless, we argue that, in a variety of settings, including importing, exporting, and investment in physical plant, there will still be substantial risk that cannot be hedged (McKinnon, 1979, 1988). The fundamental problem is that there are two interrelated risks involved - t h e currency exchange risk and commodity price risks. Complete hedging will require not only the existence of a forward exchange market, but also markets that allow forward contracting in goods and services. These observations stem directly from the insights of the Arrow-Debreu models, which highlight the importance of forward commodity markets that hedge against price risk and insurance markets that offset risks in production. But it is generally known that the setting up of a complete set of contingent future markets is not feasible in actual economies. Prohibitively high transaction costs, moral hazard associated with contract enforcement, etc. are just a few of the problems hindering the realization of forward markets that can cover all uncertain states of nature. Thus producers frequently cannot feasibly protect themselves with outside insurance policies against their inability to deliver, nor against unexpected cost changes (Greenwald and Stiglitz, 1986; Arrow, 1973). A farmer, unable to predict his own production nor able to buy insurance against a variety of possible natural disasters would not necessarily wish to sell his crops forward in advance (McKinnon, 1967). Given the uncertainty of delivery, producers typically sell forward only some of their potential output, often a short time in advance on a non-contingent basis. Instead, most goods are held in inventory and the market is cleared by spot purchase. Because of the incomplete forward markets for commodities, fixed investments that produce internationally tradable goods and services cannot be hedged completely against foreign exchange risk. We can illustrate the hedging problem by a thought experiment involving the decision of an entrepreneur who wishes to build a new plant to produce computer chips. First he must seek out the location or country where unit factor costs are low compared to expected output prices. With exchange-rate volatility, the calculations to seek out the country with lowest costs can easily be swamped by considerations of future currency movements.

228

R. I. MCKINNON AND K. C. FUNG

In order to cover fully the exchange risk, the planned production of computer chips has to be insured and sold forward over the lifetime of his plant. If the hypothetical forward prices for chips are specified in all relevant currencies over a continuum of future dates, then the forward foreign exchange market can be fully utilized. The entrepreneur could sell forward foreign currencies for the currency of the country of production at the various terms to maturity and amounts negotiated in any forward commodity contracts. However, such forward commodity contracts, which are fully insured against unexpected dislocation and delays in production, do not exist. For a merchant exporter with inventories of already produced goods, short-term exchange risk can be hedged when he sells consignments abroad to be delivered in, say, thirty days or six months. After forward selling his goods for foreign currency, he can sell foreign exchange forward to get safely back into the domestic currency, i.e., by double hedging. However, short-term hedging on a non-contingent basis covers only a small proportion of the potential longer-term risk arising from investment decisions that are made today. Effectively, exchange-rate instability together with incomplete forward commodity markets exposes entrepreneurs with long-term fixed investments to foreign exchange risks that are unhedgable. In recent years we have witnessed dramatic examples of this unhedgable intermediate-term exchange risk in the world economy. When the dollar was generally weak in the 1970s and became greatly undervalued from 1977 to 1980, American tradable goods industries looked profitable and "excessive" investment occurred in certain kinds of US mining and manufacturing, with agriculture also becoming overcapitalized. As the dollar unexpectedly rose in 1981, these industries suffered from excess capacities. There were many bankruptcies and plant closures. Judging from the experience of these companies, they obviously did not (and could not) hedge, their foreign exchange positions. Although by 1987 the dollar was no longer overvalued, exchange-rate volatility still casts a shadow over investors' decisions to commit to long-term investments. In assessing whether or not to reinvest in America's productive capacity in tradable goods industries in the early 1990s, interbloc exchange risk (from trade with partners such as Europe or Japan outside of the North American Free Trade Area) is a significant inhibiting factor. Note that investment choices were not necessarily incorrect ex ante. Rather entrepreneurs face unpredictable and unhedgable currency changes which cause havoc for investors planning to build factories for future production. Though the Japanese seem to have weathered the storm of the currency volatility in the eighties, the shock or "endaka" of the rising yen in

Floating exchange rates

229

1985-6 did prompt Japan's preeminent industrialist, Mr. Akiro Morita (1986), President and Chairman of the Sony Corporation, to call for reforms such that national money becomes "a common scale of value internationally rather than just another speculative commodity."

4

Empirical overview

In parallel with the broad trend toward tariff reductions, there has been in the past fifteen years a serious move toward making interbloc quantitative restrictions more extensive. To underscore the growing importance of non-tariff barriers, table 10.1 presents non-tariff measures for OECD countries for the years 1981, 1983, and 1986. Table 10.1 shows that, in 1981, hard-core non-tariff barriers were used in 15.1 percent of OECD non-petroleum imports. This would mean that $80 billion in trade was affected by hard-core NTBs. Table 10.1 also shows that in 1981,14.3 percent or $49 billion of OECD imports from other developed countries were affected by hard-core NTBs. More importantly, table 10.1 illustrates the rising importance of NTBs. Overall, from 1981 to 1986 the share of imports affected by hard-core NTBs rose by 2.6 percent, an increase equivalent to covering approximately an additional $14 billion in trade. Table 10.2 further disaggregates the changes in the use of non-tariff barriers in sixteen QECD countries. The individual country statistics show that the US had the greatest expansion in new trade restrictions as the share of US imports affected rose from 11.4 percent in 1981 to 17.3 percent in 1986. For the EEC, the index rises from 13.4 percent to 15.8 percent over the same five-year period. True, the data in tables 10.1 and 10.2 cover only the relatively short time frame from 1981 to 1986. But we hypothesize that the increasing tendency to use NTBs in interbloc transacting has continued from the mid 1970s into the 1990s. Only within emerging trading blocs where countries have succeeded in stabilizing exchange rates among themselves - as in the EEC or North American Free Trade Area - have quantitative barriers to trade been reduced. In the last section, we demonstrated that fluctuating exchange rates are unpredictable and cannot be fully hedged. Sharp volatility of the exchange rates translates into big fluctuations in the domestic currency prices of tradable goods. This problem of exchange-rate induced price variability is magnified in some specific sectors such as agriculture, where the goods produced are homogenous and the market structure is perfectly competitive. In the absence of quantitative restrictions on foreign trade, exchangerate pass through in agriculture is immediate and complete.1

230

R. I. MCKINNON AND K. C. FUNG

Table 10.1. The share of OECD country non-petroleum imports subject to hard-core non-tariff barriers

Importers/trade partners

1981

1983

1986

1981-6 change

OECD/world OECD/developed OECD/developing OECD/socialist

15.1 14.3 18.8 9.3

16.7 16.3 19.6 10.2

17.7 17.5 20.6 9.8

+ 2.6 + 3.2 + 1.8 + 0.5

Note: Hard-core non-tariff barriers include: variable import levies and productspecific charges; quotas; prohibitions; non-automatic import authorizations including restrictive import licensing requirements; quantitative 'voluntary' export restraints; and trade restraints under the Multifibre Agreement. OECD countries include: Belgium-Luxembourg, Denmark, France, West Germany, Great Britain, Greece, Ireland, Italy, Netherlands, Finland, Japan, New Zealand, Norway, and the US. Source: Laird and Yeats (1990). Table 10.2. Share of non-petroleum imports subject to hard-core non-tariff barriers: individual countries

Importer

1981

1983

1986

1981-6 change

Belgium-Luxembourg Denmark Germany, Fed. Rep. France Greece Great Britain Ireland Italy Netherlands EEC (10) Switzerland Finland Japan Norway New Zealand United States All above

12.6 6.7 11.8 15.7 16.2 11.2 8.2 17.2 19.9 13.4 19.5 7.9 24.4 15.2 46.4 11.4 15.1

15.4 8.0 13.6 18.8 21.0 13.4 9.7 18.7 21.4 15.6 19.6 8.0 24.5 14.7 46.4 13.7 16.7

14.3 7.9 15.4 18.6 20.1 12.8 9.7 18.2 21.4 15.8 19.6 8.0 24.3 14.2 32.4 17.3 17.7

1.7 1.2 3.6 2.9 3.9 1.6 1.5 1.0 1.5 2.4 0.1 0.1 -0.1 -1.0 -14.0 5.9 2.6

Source: Laird and Yeats (1990).

Floating exchange rates 5

231

Agricultural protectionism within the EEC

The problem of exchange-rate changes and the domestic currency prices of agriculture is vividly illustrated by the experience in the EEC before exchange rates were effectively stabilized through the European Monetary System (EMS). As we will show, a dizzying array of policy instruments, including artificial exchange rates for agriculture ("green" rates) and border agricultural taxes and subsidies called Monetary Compensatory Amounts (MCAs), were employed to shield national farm prices from the effects of intra-EEC exchange-rate fluctuations and to deal with the consequences of such insulated prices. Now, this complicated apparatus of sliding-scale import tariffs and export subsidies is mainly applied to external EEC agricultural trade with the rest of the world. Therefore, if international agricultural trade is liberalized between the US and the EEC by, say, substituting modest fixed-rate ad valorem tariffs for the system of pseudo quotas (sliding-scale tariffs and export subsidies), national food prices will be directly affected by the volatile exchange rate between the US dollar and the European currencies. The European objective of stabilizing agricultural prices will then become increasingly difficult, if not impossible. Without a stable exchange-rate system, well-meaning talks of liberalizing the world trade of agriculture, as those sponsored by the GATT within the current Uruguay Round, will always run counter to the European (or Japanese) desire to maintain price stability in agriculture. This partly explains the great reluctance on the part of the EEC to negotiate for trade liberalization in agriculture. To highlight the problems posed by exchange-rate variability for the agricultural sector, we will briefly describe the history, institutions, and policies of the Common European Market and the Common Agricultural Policy (CAP). As noted by Giavazzi and Giovannini (1990), "postwar European institutions - particularly the common agricultural market depend for their survival on exchange rate stability." The main objective of the Common Agricultural Policy (CAP) is to maintain uniform support prices in agriculture. The EEC in 1962 decided that the "European Unit of Account" (EUA) which was defined in terms of quantities of gold equal in value to 1 US dollar, would be used for the determination of agricultural prices under the CAP. The official IMF exchange rates between the European currencies and the dollar would then be used to convert the CAP prices into each member's national currencies. In the early phases of the Bretton Woods fixed exchange-rate system, the EUA-based CAP prices worked well to preserve agricultural price stability. However as exchange-rate realignments occurred more frequently in the

232

R. I. MCKINNON AND K. C. FUNG

late 1960s, domestic agricultural prices began to diverge among members of the EEC. Consequently a system of compensating border taxes and subsidies was needed. As described in detail by Gardiner and Josling (1991), the beginning of the exchange-rate problem for CAP occurred when France devalued its currency by 12.5 percent in August 1969. Before the devaluation, the minimum price floor for wheat in the EEC was 98.75 EUA/ton. Using the official exchange rates, the French and German governments can calculate their respective intervention prices. Intervention prices are domestic prices that the EEC guarantees to purchase from farmers. After the French devaluation, agricultural prices in the EEC were kept fixed in terms of the EUA. Using the new franc exchange rate would mean an increase in agricultural prices in France by 12.5 percent. At that time, the French government wished to avoid the political costs of food inflation in France. To stabilize agricultural support prices in domestic currency, France was allowed to use the pre-devaluation exchange rate to calculate the franc equivalent of the CAP prices. Thus a "green' exchange rate was "created" for the sole purpose of calculating agricultural support prices. The use of the green rate led to lower agricultural prices in France compared to other EEC countries. The price divergence created arbitrage opportunities that would eventually defeat the purpose of containing food-price inflation in France. To prevent arbitrage, Monetary Compensatory Amounts (MCAs) were adopted. MCAs are border taxes and subsidies aimed at insulating the domestic-currency prices from exchange-rate realignments. In this case for France, the MCAs took the form of export taxes and import subsidies and are called negative MCAs.2 Basically MCAs are the difference in the French intervention prices before and after the franc devaluation. These taxes and subsidies were administered at the border posts. The problem of exchange-rate realignment and agricultural prices was by no means unique to France. In October 1969, West Germany revalued the mark by 8.5 percent. In the case of wheat, this would imply a drop of the intervention price and a decline in the income of German farmers. Like the French, the West Germans decided to use the old rate to keep the food price stable in DM. As in the case of the French green rate, MCAs were needed to prevent arbitrage. Thus an import tax and an export subsidy were applied at the German border. West Germany, which maintains its support price above the uniform level, is said to have positive MCAs.3 Giavazzi and Giovannini (1990) provides a clear hypothetical example of how the MCAs work, which we modify for our illustration:

Floating exchange rates

233

suppose . . . 1 EUA = 1 DM = 1 lira, and the intervention price for milk is 10 EU A/liter. A realignment occurs, implying a 10% appreciation of the Deutsche mark and 10% depreciation of the lira, both relative to the EUA. The intervention price for milk is unaffected by the realignment. Therefore, the price for milk after the realignment should fall to 9 marks in Germany and rise to 11 lira in Italy. However, domestic-currency prices are kept unchanged. After the realignment the price of milk in Germany stays at 10 DM, thus effectively rising from 10 to 11 EUAs; in Italy, it stays at 10 lira, thus falling from 10 to 9 EUAs. The EEC eliminates the possibility of arbitrage profits by taxing milk exports from Italy and subsidizing milk exports from Germany. For each liter of milk that a German farmer exports to Italy, he receives 1 EUA (a positive MCA) from the EEC that compensates him for the lower prices on the Italian market. Conversely, when an Italian farmer ships milk to Germany he is taxed; the tax rate is also equal to 1 EUA (a negative MCA).4

The MCAs, as discussed earlier, are designed as temporary devices to help stabilize farm prices and they are supposed to be removed over the long run. But the removal of the MCAs has proven to be uneven. In weak currency countries such as France and Italy, negative MCAs are from time to time eliminated by domestic price increases. But for strong currency countries like West Germany and the Netherlands, the positive MCAs tend to persist. Positive MCAs are difficult to remove because the required adjustment is one of domestic price decline. This would lead to a drop in farm income. Well-organized domestic farmers will lobby strongly against such measures. In contrast, the removal of negative MCAs leads to price rises, which will have the support of the farm community. Indeed, such asymmetric adjustment was codified in 1979 in the EEC's "Gentleman's Agreement," stating that any removal of MCAs should not lead to a drop of domestic farm prices. Thus nominal exchange-rate changes interact with the political economy of the MCAs, leading to a real impact on agricultural production in Europe. Relatively speaking, farm production is stimulated in strong-currency countries while production is discouraged in weak-currency countries. The biased adjustments of MCAs create a ratchet effect on European farm prices. In fact, the average intervention prices have been rising over time. Overproduction of agricultural products worsens in Europe and this puts a further strain on the budget of the EEC. For example, in 1984, support for milk and dairy products was the single largest item on the EEC budget, accounting for 20 percent. Green exchange rates, and positive and negative MCAs are the direct and indirect consequences of the Europeans' attempt to stabilize farm prices in the face of exchange-rate variability. The Europeans are well aware of the

234

R. I. MCKINNON AND K. C. FUNG

fact that exchange-rate fluctuations would endanger the CAP and the creation of a customs union. As early as 1969, the Barre Report stated "the commission estimates that a widening of the fluctuation margins for the exchange rates of member countries would pose important problems in the area of the common agricultural policy and in that of trade relations in the community, and especially that it would compromise the further integration of markets." The EECs concern for currency fluctuations and their impact on domestic prices is clearly reflected in the Uruguay Round talks on agriculture. Currently, a variable levy system is used in the Community to offset any changes in the world price of agricultural products. Given the dollar gyrations in the 1980s, it is safe to say that an important source, if not the dominant source of farm price instability, is currency variability. With their experience of recent exchange-rate changes, the Europeans are keen to protect agricultural producers and consumers from future fluctuations of exchange rates. Given our discussions of the comparative insulation properties of tariffs and quantitative restrictions in section 2, one may ask why import levies are used by the EEC for external protection. However, as we will show next, a variable levy system is fundamentally different from a fixed rate tariff. In terms of preserving internal price stability, sliding-scale import taxes are equivalent to quotas where the government appropriates the license fees or quota "rents." Thus as a substitute for a quota, we can examine the price effects of a sliding-scale tariff or a variable levy. If we can condition / to depend on o and with / < 0 representing an import subsidy, then a slidingscale tariff t(a) can fully preserve domestic price stability. For example, in figure 10.4, if the exchange rate appreciates and the corresponding domestic price incipiently falls to fi — 2a, a tariff of t = 2o will restore the price to \i. Similarly, if the domestic price incipiently rises to fi+ 2<J from exchange depreciation, an import subsidy o f / = — 2<7 will lower the price back to \i. If the industry in question is instead a natural exporter and operating at an initial price of fix with exchange and price variability measured by P s[fix — 2(J,fix + 2
Manufactures

In manufacturing industries, unlike the farm sector, the goods produced range from being homogeneous (for example, cement or low-cost

235

Floating exchange rates

DD

Exports

Imports

Figure 10.4 Exchange-rate fluctuations offset by sliding-scale tariffs and export subsidies

garments) to highly differentiated (for example, consumer electronics or automobiles). The market structures in manufacturing are thus heterogeneous, with some industries having more monopolistic powers than others. Given such diversity, it is not surprising that exchange-rate volatility will have a much richer set of linkages with potential trade policy responses in manufactures. The general impact of fluctuating exchange rates on the creation and the sustaining of trade barriers is, however, similar to that discussed in earlier sections. A nominal appreciation of the dollar, given the inertia in price levels, leads to a real US exchange-rate appreciation. This has an adverse effect on the competitiveness of the US import-competing and export sectors. For the perfectly competitive manufacturing industries, the effects of the dollar appreciation are well-known. As stated before, a 10 percent exchange-rate appreciation is equivalent to a 10 percent export tax plus a 10 percent import subsidy. Even if the industry is oligopolistic, the effects are generally similar. The effects on competitiveness can be demonstrated in figure 10.5. For illustrative purposes, we focus on an international Cournot-Nash quantity-setting industry. In figure 10.5, we describe a situation where the US domestic firm,

236

R. I. MCKINNON AND K. C. FUNG

Y H \ G O

Ctl

VE'

reign

I o

F

^

^

Foreign response function after exchange appreciation

E V^V \\

\

Domestic \ response ^. function \

\

\

V \\ \

1

Domestic production

\

F\ \

Figure 10.5 Exchange appreciation in a Cournot producer duopoly producing output X, is competing with a foreign firm, producing a differentiated product Yin the US market. HH is the best response curve for the US home firm, i.e., HH traces out the locus of the best output of the domestic firm, given the foreign output. FF is the best response curve for the foreign firm. A dollar appreciation will shift the best response curve of the foreign firm out, resulting in increased foreign market share and a reduction of the US market share. At the initial equilibrium E, the foreign firm faces a given dollar marginal revenue curve but a lower dollar marginal cost curve due to the appreciation. This leads to increased foreign output. For the US domestic firm, market share, employment, and profits all decline. Figure 10.5 can also be used to describe a US exporting firm competing in the foreign market. The dollar appreciation will then have the same effect on the US firm in the export market as in the domestic market: a decline in US sales, employment, and profits. It is clear that industries are aware of the adverse competitive effects of an appreciating currency. In 1983, the US textile industry alleged that China's dual exchange rate constituted a Chinese export subsidy. In 1984, the copper industries charged that one of the ways in which Chile and other LDC were subsidizing their copper industries was by devaluing their currencies. As the dollar rose in the first half of the eighties, real textile exports fell by 40 percent while imports almost doubled. In addition, real apparel imports rose by 129 percent.

Floating exchange rates

237

The squeeze in the export and the domestic markets caused by the dollar appreciation led to intense lobbying by various industries and labor groups. Unlike the scenario of multilateral trade liberalization talks which hurt import-competing industries but benefit the exporting industries, currency appreciation hurts the entire tradable sector. This generates a strong coalition of industrialists and labor groups which lobby for increased protection. For example, highly restrictive textile bills passed Congress in 1985 and in 1987. Though the bills were vetoed, the Administration responded to the lobbying pressure from the textile industry by imposing much tighter restrictions under existing arrangements and by tightening protection under the 1986 Multifiber Agreement. Protection in autos and steel were also continued. The administration clearly understood the role of the exchange rate in increasing the protectionist sentiment. In September 1985, the Plaza Agreement was reached to coordinate actions of central banks to bring down the dollar. An appreciating currency, as discussed earlier, creates great pressures on the government to sustain, increase, and erect trade barriers. But, in periods when the currency depreciates, it is much less likely that trade barriers will go down with the currency cycle. The reason rests again primarily on political economy grounds. With a lower home exchange rate, both producers in the home-exporting and import-competing industries tend to increase their competitiveness. From the domestic viewpoint, commercial policies to assist these firms are no longer needed. However, there are really no incentives for these producers unilaterally to give up import quotas or export subsidies. In fact, existing protectionist policies are rarely removed once they are installed since the producers and workers in the affected industries develop vested interests in these programs. The factors of production in the tradable sector will spend resources (time, money, and effort) to engage in the preservation of these policies. Furthermore, domestic producers will use the legitimate argument that new investment to reequip their industry will be too risky unless quantitative restrictions remain in place to offset the effect of any future exchange appreciation. From the foreign country's viewpoint, the home currency's depreciation is the foreign country's exchange-rate appreciation. By the same logic described before, the foreign firms (whether they are perfectly competitive or oligopolistic) will lose market share. This leads to a drop in employment and profits. By the process of political lobbying, the foreign government will have to respond by increasing their own trade barriers. Thus for each cycle of currency appreciation and depreciation, there will be a tendency for worldwide trade barriers to rise. Given the frequency and the magnitude of exchange-rate volatility under the floating-rate system, it

238

R. I. MCKINNON AND K. C. FUNG

is not hard to understand the increasing pressures that governments face in pursuing and sustaining various forms of trade policies. In addition to the general tendencies of increasing pressures for adopting protectionist policies, volatility in exchange-rates has also led to the adoption of specific forms of protectionism. We can describe American trade policies in the 1980s by three major tendencies: the proliferation of quantitative restrictions and other market-sharing schemes, the increased use of anti-dumping duties and countervailing duties, and unilateral trade actions (such as US Super 301) in pursuit of "fair" trade. We argue in earlier sections that the increased use of quotas, VERs, and other market-sharing schemes since the mid 1970s is a natural outcome of fluctuating exchange rates. As shown before, exchange-rate variability has the property that promotes a general rise of trade barriers. But domestic industries that seek to shield their industries from exchange-rate shocks will prefer quantitative restrictions rather than tariffs. Unfortunately, from a welfare standpoint, the use of quantitative restrictions often leads to more pernicious effects than a fixed-rate tariff. The effects of tariffs are generally well understood. Tariffs are more transparent and it is much easier for international negotiators to negotiate for lower tariffs. In contrast, quantitative restrictions are much less understood, at least by the public. The lack of transparency makes it easier for the affected groups to sustain the barriers without much opposition. In the case of voluntary export restraints (VERs), the entire quota rents, i.e., the economic profits generated by the discrepancies between the domestic price and the world price, are given to the exporting country. This adds to the welfare loss of the home country. In the case of tariffs, the home country is able to capture the equivalence of the quota rents in the form of tariff revenue. At least for some countries, there are additional reasons why countries will opt for quantitative restrictions (QRs) instead of tariffs to protect them from the fluctuating exchange rates. As discussed in McKinnon (1979), the answer may lie in politics: "For countries with rather weak constitutional limits on the assertion of economic power by the government of the day, favorable QR allocations can reward one's political friends, and the denial of licenses - including the permission to export freely . . . can be economically devastating to groups who are politically unfriendly. And the pressures to distribute positive and negative patronage this way become more difficult to resist once a formal QR regime is put in place" (McKinnon, 1991a, p. 101). Who claims the quota rents depends on how the quantitative restrictions are administered and how the rights to import and export are licensed. In the case of VERs, the rights of the quota rents are automatically granted to

Floating exchange rates

239

the exporting country, even though who in the exporting country eventually captures the rents remains to be settled. In other instances of quotas, the importing country retains the claim to the rents. The potential competing claims to these rents create classic situations for rent-seeking activities, either at home or abroad. Important scarce resources including skilled professionals are diverted from production activities into lobbying and rent-seeking activities. Thus though QRs are rational responses by governments in a world of floating exchange rates, quotas and VERs generally distort the world economy to a larger degree than import taxes. The prevalence of VERs, furthermore, can interact with old legislation such as the anti-dumping laws to generate further protectionist outcomes. As discussed earlier, VERs create economic rents that are captured by the exporting country. To allocate the rights to export, an often-practiced rule is to link the past export market share to the extent of quota licenses. Thus, if exporters perceive that a VER is probable, as perhaps due to new legislation by the US Congress, then a multi-period profit-maximizing response would be to expand export sales immediately before any future VER is implemented. This is done in anticipation of future claims to the quota rents. For perfectly competitive firms generally and for oligopolistic firms under some conditions, this will imply that foreign exporters sell in the US market at below cost, thus constituting dumping.5 This type of dumping is not "predatory" in the usual sense, since it is not meant to drive out domestic competitors in order to establish an illegal monopoly. Rather anticipatory dumping here is a reaction to profit opportunities created by a legally sanctioned market-sharing arrangement (a VER). Dumping, as induced by prospective VERs, will nonetheless injure domestic producers since home prices will drop. This will naturally trigger filing of anti-dumping complaints. Since there are actual dumpings, there can be one of two results. One is the imposition of an anti-dumping duty. But this may not be sufficient to deter dumping, depending on the foreign producers' perceived probable gains from the VERs. This can eventually lead to a curious mix of protection in the form of anti-dumping taxes and VERs. Alternatively, the foreign firms can agree to the domestic producers' demand to raise export prices, thus settling the anti-dumping cases out of court. But this will have a potential effect of creating an implicit cartel between the domestic and the foreign firm. Indeed, out-of-court settlements of anti-dumping suits frequently take the form of the foreign exporter agreeing to limit his sale in the domestic market. Such a "legalized" cartel is thus sustained by an informal VER.

240

R. I. MCKINNON AND K. C. FUNG

In either scenario, the interplay of potential VERs and anti-dumping legislation leads to further restrictive or protectionist acts. But, as mentioned before, the root of the phenomenon may lie in the floating exchange-rate system. Another channel whereby floating exchange rates can lead to dumping is when an invoicing currency such as the US dollar depreciates unexpectedly. It is fairly well-known that, for a variety of differentiated manufactured goods, the export dollar prices of these goods in the US are relatively fixed. But translated into dollars the price of the same good in the exporters home country can fluctuate with the exchange rate. This can lead to price divergence. As a thought experiment, suppose a Japanese auto costs $10,000 in the US, 1,400,000 yen in Japan and the exchange-rate is 1 $ = 140 yen so that the law of one price holds initially. Now let the dollar depreciate to l$ = 120 yen and assume that for a variety of reasons (e.g., pricing to market), the dollar price of the auto in the US stays constant at $10,000. The yen price of the auto will likely be unchanged since Japanese producers have to pay wages and rents in yen. But the dollar price in Japan becomes $11,667. Thus the Japanese producers find themselves selling cheaper abroad than at home, i.e., they are dumping. This can cause complaints of dumping and can lead to anti-dumping duties. Various countries' attempts to shield their markets from exchange-rate uncertainties feed the growing perception that these are "unfair" trade practices. The major piece of legislation used by the US is the strengthened section 301 of the 1988 Omnibus Trade and Competitiveness Act: the so-called "Super" 301. It authorizes the Administration to retaliate against trading partners if successful negotiations are not concluded to open up their markets for US exports. As noted before, there are actual legitimate reasons why countries wish to protect their markets from exchange-rate fluctuations by means of quantitative restrictions and other market-sharing devices. On the other hand, it is also true that such attempts to shield their domestic markets can act as barriers against US exports. Thus, legislation such as the Super 301 and its resulting trade tensions are companions of the current exchange-rate system. 7

A concluding note on regional trading blocs

In the 1950s and 1960s, when par values for exchange rates were more or less fixed under a common monetary standard the industrial countries experienced an unprecedented increase in the growth rate of GNP and an expansion of world trade. Tariffs were reduced and quantitative restrictions were virtually eliminated. With the advent of the floating

Floating exchange rates

241

exchange-rate system after 1973, however, we witnessed an upsurge of "new" protectionist policies - including quotas, VERs, market-sharing schemes, anti-dumping, and countervailing duties. In this chapter, we explored the hypothesis that the volatility of exchange rates contributes to the increase in these "new" forms of protectionism. We showed that, to maintain domestic price stability, quantitative restrictions will be preferred over fixed-rate tariffs, while sliding-scale tariffs or variable levies, as^practiced in the agricultural sector in the EEC, have similar insulating properties as do quotas. Thus, does exchange-rate volatility erode the global free-trade ethic (most favored nation principle) as envisioned in the GATT? Similarly, the tendency for the world to break down into regional trading blocs - within which more stable exchange rates based on a common monetary standard can more easily be established - has been greatly strengthened. For economies closely integrated in foreign trade, a zone of exchange-rate stability now seems necessary to preserve free trade in goods and services while reducing investment risk. The EMS trading bloc in Europe is now paralleled by an emerging North American free trade area, where Canada and Mexico are currently orienting their domestic monetary policies toward keeping their exchange rates within a narrow range against the US dollar. In the not-too-distant future, a yen-bloc among countries closely linked to the Japanese economy could well emerge. While such regional trading blocs, each with its own common monetary standard, broaden the ambit of free trade (trade creation) internally, they are very much a second-best way of limiting the damage from global exchange-rate instability. Indeed, as long as untethered exchange rates among the major blocs of Europe, North America, and Japan continue to swing wildly of the order of 10 to 30 percent from one year to the next, interbloc protectionism in the form of quantitative restrictions, export subsidies, and generally heightened mercantilist rivalry may well be accentuated.

Notes 1 The added problem is that for some farm inputs (e.g., tractors), nominal contracts, menu costs, and pricing to markets can lead to delayed responses of input prices to exchange-rate changes. The differential responses of inputs and outputs in agriculture can greatly exacerbate the problem of income stability for farmers. 2 In trading with non-EEC countries, the MCA was subtracted from the external import levies and export subsidies.

242

R. I. MCKINNON AND K. C. FUNG

3 Positive MCAs are added to the external import levies and export subsidies. 4 Note that, in the original example in Giavazzi and Giovannini (1990), ECUs instead of EUAs are used. 5 For references, see Yano (1989) and Anderson (1991). References Anderson, J.E. (1992), "Domino Dumping, I: Competitive Exporters," American Economic Review, March, pp. 65-88. Arrow, K. (1973), Essays in the Theory of Risk Bearing, Chicago: Markham Press. Dornbusch, R. (1976), "Expectations and Exchange Rate Dynamics," Journal of Political Economy, 84, 1161-76. Frenkel, J. and K. Froot (1988), "Chartists, Fundamentalists, and the Demand for Dollars," in Tony Courakis and Mark Taylor (eds.), Policy Issues for Interdependent Economics, London: MacMillan. Frenkel, J. and M. Mussa (1980), "The Efficiency of the Foreign Exchange Market and Measures of Turbulence," American Economic Association Papers and Proceedings, 70, 374-81. Friedman, M. (1967), The Balance of Payments: Free versus Fixed Exchange Rates, Washington, DC: American Enterprise Institute. Gardiner, W.H. and T.E. Josling (1991), "Dismantling the EC's Agri Monetary System: Impacts on European Agriculture," mimeo, Stanford University. Giavazzi, F. and A. Giovannini (1990), Limiting Exchange Rate Flexibility: The European Monetary System, Cambridge, Mass.: The MIT Press. Greenwald, B. and J. Stiglitz (1986), "Externalities in Economies with Imperfect Information and Incomplete Markets," Quarterly Journal of Economics, 101, 229-64. Johnson, H.G. (1973), The Case for Flexible Exchange Rates, in Further Essays in Monetary Economics, Cambridge, Mass.: Harvard University Press. Laird, S. and A. Yeats (1990), Quantitative Methods for Trade-Barrier Analysis, London: Macmillan. Meese, R. and K. Rogoff (1983), "Empirical Exchange Rate Models of the 1970s: Do They Fit Out of Sample," Journal of International Economics, 14, 3-24. (1990), "Non-linear, Non-parametric, Non-essential Exchange Rate Estimation," American Economic Review Papers and Proceedings, 192-6. McKinnon, R.I. (1967), "Future Markets Buffer Stocks and Income Stability for Primary Producers," Journal of Political Economy, 75, 844—61. (1979), Money in International FLxchange: The Convertible-Currency System, New York: Oxford University Press. (1988), "Monetary and Exchange Rate Policies for International Financial Stability: A Proposal," Journal of Economic Perspectives, 2, 83—103. (1991a), The Order of Economic Liberalisation: Financial Control in the Transition to a Market Economy, Baltimore: John Hopkins Press. (1991b), "Exchange Risk and Interest Rate Volatility in Historical Perspective," mimeo, Stanford University.

Floating exchange rates

243

McKinnon, Ronald and Kenichi Ohno (1989), "Purchasing Power Parity as Monetary Standard", in O. Hamouda, R. Rowley, and B. Wolf (eds.), The Future of the International Monetary System: Change, Coordination or Instability?

Aldershot: Edward Elgar Publishing, pp. 42—67. Morita, A. (1986), "Presentation to International Working Round on Exchange Rates and Coordination," Zurich, June 19. Yano, M. (1989), "Voluntary Export Restraints and Expectations: An Analysis of Export Quotas in Oligopolistic Markets," International Fconomic Review, 30, 707-24.

CHAPTER

II

The theory of tariffs and monetary policies ROBERT MUNDELL

Ever since the Plaza Accord, when the finance ministers of the group of five largest industrial countries agreed to a depreciation of the dollar, monetary, and exchange-rate policies have been used as an alternative to protection. It was no secret that the motivation for the Plaza Accord was the intent to ward off protectionist pressures in Congress. Devaluation of the dollar was looked upon as a means of improving US competitiveness in the face of increasing foreign penetration of domestic markets and a soaring trade balance deficit. The failure of the balance of trade to respond to changes in the exchange rate puts into question the economic theory that predicted it. Classical international trade theory did not establish any connection between exchange rates or tariffs and the balance of trade. On the contrary, the classical model assumed that the balance of payments was self-adjusting at the level of the balance of trade determined by autonomous classical movements. Trade policy could affect the terms of trade and the level of prices but not the balance of trade. There was one exception that was clarified by the development of the monetary approach to the balance of payments. With unchanged terms of trade, the imposition of a tariff would raise the equilibrium level of prices of the tariffed good and hence require a larger stock of money. Under the gold standard, a tariff would increase the demand for money, reduce expenditure below income, and generate a surplus in the balance of trade sufficient to import the larger stock of gold required to back the monetary expansion. Although the relation between tariffs and the terms of trade and tariffs and domestic price ratios has been well-developed in the literature, there has been scant attention paid to the relation between tariffs, money price levels, 244

The theory of tariffs and monetary policies

245

and exchange rates under alternative monetary regimes. It seems useful, therefore, to discuss the subject through the use of a simple general equilibrium model amenable to graphical analysis. This chapter will therefore analyze the effects of export and import taxes at home and abroad under conditions of both fixed and flexible exchange rates. The final section will develop the analysis in the context of discriminatory tariff preferences and the theory of customs unions. 1

The balance of payments

The model is best expressed by a general equilibrium system based on conditions that demands and supplies are in balance in each goods market and for the money supply. In the absence of trade, the equations of (autarkic) equilibrium for a country, A, are as follows:

X=X{px,py,M) = 0 Y=Y(px,py,M) = Q where X, Y, and L, denote the excess demands for X and Y and for money. These three equations are assumed to be homogeneous of degree zero in the two prices and the money supply, so that changing both prices, and the money supply, in the same proportion leaves excess demands unchanged. When the supply of money is given, we are left with three equations in two unknowns. The system would be overdetermined were it not for Walras's Law, which requires that PxX+PyYL

=0

Consistence is restored because the homogeneity postulate makes one of the variables, and Walras's Law one of the equations, redundant. The equilibrium under autarky is portrayed in figure 11.1, which plots the prices of Xand Yon the abscissa. The configuration around the point Q depicts the intersection of the three excess demand curves on the assumption that the stock of money is constant. Thus X X and YY describe the locus of the combinations of the money prices of X and Y at which there is zero excess demand for, respectively, good X and good Y; the curve LL similarly describes the combinations of prices that will yield equilibrium between the demand for and supply of money. The three curves are drawn on the assumption that the commodities and money are gross substitutes with respect to each other. In passing, it should be noted that a given proportionate increase in the stock of money would result in an equiproportionate displacement of the equilibrium along the ray OQ extended to, say Q'.

246

ROBERT MUNDELL

Figure 11.1 Equilibrium exchange rates and the terms of trade

Now consider the opening up of trade. I shall make the assumption initially that country A is too small, relative to the rest of the world, to affect international prices. In this case we can suppose that the international prices of X and Y are fixed at the levels epx* and epy*, where the asterisked terms are prices in terms of the international currency (such as gold or pounds or dollars), and e is the exchange rate, defined as the price of the international currency in terms of A's currency. When the possibility of international trade appears, country A will find it profitable to export the commodity that is relatively cheap at home, and to import the commodity that is relatively expensive at home, compared to that abroad. Let us now abstract from transactions costs, information costs, transport costs, tariffs and non-tariff barriers. In that case, trade will increase until, in equilibrium, prices at home and abroad are identical. Under free trade, arbitrage will ensure that the law of one price, viz., thatpx = epx*, and py = epy*, will prevail. In figure 11.1, the equilibrium level of world prices is given by the point W and the terms of trade are given by the slope of the line OW.1 It may be seen that country A has a comparative advantage in commodity X and a comparative disadvantage in commodity Y; this is because the relative price of Y in country A, in autarky, is higher than the relative price of Y in international markets: the slope of the line OQ, which is the autarkic price of Y in terms of X in country A, is larger than the slope of the line OW, which is the international price of Y in terms of X.

The theory of tariffs and monetary policies

247

The equations of equilibrium under trade have to be modified to take account of imports and exports. Let E represent A's exports and 7, A's imports. Then the equations of equilibrium are now

X(Px,py,M) + E = 0 Y(Pxj>,M)-I=0 L(px,py,M) = 0

It may be seen now that there are five equations in six unknowns, px,py,E,I, Mand e. Under flexible exchange rates, M is given and the exchange rate is determined by the equations; under fixed exchange rates, e is given and Mis determined by the equations. Back now to figure 11.1. The exchange rate, the price of foreign currency in terms of A's domestic currency, is given by OV/OW; it can also be related to the home and foreign prices of either of the commodities, since . In general,

Note that this exchange-rate equation does not imply a relationship between the exchange rate and the terms of trade. The terms of trade (tot) equals

= pjpy = ep*lep * which cannot be derived from the exchange-rate equation. The exchange rate is a monetary variable and the terms of trade is a real variable; the exchange rate and money supply cannot therefore affect the equilibrium terms of trade. 2

The effect of a tariff

Let us now consider the effect of a tariff on imports in country A. Since country A imports commodity Y, the price of Y rises, at constant foreign prices, by the full amount of the tariff; our assumption that country A is small relative to the rest of the world means that the tariff has no effect on the terms of trade. It is useful now to take into account the assets backing the money supply. It is convenient to divide the assets backing the money supply, M, into a domestic component, D, and a foreign component, R, for international reserves, along the lines of the monetary approach, so that M = D - f eR. Another equation specifies the relation between foreign reserves and the money supply, R = aM, where a is the target ratio of

248

ROBERT MUNDELL

international reserves to the money supply. The equation system now becomes

Y(pX9py,M)~I=0

y

M= where T is the rate of the tariff on imports in country A. The conditions of the new equilibrium can now be determined by specifying the exchangerate/monetary system. Consider first a tariff under flexible exchange rates. Under the assumption that the central bank does not intervene in the foreign exchange market to change the level of reserves or supply new domestic credit, the money supply is constant. The tariff increases the price of import goods, initially by the full amount of the tariff, creating an excess demand for money, a reduction of domestic expenditure below income and an incipient export surplus; the exchange rate appreciates. But because the prices of export goods at home and abroad must be identical (when translated into different currencies), the appreciation of the currency means that the price of exports must fall in terms of country A's currency. In figure 11.2, the price of export goods, initially at OG, falls to OS, and the price of import goods becomes SU exclusive of the tariff and ST inclusive of the tariff. The equilibrium is determined at the point where the money requirements line LL is above the terms of trade line OW by the rate of the tariff UT/ST. The currency appreciates from OV/OW to OU/OW. The extent of the appreciation of the currency and the changes in domestic prices depends on how the demand for money is related to expenditures on the two goods. The exchange rate will be greater, the more the money is held for command over import goods. If the demand for money depends on the two prices in proportion to expenditure on the two goods, the change in the exchange rate will be greater the larger is the share of imports in total expenditure, i.e., the more open the country is. There are two extreme cases. In one case, the demand for money depends only on exportable goods and the LL curve is vertical; in this case the price of exports is unaffected by the tariff, the price of imports rises by the full amount of the tariff, and the exchange rate is unchanged. In the other extreme case, the demand for money depends only on importable goods and the LL curve is horizontal; in this case the price of imports inclusive of the

The theory of tariffs and monetary policies

249

Figure 11.2 Tariff equilibrium under flexible exchange rates

tariff remains constant and the price of exports and the exchange rate falls by the full extent of the rate of the tariff. Let us now consider the effect of a tariff when the exchange rate is fixed. In the absence of autonomous credit operations (changes in D), the money supply is determined by the balance of payments, surpluses causing an increase and deficits causing a decrease. We can assume, however, that domestic credit expansion takes place to the extent consistent with preserving a given ratio, a, of international reserves to domestic money, in conformity with the equation R = aM. In figure 11.3, the curve RR, denoting the equilibrium level of reserves, is drawn in a proportion to LL such that OC/OV = a. The imposition of a tariff by country A raises the price of the import good by the full amount of the tariff, creating an excess demand for money, a reduction of income below domestic expenditure and a surplus in the balance of trade and balance of payments. Pressure for the currency to appreciate is now offset by central bank purchases of foreign exchange or gold in foreign-exchange markets in exchange for new supplies of domestic currency. The excess demand for money is thus eliminated by an increase in the supply of money. In figure 11.3, the balance-of-payments surplus shifts the RR curve to R'R' and the increase in the money supply shifts the LL curve to L'L' where the domestic price of imports at the new equilibrium is higher than at the old by the full amount of the tariff; export prices remain unchanged.

250

ROBERT MUNDELL

"R

0

G

Px

Figure 11.3 Tariff equilibrium under fixed exchange rates The situations under flexible and fixed exchange rates are similar but not the same, even in real terms. In both cases, the relative price of imports and exports is the same, but there is a difference in the composition of wealth. In the case of flexible exchange rates, the money supply remained constant and there was no change in the balance of trade. In the case of fixed exchange rates, with accommodating monetary policy, there was an increase in international reserves that had to be financed by a temporary reduction of expenditure below income and a balance-of-trade surplus; this means that private-sector wealth has been reduced and public-sector wealth has been increased. The two systems could be made identical in real terms, however, if the central bank expanded domestic credit by enough to satisfy the excess demand for money without an increase in reserves; this would imply a reduction in a, the reserve ratio. 3

The symmetry of import and export taxes

The proposition, first advanced by Alfred Marshall in 1879, and developed by Abba Lerner in 1936, that export and import taxes have symmetrical effects, applies for the barter model of trade theory. The question is whether it also applies, or in what sense it applies, to the general equilibrium theory of trade incorporating money. To answer this question it will be convenient to write down the system allowing for both export taxes (e) and import taxes in the home country:

The theory of tariffs and monetary policies

251

Figure 11.4 Export tax and flexible exchange rates

X(px,py,M) + E = 0 Y(px,py,M)-I=0 L(px,py,M) = 0 M=D + eR It may be seen at once that the domestic price ratio is affected in the same way by export and import taxes. Both taxes drive a wedge between the domestic and international price ratios, raising the relative price of the import good relative to the export good. The monetary general equilibrium model does not alter that conclusion. There is, however, a substantial difference in the monetary effects. Whereas a tariff, under flexible exchange-rates, led to a currency appreciation with a deflationary effect on the prices of export goods, an export tax leads to a currency depreciation with an inflationary effect on the prices of import goods. Because international prices are fixed, the export tax forces down the domestic price of exportables, leading to an excess supply of money and a depreciation of the currency, raising the price of import goods. This case is illustrated in figure 11.4, where it is assumed that the initial equilibrium at V is disturbed by an export tax.

252

ROBERT MUNDELL

The tax, equal to TZ/HT, creates a wedge between the international and domestic prices of exports and results in a new, higher, equilibrium tax-inclusive price of export goods of HZ, and a fall in the home price to HT. The currency depreciates from an exchange rate of OV/OW to an exchange rate of OZ/OW. Consider now fixed exchange-rates. Whereas a tariff, under fixed exchange rates, leads to a balance-of-payments surplus and an increase in the price level, an export tax leads to a balance-of-payments deficit, an outflow of reserves and a decrease in the price level; the price of export goods at home falls by the full amount of the export tax. This case is illustrated in figure 11.5, where it is assumed that the initial fixed exchange-rate equilibrium is disturbed by an export tax. The tax, equal to DV/BD, lowers the home price of exports, and creates a balance-of-payments deficit and a reduction in the money supply until the equilibrium quantity of money, indicated by the L'L' curve, is reached. We can also consider the combined effects of tariffs and export taxes. In the classical barter theory of trade, export and import taxes, because of their symmetrical effects, can be compounded, so that the total protective effect of a tariff is the sum of export and import taxes. In this case, it would appear to make little difference whether relative prices were changed by export or import taxes. However, as we have seen, it becomes clear that export and import taxes have different monetary effects whether the exchange rate is fixed or flexible. Under flexible rates, for example, a tariff would result in the appreciation of the currency whereas an export tax would result in its depreciation. It is therefore normally possible to achieve a given protective effect, with no change in the exchange rate, by a suitable combination of export and import taxes. "Protective effect" is measured by the change in relative price of import and export goods. In figure 11.6, the combined protective effect of import and export taxes is measured by the change in relative prices in the home country, given by the difference between the slopes of OV and OS. To achieve the protective effect with no change in the exchange rate or the quantity of money requires establishment of the home price equilibrium at the point where the LL schedule intersects OS extended, i.e., at the point S itself. The establishment of such a point requires a tariff equal to SF/FE combined with an export tax equal to FV/DF, yielding a combined protective effect of SG/GE. 4

Effect of foreign tariffs and export taxes

A tariff or export tax abroad has the effect of turning the terms of trade against country A, reducing the opportunities for gains from trade.

The theory of tariffs and monetary policies

Figure 11.5 Export tax and fixed exchange rates

Figure 11.6 Protective policy mix at constant exchange rates

253

254

ROBERT MUNDELL

To analyze the impact of foreign tariffs and export taxes, it is necessary to modify the arbitrage equations in the general equilibrium system as follows

X(px,py9M) + E = 0 Y(pX9py,M)-I=0 px = epx*l(l+e)(l+T*) /> y = ( M=

where T* and e* are, respectively, the tariffs and export taxes in the rest of the world. From the arbitrage equations, it can be seen that the relative price levels in the rest of the world and at home are as follows: Px*IP,*lPxlP, = (1 + 6)(1 + T*)(l + T)(l + £*) Tariffs and export taxes in the rest of the world have the same effect on relative prices as those at home, although, of course, the welfare effects are fundamentally different. An export tax in the rest of the world has the effect of raising the price of A's imports, creating an excess demand for money and, under fixed exchange rates, a balance-of-payments surplus until a higher equilibrium quantity of money is attained. In figure 11.7, an export tax equal to TV/VS would raise the price of A's imports by VT, creating an excess demand for money that generates the balance-of-payments surplus that eventually results in a new equilibrium quantity of money indicated by L'L' to support A's new price equilibrium point at T. Under flexible exchange rates, an export tax in the rest of the world would appreciate country A's currency until the export tax just equalled the gap between home and foreign prices, compatible with the equality of money demand and money supply. In figure 11.8, the export tax of JK/JL in the rest of the world results in a set of prices in A indicated by the point K. At this equilibrium, A's currency has appreciated from OV/OW to OJ/OW. Whereas under fixed exchange rates, the export tax in the rest of the world has an inflationary effect in A, under flexible exchange rates it has a deflationary or neutral effect. Now let us consider a tariff in the rest of the world. Under fixed exchange rates the tariff has a deflationary effect in country A. With prices in the rest of the world constant, a tariff worsens down the price of exports in country A by the full amount of the tariff. At lower export prices there is an excess supply of money and a balance-of-payments deficit that lowers the supply of money until it is equal to the lower demand.

255

The theory of tariffs and monetary policies

Figure 11.7 Export taxes in ROW under fixed rates

0

L

Px

Figure 11.8 Export taxes in ROW under flexible rates

256

ROBERT MUNDELL

Figure 11.9 Tariffs in ROW under fixed rates In figure 11.9, the tariff in the rest of the world lowers country A's export prices by VS, where VS/NS is the rate of export tax. The new equilibrium at S will be achieved through a balance-of-payments deficit and a reduction in the money supply to that indicated by L'L'. Under flexible exchange rates, as might be expected, the result is different. The tariff in the rest of the world lowers domestic export prices, creating an excess supply of money and a depreciation of country A's currency. For a tariff in the rest of the world at the rate (see figure 11.10) KH/JK, equilibrium prices in country A are indicated by the point K. Country A's currency will have depreciated from OV/OW to OH/OW. The combination of tariffs and export taxes has a compounded effect on the rate of protection but, as we have seen, opposite effects on the exchange rate or the balance of payments. It is left as an exercise to the reader to prove that it is possible to achieve a given change in the terms of trade by means of a combination of export and import taxes in the rest of the world that will leave the balance of payments, the exchange rate, and the stock of money unchanged. Figure 11.11 illustrates the equilibrium when country A as well as the rest of the world have both tariffs and export taxes, for a case where the exchange rate and the quantity of money do not need to change. First note that there are three price ratios to take into account: relative prices in the rest of the world, the terms of trade, and the price ratio in country A. In figure 11.11, the export tax in the rest of the world is DC/DB, the tariff in the rest of

The theory of tariffs and monetary policies

257

Figure 11.10 Tariffs in ROW under flexible rates

A's price ratio

0

J

B

Px

Figure 11.11 Combination of tariffs and export taxes at home and abroad

258

ROBERT MUNDELL

the world is VD/ED, the export tax in country A is FC/GF, and the tariff in country A is HF/FJ. 5

Allowance for changes in the terms of trade

Up until now the analysis has assumed that the prices in the rest of the world are constant so that tariffs and export taxes in country A, hitherto assumed to be a small country, have no effect on the terms of trade. This assumption is relaxed in the present section in order to take into account mutual interaction of country A and the rest of the world in determining the terms of trade. The first step is to modify the general equilibrium system in order to allow for the fact that country A is no longer a "price-taker" in world markets. The new equations specify that the excess demands in country A are matched, in equilibrium, by corresponding excess supplies in the rest of the world. For the money market conditions, it is again true that the excess demands for money in each country must be zero, but it is now necessary to specify the nature of the international monetary system. One possibility is to introduce one or more commodities or reserve assets into which currencies may be made convertible, as under silver, bimetallic, gold, or SDR systems. Interesting as it may be, such a characterization would carry us far afield from the scope of the present chapter. I shall instead assume that the currency of one or all of the countries of the rest of the world, ROW, produces reserves, leaving up to country A the choice of fixing exchange rates or allowing them to float. In this case, the monetary authorities would be able to follow an independent monetary policy. The equation system would be as follows X(px,py,M) + X*(px*9 py*9 M*) = 0 Y(pX9py9 M) + Y*(px*9py*9 M*) = 0 L(px,py,M) = 0; L*(px*9py*9M*) = 0 px = epx*l(l+s)(l+r*) py = (l+r)(l+e*)epy* M=D + eR

where M*° indicates that the monetary policy in the rest of the world is autonomous. In this framework, under the assumption that the money supply in the rest of the world is given, we can develop the more general analysis of tariffs and export taxes that takes into account changes in the terms of trade.

The theory of tariffs and monetary policies

259

L*

Figure 11.12 Determination of the terms of trade

In figure 11.12, Q and W represent the autarkic equilibria of country A and ROW, with OQ and OW reflecting the pre-trade price levels. With the opening up of trade in goods (but not money), A will export commodity X and B will export commodity Y, lowering the relative prices of Y in country A and X in ROW. The terms of trade will settle at an equilibrium terms of trade indicated by the ray OVU. The equilibrium terms of trade will lie between the autarkic price ratios; other things equal, it will lie closer to the larger of the two trading blocs. In the cases discussed in previous sections, ROW was so large relative to country A that the terms of trade lines OW and OU coincided. But the larger country A becomes relative to ROW, the closer to OQ will the international terms of trade settle. The purchasingpower-parity equilibrium will be at an exchange rate, the value of ROW's currency relative to A's, equal to OV/OU. To analyze the effects of trade taxes, it will be sufficient to consider a tariff in country A; the conclusions for other taxes will follow, mutatis mutandis. As before, a tariff creates a wedge between home and foreign price ratios. At constant foreign prices, the tariff raises the price of imports, and shifts demand in country A from imports onto exportables and money. This reduces the demand for the rest of the world's exports and the supply of its imports, turning the terms of trade in favor of country A. In figure 11.13, assuming flexible exchange rates, a tariff equal to ST/JT results in a rise in the price of imports by less than the rate of the tariff; the "foreigner pays part of the tax." The exchange rate appreciates from OV/OU to OT/ON.

260

ROBERT MUNDELL

Figure 11.13 Tariffs and flexible rates

Under fixed exchange rates, a tariff also raises the price of imports in country A and shifts excess demand from commodity Y onto commodity X and A's currency, giving rise to a balance-of-payments surplus and an influx of money from the rest of the world (see figure 11.14). To determine the effect of the tariff in country A on prices in the rest of the world, it is necessary to take into account two effects, a terms-of-trade effect and a monetary effect. The terms-of-trade effect lowers the price of Y and raises the price of X in ROW. But the monetary effect involves a balanceof-payments deficit in ROW (the counterpart of country A's surplus), leading to an export of money and a fall in both prices. It is therefore clear that the money price of commodity Y falls in ROW, but the two effects work in opposite directions on the money price of commodity Y: its price might rise, fall, or remain unchanged as a result of the tariff in country A. A similar analysis applies to the effect of the expansion in money in country A. The price equilibria after the imposition of the tariff in country A are indicated by the (tariff-inclusive) point J in country A and the point F in the rest of the world (see figure 11.14). 6

Customs unions and discriminatory tariffs

Up until now we have considered ROW, the rest of the world, as a single trading entity. To consider the problem of discriminatory tariffs, we must have a model that includes at least three countries, one of which can again be taken as the rest of the world. It will be convenient first to specify

261

The theory of tariffs and monetary policies

L*1

L'

0

Px Figure 11.14 Tariffs and fixed exchange rates

the conditions of equilibrium for n countries and then to examine more interesting special cases. Commodity prices in the different countries are connected by the law of one price. The price of a good in country /is equal to the export prices of the cheapest producer, labeled countryy, after adjusting for tariffs and export taxes and multiplying the price in country^ by the exchange rate, the price of country j's currency in terms of country /. The monetary arrangements must now take into account the multiplicity of currencies, all of which, in principle, could be used for foreign exchange reserves. To avoid complications associated with private holdings of foreign, currencies, I shall assume that only central banks (or governments) hold reserves for the purpose of backing, in the case of fixed exchange rates, the domestic money supply. The equation system can be written as follows

//

(

)

(

Af' = V + Z

(>=1,2,..,») (i=\,2,..\n)

262

ROBERT MUNDELL

Figure 11.15 Three-country equilibrium

The final equation specifies foreign currency preferences for each central bank; these preferences are assumed to be autonomous. The formation of a free-trade area involves the elimination of tariffs between any two or more countries. It will be convenient, however, to proceed in a reverse direction, starting from free trade and analyzing the effects of one or two countries imposing a tariff that exempts its prospective partner. It will also be convenient to analyze the situation in a three-country model, one of which can again be taken to be the rest of the world. Figure 11.15 portrays the equilibrium for three countries, A, B, and C, under conditions of free trade: country A. exports commodity X to countries B and C and imports commodity Y from one or both of those countries. The equilibrium exchange rates are as follows eAB = OJ/OH; *AC = OJ/OK; eBC = OH/OK There are two types of free-trade areas that need to be distinguished. One involves an agreement between two rivals, countries producing substitute goods, such as a free-trade agreement between B and C; the other involves an agreement between two complementary countries, such as an agreement between either A and B or A and C. In the theory of customs unions, it is generally concluded that free-trade agreements between rival countries are more likely to replace inefficient industries in a rival country and therefore involve less trade diversion and

The theory of tariffs and monetary policies

263

more trade creation than is the case of free-trade areas between complementary countries. Our interest here, however, is not with the static welfare gains but with the new equilibrium terms of trade, exchange rates, and price levels after the free-trade area is formed. Consider first the effect of a tariff in country B imposed on country A's goods. The price of country A's goods rises in country B relative to those of country A. However, since no tariff is imposed on C's products, A's goods could enter B through C unless transport costs or regulations prohibited it. Certificates of origin would, however, be of little help if C also produces commodity Y; sales of commodity Y from country C to country B could be made up by increased imports in country C from country A. The unilateral imposition of a discriminatory tariff in B fails to alter the equilibrium because of the problem of trade deflection.2 Now consider a mutual and equal increase in tariffs in countries B and C on A's products as in a customs union; the problem of trade deflection is avoided in a customs union. This raises the price of commodity X in both B and C and creates an excess demand for commodity Y and money in each country, improving the terms of trade, and appreciating the currencies (under flexible exchange rates), of both B and C. In figure 11.16, the tariffs in B and C, equal to JK/HJ and NR/LN, turn the terms of trade against A from OV to OU. It depreciates A's currency from OE/OZ to OU/OJ against currency B, and from OE/OG to OU/ON against currency C. Under fixed exchange rates, increases in the money supplies of B and C would replace currency appreciation and the same change in the terms of trade would take place, but through a rise in the prices in the customs union area of both goods. If countries B and C hold reserves in A's currency, however, part of the inflationary effects will be mitigated by a fall in the price level in country A. Consider now a customs union between two complementary countries, such as A and B. The two member countries, from an initial free-trade position, levy a tariff against imports from country C, driving a wedge between the relative prices in A and B on the one hand and C on the other. In which direction do the terms of trade and other variables change? Figure 11.17 illustrates customs union between complementary countries. Because A and B are complementary, exporting different products, any change in the terms of trade means that the terms of trade of one of the member countries improves while that of the other worsens. Which of the member countries terms of trade will improve?3 It will be that country whose trade is complementary to the trade of the rest of the world. If the rest of the world is an exporter of commodity Y, as in figure 11.17, the country that exports commodity X will experience an

264

ROBERT MUNDELL

Figure 11.16 Customs union between competitive countries, B + C

0

Px Figure 11.17 Customs union between complementary countries, A + B

The theory of tariffs and monetary policies

265

improvement in its terms of trade, and the other country will suffer a worsening of its terms of trade. In figure 11.17, country B exports the same commodity as the rest of the world and will therefore experience a deterioration of its terms of trade. The terms of trade of country C deteriorate from the line OM to the line OL. Under flexible exchange rates, country A's currency appreciates from ON/OM to OK/OL, and country B's currency, from OR/OL to OJ/OL, against the currency of country C. It is not, in general possible to determine the effect of the currency union on the intraunion exchange rates; the price of A's currency in terms of the currency of country B changes from OR/ON to OJ/OK. What if the rest of the world contains many individual countries, some of which export commodity X and others, commodity Y? What matters is the //£/exports of the rest of the world. If the rest of the world, on balance, exports commodity X, the world price of X will tend to fall, and so will the terms of trade of the member of the customs union that exports commodity X. Whether the customs union is composed of complementary or competitive economies, it holds that the terms of trade of the rest of the world depreciate upon formation of the union. Those members of a union that export the same products as the rest of the world to the union will be harmed by the terms-of-trade effect, whereas those countries that are complementary to the rest of the world will be benefited by the terms-of-trade effect. Notes 1 The techniques used in this chapter were developed in Mundell (1964, 1968, and 1971); they have been applied in an intertemporal context in Mundell (1992). 2 This conclusion, of course, would not hold in the presence of transport costs or other transactions costs. 3 For general proofs of the effects of discriminatory tariff preferences in a real economy, see Mundell (1964). References Mundell, Robert A. (1964), "Tariff Preferences and the Terms of Trade," Manchester School of Economic and Social Studies, 32 (January), 1—14 (reprinted in Mundell, 1968). (1968), International Economics, New York: MacMillan. (1971), Monetary Theory: Interest, Inflation and Growth in the World Economy, Pacific Palisades: Goodyear. (1992), "Fiscal Policy in the Theory of International Trade," in Herbert Giersch (ed). Money, Trade and Competition (Essays in Memory of Egon Soh men), Heidelberg: Springer.

CHAPTER

The economics of content protection MICHAEL MUSSA

1

Introduction

This chapter analyzes economic effects of policies of content protection, such as the requirement that "domestic" automobiles sold in the United States to embody a prescribed minimum share of domestic value added. Similar policies have been implemented in a number of developing countries, as well as in Australia and Canada. Included in this broad class of policies are both requirements that final goods assembled in a country should use a minimum amount of domestic input, and requirements that final goods exported to a country should use a minimum amount of domestic input in their foreign assembly processes.1 To analyze the consequences of such policies of content protection, it is assumed that the final product is produced in accord with a neo-classical production function, specified in section 2, that employs domestic inputs and imported (or foreign) inputs. This specification allows for smooth substitution possibilities between domestic and imported inputs which, it is argued, characterize the situation of many industries that are the actual or potential subjects of content protection policies. This specification differs from that of Grossman (1981) who assumes that the imported material input is a perfect substitute for a domestically produced material input and that total material input (imported plus domestic) substitutes against domestic labor in the production of final output. 2 Using the present specification, it is shown that a domestic content requirement, specified as a required minimum share of domestic input in the value of the final product and enforced by a penalty tariff imposed against violators, raises the ratio of 266

The economics of content protection

267

domestic input to imported input and creates a production distortion by raising unit production cost above the cost-minimizing level. Because of the all-or-nothing nature of the enforcement penalty, however, content protection has the virtue of not inducing a distortion between the social cost of production (given the distortion of input choice) and the price charged to consumers of the final product. For this reason, content protection is preferable to a tariff on imported inputs or a subsidy on domestic inputs as a policy to increase the ratio of domestic to imported input in final output. Content protection has less salutory effects when account is taken, in section 3, of the effects on incentives for improvements in technical efficiency. The private and social benefit of technical improvements that save on domestic input is artificially reduced and the private and social benefit of technical improvements that save on imported inputs is artificially increased by a domestic content requirement. Taking account of the cost of finding and implementing improvements in technical efficiency, therefore, content protection will impede progress in making production processes that employ domestic inputs more efficient, and will lead to excess investment in technical improvements that will reduce required amounts of imported inputs. The effects of content protection on the equilibrium price and quantity of the domestic input are obviously of central concern to the suppliers of this input, who are frequently the most ardent advocates of content protection. In section 4, it is shown that under competitive market conditions, a small increase in the domestic content requirement above the level that firms would voluntarily choose will increase the derived demand for the domestic input and (assuming a positive but less than infinite elasticity of supply of the domestic input) will also increase the equilibrium quantity and price of the domestic input. The direct effect of further increases in the domestic content requirement above this marginally effective level are at least partially offset by reductions in demand for the final product resulting from increases in its price due to increased unit costs that are the consequence of content protection. However, provided that the price elasticity of demand for the final product is less than a critical value, the overall effect of an increase in the domestic content requirement will still be to increase the derived demand for the domestic input and, hence, the equilibrium quantity and price of that input. The condition for an increase in the content requirement to increase demand for the domestic input is modified in situations, examined in section 5, where only foreign firms (and not domestic firms) are effectively constrained by content protection. In such situations, the own price elasticity of demand for the foreign firms' product is likely to be larger than the overall price elasticity of demand for the products of the domestic and

268

MICHAEL MUSSA

the foreign firms because an increase in the price of the foreign firms' product shifts demand toward the domestic firms' product. Hence, the increase in the price of the foreign firms' product due to an increased domestic content requirement will be a more powerful force in reducing demand for the domestic input. However, since a shift of demand toward the domestic product increases the domestic firms' demands for the domestic input, the overall effect of an increase in the content requirement for foreign firms is still likely to be an increase in demand for the domestic input. The implications of non-competitive behavior in the final product and domestic factor markets are examined in section 6. Content protection does not alter the usual difference between monopolistic and competitive behavior by sellers of the final product or between monopsonistic and competitive behavior by buyers of the domestic input, unless content protection creates a monopoly or monopsony situation when one would not otherwise exist. Content protection, however, does interact in an interesting way with monopoly behavior of suppliers of the domestic input since it alters the elasticity of demand for this input in a way that can always be exploited by these suppliers. The main results of this analysis of content protection are summarized in section 7, and extensions and modifications of these results are briefly discussed. 2

Effects on input choice, production cost, and output price

The technology of the industry subject to content protection is assumed to be described by a neo-classical, linear homogeneous production function X=F(I,D)

or XjI=x=f(d)

= F(l,d\ with d=D/I

(1)

where X is the quantity of final output, I is the quantity of imported (or foreign) input, and D is the quantity of domestic input. This production function applies equally well to a final product assembled at home using domestic and imported inputs, and to a final product assembled abroad using domestic and foreign inputs and then exported back to the home country. The smooth shape of the isoquants of this production function illustrated in figure 12.1 reflect the assumption that domestic input can be substituted continuously, but with increasing difficulty, for imported (or foreign) input. The idea is that as the required share of domestic content rises, production of more components and more assembly processes must employ domestic inputs, starting first with the production activities in which domestic inputs are relatively most efficient and moving progressive-

The economics of content protection

269

D I=d

Figure 12.1 The effect of content protection on input choice and the iso-cost locus

ly to activities where these inputs are less and less efficient in comparison with imported (or foreign) inputs. This description applies fairly well to the automobile industry which is a favorite subject of such requirements in many countries. Arguably, this description also applies to a wide range of manufactured goods, from apparel to televisions, that are or might become subjects of content protection.3 Domestic nominal prices of final product, domestic input, and imported input are denoted by P, U> and V> respectively. Using the domestic input as numeraire, relative prices of the final output and the imported input are denoted by^> = P/Uand v=VIUy respectively. The zero profit condition for final goods producers requires that or p =

(2)

where aD = D/X is the amount of domestic input per unit of output and aj = I/X is the amount of imported input per unit of output. Using the fact

270

MICHAEL MUSSA

that aD = d\f{d) and al = \\f{d)^ it follows that the relative output price consistent with zero profits is given by p = {d+v)lf{d)

(3)

The cost-minimizing input ratio is determined by the requirement that the marginal technical rate of substitution equal the relative input price

(dFldI)l(dFldD)=f(J)-J-fy)

= (t*(J) = p

(4)

For example, if the input price ratio is v0, then as illustrated in figure 13.1, the cost-minimizing input ratio d0 is determined by the point A along the isoquant F(I, D) = \ at which the marginal technical rate of substitution is equal to v0. For any v, choice of d=(f)~\v) clearly results in the minimum relative output price, p(v) = [~\v) + v\ Iflfi ~1(v))> t n a t ls consistent with zero profits for that input price ratio. Domestic content requirements may sometimes be set in terms of physical units, but because of difficulties in comparing physical units of different inputs and outputs, it is more common for a domestic content requirement to be stated in terms of the share of domestic inputs in the value of final output. This share is given by <5 = aDl(aD + vaj) = d\{d+ v)

(5)

An effective domestic content requirement that sets a minimum value (5* for 5 implies that the input ratio must be d*(S*9p) = (p-d*)l(l-S*)

(6)

which is greater than the input ratio d=(f)~\v) that would minimize costs. The relative output price necessary to yield zero profits with such a content requirement />*(#•» = (d* + v)lf(d*) = [,7(1 - 6*)M(p • 5*1(1 - 3*)) 1

(

v

tnat

(7)

is greater than the price p = [(f>~ (v) + v]lf( t> ~\ )) yields zero profits when producers minimize cost. For example, if the domestic content requirement necessitates an input ratio d1=d*(v0, d*)>do = 4>~\vo), then as illustrated in figure 12.1, production occurs at the point B along the unit isoquant, rather than at the cost-minimizing point A. The relative output price that yields zero profits with this content requirement px =P*(VQ, 5*) = (d1 +^o)//(^i)> ls indicated by the vertical intercept of the line passing through B with slope equal to — P0. This price is clearly greater than the price p0 that yields zero profits under cost minimization, as indicated by the vertical intercept of the tangent to the unit isoquant (with slope equal to — v0) at the cost-minimizing production point A.

The economics of content protection

271

To enforce an effective domestic content requirement, there must be an incentive for compliance or a penalty for violation. When the final product is produced at home with imported inputs, it is natural for this incentive or penalty to be a tariff charged on imported inputs for producers violating the content requirement but rebated to producers complying with the requirement.4 The penalty tariff necessary to enforce the content requirement must make the minimum cost for violators at least as great as the cost for those who comply. For example, in the situation illustrated in figure 12.1, an input tariff Tl which makes the relative price of imported inputs for violators ^i = (1+^1)^0 *s sufficient to enforce the content requirement because the isocost line for violators with intercept px and slope vx passes below the unit isoquant. Producers who satisfy the content requirement, however, can produce at B and enjoy unit costs of only pv The content requirement and enforcement penalty make the isocost locus for unit cost of pi correspond to the kinked line that connects the pointsp x , B, E, and G. With this isocost locus, the optimum production point is clearly B. When the final good is produced abroad with exported domestic inputs, it is natural that the incentive for compliance be a penalty tariff (at an ad valorem rate /) in imports of the final product that do not satisfy the content requirement. Since \+t imported units of final product subject to this tariff generate the same revenue as one unit satisfying the content requirement, producers will satisfy the requirement when the cost of producing 1 + / units with a free choice of inputs is greater than the cost of producing 1 unit under the content requirement. In terms of figure 12.1, the penalty tariff rate tl9 for which the isoquant F(I, D) = \+t1 lies everywhere above the isocost line connecting pl9 B, E, and G, is sufficient to enforce the content requirement. It is an important property of content protection that the output price consistent with zero profits does not depend on the penalty used to enforce the content requirement, so long as the penalty is high enough to induce compliance. This property reflects the all-or-nothing nature of the penalty imposed for any violation of the content requirement. If the requirement is satisfied, no penalty at all must be paid and the zero profit price of a unit of output embodies only unit production cost and no penalty. Unit production cost, of course, exceeds the minimum achievable when producers are allowed free choice of inputs, but a content requirement generates no additional distortion by forcing a divergence between true social production cost (with the distorted choice of inputs) and the price producers must charge to earn zero profit.5 This property of content protection schemes accounts for their superiority over alternative policies for increasing the ratio of domestic input to imported input. Another policy that could achieve this same objective,

272

MICHAEL MUSSA

is an ad valorem tariff on imported inputs, with no rebates for producers. Specifically, to achieve the same input ratio dx as the content protection policy illustrated in figure 12.1, a tariff on the imported input would have to raise the domestic relative price of this input to the level v2 determined by the slope of the unit isoquant at the point B. Since this tariff is collected, producers must charge a pricep 2 , determined by the vertical intercept of the tangent to the unit isoquant at B, in order to earn zero profit. Sincep± is the true social cost of producing a unit of output using the input combination at B, the difference p2 —p\ measures an excess of price charged to consumers over true social production cost and implies a consumption distortion loss in excess of the distortion loss from content protection.6 Alternatively, the domestic input ratio could be raised to d1 by paying a subsidy on the use of domestic inputs that raises the relative price of imported inputs to v2. Since this subsidy reduces the output price for consumers (at the expense of the government) to below true social production cost, this subsidy creates consumption distortion loss in excess of the distortion loss from content protection. Because a content requirement avoids the excess consumption distortion loss generated by a tariff on imported inputs or a subsidy on domestic inputs, it is the second-best policy for achieving a prescribed increase in the ratio of domestic input to imported input. The question remains, of course, why society should tolerate the production inefficiency that inevitably results from content protection in order to raise the ratio of domestic to imported input. 3

Incentives for improvements in technical efficiency

The conclusion concerning the second-best optimality of a content protection policy requires qualification in situations where firms incur costs to maintain and improve the efficiency of their production processes. While analysis of such situations is not fully compatible with assumptions of perfect competition and a common constant-returns-to-scale technology for all firms, we can use the model of the preceding section to indicate the distortion of incentives for improvements in technical efficiency created by content protection. To this end, consider the reduction in unit production cost resulting from improvements in technical efficiency for a firm initially operating under the content protection policy described by the point B in figure 12.1, which corresponds to the points B in figures 12.2 and 12.3. An improvement in technical efficiency that allows for a small reduction AD < 0 in the amount of domestic unit used to produce a unit of output shifts the unit isoquant downward in figure 12.2 to the isoquant passing through the point G. To satisfy the domestic content requirement with the new technology, the firm must move up the unit isoquant to the point H that lies

The economics of content protection

273

D

Figure 12.2 The effects of a reduction in domestic input necessary to produce a unit of output

along the ray where DjI = d1. Substitution of domestic for imported inputs in the move from G to H occurs at the rate v2 which is equal to the marginal technical rate of substitution at G and at B. The reduction in unit production cost made possible by the improvement in technical efficiency is measured in units of domestic input by Ap =p$ —p\, where px and^>3 are the vertical intercepts of the lines passing through B and H, respectively, with slopes of —v0. To the first order of approximation, the reduction in unit production cost per unit reduction in required domestic input is given by

This result indicates the benefit to the firm and to society from an improvement in technical efficiency that saves domestic input, given that the content protection policy is in force. However, the benefit to society from this improvement in technical efficiency in the absence of the content protection policy is greater than the amount indicated by (8) because saving

274

MICHAEL MUSSA D

DII=d

Figure 12.3 The effects of a reduction in imported input necessary to produce a unit of output

one unit of domestic input in the absence of this policy would be worth exactly one unit of domestic input. Since (v2 — Vo)l( v2 + ^1) m u s t be less than 1, the distortion created by content production cannot eliminate all of the benefit from improvements in technical efficiency that save domestic input. But, if the difference between v2 and v0 is large, the distortion could substantially reduce this benefit. For improvements in technical efficiency that save imported input, the distortion created by content protection works in the opposite direction. As illustrated in figure 12.3, saving a small amount A/ of imported input shifts the unit isoquant to the left to the new isoquant passing through the point J. At J, the firm can substitute increased imported input for reduced domestic input (at a rate equal to the marginal technical rate of substitution v 2) and still satisfy the domestic content requirement by producing at the point K. To the first order of approximation, the reduction in unit production cost resulting from this increase in technical efficiency is

The economics of content protection

Vo + ^ - ^ - y j ^

+ d,)]

275

(9)

In this result, the amount v0 represents the social and private benefit of saving a unit of imported input in the absence of content protection. The additional amount (t;2~vo)' [^1/(^2 + ^1)] > 0 ls t n e distortionary effect of content protection that arises because content protection forces a differential between the marginal value of imported input in production (measured by the marginal technical rate of substitution v2) an<^ t n e price v0 of this input. Content protection does not create any distortion for a neutral improvement in technical efficiency that saves domestic and imported inputs in the same proportion as these inputs are initially used in producing a unit of output. Under content protection, however, economic incentives operate against neutral improvements in technical efficiency and in favor of improvements that save on imported input rather than domestic input. Diminished incentives for improvements in technical efficiency that save on domestic inputs provide a serious argument against use of content protection to provide temporary protection for infant industries or mature industries that need to regain international competitiveness since achievement of world standards of efficiency is likely to depend on efficiency in the use of domestic inputs. 4

Price and quantity of domestic input

To assess these effects of content protection on the equilibrium price and quantity of the domestic input, assume that demand for the final product is a declining function, H(P)y of its absolute price, with an elasticity of demand Y\ = {PjH)'H\P). The imported input is assumed to be an infinitely elastic supply at a given level of its absolute price, V. Supply of the domestic input is assumed to be an increasing function, S(U), of its absolute price, with an elasticity of supply e = (UIS)*S'(U). When competition prevails in all markets, derived demand for the domestic input is determined by multiplying X=H(P) by DjX=aD = djf{d). Using the fact that PU-[(d+v)lf(d)], it follows that D = (dlf{d)) • H[(U-d/f(d)) + (Vjf{d))]

(10)

Using a hat " " to denote a proportionate change, the proportionate responses of D, to proportionate changes in d, 17, and K, are given by 5-y)]'2

(11) (12)

-0

(13)

276

MICHAEL MUSSA

As before 8 = d\{d-\- v) is the share of domestic input in the cost of producing a unit of output. y = {djf(d))'f'(d) is the elasticity o£XjI=f{d) with respect to d=D/I. When content protection is not in force, the input ratio d is the cost-minimizing input ratio d=(j)~1(v), where <j>{d) = [f{d) — d*f\d)\l f'(d). Substituting (f)(d) for v, it is easily shown that in the absence of content protection, the share of domestic input, S = dl(d+v), is equal to y 5 = d\{d+ v) = df{d+ 4>(d)) = d-f(d)lf(

J) = y, when v = (d) (14)

Further, since in the absence of content protection d= (f) ~ \v), it follows that the proportionate change in the input ratio is given by 2=a.(0-V)

(15)

where a — (1 — y) • \f{d)\d*f"{dj\ < 0 is the elasticity of substitution between domestic and imported inputs. This result together with (11), (12), and (13) implies that D=[ti5 + (l-y)-<j]-U

(16)

D=[(l-(5).>/-(l-7).<x].F

(17)

Equation (17) reveals the condition for a tariff on the imported input, which induces a positive K, to increase demand for the domestic input: the elasticity of substitution between the domestic and the imported input must be greater in absolute value than the price elasticity of demand for the final output. Otherwise, the negative effect on the derived demand for D from the increase in the output price caused by the increased cost of the imported input will outweigh the positive substitution effect of the increase in the price of the imported input relative to the domestic input. It follows that suppliers of the domestic input will benefit from a tariff imposed on the imported input only when the elasticity of substitution between domestic and imported inputs is greater (in absolute value) than the price elasticity of demand for the final product. Under content protection, the input ratio is no longer d=(j)~1(v), but rather ^==*/*((5*,J>)==(Z>*<5*)/(1 — <5*). With a fixed content requirement, therefore d=i=V-0

(18)

Further, the proportionate response of d to a proportionate change in the content requirement is
(19)

These results together with (11), (12), and (13), determine the proportionate

The economics of content protection

277

response of demand for the domestic input to proportionate changes in U, V, and 5* when content protection is in force

*)]-tJ = r . t ;

(20)

(y-(5*)].K=A.F

(21)

d*)]-5* = Q.5*

(22)

In all of these results, the difference between y — d'f\d)jf{d) and S* is important, y measures the share of domestic input when the price of the imported input equals the marginal technical rate of substitution; whereas <5* is the minimum required share of domestic input when imported input is valued at its market price. Without a content requirement, y = 6. When the content requirement is more than marginally effective, however, the marginal technical rate of substitution is greater than the market relative price of the imported input and hence <5* is > y. It follows that starting with a content requirement that is on the margin of effectiveness, a small proportionate increase in <5* causes an equal proportionate increase in demand for the domestic input;7 that is D = [(1 - y)/(l - d*)] • S* = (5*

\ \ v )

+ v)

when

(23)

The reason for this equal proportionate effect is that with only a marginally effective content requirement, a small increase in (5* has only a second order or small effect on unit cost and output price. The direct first-order effect of an increase in (5* on demand for the domestic input, therefore, is not offset by any first-order indirect effect of a reduction in final output demand due to an increase in the final output price. When the content requirement is more than marginally effective, an increase in 8* causes a first-order increase in the final output price to the extent of P = [(<5* — y)/(l — <5*)]«<5*. This increase in P at least partially offsets the direct effect on the demand for the domestic input of the increase in <5*. When content protection is sufficiently stringent to produce a large difference between <5* and y, it is possible that further increases in <$* will reduce demand for the domestic input. For this not to happen, the price elasticity of demand for the final product must not be too large; specifically, for Q defined in (22) to be positive, it is necessary that \t,\<(l-y)l(8*-y)

(24)

Since <5*^1, it is clear that this condition is always satisfied if \r\\ < 1 . The equilibrium responses of D and U to changes in <5* depend on the elasticity of supply of the domestic input. Specifically, equating D = 8 • U to

278

MICHAEL MUSSA

D = r • t / + Q • S*, it follows that the equilibrium responses of U and D to a change in the content requirement are given by £/=[l/(e-r)].Q.(5*

(25)

£=[fi/(e-r)].Q.<5*

(26)

Since e is >0 and F is <0, an increase in (5* will increase both the equilibrium price and the equilibrium quantity of the domestic input if and only if Q is positive for which the necessary and sufficient condition is given by (24). In summary, when competition prevails in all markets, imposition of a marginally effective content requirement always protects suppliers of the domestic input by raising demand for this input thereby increasing its equilibrium price and the quantity employed in the industry subject to the content requirement. In contrast, a tariff imposed on the imported input benefits suppliers of the domestic input only when the elasticity of substitution between domestic and imported inputs is greater (in absolute value) than the price elasticity of demand for thefinalproduct. Increases in the required share of domestic content above the marginally effective level will further increase the equilibrium price and quantity of the domestic input when the price elasticity of demand for the final product is less (in absolute value) than the critical value (1 —y)l(S* — y). For moderate price elasticities of demand for the final product, this condition can be violated only when a very stringent content requirement forces a substantial divergence between <5* and y. Hence, over a fairly broad range, it is reasonable to expect that increases in the required share of domestic content will provide increased protection to suppliers of the domestic input. 5

Differences between domestic and foreign producers

In some situations where content protection is used or contemplated, such as the US auto industry, domestic firms typically use a much higher ratio of domestic to imported input than foreignfirmswho export to the domestic market. To deal with this case, it is convenient to assume that foreign firms use both domestic and imported (or foreign) inputs to produce their output X, in accord with the production technology described in section 2, while domesticfirmsproduce their output, Y, using only a fixed amount, b, of domestic input per unit of output. Outputs of domestic and foreign firms are assumed to be good but imperfect substitutes. The demand for foreign output is X=H(P, R), where P is the absolute price of the foreign product and R is the absolute price of the domestic product, CO = (P/H) • (dHjdP) < 0 is the elasticity of demand for

The economics of content protection

279

the foreign product with respect to its own price, and ^ = dR) > 0 is the elasticity of demand for the foreign product with respect to the absolute price of the domestic product. The demand for domestic output is Y=N(R,P), where v = (R/N)-(dN/dR)<0 is the elasticity of demand for the domestic product with respect to its own price, and \i = (P/ N) • (dN/dP) > 0 is the elasticity of demand for domestic product with respect to the price of the foreign product. Imported (or foreign) input is assumed to be in infinitely elastic supply at a given value of its absolute price, V, Supply of the domestic input is S(U), with an elasticity of s = (UIS(U))-S'(U)>0. Given these assumptions, the derived demand for the domestic input may be expressed as D = (d/f(d)) • H(PfR) + b • N(R,P)

(27)

where d is the ratio of domestic to imported input used by foreign producers, P(LNrf//(
+ C-R] + (\-<x)-[v-R + ii-P]

(28)

where OL = {djf{d))'H{P)R)jD is the fraction of total demand for domestic input accounted for by foreign firms; where, as before, y = d*f'(d)lf(d); and where d)-V R=U

(29) (30)

Using these results and (19), we may derive the expression for the proportionate change in the derived demand for the domestic input (at constant values of U and V) in response to a proportionate change in the domestic content requirement a). i u]}.5* (31) It follows that the condition for an increase in the domestic content requirement to increase demand for the domestic input is \
(32)

Since the right-hand side of (32), which is identical to the right-hand side of (24), has a value of infinity when ^* = 7, it follows that imposition of a marginally effective content requirement for foreign firms necessarily increases demand for the domestic input. As the content requirement rises above the marginally effective level, further increases in (5* diminish foreign firms' demand for domestic input by raising the price and decreasing the

280

MICHAEL MUSSA

demand for the output of these firms. This effect is captured by the term \o\ on the right-hand side of (32). This effect, however, is at least partially offset by the increased demand of domestic firms for domestic input resulting from the shift of final product demand toward domestic firms due to the increased price of the output of foreign firms. This effect is captured by the term — (1 — a)«/x/a on the right-hand side of (32). It follows that if the reduction in demand for the foreign firms' product due to a price increase is largely a shift of demand toward the product of domestic firms (which use a higher ratio of domestic input than foreign firms), then an increase in the required share of domestic content for foreign firms is likely to increase demand for the domestic input and hence benefit suppliers of this input. Suppliers of the domestic input can lose at the margin from an increase in the domestic content requirement only when the required share of domestic content already significantly exceeds the share foreign firms would voluntarily choose and when an increase in the price of the foreign firms' product significantly diminishes total demand for the product of foreign and domestic firms. 6

Non-competitive behavior

Returning to the simpler case where all firms produce the same product using the same technology, it is interesting to consider the effects of content protection when markets for the final product or the domestic input are not competitive. We continue with the assumption that the imported input is competitively supplied at a constant absolute price. Only minor modifications of the preceding analysis are required to deal with monopoly power in the final product market, if this power was already being exercised prior to the introduction of content protection. For a given demand function, X=H(P)y there is a marginal revenue function, X—J(M), which expresses the relationship between output sold, X, and the monopolist's marginal revenue, M.8 The monopolist equates marginal revenue to unit production cost, implying that derived demand for the domestic input is determined by D = (
(33)

This derived demand relationship is identical to the derived demand relationship for the competitive case, except that the marginal revenue function / has replaced the demand function H. All results for the competitive case, with both free input choice and with content protection, remain valid, provided that the elasticity of demand for the final product, t] = (PIH(P))'H\P); is everywhere replaced by the elasticity of the marginal revenue function, £ =

The economics of content protection

281

Of course, a monopolist will generally sell a smaller amount of final product and use a smaller amount of domestic input than would be the case under competition. Introduction of content protection, however, need not affect these differences between monopoly and competition in the final product market because content protection directly affects the cost conditions of final goods producers, not the conditions of demand for their output. Monopsony in the domestic input market can be handled in much the same way as monopoly in the final product market. The supply function for the domestic input, S(U)9 implies a marginal factor cost function, Q(W), which specifies the relationship between the quantity of this input and its marginal cost, W, for a purchaser who recognizes that supply price rises with the number of units purchased. The analysis of section 4 goes through when the price of domestic input, U, is everywhere replaced by its marginal factor cost, HF, and when the elasticity of the supply function, £ = (17/ S(U))* S'(U), is everywhere replaced by the elasticity of the marginal factor cost function, k = (WjQ(W))'£y{W). As with the case of monopoly in the final product market, monopsony in the domestic input market will generally reduce the equilibrium price and quantity of the domestic input to below the levels that would prevail under competition. However, since content protection does not directly affect the conditions of supply of the domestic input, it need not affect, in an important way, these differences between the monopsonistic and the competitive equilibrium in the market for the domestic input. Content protection interacts in an important way with monopoly in the final product market and monopsony in the domestic input market if it creates monopoly and monopsony power in these markets when it would not otherwise have existed. This might happen, for example, if a single domestic firm which used a high ratio of domestic to imported input originally competed with foreign firms that found it efficient to use lower ratios of domestic to imported input. A content protection requirement that forced the foreign firms to use much higher ratios of domestic input to imported input than they would otherwise choose might remove these firms as competitors both in the market for the final product and in the market for the domestic input, thereby allowing the domestic firm to exercise both monopoly power in the final product market and monopsony power in the market for the domestic input. Exercise of monopoly power by suppliers of the domestic input against demanders of this input is almost always affected by content protection because content protection directly affects the elasticity as well as the level of derived demand for the domestic input, as illustrated in figures 12.4 and 12.5. In these figures, the curve labeled D(U) shows the derived demand for

282

MICHAEL MUSSA

Marginal cost

Figure 12.4 Marginally effective content protection when the supplier of the domestic input acts as a monopolist and |<5| is > 1

the domestic input when producers exercise free choice of input combination. The curve labeled D*(U) shows the derived demand for domestic input when the content requirement d ^ <5* is imposed. It is assumed that this content requirement is only marginally effective, in the sense that (5* is equal to the share of domestic input that producers voluntarily choose in the equilibrium that prevails before content protection, with suppliers of the domestic input optimally exploiting their monopoly power. In figure 12.4, it is assumed that the elasticity of substitution between domestic and foreign inputs, (7, is greater than one in absolute value. This implies that the share of the domestic input grows as we move down along the D(U) curve. The D*(U) curve therefore corresponds to the D(U) curve below the initial equilibrium price and lies to the right of this demand curve above the initial equilibrium price. In contrast, in figure 12.5, it is assumed that \a\ is < 1 . This implies that the share of domestic input declines as we move down along the D(U) curve. The D*(U) curve therefore corresponds to the D(U)

283

The economics of content protection

Marginal cost

D*(U',8*)

Figure 12.5 Marginally effective content protection when the supplier of the domestic input acts as a monopolist and \d\ is > 1

curve above the initial equilibrium price, and lies to the right of this demand curve below the initial equilibrium price. Constructing the marginal revenue curves that are appropriate with and without marginally effective content protection, it is apparent from figure 12.4 that when | < T | > 1 , marginally effective content protection raises the monopoly price and lowers the monopoly quantity of the domestic input. In contrast, as illustrated in figure 12.5, when \o\ is < 1 , marginally effective content protection lowers the monopoly price and raises the monopoly quantity of the domestic input.9 Despite the differences in the direction of change in the monopoly equilibrium price and quantity of the domestic input, depending on the elasticity of substitution, the monopolistic supplier of the domestic input always gains from the introduction of marginally effective content protection. The monopolist clearly cannot lose because the initial monopoly equilibrium point remains available. The monopolist actually gains

284

MICHAEL MUSSA

from marginally effective content protection by moving away from this initial equilibrium point in whichever direction allows him to take advantage of marginally effective content protection. Increases in the domestic content requirement above the marginally effective level are clearly beneficial to the monopolistic supplier of the domestic input so long as they increase the demand for this input at the price that the monopolist is already charging. 7

Conclusions and extensions

The main conclusions of the preceding analysis may be summarized as follows. First, content protection creates a production distortion by forcing an inefficient choice of domestic and imported (or foreign) inputs, but it does not create an excess consumption distortion by forcing a divergence between social production cost (given the inefficiency in input choice) and output price. Second, content protection biases investments in improvements in technical efficiency away from improvements that save domestic input and toward improvements that save imported (or foreign) input. Third, under competitive conditions, introduction of a marginally effective content requirement increases demand for the domestic input and benefits suppliers of this input. Fourth, increases in the domestic content requirement above the marginally effective level further increase demand for the domestic input and further benefit suppliers of this input provided that the price elasticity of demand for the final product is smaller (in absolute value) than a critical value that varies inversely with the stringency of the content requirement. Fifth, the consequences of monopoly in the final product market and monopsony in the domestic input market are not materially affected by content protection unless such a policy creates a monopoly or monopsony situation when one would not otherwise have existed. Sixth, content protection does in general enhance the monopoly power of suppliers of the domestic input. Among these conclusions, those that deal with effects on the suppliers of domestic inputs are sensitive to the assumption of a single domestic input and hence a unity of interests among the suppliers of this input. Suppose instead that the domestic input D is produced by two more basic inputs, JL and K, in accord with a neo-classical, linear homogeneous production function, D = G(LD, KD). Suppose further that these two inputs are mobile between industries, are fixed in aggregate supply, and are used in the neo-classical, linear homogeneous production function Z = H(LZ, Kz) to produce the economy's other genie output, Z, whose absolute price is denoted by T. Under these assumptions, the economy may be thought of as producing two products, D and Z, with absolute prices U and T, using the

The economics of content protection

285

technology of the standard, Heckscher-Ohlin-Samuelson model.10 Content protection which increases the demand for D at given values of all prices will increase the equilibrium relative product price UjT. Hence, the combination of L and K used in D (which earns U for each unit of D produced) will gain relative to the combination of L and K used in Z (which earns T per unit of Z produced). When we look at individual factors, however, the Stolper-Samuelson theorem tells us that the reward to the factor used intensively in D will rise relative to U and T, while the reward to the other factor declines in terms of both products. With this production structure, therefore, it is the factor used intensively in D that is the beneficiary of content protection. Modifying the assumptions of the preceding paragraph, suppose that only the factor L is mobile between D and Z and that K is specific to the industry where it is employed. The production structure for D and Z is now described by the specific capital or Viner-Ricardo model.11 Content protection which raises the demand for D at given prices will, as in Heckscher-Ohlin-Samuelson structure of the preceding paragraph, increase the equilibrium relative product price UjT. The combination of L and K used in D will gain relative to the combination of L and K used in Z. With respect to individual factors, the reward to KD will rise relative to U and T, the reward to mobile L will rise relative to T but fall relative to U, and the reward to Kz will fall relative to T and U. Thus, K specific to D will be the clear winner from content protection, with K specific to Z the clear loser, and mobile L, somewhat in between. Modifying further the technology for production of X to allow for different substitution relations between imported inputs and particular domestic inputs, the production function X=F(I, D) might be written as X=J(I, LD,KD). With this technology for X production, the effect of content protection on demands for LD and KD at given prices depends on the nature and strength of complementarity and substitution relations among the factors cooperating to produce X. If, for example, LD was a strong complement for I(JIL > 0) while KD was a strong substitute for I and L (JIK < 0 and JLK < 0), then content protection might increase demand for KD while reducing demand for LD. Equilibrium effects on rewards to individual domestic factors would depend additionally on the specificity or mobility of LD and KD and the intensity with which these factors are used in other domestic production activities. Rather than enumerating the possible outcomes, suffice it to emphasize that the gains and losses of individual domestic factors from content protection depend on the details of the production structure of the economy. Finally, it is important to note two assumptions of the preceding analysis that are not always satisfied when content protection is an active issue. First,

286

MICHAEL MUSSA

foreign firms induced to establish domestic production facilities because of content protection may not pay the same price or face the same conditions in the market for domestic factors as domestic firms in their industry. In developing countries, foreign firms typically pay higher wage rates than domestic firms. It is argued that these higher wage rates attract a queue of unemployed workers who wait for a high wage job at a foreign firm rather than accepting a lower wage job at a domestic firm. If so, then content protection that increases the number of high wage jobs at foreign firms may have a significant efficiency cost in terms of higher unemployment. For the US auto industry, this situation is probably reversed: the effect of content protection on automobile prices, on wages of auto workers and on profits of US auto firms depends on whether foreign firms induced to locate production facilities in the US accept the high wage rates and work rules currently prevailing at US firms. Second, constant returns to scale is not an accurate assumption for automobile industries in developing countries that have frequently been subjects of content protection and may not be an accurate assumption for other industries in these countries or for industries in developed countries that are actual or potential subjects of such policies. Reasonable efficiency in final assembly and in manufacture of some automobile parts and components may be achievable at a scale commensurate with final sales (and desirable product diversity) in the markets of the largest developing countries. Efficient manufacture of some automobile components (with allowance for desirable product diversity), however, apparently requires a scale achieved only in the largest automobile markets (North America, Europe, and Japan).12 Stringent domestic content requirements that force small-scale production of these automobile components necessarily have significant efficiency costs beyond the increase in unit production cost indicated by the analysis in section 2. The same point applies to content requirements in other industries where scale economies remain important for some components of the final product at scales beyond the size of the domestic market. Moreover, when scale economies are important, there is the danger that content protection will artificially create a "natural monopoly" in which a single, efficient domestic producer keeps out domestic rivals (who cannot reach efficient scale) and is freed of competition from foreign firms who do not meet the content requirement. Notes 1 The United States, for example, permits foreign assembled products normally subject to tariffs or other import restraints to enter the US with reduced duties if they embody a minimum specified value of US-produced components. 2 Analyses of content protection schemes are also provided by Corden (1971,

The economics of content protection

287

45—50), by Johnson (1972, 285—306) for the case of the Canadian auto industry before the Canada-US auto pact, and by Munk (1969) for automobile industries in several Latin American countries. 3 Grossman (1981) writes the production function as X= F(L, M+ M*) where L is a domestic primary input called "labor" and M and M* are domestic and imported material inputs that are perfect substitutes in producing the final product. Grossman specifies that a physical content requirement sets a maximum ratio of M* to M and that a value (or share) content requirement sets a maximum value of P* • MjP • X where P* is the price of the imported material input and P is the output price. While this formulation may be reasonable for some industries subject to content protection, I would argue that the production function (1) which allows a more flexible representation of substitution possibilities between domestic and imported inputs is more appropriate for a general analysis of content protection. Moreover, in most domestic content schemes, the cost of primary domestic inputs, as well as purchased domestic material inputs or components, is allowable in the calculation of domestic content. The present analysis of content protection is consistent with this practice rather than with Grossman's assumption that only domestic material inputs count as domestic content. 4 To protect domestic firms whose costs are raised by the content requirement from firms that produce entirely abroad it will generally be necessary either to prohibit imports of foreign goods that fail to meet the content requirement or charge a suitable tariff on such goods. In some cases, protective tariffs on the final product may already be in force before content protection is introduced. Then, as Corden (1971) notes, introduction of content protection will reduce effective protection for domestic producers of the final product by raising the price they must pay for purchased domestic inputs. 5 In his analysis of the content protection scheme for the Canadian auto industry, Johnson (1972) concludes, "The content requirement, when effective, forces the manufacturer to spend the increased profit he would obtain [from remission of duties on imported inputs], not on lower prices to the consumer, but on the excess cost of parts production in Canada." This conclusion is only partially correct. Some fraction of the remitted penalty tariff is effectively absorbed by the increase in unit production cost for firms that satisfy the domestic content requirement. However, provided that competition prevails, the part of the remitted penalty tariff that is not absorbed by this increased cost is passed through to purchasers of the final product. 6 There is some consumption distortion loss from content protection because the price to consumers is raised above the price that would prevail in the absence of such protection. There is not, however, any excess consumption distortion loss from content protection due to a divergence between true social domestic production cost and the price charged to consumers. 7 This result differs from Grossman's conclusions (see his propositions 3 and 5) that a "small" (marginally effective) content requirement could raise or lower domestic value added. The reason for the difference is in Grossman's

288

8 9

10 11 12

MICHAEL MUSSA

specification of technology and in his definition of the magnitude to which the domestic content requirement is applied. Specifically, Grossman includes only domestic material inputs in his definition of domestic content that satisfies the content requirement. A marginally effective content requirement in his model increases the quantity of the domestic material input, but may reduce value added from the primary domestic input sufficiently to reduce total domestic value added. For the purpose of this discussion, it is assumed that marginal revenue, M, is a well-defined, monotonically decreasing function of output, X. Grossman (proposition 6) concludes that imposition of a marginally effective content requirement will always induce a monopolistic supplier of the domestic material input to raise his price and reduce his quantity. This is because in Grossman's model the demand curve facing the supplier of the domestic material input is infinitely elastic at the world market price of the imported input before content protection is imposed. When a marginally effective domestic content requirement is imposed, the demand curve facing the domestic supplier of the material input has finite elasticity at and in a region to the left of the old equilibrium point without this requirement. At the old equilibrium, therefore, marginal revenue must be less than marginal cost when the marginally effective content requirement is imposed, implying that the monopolistic supplier of the domestic material input will increase his price and reduce his quantity. See Jones (1965) for a clear discussion and summary of the properties of this model. See Jones (1971), Mayer (1974), or Mussa (1974) for an exposition of the essential properties of the model. Munk (1969) cites evidence from the 1950s and 1960s that indicates differing degrees of scale economies for different processes in automobile manufacturing, with some meaningful economies for total vehicle production remaining at a scale of three or four hundred thousand vehicles per year. The shift toward production of "world cars" by leading manufacturers suggests that scale economies in automobiles today are at least as important as they were a decade or two decades ago. References

Corden, W.M. (1971), The Theory of Protection, Oxford: Oxford University Press. Grossman, Gene M. (1981), "The Theory of Domestic Content Protection and Content Preference," Quarterly journal of Economics, 96(4), November, 583603. Johnson, Harry G. (1972), Aspects of the Theory of Tariffs, Cambridge, Mass.: Harvard University Press. Jones, Ronald W. (1965), "The Structure of Simple General Equilibrium Models," Journal of Political Economy, 73(4), August, 557-72.

(1971), "The Three Factor Model in Theory, Trade and History," in Bhagwati, et al. (eds.), Trade, the Balance of Payments and Growth, Amsterdam: North-Holland.

The economics of content protection

289

Mayer, Wolfgang (1974), "Short-Run and Long-Run Equilibrium for a Small Open Economy," Journal of Political Economyp, 82(5), September, 955-67. Munk, Bernard (1969), "The Welfare Costs of Content Protection: The Automobile Industry in Latin America," Journal of Political Economy, 77(1), February, 85-98. Mussa, Michael (1974), "Tariffs and the Distribution of Income: The Importance of Factor Specificity, Substitutability and Intensity in the Short and Long Run," Journal of Political Economy, 82(6), November, 1191-203.

CHAPTER

The relationship between exchange-rate variability and protection JOSEPH ASCHHEIM, GEORGE S. TAVLAS, AND MICHAEL ULAN

1

Introduction

In the economics profession, opinion on whether exchange-rate variation encourages protection has undergone a substantial turnaround since the move to a managed-floating exchange-rate regime. Prior to the adoption of flexible rates in 1973, the foremost proponents of floating rates expressed the prognosis that floating rates would result in a diminution of pressures for protection. In contrast, some twenty years after the breakdown of the Bretton-Woods system, a number of writers have attributed rising protectionist pressures to the very system that was supposed to see the abatement of such pressures. For example, as Goldstein (1984, p. 42) has noted that, "if transitory or excessive movements in exchange rates are perceived as being a primary cause of the decline of certain export- or importcompeting industries, they can generate pressures for protectionism." What has actually happened over the past twenty years to the relationship between exchange-rate variability and protection? Have the foremost proponents of floating been vindicated, or has floating exacerbated pressures for protection? This chapter attempts to address these and other issues regarding the relationship between exchange-rate variability and protection in terms of trade performance. 1 This chapter is divided into five sections, including this introduction. The next section traces the changing nature of academic discourse on the subject of exchange-rate flexibility and protection. As will be shown, not all writers view the variations in exchange rates that have occurred since the adoption of floating rates as harmful to trade or conducive to protectionist pressures. Section 3 investigates whether recent years have seen a 290

Exchange-rate variability and protection

291

diminution in exchange-rate variation. Additionally, in view of the argument espoused by the early proponents of a floating-rate regime that such a regime would facilitate external-payments adjustment, the section also examines whether recent years have seen reductions in the currentaccount imbalances of the major industrial countries. Section 4 presents empirical estimates of the relationship between exchange-rate volatility and trade performance. Concluding comments are provided in section 5. 2

The exchange-rate system and protection

Prior to 1973, academic discussions made the case for, and paved the way to, a more flexible exchange-rate system.2 While a number of elements underpinned the case for a flexible exchange-rate system,3 several elements related directly to the issue of protection. In this regard, it was argued that a floating-rate system would provide stable exchange rates and would facilitate external adjustment. Both features - i.e., stable exchange rates and ability to facilitate external adjustment - would, it was argued, decrease trade barriers. In the following paragraphs, we consider each of these factors in turn. The early advocates of floating rates argued that a floating-rate regime would also be a stable regime. There were several strands to this argument. First, in line with the flow model of exchange-rate determination that was prevalent in the 1960s, exchange rates were perceived as moving to equilibrate the prices of traded goods - i.e., exchange rates were determined by purchasing power parity (PPP). Second, while it was acknowledged that deviations from PPP could occur, under floating rates the behavior of speculators would stabilize exchange rates. Thus Friedman (1953) argued that floating rates would lead to more stable rates than a fixed-rate regime because a floating-rate regime does away with one-way bets. Additionally, Friedman posited that speculators who destabilized an exchange rate would lose money and eventually would go out of business, leaving the stabilizing speculators to determine exchange rates (see MacDonald, 1988, p. 3). Moreover, as Goldstein (1984, p. 5) has noted, the early advocates of floating rates predicted that "the large variability experienced at the onset of floating exchange rates would fade away as traders and policy-makers learned how to operate under the new regime." Since the flow model of exchange-rate determination postulated that exchange rates adjusted to equilibrate the prices of tradable goods, the early advocates of floating rates also argued that flexible rates would provide a speedy and effective mechanism of external adjustment. Consequently, "a country's policy-makers should not have recourse to protectionist devices for balance of payments reasons" (MacDonald, 1988, p. 4). In contrast to

292

J. ASCHHEIM, G. S. TAVLAS, AND M. ULAN

the situation that was expected to prevail under floating, the early advocates of floating argued that, under the Bretton-Woods system, "balance of payments disequilibria were allowed to grow generating crises of confidence and, therefore, inducing large speculative flows of capital. When balance of payments deficits ceased to be sustainable, primarily because of foreign exchange reserve losses, countries . . . would resort . . . to the introduction of trade and other controls and, in the more extreme situations, to devaluations of their currencies" (Zis, 1989, p. 2). Since the adoption of a managed-floating regime in 1973, world financial markets have been characterized by large movements in nominal exchange rates. These movements have been considerably larger than the early advocates of floating had expected, and they have been accompanied by large swings in real exchange rates, reflecting the fact that nominal exchange-rate variations have not closely followed changes in relative prices of traded goods. Two schools of thought have emerged concerning the effects of the large movements in nominal and real exchange rates on protection. Advocates of managed exchange rates argue that movements in real exchange rates produce adverse real effects on the economy. As explained by De Grauwe (1989, p. 228), the so-called "misalignments" lead to a boom in the traded-goods sectors of countries whose currencies have become undervalued. Correspondingly, in the countries whose currencies have become overvalued, the traded-goods sector is squeezed, leading to losses of output and employment that are not easily absorbed in the short run by other sectors of the economy.4 Thus, "individuals hurt by these developments organize themselves to pass protectionist legislation. As a result, markets become more protected, so that international trade is negatively affected" (De Grauwe, 1989, p. 229).5 The other school of thought - that expressed by proponents of floating argues that the fact that exchange-rate movements have been larger than had been expected and have not been very predictable does not necessarily imply that they have been harmful. In this connection, advocates of floating rates rest their case, in part, on the demise of the flow model of exchange-rate determination. Thus, Frenkel (1983, p. 8) has argued that, "The central insight of the modern approach to the analysis of exchange rates is the notion that the exchange rate, being the relative price of two durable assets, can best be analyzed within a framework that is appropriate for the analysis of asset prices." In line with this approach, proponents of floating rates point out the following. First, since exchange rates are financial-asset prices, they are flexible and forward-looking, unlike many goods prices that are sticky and backward-looking, reflecting previous contractual agreements. Hence,

Exchange-rate variability and protection

293

Table 13.1. Average current-account imbalances relative to GNP/GDP cent)

1975-1990 1975-1982 1983-1990 /-ratio of difference

(per-

US

UK

Can.

Fr.

Ger.

It.

Jap.

1.47 0.41 2.53 6.44

1.38 1.08 1.67 1.01

1.63 1.71 1.55 -0.39

0.71 0.98 0.45 -2.16

2.10 0.98 3.21 4.22

1.05 1.55 0.64 -3.24

1.86 0.88 2.98 5.22

volatility is to be expected in an auction market, such as the exchange market, under floating rates because of the continuous arrival of new information. Second, the ability of flexible exchange rates to absorb shocks has eased quantity and price adjustments in goods and labor markets. Third, it is doubtful that the fixed exchange-rate system would have survived the shocks to the world environment after 1973, such as oil-price shocks, without the imposition of controls on capital movements and restrictions on trade. In sum, two schools of thought have emerged regarding the effects of floating (and variable) exchange rates on trade and protection. In what follows, we turn to the evidence on the extent of exchange-rate variations under floating, the effects of floating on the external imbalances of the larger industrial countries, and the linkages (if any) between exchange-rate variability and trade performance. 3

Some stylized facts

The period analyzed in this chapter runs from 1975 through 1990. In absolute value terms, relative to GNP, the annual average currentaccount imbalances of five of the G-7 nations - the United States, France, Germany, Italy, and Japan - differed significantly between the first and second halves (1975-82 and 1983-90, respectively) of the period. 6 Two of the five countries, France and Italy, saw their current-account imbalances shrink; the other three, the United States, Germany, and Japan, saw them grow. There were no significant changes in the average current-account balance to GDP ratios of the United Kingdom and Canada between 1975-82 and 1983-90 (see table 13.1). Some of the increases in imbalances noted in table 13.1 may stem from the removal of restrictions on exports of capital from some European countries and Japan during the late 1970s and early 1980s, permitting residents of these nations to invest abroad (e.g., in the United States) more easily. As noted, some of the early advocates of floating rates argued that exchange-rate variability would decline as traders and policy-makers gained

294

J. ASCHHEIM, G. S. TAVLAS, AND M. ULAN

experience with floating rates. In order to investigate this hypothesis, we examined the behavior of the exchange-rate-volatility variables used in this study to determine whether the average volatility differed between the first and second halves of the period under study. The average log of the eight-quarter moving standard deviation of the real effective exchange rate of the dollar rose significantly, indicating rising volatility as measured by the moving standard deviation, between the first and second halves of the period.7 The average logs of the eight-quarter moving standard deviations of the real effective exchange rates of the franc, DM, and lira fell significantly between the first and second halves of the period, indicating falling volatility as measured by the moving standard deviation. Turning to the logarithms of the absolute values of the quarterly percent change in real effective exchange rates, the analysis is somewhat more complicated. Since values of this variable beginning in 1973:Q2 are used in the first polynomial distributed lag that enters the regression analysis, it is appropriate to begin the first half of the analysis in that quarter. Hence, the first half of the period studied runs from 1973:Q2 through 1981:Q4, and the second half from 1982:Q1 through 1990:Q2.8 Only in the case of the DM did the logarithm of the average absolute value of the quarter-to-quarter percentage change in the exchange rate differ significantly between the two periods. The volatility of the DM was lower during the second half of the period than during the first. Analysis of the period 1975:Q1 through 1990:Q29 revealed no statistically significant increases in the average volatility of any G-7 currency between 1975:Q1-1982:Q2 and 1982:Q2199O.Q2. To summarize briefly, the current-account imbalances of three of the G-7 nations - the United States, Germany, and Japan - increased significantly relative to GNP between the first and second halves of the period under study. Measured as the absolute value of the quarterly percent change in real effective exchange rate, none of the G-7 currencies exhibited significantly greater exchange-rate volatility during the second half of the period than during the first. In turn, measured as the moving standard deviation of the effective exchange rate, only the US dollar exhibited significantly greater volatility during the second half of the period. 4

Effects of exchange-rate variability on trade

4.1

Theoretical aspects

Exchange rates are an equilibrating mechanism. Nominal exchange rates change to bring the current account into line with net foreign investment. Real exchange rates change with the change in nominal rates

Exchange-rate variability and protection

295

and changes in relative prices at home and abroad. It is real rates that affect trade flows since they reflect relative prices of traded goods and domestic substitutes. In a recent paper with Bailey (Bailey, Tavlas, and Ulan, 1987), we reviewed arguments pertaining to the hypothesis that short-term volatility in exchange rates reduces trade. A priori, one might expect exchange-rate volatility to reduce trade since the possibility that exchange rates will change between the time that a sales contract is signed and payment is effected represents a risk to the party (buyer or seller) in whose currency the contract is not denominated, that is the party who must pay or receive foreign currency in the transaction. To a risk-averse trader, this risk represents a cost, a cost that can be offset only at a price in the forward market. The bid-ask spread may be low for short-term transactions, but it tends to increase with the volatility of the exchange rate. Nonetheless, small or not, an added cost tends to raise the prices of both exports and imports, thus tending to decrease the quantities of traded goods demanded. So goes an obvious argument about the effect of exchange-rate volatility on the volume of trade. There are, however, situations in which volatility of exchange rates could be expected to increase the volume of trade. If, for example, traders gain knowledge through trade enabling them to anticipate changes in exchange rates better than the average participant in the foreign-exchange market, they can profit from this knowledge. That profit may offset the risk represented by movements in the exchange rate. Perhaps more important in weighing the effect of exchange-rate volatility on trade is the specification of the alternative to volatility. If the movements in the foreign-exchange value of a currency reflect shifts in underlying fundamentals, the cost of maintaining stability may be intervention in the market by the monetary authority. In the long run, market intervention by the monetary authorities would be unsustainable. Maintaining the exchange rate would require the introduction of trade or capital controls or the adoption of monetary and fiscal policies that are not desired for domestic economic reasons. These controls, or the adoption of monetary and fiscal policies, could be more costly to both the domestic economy and trade than is exchange-rate volatility. 4.2

Exchange-rate movements and export performance

In recent years, a number of empirical studies dealing with the post-1973 period have been produced, examining the issue of whether short-term exchange-rate volatility hampers trade. A number of recent empirical studies have supported the proposition that short-term volatility does indeed impede trade (Abrams 1980; Cushman 1983, 1986, 1989; Coes

296

J. ASCHHEIM, G. S. TAVLAS, AND M. ULAN

1981; Akhtar and Hilton 1984; Kenen and Rodrik 1986; Maskus 1986; Thursby and Thursby 1987; De Grauwe and de Bellefroid 1987; De Grauwe 1988; Caballero and Corbo 1989; and Kumar and Dhawan 1991). The coverage of these studies has been impressive. It has encompassed both total and bilateral trade flows, differences in sampling data (i.e., time series and pooled time series cross sectional), bilateral and trade-weighted measures of exchange rates, real and nominal exchange rates, and a range of industrial countries. In turn, studies that have rejected the hypothesis that volatility has had an adverse impact on trade include the IMF (1984), Gotur (1985), Koray and Lastrapes (1989), Medhora (1990), and several papers with which we have been associated - Aschheim, Bailey, and Tavlas (1985); Bailey, Tavlas, and Ulan (1986); Aschheim, Bailey, and Tavlas (1987); and Bailey, Tavlas, and Ulan (1987). The most empirically comprehensive of our studies - Bailey, Tavlas, and Ulan (1987) - tested for the impact of exchange-rate volatility on real exports of eleven OECD countries, using for most countries two measures of volatility for both real and nominal exchange rates.10 In all, over the managed floating period, there were presented thirty-three regression equations. In addition to exchange-rate volatility, the factors that were posited to affect exports of these countries were real GNP in partner industrial countries, real export earnings of oil-producing countries, and relative prices (defined as the ratio of the dollar-denominated export unit values of each country relative to the dollar-denominated export unit values for the IMF's "industrial country" aggregate). Of the thirty-three regressions estimated, only three showed a significant and negative impact of volatility on exports. These three equations involved real volatility. So perhaps real volatility is the culprit. Considering only those equations with real exchange-rate volatility variables, that still left only three instances out of sixteen in which exchange-rate volatility negatively and significantly affected real exports. Despite the diversity of empirical results, some generalizations can be drawn from the current status of empirical work. First, most studies (including our work) that find a significant effect of volatility on tradefindit only for real exchange-rate volatility. But as our aforementioned results indicate, even in the case of real volatility, the evidence is anything but overwhelming. Second, of the studies that do find a negative effect of exchange-rate volatility on trade, most do so using bilateral trade data (e.g., Cushman 1983; Akhtar and Hilton 1984; Maskus 1986; and Thursby and Thursby 1987). Thus it may be that volatility affects the pattern of trade, but not its overall level. Indeed, Cushman (1986) argues that a purely bilateral approach to the analysis may result in specification problems since the volatilities of the bilateral exchange rates between a nation's currency and

Exchange-rate variability and protection

297

those of its trading partners may not be equal. One could obtain a spurious finding about the effect of exchange-rate volatility on trade in general if potential trade-diverting effects of differences in the volatilities of bilateral rates are ignored. Nonetheless, not all authors analyzing bilateral trade flows find a strong negative effect of exchange-rate volatility on trade. Thus Koray and Lastrapes (1989) find that the effect of exchange-rate volatility on US imports from the United Kingdom, France, Germany, Japan, and Canada is weak. Regarding the aggregate trade studies that find a negative impact of volatility on trade, Kenen and Rodrik (1986) examine the effects of exchange-rate volatility on imports - not exports. Still, in only four of the eleven countries examined did their results show a negative and significant impact. On the other hand, De Grauwe and de Bellefroid (1987) find less ambiguous effects of volatility on exports. However, their study does not include a relative price term. In their words, "The reader may wonder why no relative price (or competitiveness) variables appear in the equation. The reason is that we concentrate here on the determinants of the long-run growth rates of trade. . . . Over very long periods . . . these relative price effects are likely to have disappeared" (De Grauwe and de Bellefroid 1987, p. 195). The theoretical motivation behind this argument escapes us. At the very least, the effect of relative prices should have been empirically tested. Because the authors failed to test for this effect, it is likely that in the results obtained by De Grauwe and de Bellefroid, the effects of relative prices are comingled with those of exchange-rate volatility, with the result that the estimated impact of the latter is exaggerated or spurious. The final generalization to be drawn from empirical work is that the primary determinants of trade are real output in trading partner countries and the terms of trade. In a recent paper, Qian and Varangis (1992) examine the effect of exchange-rate volatility on trade volumes in different markets (primary commodities and manufactured goods) and under different trade regimes (free versus restricted trade). They depart from the approaches of other authors chiefly by modeling export volumes and prices simultaneously. They note that, like the effect of exchange-rate volatility on the volume of exports, the effect of volatility on the foreign currency denominated price of the export is also uncertain, a priori, since the volatility could induce exporters to increase or to decrease supply, courses of action that, given a foreign demand curve, would have opposite effects on the price of the export. Examining the aggregate exports of Australia, Canada, Japan, the Netherlands, Sweden, and the United Kingdom as well as Japanese exports to the United States, over the period 1974 to 1990, Qian and Varangis found that in only the equations relating to Japanese exports to the United States

298

J. ASCHHEIM, G. S. TAVLAS, AND M. ULAN

and Swedish exports to the world was exchange-rate volatility statistically significant. In the former equation, the coefficient was negative; in the latter, it was positive. 4.3

Some further results

In this chapter, real export and import equations are estimated for the Big Seven nations - the United States, the United Kingdom, France, (West) Germany, Italy, Japan, and Canada. Exports of country /are taken to be a function of real income in other industrial-country trading partners, relative prices, and exchange-rate volatility. For all countries except Canada, the relative-price variable is a real effective exchange-rate index for the exporting nation's currency. Because of the large proportion of Canada's exports that go to the United States, the ratio of Canada's export-unit-value index to the US producer price index is substituted for the real effective exchange rate of the Canadian dollar. Each nation's exports are modeled with two different exchange-rate volatility variables, the absolute value of the quarter-to-quarter percentage changes in the real effective exchange rate of the exporter's currency and the moving standard deviation of the currency's real effective exchange rate:

where: Xt = exports of country /, Yj = real GDP/GNP of trading-partner nations, Pt = relative prices, V. = volatility of the exchange rate of country /*s currency. Imports of country / are modeled as a function of the importing nation's real income, relative prices, and the volatility of the exchange rate of country /'s currency. Real income is represented by the importing nation's real GNP/GDP. As in the export equation, for each nation except Canada, relative prices are represented by the importing nation's real effective exchange-rate index. For Canada, the nation's import unit value index divided by the domestic producer (wholesale) price index is the proxy for relative prices. The same volatility variables used in the export equations are also employed in the import regressions: where: M( = imports of country / Yt = real GDP/GNP of the importing nation,

Exchange-rate variability and protection

299

Pf = relative prices, Vt = volatility of the exchange rate of country /'s currency. The equations are estimated using quarterly data. The data and the sources are described in the appendix. Generally, the variables VI, RP, and RXR enter the equations as eight-quarter, second-degree polynomial distributed lags (PDL's). In those equations where these PDL relative-price variables are not statistically significant, equations containing the price variables with a one-quarter lag were estimated. The equations containing the one-quarter lags on the price variable are designated with an asterisk (*). Except for those pertaining to France, all the equations are estimated over the period 1975:Q1 through 1990:Q2. Since French trade data for 1990:Q2 were not available, the equations pertaining to French trade cover the period 1975:Q1 through 199O:Q1. All of the variables enter the equations in logarithmic form; thus, all the parameter estimates represent elasticities. The empirical results are reported in tables 13.2 and 13.3. Of the twenty-eight equations estimated, only seven (three export equations and four import equations) show a statistically significant effect of exchange-rate volatility on trade volumes. Two of the significant volatility elasticities in the export equations are negative; one is positive. Three of the significant volatility elasticities in the import equations are negative; one is positive. In sum, only five of the equations contain significant negative volatility elasticities. All of the estimated elasticities are very small, however, indicating that the responsiveness of trade flows to changes in exchange-rate volatility is much smaller proportionately than the responsiveness to changes in relative prices or incomes. In absolute value, the largest statistically significant volatility elasticity is 0.05. The number of equations in which each of the volatility variables is significant is nearly equal (four equations containing the PDL of absolute percentage change of the exchange-rate index and three equations containing the moving standard deviation of the real exchange-rate index), but for no country except Germany are both of the volatility measures statistically significant in equations modeling either a nation's exports or imports. In addition, only for Germany does exchangerate volatility appear to depress both exports and imports. German exports are reduced four hundredths of a percent for each 1 percent increase in volatility, whether the volatility is represented as the moving standard deviation or the absolute percentage change of the real exchange rate of the DM. German imports are reduced three hundredths of a percent for each 1 percent increase in volatility when volatility is represented by the moving standard deviation; volatility is not quite a significant influence on German imports when it is modeled in absolute percentage-change form.

Table 13.2. Exports (t-ratws in parentheses) Country

K

Y

-4.60 (-5.41) -4.33 (-5.04)

RP

RXR

VI

1.05 (11.20) 1.02 (10.59)

-1.04 (-6.95) -1.05 (-6.64)

-0.02 (-0.82)

1.33 -7.64 (-23.50) (38.82) 1.37 -7.99 (-19.23) (29.15)

-0.24 (-4.85) -0.24 (-3.76)

0.02 (2.27)

1.32 (13.76) 1.36 (13.41)

-0.19 (-1.55) -0.20 (-1.69)

-0.02 (-1.23)

1.34 -8.08 (-29.57) (40.60) 1.37 -8.24 (-24.81) (34.01)

-0.49 (-3.40) -0.58 (-3.06)

-0.04 (-3.57)

1.48 -9.24 (-14.74) (20.21) 1.53 -9.54 (8.04) (-6.85)

-1.09 (-2.98) -1.24 (-6.63)

-0.02 (-0.98)

-7.79 (-1.33) -7.70 (-1.40)

1.38 (2.18) 1.39 (2.32)

-0.70 (-2.69) -0.78 (-3.30)

-0.008 (-0.34)

-9.29 (-8.61) -9.64 (-8.54)

1.51 (13.28) 1.57 (13.71)

V2

R2

DW

0.75 (8.94) 0.77 (9.67)

0.97

1.86

0.97

1.81

-0.30 (-2.42)

0.97

1.85

0.96

2.30

0.99

2.19

0.99

2.15

0.98

1.48

0.98

1.89

0.95

1.91

0.95

1.82

0.99

1.94

0.99

1.94

0.95

2.15

0.95

2.16

Pi

Pi

United States -0.001 (-0.05)

United Kingdom

France *

-7.63 (-8.91) -7.96 (-8.70)

0.02 (1.04) 0.84 (12.22) 0.86 -0.003 (-0.26) (13.43)

Germany -0.04 (-2.50)

0.29 (2.38)

Italy

Japan

Canada

-0.80 (-3.74) -0.65 (-3.09)

0.00001 (0.001)

0.02 (1.08) 0.03 -(1.25) 0.07 (1.80)

0.66 (5.01) 0.69 (5.13)

0.29 (2.28) 0.26 (2.01)

Table 13.3. Imports (t-ratios in parentheses) Country United States

United Kingdom * * France

Germany

Italy

K

Y

RP

RXR

VI

-12.50 2.04 (-14.61) (20.41) -12.06 2.01 (-12.30) (17.07)

0.31 (2.12) 0.29 (1.62)

-0.03 (-1.27)

10.54 2.31 (-18.62) (26.70) -10.13 2.30 (-15.91) (22.78)

0.17 (1.77) 0.04 (0.42)

-0.05 (-2.50)

1.67 -6.61 (-11.74) (18.75) -7.19 1.78 (-10.82) (17.23)

0.47 (2.22) 0.54 (2.10)

-0.03 (-2.33)

-7.91 1.83 (-26.75) (37.86) -7.92 1.82 (-30.29) (42.26)

0.80 (5.30) 0.74 (5.23)

-0.02 (-1.79)

-5.25 1.53 (-6.42) (10.98) -5.72 1.62 (-5.95) (9.84)

1.42 (2.17) 1.10 (1.58)

-0.009 (-0.25)

-2.26 (-2.37) -2.09 (-2.37)

0.74 (3.95) 0.85 (4.07)

0.01 (0.57)

R2

DW

0.98

2.17

0.02 (0.66)

0.59 (5.80) 0.68 (7.41)

0.98

2.28

0.97

1.75

0.03 (1.40)

0.38 (3.23) 0.45 (3.98)

0.97

1.80

0.99

2.03

0.002 (0.19)

0.64 (6.53) 0.71 (8.00)

0.99

2.08

0.99

1.98

-0.03 (-2.12)

0.29 (2.37) 0.26 (2.14)

0.99

2.01

0.43 (3.77)

0.94

2.17

0.94

2.17

0.98

1.92

-0.01 (-0.99)

0.85 (13.30) 0.83 (11.89)

0.98

1.92

0.97

2.07

-0.03 (-0.60)

0.51 (4.63) 0.62 (6.22)

0.97

2.18

V2

Pl

0.01 (4.05)

Japan

Canada

0.87 (6.96) 0.84 (7.27)

-7.69 2.02 (-13.55) (21.64) -7.69 2.03 (-10.28) (17.93)

-1.66 (-8.62) -1.61 (-6.74)

-0.05 (-1.44)

p2

304

J. ASCHHEIM, G. S. TAVLAS, AND M. ULAN

When defined in absolute percentage-change form, volatility of the exchange rate has a small negative effect on UK and French imports - but a small positive effect on UK exports. When defined in moving standarddeviation form, there is a very small positive relationship between volatility and Italian imports. Among the G-7 nations, the only country whose trade (exports and imports) appears to be depressed by increased volatility of the exchange rate of its currency is Germany - and the response of German exports and imports to rises in volatility is quite small. For the most part, our work shows an overall absence of a relationship between the short-term volatility of the exchange rates of their currencies and the aggregate trade of these seven countries. V

Summary and conclusions

Theoretical arguments concerning the relationship between exchange-rate volatility and protection are inconclusive. Some writers argue that volatile exchange rates lead to protectionist pressures which inhibit world trade. Other writers, however, argue that there is no clear theoretical reason for exchange-rate volatility to reduce trade. Ultimately, the issue is an empirical one. The empirical evidence presented in this chapter shows that, contrary to the views of the early proponents of floating, exchange-rate volatility and external-account imbalances do not appear to have diminished as market participants have learned how to operate under a floating-rate regime. Yet exchange-rate volatility does not appear to have reduced trade. Thus, our results suggest that exchange-rate volatility has not affected trade and, therefore, has not provided grounds for protection. Appendix The data The chief sources of the data were the data tapes of the International Monetary Fund's International Financial Statistics (IFS) and Directions of Trade (DOT). The Italian wholesale price index was not available beyond 1989:Q4 on the IFS tape; data for the first two quarters of 1990 are taken from the April 1990 and May 1991 issues of 1STAT Indieatori Mensili and linked to the series on the IFS tape. On the IFS tape, Canadian data on gross domestic product, import and export volumes and import and export unit values terminate in 1988:Q4; data for 1989 and the first two quarters of 1990 are taken from the May 1991 issue of the Bank of Canada Review and linked to the earlier data. The data used in this study are available from the authors.

Exchange-rate variability and protection

305

Definitions of variables

Dependent variable: Real exports or imports. Y: In export equations, the sum of the US dollar-denominated GNP/GDP's of nine major developed-nation trading partners, the United States, United Kingdom, Canada, France, Germany, Italy, Austria, Australia, Japan, and Sweden. National-currency data were converted to US dollars using 1988:Q4 exchange rates. In import equations, the US dollar-denominated GNP/GDP of the importing nation. RP: In export equations for Canada, since the United States accounts for a high proportion of the nation's exports, the ratio of Canada's US dollar-denominated export unit-value index to the US wholesale (producer) price index. In import equations for Canada, the ratio of the domestic currency-denominated import unit-value index to the domestic wholesale (producer) price index. RXR: The real effective exchange-rate index of the currency of the nation in question. The trading partners were those in the MorganGuaranty list of industrial countries. Except for France, which no longer publishes a wholesale (producer) price index, domestic wholesale-price indices were used to construct the bilateral real indices. For France, the consumer price index was substituted. The weights for each nation (x) in every other nation's (j) trade are the geometric means of the proportion of j ' s annual trade with (exports to plus imports from) each of the fourteen partners over the time 1973-89. To get around the fas/cif valuation difference, in calculating trade weights, imports of country j from country x were taken as country x's exports to country j . V\: The absolute value of the quarterly percentage change in the real effective exchange rate of the national currency. 1/2: The moving eight-quarter standard deviation of the real effective exchange-rate index of the national currency. All of the data series used to construct the variables were tested for seasonality using the Census X-ll program. The data exhibiting seasonality were seasonally adjusted. The data are available from the authors upon request. Notes The views expressed are those of the authors and should not be interpreted as those of the International Monetary Fund or the US Department of State. 1 In an earlier paper (see Aschheim, Bailey, and Tavlas, 1987) the relationship

306

2 3 4 5

6

7

8 9 10

J. ASCHHEIM, G. S. TAVLAS, AND M. ULAN

between exchange-rate variability and protection is assessed in terms of financial performance. Studies by Friedman (1953), Meade (1955), Sohmen (1961), and Johnson (1969) were particularly influential. See Goldstein (1984) for an overview of the issues involved. See Cline (1984) and Salvatore (1987) for empirical evidence on the determinants of protectionist pressures. However, as De Grauwe (1989, p. 229) also points out, "this hypothesis only makes sense if there is some asymmetry in the protectionist tendencies - for example, when protectionist legislation passed when the currency tended to be overvalued is kept in place when the currency is in the undervaluation part of the cycle." In fact, as pointed out by Aschheim, Bailey, and Tavlas (1985), the so-called "New Protectionism" in the United States emerged during a period of dollar depreciation in the 1970s. Since Italy and Japan had not published 1990 current-account data when this study was begun, the second half of the time period runs from 1983 through 1989 for these two countries. Goldstein (1984) shows that the average payments imbalances of the larger industrial countries were smaller during the first ten years of floating than during the last ten years of the adjustable-peg system. For the moving standard-deviation measure of volatility, the first half of the period runs from 1975:Q1 through 1982:Q2. The second half of the period runs from 1982:Q3 through 1990:Q2 for all currencies except the franc. Since French trade data for 1990:Q2 were unavailable, the second half of the period is defined as 1982:Q3 through 199O:Q1 for the franc. Once again, through 199O:Q1 for the franc. Again through 199O:Q1 for the franc. The countries examined are Australia, Canada, France, Germany, Italy, Japan, the Netherlands, New Zealand, Switzerland, the United Kindom, and the United States. References

Abrams, Richard K. (1980), "International Trade Flows Under Flexible Exchange Rates," Economic Review, Federal Reserve Bank of Kansas City, 65, 3—10. Akhtar, M.A. and R. Spence Hilton (1984), "Effects of Exchange Rate Uncertainty on German and U.S. Trade," Federal Reserve Bank of New York Quarterly Review, 9, September, 7—16. Aschheim, Joseph, Martin J. Bailey, and George S. Tavlas (1985), "Dollar Appreciation, Deficit Stimulation, and the New Protectionism," Journal of Policy Modeling, 1, 107-21.

(1987), "Dollar Variability, the New Protectionism, Trade and Financial Performance," in D. Salvatore (ed.), The New Protectionist Threat to World Welfare, New York: North-Holland, pp. 425^4. Bailey, Martin J., George S. Tavlas, and Michael Ulan, "Exchange Rate Variability and Trade Performance: Evidence for the Big Seven Industrial Countries,"

Exchange-rate variability and protection

307

Weltwirtschaftliches Archiv, 122, 466-77. (1987), "The Impact of Exchange Rate Volatility on Export Growth: Some Theoretical Considerations and Empirical Results," Journal of Policy Modeling, 9, 225-43. Caballero, Ricardo J., and Vittorio Corbo (1989), "The Effect of Real Exchange Rate Uncertainty on Exports: Empirical Evidence," The World Bank Economic Review, 3(2). Cline, William (1984), Exports of Manufactures from Developing Countries, Washington, DC: The Brookings Institute. Coes, Donald (1981), "The Crawling Peg and Exchange Rate Uncertainty," in John Williamson (ed.), Exchange Kate Rules: The Theory, Performance, and Prospects of the Crawling Peg, New York: St. Martin's, pp. 113-36. Cushman, David O. (1983), "The Effects of Real Exchange Rate Risk on International Trade," Journal of International Trade, 15, 45-63. (1986), "Has Exchange Risk Depressed International Trade? The Impact of Third-Country Exchange Risk," Journal of International Money and Finance, 5, 361-79. (1988), "U.S. Bilateral Trade Flows and Exchange Risk During the Floating Period," Journal of International Economics, 24, 317-30. de Grauwe, Paul (1988), "Exchange Rate Variability and the Slowdown in Growth of International Trade," IMF Staff Papers, 35, 63-84. (1989), International Money: Postwar Trends and Theories, Oxford: Clarendon Press, de Grauwe, Paul, and B. de Bellefroid (1987), "Long-run Exchange Rate Variability and International Trade," in S. Arndt and J.D. Richardson (eds.), RealFinancial Linkages among Open Economies. Cambridge, Mass: MIT Press. Frenkel, Jacob (1983), "Turbulence in Foreign Exchange Markets and Macroeconomic Policies," in D. Bigman and T. Taya (eds.), Exchange Rate and Trade Instability: Causes, Consequences and Remedies, Cambridge, Mass.: Ballinger. Friedman, Milton (1953), "The Case for Flexible Exchange Rates," in Essays in Positive Economics, Chicago: University of Chicago Press, pp. 157-203. Goldstein, Morris (1984), "The Exchange Rate System: Lessons of the Past and Options for the Future," IMF Occasional Paper No. 30, Washington, DC. Gotur, Padma (1985), "Effects of Exchange Rate Volatility on Trade," IMF Staff Papers, 32, 475-512. International Monetary Fund (1984), "Exchange Rate Volatility and World Trade," IMF Occasional Paper No. 28, Washington DC. (1984), "Exchange Rate Volatility and World Trade," IMF Occasional Paper No. 29, Washington DC. Johnson, Harry G. (1969), "The Case for Flexible Exchange Rates, 1969," Review, Federal Reserve Bank of Saint Louis, 51, 12-24. Kenen, Peter B., and Dani Rodrik (1986), "Measuring and Analyzing the Effects of Short-Term Volatility in Real Exchange Rates," The- Review of Economics and Statistics, 68, 311-15. Koray, Faik and William D. Lastrapes (1989), "Real Exchange Rate Volatility and U.S. Bilateral Trade: A VAR Approach," The Review of Economics and Statistics,

308

J. ASCHHEIM, G. S. TAVLAS, AND M. ULAN

71, 708-12. Kumar, R. and R. Dhawan (1991), "Exchange Rate Volatility and Pakistan's Exports to the Developed World, 1974-85," World Development, 19(9), 1225-40. MacDonald, Ronald (1988), Floating Exchange Kates: Theories and Evidence, London: Unwin Hyman. Maskus, Keith E. (1986), "Exchange Rate Risk and U.S. Trade: A Sectoral Analysis," Economic Review; Federal Reserve Bank of Kansas City, 71, 16-28. Meade, James E. (1955), "The Case for Variable Exchange Rates," Three Banks Review, 27, September, 3-27. Medhora, Rohinton (1990), "The Effect of Exchange Rate Volatility on Trade: The Case of the West African Monetary Union's Imports," World Development, 18(2), 313-24. Qian, Ying and Panos Varangis (1992), "Does Exchange Rate Volatility Hinder Export Growth? Additional Evidence," World Bank Working Paper. Salvatore, Dominick (1987), "Import Penetration, Exchange Rate, and Protectionism in the United States," Journal of Policy Modeling, 9, 125—42. Sohmen, E. (1961), Flexible Exchange Rates, Chicago: University of Chicago Press. Thursby, Jerry G. and Marie C. Thursby (1987), "Bilateral Trade Flows, the Linder Hypothesis and Exchange Rate Risk," The Review of Economics and Statistics, 69, 488-95. Zis, George (1989), "Is There Still a Case for Flexible Exchange Rates?" British Review of Economic Studies, 11, 1-20.

PART

IV

The new protectionism in the world economy

CHAPTER

Trade protectionism and welfare in the United States DOMINICK SALVATORE

During the past decade and as a result of the huge US trade deficits and increased globalization of the international economy, US trade policy has become much more controversial as it shifted in emphasis away from non-discriminatory multilateralism and toward aggressive bilateralism and protectionism. In this review, I will (1) briefly review the background for the increase in US protectionism in recent years, (2) summarize present US trade laws and their implications for protectionism, (3) examine current US trade policies and actual protectionism, (4) summarize the most notorious examples of increased US protectionism (that on textile and apparel, steel, and automobiles), (5) examine the political economy of protectionism, (6) discuss the relationship between the new protectionism and strategic trade policies in the United States, (7) evaluate the importance of the formation of three major trading blocs for the future of US protectionism. 1

Background of present US protectionism

Trade policy in the United States over the past thirty years has been marked by continuous tension. While the United States has generally adhered to the principle of free trade, it has made an increasing number of compromises or exceptions to protect textiles, steel, automobiles, and other industries in exchange for political support for the general principle of free trade, in the general context of increased global competition, unemployment, and lagging growth. The instruments used to restrict imports were also different from and less transparent than traditional import tariffs and 311

312

DOMINICK SALVATORE

quotas. These non-tariff barriers (NTBs) refer to "voluntary" export restraints (VERs), orderly marketing arrangements, anti-dumping measures, countervailing duties, safeguard codes — collectively known as the "new protectionism" (Salvatore, 1987a). More recently, talk of "free" trade in the United States has given way to the stated goal of "fair" trade. Free trade is based on the notion of laissez-faire and lack of interference with the trading system. Fair trade, on the other hand, refers to unilateral rules as to what is permissible and what is not. Procedures to enforce fair trade often end up in protectionist measures. As Finger (1991, p. 21) points out, "in domestic politics 'fair trade' has come to mean the right to protection." Talking of fair trade, however, has put the defenders of a free multilateral trading system on the defensive since no one can really argue against fairness in trade. Thus, at the same time that tariffs were reduced and are presently very low on most industrial goods in the United States and in other industrial countries as a result of several successful rounds of multilateral trade negotiations through GATT (General Agreement on Tariffs and Trade), the number and importance of NTBs have grown rapidly since the mid 1970s to the point where they have now become more important than tariffs as obstructions to the flow of international trade. In the process, the United States and the world have slipped more and more toward bilateralism and managed trade. Unless arrested, this process can undermine the entire world trading system - a system that was so painstakingly put together and served the world so well since the end of the Second World War (Finger, 1992; Salvatore, 1992). 2

Present US trade laws and protectionism

Concern over fairness in international trade has led to the revision of US trade laws and their interpretation in recent years away from non-discriminatory multilateralism and toward aggressive bilateralism. Specifically, current US trade laws often demand "full" reciprocity, in the sense of equal access to foreign markets for US products, country by country and product by product, and require retaliation against countries that do not provide market access to US products equal to that which foreign countries and products enjoy in the United States. In short, full reciprocity requires that two countries allow identical access to their markets, implying identical trade restrictions, and thus provide a "level playing field." "Full reciprocity" must be distinguished from "firstdifference" reciprocity, which is the traditional interpretation of reciprocity under GATT rules and refers to the rough balance or equal benefit stemming from tariff concessions by two countries, rather than insisting on

Trade protectionism and welfare

313

identical access and trade restrictions in the two countries (Bhagwati and Irwin, 1987). Since present US trade laws require retaliation against countries that do not provide full reciprocity, they are often referred to as "trade-remedy laws" because they are intended to remedy the hardships that US firms allegedly suffer from less than full reciprocity by foreign nations and thus provide a fair trading environment and a level playing field. The attempt to enforce unilaterally full reciprocity on the part of the United States, however, often results in protectionist actions against foreign countries, which invites retaliation (Bhagwati, 1988, 1991). Table 14.1 lists the major trade-remedy laws currently on the books in the United States and summarizes their key features and the required government responses to foreign trade that results or can lead to injury to US producers or that is unfair.

Section 201: the escape clause The escape clause provision of current US trade law provides for temporary assistance in the form of trade restrictions when fairly traded imports are a substantial cause of serious injury to a domestic industry. The assistance is to be temporary as the domestic industry adjusts to increased imports. Increased imports are a substantial cause of injury if they are no less important than any other cause of injury. Evidence of substantial injury is provided by significant declines in sales, production, profits, wages, or employment. Petitions for relief can be filed by individual firms, trade associations, labor unions, the US Trade Representative, the House Ways and Means Committee, and the Senate Finance Committee. The petitioner submits its case to the International Trade Commission (ITC) - a six-member appointed body - which then undertakes an investigation, including public hearings, to determine if increased imports are a substantial cause of serious injury. An affirmative decision by the ITC is forwarded to the President together with a recommendation of the appropriate trade and/or adjustment assistance. The President must then provide relief and decide what method and amount of relief to provide, unless this is not deemed in the national interest. The Omnibus Trade and Competitiveness Act of 1988 strengthened Section 201 by adding the loss in market share (even if the domestic industry is not harmed in absolute terms by increased imports and even if its sales actually increase) as an acceptable reason for determining to grant protection (Council of Economic Advisers, 1988; Salvatore, 1987a; Grinols, 1989; Bhagwati, 1988, 1991; Coughlin, 1991).

Table 14.1. Principal US trade laws provisions" Statute

Focus

Criteria for action

Response

Responsibility

Section 201: fair trade (escape clause)

Increasing imports

Increasing imports are substantial cause of injury

Duties, quotas, tariff-rate quotas, orderly marketing arrangements, adjustment assistance

President (ITC recommendation)

Section 301: unfair trade Foreign practices violating a trade agreement or injurious to US trade

Unjustifiable, unreasonable or discriminatory practices burdensome to US commerce

All appropriate and feasible action

US trade representative subject to direction by the President

Section 701: subsidized imports

Manufacturing, production or export subsidies

Material injury or threat Duties of material injury*

ITC-injury determination; ITA-subsidy determination

Section 731: dumped imports

Imports sold below cost Material injury or threat Duties of production or below of material injury foreign market price

ITC-injury determination: ITA-dumping determination

Notes: Origin of current provisions: Tariff Act of 1930 (Smoot-Hawley), as amended; Trade Act of 1974, as amended: Trade Agreements Act of 1979, as amended; Trade and Tariff Act of 1984; Omnibus Trade and Competitiveness Act of 1988. b The material injury test is extended only to countries that fulfill certain conditions. Source: Council of Economic Advisers (1988, p. 152), as modified by author. a

Trade protectionism and welfare

315

Section 301: unfair trade The Omnibus Trade and Competitiveness Act of 1988 grants the US Trade Representative (USTR) authority, subject to any direction from the President, to take all "appropriate and feasible action" to force a foreign nation to remove all "unjustifiable, unreasonable, or discriminatory practices, burdensome to US commerce" (i.e., that restrict US exports). This refers not only to foreign measures in violation of trade agreements (already provided by the Trade and Tariff Acts of 1984 and earlier statutes), but also to any action by a foreign government that tolerates domestic anti-competitive activities that leads to the restriction of the purchase of US products by foreign firms. Also actionable under the new Section 301 is export targeting or any coordinated foreign action "to assist the enterprise, industry, or group, to become more competitive in the export of a class or kind or merchandise." Denial of workers' rights, such as the right to bargain collectively, the use of compulsory labor, and the failure to provide standards for minimum wages abroad, are also actionable under the new Section 301 code. The new Section 301 code, often referred to as "Super 301", further (1) calls on the USTR to designate priority countries that maintain numerous and pervasive trade barriers against US exports, (2) sets a rigorous schedule for negotiations to be held on eliminating those barriers, and (3) mandates retaliation in the form of 100 percent ad valorem import tariffs on selected exports to the United States by these countries if they did not relax their trade restrictions against US exports within twelve months (eighteen months for trade agreement cases). The Omnibus and Competitiveness Trade Act of 1988 also transferred authority to determine unfairness from the President to the US Trade Representative, who is more likely than the President to be responsive to protectionist pressures. Super 301 is thus more protectionist in nature than the previous Section 301 because it eliminates the elements of judgment and discretion that characterized the earlier law and leads to more rigid and confrontational actions, and by so doing encourages foreign imitation of mirror legislation (see Council of Economic Advisers, 1988; Salvatore, 1989a; Grinols, 1989; Bhagwati and Patrick, 1990; Coughlin, 1991). Figure 14.1 shows the timing for a representative Section 301 process. Section 701 and Section 731: subsidized and dumped imports Section 701 of the Tariff Act of 1930, as amended, directs the Department of Commerce to examine claims that specified imports have received manufacturing, production or export subsidies by foreign govern-

316

DOMINICK SALVATORE

Lesser of 12 months (18 for trade agreement cases) or 30 days after procedures end 45 days

t

Petition filed

'

30 days

30 days

!

USTR takes case Federal register

Retaliation 30 days

Unfairness decision by USTR

Public Hearings

Notice for public hearings

Public Hearings

Federal register

I

President gives "Specific direction, if any" to stop retaliation

Figure 14.1 A representative case for the Section 301 process Source: Grinols (1989), p. 511

merits, while Section 301 of the Antidumping Act of 1921, as amended, directs the Department of Commerce to investigate claims of dumping by private firms. Dumping refers to the sales in the United States at prices below foreign long-run costs of production or below home-market prices abroad. US imports subsidized by foreign governments or dumped by foreign firms are referred to as "less-than-fair-value" (LFV) cases because they involve the sale in the United States at less than fair value. US law allows exports to the United States that are subsidized or dumped to be counteracted with tariffs called countervailing or anti-dumping duties. This is to protect US producers from unfair or predatory competition from abroad. A countervailing or anti-dumping petition can be filed by a firm, a trade association, a labor union, or the Department of Commerce itself by presenting some evidence that subsidized or dumped imports have materially injured, or threaten to injure materially, or threaten to retard materially the establishment of a domestic industry. The petition is filed simultaneously with the International Trade Administration (ITA) of the Department of Commerce and the ITC. The ITA then conducts an extensive investigation to determine if imports have, in fact, been subsidized or dumped in the United States and to calculate the amount of the subsidy or dumping margin (i.e., the difference between the price at which the product is sold in the United States and the fair value of the product, defined as the price or cost of the product in the exporting country). At the same time that the ITA is conducting its investigation, the ITC

Trade protectionism and welfare

317

investigates whether the domestic industry has been injured by subsidized or dumped imports. After forty-five days of the filing date (and while the ITA is continuing its investigation) the ITC must file a preliminary report on the case. Only if the result of this preliminary investigation is positive (i.e., the domestic industry has been injured or is threatened with injury, the investigation continues. If the domestic industry is not found to have been injured, the case is closed even if imports have been subsidized or dumped. Finally, after 120-210 days (depending on the complexity of the case) from the date of a preliminary affirmative ITC decision, the ITC must submit the final ruling on the case. If in its final report the ITC rules that the domestic industry has in fact been injured by subsidized or dumped imports, a countervailing or anti-dumping duty equal to the amount of the subsidy or dumping margin is automatically imposed by the ITA on foreign exporters (ITC, 1990; Palmeter, 1987). In 1980, the administration of "lessthan-fair-value" cases was shifted from the Treasury Department to the Commerce Department partly because of the feeling that the former was overly sensitive to the interests of consumers and importers as opposed to the interest of domestic producers. This encouraged the filing of lessthan-fair-value complaints (see Salvatore 1989a, Coughlin 1991). 3

Current US trade policies

Table 14.2 shows the total number of trade-remedy petitions filed in the United States from 1979 to 1990, as well as its breakdown into escape clause (Section 201), Section 301 (unfair trade), Section 701 (countervailing duty), and Section 731 (anti-dumping) petitions. From the table we see that a total number of 983 petitions of all kinds were filed in the United States in the twelve years running from 1979 to 1990. The table also shows that the total number of petitions was lower during the last six years than in the previous six years. This, however, cannot be construed as indication that the level of protectionism in the United States is declining. It is rather due to two other reasons. One is that the number of petitions increased sharply during the deep 1982 US recession. The second is that the voluntary export restraint agreements that the United States negotiated for steel with most exporting nations in 1984 and 1985 led to a sharp reduction in the number of countervailing and anti-dumping petitions filed. If we look at the distribution, we see that the number of escape clause and Section 301 petitions filed represented only a small percentage of the total. Also, while the number of escape clause cases declined during the last six-year period as compared with the first period, the number of 301 petitions increased. Table 14.2 shows that the largest number of petitions filed were countervailing duty and anti-dumping complaints, and that the former declined while the

Table 14.2. US trade-remedy petitions filed

Section 301

Escape clause

Countervailing duty

Anti-dumping

Total petitions

Year

No.

%

No.

%

No.

%

No.

%

No.

1979 1980 1981 1982 1983 1984 1979-84 1985 1986 1987 1988 1989 1990 1985-90 1979-90

4 2 6 1 5 6 24 3 3 2 2 0 1 11 35

6.5 5.4 11.3 0.5 6.2 4.5 4.1 2.7 2.9 7.7 3.2 0.0 1.8 2.7 3.5

5 0 10 7 7 2 31 5 6 5 7 9 4 36 67

8.1 0.0 18.9 3.2 8.6 1.5 5.3 4.5 5.7 19.2 11.3 22.0 7.1 9.0 6.8

37 11 22 145 22 52 289 38 26 5 13 9 8 99 388

59.7 29.7 41.5 67.1 27.2 39.1 49.7 34.2 24.8 19.2 21.0 22.0 14.3 24.7 39.5

16 24 15 63 47 73 238 65 70 14 40 23 43 255 493

25.8 64.9 28.3 29.2 58.0 54.9 40.9 58.6 66.7 53.8 64.5 56.1 76.8 63.6 50.1

62 37 53 216 81 133 582 111 105 26 62 41 56 401 983

Source: Messerlin (1990) and Office of United States Trade Representative (1991).

Trade protectionism and welfare

319

latter increased during the more recent period. In order to understand both the frequency and the change in the frequency of the various types of trade-remedy petitions filed we must look at the political economy of protectionism. The political economy of protectionism leads to the following observations. First, it seems that the United States and other industrial nations have increasingly relied on NTBs as the preferred method of protection as a substitute for tariff protection, in order to preserve at least the appearance of continued support and movement toward trade liberalization in the form of negotiated tariff reductions (Salvatore, 1987a). Second, while escape clause cases have been much fewer in number than LFV cases, they were much larger. The average escape clause case is about $600 million as compared with $40 million for the average LFV case. Third, the legal criteria for determining LFV cases are more technical and precise and less political than the criteria for deciding escape clause cases. The more technical "lowertrack" LFV cases minimize the political costs of making decisions, whereas the "higher-track" escape clause cases are decided in the national spotlight and are, therefore, inevitably more political. Finger et al. (1982) point out that escape clause cases are nearly zero-sum decisions for the President, with political gains more or less balancing losses. Specifically, the small political gains from an affirmative decision on the part of the President will encourage more petitions to be filed and subsequently lead to the need for him to have to make more politically difficult decisions. Thus, the filing of escape clause petitions is somewhat discouraged. Furthermore, in affirmative escape clause cases, the President is likely to diffuse political accountability by ordering the negotiation of an orderly marketing arrangement by the US Trade Representative rather than deciding himself what action to take. Fourth, affirmative escape clause petitions, by not charging unfairness on the part of foreigners, are more likely to lead to retaliation. Fifth, changes in US trade laws are making it easier to file anti-dumping petitions and to receive affirmative rulings. Econometric research on NTBs also shows that (1) recession or a lower level of economic activity in the United States leads to more trade-remedy petitions filed (Takacs, 1981; Salvatore, 1987a, 1989a); (2) an overvalued dollar in a given year generally leads to a larger number of petitions filed in the subsequent year (Salvatore, 1987a, 1989a); (3) the very threat of filing a complaint has a protectionist effect (i.e., discourages exports and leads foreign firms to reduce the dumping margin) and is referred to as the "harassment thesis" (Finger, 1981; Herander and Schwartz, 1984). To be pointed out is that the ability to file a complaint is in and of itself (and independently of actual retaliatory action or countervailing action to which it may lead) likely to discourage trade and can reasonably be regarded as

320

DOMINICK SALVATORE

protectionistic because of the cost imposed on the foreign exporter to respond to the complaint. While the domestic industry also incurs the costs of riling a petition, these costs are usually spread over many more units of the product since imports are generally less than one half of domestic consumption for most importable commodities. Domestic firms are also more likely to have a further advantage over foreign firms because of their greater experience in filing and pursuing such petitions. On balance, therefore, the very act of filing a trade complaint is likely to have a protectionist effect. The actual application of Section 301, and especially Super 301 (the latter introduced by the Omnibus Trade and Competitiveness Act of 1988) has also led to increased protectionism on the part of the United States and has encouraged other countries, especially the European Community, to introduce similar legislation and apply it in a similar protectionist manner. The introduction of Super 301 generated sharp protests by the US major trade partners in Europe and Japan, but it produced only a small number of offenders and market-opening actions. In 1989, the United States named Japan, Brazil, and India as the most unfair traders. Japan was cited for the refusal of its public authorities to purchase US satellites and supercomputers, and for excluding US manufactured forest products. Brazil was cited for the licensing requirements it imposes on practically all imports, and India for the restrictions on foreign investments and curbs on foreign-based insurance companies. There is a great deal of disagreement regarding the effectiveness of Section 301 and Super 301 in opening foreign markets. The complaints brought under Super 301 against Japan and Brazil in 1989 did lead to some market-opening responses by these two countries and, as a result, they were removed from the list in 1990. India, on the other hand, remains on the list. Taiwan and South Korea did make some marketopening measures in favor of US exports (and at the expense of exports from other nations) in order to remain off the US list. In the vast majority of Section 301 petitions, however, this "crowbar approach" to opening foreign markets more widely to US exports has generally failed. Despite this, the use of this trade-remedy approach is very likely to be used with increased frequency in the future and to induce other nations to respond in kind. The overall conclusion that emerges is that while modern LFV codes have been on the books for many decades, and their alleged purpose is to defend domestic producers against unfair trade practices by foreign exporters and governments, the way these codes have been interpreted and enforced leaves no doubt that they have increasingly been applied for protectionist purposes and have, in fact, led to greater US protectionism. Indeed, they are the most important expressions of the new protectionism in

Trade protectionism and welfare

321

the United States and abroad. The same is true for Section 301 and Super 301 petitions and actions. By also discriminating among a country's trade partners, these LFV and 301 codes violate the most-favored-nation principle, which is the cornerstone of the GATT, the multilateral agreement that governs international trade. Most trade-remedy legislation has introduced what is often referred to as "procedural protectionism" because it relies on procedural rules to hide the protectionist impact of these laws. Thus, rather than facilitating free trade, fair trade laws have often been used for protectionist purposes and have in fact led to more protected international trade in the United States and abroad. Today, almost one-fifth of US imports are subject to this new protectionism. The same is generally true for most other major industrial nations. 4

Examples of new protectionism in the United States

The three most important and notorious examples of increased US protectionism are in textiles and apparel, steel, and automobiles. A summary of each is given below, followed by estimates of the benefits to US consumers that would result from removing protection or from changing its form so as to capture the rents arising from protection in these industries. Textiles and apparel Control over the textile trade started in 1956 when the United States and other industrial nations began to restrict cotton exports from Japan. Over time, export restrictions became more and more complex and comprehensive. In 1962, the Long-Term Arrangement (LTA) on cotton and textiles, negotiated under the leadership of the United States, restricted imports of cotton textiles from Japan and most developing-country exporters. This was followed in 1974 by the Multifiber Arrangement (MFA), which limited the growth of all textile and clothing imports into the United States to 6 percent per year. The renewal of the MFA in 1977 and then again in 1981 reduced still further the growth of textile and clothing imports quotas into the United States, and reduced imports from "dominant" suppliers even more. Subsequent renewals and extensions of the MFA limited imports of all synthetic and nearly all natural fiber textiles and apparel into the United States and most other industrial countries from nearly all developing countries. Today, the United States has bilateral textile export restraints with forty-six countries and territories, and the industry is getting a higher degree of protection than any other major US industry, with equivalent tariffs averaging 22 percent, compared with less than 5 percent for all other industries. Nevertheless, employment in the

322

DOMINICK SALVATORE

industry declined by more than 250,000 since 1980 under the pressure of imports and automation. Steel The United States negotiated voluntary export restraints with all the leading steel exporters from 1969 to 1974, orderly marketing arrangements that reduced imports of specialty steel from 1976 to 1980, and a trigger-price mechanism to speed up and facilitate dumping investigations in 1978. Steel imports into the United States, however, continued to increase, and in early 1982 US steel producers filed more than 100 anti-dumping and countervailing duty petitions against steel exporters from eleven countries. In late 1982, the United States negotiated a new voluntary export restraint limiting the European Community's steel exports to about 6 percent of the US market. Japan also unilaterally kept its steel exports from exceeding 5 percent of the US market. In spite of these restrictions, steel imports into the United States swelled from 16 percent of the domestic market in 1980 to 26 percent in 1984. Most of this increase came from the newly industrializing countries of Korea, Brazil, Mexico, as well as from Spain, Australia, and South Africa. By November 1985, the United States had negotiated a total of fifteen export restraint agreements which limited for five years imports of steel into the United States to 20.5 percent of the domestic market. Similar import restraints were imposed by the European Community. In 1989, President Bush, honoring a campaign promise, extended the agreement until 1992. When they expired on March 31 1992, US steel producers immediately filed nearly 100 anti-dumping and countervailing duty petitions against foreign steel exporters of more than twenty countries - foreshadowing a return to the conditions prevailing in the early 1980s before the voluntary export restraints were negotiated. During the ten years that the voluntary export restraints were in effect, US steel producers spent nearly $25 billion on modernization, doubling productivity, and bringing parts of the industry up to world standards. This, however, resulted as much, if not more, from the elimination of the large dollar overvaluation during the second half of the 1980s and from increased domestic competition from the low-cost and non-union mini mills that sprang up during the 1980s as from the voluntary export restraints, since foreign producers often failed to fill their steel export quotas after 1985. Automobiles From 1977 to 1981, US automobile production fell by about one third, the share of imports rose from 18 to 29 percent, and nearly 300,000

Trade protectionism and welfare

323

auto workers in the United States lost their jobs. In 1980, the Big Three US auto makers suffered combined losses of $4 billion. As a result, the United States negotiated an agreement with Japan which limited Japanese automobile exports to the United States to 1.68 million units per year from 1981 to 1983 and to 1.85 million units for 1984 and 1985. Japan "agreed" to restrict its automobile exports out of fear of still more stringent import restrictions by the United States. US auto makers generally used the time from 1981 to 1985 wisely to lower break-even points and improve quality, but the cost of improvements was not passed on to consumers, and Detroit reaped profits of nearly $6 billion in 1983, $10 billion in 1984, and $8 billion in 1985. Japan gained by exporting higher-priced autos and earning higher profits. The big loser, of course, was the American public, who had to pay substantially higher prices for domestic and foreign automobiles. The ITC estimated that the agreement resulted in a price $660 higher for US-made automobiles and $1,300 higher for Japanese cars in 1984. The ITC (1985) also estimated that the total cost of the agreement to US consumers was $15.7 billion from 1981 through 1984, and that 44,000 US auto jobs were saved at a cost of more than $100,000 each. This was more than two to three times the yearly earnings of a US auto worker. Since 1985, the United States has not asked for a renewal of the VER agreement, but Japan unilaterally limited its auto exports to avoid more trade frictions with the United States. During the late 1980s and early 1990s, Japan invested heavily to produce automobiles in the United States in so-called transplant factories and by 1992 it was producing more than one million cars in the United States. This type of investment made by a nation in its export market to overcome trade restrictions and future trade controversies was branded quid pro quo investments by Bhagwati (1987). By 1992, Japan has captured 31 percent of the US automarket (18 percent from exports and 13 percent from production in the United States). Following the US lead, Canada and Germany also negotiated restrictions on Japanese automobile exports (France and Italy already had very stringent export quotas on Japanese automobiles). In 1911, an agreement was reached to limit the Japanese share of the European Community auto market to 16 percent by the end of the decade. During the 1990-2 recession, US auto makers came under increased pressure and faced huge losses, and once again began to demand increased protection against Japanese auto exports and production in Japanese-owned transplanted factories in the United States. In a gesture clearly aimed at avoiding increased US protection, Japanese auto makers announced in March 1992 that they would reduce their self-imposed export limit of 2.3 million cars per year to the US market to 1.65 million per year. However, Japan had failed to fill its 2.3 million quota since 1986 and exported only 1.76 million cars to the United States in 1991.

324

DOMINICK SALVATORE

Table 14.3. Welfare benefits and employment effects of removing or changing quantitative restrictions (QRs) on exports of textile and apparel, automobiles, and steel to the United States Employment change in rest of economy (000 work-years)

Welfare gain (billion dollars)

Employment change in industry (000 work-years)

11.92

-157.2

157.2

6.05

-3.1

28.4

7.50

-1.2

1.3

7.15

1.3

37.2

0.86

-20.7

22.3

0.74

-0.1

3.5

Textile and apparel

Remove QRs Capturing rents from foreigners Automobiles

Remove QRs Capturing rents from foreigners Steel Remove QRs Capturing rents from foreigners

Source: J. de Melo and D. Tarr, "Welfare Costs of US Quotas on Textile and Apparel, Automobiles, and Steel", Review of Economics and Statistics^ August 1990.

Welfare effects of removing or changing US trade restrictions There have been several attempts at measuring the welfare effect of removing or changing US trade restrictions. One of these (by de Melo and Tarr, 1990) is summarized in table 14.3. Table 14.3 shows that removing all quantitative restrictions (QRs) on textile apparel exports to the United States would result in a gain of $11.92 billion to the United States at 1984 prices. Retaining QRs but capturing the rents from foreigners (for example by auctioning off export quotas to foreign firms) would result instead in a gain of $6.05 billion to the United States. The gains are smaller for automobiles and much smaller for steel. Removing QRs would also lead to employment losses in the industry losing the QRs, but these employment losses would be matched or more than matched by economy-wide employment gains. Removing QRs on all three products would thus lead to a total welfare gain of $20.28 billion to the United States. Eliminating also all tariffs on industrial products after all the above QRs have been removed would result in a further gain of $0.6 billion to the United States (not shown

Trade protectionism and welfare

325

in the table). Other studies measuring the welfare effects of removing or changing US trade restrictions are by US International Trade Commission (1985), Hufbauer, Berliner, and Elliot (1986), Rouslang (1988), US International Trade Commission (1989), Tarr (1989), and by de Melo and Tarr (1992). 5

The political economy of protectionism in the United States

In recent years there have been many attempts to explain the formation of trade policy in general and protectionism in particular in the United States and in other industrial nations in terms of the interests of those who benefit from protection and those who lose (Olson, 1965; Caves, 1976; Baldwin, 1984 and 1985; Magee and Young, 1987; Marks and McArthur, 1990; Krueger, 1990; Eithier and Horn, 1991). These models of endogenous protection postulate that legislators are restrained from yielding to demands for protection only by the opposing political demands of those who stand to lose from protection. That is, legislators act as intermediaries between the strong narrow protectionist interests of producers and the broad anti-protectionist interests of voters (consumers). However, since producers are few and stand to gain a great deal from protection, they have a strong incentive to lobby the government to adopt protectionist measures. On the other hand, since the losses are diffused among many consumers, each of whom loses very little from the protection, they are not likely to organize effectively to resist protectionist measures. Thus, there is a bias in favor of protectionism. For example, the sugar quota raises individual expenditures on sugar by only a few dollars per person per year in the United States. But with nearly 250 million people in the United States, the quota generates more than $600 million in rents to the few thousand sugar producers in the United States. Neutralizing the protectionist bias in the United States Since the late 1930s, the protectionist bias inherent in the gainers-losers accounting has been neutralized in the United States by three important forces. The first was the delegation of authority over trade policy from Congress (which is more susceptible to pressure groups) to the President (who is more likely to consider the broader national interest) after the disastrous Smoot-Hawley Tariff Act was passed by a log-rolling Congress in a protectionist frenzy in 1930. To be sure, Congress continued to set guidelines in trade matters and individual Congressmen spoke loudly on behalf of producers' interests secure in the knowledge that actual trade

326

DOMINICK SALVATORE

decisions would be made elsewhere (Destler, 1992, pp. 14-15). As Baldwin (1985) points out, the President also tends to be more liberal on trade policy because this makes the conduct of foreign policy easier. To insulate itself further from pressure groups, Congress also created in 1962 the post of the Special Trade Representative with cabinet rank to deal with trade matters, passed "trade remedies" legislation designed to offer recourse to US firms seriously injured by imports or facing "unfair" foreign competition, and provided for trade adjustment assistance. The second force that helped overcome the otherwise protectionist bias in US trade policy in the past resulted from turning the policy agenda from limiting imports to reducing trade barriers and promoting exports. As Destler (1992, p. 16) points out, this enlisted the help of the public in general and of exporters in particular in promoting trade liberalization, thus creating a political counterweight on the liberal trade side. The third important force that favored trade liberalization resulted from the fact that both political parties in the United States did not make trade policy a major source of conflict during the early post-war period. This meant continuity of trade liberalization across administrations in Washington. These three forces allowed the trade liberalization that started in the late 1930s and gained momentum in the 1950s and 1960s, and which led to the rapid expansion of world trade, increased international specialization and growth, and served as the pillar of US global leadership in the trade field. Reasons for increased protectionism during the past two decades During the 1970s and 1980s, the balance tilted in favor of the protectionist forces and this resulted in sharply increased protectionism in the United States and in other industrial nations. There were several reasons for this. First, there were the more difficult macroeconomic conditions of the last two decades. As pointed out earlier, high unemployment leads to increased calls for protectionism. Furthermore, since the volume of international trade more than doubled as a share of total output in the United States during the past two decades, the number of potential losers (firms and individuals) demanding protection against foreign competition more than doubled. Second, there were the slower growth and reduced international competitiveness of the United States relative to Japan and Europe. While successive administrations in Washington sought to blunt the resulting demands for increased protection with demands for increased access to foreign markets, the net effect was nevertheless one of increased protectionism. The demands for protection in the United States were particularly strong during the middle 1980s on account of the large

Trade protectionism and welfare

327

overvaluation of the dollar. At the peak of its value in February 1985, the dollar was overvalued by about 40 percent on a trade-weighted basis with respect to the currencies of the major trade partners of the United States. This was equivalent to a 40 percent subsidy on US imports and tax on US exports, and resulted in sharply increased US trade deficits. Labor and many industries, which had supported free trade in the early post-war period, now turned toward protectionism, or at least toward demanding fairness and a level-playing field in trade. Third, intellectual support for free trade seemed to be wavering. Strategic trade theories (see section 6) showed that a nation could gain at the expense of its trade partners by supporting strategic industries, such as telecommunications and computers, and nations such as Japan that seemed to practice such strategic trade policies outperformed the United States at home and abroad. Finally, trade politics has grown more partisan during the past two decades, as Democrats saw greater electoral opportunities from adopting a more protectionist stand and Republicans followed suit. Congress also began to battle the White House to regain control over trade policy, and this forced the White House to grant special protection to steel, automobilies, and some other industries, and to make stronger marketopening demands on foreigners. As evidence of the Congressional desire to regain control over trade matters is the passage of the Omnibus Trade and Competitiveness Act of 1988 - the first major trade legislation initiated by Congress since 1930. Haufbauer, Berliner and Elliot (1986, p. 21) found that, as a result of these protectionist pressures, US imports subject to protection increased from 12 percent in 1980 to 21 percent in 1984 and de Melo and Tarr (1992, chapters 5 and 15) estimated that this was equivalent to an average nominal tariff of 24 percent on all imports. This new protectionism thus completely reversed the trade-liberalizing progress made during the earlier post-war years. Things could have been worse, however, with protectionism rising much more than it actually did and becoming more explicit and overt. Specifically, by demanding fair trade and a level-playing field, and imposing trade restrictions only in response to alleged foreign restrictions and in the absence of greater trade-opening measures by foreigners, the principle that freer trade is better than more restricted trade was at least saved. There were two reasons for the absence of even greater protectionism during the past decade. One is that the overvaluation of the dollar was eliminated during the second half of the 1980s and this helped reduce the pressure on US exporters and resulted in smaller US trade deficits. The second reason is that in the much more integrated and globalized world economy of the late 1980s and early 1990s, there developed in the United States stronger special interests against protectionism. These are the multinational or global firms,

328

DOMINICK SALVATORE

which rely on imported inputs and global markets for sales. For example, users of steel opposed the renewal of negotiated steel quotas in 1989 and were certainly a factor in President Bush's refusal to renew those quotas in 1992. Similarly, strong domestic opposition to the imposition of restrictions on Toshiba's exports to the United States for selling sensitive equipment to the former Soviet Union in 1989 was essentially blocked by a coalition of US producers using many of Toshiba's inputs in their production. Who gets protected? In recent years economists have developed several theories regarding which group, and industries get protected, and some of these theories have been empirically confirmed. In the United States and other industrial countries, protection is more likely to be provided to laborintensive industries employing unskilled, low-wage workers who would have great difficulty in finding alternative employment if they lost their present job (Olson, 1965; Caves, 1976). Some empirical support has also been found for the pressure-group or interest-group theory, which postulates that industries that are highly organized (such as the automobile industry) receive more trade protection than less organized industries. An industry is more likely to be organized if it is composed of only a few firms. Magee and Young (1987) found that industries that produce consumer products generally are able to obtain more protection than industries producing intermediate products used as inputs by other industries because the latter industries can exercise countervailing power and bloc protection (since that would increase the price of their inputs). Furthermore, as pointed out by Baldwin (1984), more protection seems to go to geographically decentralized industries that employ large numbers of workers than to industries that operate in only some regions and employ relatively few workers. The large number of workers have strong voting power to elect government officials who support protection for the industry. Decentralization ensures that elected officials from many regions support trade protection. Another theory suggests that trade policies are biased in favor of maintaining the status quo. That is, it is more likely for an industry to be protected now if it was protected in the past. Governments also seem reluctant to adopt trade policies that result in large changes in the distribution of income, regardless of who gains and who loses. Finally, Michaely (1987) points out that protection seems to be more easily obtained by those industries that compete with products from developing countries since these countries have less economic and political power than industrial countries to resist trade restrictions against their exports

Trade protectionism and welfare

329

successfully. To be noted is that some of the above theories are overlapping and some are conflicting, and that they have been only partially confirmed empirically. 6

The new protectionism and strategic trade policy in the United States

Strategic trade policy is another qualified argument that is being advanced in favor of an activist trade policy and protectionism in the United States and in other industrial countries. According to this, a nation can create a comparative advantage (through temporary trade protection, subsidies, tax benefits, and cooperative government-industry programs) in such fields as semiconductors, computers, telecommunications, and other industries that are deemed crucial to future growth in the nation. These high-tech industries are subject to high risks, require large-scale production to achieve economies of scale, and give rise to extensive external economies when successful. Strategic trade policy suggests that, by encouraging such industries, the nation can reap the large external economies that result from them and enhance its future growth prospects. This is similar to the infant industry argument in developing nations, except that it is advanced for industrial nations to acquire a comparative advantage in crucial high-tech industries. Most nations, including the United States, do some of this. Indeed, some economists in the United States would go so far as saying that a great deal of the post-war industrial -and technological success of Japan is due to its strategic industrial and trade policies. Examples of alleged strategic trade and industrial policies in Japan are in the steel industry in the 1950s and semiconductors in the 1970s and 1980s, and, in Europe, the development of the supersonic aircraft, Concorde, in the 1970s and the Airbus aircraft in the 1980s. Semiconductors in Japan are usually given as the textbook case of successful strategic trade and industrial policy. The market for semiconductors (such as computer chips that are used in many new products) was dominated by the United States in the 1970s. Starting in the mid 1970s, Japan's powerful Ministry of Trade and Industry (MITI) targeted the development of this industry by financing research and development, granting tax advantages for investments in the industry, fostering government-industry cooperation, while protecting the domestic market from foreign (especially US) competition. These policies are credited for Japan's success in wresting control of the semiconductor market from the United States in the mid 1980s and dominating world markets thereafter, and are often used as justifications for advocating an activist trade policy and protectionism for the United States (Prestowitz, 1988; D'Andrea Tyson, 1992). Most economists in the United

330

DOMINICK SALVATORE

States remain skeptical, however, and attribute Japan's stunning performance in this field primarily to other forces, such as greater educational emphasis on science and mathematics, higher rates of investment, and a willingness to take a long-run view of investments rather than stressing quarterly profits as in the United States. In steel, the other targeted industry in Japan, the rate of return was lower than the average return for all Japanese industries during the post-war period. In Europe, the Concorde was a technological feat but a commercial disaster, and Airbus Industrie would not have survived without continued heavy government subsidies. While strategic trade policy can theoretically improve the market outcome in oligopolistic markets subject to extensive external economies and increase the nation's growth and welfare, even the originators and popularizers of this theory in the United States recognize the serious difficulties in carrying it out (Brander and Spencer, 1983; Krugman, 1986; Helpman and Krugman, 1989). First, it is extremely difficult to pick winners (i.e., choose the industries that will provide large external economies in the future) and devise appropriate policies to successfully nurture them. Secondly, since most leading nations undertake strategic trade policies at the same time, their efforts are largely neutralized so that the potential benefits to each may be small. Thirdly, when a country does achieve substantial success with strategic trade policy, this comes at the expense of other countries (i.e., it is a beggar-thy-neighbor policy) and so other countries are likely to retaliate. Faced with all these practical difficulties, even the intellectual supporters of strategic trade policies in the United States grudgingly acknowledge that free trade is still the best policy, after all (Krugman, 1987). This, however, does not discourage advocates of protectionism from using this argument for protectionism in the United States. Pro-protectionistic forces in the United States will take any argument that can potentially be used to further their cause - wherever it comes from — and this, introduced by generally liberal economists in the United States, is particularly welcomed by them. One US trade action based on a strategic (both commercial and security) industry rationale was the 1986 Semiconductor Agreement between the United States and Japan. This resulted from the anxiety in commercial circles and in the Pentagon about the erosion of US technological leadership in the semiconductor industry due in part to Japanese dumping of computer chips on the American market and around the world in the face of allegedly unreasonable barriers to US exports of computer chips by Japan. The 1986 Agreement prohibited the dumping of Japanese chips in the United States and around the world by requiring MITI to operate a minimum export price system and also requiring the opening of Japanese markets to US chip exports to eventually supply 20 percent of the Japanese market by 1991 (up

Trade protectionism and welfare

331

from less than 10 percent in 1986). In 1987, the United States imposed a 100 percent tariff on $300 million of Japanese consumer electronics exports to the United States for the failure of Japan to enforce the agreement adequately. These restrictions were reduced later that year and removed in 1991 when a new semiconductor agreement was signed. As in the case of automobiles, however, this agreement increased the profits of Japanese semiconductor firms, allowing them to invest more and become even more competitive, while at the same time forcing US firms to pay higher prices for computer chips. On a broader scale, the United States and Japan are engaged in negotiations (called the Structural Impediments Initiative or SII) aimed, among other things, to open the entire Japanese distribution system more widely to US firms. The United States is also demanding a sharp reduction in subsidies granted to the Airbus Industrie by the governments of France, Germany, England, and Spain and in agricultural subsidies by the European Economic Community. Furthermore, the United States has also requested that other countries, such as Brazil, China, and India remove excessive restrictions against specific US exports and demanded protection for its intellectual property (such as patented materials) from unauthorized and uncompensated use. 7

Trading blocs and the future of US protectionism

As support for a multilateral trading system weakened, the United States moved more and more toward bilateral deals and regionalism. One reflection of this is the pursuit by the United States of a North American Free Trade Area (NAFTA) with Canada and Mexico. This was also stimulated by the movement toward the completion of a unified market in the European Community (EC) by the end of 1992. Such breaking up of the world into three major trading blocs (the EC, NAFTA, and an Asian trading bloc centered on Japan) is inherently dangerous and could lead to increased protectionism, managed trade, and even trade wars among the blocs - to the detriment of all. The likelihood of such an outcome would certainly increase if the stalled Uruguay Round fails or reaches only meager results. The increased level of protectionism during the past two decades does not make the United States the worst trade offender, but by moving toward bilateralism and regionalism, the world has lost the most enthusiastic supporter of multilateralism and free trade. Furthermore, the move toward greater negotiating activism, the stressing of fair rather than free trade, and the increased willingness to threaten confrontation and retaliation on the part of the United States does not bode well for the future of the

332

DOMINICK SALVATORE

international trading system. They certainly make the conduct of trade policies more unpredictable and the maintenance of an open trading system more difficult. This increase in "process protectionism" also restricts the room for maneuver for liberal-minded administrations and congressional leaders in the United States. It is true that by stressing foreign market opening measures, the United States has avoided greater more direct and traditional forms of protectionism, but the movement toward unilateralism, bilateralism, and regionalism remains, nevertheless, very dangerous. It is, of course, possible that the present US aggressiveness in trade matters is only a temporary response to the large US trade deficits and generally difficult economic conditions of the last decade, and that as conditions improve the United States will lean back toward its traditional support for multilateralism. But this requires strong domestic policies to reverse the declining growth trend in the United States relative to its most important trade partners - specifically Japan and Germany. It is generally agreed that in order to increase its growth and improve its international competitiveness, the United States needs to increase domestic savings, improve education, and encourage research and innovations. To the extent that these measures are adopted and bear fruit, the trade deficit of the United States will stop being a problem, anxiety about falling behind Japan and Germany will diminish, and so will the demand for protection. Only then the United States might resume its traditional role of promoter of a free, multilateral trading system. Note This chapter is a revised version of an article that appeared in Vol. 3, No. 3, 1992 issue of the Open Economies Review. References Baldwin, R.E. (1984), "Trade Policies in Developed Countries," in R.W. Jones and P.B. Kenen (eds.), Handbook of International Economics, Vol. 1. Amsterdam: North-Holland. (1985), The Political Economy of U.S. Import Policy, Cambridge, Mass.: The MIT Press. (1987), "The New Protectionism: A Response to Shifts in National Economic Power," in D. Salvatore (ed.), The New Protectionist Threat to World Welfare. New York: North-Holland, pp. 95-112. Baldwin, R.E., C.B. Hamilton, and A.Sapir (eds.) (1988), Issues in U.S.-E.C. Trade Relations, Chicago: University of Chicago Press. Bhagwati, J.N. (1987), "Political-Economy-Theoretic Analyses of International Trade: VIEs, Quid Pro Quo and VIEs," Journal of International Trade, 22,1-15.

Trade protectionism and welfare

333

(1988), Protectionism, Cambridge, Mass.: The MIT Press. (1991), The World Trading System at Risk, Princeton, NJ: Princeton University Press. Bhagwati, J. and D.A. Irwin (1987), "The Return of the Reciprocitarians - U.S. Trade Policy Today," World Economy, 2, 109-30. Bhagwati, J.N. and H.T. Patrick (eds.) (1990), Aggressive Unilateralism: America's 301 Trade Policy and the World Trading System, Ann Arbor, Michigan: University of Michigan Press. Boadu, F.O, and E. Wesley F. Peterson (1990), "Enforcing United States Foreign Trade Legislation: Is There a Need for Expanded Presidential Discretion?" Journal of World Trade 24, 79-93. Brander, B.J. and B. Spencer (1983), "International R&D Rivalry and Industrial Strategy," Review of Economic Studies, 50, 707-22. Caves, R.E. (1976), "Economic Models of Political Choice: Canada's Tariff Structure," Canadian Journal of Economics, 9, 215—33. Corden, M.W. (1987), Protection and Liberalisation: A Review of Analytical Issues, Washington, DC: International Monetary Fund. Coughlin, C. (1991), "U.S. Trade-Remedy Laws: Do They Facilitate or Hinder Free Trade?" Federal Reserve Bank of St. Louis Review, 73, 3-18. Council of Economic Advisors (1988), Economic Report of the President, Washington, DC: United States Government Printing Office. (1992), Economic Report of the President, Washington, DC: United States Government Printing Office, de Melo, J. and D. Tarr (1992), U.S. Foreign Trade Policy, Cambridge, Mass.: The MIT Press. (1990), "Welfare Costs of U.S. Quotas on Textile and Apparel, Automobiles, and Steel," Review of Economics and Statistics, 72, August. Destler, I.M. (1992), American Trade Politics, Washington, DC and New York: Institute for International Economics and Twentieth Century Fund. Dornbusch, R.W., A.O. Krueger, and L. D'Andrea Tyson (eds.) (1990), An American Trade Strategy: Optionsfor the 1990s, Washington, DC: The Brookings Institution. Eithier, W.J. and H. Horn (1991), "Modern Mercantilism and a Liberal Trade Order," Seminar Paper No. 498, Stockholm: Institute for International Economic Studies, September. Finger, J.M. (1981), "The Industry-Country Incidence of 'Less than Fair Value' Cases in U.S. Import Trade," Quarterly Review of Economics and Business, 21, 260-79. (1991), "The Meaning of'Unfair' in U.S. Import Policy," World Bank Working Paper WPS 745, July. (1992), "Trade Policies in the United States," in D. Salvatore (ed.), Handbook of National Trade Policies, New York and Westport Conn.: North-Holland and Greenwood Press. Finger, J.M., H. Keith Hall, and Douglas R. Nelson (1982), "The Political Economy of Administered Protection," American Economic Review, 72,452—66.

334

DOMINICK SALVATORE

Grinols, E.L. (1989), "Procedural Protectionism: The American Trade Bill and the New Interventionist Mode," Weltwirtschaftliches Archiv, 125, 501-21. Helpman, E. and P.R. Krugman (1989), Trade Policy and Market Structure, Cambridge, Mass.: The MIT Press. Herander, M.G. and J.B. Schwartz (1984), "An Empirical Test of the Impact of the Threat of U.S. Trade Policy: The Case of Antidumping Duties," Southern Economic journal, 51, 59-79. Hufbauer, G.H., D.T. Berliner, and K.A. Elliot (1986), Trade Protection in the United States: 31 Case Studies, Washington, DC: Institute for International Economics. Jackson, J.H. and E.A. Vermulst (1989), Antidumping Law and Practice, Ann Arbor, Michigan: Michigan University Press. Krueger, A.O. (1990), "The Political Economy of American Protection in Theory and Practice," National Bureau of Economic Research, Working Paper No. 3544, December. Krugman, P.R. (1987), "Is Free Trade Passe?" Journal of Economic Perspectives, 1, 131-44. (1986), Strategic Trade Policy and the New International Economics, Cambridge, Mass.: The MIT.Press. Magee, S.P. and L. Young (1987), "Endogenous Protection in the United States, 1900-1984," in R.M. Stern (ed.) U.S. Trade Policies in a Changing World Economy. Cambridge, Mass.: The MIT Press. Marks, S.V. and J. McArthur (1990), "Empirical Analyses of the Determinants of Protection: A Survey and Some New Results," in J.S. Odell and T.D. Willett (eds.), International Trade Policies, Ann Arbor, Michigan: Michigan University Press. Messerlin, P.A. (1990), "Antidumping," in J.J. Schott (ed.), Completing the Uruguay Round: A Results-Oriented Approach to the GATT Trade Negotiations, Washington DC: Institute for International Economics. Michaely, M. (1987), "The Demand for Protection Against Exports of Developing Countries," in D. Salvatore (ed.), The New Protectionist Threat to World Welfare, New York and Amsterdam: North-Holland. Odell, J.S. and T.D. Willett (eds.) (1990), International Trade Policies, Ann Arbor, Michigan: Michigan University Press. OECD (1985), Costs and benefits of Protection, Paris: OECD. Office of the United States Trade Representative (1991), Foreign Trade harriers, Washington DC: United States Government Printing Office. Olson, M. (1965), The Logic of Collective Action, Cambridge, Mass.: Harvard University Press. Palmeter, D.N. (1987), "Injury Determinations in Antidumping and Countervailing Duty Cases - A Commentary on U.S. Practice," Journalof World Trade Law, 21, 7-45. Pearson, C. and J. Reidel (1990), "United States Trade Policy: From Multilateralism to Bilateralism?" in E. Grilli and E. Sassoon (eds.), The New Protectionist Wave, London: Macmillan, pp. 100—19. Prestowitz, C.V. (1988), Trading Places, New York: Basic Books.

Trade protectionism and welfare

335

Richardson, J.D. (1991), "U.S. Trade Policy in the 1980s: Turns and Roads Not Taken," National Bureau of Economic Research, Working Paper n. 3725, June. Rousslang, D.J. (1988), "Import Injury in U.S. Trade Law: An Economic View," International Review of Law and Economics, 8, 117—22. Salvatore, D. (1987a), "Import Penetration, Exchange Rates, and Protectionism in the United States," Journal of Policy Modeling, 8, 125-41. (ed.) (1987b), The Protectionist Threat to World Welfare, New York: NorthHolland. (1988), "The New Protectionism with Non-tariff Instruments," in C. Saunders (ed.), Macroeconomic Management and the Enterprise in the East and the West, London: Macmillan, pp. 155—82. (1989a), "A Model of Dumping and Protectionism in the United States," Weltwirtschaftiches Archiv, 125, 763-81. (ed.) (1991a), Handbook of National Economic Policies, Amsterdam and West Port, Conn.: North-Holland and Greenwood Press. (1991b), The Japanese Trade Problem and the U.S. Response, Washington, DC: Economic Policy Institute. (1991c), "The U.S. Trade Problem with Japan," Challenge, 34, 40-6. (1991d), "Trade Protection and Foreign Direct Investments in the United States," Annals of the American Academy of Social Sciences, 516, 91—105. (ed.) (1992), Handbook of National Trade Policies, Amsterdam and West Port, Conn.: North-Holland and Greenwood Press. (1993), "Trading Blocs and Protectionism," in M. Kreinin (ed.), International Trade and Finance in the 1990s, New York: Francis & Taylor. Stern, R.M. (ed.), U.S. Trade Policies in a Changing World Economy, Cambridge, Mass.: The MIT Press. Takacs, W.E. (1981), "Pressure for Protectionism," Economic Inquiry, 19, 687-93. Tarr, D. (1989), A General Equilibrium Analysis of the Welfare and Employment Effects of U.S. Quotas in Textiles, Auto and Steel, Washington DC: Bureau of Economics, Federal Trade Commission. Tyson, L.D. (1992), Who's Bashing Whom? Trade Conflicts in High-Tech Industries, Washington, DC: Institute for International Economics. United States International Trade Commission (1985), A Review of Recent Developments in the U.S. Automobile Industry Including an Assessment of the Japanese Voluntary Restraint Agreements, Washington, DC: International Trade Commision. (1989), The Economic Effects of Significant U.S. Import Restraints, Phase I: Manufacturing, Washington, DC: International Trade Commission. (1991), Annual Report, Washington, DC: International Trade Commission.

CHAPTER £

^

Protectionism and growth of Japanese competitiveness RYUZO SATO, RAMA RAM AC HAND RAN , AND SHUNICHI TSUTSUI

1

Introduction

The idea that the United States won the Second World War only to lose the subsequent economic war is widely repeated in popular and professional press. Business Week (June 15,1990, p. 35) begins the lead article in its special issue on "Innovation; The Global Race" with a story on the paintings in the corridors of Pentagon that depict the great American battles. The plaque commemorating the last global war reads, "World War II, 1941—"; the missing date, it is suggested, is prophetic of the continuing epic struggle for economic leadership. Fortune (December 2, 1991, p. 56) argues in its cover story that America is not innovating fast enough for global competition just when innovations are more important than ever. What is this global technological competition that America is in danger of losing? Hardly three decades ago, the United States had the highest standard of living and an enviable growth rate. Its firms had a dominant share of the world market, selling cars, computers, heavy machinery, office and photographic equipment, and televisions. As foreign firms acquired the technology to compete with American firms - through licensing, imitation, or innovation - many US firms were forced to concede market share, to quit a particular line of business or even go out of existence. The resulting restructuring imposed considerable hardship on certain sections of the population at the very time when the general public was noticing a slowdown in the growth rate of the standard of living. The globalization of the economy was seen not only as a threat to the individual firms but to the American way of life itself. But Krugman (1991, p. 811) notes the fallacy of identifying competition 336

Japanese competitiveness

337

among nations with that among firms; indeed they are so dissimilar that many economists believe the word "competitiveness" to be meaningless when applied to countries. While firms can be competed out of existence, international trade cannot lead to the liquidation of an economy. The Ricardian proposition that an economy will have comparative advantage in some commodities even if it has no absolute advantage in any and that the economy can export those commodities in which it has a comparative advantage remains valid even today. Classical international trade theory also showed that trade will in general raise the welfare of both nations. It is these propositions that convinced economists of the superiority of free trade. But countries that lose the productivity race will have to compete on the basis of lowering their wages relative to those that win it; but classical theory did not derive conclusions on the relative welfare implications of different growth rates in productivity. The classical models of international trade differ in detail but they assume that technology, resources, and the pattern of comparative advantage are given to an economy. It is true that some exceptions were noted: the infant industry argument has that some industries disadvantaged in the initial stages of growth would, given protection from the ravages of foreign competition for a limited period of time, transform themselves into those that have comparative advantage in the world market. Existence of externalities was justification enough to impose tariffs or subsidies. It is only recently that the theoretical tools for discussing international trade in a non-competitive environment were developed and it is now possible to abandon the three assumptions common to the traditional trade theories - constant returns to scale, perfect competition, and absence of externalities - and to endogenize technology. Derivation of welfare results is complicated by a host of factors that must be taken into consideration. First, there is a degree of indeterminacy in the direction of trade. If one country had somehow developed an industry and established comparative advantage, then it would be difficult for another nation to establish a competitive position in that industry, irrespective of the availability of the technology or of an advantageous position in resources. Second, it raises the question of how the economy achieved a strong position in the industry in the first place. If the firms there had developed the technology, they can claim the right to appropriate. Since innovation is a costly process, the firms have no incentive to invest in R&D unless they are assured that the profits from the new product or process will not be quickly eroded by new competitors. Schumpeter (1942) and Arrow (1962) pioneered the study of the relation between market structure and the incentives to develop new technologies. Third, if firms in an economy are able to appropriate technologies

338

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

developed in another country and so affect the comparative advantage between the two economies, then the process becomes a source of international friction. It is worth remembering that this problem existed long before the much-heralded globalization of the technology. In the early years of industrialization, the Germans criticized the English and were criticized in turn. Recently, Japan has been accused of systematically transferring Western technology to boost her competitive position. Finally, the recognition that the rate of generation and transfer of technologies affect national welfare raises the question of appropriate public policies to enhance the degree of technological competitiveness. One view is that innovation is the product of individual inspiration and is influenced by cultural, religious, and social factors as well as by rational considerations. A recent survey (Mokyr, 1990, pp. 151-92) listed the following determinants of the propensity to invent: life expectancy, nutrition, willingness to bear risks, geographical environment, path dependency, labor cost, science and technology, religion, values, property rights, resistance to innovation, politics and the state, war, openness to new information, and demographic factors. But many innovations and definitely their commercial utilization are products of nurturing. If this is true then there are opportunities for organizational innovations that would make the nurturing process more effective and influence the directions of change in comparative advantage to the benefit of a nation. This is the crux of the controversies on the role of industrial and strategic trade policies. In the next section we examine some specific issues on the role of these policies in Japan and in the United States and in the last section will examine analytical models that throw light on our understanding of technological competition. 2

Lessons from the Japanese use of Western technology

Modern economic theory provides a rigorous foundation for the insight of Adam Smith about the efficiency of the invisible hand. In the Western world it put any attempt to interfere with the market on the defensive; policy-makers scrambled to find an economic justification (like the presence of externalities) or a political justification (like self-sufficiency in defense industries) and even then they were challenged on their ability to determine the appropriate degree of intervention with the free market system. Accordingly the role of government in the promotion of knowledge was confined to the support of fundamental research. Because of the lack of appropriability of basic research, it has to be undertaken by universities and research institutes supported by the government. But profit-making

Japanese competitiveness

339

organizations have an interest in applying knowledge to the development of new processes and products and the role of government in this process is confined to providing the necessary legal institutions. The legal structure should permit the firms to earn enough profit to provide the needed incentive to undertake R&D without conferring socially undesirable monopoly power. Yet such arrangements to protect intellectual rights made many newly independent nations in Asia and Africa suspicious of all voluntary channels of technology transfer. They fear that the Western firms will try to perpetuate their technological superiority by transferring only obsolete technologies. Further these countries charge that the licensing fees are generally excessive and that they are made to bear the overheads for developing products or processes that have no value to them.1 It is also claimed that, even if Western firms are willing to make the technology available to the emerging nations, they are not appropriate for them because of differing factor resources and costs. There is also the fear that multinationals will undermine newly won sovereignty just as the trading companies in the colonial era did. In general many developing nations showed hostility to commercial transfer of Western technologies and adopted industrial policies that inhibited foreign technological collaboration in many areas.2 Aharoni (1991, p. 86) notes that the licensing fees are minuscule compared with the total cost of overcoming technological barriers and achieving technological transfer. Other costs usually not taken into consideration include the underutilization of the equipment, delivery mix-ups, and paying excessive prices because of the import substitution program. He argues (1990, p. 96) that an efficient technology strategy involves first and foremost the development of technological and vocational education and next the adoption of an intensive search for global technologies to be transferred. Japan, according to him, scored well on both these approaches. He also points out that the transaction costs of transferring technology are less for multinational firms. The discussion shows that the rate of innovation within a nation and its willingness to absorb foreign technologies depend on a complex of cultural, economic, and political factors. Earlier writers like Gerschenkron (1962) assumed that the greater is the backlog of innovations that a country can take from another and the larger the gap between economic potential and actualities of the economy, the greater is the speed with which it can go through the development process. But the above discussion shows that the rate can be crucially affected by the policies adopted by the country. Japan has historically welcomed foreign cultural influences, first from China and then from the West. In an historical study, Sugimoto and Swain

340

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

(1989) divide the thirteen centuries from 600 to 1894 A.D. into three periods. In the first period (600-894), direct contact with China was established through kentoshi missions and this was followed by the establishment of the University, the Institute of Divination, the Institute of Medicine, and the provincial colleges under the ritsuryo system. It was followed by a period of seclusion (894-1401). In this period the University ceased to be a means for appointment to higher administrative offices and was not rebuilt after the fire. Under the new arrangement which included establishment of houses of learning based on hereditary professorships, the objective of learning became the preservation of the earlier knowledge in a formalistic way. The final phase began with the Tally Trade Agreement of 1401. The rise of the castle towns under the daimyo (local lords) led to a pragmatic culture in the towns which encouraged secularization and specialization and to the adoption of innovative techniques learned through the importation of books from China. By the middle of the seventeenth century, there was not only a flowering of Confucian scholarship but also the rise of indigenous astronomy and mathematics. But the pedantic Confucian tradition lost its vitality within a century and the need to reform the calendar led to the reintroduction of Western learning, particularly through the import of astronomy books translated into Chinese by the Jesuit priests. With the Meiji restoration of 1868, Chinese learning was replaced by Western learning in the spirit of the Charter Oath, that knowledge must be sought throughout the world (Masao, 1985, p. 372). Assimilation of Western technology was expedited by sending students abroad and inviting scholars from America, England, and Germany. English replaced Chinese as the language for the transmission of foreign knowledge into the country. The Western science at the end of the nineteenth century could still be explored by individual scholars and the linguistic unity of Japan enabled the few Japanese scholars who went abroad or foreign scholars who came to Japan to effectively disseminate the knowledge. But there was criticism that the Japanese strictly viewed science and technology as a machine that delivered results and did not absorb the other liberalizing aspects of Western culture (Masao, 1985, pp. 386-8). Still, Japan managed by the end of the nineteenth century to establish an educational system; scientific societies were formed and publications incorporating original research were produced. The attempt by Japan in the post-war period to catch up with Western technology through the use of industrial policy has been widely discussed. One of the important instruments of governmental policy was the direct control of a large share of national savings in the 1950s and early 1960s by the Ministry of Finance; this permitted it to ration credit. Another

Japanese competitiveness

341

instrument at that time was the power that the Ministry of Trade and Industry (MITI) had to allocate foreign exchange selectively among the firms. In the 1960s, industrial policies became more selective with the liberalization of foreign trade and exchange. The emphasis now was to replace the labor-intensive imports with capital-intensive ones. In the same period, Japanese corporations began to play an active role in the development of the computer and information technology industries. The growth of this industry under the guidance of MITI and NTT (Nippon Telegraph and Telephone) is analyzed by Fransman (1990). The responsibility of the government to provide a technological infrastructure in which the individual firms can function is widely accepted (Tassey, 1991). As long as the economy is in the process of catching up, the functional responsibility of the infrastructure is to identify technologies that can be transferred and to provide an institutional basis that will reduce the cost of transfer; in this sense the Gerschenkron thesis is valid. Fransman (1990) has undertaken a detailed analysis of the efforts by the Electrotechnical Laboratory (ETL) of MITI and Electrical Communications Laboratory (ECL) of NTT in providing resources that encouraged individual firms to pursue technological development that led to the establishment of a viable computer and communications industry in Japan. He points out (1990, pp. 277-80) that the Japanese companies are as concerned as the Western ones about information leakage. This concern is reflected in the greater willingness to participate in efforts by ETL to coordinate research done by various firms in the in-house laboratories of the companies than in joint research laboratories. But the government was successful in achieving greater diffusion of the results of the programs by including weaker firms that the industry leaders would have preferred to exclude. Japan also used government procurement as a means to achieve technological targets. The four companies constituting the Den-den family (NEC, Fujitsu, Hitachi, and Oki) have benefited from the research guidance from NTT and an assured market that NTT provided for their products. But once the catching up process is over, the uncertainties facing the policy-maker and the firms become much more complex. It is now necessary to increase the scope and depth of the technology infrastructure (Tassey, 1991, p. 345). Fransman (1990, p. 273) argues that in the late 1970s the emphasis both by the companies and the government agencies shifted to oriented basic research. Indicators so far show that Japan has been able to meet the challenge of being a technology leader instead of a follower. Business Week (June 15, 1991, p. 46) considers the relative standing in technology as well as the direction of change for twelve emerging technologies. In advanced semiconductor devices, high-in tensity data

342

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

storage, and optoelectronics, the United States and Japan are even and the position is not changing. In advanced materials, digital imaging, and superconductors, they are even in their current position but the trend is against the US. In artificial intelligence, flexible computer integrated manufacturing, high performance computing, and medical devices and diagnostics, the US is able to sustain its leadership. Finally in biotechnology and sensor technology, US leadership is being eroded. Frame and Narin (1990) examine the US and Japanese company patents issued in the United States in the two periods 1976-80 and 1981-5. This provides them with a consistent data set to examine the trend in patenting and the patent citations by the patent examiners provide a measure of the importance of the patents. The ratio of patents taken in a year by US companies to that taken by Japanese companies fell from 5.7 to 2.5 in that period. This is valid across most industrial classifications and is particularly valid in areas with the most visible commercial applications. Frame and Narin (1990, pp. 454-5) found that the ratio of 1983 Japanese patents which are highly cited is 1.17 times the ratio of worldwide 1983 patents which are highly cited. They take this to indicate that quality patents exceed what should be expected from the general level of patenting activity by 17 percent. But the US policy toward technological infrastructure is also changing. Here as in other countries committed to the free-market economy, the favored position used to be that the government should confine its involvement to provision of basic education and support of fundamental research. The US university system has been very successful in training scientists and engineers and in graduate education and research. There is much concern that the fragmented school system, administered by the local governments, does not have the curricular cohesion and quality of the European and Japanese systems in providing trained personnel to the increasingly technologically oriented industries. Another component of the infrastructure is the national laboratories formed under the Atomic Energy Act of 1946. While the individual functions of these laboratories varied, they all sought to concentrate in the high-risk, long-term research and development (Schriesheim, 1990-1, p. 52). Faced with the challenge from foreign technological competition, the Congress passed in the 1980s a series of acts to improve the technology transfer to industries. The hands-off policy that the results of research by these institutions should be in the public domain and available to all users, was replaced by one that permitted directors of national laboratories to enter into cooperative research and development agreements with private firms and to obtain title to and license intellectual property. This was in recognition of the fact that technology transfer from the laboratory to the

Japanese competitiveness

343

market place is costly and no firm will have the incentive to make this kind of investment unless assured of a right to appropriate profits. The 1984 National Cooperative Research Act addresses the monopoly question by permitting the formation of industry-level research consortia and the 1986 Technology Transfer Act allowed Federal government scientists to patent technology with commercial potential. The 1988 Omnibus Trade and Competitiveness Act provided for many additional institutions to strengthen the technology infrastructure (Tassey 1991, pp. 350-2; Schriesheim, 1990-1, pp. 53-4). American firms are also moving away from blind faith in mass production and adopting sophisticated manufacturing processes and inventory control techniques. The manufacturing productivity in the 1980s is higher than in the previous decade and the restructuring of the service sector in the 1990s will contribute to higher productivity growth for the whole economy. The technology policy in any nation must face some basic set of issues. It involves the degree of uncertainty that any R&D program faces; the extent of this uncertainty depends on whether a country is a technological follower or leader. It involves meeting the cost of commercializing the innovations. Firms will not have the incentive to incur these costs and face the uncertainties unless they have the means to appropriate the profits from successful inventions; an appropriate legal structure must protect intellectual property rights without undermining the competitiveness of the market. It is true that the Western nations in the period of industrialization, the newly independent nations of Asia and Africa, and the Pacific Basin nations each adopted their individualistic approaches to solutions to these problems. Some like Japan have an enviable track record; others failed dismally. Does this mean that the technology policy of each nation is a reflection of the cultural factors that are unique to it and not reproducible elsewhere? How different is a question that can be debated forever. But a case can be made that technology is converging faster than in previous centuries as the nations observe the successes and failures of each other. 3

Role of interventionist policies in strategic trade

This section is divided into two subsections. In subsection 3.1, we discuss the factors which are not consistent with traditional theories but play an important role in explaining two-way trade (the exchanges of similar goods) which is a large and growing part of the current trade among industrialized countries. In subsection 3.2, we discuss four strategic trade

344

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

models developed by Spencer and Brander (1983), Beath et al. (1989), Sato and Tsutsui (1987), and Sato and Tsutsui (1992) in order to deepen our insight into the complex issues of competitiveness and trade policies. In the first model, we discuss the role of R&D subsidy in a highly stylized framework, focusing on how R&D subsidy changes the outcome of trade under duopolistic competition. In the second model, we investigate the incentive for R&D investment and the role of R&D policies whenfirmsare engaged in a risky R&D race to be the first to introduce new technology. In the third model, R&D is thought of as a continuous process and a distinction is made between basic knowledge and applied knowledge. The firms may differ in their efficiency of applied knowledge production. Given the information structures of the firms, we study how the diffusion of basic knowledge and the difference in efficiency of applied knowledge production affect the outcome of trade. The last model takes up an issue of how the success of an interventionist trade policy is influenced by the relationship between government and the private sector. Consider that the interventionist trade policy is aimed at assisting a domestic firm that is an entrant to the international market that is monopolized by a foreign incumbent. As a newcomer, the domestic firm is subject to technical uncertainty. The similarity between the domestic firm's forecast of the technology uncertainty and the domestic government's is considered to measure the degree of consensus between them. We study how the effectiveness of a production subsidy given to the domestic firm is affected by the degree of consensus. 3.1

Departure from traditional theories

It is not difficult to see that non-interventionist trade policy has its foundations in one of the important results in economics. That is, perfect competition is efficients or a system of markets, when each market is perfectly competitive, can achieve the most (Pareto) efficient allocation of resources. There are a number of conditions which have to be met for a perfectly competitive market to prevail. They include the access to relevant information and the absence of externalities and public goods. Further, each firm is required to take prices as exogenously given by the market, believing that it has no influence on market prices so that it can sell whatever amount it wants to. Furthermore, free entry and free exit are required so that no excess profits will be reaped by firms. The assumption of perfect competition and its derivative that markets do not deviate from it very much have been the core assumptions used in most of traditional economic analysis. Of course, no one denies that such assumptions remain a good approximation for some markets and industries.

Japanese competitiveness

345

However, as noted before, the recent pattern of international trade makes it increasingly difficult to accept them as appropriate working assumptions. A large and growing part of current trade, as evidenced by the extensive two-way trade among industrialized nations, seems to be generated by factors which are not quite consistent with the conditions of perfect competition. Among them are the advantages of large-scale production, the advantages of cumulative learning and experience, the temporary advantages created by technological innovation, and imperfect competition. These factors are important characteristics of many manufacturing industries and are easily found in the real world. It is instructive as well as informative to consider how the incorporation of these factors affects our understanding of trade. First, we need to take into account a dynamic aspect of trade. The advantages of cumulative learning and experience, and the advantages generated by technological innovation not only depict how the cost structure of a firm changes over time but also make it possible that the firm can actively influence its cost structure over time through some kinds of economic activities. We consider R&D one of the most important among those activities. The incorporation of a dynamic dimension into trade makes a stark contrast with the traditional static view on this matter, and consequently, at least in the short run, there arises the possibility that firms earn excess profits above the normal return. But the presence of such excess profits need not be regarded as a manifestation of inefficient resource allocation since it can assume a new role of financing R&D to further cost reduction in the future - a view that was vigorously advanced by Schumpeter (1942). Second, the consideration of imperfect competition opens up the possibility that firms can act strategically. Under imperfect competition, an individual firm faces a few identifiable rivals in the market or industry and no longer acts as a price-taker because it knows what it does affects the market price and thus the behavior of rival firms in the same market. Thus, a firm can act strategically in the sense that it takes into account how its behavior affects the behavior of its rivals and how the behavior of the rivals affects its behavior. The incorporation of strategic interaction makes it necessary for us to adopt the game theoretic approach as in industrial organization in which oligopoly - an intermediate case between perfect competition and pure monopoly - is a main subject. A type of game is classified in a number of ways, depending on what aspect of the game's structure one looks at. For example, static versus sequential games; games with perfect information versus games with imperfect or incomplete information; continuous-time versus discrete-time

346

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

games; deterministic versus stochastic games. In a static game, every player (the term for a decision-maker in the game theory) moves simultaneously and only once, while in a sequential game, some or all players are allowed to move sequentially or more than once in the course of play. In a game with perfect information, every player can perfectly observe all of the past moves including chance moves, while in a game with imperfect or incomplete information, some or all players cannot. In a continuous-time game, moves can be made continuously in time, while in a discrete-time game, moves can only be made discretely in time. In a deterministic game, no random element is present in the relationship between moves and payoffs, while in a stochastic game, there are some elements which affect players' payoffs randomly. The choice of particular type of game, of course, depends on what the researcher regards as the most important element in the strategic interaction under investigation. Game theory has proved to be a very powerful tool in the analysis of strategic interaction, being applied to many areas of economic studies including industrial organization, macroeconomics, and resource economics. However, it is worth noting that the game theoretic approach is not without problems. One problem is concerned with the limitations of game theory. As Binmore (1990) discusses, the standard assumptions used in game theory, regarding common knowledge and the rationality of players, have difficulties in their relevance to reality, yet play a crucial role in analyzing games. Another problem is concerned with how to model strategic interaction. It matters greatly what elements one incorporates in the game. Adding new elements such as sequential moves and imperfect or incomplete information makes the game more realistic, but it can give to results which are significantly different from those without them. 3 As a result, great care has to be taken in modeling strategic interactions and interpreting the results of the game. 3.2

Strategic trade models

3.2.1

R&D subsidy

We first discuss the model of Spencer and Brander (1983).4 This model illustrates the important and interesting role of R&D subsidy in strategic international trade in the simplest possible framework. Suppose that there are two firms, one domestic and one foreign, producing a homogeneous product and selling all of their output in a world market which is entirely contained in other countries. Note that, with this assumption, we can equate the domestic firm's profits to the domestic country's welfare, and hence the objective of the domestic government is

Japanese competitiveness

?>A1

reduced simply to promote the domestic firm's profits, assuming that the domestic government is concerned only with the national welfare. With only two firms in the market, each firm acts strategically, recognizing that its profits significantly depend on what its rival does - the emergence of strategic interaction. To proceed with the analysis, more has to be specified. It is supposed that output is a firm's choice or strategic variable. In this case, the situation facing the firms is called a quantitysetting (Cournot) game. The consideration of output as a strategic variable, however, is not the only way to model the strategic situation. It may be appropriate if quantities are set by physical capacity constraints but inappropriate if firms can choose prices as their strategic variables. It is also supposed that the firms set their output once and for all. This amounts to neglecting the possibility that the firm changes its output at a future time and thus any possibility that the firm can induce collusion with its rival by means of future retaliation. Finally, it is assumed that both firms make their output decision simultaneously. Under these assumptions, we consider our quantity-setting game. As Brander (1986) argues, each firm knows that it could earn greater profits if it could persuade or force its rival to cut back on output, since the output contraction of its rival firm raises the market price and thus the firm's profits even without any change in the firm's output. So the firm might try to induce the contraction of output or for that matter the exit of its rival from the market by threatening to produce a large output. However, it is difficult to see why one firm, intimidated by such a threat, should be forced to contract the output, since the other firm is well aware that, by simply matching the increase, it can induce a price war - the outcome which is not in the best interest of the threatening firm. This argument shows that it is necessary in equilibrium that no aggressive threat by one firm should be believed by the other. In the terminology of game theory, it implies that we need to consider a Nash equilibrium of the quantity-setting game. Formally, Nash equilibrium requires that, at this equilibrium, a unilateral deviation of the firm does not strictly improve its profits. Now we modify the quantity-setting game so that both firms are engaged in R&D as an effort to improve their production efficiency. Suppose that only the domestic firm succeeds and that the successful R&D makes it possible for the costs of producing an additional unit of output to be substantially reduced (i.e., lowered marginal cost). Suppose further that the foreign firm knows its rival's success. We are interested in how the Nash equilibrium will be affected by this. It is straightforward to see that it is in the best interest of the domestic firm to expand its output. In fact, given the output of the foreign firm, the marginal profits, the difference between the additional revenue from selling an additional unit of output (i.e., marginal

348

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

revenue) and the marginal cost, are now positive for the domesticfirmdue to the lowered marginal cost. This gives the domestic firm an incentive to increase its output. The process of output expansion continues as long as the domestic firm's marginal profits are positive. The next question is how the foreign firm reacts to this output expansion. For this firm, the central issue is whether this output expansion constitutes a non-credible threat. However, knowing the success of the domestic firm and thus its underlying incentive, the output expansion by the domestic firm is no longer mistaken as a non-credible threat to the foreign firm. The domestic firm's success induces the foreign firm to contract it output. To see this, suppose on the contrary that the foreign firm does not respond to the output expansion. Since the output expansion increases total output, it reduces both firms' marginal revenue. Because the foreign firm's marginal costs remain the same as before and because the foreign firm's marginal profits were zero before the output expansion, the foreign firm's marginal profits become negative after the output expansion. This gives the foreign firm an incentive to contract its output. Combined with these two effects, we can conclude that the domestic firm produces more and the foreign firm produces less in the new Nash equilibrium than in the original Nash equilibrium. Suppose now that both firms are not engaged in R&D originally, but the domestic government can always entice the domestic firm to engage in R&D by providing R&D subsidy. Moreover suppose that there is no uncertainty in the link between R&D subsidy and the outcome of R&D, so that the domestic government can make it possible to reduce the domestic firm's marginal cost through the R&D subsidy. There are two effects of the R&D subsidy. One effect is the cost saving but in effect is just a transfer from the domestic government to the domestic firm. Thus, this effect has no impact on national welfare. But there is another effect to be taken into account. The R&D subsidy, in the eye of the foreign firm, works exactly the same as the domestic firm's discovery of producing output at lower marginal cost.5 Hence the expansion of output by the domestic firm is credible to the foreign firm.6 As a result, the domestic firm earns more profits and the foreign firm earns less profits. The net effect of the R&D subsidy is that the profits of the domestic firm rise more than the amount of the R&D subsidy - the phenomenon known as profit shifting. 3.2.2

Re^D competition and uncertainty

We now consider the model of Beath etal. (1989). This model sheds a light on the nature of R&D investment and its relationship to the success of R&D. In particular, it focuses on the two important factors which are

Japanese competitiveness

349

absent in Spencer and Brander (1983). The first is about the risky character of R&D investment. In Spencer and Brander (1983), there is no qualitative difference between R&D investment and ordinary capital investment. The second is that firms are frequently engaged in an R&D race to be the first to introduce some new technology or product. These two factors significantly affect the incentive of individual firms to participate in R&D investment and hence the role of R&D subsidy. Beath et al. (1989) examine the relation between R&D subsidy and the outcome of the R&D race in the presence of those two factors. The model distinguishes two types of the firm's R&D investment incentives, competitive and non-competitive incentives. The competitive incentive is defined to be the rate of R&D investment chosen by a firm when its rival firm chooses the infinite rate of R&D investment. It measures how much R&D investment a firm is willing to make when its rival firm tries to innovate successfully instantaneously. On the other hand, the non-competitive incentive is defined to be the rate of R&D investment chosen by a firm when its rival firm chooses the zero rate of R&D investment, measuring how much R&D investment a firm is ready to make when its rival firm is not engaged in the R&D race. A determinant of the relative magnitude of these incentives is found to be the ease of imitation which is defined to be the difference between a firm's profits if it innovates before its rival, and the profits it would make if its rival innovated first. If the ease of imitation is very low (for example, the new invention is protected by a highly effective patent), then the competitive incentive is greater than the non-competitive one. In this case, Beath et al. (1989) find that a firm's R&D investment rate is positively related to its rival firm's and that, given the firm's R&D investment rate, the firm prefers its rival firm to do less R&D because less R&D by the rival firm enhances the firm's chance of becoming the first to invent and thus secure the relatively high benefits of the new invention. On the other hand, if the ease of imitation is very high (for example, the new invention is not patentable), then the non-competitive incentive outweighs the competitive one. It is shown that the firm is motivated to reduce R&D investment as its rival firm increases its R&D investment and that, the firm prefers its rival firm to do more R&D because this makes it possible for the firm to reap the relatively high benefits of being an imitator by letting the rival firm bear the costs of the new invention. The characterization of a Nash equilibrium is fairly straightforward. Suppose that the competitive incentive dominates the non-competitive one for both firms. Then both firms' R&D reaction curves are upward sloping and the outcome of overinvestment in R&D emerges in the Nash equilibrium. Subsidizing the domestic firm in this case will obviously lead

350

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

to the expansion of the domestic firm's R&D investment. But it also leads to the expansion of the foreign firm's R&D investment instead of to the contraction of the foreign firm's R&D investment. The reason is that the foreign firm tries not to be outcompeted in the R&D race. Since the expansion of both firms' R&D investment will reduce both firms' (expected) profits, there arises no profit shifting. Clearly taxing the domestic firm is called for to improve the domestic welfare in this context. Suppose now that the non-competitive incentive dominates the competitive one for both firms. Then both firms' R&D reaction curves are downward sloping, so that the outcome of underinvestment in R&D appears in the Nash equilibrium. The domestic government is motivated to let the foreign firm do more R&D and bear the costs of the new invention so that the domestic firm will be able to enjoy the benefits of being an imitator. This is not, however, achievable by subsidizing the domestic firm. We notice that subsidizing the domestic firm will lead to the expansion of the domestic firm's R&D investment and to the contraction of the foreign firm's R&D investment. Unlike Spencer and Brander (1983), these changes in the firms' R&D investment will not necessarily increase the domestic firm's (expected) profits, because each firm prefers to let the other firm bear the costs of the new invention. In fact, subsidizing the domestic firm allows the foreign firm to relax its R&D and effectively lets the domestic firm bear the costs. Hence subsidizing the domestic firm is not necessarily welfareimproving. Finally, Beath et al. (1989) point out that subsidizing the domestic firm is potentially beneficial for the domestic country only when two countries differ in their ease of imitation. 3.2.3

Diffusion of technology

We consider the model of Sato and Tsutsui (1987) (see also Sato, 1988). They depart from Spencer and Brander (1983) and Beath et al. (1989) by taking the view that a firm's long-run competitiveness is determined by incremental improvements in production efficiency rather than by one-time gains brought about by a single successful new invention. Consequently, in the model, R&D activities are thought of as the process of creating new knowledge and the creation of new knowledge is supposed to take place continuously. There are two important characteristics in the model that are absent in the previous two models. First, with newly created knowledge, the outcome of R&D investment has all the complications associated with information goods. Among them, the model focuses on spillovers of knowledge. The spillovers of new knowledge benefit not only the producer of the new knowledge but also those who are not directly engaged in that

Japanese competitiveness

351

knowledge production and hence affect the firm's incentive to do R&D investment, creating a free-rider problem. It is to be noted that the model considers the spillovers in the context of continuous knowledge production. Second, two types of knowledge, basic and applied, are considered. The distinction between them is made based on how easily knowledge can be diffused. The introduction of two types of knowledge is motivated by capturing the fact that the government's R&D subsidy is often concentrated on the technology areas which have the broadest social applicability (i.e., the greatest spillover effect) and the fact that, as the development of video cassette recorders for household use illustrates brilliantly, the original idea of VCR (basic knowledge) is quite different from the production technology (applied knowledge) that makes it possible to produce VCR inexpensively. The firms are engaged in both basic and applied knowledge production. Basic knowledge is characterized as having a strong tendency to spill over, while applied knowledge, regarded more as firm specific knowledge, has a weak tendency to spill over. The production of knowledge as well as output is assumed to take place continuously. In order to focus on the impact of spillover effect of the basic knowledge, it is assumed that only the domestic firm is engaged in the production of the basic knowledge. But the model allows for the possibility that the costs of the basic knowledge production are shared by both the domestic and foreign firms and for the possibility that the efficiency of producing the applied knowledge may vary between the firms. These assumptions in the model are intended to reflect the reality that some firms tend to excel relatively in basic knowledge production while others are known to excel relatively in applied knowledge production. Sato and Tsutsui (1987) examine how the long-run competitiveness of firms is determined when the firms differ in their abilities to produce applied and basic knowledge and when basic knowledge spills over. Since firms are engaged in output and knowledge production, the instantaneous profits of a firm are equal to revenue minus costs of output production minus the costs of applied knowledge and basic knowledge production. Both the domestic and foreign firms maximize their discounted sum of instantaneous profits over an infinite planning horizon. In particular, the domestic firm maximizes /•oo

Jd

Jo

and the foreign firm maximizes

352

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI /*Q

=

Jo The notations are as follows, r is a common discount rate. qt is firm /'s output (/ = */,/). Ct is firm /'s effective marginal cost, which is inversely related to the level of the applied knowledge. B is the level of the basic knowledge of the domestic firm. u{ is the rate of change in Ct (i.e., ut = deceit), v is the rate of change in B (i.e., v = dBjdt). Since it is assumed for simplicity that the firms produce homogeneous output, P(-) denotes the inverse demand. Lastly, S(-,uh-) is the cost of reducing the effective marginal cost in terms of ui and T(*,p) is the costs of producing the basic knowledge in terms of v. It is assumed that S(',-,-) and T(v) a r e convex in their arguments, so that as the levels of applied knowledge and basic knowledge become higher, their incremental increase becomes increasingly more costly. The parameters <7, S, and £ are introduced to specify, respectively, how the foreign firm shares the costs of basic knowledge production with the domestic firm, to what extent basic knowledge diffuses from the domestic firm to the foreign firm, and how relatively efficient is the applied knowledge production of the foreign firm. The parameters a and 5 are required to satisfy 0 ^ a ^ 1 and 0 ^ 3 ^ 1, while the parameter £ needs only to be non-negative. Comments are useful for some values of these parameters in order to clarify their meanings. If (7 = 0, the foreign firm does not share the costs of basic knowledge production with the domestic firm; if a= 1/2, the foreign firm shares the costs equally with the domestic firm; if a = 1, the foreign firm pays all the costs of basic knowledge production. If £ = 0, there is no diffusion of basic knowledge from the domestic firm to the foreign firm; if £ = 1, the diffusion is perfect in the sense that the level of basic knowledge of the foreign firm is always equal to that of the domestic firm. If £< 1, then the efficiency of the applied knowledge production of the foreign firm is relatively higher than that of the domestic firm; if e = 1, then both firms have the same efficiency; if £ > 1 , the domestic firm has relatively higher efficiency. The technology game that is discussed in the model can be divided into two stages. In the first stage, the value of the parameter a can be determined. In the second stage, the firms choose their control variables over an infinite planning horizon. The domestic firm's control variables are qd (output), ud (the rate of change in the effective marginal cost), and v (the rate of change in the level of basic knowledge), while the foreign firm's control variables are qj- (output) and Uj- (the rate of change in the effective marginal cost). The parameter o is conceptually an endogenous variable and its value is likely to be determined as an outcome of the negotiation between the firms. However, the model treats a as exogenously given in order to concentrate

Japanese competitiveness

353

on the analysis of the second-stage technology game, which is already quite complicated. The technology game outlined above constitutes a differential game. Differential games are essentially the same as sequential games but the former are distinguished from the latter in three respects. First, the dynamics of a differential game is represented solely by so-called state variables. Note that, in our technology game, the state variables are the level of basic knowledge B and the firms' effective marginal costs Cd and Cf. Second, the evolution of state variables is governed by a system of differential or difference equations. Finally, the objective of each player is given by the integral or sum of instantaneous payoffs over a planning horizon. Despite these stylistic properties, the same equilibrium concept (i.e., Subgame perfect Nash equilibrium) is applied to solve a differential game.7 What remains to be determined is the information structure of a firm. The information structure specifies the extent to which each firm can observe the start variables. Perhaps, a word of warning is called for. Observation must not be confused with spillovers or diffusion. To clarify the matter, let us suppose that the effective marginal cost of the foreign firm is lower than that of the domestic firm and that the domestic firm observes the effective marginal cost of the foreign firm. In this case, the domestic firm simply knows what level of effective marginal cost the foreign firm has at each moment in time. Thus having such knowledge does not help reduce the effective marginal cost of the domestic firm down to that of the foreign firm. On the contrary, the diffusion of basic knowledge means that the foreign firm's basic knowledge automatically advances whenever the domestic firm's does. The importance of observation lies in the fact that a firm's observation determines what strategy of the firm constitutes a credible or non-credible strategy. If a firm observes the current levels of the state variables, then the firm has no incentive to sustain a non-credible strategy, because, when new information arrives, it is in the interest of the firm to adjust its action rather than to cling to a predetermined one. On the other hand, if a firm has no observation, then the firm cannot adjust its action but has to stick to a predetermined action. Hence the firm is able to sustain a strategy that otherwise may not be credible. It seems that, in the technology game, there arise some instances in which the firms may make only partial observations of the state variables. In particular, the effective marginal cost of a firm may not be observable to the other firm but only to the firm itself due to the firm-specific nature of the applied knowledge. Sato and Tsutsui (1987) consider two types of the information structure for each firm depending on whether or not the firm

354

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

observes the level of the other firm's effective marginal cost. Following the custom in the literature, we say that a firm has a closed-loop information structure when it has such an observation, while a firm has an open-loop information structure when it has not. The outcome of the technology game is measured by the steady-state market shares of the firms. It is shown to be dependent upon what type of information structure each firm has and how the parameters <7, S, and £ are configured. Given the fixed a, a ^ 1, the iso-share curve can be drawn in the 8-d parameter space (see figures 15.1 and 15.2). The curve has the following interpretations. If the combination of £ and S happens to be in the area above the iso-share curve, then the domestic firm (resp. the foreignfirm)has more (resp. less) than 50 percent of the market; if it happens to be on the iso-share curve, the domestic firm and the foreign firm have 50 percent of the market; if it happens to be in the area below the iso-share curve, the domestic firm (resp. the foreign firm) has less (resp. more) than 50 percent of the market. We now examine the symmetric cases in which the information structures of both firms belong to the same type. The iso-share curve in figure 15.1 (resp. figure 15.2) corresponds to the case where both firms have an open-loop (resp. closed-loop) information structure. Notice that both iso-share curves are upward sloping and are going through (1,1). Some remarks are in order. First, the upwardness of both iso-share curves shows the existence of a trade-off between the rate of basic knowledge diffusion and the efficiency of the foreign firm in applied knowledge production. Namely, in order to maintain the equal market share, the efficiency of the domestic firm in applied knowledge production has to increase relatively to counter the adverse effect caused by a high rate of basic knowledge diffusion. Second, both iso-share curves lie below the equal efficiency line (i.e., £ = 1). This shows that the domestic firm can capture more than 50 percent of the market regardless of the rate of basic knowledge diffusion, as long as the domestic firm has relatively higher efficiency in applied knowledge production (i.e., £>1). As a result, for the foreign firm to capture more than 50 percent of the market, it is necessary that the foreign firm has relatively higher efficiency in applied knowledge production (i.e., £<1) and the rate of basic knowledge diffusion is sufficiently high. Let us turn to the asymmetric cases in which the firms' information structures are different in type. Figure 15.3 shows the iso-share curve for the case where the domestic firm has a closed-loop information structure, while the foreign firm has an open-loop information structure. The corresponding iso-share curve is upward-sloping and going through (1,£), where £> 1. As in the symmetric cases, there exists a trade-off" between the rate of basic knowledge diffusion and the efficiency of the foreign firm in applied

Japanese competitiveness

355

Figure 15.1 Symmetric information structure with open loops: region above iso-share curve represents domestic share >50%

Figure 15.2 Symmetric information structure with closed loops: region above iso-share curve represents domestic share >50%

Figure 15.3 Asymmetric information structures: domestic firm has closed loop, foreign firm has open loop. Region above iso-share curve represents domestic share >50%

knowledge production. In this asymmetric case, however, the entire iso-share curve does not necessarily lie below the equal efficiency line but some portion of the iso-share curve lies above it. Hence there arise cases where even if the foreign firm (resp. the domestic firm) has relatively lower (resp. higher) efficiency in applied knowledge production, its market share

356

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

is more (resp. less) than 50 percent, when the rate of basic knowledge diffusion is sufficiently high. The reason is the following. Under these information structures, the foreign firm has to commit itself to a predetermined action but the domestic firm can adopt a so-called feedback strategy which permits adjustment of action in the course of play. In a non-strategic context, it is easily seen that the domestic firm is in a more advantageous position than the foreign firm because of the flexibility of a feedback strategy. However, in a strategic context, the domestic firm can be in a less advantageous position than the foreign firm. The foreign firm does not have to revise its action in response to a change in the domestic firm's action, while the domestic firm has to do so. The difference in the firms' responses gives the foreign firm a credible way to impose its action on the domestic firm. This leads to the expansion of the foreign firm's R&D and hence to the contraction of the domestic firm's R&D, as in Spencer and Brander (1983). As a result, the foreign firm, even with relatively lower efficiency in applied knowledge production, may be able to achieve the same level of applied knowledge production and thus 50 percent of the market. The other asymmetric case arises if the domestic firm has an open-loop information structure, while the foreign firm has a closed-loop information structure. As one can expect from the previous discussion, the iso-share curve is upward sloping but going through (1,2), where 8
Japanese competitiveness

357

the firms belong to the same type. It then follows that the value of 8 that gives the equal market share is 1 (i.e., both firms have the same efficiency in applied knowledge production). Suppose that subsidizing basic knowledge production reduces its costs. It can be shown that the effect of such subsidy, however, does not change the value of £ that warrants the same market share. In other words, subsidizing basic knowledge production does not increase the domestic firm's market share. The reason is that, since the diffusion is perfect, an increase in basic knowledge, generated by subsidy, benefits both firms equally. Suppose now that subsidizing applied knowledge production works in such a way that it effectively raises the value of £ (i.e., it reduces the foreign firm's efficiency in applied knowledge production). It is clear that such subsidy raises the domestic firm's market share. As a result, it can be concluded that subsidizing applied knowledge production is a more effective policy. Another issue is the control of the diffusion of basic knowledge and of the firms' observations of each other's effective marginal cost. We can draw two policy implications on this matter. First, from the existence of a trade-off between the foreign firm's efficiency in applied knowledge production and the rate of basic knowledge diffusion, it follows that curtailing the diffusion of basic knowledge, if feasible, constitutes a viable policy, even if the foreign firm's efficiency in applied knowledge production is sufficiently higher than the domestic firm's. Second, since the firm's observation of the other firm's effective marginal cost deprives it of an ability for commitment, it follows that a policy that facilitates the foreign firm's observation of the domestic firm's effective marginal costs turns out to be more shareincreasing than a policy that obstructs it. Here it is interesting to note that a difference in the nature of information between diffusion and observation makes such contrast of the two policies, i.e., the curtailment of diffusion vs. the promotion of observation. 3.2.4

Uncertainty and the role of consensus

We consider the model of Sato and Tsutsui (1992). The model is concerned with two prevailing views regarding the role of interventionist trade policies in improving the competitiveness of home industries in international markets. 9 Inspired in part by successes such as the Ministry of International Trade and Industry's (MITI) development of high-capacity memory chips in the 1970s, proponents argue for the virtue of close cooperation between the government and the private sector, which may include the implementation of interventionist trade policies. Opponents, on

358

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

the other hand, stress inherent difficulties in implementing interventionist trade policies efficiently. In particular, they point to the lack of close cooperation between the government and the private sector in the United States and stress that it may not be possible for the government to pick a winner given so many alternatives and so much uncertainty. As we saw in subsection 3.2.1, interventionist trade policies can be effective not only in promoting the exports of a domestic firm but also in improving domestic national welfare. But an interesting issue arises as to whether the manipulation of the strategic effect is always possible with interventionist trade policies. Although the issue has been previously studied, it does not seem to have been fully analyzed in the presence of uncertainty.10 Taking uncertainty into account in the analysis is important in at least two respects. First, as typified by the debate surrounding the development of HDTV, the very presence of uncertainty seems to be largely responsible for creating the two opposite views, since the difference reflects primarily their views on how to deal with the uncertainty.11 It is worth noting that in one view, because of uncertainty, interventionist trade policies are often called for and perhaps justified if they are intended to assist the domestic firms facing the uncertainty, while in the other, because of uncertainty, the strategic effects of interventionist trade policies are hard to predict and could be counterproductive. The introduction of uncertainty makes it possible to investigate these two opposite views. Secondly, the incorporation of uncertainty provides insights into the issue of how important the "consensus" between a government and a private sector is in determining the success or failure of interventionist trade policies. In fact, facing the uncertainty, the government and the private sector are likely to have assessments (or forecasts or private information) of the uncertainty. The similarity of their assessments provides a convenient way to measure the degree of "consensus" on the likely outcome of the uncertainty between the government and the private sector.12 According to this view, it seems too sensible to assert that the government and the private sector have a high degree of consensus on the uncertainty if these assessments are highly correlated. This notion of consensus may also be useful to classify types of relations between the government and the private sector. For instance, in a country like Japan where the society is consensus oriented, it may well be that the degree of consensus is relatively high, while in a country like the United States where taking initiatives and independent thinking are highly valued, it may well be that the degree of consensus is relatively low. In this way, a country's orientation towards consensus can be incorporated into the analysis of interventionist trade policies. To investigate systematically the effectiveness of interventionist trade policies in the presence of uncertainty, Sato and Tsutsui (1992) consider an

Japanese competitiveness

359

international market where two firms, one domestic and one foreign, play a Cournot game. We assume that the production technology of the domestic firm is subject to uncertainty, while that of the foreign firm is not. The asymmetric nature of the uncertainty may arise if the foreign firm is an incumbent monopoly and the domestic firm is an entrant and if the technology is such that its mastery involves trial and error and/or requires many years of production experience, so that the domestic firm is technologically less proficient.13 Since the domestic firm faces the uncertainty, the domestic government assists the domestic firm with an interventionist trade policy. The model considers the policy of giving a production subsidy to the domestic firm. The subsidy is a specific type and is given by the domestic government before the two firms decide their outputs in order for it to generate the strategic effect. The game described above extends the trade game of Spencer and Brander (1983) by incorporating uncertainty. Note that there are two reasons why this extension is chosen. First, without foreign firms, there would be no incentive to employ interventionist trade policies in the model. Rather the incentive to do so arises solely from the fact that they generate the strategic effect. In this sense, the model allows us to focus on the impact of uncertainty on the strategic effect. Second, with no uncertainty, interventionist trade policies always produce an advantageous strategic effect. Hence the comparison between the results of this chapter with those of Spencer and Brander (1983) highlights how the presence of uncertainty alters the direction and the magnitude of the strategic effect. More specifically, the objective function of each firm is to maximize its profits denoted by Tli, i=d (domestic) and f (foreign), while that of the domestic government is to maximize its national welfare denoted by W. As with Spencer and Brander (1983), it is assumed that the domestic firm's output is only for export and thus not for domestic consumption. Hence the national welfare W is simply reduced to the domestic firm's profits Wd minus the costs of giving a production subsidy to the domestic firm. Then the objective functions of the firms and the domestic government are expressed as

,

d(z,JdJf)

= P(Jd +j f )Jd - Cd(Jd,9) + zyd,

n ' = nf(Jd>Jf)=P(Jd +Jf)Jf - cf(Jf), fi

d-

Cd(jd,9),

where P(j) is the inverse demand function of the international market, j t is firm /'s output, C{ is firm /'s cost function (/= d and f), 9 denotes a random variable representing uncertainty with E[0] = 0 and Var[0] = a, and % is the unit production subsidy given to the domestic firm. The firms as well as the government forecast 9 and their forecasts

360

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

constitute their private information regarding uncertainty. Let 9g denote the government's forecast of 9 and 9t firm /'s forecast of 0(7= d, f). It is assumed that (a) the forecast Oj is an unbiased estimate of 0(/ = g,d,f); (b) player/s forecast errors 9j-9 are independent of 0(J = g,d,f); (c) the forecast errors 9j-9 of the foreign firm are independent of the forecast errors 9g-9 and 9d-9 of the domestic government and firm; but the forecast errors of the domestic government need not be independent of the forecast errors of the domestic firm. Let »j denote the variance of the forecast errors Oj — 0(J = g,d,f) (i.e., rij = Var[0y-0]) and a the covariance between 9g-9 and 9d-9 (i.e., a = Cov[(0g-0), (0d-0)]). Note that from the definition of covariance, it follows that \(x\^^ngfid. Since the forecasts are unbiased estimates of uncertainty, the mean of uncertainty is zero, and a high value of 9 means low marginal cost for the domestic firm, it seems reasonable to divide the forecasts into two groups, depending on their signs. Specifically, for eachy = g, d, and f, if playery's forecast 9j is non-negative (i.e., 0y^O), then the forecast 9J is said to be optimistic but if player/' s forecast 9J is negative (i.e., 9} < 0), then the forecast 9j is said to be pessimistic. Moreover, since a measures how similar the opinions of the domestic government and the firm are regarding the uncertainty, it seems natural to interpret a as the degree of consensus concerning the uncertainty between the domestic government and the domestic firm. It is interesting to note that if a = >Jngnd, the forecasts of the domestic government and the firm are conditionally correlated perfectly and thus the degree of consensus is highest, while if a = 0, their forecasts are conditionally independent and therefore there is not consensus between them. On the other hand, if 0 < a < ^/ngnd, their forecasts are conditionally but not perfectly correlated, so there is a corresponding degree of consensus between them. Assuming linear functional forms for the inverse demand and cost functions, Sato and Tsutsui (1992) obtain a sequential Nash equilibrium for this strategic trade game within a linear class.14 More specifically, they look for Z = q{9g\jd = qd{^9d) andj/ / = ^ / f c 0 / ) such that

where the coefficients Ao, A^ A2, Bo, Bl9 B2, h, and x are to be determined by the equilibrium conditions.15 The signs of the coefficients Al9 A2, Bu B2, and x determine how the firms as well as the domestic government respond to their forecasts and the observation of the unit subsidy. In order to minimize the confusion in

Japanese competitiveness

361

describing the effect of ^ and 0t on the firms outputs (/ = d,f) and the effect of 6g on the unit subsidy, we introduce the following terminology. If Ax is positive (negative), we say that the domestic firm's output responds positively {negatively) to the unit subsidy %. Similarly, we characterize the response of the foreign firm to the unit subsidy. If A2 is positive (negative), we say that the domestic firm's output responds aggressively (counteractive ly) to its forecast 9d. Similarly, we characterize the responses of the foreign firm and the domestic government to their forecasts 9f and 9g. A positive (negative) response of a firm's output to the unit subsidy implies that the firm increases or decreases its output with increases or decreases in the unit subsidy. The implication of an aggressive (counteractive) response of a firm's output to its forecast is that the firm increases its output in response to an optimistic (pessimistic) forecast but decreases it to a pessimistic (optimistic) one. Similarly, if the response of the unit subsidy to 6g is aggressive (counteractive), then the domestic government raises the unit subsidy in response to an optimistic (pessimistic) forecast but lowers it to a pessimistic (optimistic) one. The trade game described above is shown to have two types of equilibrium, aggressive and counteractive, depending on how the domestic government's subsidy responds to its forecast of the uncertainty; i.e., the sign of x. In an aggressive (counteractive) equilibrium, the domestic government increases (decreases) the subsidy with optimistic forecasts; i.e., x > 0 ( < 0 ) . It is shown that in every aggressive equilibrium interventionist trade policy is effective since the subsidy increases the domestic firm's expected output relative to the non-intervention case, while in every counteractive equilibrium interventionist trade policy is detrimental since the subsidy decreases output relative to the non-intervention case. Sato and Tsutsui (1992) show the crucial role played by the degree of consensus a and the relative accuracy of the domestic government's forecast CO (defined by CO = ndjng) in determining what type of equilibrium exists and how effective interventionist trade policy is in each type of equilibrium. They show that a high degree of consensus acts like a double-edged sword. Having a high degree of consistency is definitely advantageous in interventionist trade policy if the relative accuracy of domestic government forecast is high; it unambiguously promotes the domestic firm's share in the international market. In contrast, having a high degree of consistency is definitely disadvantageous in interventionist trade policy if the relative accuracy of domestic government forecast is low; since it unambiguously reduces the domestic firm's share. The double-edged nature of a high degree of consensus highlights not only the costs but also the benefits of forging the consensus between the domestic government and firm. The costs reflect the fact that having a high

362

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

degree of consensus is counterproductive for risk diversification. In a risky environment, having a high degree of consensus increases the chance that the forecasts of the domestic government and firm are wrong. The increased chance of their having wrong forecasts makes the foreign firm's output respond less or differently to interventionist trade policy. This changed response of the foreign firm's output in turn reduces the effectiveness of interventionist trade policy. We consider that this reduction in effectiveness represents the risk-concentration effect of a high degree of consensus. On the contrary, the benefits reflect the fact that having a high degree of consensus increases the efficiency of interventionist trade policy by improving the coordination between the domestic government and the firm. That is, having a high degree of consensus increases the chance that the forecasts of the domestic government and the firm are similar. The increased chance of their having similar forecasts makes the domestic firm's output more responsive to interventionist trade policy. The increased responsiveness improves the efficiency of interventionist trade policy in two respects. First, the domestic government needs a lower unit subsidy to induce the same increase in the domestic firm's output. Second, the increased response of the domestic firm's output makes the foreign firm's output respond more negatively to the unit subsidy. These two phenomena increase the efficiency of interventionist trade policy. We consider that this increased efficiency represents the coordination effect of a high degree of consensus. The effectiveness of interventionist trade policy when the degree of consensus is high is determined by the interaction of the coordination effect and the risk-concentration effect. An important determinant of their magnitudes is the relative accuracy of the domestic government's forecast co. The preceding discussion suggests that if the relative accuracy (O is high (i.e., the forecast of the domestic government is more accurate than that of the domestic firm), the coordination effect is expected to dominate the risk-concentration effect. In fact, if (O is high, then the domestic firm wants to be more responsive to the interventionist trade policy, since it reflects the domestic government's forecast, which is more accurate than the domestic firm's. This increased responsiveness increases the coordination effect. As a result, the coordination effect dominates the risk-concentration effect and interventionist trade policy becomes effective. On the other hand, if the relative accuracy CD is low (i.e., the forecast of the domestic firm is more accurate than that of the domestic government), the risk-concentration effect is expected to dominate the coordination effect. The reason is as follows. If CO is low, then the domestic firm wants to be less responsive to interventionist trade policy, since it reflects the domestic government's forecast, which is less accurate than the domesticfirm's.The

Japanese competitiveness

363

decreased responsiveness decreases the coordination effect. Consequently, the risk-concentration effect dominates the coordination effect and thus interventionist trade policy becomes detrimental. The analysis also shows that the characterization of equilibrium is independent of how much or how little the foreign firm knows about uncertainty. In particular, this result holds if the foreign firm knows about uncertainty perfectly (i.e., nj- = 0) or if the foreign firm is not engaged in forecasting uncertainty (i.e., n^= oo). These two extreme cases show that for interventionist trade policy to be effective or detrimental, what the domestic government and firm know about uncertainty is more important than what the foreign firm knows about uncertainty. The reason is as follows. First, the foreign firm's output decision depends not on the foreign firm's forecast but on its assessment of the domestic firm's output, which in turn depends only on how the domestic firm's output responds to its forecast and interventionist trade policy. Since interventionist trade policy depends only on the domestic government's forecast, it follows that the foreign firm's output depends only on the forecasts of the domestic government and the firm. It is perhaps worth noting the difference between these results and those of Spencer and Brander (1983). In the deterministic version where 6 = 0 identically, the domestic government's equilibrium subsidy is equal to ^ and the domestic firm's output and the foreign firm's output are equal to \ and ^, respectively. When only an aggressive equilibrium exists, the expected subsidy is less than in the deterministic case and hence the domestic (foreign) firm's expected output is less (greater) in this aggressive equilibrium than in the deterministic case. However, when both equilibria exist, the expected subsidy is greater in an aggressive equilibrium than in the deterministic case, so that the domestic (foreign) firm's expected output is greater (less) in this aggressive equilibrium than in the deterministic case. This difference in the level of the expected subsidy seems to stem from the fact that when two equilibria exist, the aggressive domestic government tries to distinguish itself from the counteractive government. The similar phenomenon that appears in the monetary policy game is known to be the reputational effect (see Backus and Driffill, 1985 and Mino and Tsutsui, 1990). The foreign firm's output responds negatively in every aggressive equilibrium, while it responds positively in every counteractive equilibrium. This implies that B1 < 0 in every aggressive equilibrium and Bl > 0 in every counteractive one. The response of the foreign firm's output to the unit subsidy is determined by two effects. One is a profit-shifting effect of the unit subsidy on the foreign firm's output as discussed in Spencer and Brander (1983). The profit-shifting effect, in response to a positive unit

364

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

subsidy, always induces the domestic firm to produce more and the foreign firm to produce less. In our trade game, however, there arises a signaling effect of the unit subsidy on the foreign firm's output. The signaling effect stems from the fact that the unit subsidy signals to the foreign firm what forecast the domestic government has. In fact, the observation of the unit subsidy allows the foreign firm to infer the domestic government's forecast 9g. This inference induces the foreign firm to revise its assessment of the contribution of the domestic firm's forecast on the domestic firm's output and hence affects the foreign firm's output. Sato and Tsutsui (1992) show that the signaling effect of the unit subsidy can reinforce or weaken the profit-shifting effect of the unit subsidy. They find that, when both equilibria exist, the signaling effect in an aggressive equilibrium reenforces the profit-shifting effect and thus B1<0, while, in all other cases, the signaling effect weakens the profit-shifting effect. The (total) strategic effect is given by the sum of these effects. In every counteractive equilibrium, the signaling effect weakens the profit-shifting effect to such a degree that the strategic effect becomes positive. On the contrary, when only an aggressive equilibrium exists, the signaling effect weakens the profit-shifting effect but the strategic effect remains negative. Finally, the study of the expected social welfare of the domestic country reveals that if the relative accuracy of the domestic government's forecast is the same or high (i.e., a>^l), then the expected social welfare is always greater in an aggressive equilibrium than in a counteractive one whenever both exist. On the contrary, (i) if the relative accuracy of the domestic government's forecast is low and the degree of consensus is high but not too high, then the expected social welfare is greater in a counteractive equilibrium than in an aggressive one; (ii) if the relative accuracy of the domestic government's forecast is low and the degree of consensus is low, then the expected social welfare is greater in an aggressive equilibrium than in a counteractive one. Note in (i) that the effectiveness of interventionist trade policy is not necessarily related to the expected social welfare, so that although interventionist trade policy is detrimental in a counteractive equilibrium, the expected social welfare can be higher in this equilibrium than in an aggressive equilibrium. Also it is interesting to note that, as long as the degree of consensus is low, the expected social welfare is greater in an aggressive equilibrium than in a counteractive one, regardless of the value of the relative accuracy of the domestic government's forecast. One economic interpretation is as follows. Suppose that when two equilibria exist, the domestic government chooses an equilibrium that generates the highest expected social welfare. Then to each configuration of the parameters there corresponds a unique equilibrium in our trade game. Moreover, if the relative accuracy of the domestic government's forecast is

Japanese competitiveness

365

low, then an aggressive equilibrium arises and interventionist trade policy is effective if the degree of consensus is low, but a counteractive equilibrium emerges and the interventionist trade policy is detrimental if the degree of consensus is high. This observation has an interesting implication: if the forecast of the domestic firm is more accurate than that of the domestic government, then having a low degree of consensus is more advantageous than having a high degree of consensus. This result is in stark contrast with the other result that if the forecast of the domestic firm is less accurate than that of the domestic government, a high degree of consensus is more advantageous than having a low degree of consensus. 4

Conclusion

The theory of technological competition is a new branch of economics but, apart from its infancy, it suffers from intellectual incompatibility with its progenitors. Until recently, the dominant paradigm of international trade was that of laissez-faire. In contrast, the theory of innovation recognizes the need to internalize the benefits from R&D investments; the possibility of market failure became widely recognized and the need for government intervention vehemently argued. Strategic trade theory recognizes that one of the sources of externalities that nullifies the assumptions of traditional trade theory is the incomplete appropriability of R&D. As a result, an acrimonious debate has arisen as to whether any nation has developed an advantageous position in international trade through manipulation of technology policy. Ergas (1987) divides the technology policies of six Western countries the United States, the United Kingdom, France, Germany, Switzerland, and Sweden - into two categories: "mission oriented" and "diffusion oriented." He relates the policies to two possible purposes, national sovereignty and provision of public goods. He argues that the technology policies of the US, UK, and France are mission oriented and have close links to national sovereignty goals. The decision process is concentrated and the projects which receive funding are few and biased towards industries in the early stages of the technology life-cycle. Further they are concentrated on radical innovations. Ergas considers Germany, Switzerland, and Sweden to be diffusionoriented countries where centralized policy-making is rather limited and funds are widely distributed to industry associations or cooperative research groups. These benefit a large number of small- and medium-sized firms. Moreover, these countries emphasize education push. The chemical and electric industries, which were highly concentrated, needed a highquality university system which in turn needed a superior school system,

366

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

and they lobbied for these objectives. In contrast, the mechanical industry was highly decentralized and lobbied for vocational training, product standardization, and cooperative research. The German-speaking countries developed a comprehensive system of vocational apprenticeship and skill certification. The government provided subsidies to firms to hire research scientists; the program was administered through industry organizations and had very precise eligibility rules to prevent arbitrary selection. Overall the diffusion-oriented policy leads to quick dissemination of incremental technologies but firms tend to be conservative in adopting radical alternatives. The Japanese system seems to be an amalgam of the two approaches. Sato (1988) notes that the government plays a dual role. In supporting a mission-oriented policy, it selects a small number of important and powerful companies in the industry so as to protect them from domestic and foreign competition. In the early stages, these are the only companies permitted to import foreign technologies and they are encouraged to develop a cooperative research program and received tax credits and subsidies. Once a technology is developed, the government encourages competition. The public school system is well known for its emphasis on science education at school levels and the universities graduate a high percentage of engineers. The debate whether government intervention is needed to correct market failure or would only lead to market failure is not confined to strategic trade theory alone. The probability is that the debate will not be resolved in the near future. But one has to be careful when inconclusive arguments are used to evaluate the performance of a nation or, even more, if they are used to compare two nations. Without proper care, any desirable conclusion can be derived depending on a careful marshaling of assumptions and data. Much of the debate on Japan has concentrated on whether MITI has superior signaling abilities in directing technology trade by firms and whether this requires a response from the Western nations. We seek to show that the channels of transfer of technology to her and her method of utilizing it differ at best only in degree from what prevails in other countries. This will not resolve the debate but hopefully it can bring some perspective to it. Notes 1 One of the sources of friction between the United States and countries like India is the unwillingness of the latter to recognize the patents of pharmaceutical companies. It is argued that these companies keep the prices of essential

Japanese competitiveness

2 3 4

5

6 7 8 9 10

11

12

13

14

367

life-saving products high to recover their research cost on expensive medical technologies that have no value to developing nations. See surveys by Khatkhate (1991, pp. 237-52) and Adedeji (1991, pp. 288—203). See Kreps and Wilson (1982), Milgrom and Roberts (1982) for example. Strategic trade theory has attracted much attention from many economists lately. Many issues have been investigated in international trade in the framework of strategic trade. The recent studies include Brander (1986), Brander and Spencer (1984, 1985), Clemenz (1990), Dixit (1984), Fung (1989), Krugman (1984), Levinsohn (1989), Meza (1989), Ono (1989), Tsutsui (1989), and Sato and Tsutsui (1991). This is only a partial list and is not meant to be inclusive. For this to happen, it is necessary that both firms make a full observation of the domestic government's R&D subsidy. Therefore, it is to be noted that if the foreign firm makes no such observation its reaction to the R&D subsidy cannot be the same. The domestic government's R&D subsidy essentially provides the domestic firm with first-mover advantages. See Tsutsui and Mino (1990) for the derivation of a subgame perfect Nash equilibrium in a differential game with an infinite horizon. If the foreign firm pays the entire costs of basic knowledge production (i.e., (T=l), then the iso-share curve coincides with the horizontal axis. Consider, for example, the fierce debate over the government role in developing High Definition Television (HDTV) in the United States. The following results are obtained using deterministic models. Grossman (1986) shows that, when the number of domestic firms is sufficiently large, the conclusion of Spencer and Brander (1983) does not hold, since the subsidies to the domestic producers intensify the competition among them and that offsets the gains from the reduced foreign output. Consider, on the contrary, that the outcome of interventionist trade policies is always manipulative to the advantage of a domestic country. Then, aside from the obvious concern that such policies are hardly justifiable in view of the rules of GATT (The General Agreement on Tariffs and Trade), there should be no differing opinions on their effectiveness. The issue of consensus is part of the much larger and more important issue of what role "close cooperation" between the government and the private sector plays in determining the effectiveness of interventionist trade policies. It is the widely accepted view that close cooperation is the key to the success of such policies in Japan. Consider, for instance, the situation where a newcomer from a less developed country enters the international market that has been monopolized by a firm from a more developed country. The specification used in Sato and Tsutsui (1992) is that P(j) = 2—j, CdUdfi) ~ 0- ~ Q) Jd> a n d CfUf) = Jf- Note that, in this specification, both firms have constant marginal costs and only the domestic firm's constant marginal cost is subject to the uncertainty. Also, note that the domestic firm's

368

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

marginal cost coincides with the foreign firm's when 6 = 0, so that the domestic firm's expected marginal cost and the foreign firm's marginal cost are the same and equal to unity (i.e., E[dCd(jd,6)ldJd] = E[l - 0 ] = 1 =dC f(jf)ldy f). 15 It goes without saying that linear equilibrium strategies are consistent with quadratic objective functions only when a class of distributions is such that the conditional expectations E[0\6g], E[0d\0g\, E[0f\Og], E[0\0g,0d], E[0f\6g,0d) and E[6d\6gy0j] satisfy a linearity property, so that all of these conditional expectations are linear in their argument(s). It is important to note that the linearity of these conditional expectations is satisfied for various classes of distributions. Ericson (1969) shows that if distributions are natural conjugates to one another, then in many distribution combinations the conditional expectations satisfy a linearity property. 16 The linear structures of the cost and the inverse demand functions and the linear property of the conditional expectations enable us to derive a sequential equilibrium within a linear class. This in turn makes it possible to analyze an equilibrium in great detail. See, for instance, Basar (1978), Gal-Or (1985, 1987), Li (1985), Novshek and Sonnenshein (1982), and Mino and Tsutsui (1990). References Adedeji, A. (1991), "National Economic Policies in Africa; Egypt, Kenya, and Nigeria as Case Studies," in D. Salvatore (ed.), National Economic Policies, West Port CT and Amsterdam: Greenwood Press and North-Holland. Aharoni, Y. (1991), "Education and Technology Transfer," in T. Agmon and M. A.V. Glinow (eds.), Technology Transfer in International Business, New York: Oxford University Press. Arrow, K. (1962), "Economic Welfare and the Allocation of Resources for Invention," in R.R. Nelson (ed.), The Kate and Direction of Inventive Activity, Princeton, NJ: Princeton University Press. Backus, D. and J. Driffill (1985), "Rational Expectations and Policy Credibility Following a Change in Regime," Review of Economic Studies, 52, 211—21. Basar, T. (1978), "Decentralized Multicriteria Optimization of Linear Stochastic Systems," IEEE Transactions on Automatic Control, 23, 233—43. Beath, J., Y. Katsoulacos, and D. Ulph (1989), "Strategic R&D Policy," Economic journal, 99, 74-83. Binmore, K. (1990), Essays on the Foundations of Game Theory, Oxford: Basil Blackwell. Brander, J.A. (1986), "Rationales for Strategic Trade and Industrial Policy," in P.R. Krugman (ed.), Strategic Trade Policy and the New International Economics, Cambridge, Mass.: The MIT Press. Brander, J.A. and B.J. Spencer (1984a), "Tariff Protection and Imperfect Competition," in H. Kierzkowski (ed.), Monopolistic Competition and International Trade, Oxford: Oxford University Press. (1984b), "Export Subsidies and International Market Share Rivalry," Journal of International Economics, 18, 83—100.

Japanese competitiveness

369

Business Week, June 15, 1990. Clemenz, G. (1990), "International R&D Competition and Trade Policy," Journalof International Economics, 28, 93-113. Dixit, A. (1984), "International Trade Policy for Oligopolistic Industries," Economic Journal, 94, 1-16. Ergas, H. (1987), "The Importance of Technology Policy", in P. Dasgupta and P. Stoneham (eds.), Economic Policy and Technological Performance, Cambridge: Cambridge University Press. Ericson, W.A. (1969), "A Note on the Posterior Mean of a Population Mean," Journal of the Royal Statistical Society, 31, 332-4. Fortune, December 2, 1991. Frame, J.D. and F. Narin (1990), "The United States, Japan and the Changing Technological Balance," Research Policy, 19, 447-55. Fransman, M. (1990), The Market and Beyond: Cooperation and Competition in Information Technology Development in the Japanese System, Cambridge: Cambridge University Press. Fung, K.C. (1989), "Tariffs, Quotas, and International Oligopoly," Oxford Economic Papers, 41, 749-57. Gal-Or, E. (1985), "Information Sharing in Oligopoly," Econometrica, 53, 329-43. (1987), "First Mover Disadvantages with Private Information," Review of Economic Studies, 54, 279-92. Gerschenkron, A. (1962), Economic Backwardness in Historical Perspective, Cambridge: The Belknap Press. Grossman, G.M. (1986), "Strategic Export Promotion: A Critique," in P.R. Krugman (ed.), Strategic Trade Policy and the New International Economics, Cambridge, Mass.: The MIT Press. Khatkhate, D. (1991), "National Economic Policies in India," in D. Salvatore (ed.), West Port CT and Amsterdam: Greenwood Press and North-Holland. Kreps, D. and R. Wilson (1982), "Reputation and Imperfect Information," Journal of Economic Theory, 27, 253-79. Krugman, P.R. (1984), "Import Protection as Export Promotion: International Competition in the Presence of Oligopoly and Economies of Scale," in H. Kierzkowski (ed.), Monopolistic Competition and International Trade, Oxford: Oxford University Press. (1987), "Is Free Trade Passe?" Journal of Economic Perspectives, 1, 131-44. Krugman, P.R. (1991), "Myths and Realities of US Competitiveness," Science, 254, 811-15. Levinsohn, J.A. (1989), "Strategic Trade Policy When Firms Can Invest Abroad: When are Tariffs and Quotas Equivalent?" Journal of International Economics, 27, 129-46. Li, L. (1985), "Cournot Oligopoly with Information Sharing," Rand Journal of Economics, 16, 521-36. Masao, W. (1985), "Science across the Pacific: American and Japanese Scientific and Cultural Contacts in the Late Nineteenth Century," in A.W. Burks (ed.), The Moderniters: Overseas Students, Foreign Employees, and Meiji Japan, Boulder:

370

R. SATO, R. RAMACHANDRAN, AND S. TSUTSUI

Westview Press. Meza, D.D. (1989), "Not Even Strategic Trade Theory Justifies Export Subsidies," Oxford Economic Papers, 4 1 , 720—36.

Milgrom, P. and J. Roberts (1982), "Limit Pricing and Entry Under Incomplete Information: An Equilibrium Analysis," Econometrica, 50, 443—59. Mino, K. and S. Tsutsui (1990), "Reputational Constraint and Signalling Effects in a Monetary Policy Game," Oxford Economic Papers, 42, 603—19. Mokyr, J. (1990), The Lever of Ricbes: Technological Creativity and Economic Process,

New York: Oxford University Press. Novshek, W. and Sonnenshein, H. (1982), "Fulfilled Expectations in Cournot Duopoly with Information Acquisition and Release," Bell Journal oj'Economics, 13, 214-18. Ono, Y. (1990), "Foreign Penetration, New Entry and National Welfare Under Oligopoly," Japan and the World Economy, 2, 141-54. Porter, M.E. (1990), The Competitive Advantage of Nations, New York: The Free Press. Sato, R. (1988), "The Technology Game and Dynamic Comparative Advantage: An Application to U.S.-Japan Competition," in A.M. Spence and H.A. Hazard (eds.), International Competitiveness, Cambridge, Mass: Ballinger Publishing Company. Sato, R. and S. Tsutsui (1987), "Information Strategies, Market Barriers and Trade Performance," Center for Japan—U.S. Business and Economic Studies, Stern School of Business, New York University. (1992), "Uncertainty, the Role of Consensus, and Interventionist Trade Policy," Working Paper, Center for Japan-U.S. Business and Economic Studies, Stern School of Business, New York University. Schriesheim, A. (1990-1), "Towards a Golden Age for Technology Transfer," Issues in Science and Technology, 7, 52-8.

Schumpeter, J.A. (1942), Capitalism, Socialism and Democracy, New York: Harper and Row. Spencer, B.J. and J.A. Brander (1983), "International R&D Rivalry and Industrial Strategy," Review of Economic Studies, 50, 707—22. Sugimoto, M. and D.L. Swain (1989), Science and Culture in Traditional

Japan,

Rutland, VT: Charles E. Tuttle Company. Tassey, G. "The Functions of Technology Infrastructure in a Competitive Economy," Research Policy, 20, 345—61. Tsutsui, S. (1991), "Cost Differential and Welfare Effects of Interventionist Trade Policies in Oligopolistic International Trade," Japan and the World Economy, 3, 341-55. Tsutsui, S. and K. Mino (1990), "Nonlinear Strategies in Dynamic Duopolistic Competition with Sticky Prices," Journal of Economic Theory, 52, 136—61.

CHAPTER

Protectionism in Western Europe NORMAN SCOTT

By means of glasses, hotbeds and hotwalls, very good grapes can be raised in Scotland, and very good wine can be made of them at about thirty times the expense for which at least equally good wine can be brought from foreign countries. Would it be a reasonable law to prohibit the importation of all foreign wines merely to encourage the making of claret and burgundy in Scotland. [Adam Smith, The Wealth of Nations, 1776]

The debate about the relative merits of free trade and protectionism has a long history in Europe, stretching back at least four hundred years to the argument between the mercantilists and their eloquent and persuasive adversaries in the eighteenth century, David Hume and Adam Smith. The debate goes on in the closing decade of the twentieth century in the wider arena of worldwide negotiations in the Uruguay Round to reduce or remove restrictions on international trade in goods and services. A majority of economists have long since espoused the advocacy of Adam Smith in favor of liberalization, but decision-makers are still attentive to the special pleading of constituencies of interests clamoring for protection from "unfair" or "disruptive" foreign competition. In contemporary Europe the battle lines between the adverse camps have been redrawn, and some of the issues at stake obscured, by the accelerating drive toward economic integration within the European Community (EC). Not only has the EC carved out for itself the lion's share of world trade. It is also now enlarging its scope, and potential strength, by simultaneously striving toward closer economic, monetary, and political union (the single market program and the Maastricht Treaty) and toward a wider sphere of membership and influence (the European Economic Area and prospective enlargement of membership and associate membership). The objective of consolidating and deepening the degree of economic integration already achieved, by advancing further down the road to a single internal market 371

372

NORMAN SCOTT

cannot be readily reconciled with progress in dismantling explicit and implicit trade controls. Often, to the dismay and irritation of its trade partners, the EC appears to be setting greater store on the achievement of its internal cohesion than on the liberal international trading objectives set out in the Treaty of Rome. Hence, the allegations (and fears) that the EC is moving toward the establishment of a "Fortress Europe," inward-looking and difficult of access to the outside world. Even in quarters where the political dimension of European integration is well understood and approved, the trade policy stance of the EC has come in for stern criticism - as when President Bush in 1992 stated that "sooner or later the EC must stop hiding behind its iron curtain of protectionism."1 The EC with twelve member states, and about the same number waiting in its ante-chambers for admission, is not the whole of Western Europe. Nonetheless, in what follows attention is focused on the EC because it accounts for one-fifth of world trade, and roughly one-fourth of world GDP. 2 The type and extent of protectionist measures applied by the EC should be assessed in the context of the evolution of the interaction process and also in comparison with the policies of the Community's trading partners. The latter are dealt with elsewhere in this volume. Here, available space does not permit more than passing references to the historical background. Rather, some salient economic features of the EC's changing position in world trade are recalled, then the main measures of Western European trade regulation are described and assessed. The place of the EC in world trade A basic aim in forming a customs union or free-trade area is to bolster economic growth by taking advantage of the economies of scale offered by a larger, duty-free market and by obtaining an improved allocation of resources. The internal dismantling of barriers to trade is part and parcel of that process within the EC and has generated a faster growth of the internal than of the external trade of the EC member countries. Thus, the share of external transactions in the total merchandise trade of the twelve contracted from 63 to 40 percent between 1958 and 1989. External trade has nonetheless grown strongly - imports by over 10 percent a year in the same period - and has remained important, with a ratio to GDP of about 10 percent in 1990. Expressed in volume terms, and abstracting from exchange-rate changes, EC external imports have grown more than internal imports since the mid 1980s, despite the measures taken during that period to create a single internal market.

Protectionism in Western Europe

373

The EC adduces statistics such as these to support its claim that integration has generated more trade-enhancing than trade-diverting effects. The trade partners of the EC do not share this positive assessment. They point out that the Community pursues policies which limit their access to the EC market and thus impose costs both on them as suppliers (also on export markets, because of EC subsidies) and on EC consumers who are obliged to pay more than world-market prices for many of their purchases. Thus, the import penetration of EC markets by third countries, as measured by their share in apparent consumption (defined as EC gross domestic production plus imports minus merchandise exports) declined during the 1980s for agricultural products from 18 percent to 15 percent, while that of manufactures stagnated at around 11 percent. At the same time, the share of internal EC imports in apparent consumption rose in both these categories. Because of its common external tariff the EC is regarded as a single trading partner by other trading nations. As such, and in terms of the volume of its trade with the rest of the world, the EC is the largest entity in international trade: averaging imports and exports the Community accounted for 39.0 percent of merchandise trade, the United States for 17.5 percent and Japan for 7.5 percent. In recent years the twelve member countries of the Community have conducted almost 60 percent of their trade among themselves. The above figures go to show why the aims and methods of the common commercial policy of the Community are so important for the trading position of its trading partners. Amongst these, the largest is the EFT A group of countries (taking nearly one-third of EC exports in 1989 and providing one-fourth of merchandise imports), to which the EC is linked by agreements abolishing customs duties on industrial goods. The United States comes next in importance as a trading partner taking 20 percent of Community exports (which in turn amount to 20 percent of total US imports; the corresponding share of the Community in US exports is around 23 percent) and providing 20 percent of EC inputs. Japan supplies 11 percent of EC imports, but takes only 5 percent of EC exports (this imbalance being at the root of various restrictive policies applied by the Community to its trade relationships with that country). Trade with developing countries (other than OPEC) stayed throughout the 1980s at around the level it had reached by 1972 - i.e., 21 percent, although some sub-groups have fared better - notably the ASEAN countries and the Mediterranean countries which are also linked to the EC countries by cooperative agreements, while African and Latin American countries lost EC market shares. The structure of the EC's trade - characterized by the predominance of energy and raw materials in imports and, even more, of manufactured

374

NORMAN SCOTT

goods in exports - underlines its position as the world's second largest industrial power and its comparative scarcity of natural resources. The Common Agricultural Policy has significantly reduced the contribution of food imports to the EC's overall external trade deficit (although the food balance itself remains in deficit). However, although the EC remains the world's leading importer of agricultural produce (especially of tropical foodstuffs and animal feed) the share of such imports in total imports has fallen steeply (to less than 10 percent) as the share of industrial goods (particularly, machinery and transport equipment) has increased significantiyThe EC acknowledges, rather grudgingly, that important sectors of the EC economy are the objects of specific policies which, in their effects, restrict, deflect or "manage" the flows of its external trade. It maintains that these policies are necessary to achieve internal growth or cohesion, or to preserve the stability of markets, incomes, or employment. Some measures (especially those affecting textiles and steel) are said to be justified on the grounds of assisting structural adjustment; others because they are designed to provide "fair" levels of income in agriculture. Not all of those measures are protectionist in the narrow sense of the term implying restrictions imposed on market access for external imports. Many - such as procurement policies or exchange-rate support measures - are not explicitly protectionist in intent, but nevertheless limit trade opportunities. Export subsidies are a particular bone of contention for EC trade partners because of their effects on the prices and volumes of world trade in agricultural products. Commercial policy aims and instruments Any attempt at a general characterization of the trade-related policy instruments applied by the EC and its member states has.to begin by acknowledging that they constitute a highly complex foreign trade regime. The common customs tariff (averaging 5.7 percent in 1990 on industrial goods, of which 99.8 percent are bound) is only one - and not the most decisive - amongst a whole array of instruments affecting trade flows. Other measures include variable import levies, quantitative restrictions, import controls and prohibitions, minimum price regulations, and export restraint arrangements. Such measures are applied to a wide range of product sectors. Since the EC operates a complex hierarchy of preferential arrangements (some 60 percent of EC external imports originate from countries participating in such arrangements) and has sought to conclude many sector-specific trade arrangements, levels of protection vary widely amongst product groups and partners. Consequently, an aim of the

Protectionism in Western Europe

375

Uruguay Round has been not only to lower the general level of EC protection but also to reduce the disparities in the treatment of imports from third countries. The ingredients of the common commercial policy (CCP) of the EC have flowed quite logically from its economic raison d'etre and its broader political aspirations. They have been as follows: (a) a common customs tariff (CCT) whose level has been negotiable with other members of GATT; (b) a common classification of goods for customs purposes (customs nomenclature), with supporting common rules for customs valuation, rules for suspending or reducing customs duties, rules of origin, and rules whose purpose - as defined in Article 113 of the Treaty - is to ensure that "uniform principles" should be applied, in essentially the same manner, in all EC member states in their trade relations with non-member states. The article in question lists some of the issues where such uniform principles or rules should apply. They encompass tariff amendments, existing import quotas, protective commercial measures (including anti-dumping cases and those involving subsidies), export policy, and the negotiation of international tariff or trade agreements. The elements of the CCP mentioned above stem from the very nature of the EC as a customs union. Without them, even the narrowest commercial and economic aims of the EC would be difficult — indeed, impossible — to achieve. In addition, the EC has entered into a whole series of international agreements governing particular sectors of international trade (e.g., textiles, steel) or trade with specific countries or groups of countries (developing countries, state-trading countries, members of other regional trade groupings such as EFT A). In the negotiation and implementation of these numerous agreements there has also been a need for a CCP and unified negotiating stance. Given the preponderant share of the EC in world trade, the EC institutions have had to assume a major role in shaping the international trading order. Although the EC Commission has been entrusted with an important role in shaping and implementing commercial policy since the inception of the EC, and has grown considerably in stature and influence over the past three decades, the main decision-making body is the Council of Ministers on which each member-state has one member, the ministerial composition of the Council varying according to the subject under discussion (Foreign Affairs, Agriculture, Trade, Transport, etc.). It is at meetings of the Council (which has its own secretariat) that policy issues are threshed out and a balance struck between competing national positions. Major initiatives and

376

NORMAN SCOTT

policy guidelines are now invariably discussed at "EC summit meetings" at head of government level in which the President of the Commission also participates. These summit meetings take place on average twice each year, being hosted by each country in turn during its six-month presidency of the Council. So far as tariff amendments were concerned, since 1968 tariff changes can only be made through the Council and with the agreement of all members. Changes in rates are negotiated internationally by the Commission, mainly through GATT, where it speaks on behalf of the Community as a whole. Negotiated reductions in the level of the CCT Article 18 of the Treaty signified the willingness of EC member states to negotiate reductions in the CCT "below the level which they could claim as a result of the establishment of a customs union between themselves." Such reductions in the CCT have been agreed during successive rounds of negotiations held under the auspices of the GATT. The cumulative effect of the Dillon and Kennedy Rounds was a 35 percent reduction in the average level of the CCT, with the result that the trade diversion effects which had been feared as a result of the formation of the EC were limited. Further adjustments were made to the CCT following the admission to membership in 1973 of the United Kingdom, Ireland, and Denmark and the resultant need to align their tariffs upwards or downwards to the CCT. Non-EC countries affected by these adjustments then negotiated a compensatory reduction of the CCT under Article XXIV of the GATT. The subsequent rounds of reduction of trade barriers within GATT (the Tokyo and Uruguay Rounds) have also dealt with non-tariff barriers and previously exempt segments of international trade such as services, agriculture, and intellectual property. The Tokyo Round resulted in an agreement on a substantial cut of over 30 percent in tariffs on industrial goods, phased over several years during the 1980s, thus reducing the CCT nominal level to about 6 percent. This observation must be qualified, however, by the fact that effective protection in particular product groups in certain member countries is substantially higher than the apparently low average of 6 percent for industrial goods in the CCT. Preferential trade agreements with many countries or groups of countries also confer reductions or suspensions of duties on imports from their signatories. The Generalized Scheme of Preference (GSP) belongs to this category. Other reductions or suspensions are granted when particular products are in short supply in the Community (such tariff suspensions having averaged about 1,000 a year in the 1980s and having affected such

Protectionism in Western Europe

377

products as chemicals, plastics, electronic products, aircraft, and medical supplies). For some countries (e.g., Mediterranean countries which have cooperative agreements with the EC) or products the EC applies tariff quotas admitting duty-free entry of imports up to a specified limit. Article 113 of the Treaty of Rome requires, as noted earlier, that the CCP should be administered according to "uniform principles," implying that the rules governing trade with non-member countries should be essentially the same in each member state. The Article goes on to list some of the fields of application in which a common approach should be used. These range from tariff amendments to import quotas, safeguards and anti-dumping measures, and export credits. The import quotas introduced by the EC from the pre-Community period have proved less amenable to a common approach since member states have been reluctant to give them up. The "liberalization" list of products not subject to quotas was consolidated at the beginning of the 1980s, when some 500 quota notifications were still in operation. By the end of the decade, and leaving aside the national quotas imposed by Portugal and Spain under the Act of Accession, 120 non-textile categories of imports were subject to quantitative restrictions or import prohibitions in the use of which Italy and France were most active. The rules of application of Article 113 concerning measures (safeguard mechanisms) which member states may take to protect themselves against a surge of imports were revised in 1982 in such a way as to restrict the scope for individual national action and to vest greater power in Community institutions. The revised rules provide for Community procedures to deal with a sudden and rapid growth in imports of a particular product. Community monitoring - or surveillance - of imports through the use of import licenses, issued freely, has frequently been applied during the 1980s to major Japanese exports such as motor vehicles, video-tape recorders, TV sets, and machine-tools. Under GATT rules, surveillance measures must not discriminate against contracting parties, but the EC has pleaded repeatedly for the right to take selective action against suppliers which "disrupt" the EC market. Since this stand has been vigorously opposed by the NIC's and Japan, the EC has seldom had recourse to safeguard measures, preferring in their stead voluntary restraint agreements (VERs).

Anti-dumping procedures and policies From 1968, when the administration of anti-dumping legislation passed from individual member-states to the Commission (supervised by a consultative Committee of representatives of member governments), there has been a steady increase in the technicality and complexity of the

378

NORMAN SCOTT

applicable rules and procedures. This has come about because of the cumulation of decisions following particular investigations and also by additions to the body of legislation. In broad terms, EC anti-dumping rules follow the provisions of Article VI of GATT which lays down the circumstances in which contracting parties can take anti-dumping measures as well as the form these measures may take. Article VI was enlarged by the Anti-Dumping Code concluded at the end of the Kennedy Round and revised in 1979 following the Tokyo Round. Further changes have been negotiated during the Uruguay Round. The Community's controversial rate on components, introduced in 1987, was the subject of a formal complaint by Japan under Article XXIII of the General Agreement, the panel finding in favor of Japan. Following GATT principles, the EC imposes anti-dumping duties only in cases where EC industries have been injured by the impact of "dumped" products. Products are defined as being dumped if sold at a lower price in the export market (the "export price") than in the home market (the "normal value"). Problems arise in calculating those price comparisons. As regards the "normal price" the EC invariably bases its calculations on the cost of production plus a margin of profit (the "constructed price"). That procedure, in turn, depends on what is considered to be an appropriate level of profit. In recent years the EC has tended to set profit levels on a higher scale, sometimes reaching as much as 30 percent. Once the export prices and normal values have been calculated, the two have to be compared in order to establish the margin of dumping. EC legislation allows adjustments to be made for three categories of differences - namely, physical characteristics of the products, import charges and indirect taxes, and selling expenses such as transport, sales commissions, etc. However, the EC has been criticized for its refusal to take into account higher differences in the quantities sold on export and domestic markets or in the level of marketing (wholesale or retail) at which the sales are made. EC comparison rules have also been criticized (mainly by Japan) because of the different treatment accorded to the costs of sales organizations on Community markets from those incurred on domestic markets: these are deducted from the former but not from the latter, thus creating an artificial dumping margin. When the anti-dumping margin has been calculated the next requirement under GATT rules is to establish that the dumped goods have caused injury to domestic industry - the extent of injury being measured by reference to such factors as profit levels, capacity utilization, and market shares. For this to be established a considerable volume of economic, financial, and commercial information has to be collected during the investigation of a complaint by the anti-dumping division of the EC Commission.

Protectionism in Western YLurope

379

Once dumping and injury have been established the EC can either impose anti-dumping duties or accept an undertaking from the exporter that he will maintain his export price above a specified level. In some cases, undertakings regarding the quantity rather than the minimum price of exports are accepted. Appeals against anti-dumping duties can be lodged with the European Court of Justice. Amongst the controversial issues with which the Court has been seized, the interpretation of Community Regulation 802/68 has attracted much criticism since it concerns a restrictive use of rules of origin. In essence, the rule provides that, when two or more countries are involved in processing a product imported to the EC, the country of origin is considered to be the one in which the last substantial process or operation that is economically justified takes place — provided it was carried out in an enterprise equipped for that purpose and results in the manufacture of a new product or represents an important stage of manufacture. Disputes have arisen over procedures initiated by the EC against integrated circuits and photocopiers which made it difficult for an assembler or processor to give a new origin to a product. As mentioned earlier, Japan has successfully contested, through GATT dispute settlement procedures, Community rules on imported components. Various aspects of EC anti-dumping law have come under criticism over the years and during the Uruguay Round. They fall under two broad headings; first, those which claim that EC procedures are in breach of GATT obligations; second, those which relate to specific EC practices. So far as the former are concerned, it has to be admitted that there is no coherent set of legal principles under the GATT anti-dumping rules which permit a systematic confrontation of EC departures from them. As for the second group, they relate to EC methods and time-limits for anti-dumping investigations as well as to methods of calculation or "fair comparison" of export prices and "normal values." Other procedures which are contested concern adjustments made when proceeding to comparisons (including the exchange rates used), the determination of the "injury threshold," the calculation of "residual duties," and the method of making refunds. During the decade of the 1980s there were over 360 anti-dumping proceedings initiated in the EC, with a decline over the years in the number of anti-dumping measures in force, from 187 at the end of 1986 to 120 in 1989. The product groups most frequently affected were chemicals and allied materials (almost one-half of all cases) followed at some distance by mechanical engineering, textiles, iron and steel, and office equipment and consumer electronics. Price undertakings offered by suppliers were twice as frequent as the imposition of definitive duties. The supplying countries most frequently involved in anti-dumping proceedings were Japan,

380

NORMAN SCOTT

Yugoslavia, the United States, and Canada and the four "Asian dragons" (GATT, 1992). Changes in the number of proceedings provide no guidance to the amount of trade affected by them. According to EC Commission estimates only 0.9 percent of total EC imports were affected by anti-dumping duties in 1987. However, it is unknown to what extent the price behavior of suppliers was influenced by the possible use of these procedures. Since third world countries claim that their use by the EC has been increasingly restrictive and discriminatory their exporters may have been discouraged from taking advantage of trade opportunities because of the risk of being subjected to anti-dumping procedures. As mentioned earlier, the degree of protection provided by the combination of tariffs and non-tariff measures varies amongst product groups or branches of industry. Some illustrations of the incidence of those measures in combination are provided below by reference to selected sectors. Sectoral arrangements for managed trade Textiles and clothing With the stated aim of permitting "a smooth structural adjustment" in the textiles sector, the EC has entered into a large number of bilateral agreements with MFA and non-MFA countries as well as arrangements in the framework of cooperation or assistance agreements limiting imports of textiles and clothing. It also maintains some autonomous restrictions vis-a-vis non-GATT contracting parties. Within the framework of MFA IV (the 1986 Protocol extending the Arrangement on international trade in textiles) the EC had negotiated by the end of 1990 twenty-eight bilateral agreements on restraints on the import of textile products in order to reduce the risk of market disruption. The agreements provide for quantitative limitations and, under specific circumstances, for basket exit mechanisms. The number of categories of textile products subject to limits ranges from forty in the case of large suppliers (e.g., Republic of Korea and Taiwan) to only two or three for small suppliers (e.g., Peru and Argentina). Under the "autonomous regime" provisions of Regulation 3420/83, member states can apply QRs to certain textile and clothing imports from state-trading countries - applicable mainly to outward processing trade and to categories not covered by MFA-type agreements. These restrictions have been relaxed following the demise of state-trading regimes in Europe since 1990.

Protectionism in Western Hurope

381

Categories not under restriction are subject to the "basket exit mechanism" under which a trigger mechanism is applied when imports from a supplying country reach a specified share of total imports. Consultations must then be held to achieve mutual agreement on limitation of further imports. If agreement is not reached, the EC reserves the right to impose unilateral restrictions. The basket exit mechanism is mainly used at the level of member states and has been applied on average ten times each year since 1986 (as compared with eighteen cases each year under MFA III and forty each year under MFA II). Most MFA import quotas are negotiated for the EC as a whole, quotas being allocated to individual member states in line with their traditional trade pattern in textiles and clothing. Under the five preferential economic cooperation or association agreements with Mediterranean countries provision has been made for surveillance and, if need be, for safeguard arrangements to prevent disruption of Community markets caused by "sensitive" imports of certain categories of textiles and apparel products. In 1988 66 percent of textile imports from MFA countries were subject to bilateral quotas and one-third were free of restrictions. For preferential or restrained countries (amongst which Turkey is the second largest supplier, after Hong Kong and ahead of China) 79 percent of imports were in categories subject to surveillance. The Community claims that since 1966 it has practiced a more liberal approach to textile trade, citing in support of this claim the 25 percent reduction in the number of QRs and the 56 percent increase in the volume of imports from MFA countries between 1985 and 1989. With a view to easing the structural adjustment of the textiles sector, the EC has since the 1970s discouraged specific national support measures for those industries (EC Commission, 1990a). Aid to industries with excess capacity is limited to new activities and since 1977 all aid for the synthetic fibres branch which results in expansion of capacity is banned. Approval is granted only to support measures leading to conversion or reduction of capacities. What has been the outcome of this protection? For the Community as a whole, the production of textiles remained fairly stable throughout the 1980s while output of clothing was declining. Both industries lagged behind the expansion of manufacturing production as a whole. On average, employment in the two industries accounted for 9 percent of manufacturing employment as a whole in 1988, but with significant differences amongst member states, ranging from 4 percent in the Netherlands in 1986 and almost 30 percent in Greece and Portugal. Exports increased moderately in the 1980s, and the traditional export surplus in textiles was maintained, although at a lower level. By contrast there was a sharp rise in imports of

382

NORMAN SCOTT

clothing to twice the volume of exports so that the deficit rose appreciably. Many attempts have been made to evaluate the economic effects of EC textile policies. To take one example, it has been estimated that in the United Kingdom, if there were no MF A, clothing prices would be about 5 percent lower and those of textiles 2 percent lower. The resultant loss of jobs has been put at around 33,000. The costs to the consumer of protecting these jobs has been estimated at three to four times the average annual earnings per employee in the textiles and clothing sector. Iron and steel The process of European integration started before the Treaty of Rome with the ECSC Treaty of Paris which conferred much stronger powers of intervention than its successor. When the iron and steel sector was in acute crisis in the early 1980s to which member states were responding in divergent ways, a system of EC-wide production quotas and price controls was introduced under the Davignon Plan, based on Article 58 of the Treaty. The crisis itself had started to take shape in the 1960s and was caused by changes in demand patterns, under the influence of technical progress toward less steel-intensive products, steel-saving substitution processes in traditional user industries such as automobiles and competition from new sources of supply outside the Community. During the 1970s apparent consumption stagnated, capacity utilization in crude-steel dropped from 87 percent in 1974 to 56 percent in 1982 and there was a loss of roughly 260,000 jobs. Since EC tariff rates on imports of iron and steel are low (between 3.5 and 5.4 percent, depending on the degree of processing) the initial EC response was to reduce external pressures by voluntary restraint arrangements under which some sixteen major external suppliers accepted auto-limitation of their exports. During the 1980s the number of countries and products covered by those arrangements has been greatly reduced as the closure or adjustment of productive capacities was achieved under a programme of EC sectoral assistance (the steel-aid Code). Against this background the Commission envisages further liberalization of its protectionist policies, a less interventionist policy stance, and more effective use of anti-dumping procedures. In general, the EC foresees "a progressive liberalization of its external trade system to facilitate a smooth transaction to a post-1992 environment" (EC Commission, 1990b). In 1988 the Community exported 21 million tons of ECSC steel products (excluding pipes and tubes) despite restraint on its exports to markets such as the United States, and imported 10 million tons.

Protectionism in Western Hurope

383

Motor vehicles Protection in the automobile industry attracts more attention than in possibly any other manufacturing sector because of the unique importance of motor vehicles in the contemporary consumer's notion of well-being. Despite (and partly because of) numerous barriers to trade in the main sources of supply - the United States, Japan, and the EC competition between domestic and international manufacturers is fierce. The sector is also large in economic terms. Some estimates suggest that one in ten of all jobs in the EC is in the motor-vehicle industry and ancillary branches. Within manufacturing industry, it accounts for 10 percent (2.4 million jobs) of total employment and 9 percent of EC value added if production of parts and accessories is included. In 1988 13.4 million new vehicles were registered in the EC, making it the largest motor vehicle market in the world. International trade in motor vehicles (and components) is large and rising and international mobility of the factors of production is increasing. Assuring the competitiveness of national industries is understandably high on the list of priorities of EC member states and this aim has been strengthened by the fall in demand associated with the recession of the early 1990s. Access of external suppliers to EC markets is governed by a common customs tariff of 10 percent for passenger cars (some exceptions are granted under preferential arrangements) and by the requirements imposed by member states or the EC for conformity to certain technical specifications (type approval). Since there is no Community-wide system of type approval, the freedom of trade in passenger cars within the Community is thereby limited. Several member states restrict imports of Japanese automobiles. Neither of these trade-restricting arrangements can be reconciled with the envisaged completion of the Single European Market by January 1,1993. The effects of the recession in the early 1990s, having hit automobile producers particularly hard, are likely to retard achievement of the Commission's stated policy of opening up the Community's market as a whole to unfettered external competition. In these circumstances, a transitional period of adjustment is envisaged. This industry is a case par excellence of strongly restrictive national import policies diverging from the Commission's view of the Community-wide benefits from more open markets (and is thus in contrast with the respective positions of the Commission and member states in respect of agricultural production and trade). As a result, the EC market for imported motor vehicles is highly fragmented. Most import restrictions are directed against Japan. Thus, since 1960 Italy has admitted only about 2,500 direct car imports, and 750 light

384

NORMAN SCOTT

commercial vehicles, annually from Japan. In France the market share of Japanese vehicles is limited to about 3 percent. In the United Kingdom the corresponding share is limited to 11 percent under a bilateral arrangement with Japanese suppliers. A similar auto-limitation agreement between Germany and Japanese suppliers has limited the share of the latter in that market to about 15 percent. Indirect imports tell another, less restrictive story. Article 115 of the Rome Treaty governing nationally imposed quantitative limits has been administered in such a way as to permit an increase in total EC imports from Japan from 964 thousand in 1984 to 1.2 million units in 1989, giving Japan a share of 58 percent in the total EC import market for motor vehicles. Consumer electronics The electronics and information technologies industries have enjoyed very dynamic growth in the past decade in both production and international trade. Within the EC large sums have been spent to encourage research and development in these industries. At the same time a broad array of measures has been applied by the EC and member states to limit competition from external suppliers - notably in the Far East. Customs tariffs are high - around 14 percent for most integrated circuits, television receivers, radios, and video recorders - but the main restrictions on market access take the form of non-tariff measures (NTMs). Amongst these, anti-dumping duties and price undertakings have been applied to a broad range of office machines (electronic typewriters and word processors, printers, photocopiers, etc.) imported from Japan, Hong Kong, and the Republic of Korea, as well as to video-cassette recorders, cassette tapes, and disc players. In addition, imports of Japanese and Korean color TV sets and video-tape recorders are subject to voluntary restraint agreements and to EC surveillance. Further measures in restraint of imports (import quotas, export restraint arrangements, or monitoring procedures) are used by individual member states (France, Germany, Italy, and Spain). Measures taken to limit access to the EC market of Japanese manufactured semi-conductor chips have attracted much attention and stirred up controversy. After the EC had initiated in 1987 an anti-dumping investigation, a price undertaking was offered in 1990 by thirteen Japanese suppliers of dynamic random access memory (DRAM) chips. The floor price is established every three months. In addition, an anti-dumping duty of 60 percent was imposed by the EC on "grey market" imports. Estimates of the costs to the consumer in the EC area of anti-dumping measures suggest that they are high, amounting to 513 million ECUs in 1988 for just one item - dot matrix printers. The corresponding additional

Protectionism in Western Europe

385

costs to consumers for video-cassette recorders and compact-disc players in 1989 were estimated at ECU 272.5 million and ECU 146.1 million respectively. Civil aircraft The EC aerospace industry provides a good example of a sector where internal support (largely in the form of government subsidies) is a more efficient method of winning a larger share of the international market than border protection. Moreover, Europe's transport and tourism would be disadvantaged if the costs of acquiring imported aircraft, and jumbo-jets in particular, were higher than those paid by competitors. Hence, the simple average tariff applied by the EC to aircraft and parts is only 3.7 percent, the weighted average being as low as 0.5 percent. Leaving aside the substantial share of military production in the EC aerospace sector (65 percent in 1988), the civil aircraft sector is just under one-half of the size of the United States industry. Nevertheless, EC external trade in civilian aircraft has roughly been in balance in recent years, largely owing to the success of the Airbus program, which accounts for about 50 percent of total civilian aerospace turnover in the EC. It is precisely the competitive success of the Airbus which has been the cause of a longstanding dispute between the EC and the United States, the latter claiming that extensive subsidization by the four partner Governments (France, Germany, Spain, and the United Kingdom) constitutes unfair competition for its own industry. The unresolved dispute hinges on differing estimates of the scale of support, including indirect subsidies, for the project and the guarantees against exchange-rate losses provided to the Airbus consortium. EC policy measures affecting agricultural trade The EC is the world's largest importer of agricultural products accounting for about 23 percent of world trade in 1987, but with a large and increasing deficit in agricultural trade amounting to over US$ 27 billion in 1988. About 55 percent of agricultural imports enter duty free; nearly 34 percent are subject to customs duties, or a combination of duties with levies and countervailing charges; and 11 percent of imports are subject to levy. Over 80 percent of EC agricultural production is subject to common market organization, which means that all rules governing production and marketing (basic policy mechanism, quality standards, price, etc.) are applied or coordinated on an EC-wide basis. A shift in policy which has implications for international trade has been

386

NORMAN SCOTT

the increasing recourse in recent years to subsidized exports for major agricultural exports such as cereals, dairy products, and beef instead of intervention storage as the main means of regulating EC markets. In view of the size of the EC's trade in many agricultural commodity sectors, the Common Agricultural Policy (CAP) has an important influence on the quantities and prices at which agricultural products trade on international markets. Hence, any change in EC policies that results in changes in trade flows also has an effect on world prices of traded goods. These effects, and the levels of subsidization and protectionism associated with the CAP, have become highly sensitive issues in the trade relations of the EC with the rest of the world. The CAP has, according to Article 39 of the Treaty of Rome, the following objectives: increasing agricultural productivity, ensuring a fair standard of living for farmers, stabilizing markets, ensuring supplies, reasonable prices for consumers. It is evident that conflicts can arise in the attainment of such disparate objectives and that over time priority has to be given to some over others (EC Commission, 1989). There is also considerable scope for conflict between these objectives and those of the common commercial policy. Moreover, the agricultural structures, priorities, and lobbies in the member states vary greatly in nature and strength. As a result, compromises have to be worked out laboriously and, once reached, are difficult to change. The Commission thus has to bargain in international trade negotiations within strict limits imposed by EC member states whose aims and interests frequently diverge. Member states which do not agree with EC-majority views (or Commission proposals) and which fail to change them in their favor can set up their own national support programs subject to Articles 92 and 93 of the Treaty. In relating to agricultural products, the commercial policy is more complex than for industrial products. Comprehensive border measures are applicable to the principal agricultural products produced in the Community, but some groups of third countries are granted preferential access (by means of specific agreements) to the Community market. Such agreements involve lower rates of levy or duty limited in some cases to certain periods or to certain quantities of exports to the Community. The inclusion of the agricultural sector in the process of Western European integration was considered to be politically and economically essential. Hence, agricultural policy is included in Part Two of the Treaty of

Protectionism in Western Europe

387

Rome ("Foundations of the Community") and is considered to be a cornerstone of the Community's institutions (Tracy, 1990). However, the CAP has to take shape against a background of widely divergent national structures and interests as well as of serious structural maladjustments which impeded the development of common policies agreed by consensus. A major underlying problem, which still persists, was the inherited national policies of member states to protect domestic producers from competition from more efficient overseas producers and the desire to maintain, for social and political reasons - many of which are inscrutable - farm incomes and populations at a higher level than could be justified on economic grounds alone. As a result, the elaborate and costly array of measures of agricultural support have been a source of continuing conflict between the EC and its trading partners as well as amongst the more and less efficient producing countries within the Community itself. In general, short-term measures of price support have prevailed over longer-term programs of structural adjustment. Agricultural trade policy instruments The operation of the basic instruments of arrangements for agricultural trade with non-EC countries may be summarized as follows. The instruments used vary with the market situation of the products to which they apply, but they consist essentially of import charges - variable levies and/or customs duty - together with export subsidies or refunds. When the prices on non-member markets are higher than Community prices, the levy/restitution system may be inversed - taxes may be applied to exports and subsidies to imports. Measures affecting imports In the case of most Community agricultural products there is no customs duty on imports from non-member countries and the common organization of markets and prices is based on the setting of a target price, an intervention price, and a threshold price. In order to guarantee Community producers a level of prices generally above world market prices and to protect internal markets against the erratic variations of world prices, a variable levy is charged on imports from outside the Community. These levies which match the difference between c.i.f. import prices and the threshold prices bring world supply prices up to the target prices, taking account of internal transport costs. The products, to which this system applies in full, are cereals, rice, olive oil, sugar, milk, and milk products.

388

NORMAN SCOTT

The import arrangements for certain products are based on the combined use of levies and customs duty. Variable levies are also charged to cover the difference between the guide price and the import price plus customs duty the basic levy. The levy calculated in this way can, however, be raised or lowered according to whether internal market prices are below or above the guide price. Because of the special character of their markets, wines and certain fruits and vegetables are protected by a reference price in addition to customs duties. The common customs tariff is applied to some other products subject to the common organization of markets and prices (oilseeds, tobacco, hops, etc.) and to all products not covered by the common organization. For most of the products covered by the common organization of markets, a licensing system is applied to trade with non-member countries. Under this system both importers and exporters have to obtain a license for the proposed transaction. There are no quantitative restrictions on imports but agreements on voluntary restraint (VERs) have been concluded with non-member supplying countries whereby they agree that their exports to the EC will not exceed certain limits. In addition, provisions are included in all agricultural regulations, enabling the Community to adopt promptly, in exceptional circumstances, any measures needed to defend the Community market against serious disturbance. This is a safeguard clause which can only be activated with due regard to international undertakings. Measures influencing exports Refunds are the basic instruments for exports. They are to some extent the counterpart of import levies and are used to ensure Community products a competitive position for outlets in non-member countries. European producers are refunded the difference between the internal market price (including cost of transport to the Community port of export) and the selling price obtainable on the world market. However, in addition to this price difference, the amount of the refund also depends on the destination and on the market situation at the time. Refunds are the same for all the Community and are met out of the Community's agricultural budget (except for sugar). Refunds are fixed at intervals by the procedure used by the Management Committee for the particular product. This procedure varies with the product, but is most complete for the cereals sector. Implications of border measures for third world countries In analyzing the effects of these measures on trade with third world countries, a distinction has to be made, first, between products subject to

Protectionism in Western Hurope

389

levy and refund systems and those not so subject and, secondly, between products which compete with Community products and those which do not. Since such systems are applied to practically all Community products, they protect internal production from external competition and, in association with price and intervention systems, they encourage production and exports of Community products. This has marked effects on trade in these products with third world countries. Levies usually restrict trade more than do customs duties. Exporting countries can, in the medium or long term, reduce the effect of customs duty by improving their relative efficiency and by thus reducing their production costs can surmount the customs barrier. This is not possible with variable levies, because any lowering of the offer price is completely offset by an equal increase in the levy. Therefore, whatever the increase in productivity or the amount of export aids, exporting countries taken together cannot hope to increase their overall share of the Community market by these means even though the shares of the different exporters may vary. In general the immediate effect of this protection is a gain for producers and a tax on consumers in the EC. Production and trade effects of the CAP Agriculture attracts the preponderant share of EC aid to specific sectors. Agricultural price guarantees alone account for about 55 percent of the EC budget (1990). Since 1988 the European Council has used a reference framework for annual agricultural expenditure within which the increase in EC expenditures on guaranteed prices is limited to 74 percent of GDP growth. Agriculture also benefits from large expenditures from structural funds such as the European Regional Development Fund and the Guidance Section of the Agricultural Guidance and Guarantee Fund (EAGGF) which taken together account for another 24 percent of the EC Budget (1990). However, more than 90 percent of EC expenditure on agriculture is still allocated to the Guarantee Section of the Fund and is thus expended, for the most part, on price support. During the past three decades the EC agricultural sector has gone through three main stages of development which have affected the policies put in place to achieve the above objectives. First, there was a reduction in production deficits; then, a transition to self-sufficiency and the emergence of exportable surpluses of several major commodities; and, thirdly, in the 1980s, imbalances between supply and effective demand and the accumulation of large unsold stocks. These developments should be seen against the background of the rapid increases in agricultural productivity propelled by technical improvements

390

NORMAN SCOTT

in farming methods, the continuing shift of labor from agriculture to expanding industries, and slackening growth of demand for foodstuffs. The rise in production was largely due to exogenous factors such as technical progress, but the highly protectionist policies (and as a concomitant high EC-internal prices) must also have played a major part. Notwithstanding the gains in output and efficiency just mentioned, the EC agriculture sector has conspicuous weaknesses, the scale of which varies greatly from country to country, and is a less efficient producer than many of its competitors on international markets. Thus, average farm size at 14 hectares is too small to be economic; many of the small units cannot provide full employment for even one person; and over half the EC farmers are fifty-five years of age or older. Some 20 percent of the farms provide 80 percent of agricultural output. On average, farms in the United Kingdom are 2.5 times as large as those in Germany and 5 times as large as farms in Italy, Spain, or Portugal. Earlier, an indication of the cost of the CAP was provided in terms of its share of the EC budget. A broader measure is obtained by aggregating the transfers made by the Community with those of its member states. In 1989 these transfers are estimated to have totaled ECU 89 billion, of which ECU 49 billion was provided by consumers through the price effects of border measures, while taxpayers contributed ECU 40 billion in the form of production and export subsidies, income support for farmers, structural improvement programs, and general support measures such as investment and training. An alternative estimate, much used in the Uruguay Round, is based on producer subsidy equivalents (PSE). This measures the absolute amount (or share) of farm income attributable to policy intervention in the form of subsidies or border measures. (It should be noted that estimation of income transfers due to border measures relies on the price differential between internal and world market prices. When the trading entity is as large as the EC, world market prices themselves are affected by EC policies, thereby distorting to some extent the basis of the estimation.) Using the PSE measure, we see from table 16.1, the dairy sector is the most heavily protected in terms of the amount of assistance received (ECU 16.5 billion in 1989, or about one-third of total EC support to agriculture), followed by beef and veal (over one-fourth) and coarse grains and wheat (about one-fifth, taken together). When the amount of assistance given is expressed as a percentage of the value of output for each product-group, it provides a measure of the intensity of support. The following tabulation shows both types of PSE measurement by major categories for 1978-85 (EC 10) and for 1989 (EC 12).

Protectionism in Western Europe

391

Table 16.1. (PSE values) for EC agricultural support by main product groups percentages and EC values

All products Livestock sector Beef & Veal Milk Crops Soyabeans Course grains Wheat Sugar

1978-85

1989

Million ECU in 1989

35 36 47 52 33 48 32 28 54

38 39 55 52 35 54 35 24 52

48,195 36,649 12,937 16,542 11,446 523 4,173 3,244 2,012

Source: OECD, PSE, and CSE tables 1979-89. Trade in specific product groups From these figures it is clear that beef and veal, soyabeans, and dairy products all benefit from an intensity of support exceeding 50 percent. Not surprisingly, support on this scale has provoked the ire of other producers. Thus, the American Soyabean Association complained to GATT in 1987 and won decisions by two dispute settlements panels in 1989 and 1992 to the effect that the EC oilseeds subsidies impair the duty-free status accorded to soyabeans in the early 1960s, since the EC subsidy schemes completely protect domestic producers from price changes in imports. This infringes Article III.4 of GATT which provides for national treatment. It should be noted in this connection that the EC is the world's largest market for oilseeds and derivatives, importing some 24 million tonnes in 1988 (as compared with domestic production of one-half of that amount). Imports of soyabeans, rapeseed, sunflower seed, and oil-cake are admitted duty free. Since EC internal prices are higher than those on world markets, support for EC producers mainly takes the form of compensatory payments to processors to offset the additional cost of purchasing domestic inputs. Talks between the USA and the EC were continuing on the issue in the summer of 1992, under the threat of retaliatory US tariff action affecting US imports of $1 billion of European wines, liqueurs, cheeses, and other goods. The EC was apparently willing to change its support system within the framework of a successful outcome of the Uruguay Round. An array of other EC measures affects international trade in agricultural products. They include packaging and labeling requirements and sanitary

392

NORMAN SCOTT

and phytosanitary regulations. The so-called "Hormone Directive" banning the use of hormones in meat production, and the import of such meat, is an example of this type of regulation - and one which created friction with, and prompted retaliatory action by, the United States. Another source of protection of an unusual trend is the specific currency mechanism used by the EC in the monetary sector to cushion the effects on agricultural incomes and prices of exchange rate fluctuations in member states. It is estimated that in 1990 agricultural support prices in the EC were 14.5 percent higher than they would have been in the absence of such a mechanism. In the livestock sector, sheepmeat is the only product of which there are substantial net imports. Protection is provided by export restraints observed by the main suppliers (Australia, Argentina, New Zealand, Hungary, and some Central European countries) who in turn receive duty-free entry to EC markets for these products. Although the EC strives actively to reduce its cereals production by means of early-retirement and set-aside schemes (taking land out of production) this sector continues to account for about 10 percent of EC agricultural production and attracts 16 percent of total guaranteed expenditure on agriculture (1988). Net exports rose markedly during the 1980s to about 20 million tonnes annually at the end of the decade. Since cereal yields are forecast to continue to rise by about 2 percent each year until the mid 1990s, there is a risk of considerable oversupply and consequently heavy expenditures by the EC on accumulating large intervention stocks and supporting subsidized exports. Agricultural protection and the Uruguay Round Given the range of border measures applied to different agricultural products and the variability of the import levies over time, as well as the range of preferences to third world countries, it is difficult to determine precisely the levels of protection accorded to agricultural products. What is clear, however, is that on average the tariff equivalents of the border measures applied to temperate zone agricultural products have been substantially higher than the average level of tariff" for manufactured products. This large disparity in levels of protection has clear implications for the level of agricultural production, for the efficiency of resource allocation in the Community, and for an open and equitable trading system for agricultural produce. The CAP greatly strengthened the tendency toward self-sufficiency in foodstuffs to the point where the growth of EC exports was three times as

Protectionism in Western Hurope

393

great as that of imports. As a result, many third world countries have suffered important losses of markets in the Community and market shares elsewhere because of the inroads made by EC exports on world markets. Another result has been that EC consumers have had to pay much more for their admittedly abundant supply of foodstuffs than they would have had to under a less protectionist agricultural trade regime. An additional opportunity cost has been the severe limits placed on EC budgetary expenditures in other fields because of the disproportionate share devoted to agricultural support. The conflict between farm support policies and trade liberalization came to a head during the Uruguay Round. For the first time progress in liberalizing international trade in agricultural products was included in the agenda for a round of multilateral negotiations held under GATT auspices, largely at the insistence of the United States and the fourteen "fair-trading" nations known as the "Cairns group." Export subsidies were at the heart of the conflict and without resolution of this problem the negotiations dragged out for over six years. Numerous studies and qualitative estimates were made during the negotiations of the immediate and indirect economic consequences of reductions in export subsidies and other support measures. The general thrust of these studies was that liberalization would bring benefits to everyone except farmers; manufacturing output in the EC and exports would be expected to rise and unemployment to fall. Overall, the prospects would be for appreciable net gains in economic welfare from reductions in support levels. The drawback is that these benefits would be distributed so widely as to be barely perceptible to beneficiaries, whereas EC farmers would bear immediate costs in the form of lower prices and incomes. The approach to the resultant policy dilemma of making such a policy change in a manner acceptable to farmers and their vocal lobbies was that of "decoupling" - in other words, switching from price supports for farm produce to direct payments to boost farmers' incomes. Such an approach had no realistic chances of success unless applied in a series of coordinated reductions of support level by the EC and the United States. Although support for the approach was stronger amongst its trading partners than in its member countries, the EC introduced a plan in 1991 providing for steep cuts in subsidized prices, with full compensation to be paid directly to the farmer, contingent on medium to large farms taking land out of production (for which the larger farmers would not be compensated). However, the EC offer of a 30 percent cut in subsidies was rejected as inadequate by the United States, which held out for twice as deep a cut in real terms.

394

NORMAN SCOTT

Notes The views expressed in this Chapter are those of the author and do not necessarily reflect those of the UN-ECE secretariat. 1 Speech by President Bush in Kansas City, January 13, 1992. 2 Complete coverage of protection in "Western Europe" would call for consideration of the trade policies of the seven EFTA countries (Austria, Finland, Iceland, Liechtenstein, Norway, Sweden, and Switzerland), linked with the EC twelve in the European Economic Area (EEA), as well as of countries in the region outside these groupings (notably, Turkey and the successor states of Yugoslavia). Such extension of the coverage would inevitably result in an excessively long text. Although the omission of consideration of those countries is regrettable, it may be observed that the level of protection in the EFTA countries is broadly comparable to that of the EC countries so far as trade in industrial goods is concerned, although several of them use a gamut of trade barriers to shelter from foreign competition such sectors as textiles and clothing, footwear, furniture, transport equipment, the food-processing industry, and the electronics and electrical industries. In addition, the EFTA countries in general provide heavy protection for agriculture which is largely insulated from international market signals. Protection is provided by internal aid programs ranging from production subsidies and marketing aids to production quotas strengthened by restrictive border measures including import and export licensing, import quotas, variable import levies, high tariffs, minimum import price requirements, and export subsidies. This policy results in high prices for agricultural produce and large price differentials - ranging between 50 and 100 percent - compared with international prices. The average level of producer subsidy equivalent in EFTA countries is near 70 percent, well above the OECD average. The EEA agreement, understandably in these circumstances, does not provide for a common agricultural policy nor for a common external trade policy. The seven countries of EFTA have agreements with EEC countries which allow for almost complete free trade in industrial products between the two groups of countries. Apart from some very minor concessions, agricultural trade is excluded from these agreements. There are other agreements between EFTA countries and the EC that involve agricultural trade, as indicated above. In 1990 formal negotiations were opened between EFTA and the EC on a significant enlargement of cooperation between the two groupings in order to create a "European Economic Area." The negotiations ended in agreement at the intergovernmental level in May 1992, after which the accords were submitted to the national legislations of the EFTA countries for approval by the end of the year, in time for the entry into force of the Single Market in January 1993. References EC Commission (1989), "A European Agricultural Policy for 1995," European Documentation, No. 5.

Protectionism in Western Europe

395

(1990a), Industrial Policy in an Open and Competitive Environment, Brussels. (1990b), General objectives steel-1995, Brussels. GATT (1991), Trade Policy Reviews, Reports by the secretariat and governments of contracting parties on the European Communities and on the individual EFTA member countries (except Iceland), 1991-1992. (1992), International Trade, 1990-1991, Geneva. Tracy, Michael (1990), Government and Agriculture in Western Europe 1880-1988, Third edition, New York: Harvester Wheatsheaf.

CHAPTER

Protectionism and the developing countries G.

K.

HELLEINER

The developing countries suffer greatly from the practice of protectionism by their trading partners and shortfalls from the multilateral, nondiscriminatory, and transparent trade regime to which the GATT ostensibly aspires. Trade barriers in the industrialized countries impose costs upon the developing world which substantially exceed, in the aggregate, the total flow of official development assistance. These costs seem certain to remain largely in place, regardless of new arrangements in the GATT or elsewhere, throughout the 1990s. Section 1 of this chapter addresses the issue of the developing countries' interest in reducing protectionism in the industrialized world. Protectionism is also found in the developing countries. Its logic, in the context of overall development efforts, and provided it is not taken to extremes, has traditionally received some professional backing. In the 1970s and 1980s, however, there have been significant trade policy changes in much of the developing world - typically in the direction of liberalization and reduced industrial protectionism. Section 2 addresses the continuing debate over the appropriate role of trade policy in the industrialization of low-income countries in the context of current liberalization initiatives. 1

The new protectionism, the GATT and the developing countries1

The post-war international economic order was on the whole created by and for the victorious powers. Memories of the protectionist and discriminatory practices of the 1930s were fresh. It was a time of unusual 396

Protectionism and the developing countries

397

and worldwide commitment to the building of a more stable, efficient, and equitable international order. In the trade sphere, the vision of the architects of the new order was of a more liberal and non-discriminatory system within which independent and equally treated partners would trade in open and competitive markets. Macroeconomic events were to be more effectively managed than previously, it was hoped, by governments imbued with a new confidence in their capacity to maintain full employment and buttressed by the new International Monetary Fund (IMF). Africa, Asia, and the Caribbean were still mostly under colonial rule and were largely absent at the creation of the IMF, the World Bank, and the GATT. Latin American countries and India participated actively in some conferences, including that at Havana which framed the stillborn International Trade Organization (ITO), but their arguments found few echoes in the institutions that later emerged during the 1950s and 1960s. Bargaining at the GATT was dominated by countries seeking access to large industrial markets and offering reciprocity in kind; less developed countries (LDCs) until recently remained passive spectators of this process. While the rules of the tariff bargaining game showed little concern for LDC interests, the great post-war expansion nevertheless resulted in new trade opportunities for many third world countries. Their exports grew at rates faster than during the dismal interwar period, and their export bills became more diversified, both as to products and geographical destinations. Building upon the GATT To the founding fathers of the GATT, the principle of nondiscriminatory multilateral trade, embodied in the unconditional mostfavored-nation clause, was as important as the search for a freer international flow of commodities. The motivation was not merely, or even mainly, consideration of economic efficiency. Rather, it was the product of memories of international political frictions generated before the Second World War by the discriminatory preferential arrangements enforced by imperial powers and other countries aspiring to hegemonic preeminence. In the view of many historians, the scramble for preferential trading arrangements and the exclusion of rivals from both sources of raw materials and promising markets contributed to the tensions that led to both world wars. It was the intention of those who framed both the GATT and the ITO to devise trading rules that would allow a maximum of predictability regarding the openness of foreign markets, in order to stimulate long-term investments. Naturally, no nation is able to commit itself to maintain the same degree of openness to trade under all circumstances, so allowances were

398

G. K. HELLEINER

made for escape clauses, such as Article XIX in the GATT. The intention was to make the adoption of any departure from commitments as transparent as possible so it could be subjected to international review and surveillance. These principles of predictability and transparency maintain their attractiveness, even if, particularly during the last two decades, they have been honored more in the breach than in practice. Their attractiveness rests not just in the inducement they provide for efficient trade expansion, but also in the contribution they make to the creation of an atmosphere of fair play among competing economic agents, an atmosphere without which political pressures could again arise, leading to trade wars. The original norms and expectations of the GATT's founders have been breached in many spheres but nowhere more than in terms of the rise of preferential and discriminatory trading areas and arrangements. A particularly severe blow to non-discriminatory principles was the negotiation, under GATT auspices, of discriminatory quotas against "low-cost" suppliers of textiles and textile products. Originally negotiated on a short-term basis for cotton textiles thirty years ago, they have since been greatly extended in time and in product and country coverage. The Multi-Fiber Arrangement (MFA), the successor to the original short-term cotton agreement, continues to legitimize some of the most egregious of the discriminatory practices of the industrialized countries and constitutes a standing repudiation of the GATT's purported first principle of nondiscrimination. It has been enormously harmful to the GATT's image and credibility - particularly, of course, among developing countries. The multilateral and non-discriminatory aspirations implicit in the GATT were also confounded in the arrangements within the European Economic Community (EEC), and between it and many of its trading partners. More recently, with the US-Canada free-trade agreement, proposals for a Western Hemispheric preferential deal, and the consolidation and expansion of the European Community, there has been acceleration in the potential splintering of the multilateral trading system into regional trading blocs, with the developing countries always as very junior partners. No less contrary to the non-discrimination principles of the GATT, although of relatively small overall significance, is the multilaterally agreed system of tariff preferences granted by industrialized countries since the early 1960s on semi-processed, processed, and manufactured products from developing countries. Only about two-thirds of world trade now takes place on the basis of non-discriminatory (most-favored-nation) tariffs that are equally available to all under firm and contractual commitments in the GATT. Also damaging to the status and credibility of established multilateral institutions like the GATT was the increasing resort to international codes

Protectionism and the developing countries

399

and agreements of a much less than universal character. In such areas as capital liberalization, the terms of export credit, and the appropriate conduct of transnational companies, the Western industrialized countries negotiated their own codes or agreements within their Organization for Economic Cooperation and Development (OECD) without waiting for more universal accommodations to be reached. Even within the GATT, some of the codes governing non-tariff measures negotiated during the Tokyo Round, notably those on export subsidies and on government procurement, were applied, contrary to previous principle and practice, only to the exports of some countries; eligibility for benefits was conditional on signature of the codes. In the 1970s and 1980s there was a quantum leap in the extent to which industrialized countries employed non-traditional (i.e., non-tariff) measures to restrain imports. Many of these measures are neither specifically authorized nor explicitly forbidden by the GATT. The most visible of these much-discussed measures of "new protectionism" are so-called voluntary export restraints and orderly marketing agreements (similar in type to the MFA). These are bilaterally negotiated against the threat of unilateral imposition of restrictions by the importing countries; they are thus inherently discriminatory. It is widely feared that such arrangements, now applied to footwear, electronics, machine tools, steel, motor vehicles, textiles and clothing, and agriculture - sectors which together constitute close to half the total value of world trade - will gradually intrude upon more and more segments of international exchange. Particularly vulnerable to these new instruments of protectionism are the newly expanding manufactured exports of the developing countries. A variety of other new non-tariff measures were also increasingly deployed. These include local content requirements, government procurement policies, laws concerning standards, variable levies on imports, and subsidies for local production under the terms of "industrial," "regional," or other national policies. At the same time there was increasing resort to anti-dumping and countervailing duties, on terms decided upon by the governments of the importing countries, to counter what these governments judged to be export subsidies doing injury to its firms. In the case of the US, measures were also unilaterally introduced to punish foreign trading partners, in ways not authorized in the GATT, for what the US government judged, on the basis of norms unrecognized by the GATT, to be "unfair" trading practices. Bilateral negotiations on terms prescribed by US laws replaced the existing multilateral machinery of trade negotiations under the GATT. Increasing Northern concern with environmental issues threatens to add a new rationale for restrictions on developing country exports. Administrative discretion, complexity, and obscurity substitute in

400

G. K. HELLEINER

these measures for the relative simplicity and automaticity of the import tariff. With these highly contingent and lawyer-intensive protectionist instruments inevitably also comes discrimination among trading partners. Where complexity and discretion take over from simple rules, the weaker firms and countries invariably lose most. In the 1970s and 1980s, pressures arising from serious worldwide macroeconomic dislocation were added to the deteriorating environment for world trade and investment created by changing politico-economic circumstances and declining respect for the previously agreed (GATT) norms. These macroeconomic difficulties not only had direct effects upon world trade but also generated sharp new pressures for protectionism to save jobs from competitive imports. The smaller industrialized countries and the developing countries, while typically highly vulnerable to events in the world economy, have rarely been able to exert much influence upon them. As in previous such periods, the weakest were hit the hardest. The immediate task must be to prevent any further deterioration in world trade and to arrest the dangerous downward slide into bilateralism, ad hoc protectionism, and the flouting of multilaterally agreed norms. Uncertainties as to market access are prejudicial to investment and therefore the prospects for adjustment to the current difficult circumstances in the developing countries and their eventual recovery. Current discriminatory practices, bilateralism, and demands for "reciprocity" are reminiscent of the bullying tactics and retaliatory practices of the 1930s, the very circumstances that led to the creation of the GATT in the first place. It is fundamental to the restoration of the credibility of the GATT that a satisfactory "safeguard" mechanism be agreed. Its principal elements must include arrangements for detailed international monitoring and surveillance of trade-restricting safeguard measures and related adjustment, together with equitable procedures for dispute settlement; specific and strict time limits and phasing-down procedures; strictly defined and internationally sanctioned objective criteria, based on economic principles, for the circumstances, scope, and terms of their use; with the onus of proof of the need for such measures resting upon the importing country. (The "serious injury" that authorizes safeguard action in the current Article XIX has never been defined, so it is in fact still unilaterally determined by the importing country on its own terms.) If the traditional GATT principles are to be honored, safeguard action should be permissible only on a non-discriminatory basis, unless discrimination - say, in favor of the poorest countries - is itself multilaterally agreed. The principle of non-discrimination is so fundamental to a well-functioning world trading system that it must be retained, and deviations therefore - actual or proposed - systematically rejected. Only

Protectionism and the developing countries

401

temporary derogations, in narrowly defined and strictly timebound circumstances, should ever be tolerated; and, even then, only if the maintenance of trading "order" - the return to some set of agreed rules absolutely requires it. Agreement on a revised safeguard system should logically be followed by a negotiated "winding down" and "phasing out" of voluntary restraint and orderly marketing agreements that do not meet its terms, not the least of which must be the Multi-Fiber Agreement. It is striking that, while the industrialized countries' trade authorities debate whether their textile and clothing sectors require fifteen or twenty years to adjust to a phasing-out of the MFA, their financial officials at the same time demand, in the IMF and World Bank, instant change and "shock" treatment in the trade (and other) policies of the developing countries. While a strengthened GATT secretariat may ultimately be the most important element in a workable multilateral system it will also be critically important to resolve remaining definitional and procedural ambiguities such as the precise meaning of the "material injury" that authorizes anti-dumping and countervailing duties, the means of establishing a causal link between "injury" and imports, and the basis for measuring the extent of export subsidies in different countries and circumstances and appropriate offsetting duties. Such agreements must accompany renewed efforts to bolster the development of open, efficient, and impartial dispute-settlement procedures, without which the existing system of trading rules, even with new codes, is likely to generate outcomes consistently biased against the smallest and weakest partners. Whatever else is done, it is vital, at a time of increasing resort to opaque, contingent, and discretionary instruments of trade policy, to increase the transparency of international trade-restricting practices. The visibility and predictability of the GATT "bound" import tariff were major advantages that are perhaps only now being fully appreciated. The GATT, the UNCTAD, and the IMF all seek to monitor trade-restrictive practices in their own ways. It is important, regularly and systematically, to collect and publicly disseminate information concerning the new instruments of trade policy - their extent, incidence, effects, and changes, country-by-country, over time. Discriminatory practices should be particularly sought out and publicized in the light of their uniquely damaging effects upon the system. The newly launched surveillance reports are an important step in this direction. "New issues" in the GATT and developing countries The international trading patterns of the future are determined by today's investment decisions; and, conversely, international investment

402

G. K. HELLEINER

decisions today are governed to a substantial degree by expectations as to the openness or restrictiveness of trading opportunities in the future. Both private firms and national governments engage in practices that are restrictive of others' trading and investment possibilities. National policies respecting rights of establishment, national treatment of foreign firms once established, and the international flow of goods, non-factor services, data, and factors of production are all obviously matters for independent sovereign judgment. The rights and obligations of private firms in the national and international economic arena are large and complex topics on which governments are bound to take divergent views. The object must nevertheless be to seek international accommodation and consensus among sovereign governments so that at least the international rules of the trading and investment game are subject to certain agreed principles, however small in number, which are understood and, to the maximum extent possible, accepted by all. Few services were included in the original GATT rules. Some were regarded as involving national security concerns (e.g., telecommunications); others were at that time heavily controlled by nations who wanted no constraints on their freedom of action (e.g., shipping, which LDCs wanted included under GATT rules). Statistically speaking, international trade in "services" is by now of significant global proportions and has been growing at a more rapid rate than goods trade in recent years. The industrialized countries, particularly the United States, are substantial and growing net exporters of services of both the factor and non-factor sort, while the developing countries' net deficits in services continue to grow. But this category, a catch-all for that international exchange which is clearly not goods trade, is so all-inclusive as to be nearly meaningless for interpretation. Some international non-factor services flows are closely connected to goods flows; others are not. Many involve a high degree of intrusiveness into local cultures and even sovereignty, or at least are so perceived by various countries. Coastal shipping, radio and television, domestic air traffic, local telephones and telegraphs are some of the service activities many countries regard as out of bounds for foreign economic agents. Insurance, banking, and data flows, which figure prominently in the United States' push for the extension of GATT into services, are also so regarded by many countries. Many of these activities and concerns are at the very core of national development strategies and are intimately related to policies on technology (both imported and domestic) and foreign investment. To many, the thrust to place these matters upon the GATT agenda seems tantamount to launching an effort to dismantle various screening and control measures

Protectionism and the developing countries

403

painstakingly constructed over the post-war period - not only in LDCs but also in many industrialized countries - in order to rein in burgeoning US-based and other transnational. While the origins of the US push for GATT involvement in international trade in services seem to lie primarily in the finance and telecommunications sectors, it is possible to interpret it, as many do, as an attempt much more generally to "open up" the markets of the GATT contracting parties to the direct foreign investment and other activities of transnational corporations. In this scenario, access for developing countries' goods in the markets of the rich may in the future be traded off under legitimate GATT auspices, instead of via bilateral pressure, against improved access for industrialized countries' "services" and investment in the LDCs. Transnational corporations have been very much involved in the post-war boom in international trade. In the ITO charter, attention was given to possible restrictive business practices, an issue also addressed by LDCs and smaller industrialized countries in recent years. Scholars have also remarked on the lack of international anti-trust rules and, more generally, on the lack of internationally agreed rules of the game with respect to international flows of capital and labor. When an appropriately broad approach to these issues is taken, explosive issues are thrust upon the table: if Tokyo is to be made just like home for US lawyers and bankers, why not have Texas give "national treatment" to Mexican maids? Will New York City be opened up freely to Indian doctors and South Korean construction crews? Which services and factor flows, in short, are to be "opened up," and what principles are to be followed in those decisions? The appeal of new codes or arrangements in respect of the "new issues" will, of course, depend on their attractiveness to various types of countries, and on how balances between rights and obligations are struck. A "GATT for investment," for example, should involve more than just an opening of doors. It should seek as well to clarify issues of jurisdiction, extraterritoriality, and restrictive business practices engaged in by transnational corporations, especially in the transfer of technology. Balance in overall coverage will also be important. The provision of greater security for direct foreign investors may be more easily granted by some, for instance, if accompanied by codes giving greater security to migrants. The benefits of multilateralism to the developing countries The protection offered to smaller countries by a multilateral rules system is far from perfect, but it is certainly greater than that available from the interplay among more powerful international actors pursuing their own

404

G. K. HELLEINER

interests in a world without rules. Is there any hope for reining in unilateral declarations by the US of others' "unfair" trading practices, for instance, in a world without a strong international trade authority? What hope can there be for a balanced and satisfactory outcome to individual developing countries' bilateral negotiations with more powerful OECD members on subsidies and countervailing duties in the absence of a multilateral agreement that sets out fair, transparent, and non-discriminatory rules? Bilateral - and sometimes GATT-illegal - pressures have recently been vigorously deployed by the US, most vigorously against some of the better-off developing countries, e.g., Korea, Brazil, Thailand. Is the prospect for an effective, efficient, and equitable multilateral trading regime better furthered by catering bilaterally to the new US approaches "knuckling under" on pharmaceutical patents, favors for US financial corporations, and the like in return for special trading favors - or by joining a multilateral effort to bring all such trade bullying to heel? While it is true that all countries benefit from a stable, transparent, and non-discriminatory system of trading rules - and that all should therefore cooperate in its pursuit - it is not true that all benefit from it equally. The importance of a strong multilateral system is much greater for smaller, weaker countries than it is for the strong. The strong, after all, will broadly be able to have their own way in a lawless environment. In order to induce the strong to support a rules system that disproportionately benefits the weak they are provided a "side payment" in the form of disproportionate influence in the establishment and management of the rules. Even after these concessions to the strong, however, the multilateral system is still likely to benefit the weak the most. In the "new" areas of services, intellectual property, and rules for foreign investment, just as in goods trade, developing countries are likely to fare better with multilateral alliances within a global system of checks and balances than in bilateral deals with more powerful partners. Bilateral trade arrangements with large countries may seem to offer great opportunities to small countries. The presumption of simple-minded two-country two-good neoclassical trade theory is that, when a small country "opens" to trade with a large country, the resulting gains flow disproportionately to the small country. This is basically because the large country's domestic price structure will dominate in the determination of post-trade prices and thus leave it relatively indifferent to the possibility of international exchange which does not significantly alter its incentive structure one way or the other. The small country, on the other hand, can expect a major change from its pre-trade price structure and therefore much greater gains from the reallocations of production and consumption that result from expanded trade opportunities (always allowing for greater

Protectionism and the developing countries

405

adjustment costs). Where there are scale economies, or other productivityenhancing effects, from which a small autarkic country could not previously significantly benefit, gains also are presumed to flow disproportionately to the smaller country. A more modern version of this basic proposition often offered with a neomercantilist export-oriented flavor - rests on the presumption that if one country's market is ten times as large as the other's, assuming both initially have roughly equal trade barriers, then reciprocal liberalization will provide (market) gains to the small country that are roughly ten times those of the larger country. If the smaller partner in a bilateral trading agreement is expected to derive a disproportionate share of the gains from increased trade, then the obvious question is why the larger (and presumably more powerful) partner should permit this outcome to occur. It is possible that the positive-sum nature of the game leaves the larger country content with its gains even though a larger share of the "increased pie" is claimed by its smaller trading partner. A better a priori hypothesis than that of "positive-sum contentment," however, is that the larger country will not consent to a game that is disproportionately beneficial to a smaller (weaker) trading partner unless it extracts a "side payment" from this partner that is roughly the size of that partner's disproportionate gain from trade alone. There is also the question of how the game is actually played. There must always be anxiety over the degree to which powerful actors will eventually submit to disagreeable developments, whatever they may have previously agreed, if they do not suffer greatly in consequence of failure to do so. Can the large country be credible in its commitments to comply with the terms of an agreement with a much smaller and weaker country? Must there not be a better balance of disadvantage from non-compliance, such as is approached in a multilateral system, or an alliance of small countries, for there to be hope that the large will eventually comply with what was originally agreed? Forty-five years after its inception, and following a period of dramatic changes in the international economy, the now weakened GATT-based system is in need of major strengthening. The case for a more liberal and multilateral trading order is as powerful as it ever was; yet the world is in grave jeopardy of sliding back into the discredited mercantilist practices of earlier periods. The basic principles of non-discrimination, predictability, and transparency are fundamental to an efficient and equitable trading order, and it is crucial, particularly for the weakest countries, to pursue them in a fully multilateral manner. Some may doubt that the many complex items now on the GATT agenda can be successfully negotiated in a world composed of over 150 heterogeneous nation states. It could be a profound mistake, however, to think only in

406

G. K. HELLEINER

terms of agreements on a bilateral or region-by-region basis or among those with which it might in the first instance be a little easier to agree. The temptations of that route must be resisted, lest the multilateral trading system to which the world must eventually again turn be set back for decades and the world sink into a period of trade blocs, spheres of influence, and bilateral bullying. 2

Protection and import liberalization in developing countries

The appropriate role of trade policy in the promotion of economic development has long been controversial. Traditional economic theory has emphasized the importance of utilizing comparative advantage in the pursuit of overall efficiency at both national and global levels. Emphasis upon comparative advantage need not, however, imply a policy of totally "free" trade. There exist sound and universally agreed theoretical grounds for some forms of activist trade policies. Moreover, the historical experiences of countries as diverse as Germany, Japan, the United States, and, more recently, Korea suggest that governmental interventions in trade and in other spheres may sometimes assist, rather than hinder, efficient industrialization and development processes. Most of the theoretical arguments for government interventions in trade are based upon "second-best" reasoning in which it is conceded that an alternative policy-mix in which trade policies play a minor role would be preferable if only it were attainable; political constraints, institutional rigidities, and other factors necessitate resort to interventionist trade policies, it is argued, even though they are known to be theoretically less than optimal, in pursuit of welfare improvements. Debate therefore often centers on rival interpretations of the importance of these constraints and other influences. Sometimes, in consequence of conceptual complexities and inadequate data, there are disagreements on facts relating to such matters as the size of externalities, the slopes of learning curves, the social rate of discount, etc. More recently, in the literature on so-called "strategic" trade policy, a new theoretical case for governmental intervention in trade has been built on assumptions, relating particularly to market structure and learning phenomena, alternative to those traditionally dominant in the literature. Whatever the merits of these newer theoretical arguments, as seen in the academic institutions of the industrialized world where these issues have been hotly debated, there still is a great deal of life left in the more traditional trade policy debates in the developing countries. Only very rarely, in any case, is the trade policy choice anything so bald as

Protectionism and the developing countries

407

a choice between free trade and "protectionism." The general case is rather that of partial trade policy reform - frequently also undertaken in conjunction with major macroeconomic policy changes - where the analysis of its effects can be complex and the results less than totally clear. In recent years, pressures upon developing countries to liberalize and simplify their trade policies have sharply increased. There can be little doubt that a good deal of governmental intervention in trade in developing countries has previously been too complex or that much of it has "misfired" in terms of promoting development. Other things being equal, both governments and international institutions prefer simple rules to complex ones. So do analysts of government behavior who most fear their misuse by special interest groups. The new orthodoxy of trade liberalization - in the sense of shifting from direct controls to market instruments of policy, from anti-export bias toward neutrality, and from diversity toward greater uniformity of interindustry incentives - offers what seems to be a simple, familiar, and universally applicable set of policy rules. But the models from which the rules are derived are subject to challenge; and the empirical evidence as to the impact of such policy reform is limited. The traditional policy conclusion of the orthodox trade economist - that free, or at least fairly liberal, trade policies are best for development - may still be correct. Indeed some argue that the debate has already been "settled" in that very way. But that conclusion cannot be based, without considerable further examination, upon theory that so neglects or severely plays down such real-world phenomena as oligopoly, learning by doing, externalities, scale economies, domestic non-price or institutional constraints, and foreign ownership. The empirical significance, theoretical implications, and policy relevance of these realities are matters that must be squarely addressed, rather than simply assumed away. Recent efforts to do so generate outcomes that are far less conclusive than those of the traditional approaches (Baldwin, 1992; Buffie, 1991; Grossman and Helpmann, 1991; Hazledine, 1990; Levy and Nolan, 1991). The lessons from recent historical experience include the fact that there can be heavy costs from significantly overvalued currencies and excessive levels of protection. They do not include, however, the efficacy of totally neutral incentives. Moreover, in poor countries where import tariffs are an important source of government revenue and fiscal constraints are daunting, one cannot easily contemplate major across-the-board tariff cuts unless one has first found alternative sources of revenue. The incorporation of technical efficiency and scale effects in recent simulation analysis of the microlevel consequences of import liberalization indicates that they can easily dominate those relating to allocative efficiency. Simulations of the effects of import liberalization where there is frictionless

408

G. K. HELLEINER

entry and exit can generate quite large productivity improvements (for a survey of results, see Havrylyshyn, 1990); but the assumption of freedom of entry and exit is critical to these conclusions (Rodrik, 1992), and the microevidence on these issues is unavailable. Only when import-competing entrepreneurs can be assumed previously to have been "satisficing" rather than optimizing, or to have backwardbending labor supply curves, are x-efficiency improvements from import liberalization theoretically plausible; and, even then, parallel exporter behavior would severely damage the overall case for it. Scale economy effects from import liberalization may, similarly, go in either direction. Those deriving from export expansion can be achieved via appropriate exchange rates (unified or dual) or export subsidies, and are not necessarily related to the import regime (Rodrik, 1992). There is certainly far more to the analysis of trade policy than the dichotomous characterizations that have been emphasized in much of the mainstream literature in recent years - in terms of so-called "inward" or "outward" orientation. There is also far more to industrialization policy than simply trade policy. These would have seemed banal and obvious things to say had there not been such inordinate attention to the role of anti-export bias in trade policies in recent writing on industrialization. The case for openness: current emphases in developing countries International trade has never been an end in itself. It must be functional - to the expansion of output and incomes. There is now a mainstream consensus that openness in economic relations with the rest of the world - in trade and other spheres - is likely to be conducive to economic growth. This is, of course, not a new proposition; but the emphasis in the mainstream logic behind it seems recently to have changed. Rather than focusing upon the static social costs of resource misallocation because of failure to utilize trading opportunities - which generally do not seem to add up to much in terms of readily calculable values (the areas of little triangles) - the current trade policy literature emphasizes instead the "dynamic" role of trade (and openness more generally) in stimulating productivity growth. In this respect, it reflects the current wisdom on the important role of technical change in overall economic growth. The early literature on trade and development stressed three separate kinds of potential or actual gains from trade in developing countries: (i) those relating to static comparative advantage, (ii) those associated with increased capacity utilization (and a "vent for surplus"), and (iii) those relating to productivity growth (and an "engine of growth").

Protectionism and the developing countries

409

(i) Exchange allows the separation of production from consumption, and hence the utilization of comparative advantage. Assuming that capacity remains always fully utilized, that the country is too small to influence world prices, that prices accurately reflect social benefits and costs, and that social welfare can be represented by a well-behaved utility function, national gains can be expected in a move from autarky to trade, and free trade can be shown to be socially optimal. Movement toward free trade involves a once-for-all gain. This potential gain from increased trade may be realizable in circumstances where policy has previously discouraged export production, for instance, by taxation and other disincentives that generated a domestic price structure less favorable to exports than the international one (anti-export bias). Where countries have actively encouraged local production of importables, they have thereby implicitly "taxed" exports and/or nontradables. Many have also, often inadvertently, taxed both importables and exportables by failing sufficiently to depreciate the official value of their currencies in response to rapid domestic price inflation, thus generating real currency appreciation. "Getting prices right," in the sense of eliminating anti-export and anti-tradable biases, will, in such circumstances and under the above-stated assumptions, generate welfare-improving reallocation of production and consumption, and trade expansion. (ii) The gains from trade deriving from increased capacity utilization are quite different in origin. In the traditional literature of developing countries, it was argued that local natural resources, including cultivable land, were frequently underutilized, as was, to some degree, labor. Production therefore took place well inside the frontier of production possibilities. The "opening" of the domestic economy to world demand altered the incentives facing owners of previously idle (or surplus) resources and labor sufficiently to induce them, via an aggregate supply response, to bring them into production. The potential for significant expansion of total production via voluntary responses to the provision of a "vent for surplus," in the sense of the historical literature, is today probably fairly limited. Surpluses of land or other natural resources are today much harder to find. Surplus labor is usually available, and can be obtained, since it is often involuntarily un- or underemployed, without the necessity of bidding it away from "leisure" or non-tradable production; but the reason for current labor surpluses is not demand deficiency, so increased (world) markets will not necessarily help to overcome them. The argument for gains from trade via increased capacity utilization remains an extremely important and powerful one; but its details today run

410

G. K. HELLEINER

rather differently. Capacity utilization in developing countries in the 1990s depends heavily upon the availability of critically important and nonsubstitutable imports - fuel, other intermediate inputs, spare parts, etc. When such imports cannot be financed at the levels necessary for full utilization of capacity, the result is underemployment of labor, capital, and resources in the import-dependent sectors. Since factor inputs cannot typically be redeployed quickly into other activities the entire economy is, in the short to medium term, if not longer (particularly where investment activity is also highly import dependent), also driven to production levels that are well below possibilities. Foreign exchange contraints and severe import compression have bedeviled most of Latin America and SubSaharan Africa over the past decade. Increased exports can finance increased import of critical inputs and thus achieve increased overall capacity utilization and gains in social welfare. Increased capital inflows or reduced debt servicing obligations can generate the same effects. These gains from trade derive from the role of imports, and increases in capacity utilization, not from increased allocative efficiency. Like reallocative gains, these import-related gains are once-for-all in nature. Unlike them, they can be achieved without export expansion or trade policy changes, if increased capital inflow (even on a temporary basis) can be obtained. There is another important policy dimension to the distinction between gains from trade from increased capacity utilization and those from reallocation. Liberalization of imports is generally considered likely to increase allocative gains but, if it increases the proportion of free foreign exchange that is spent on non-critical inputs, as it probably will, it may reduce those relating to capacity utilization in a foreign exchange constrained economy. It can also reduce them, even in "normal" un-foreign exchange constrained circumstances, if resources, particularly labor, are slow to adjust in the reallocative process (i.e., if the full employment assumption is violated). (iii) Discussions of the third kind of gain from trade hypothesize that there are different growth effects from different output mixes; that, in particular, given a particular factor endowment, productivity grows faster in more "open" economies. Traditionally, this was expressed in terms of the "engine" that exports can be. According to "staple" theory, some kinds of exports create greater such effects than others. Presumably, primary product exports, which are all some low-income countries can achieve in the foreseeable future, would usually generate fewer such positive externalities than industrial exports; though adequate empirical support for this

Protectionism and the developing countries

411

proposition is lacking. Some have formulated this argument in terms of purported positive externalities that are generated by the export sector to the benefit of the non-export sector (deMelo and Robinson, 1990) and have even sought to distinguish empirically between the (one-off) growth effects of reallocation and the (presumably continuing) growth effects of such externalities (Feder, 1983 and 1986). It seems more plausible, however, to associate the main externalities associated with trade with imports^ particularly those of capital goods and intermediate inputs (when they embody new technology) and technological and other non-factor services, rather than exports. Such imports bring the knowledge upon which long-term growth itself is now seen to be heavily dependent. In the words of the World Development Report, 1991: Openness — the free flow of goods, capital, people and knowledge transmits technology and generates economic growth across nations . . . First, increasing global competition raises the demand for new technology. Second, the supply of new technology for industrializing countries is determined largely by the degree to which they are integrated with the global economy. New products and processes are transmitted through imported inputs and capital goods, sold directly [or?] through licensing agreements, and transmitted through direct foreign investment or export contacts with foreign buyers. (p. 88) By affecting the nature of inputs as well as production processes, trade could generate gains which greatly exceed the short-term benefits from improved resource reallocation . . . The accumulated evidence suggests that the long-run gains from increased competition and the spillover of technology are likely to be much greater than the short-term gains. (p. 98) Growth depends, in part, according to this argument, upon the productivity-enhancing effects of particular kinds of imports, many of which are services. As with export staples, some kinds of imports are more "valuable" than others. No doubt the technology embodied in some is wholly inappropriate. Policies on goods trade may therefore not be as important for growth as those relating to imports of foreign direct investment, intellectual property, and services. Others, however, have emphasized the knowledge-enhancing role of some kinds of domestic production and the dependence of continuing growth upon them. Production of some tradables (whether importcompeting or exportable) may generate learning or technological spillovers (positive externalities) which also impact importantly upon growth. The

412

G. K. HELLEINER

unskilled labor-intensive activities in which developing countries usually have comparative advantage are often characterized as "traditional," lacking in positive externalities, and stagnant in terms of productivity (e.g., Grossman and Helpmann, 1991, chapter 9). Whether welfare can be improved by subsidizing more "dynamic" non-traditional activities certainly depends on a number of other influences, but the possibility, on these arguments, seems to exist. What is at issue is whether, or to what degree, productivity growth is influenced by the composition of output, or imports. The relationship between trade, output-mix, growth, and social welfare is complex and has attracted a good deal of theoretical attention (e.g., BufBe, 1991; Grossman and Helpmann, 1991). Empirical verification of theoretical propositions or presumptions in this sphere, however, has proven very difficult. The empirical evidence on the relationship between total factor productivity growth and output-mix, imports, or the trade regime remains inconclusive. Comparisons across countries are unpersuasive since there are so many other influences for which it is difficult to control. Nor are comparisons within countries over time always easy to interpret, since macroeconomic influences upon capacity utilization typically dominate the effects of changing output-mix or incentive structure over the short and medium run; long-run data are rarely available for developing countries. There is some evidence suggesting that, within countries, periods of protectionism and foreign exchange controls coincided with periods of only limited overall total factor productivity (TFP) growth. TFP growth in the industrial sector as a whole has been associated more with historical periods of export expansion than with those of import substitution in Korea, Turkey, Yugoslavia, and Japan (Nishimizu and Robinson, 1984). On the other hand, overall productivity growth in import-substituting Latin American countries was relatively high during the 1930s, and in the "protected" wartime years (Bruton, 1967). The impact of liberalization upon ongoing technical change - the principal element in measured economic growth over longer periods - is thus theoretically ambiguous and empirically uncertain. "There is as yet no convincing empirical evidence for developing countries that shows liberalization to be conducive to industry rationalization" (Rodrik, 1992, p. 170). For that matter, "to date there is no clearcut confirmation of the hypothesis that countries with an external orientation benefit from greater growth in technical efficiency in the component sectors of manufacturing" (Pack, 1988, p. 353). Even the World Bank, noting that "the relation between imports and productivity growth is sometimes positive and sometimes negative," concludes that "the debate is not fully resolved" (World Bank, 1991, p. 99; see also Havrylyshyn, 1990; Tybout, 1991; Rodrik, 1992; Pack, 1992).

Protectionism and the developing countries

413

Activist trade policy in developing countries: pros and cons The potential costs of protection are much more clearly understood in developing countries today than they were thirty years ago. The cost of that protection now actually being practiced in developing countries is, in consequence, almost certainly much lower than it was during the heyday of "inward-oriented" industrialization. Many of the extremely inefficient and socially costly governmental interventions in international trade have been eliminated or eased. At the same time, as seen above, these costs are no longer perceived so much in terms of static misallocation (including resources misused in rent-seeking) as in terms of the more "dynamic" costs of foregone growth. (To the extent that static misallocation is still at issue, scale economies in industrial activities are now featured in more sophisticated attempts to estimate protectionism's cost; and, although the outcome need not be so, they typically raise its size.) "Dynamic" effects are difficult to model. Much of the early trade policy literature employed comparative static analysis in a two-good two-country framework. One can easily show, using such an approach, that growth under incentives distorted by protectionist policies can be "immiserizing" (Johnson, 1967). But it is well-known that "undistorted" export expansion may also "immiserize" the growing economy via consequent deterioration in the international terms of trade (Bhagwati, 1958). The prospect of the latter kind of immiserizing growth is today a very real one in some sectors; many have noted the probability of welfare deterioration in the exporting countries if, for instance, world production of cocoa and coffee expands at more than very modest rates. Such simple exercises in comparative statics do not, indeed cannot, incorporate the learning effects and externalities deriving from different degrees of exposure to external influences or from different domestic output-mixes, that are so prominently featured in current debate over "openness" policies. With so much uncertainty in the realms of theory and fact, it is small wonder that there continues to be such active debate over trade policy. On the one hand we have caution and caveats as theoreticians and econometricians grope their way toward more realistic models and empirical tests thereof, with those on the frontier typically the most conscious of all of how much remains to be explored. On the other hand, despite considerable professional agnosticism about their developmental efficacy, advocates of both liberal and interventionist orientations continue to press confidently for various kinds of trade policy change. In the context of today's developing countries, as has been seen, the most important arguments for activist government trade policies, and as a subset,

414

G. K. HELLEINER

protection against imports, are those based on expected effects on long-term productivity growth, in particular, via learning effects and externalities. As far as learning effects are concerned, the argument is usually described as "infant industry protection." The infant industry argument posits that in certain circumstances governmental support for an industry may be justified on a temporary basis to permit it to "mature" to the point where it can stand on its own feet. In order for this to be theoretically defensible it is not enough to demonstrate that learning takes place in protected industries: the discounted present value of the future social benefits from the industry must also exceed those of the costs during the period of governmental support. Moreover, there must be reasons for believing that private decision-makers are unlikely themselves to act in such a way as to generate the potential excess of social benefits over social costs. One possibility is that the industry in question generates future positive externalities that cannot be captured by the owners. These externalities may be external to the firms but within the industry, or within the industrial sector as a whole, or they may be enjoyed by the national economy outside the industry in question. (The first-best policy for addressing such externalities is to direct pinpointed subsidies at the source of the positive external effects.) Another possibility is that governmental decision-making in the social interest may be governed by factors different from those governing private decisions. There are several possible reasons for this; for instance, the government may possess more information about future potential profitability, perhaps because it also knows about others' or its own investment programs (it is also possible, of course, that bureaucrats, in fact, know less than market participants); or the government may be less risk-averse than individual private decisionmakers because of its much wider (social) portfolio; or the government may employ a lower rate in the discounting of the future. The efficacy of governmental intervention via subsidies or trade policy based upon the infant industry argument hinges critically upon the degree to which learning and therefore productivity improvement occurs with experience and/or time. The same is true of governmental interventions motivated by the newer "strategic" arguments. The optimal instrument of government policy for the support of "infants" in such cases is, in any case, a direct subsidy, not a trade policy intervention that unnecessarily also influences consumption. If the correct instrument is chosen the current costs to society are theoretically reduced and intervention is more likely to be defensible. The long length of time over which many industries have enjoyed subsidies or protection creates a presumption that the ex post benefits have often fallen short of the costs (unless there has also been a steady flow of positive externalities). Certainly

Protectionism and the developing countries

415

the infant industry argument has frequently been abused. On the other hand, subsidies and trade barriers do sometimes come down as well as rise, and the record of the past, even if effectively demonstrated to have been a bad one, need not be relevant to the deployment of the appropriate policies in the future. The infant industry argument has a parallel at the level of the entire manufacturing sector in low-income economies; "infant manufacturing" merits protection in "infant economies" not merely because of learning effects, it is argued, but also, as noted above, because of its purported positive externalities for the rest of the economy, e.g., "spillovers" of knowledge, organizational structures, infrastructure, training of labor, etc. If manufacturing activity truly possesses such attributes many of them are presumably found in exporting as well as import-competing activities. Indeed in small countries exporting offers the only long-term hope for efficient expansion of manufacturing. But this "neutral" pro-manufacturing argument was frequently converted into a case for protection against industrial imports and thus import-substituting industrialization. Export pessimism, the GATT prescription of export subsidies and, perhaps, suspicion that externalities were less prevalent in the unskilled laborintensive activities most likely to be found in successful exporting, argued against much encouragement of export-oriented manufacturing. Import substitution was the natural first stage of industrialization, accounting for the vast bulk of recorded industrial expansion in the developing countries in the twentieth century. Some of this was the product of natural evolution, periodically offered a fillip by wars or balanceof-payments crises. That portion which was the product of conscious governmental "import substitution strategy," particularly in the postSecond World War period has been the object of severe criticism. Properly interpreted and implemented, however, the strategy made and continues to make considerable economic sense in terms of efforts to create "an economy that is sufficiently flexible, diversified, and responsive that it can weather shocks, can respond to and indeed create opportunities for growth, and can, on its own, generate continually increasing welfare for its people" (Bruton, 1988, p. 1602). Virtually all successful exporters of manufactures except Hong Kong began their industrialization with import substitution under significant protection. Protection in support of import substitution may be usefully thought of as a temporary, and obviously costly, device to assist in the restructuring and development of an infant economy. Its costs are unfortunately, in practice, more certain than its ultimate benefits. Protection ... is a means of inducing diversification and the learning upon which development is based. More accurately, perhaps, it is a means of

416

G. K. HELLEINER

creating a process of development that builds on research and learning. The goal is to create an economy with the capacity to move in various directions as opportunities are provided and new knowledge is accumulated. When such capacity exists, the economy can then seek to concentrate or specialize, because with such capacity it can more readily opt out of a declining sector into one that is recognized to be expanding. If that capacity is lacking, rapid, unanticipated changes are likely to impose major costs on the community. (Bruton, pp. 1605-6) protection generally imposes short-run costs on the economy . . . The shorter the period of import substitution as a development strategy ... the less its cost and the higher the return on the investment. This . . . means that the ending or phasing out of import substitution is an essential part of the strategy itself. (Bruton, p. 1607) "Sound" import substitution policies therefore avoid needless distortions and defiance of traditional static allocation criteria, and promote learning processes. The simple replication of foreign productive processes on domestic soil under heavy protection is unlikely to be fruitful. That policy-makers typically know little about scale economies, externalities, or the prospects for learning and productivity growth in particular industries is frequently taken as argument for their offering uniform incentives for all. Moreover, first-best policy would involve targeting subsidies directly toward the source of social gain. But some knowledge does exist in autonomous and well-run government bureaucracies in some countries; and first-best policies may be fiscally or otherwise difficult. The interindustry dispersion of incentives in the "successful" East Asian industrializing countries was high; industrial policies have sometimes been both selective and successful (Westphal, 1982; Pack and Westphal, 1986; Wade, 1990). Still, protectionism in developing countries has too frequently been unjustifiable on any reasonable economic arguments, and recognition of these policy errors has resulted in strenuous efforts in many countries significantly to reduce its costs. It is ironic that at the very time that so many developing countries were shifting to more "open" and less protectionist trade policies, the developed countries were embarking upon their "new protectionism." Notes 1 Part I draws heavily upon C.F. Diaz-Alejandro and G.K. Helleiner, "Developing Countries and Reform of the World Trading System" in D. Salvatore (ed.), The New Protectionist Threat to World Welfare (North-Holland, 1987).

Protectionism and the developing countries

417

References Baldwin, R. (1992), "High-Technology Exports and Strategic Trade Policy in Developing Countries: The Case of Brazilian Aircraft" in G. Helleiner (ed.), Trade Policy, Industrialisation and Development: New Perspectives, Oxford: Clarendon Press. Bhagwati, J.N. (1958), "Immiserizing Growth: A Geometrical Note", Review of Economic Studies, 25, June. Bruton, H.J. (1967), "Productivity Growth in Latin America," American Economic Review, 57, 1099-116. (1988), "Import Substitution," in Hollis B. Chenery and T.N. Srinivasan (eds.), Handbook of Development Economics, Amsterdam: North-Holland. Buffie, E. (1991), "Commercial Policy, Growth and the Distribution of Income in a Dynamic Trade Model," Journal of Development Economics, 37(1/2), November. DeMelo, J. and S. Robinson (1990), "Productivity and Externalities: Models of Export-Led Growth," World Bank Working Paper 387. Feder, G. (1983), "On Exports and Economic Growth," Journal of Development Economics, 12, February/April. (1986), "Growth in Semi-Industrial Countries: A Statistical Analysis," in H. Chenery, S. Robinson, and M. Syrquin (eds.), Industrialisation and Growth, A Comparative Study, Oxford: Oxford University Press. Grossman, G.M. and E. Helpmann (1991), Innovation and Growth in the Global Economy, Cambridge, Mass.: The MIT Press. Havrylyshyn, O. (1990), "Trade Policy and Productivity Gains in Developing Countries, A Survey of the Literature," World Bank Research Observer, 5(1), January. Hazledine, T. (1990), "Trade Liberalization and Economic Welfare with Endogenous Market Structure," Department of Agricultural Economics, University of British Columbia Working Paper. Johnson, H.G. (1967), "The Possibility of Income Losses from Increased Efficiency or Factor Accumulation in the Presence of Tariffs," Economic Journal, 11, March. Levy, S. and S. Nolan (1991), "Trade and Foreign Investment Policies under Imperfect Competition: Lessons for Developing Countries," Journal of Development Economics, 37(1/2), November. Nishimizu, M. and S. Robinson (1984), "Trade and Productivity Change in Semi-Industrialized Countries," Journal of Development Economics, 16, 177— 306. Pack, Howard (1988), "Industrialization and Trade," in Hollis B. Chenery and T.N. Srinivasan (eds.), Handbook of Development Economics, Amsterdam: NorthHolland. (1992), "Learning and Productivity Change in Developing Countries," in G. Helleiner (ed.), Trade Policy, Industrialisation and Development: New Perspectives, Oxford: Clarendon Press. Srinivasan, T.N. and L. Westphal (1986), "Industrial Strategy and Technological

418

G. K. HELLEINER

Change: Theory Versus Reality," Journal of Development Economics, 22(1), June, 87-128. Rodrik, D. (1992), "Closing the Technology Gap: Does Trade Liberalisation Really Help?" in G.K. Helleiner (ed.), Trade Policy, Industrialisation and Development: New Perspectives, Oxford: Clarendon Press. Tybout, James (1991), "Researching the Trade-Productivity Link: New Directions," PRE Discussion Paper, World Bank. Wade, R. (1990), Governing the Market, Economic Theory and the Role of Government in East Asian Industrialisation, Princeton University Press. Westphal, L.E. (1982), "Fostering Technological Mastery by Means of Selective Industry Promotion," in M. Syrquin and S. Teitel (eds.), Trade, Stability, Technology and Equity in Latin America, New York: Academic Press. World Bank (1987), World Development Report, Washington, DC. (1991), World Development Report, Washington, DC.

CHAPTER

18

Trade liberalization - the new Eastern Europe in the global economy JOZEF M. VAN BRABANT

This chapter draws the basic markers for the potential role that the new Eastern Europe, 1 and in some respects the East2 more generally, can prospectively play in the global economy. It also sketches the backdrop against which the concept of European economic integration - ultimately Europe tout court - can be renovated, giving full weight to the East. I am doing so mainly in connection with the concerns that arise from the modifications in trade and payment regimes during the transition. They cannot but exert a determinant role on the eventual place of these countries in the global economy. Section 1 sets the broad context of the revolutions in the East and what they portend with regard to the "new Europe." Next I examine the collapse of the Council for Mutual Economic Assistance (CMEA3), which was formally acknowledged on June 28, 1991, and the transferable-ruble (TR) trade and payment facilities that these countries maintained for over forty years.41 also point out there the options from among which the PETs could choose with a view to maintaining order in their already convoluted societal transition processes. Section 3 explores one way in which the post-CMEA can be dealt with by the former members individually and in some regional concert as well. This way, as further detailed in the next section, would also better prepare most of the East for smoother assimilation into the Western integration schemes and for withstanding global competition. By way of conclusion, I evaluate the chances of the preferred option advanced here being pursued any time soon to the benefit of integrating the East on an equal basis into the global economy through closer association with, perhaps even membership in, the European Community (EC). 419

420

JOZEF M. VAN BRABANT

1

The Eastern revolutions

Late 1989 provided a uniquely euphoric moment in post-war Europe. The annus mirabilis for the smaller six countries of Eastern Europe was remarkable by any measuring rod. Communism as implanted there utterly collapsed, an awkward acknowledgment after a protracted agony. The new political leaders there immediately declared themselves bent on realizing soonest two overriding goals: political pluralism and marketbased decision-making with firm property rights in an open trading environment. These commitments augured well for enacting measurable changes in the political and economic makeup of Europe, and perhaps the world economy. For some time, until early 1991, optimism with regard to the management of global affairs in general and the bright outlook of the East in particular were pervasive in policy stances. The scarcely bounded passions around the exultant political changes have since markedly waned and the nitty-gritty aspects of how best to conceptualize, introduce, and steer the transformations in the context of feasible societal changes now constitute the core of political discourse. As a result, a third variable has been introduced: the whole array of transitional issues of how best to get from where these societies find themselves at this juncture to a state with democratic stability and orderly markets. One aspect of this transition is how the East fits into the global scheme of things. The nature of the trade and exchange regimes of the more radically reforming countries, here referred to as the planned economies in transition (or PETs - no pun intended), after bridging the transition is one useful area of inquiry. But this is neither the purpose of this chapter nor would it be the most interesting of topics to study. Because the goals of establishing functioning market systems revolve around complete submersion into the global economy, and hopefully into the EC, the critical question following the transition is how "open" these economies will be. For the rest, they will presumably blend into the spectrum of trade and payment regimes of moderately mature economies. A more interesting area of inquiry, then, concerns the trade and foreign exchange regimes that will crystallize in the process of erecting the bridge between past and future. Certainly, this issue received a tentative answer in the context of the political ambiance of 1989 and early 1990. But the framework within which the matter was then debated, in retrospect, proved to have been far too simplistic, if not downright naive. This is particularly so now that the wherewithal for CMEA collaboration has completely collapsed and the TR trade and payment facilities have vanished. Finding a plausible and workable alternative to the backbone of the principal trade

The new Eastern Europe in the global economy

421

and payment modalities of the past forty years merits high priority on the policy agenda of those managing the transition. Overdue is some perspective on what is feasible and how finding a plausible alternative to the trade and payment modalities of the past forty years could be helped along through international assistance, so that the PETs could begin to play a constructive role in the global economic context. That should benefit not only the region. It would also open up markets for other countries. The salutary advantages would become visible rather rapidly in Western Europe. If the policy does not explicitly deny access to the new Europe to outside competitors on economic grounds, sizeable benefits should also accrue fairly soon for these other partners, provided they are willing — and able — to compete openly for such a slot. For those with a genuine interest in economic stability in the East, the inconclusiveness of the drawnout deliberations about CMEA reform that began around 1983, culminated in promising decisions in principle in 1986-7, and then stalled around 1988, has been a real challenge. Suggestions on how progress could be carved out in spite of the many inhibitions provided by the variety of parameters within which compromises, by necessity, had to be devised have been legion. When I first became seriously concerned in 1989 about the economics of the needed changes in the East's trade and payments arrangements, given the imminent failure of the reform movement of the preceding decade or so, little did I suspect that three seminal changes would very soon come to the fore. Not only that, they would radically alter in several respects the potential for international economic cooperation. One was embodied in the sequence of unprecedented political revolutions in Eastern Europe proper in late 1989 and early 1990, and the desire of these countries soonest to install market economies (MEs) broadly anchored to pluralistic decision-making. Because this could not but modify the framework of broader policy-making, it too was bound to change the fundamental character of relations with other countries. At the very least, it would call for less politically motivated, more "businesslike" transactions. The second was the complete collapse of the CMEA as a regional economic organization that had provided some coherent, if rather primitive, framework for the vast bulk of the East's trade and payments. Although this was formally acknowledged only in mid 1991, the pervasive effects of the abrupt disintegration of the CMEA framework on feasible levels of trade and payments have been felt since early 1990 at the latest. Any of the constants of east-south-west relations for so long accommodated through the mechanisms of the CMEA were simply relegated to the scrapheap of history. This new character of international economic relations, therefore, deserves to be briefly reexamined. Finally, the utter disruption of normal economic, political, and social

422

JOZEF M. VAN BRABANT

relations in the Soviet Union was in the late 1980s being entertained only as a rather remote possibility. It certainly was then not yet the prevailing reality that it is today. Regardless of whether the union (excluding the Baltic republics) can be held together in the years ahead, the uncertainty emanating from instability in the remaining Soviet republics is exerting a powerful influence on the feasible levels of the East's intragroup trade. I shall not delve into the problems aroused by the Soviet economic and political conundrum, beyond noting that the setbacks with perestroyka, however defined, have considerably complicated maintaining desirable levels of trade with Eastern partners. Also the nature of the political revolutions in the East will be touched upon only when they have a bearing on the CMEA's collapse. The paramount feature of the East is that these countries have been passing through an unprecedented wave of political and institutional changes since mid 1989. Most are now envisaging or have already embarked upon genuine market-oriented reforms. Many are hoping to find common ground with the integration schemes of Western Europe with a view to becoming full EC members. Whereas political achievements have been breathtaking and many of the aspirations of the PETs' leaders are very far-reaching, economic realities argue strongly in favor of a more sober approach, especially in the short to medium run. As experience to date has amply underlined, before the PETs can fully fuse themselves into the Western community they must enact major changes in economic structures broadly defined by revamping the entire traditional framework of economic decision-making. Given the lingering mutual interdependence of these economies for the core of industrial production, in spite of the recent erosion, activating a meaningful economic transformation at a high social cost in the presence of dysfunctional relations stemming from fundamental economic and political disarray in partner countries is not at all an easy task. The envisaged changes have very important consequences for how the East can eventually merge into the world economy. Some exert an overwhelmingly negative impact on economic relations. Others provide new opportunities for mutually advantageous economic interactions, certainly in the medium term, relative to what could have been intimated if these countries had continued to muddle through with degenerative administrative planning. These beneficial impacts of the revolutions in the East will challenge not only Western Europe but also the more dynamic of the developing countries into innovating imaginative arrangements for clearing trade until a "normal" currency can be embraced. Finally, the events in the East provide an opportunity not only for remaking Europe but also for reflecting on the basics of international economic management as it has limpingly evolved during the post-war period.

The new Eastern Hurope in the global economy

423

The negative implications are coming to the fore under several headings. First and foremost is the renunciation of most development assistance. This is exerting tremendous external shocks especially on countries that traditionally benefited from the lion's share of the CMEA's bilateral and multilateral development assistance. Cuba, Mongolia, and Vietnam in particular, but others too, are now confronted with sharply reduced assistance levels on top of considerable problems in clearing existing levels of trade because of the demise of the TR regimes. With the prospect of the creation of functioning markets in much of Eastern Europe over the next few years, barring all too onerous untoward circumstances, the old and new countries affected find themselves in a peculiar bind. They would like to promote indirect economic coordination as an essential ingredient of the transition toward MEs. But they are seriously handicapped in doing so quickly by domestic and external constraints. As far as their reciprocal trade and finance are concerned, the countries desire henceforth to conduct all of their commerce in "world market" conditions. This includes the use of world prices, payment in convertible currency, and conventional payment terms. As the experience so far in 1991 underlined, such a move is being inhibited by the lack of foreign exchange and the deep-seated reluctance to earmark it for transactions formerly denominated and settled in TRs. This entails serious obstacles to conducting trade that is otherwise economically warranted. 2

The collapse of the CMEA

Since about mid 1989, the range, speed, and depth of changes in political structures and economic aspirations in Eastern Europe in particular have been extraordinary by any measure. One of the surprising common denominators of the extraordinary events of late 1989 was the CMEA's de facto disintegration. Whatever role it played as the regional economic organization entrusted with enhancing socialist economic integration and ensuring stability in trade and payment flows among its members over the preceding four decades, this came abruptly to a stunning halt. This was in marked contrast to what should have emerged during 1989 (Brabant, 1991b), namely, seminal steps toward further refinement and gradual implementation of major changes in integration strategy and mechanism. Not only did these measures fail to emerge as planned, by the time the revolutions were under way whatever had been worked out by way of renewing the CMEA was rejected out of hand, most notably at the 45th Council Session (Sofia, January 9-10, 1990). CMEA members had also begun to sharply constrain trade. Private arbitrage was inhibited from exploiting the marked relative price differentials and consumer-market

424

JOZEF M. VAN BRABANT

imbalances among the various countries from aggravating already strained supplies. Regular commercial transactions were being rescaled because of emerging TR imbalances that neither the surplus country nor the deficit country was bent on eliminating swiftly. The result has been a gradual contraction of within-group trade in 1989, a precipitous fall of at least one fifth in 1990, and a rapid erosion of the residual trade by about half so far during 1991. There are many reasons for this overt flouting of trade commitments. They can be classified under four headings: the collapse of the TR regimes, the economic transition in PETs, the post-CMEA chaos, particularly in the Soviet Union, and political obstacles.5 Regarding the replacement of the TR regimes, the substitution of economic agents for central ministerial powers has shifted demand and supply schedules. The switch to world prices has entailed sharp one-off terms-of-trade adjustments. And the replacement of TR clearing with "world-market" conditions has been much more difficult than it had originally been anticipated. These autonomous changes in the trade and payment regimes have, of course, been aggravated by the three other factors mentioned. I shall only look into the direct effects of the transition in PETs. The ongoing reforms in individual countries are cutting not only purchasing power but also changing demand and supply schedules, beyond the effects already noted, as independent economic agents are increasingly engaging in foreign trade on their own account. Also, volatility in prices in world and CMEA trade and indeed in individual PETs, including on account of exchange-rate manipulations, have been complicating decision-making, whatever its basis. Further, unanticipated shifts in output and export priorities, including that of energy in the Soviet Union, have aggravated bilateral imbalances. The unprecedented domestic policy measures inaugurated to stem exports to the former CMEA have presented a major hurdle in maintaining trade that has a solid economic rationale. Debilitating though the above events were for trade and domestic economic stability, the CMEA malaise ran much deeper. The probability of any major change being enacted to foster more buoyant economic relations for the group as a whole quickly evaporated. There were political realities for this, such as the protracted deterioration in the Soviet political climate and politicians of several PETs seriously questioning - and quickly rejecting - the CMEA's very raison d'etre. This was not just frivolity emanating from the ambience to publicly question at length virtually any sociopolitical issue. Rather, it recognized that the CMEA as a regional organization and much of what it had stood for in the post-war period had not really figured among the key factors that determined the well-being of

The new Eastern Europe in the global economy

Alb

the members. But that was distinct from the real trade dependence among the member countries. With the political shifts, the PETs began to pursue market-oriented mutations in structures and institutions. These could not but change the level, commodity composition, and geographical direction of trade. In the medium to long run, these transformations were bound to entail significant trade creation. While laying the foundations of the restructuring, however, difficulties of clearing intragroup trade could linger as these countries attempted to regain some semblance of economic stability. Averting either a collapse in economic buoyancy, because of the inability to maintain economically warranted trade, or a major rise in trade diversion, which cannot but complicate adjustment problems elsewhere, would have been paramount in working off the worst outcome. It is for that reason that from the very beginning of the transformation I have advocated (Brabant 1990a, b; 1991a) the usefulness of exploring ways of reforming intragroup trade and payments from within. The real room for maneuver in Eastern Europe after the events of late 1989 could not have been a singular function of economic factors. With the de facto collapsing of TR regimes, the CMEA countries in early 1990 resolved in principle to scrap the TR and replace it with "world-market conditions," as explained. Also, they decided for themselves to accelerate their integration into the world economy by forcefully reducing their reliance on intragroup trade. This gave rise to two vexing conundrums. One concerned how to accomplish this trade diversion at minimal cost in terms of feasible economic buoyancy. The other questioned how henceforth to conduct intragroup trade in convertible currency. Trade diversion could not but be disruptively costly in the short run, given that these countries were tightly interlinked in their trade - roughly 40 to 80 percent of overall trade until 1989 was cleared among themselves. 6 Furthermore, the economic structures in place, including the capital stock, could be redesigned for Western markets and many of the skilled workers retrained in support of such a renovated production-f#/ff-export profile only at a high cost. Much of the available knowledge and embodied technology would need to be scrapped altogether if significant trade diversion were to be sought in the short to medium run. Furthermore, the transportation infrastructure (including railroads, oil and gas pipelines, roads, electrical powergrids, and ships) is still overwhelmingly oriented toward intragroup, particularly Soviet, relations. The alternative to this predicament was to amortize gradually those assets that could still produce competitive goods for ex-CMEA clients but for which there was little, if any, demand in world markets and to retrain employees over an affordable period of time. Even if these countries wanted to incur such a cost immediately because they

426

JOZEF M. VAN BRABANT

thought it politically opportune to assiduously pursue this goal, for practical purposes the adjustment could only be enacted over time. As a result, at buoyant activity levels, there was bound to remain a considerable demand, for example, for components, spare parts, maintenance, and service, to keep the relatively obsolete machine park running, at least to supply intragroup demand. Furthermore, unloading and processing facilities for critical fuels and raw materials other than those directly linked to regional supplies simply could not accommodate quick trade diversion. Even if a dramatic change in the geographical distribution of trade had been an explicit policy objective in its own right, the paramount technical issue of how to accommodate intragroup relations in the interim remained to be addressed. This immediately elicited concerns about the transition away from the trade and payment regimes that were the hallmark of intragroup relations throughout the post-war period. Regarding the transformation of the TR regimes, the preferred solution should have separated transactions among truly reforming PETs from transactions among and with lukewarm reformers. Only if enterprises of market-oriented countries, once domestic prices were aligned with those observed in East-West trade (to avoid excessive arbitrage that is undesirable from the macroeconomic point of view), have the authority to negotiate genuine prices7 and quantities of transactions on the basis of their own profit motive could one properly speak of market-clearing or "world" prices in intragroup trade.8 These would have meshed with the structure of world prices since the latter were slated to filter through directly into the reforming domestic economy. Microeconomic agents in such an economy should not be concerned about the selection of exchange rates and the settlement of imbalances - both tasks for macroeconomic management. Placing such trade on a dollar-cash9 basis provoked two apprehensions in the short run. One stemmed from the need to raise the size of current transaction balances of scarce foreign exchange to settle and finance trade and effect non-commercial transactions in the post-TR world.10 The other arose from the distinct possibility that the country might exacerbate its deficit, perhaps because of terms-of-trade losses or changes in the level, composition, and direction of trade emanating from the reform itself (Brabant, 1987a). The severity of both policy concerns could be eased through some multilateral clearing that enforces the discipline of moving steadily toward convertible-currency relations, yet helps in economizing with scarce foreign exchange in the short to medium run. Perhaps much more fundamental than short-term dependence was the presence of static comparative advantages within the ambit of the CMEA. Because producers previously had had only limited opportunities to explore

The new Hastern Hurope in the global economy

All

each other's markets, trade revolved essentially around the administratively determined needs of the Soviet Union, and ministerial bureaucrats tended to favor trade in investment goods and raw materials, considerable unexplored static comparative advantages continued to exist.11 To exploit them, proper institutions, policies, and commercial-policy instruments were required to guide the PETs into choosing trade lines in which they had or could hope soon to have a true comparative advantage. Furthermore, as national reforms stimulate structural change, also dynamic comparative advantages could have been expected to crystallize for independent firms to exploit. Processes that were really hard to justify on solid economic grounds had to be phased out as quickly as circumstances allowed.12 Others that ought to have been fostered were in need of more suitable accommodation than these countries had until then been able to muster. Diverting trade to the West was a variable in the hands of the PETs' policy-makers, provided they were willing to bear the consequences. For intragroup relations, however, individual PETs could only filter through into trade negotiations as much as possible the emerging rationality in the domestic economy and take advantage of opportunities for reform (see Brabant, 1990b). These are particularly important in fostering genuine economic cooperation within Central Europe, provided a solution could be found for the imbalances associated with the emergence of full-fledged markets. The argument for greater regional economic and payment cooperation with Western assistance could be enormously strengthened if other Eastern countries were to commit themselves fully to instituting market-oriented reforms. This would offer the international community a unique opportunity to look for a nearly regionwide solution for ushering all of these countries into the global economic framework. Perhaps the key issue then as now is reaching a common understanding of what the current trading and payment constraints of the PETs are. The economies bent on rapidly creating markets could disengage from their primary trading partners through trade diversion or by placing their intragroup trade on a convertible-currency basis without severely cutting absorption. This proposition has since been fully borne out: the PETs have been encountering marked problems in clearing reciprocal trade even if based on "world" conditions. At the very least, the post-TR world has been giving rise to potentially serious current-account problems, including because of the shifts in terms of trade with the Soviet Union and the worsening outlook for current trade and payments with MEs. Not only that, the switchover is being "managed," involving bilateral barter and clearing transactions at various levels, as well as cash-based deals. These various ad hoc arrangements currently in effect cannot but debilitate proper decisions on restructuring the PETs, hence regaining a more promising

428

JO2EF M. VAN BRABANT

economic growth path than the one just abandoned largely for political reasons. The result of this rebilateralization of trade and payments in the East, in some cases down to the enterprise level, has been trade destruction, dragging feasible levels down to the lowest common denominator and inhibiting the potential usefulness of abandoning the TR regimes. But the external-payment problems of the East are by no means confined to shifts away from the TR regimes. Some external payment problems could have been resolved through a payments union; those of structural origin would have had to be taken care of by other means. A payments union, especially among PETs, would have multiple purposes. The core short-run objective should be bolstering opportunities for trade through multilateral arrangements, even if for now that means only containment of the trade destruction that would otherwise occur. The essence would be the replacement of bilateralism with a framework that accommodates trade considered profitable to the economic agents that are willing to engage in it. For the PETs, it would be critical to find support for revamping intragroup economic relations (level and commodity composition as well as direction of trade), supporting more buoyant economic interchanges on the basis of rational economic decisions at the same time that structural adjustments in domestic economies are introduced at a measured pace, and facilitating the stepwise integration of these economies into the global-policy framework, including through convertibility. One obstacle in fostering such transition in CMEA relations in favor of more flexible trade is precisely the shortage of foreign exchange. Another is the absence of the institutional infrastructure to clear such trade, including settlement facilities, export-credit guarantees, and rediscounting. These are inauspicious features for moving toward full-fledged markets, including currency convertibility. As a technical device, especially if anchored to a customs union among PETs, a payments union is by definition discriminatory, hence possibly inimical to facilitating the integration of participants into the global framework. That would be so if potential participants had a clear choice between multilateralism and a payments union. As developments since early 1990 have borne out, this is not now the case. The option facing PETs at this juncture continues to be largely either a modest external financial facility or a combination of trade destruction with a fairly generalized rebilateralization of intragroup trade. A payments union can help to reach the goals set only through tight surveillance to ensure that the discrimination remains within bounds and is eliminated over time, while addressing temporary foreign exchange constraints. If a payments union can do it at an expected cost inferior to the anticipated burden of adjusting otherwise, it might be worth exploring.13 A clearing institution with some supervisory

The new Eastern Hurope in the global economy

429

muscle would seem to offer a useful intermediate policy choice for PETs, however confined the initial arrangements might have to be. 3

Trade and payments in the new East

There can be no doubt that the East is facing very serious constraints on external payments for several years to come, regardless of the economic policies that can be pursued or the largesse of Western "donors," which are for the most part really lenders. There could only be four solutions to this quagmire. One would be simply to cut out most of the traditional intragroup demand and supply through drastic deflationary policies. A second avenue would be rapid trade diversion to the West, regardless of the sociopolitical costs arising from any recession that might be engineered. Third, the latter predicament could conceivably be averted through structural-adjustment loans either from the official multilateral organizations or contracted in commercial markets with perhaps Western government guarantees. This could work if the absorptive and adaptive capacity of the PETs in the 1990s were noticeably better than in the preceding two decades. Otherwise, borrowing would only postpone - and exacerbate - much-needed adjustments by piling up foreign debt. Finally, room could be explored in which the conventional trade and payments of the PETs warranted on economic grounds would be maintained while partners undertake rapid adjustment in foreign trade and exchange mechanisms. One way to foster such a process and assist with the establishment of markets in the PETs would be setting up a constructive mechanism to buttress efforts to establish current-account convertibility. Although the latter option could be pursued through a variety of payment facilities, I have been arguing (notably Brabant, 1990a, b; 1991a, c, d, e) that it might be highly constructive for East and West alike to set up a payments union preferably in conjunction with a customs union (or even higher form of intragroup integration) for the PETs. Clearly, a payments union can only function properly when participants do not encounter chronic structural imbalances, regardless of their origin. They may arise from an irreversible worsening of the terms of trade, a cataclysmic change in export revenues because of forced trade diversion, or a sharp slowdown in economic activity in MEs, and the general climate of uncertainty engendered by the transition. Remaining, then, is the situation where the transition is associated largely with its own difficulties as well as those emanating from changing the TR regimes. The central purpose of a payments union is to accommodate bilateral imbalances by transforming them into multilateral ones and ensuring that the net surpluses or deficits vis-a-vis all participants remain manageable.

430

JOZEF M. VAN BRABANT

Manageable in this context means that the clearing agency should have sufficient funds to finance imbalances and ample supervisory clout to "guide" partners into mutually reinforcing trade behavior. In this connection, it needs to be reemphasized, as all past attempts to multilateralize CMEA trade and payments (Brabant, 1987a, pp. 273—8) have amply shown, that any clearing scheme that does not allow for the rapid emergence of more rational trade prices will undermine any multilateral clearing, regardless of its technical sophistication. Of course, many grave questions in connection with the proposal to establish a Central European Payments Union (CEPU) remain. I shall not delve into them here, if only because I have had ample opportunity to air my views elsewhere and details, while important, should not deflect from the major purpose of the proposal. This is twofold. One is to provide (partial) financing of trade that cannot now be conducted for lack of adequate foreign exchange and institutional settlement mechanisms, including export credits. The other is to keep the process of market-oriented transition going as fast as circumstances - not just the situation regarding external payments - permit so that the PETs can rapidly integrate as MEs into "Europe" and indeed global trade and finance. The main features of such a union are the following. First of all, it should be strictly temporary with a fixed duration of, say, five years. Regarding the technique of clearing, the tasks of the Clearing Agent could be entrusted to an existing institution, such as the Bank for International Settlements. Participants would grant each other unlimited credit until net balances are periodically reported to the Clearing Agent; with modern telecommunications, this could be done daily, if desired. If so, only the Clearing Agent would have to levy proper interest charges, essentially those of international markets with perhaps a marginal edge to encourage surplus countries to keep their balances on deposit with the union. Third, settlements of net imbalances would be conducted in a combination of loans and payments within the provisions of an overall quota against which participants can draw. The quota should usefully allow for multiple tranches of hardness and asymmetry for creditors and debtors to encourage intragroup exports. The precise shares and their movement over time toward full convertiblecurrency payment need to be negotiated. To make such a scheme manageable, as defined, there must be external financial support and supervision. Financial support is required to fund the asymmetry of settlements and thus to offset net claims on the union. Both TR debts and current imbalances existing prior to the inception of the union should preferably be worked off outside the union, given the disarray in TR prices at which these imbalances were originally incurred, or be revalued at more realistic prices. Also prior convertible-currency debts should be addressed

The new Hastem Hurope in the global economy

431

outside the payment framework, as argued. At the same time, it should be conceived in such a way that, on balance, the demand for resources from the central fund required for the orderly conduct of flexible trade decreases over time as a steadily rising proportion of the cumulative imbalances is being paid in convertible currency to augment the resources available to the fund. In my proposal, the external support and supervision should be under the aegis of the EC, as argued in the last section. Regarding the discriminatory character of the payments union, it is important to stress that the PETs presently maintain trade controls that discriminate more against some partners than others. Inasmuch as a payments union would help to eliminate the perfidious influence of bilateralism, it could not but raise multilateralism. This would reduce the absolute level of discrimination in favor of domestic economic activity, including by guiding participants to free remaining obstacles to trade as much as possible. The incidence of relative discrimination may temporarily favor participants in the union, either at the start of the union or to keep it manageable. This would be especially pertinent to coming to grips with negative imbalances with the union that are associated with positive ones with outside partners. Tight supervision would be required to minimize the worsening of relative discrimination (Brabant, 1991a, pp. 172-8). Neither can it be denied that a customs union, or higher form of economic union, among PETs would in principle be discriminatory. However, the basic aims of such a union would be to facilitate the clearing of trade volumes that are economically warranted, to encourage economic agents to exploit existing comparative advantages, and to foster dynamic trade gains through economic reform. In that sense, the union would essentially be trade creating, once the contraction on account of weeding out unprofitable intragroup trade has been completed. Trade creation would initially be rather strong for union members. But in time, successful reform as well as the prodding of the supervisory authority over the payments union could not but generate additional trade for outsiders too, thus encouraging further integration of participants into the global economy. The lure of eventual full accession to the EC figures prominently in this scheme of things, of course. The emerging trade regime can, therefore, be critical in the orderly preparing of these countries for accession. 4

The trade regime of the transition

Inasmuch as the PETs desire to integrate themselves quickly and fully into the international economy, they must pay special attention to molding the proper (and manageable) trade regime. As pointed out, the

432

JOZEF M. VAN BRABANT

anchoring of more effective relative prices and the rapid strengthening of competition in domestic markets also play a critical role in moving the transition forward at minimal cost. For that, measures have to be embraced that liberalize the trade and foreign exchange regimes. For one thing, the overwhelming use of export and import quotas, possibly through explicit foreign exchange licensing, will have to be replaced by a trade regime in which normal instruments of commercial policy, foremost tariffs and transparent exchange-rate policies, play a core role. The same applies to a plethora of opaque instruments that provided protection under central planning. The advice given to reformers has ranged from opting soonest for the freest kind of trade and foreign exchange policies (for example, Bofinger, 1991a, b, c; Polak, 1991) to instituting considerable trade controls on a precommitted temporary basis (for example, McKinnon, 1991; Williamson, 1990). Rapid trade liberalization of rather rigid economic systems harbors the risk of destroying much of the productive capacity of the PETs, as the example of the ex-GDR (Protzman, 1991; Schill; 1991) has so tragically demonstrated. Not only that, it might considerably weaken the sociopolitical consensus upon which the acceptable hardship of the transition is predicated, especially in countries that will have to shoulder the burden of the transition themselves. Maintaining that rock bottom as robust as possible is as critical to credible reform as a full understanding of what markets can and cannot accomplish for PETs. On the other hand, suggestions for heavy protection would seem to lean to the other extreme, namely slowing down reform too much for fear of jeopardizing the transition. McKinnon's scheme is the most elaborate and articulate. It suggests converting very early on in the transition all protection into explicit ad valorem tariffs restructured into several - perhaps half a dozen - cascading categories. The absolute level of protection would initially remain all but unchanged. Furthermore, the PET would commit itself in a precise time schedule to reduce these tariffs and eventually narrow their dispersion. The objective of such an exercise would be to forestall the collapse of large segments of industry (especially the so-called "valuesubtracting" or "negative value-added" enterprises) early on during the transition. Whereas some protection of existing economic structures could usefully be entertained, it would be counterproductive to allow the PETs to erect formidable trade barriers. Whereas it would permit policy-makers to alter economic structures gradually, hence avoid sudden dislocations with all attendant consequences, it would also inhibit competition and the transfer of an effective system of relative prices for traded goods. Furthermore, in view of the weak central actors in PETs, credibility for the phasing-out of

The new Eastern Hurope in the global economy

433

customs duties, even if a waiver under the GATT rules for all Contracting Parties (all but Albania, Bulgaria, and the Soviet Union), could be secured, would be rather low. Moreover, it is probably an insuperable task to make explicit the degree of protection that wide segments of the planned economy traditionally enjoyed. Rather than aggravate price distortions and introduce levies that may be hard to repeal later on according to a set schedule, the PETs could resort to a "regular" trade regime anchored to comparatively low ad valorem tariffs as quickly as administratively feasible and enact income transfers to sectors temporarily adversely affected. This would permit sectors potentially threatened by bankruptcy to operate below real cost, yet the government would not have to support full unemployment claims. By proceeding in this way also, valuable assets that could survive on their own strength under the reformed principles, once in place, would not be lost during the transition. There is, of course, little hope that macroeconomic policy-makers of the transition can neatly separate the potentially profitable firms most deserving of subsidies from others that in any case have to be fundamentally restructured or eliminated altogether. But some guidance can be taken from firms that are able to cover their variable costs under changing circumstances. Regarding the foreign exchange regime, the perplexing multiplicity of exchange rates must urgently be replaced by a single rate. Whether this should be fixed or flexible is one of those "either-or" choices that many economists and inexperienced policy-makers are fond of. In practice, the policy question will be more about what kind of crawling peg should be entertained and at what intervals adjustments in the exchange rate should be undertaken to avoid imparting adverse credibility effects. The issues at stake are paramount to gage progress toward meeting the conditions for current-account convertibility. Only under those circumstances is there a realistic chance of attracting the kind and volume of foreign direct investment (FDI) that most PETs are hoping to muster.14 One key item on the agenda for transition is whether the exchange rate should be undervalued to foster exports and discourage imports or devalued by just enough to anticipate the "corrective inflation" likely to ensue from price liberalization. Should it instead be set closer to the purchasing-power level to avoid a sharp cut in the real value of wages and incomes? Whereas there are no unambiguous answers to these questions, most PETs have opted for an exchange rate that is undervalued as compared to the rate that would maintain balance between demand for and supply of foreign exchange for merchandise and related transactions. But it is less clear how much that undervaluation should be. Policy-makers appear to be opting for erring on the side of undervaluation.

434

JOZEF M. VAN BRABANT

Inasmuch as the PETs desire to integrate themselves quickly and fully into the international economy, special attention must be devoted to the reconstruction of their trade and foreign-exchange sectors. But this is desirable also for other reasons. The anchoring of a more effective relative price system, at first for traded goods, and the rapid increase of competition in domestic markets, both of which can be acquired from international markets, also play a critical role in moving the transition forward at minimal cost. For that, measures have to be taken to liberalize the trade and foreign exchange regimes. In the former economic setup, the planned economies practiced pervasive protection against import competition through a variety of instruments, overwhelmingly in the form of export and import quotas. These were at times embedded in and implemented through explicit foreign exchange licensing, such as in the habitual bilateral trade and payments agreements of the post-war period. Many other forms of protection were even less transparent. Virtually all such trade impediments, with minor exceptions, will have to be replaced by a trade regime in which normal commercial-policy instruments, foremost tariffs and exchange-rate policies, play a critical role. The key questions to be addressed in transforming the foreign exchange regime will, therefore, revolve around the kind of crawling peg to be chosen and at what intervals adjustments in the exchange rate according to preannounced criteria should be undertaken to avoid imparting adverse credibility effects. The issues at stake are paramount in eventually meeting the requirements of credible current-transaction convertibility (Brabant, 1991c). It is less clear how much that undervaluation should be at the outset. Again, conditions of time and space will probably determine the margin to be chosen. Thus, if the corrective inflation is likely to be substantial, then the margin by which the exchange rate will be devalued beyond its perceived instantaneous equilibrium level would presumably have to be greater for any set of time horizons for exchange-rate stability. The commitment to a firm crawling peg is not yet in place. Thus, Poland utilized the nominal exchange rate and nominal wages as anchors for the entire transformation process; Yugoslavia did likewise for the first half of 1990. Poland initially committed itself to defend the newly chosen exchange rate (9,500 zloty to the dollar) for several months. In the end, it kept that rate unchanged for much longer (until May 13, 1991, when the first modest devaluation of 17 percent was enacted) in spite of a very sizeable differential in movements of domestic and trade prices. This suggests that the initial devaluation was much too large. Also, the rate could be maintained because of the extraordinary fortuitous confluence of factors, including the sharp

The new Hastern Hurope in the global economy

435

depreciation of the real wage without eliciting a social response, the de facto moratorium on servicing the debt, agents having sizeable foreign-exchange reserves, the very sharp contraction in industrial activity, a once-and-for-all gain in access to markets in Western Europe, and, of course, the extraordinary political situation in Poland since late 1989. These changes in the foreign exchange regime may form part and parcel of what is necessary to transit to sustainable current-account convertibility. If such a credible move can be made, perhaps with financial assistance from regional and international financial institutions, that would be a desirable policy choice. But this can be so only in two cases. Countries that have utterly lost control over their economies and where agents no longer have faith in the leadership may make current-account convertibility one component of the "shock therapy." Current-account convertibility would also be desirable for countries that have gone a long way toward relaxing domestic pricing, wholesale trading, foreign-trade licensing, and foreign exchange allocation (perhaps through retention quotas and auctions). On the other hand, countries that enter their transition with substantial imbalances or will soon face up to sizeable structural adjustments under impulse of price liberalization and enterprise reform may well elect to postpone the inauguration of current-account convertibility, as distinct from liberalizing foreign exchange allocation mechanisms, until completion of the most important adjustments. Since the old CMEA system anchored to TR clearing and pricing has died, it is of more than passing interest for PETs to avoid falling into another trap, namely to try to bilaterally balance their mutual transactions. A rebilateralization of the trade among the former CMEA members, let alone resorting to a whole array of ad hoc bilateral arrangements, in some cases down to the enterprise level (that is, pure barter), would be utterly counterproductive to the introduction of markets in these economies.

Conclusions There is no doubt that the sociopolitical revolutions in the East have resulted in an unprecedented shift in relations on the European continent. These offer excellent opportunities for reassessing the context of "European" integration and reinvigorating the movement. They also offer, or perhaps impel, us to look objectively at what is feasible in PETs and to what degree these countries could play a constructive role in the global economic system in general and the remaking of Europe in particular. This might call for a different impetus to European integration than the one that the EC has recently been most concerned about. In addition to recasting this PET integration scenario into a much wider framework, dovetailing the

436

JOZEF M. VAN BRABANT

desirable with the feasible in the East itself may call for a different approach to all-European integration. From the outset of the East's revolutions, a core ambition has been the "return" to Europe. Although left rather vague, this goal included quick fusion into the Western integration schemes, the EC's in the first instance. Although such a continent-wide economic union might be very desirable and its realization might be posited as the horizon that current policy should hold, it cannot be accelerated beyond the pace allowed by real conditions. There are many, but here I touch only on three: politics, governance, and economic compatibility. True, from the very beginning Western policy-makers in general and the EC in particular have been nominally very supportive of the democratization of the East and their ambition to put market systems into place. These supports have included financial and other assistance, lending greater credibility and strength to the reform. But the irrevocable commitment to ushering most of the East into the EC has yet to be made. Whether the EC will be prepared to opt for widening itself, regardless of the speed and plumbing of the preferred deepening, has not yet been decided. Even once such a decision will have been made, practical issues of decision-making within the EC will have to be faced anew. Consensual governance will become more difficult with a community encompassing two dozen or even more members. Another constitutional reform of the EC will be required to allow some "representative government" to emerge. It is unlikely that this concept will soon be supported by the twelve. Pouring it, then, into an operational mold, as experience has shown, will be an even more delicate and tedious task. Finally, bringing new members into the EC has economic consequences. Though the opportunities for trade creation and the diffusion of Western technological capabilities throughout the East are bright, they cannot bear fruit immediately. Bringing poorer countries into a wealthy club impoverishes the founders asked to share the burden of levelling. So far, the EC has resorted to the option of income distribution, notably in the case of the accessions of the countries on Europe's southern rim. For as long as the Cold War reigned, there were good strategic and political reasons to support such burden sharing, especially if its magnitude could be confined. Thus, it is comparatively easy to subsidize a Greece or Portugal, or even a Spain. But it is vastly more difficult altogether to do likewise for a group of countries that have a population of one third to more than 100 percent of the EC's, depending on where the EC's geographical boundary will eventually be drawn. Not only will there have to be burden sharing of the type engaged in within the context of German unification, chiefly by Western Germany.

The new Hastern Europe in the global economy

437

Promoting open markets in the East can succeed only if the EC relaxes its constraints against many Eastern products. This would be the case implicitly by bringing the East into the EC, although gradual assimilation into the EC's regimes may be one condition for membership. It would have to be explicitly recognized in any lesser form of association than membership. The tortuous negotiations of the "European" association agreements with the Central European countries have demonstrated the root obstacles to the EC's relaxing its trade restraints and fostering on a minor scale the kind of adjustments that the East has been advised to undertake on a major scale. Some compromise solution will necessarily have to emerge, lest the revolutions in the East will be dissipated for lack of political foresight and will. As I have advocated here and elsewhere, a combination of greater self-help by the PETs individually and in their groupwide concert with a gradual, if bold, revision of the EC's philosophy would seem to offer the only realistic course. Whether such a cooperative strategy will suffice to prop up the transitions of the East to pluralistic market systems that benefit from a sociopolitical consensus remains to be seen, however. Strategic support from the EC for measures that assist the PETs to exploit self-help from within each country and especially among these countries (Brabant, 1991f, g, h) would be one paramount way to ensure greater credibility of the transformation strategy. That in and of itself might be worth striving for at the earliest opportunity.

Notes The opinions expressed here are my own and do not necessarily reflect those that may be held by the United Nations Secretariat. This paper was completed in October 1991. 1 In what follows, Eastern Europe denotes Bulgaria, Czechoslovakia, the German Democratic Republic (GDR), Hungary, Poland, Romania, and the Soviet Union. Occasionally, I use the concept of the first six countries, in which case the context makes it clear that I have that subgroup in mind. Of course, in considering the future of Eastern Europe as a geographical entity with distinct political characteristics, the ex-GDR as a country is no longer to be reckoned with and the Soviet Union is unlikely to revert to the situation existing before August 19, 1991. However, the lingering economic problems between Eastern Europe and the ex-GDR have not yet been satisfactorily resolved. And the problems of the successor Soviet Republics, including the Baltic states, are similar to the issues dealt with here. 2 In what follows, East is understood as Eastern Europe (including the Baltic republics) plus Albania and Yugoslavia (including any successor republics). The latter has been among the reforming countries from the very beginning. The

438

3

4

5 6

7 8

9

10

11

12

JOZEF M. VAN BRABANT

former has been emerging as the latest member of that club. Neither played a significant role in "Eastern Europe" for at least three decades. However, in redesigning the East and in considering how the group fits into the European scheme of things, neither Albania nor Yugoslavia should be ignored. I prefer this acronym to the admittedly more euphonic Comecon. The latter is not only a politically loaded term chosen in analogy with Comintern and Cominform, it is also an incomplete abbreviation. In what follows, I am concerned solely with the active European members - Eastern Europe as defined. Certainly, the TR was created only in 1963. In many ways, however, it was a simple successor to the former clearing ruble used in CMEA relations from early on in the post-war period. The anatomy of the CMEA's collapse is very complex indeed. All I can do here is flag up key elements. For a broader picture, see Brabant (1991a, g, h). Trade data at official exchange rates, especially of reporters that clung to national exchange rates (mainly Bulgaria and the Soviet Union), tend to overstate the CMEA's share. Even if corrected for that distortion, Eastern Europe was extraordinarily interdependent. Not just emulating prices charged by "leading" Western firms, as painfully practiced in the past (Brabant, 1987b, pp. 113-30). Clearly, if firms from unreformed countries had full authority over prices and quantities, firms from PETs should not be inhibited from arbitrating the irrational prices in partners. But macroeconomic policy-makers must still be concerned about settlements arrangements. This normally means cash payments for fuels, raw materials, and foodstuffs but 90 to 120 day supplier credit for exports of most manufactures, as distinct from the immediate payment system for transactions conducted in TRs in effect until late 1990 (Brabant, 1987a, b). Either transaction could be placed on a long-term loan basis, however. But this would not fundamentally change the asymmetry in payment relations. However, financing would be required for which the present infrastructure is woefully inadequate (Bond, 1991). Very special arrangements for such transactions, ranging from tourism to unrequited transfers, were in effect until late 1990 (see Brabant, 1987a). The simplest solution would have been to satisfy that demand in parallel markets and to ensure that the technical infrastructure to effect such payments be instituted soonest. My favorite simile is the Benelux of the late 1940s, whose market size was only about one third of a viable Central Europe consisting of Czechoslovakia, Hungary, and Poland; slightly more than one fourth if some accommodation could have been found for the trade of the former GDR with Eastern Europe. It is true, however, that in the post-war period the Benelux countries had limited opportunity to engage in free multilateral trade and payments, given Western Europe's constraints. That is not to say that such activities should have been suppressed forthwith. The reforming economy might have found it cheaper to subsidize such production over some period of time than to provide social welfare benefits in a

The new Eastern Europe in the global economy

439

situation where supply is rather inflexible in the short run and the danger of chronic unemployment, as a result, serious. 13 Whether the gain might be of the same order of magnitude as achieved with the European Payments Union is immaterial to the key issues at hand. 14 Note, however, that current-account convertibility is neither a necessary nor a sufficient prerequisite for attracting FDI, as suggested by the small number of countries, especially developing ones, that have a convertible currency, yet have attracted FDI (see Brabant, 1991a, pp. 101-3). References Bofinger, Peter (1991a), "The Transition to Convertibility in Eastern Europe: A Monetary View," in John Williamson (ed.), Convertibility in Eastern Europe, Washington, DC: Institute for International Economics. (1991b), "A Multilateral Payments Union for Eastern Europe?" Banca Na^ionale del Lavoro Quarterly Review, 1, 69-88. (1991c), "A Payments Union for Eastern Europe? - A Reply," Banca Na^ionale del Lavoro Quarterly Review, 1, 99-100. (1991d), "Options for a Payments and Exchange-Rate System in Eastern Europe," European Economy, special issue No. 2, 243—61. Bond, Daniel L. (1991), Trade or Aid? Official Export Credit Agencies and the Economic Development of Eastern Europe and the Soviet Union, New York: Institute for East-West Security Studies. Brabant, Jozef M. van (1987a), Adjustment, Structural Change, and Economic Efficiency - Aspects of Monetary Cooperation in Eastern Europe, New York and Cambridge: Cambridge University Press. (1987b), Regional Price Formation in Eastern Europe — Theory and Practice of Trade Pricing, Dordrecht-Boston-Lancaster: Kluwer. (1990a), Remaking Eastern Europe - On the Political Economy of Transition, Dordrecht-Boston-London: Kluwer. (1990b), "On Reforming the Trade and Payment Regimes in the CMEA," Jahrbuch der Wirtschaft Osteuropas — Yearbook of East-European Economics, 14/2,

7-30. (1991a), Integrating Eastern Europe into the Global Economy - Convertibility Through a Payments Union, Dordrecht-Boston-London: Kluwer. (1991b), "Trade Policies in the Council for Mutual Economic Assistance (CMEA)," in Dominick Salvatore (ed.), Comparative Handbook of Regional Economic Policies, Westport, CT: Greenwood Press. (1991c), "Convertibility in Eastern Europe Through a Payments Union," in John Williamson (ed.), Convertibility in Eastern Europe, Washington, DC: The Institute for International Economics. (1991d), A Central European Payments Union: Technical Aspects, New York: Institute for East-West Security Studies. (1991e), "Key Problems of Creating a Central European Payments Union," Banca Na^ionale del Lavoro Quarterly Review, 2, 119-49.

440

JOZEF M. VAN BRABANT

(1991f), "Assistance to Eastern Europe and European Economic and Monetary Integration," De Pecunia, 3, 65-94. (1991g), "The Demise of the CMEA - The Agony of Inaction," OsteuropaWirtschaft, 3, 234-54. (1991h), "Renewing Economic Cooperation in Eastern Europe and Foreign Assistance," Economic Systems, 15(2), 243—64. McKinnon, Ronald I. (1991), "Liberalizing Foreign Trade in a Socialist Economy: The Problem of Negative Value-Added," in John Williamson (ed.), Convertibility in Eastern Europe, Washington, DC: Institute for International Economics. Polak, Jacques J. (1991), "Convertibility: An Indispensable Element in the Transition Process in Eastern Europe," in John Williamson (ed.), Convertibility in Eastern Europe, Washington, DC: Institute for International Economics. Protzman, Ferdinand (1991), "Germany's Top Banker Gives Europe a Warning," The New York Times, March 20, D2. Schill, Wolfgang (1991), "Pioneers in Establishing Convertibility: The Case of East Germany," in John Williamson (ed.), Convertibility in Eastern Europe, Washington, DC: Institute for International Economics. Williamson, John (1990), "Convertibility, Trade Policy and the Payments Constraint," paper prepared for conference "The Transition to a Market Economy in Central and Eastern Europe," organized by Centre for Cooperation with the European Economies in Transition and the World Bank, Paris, November 28-30.

Index of authors

Adedeji, A., 368 Aharoni, Y., 368 Akhtar, M.A., 306 Anderson, J.E., 242 Aoki, M., 158 Arrow, K., 242, 368 Aschheim, J., 306 Backmann, H., 217 Backus, D., 368 Bailey, M., 306 Balassa, B., 97, 198 Balassa, C , 97 Baldwin, R., 79, 97, 178, 332, 417 Basar, T., 368 Bauer, D., 198 Beath, J., 368 Bergsten, C , 79 Berliner, D.T., 334 Beveridge, W., 51 Bhagwati, J.N., 12, 51, 52, 332, 333, 417 Binmore, K., 368 Boadu, F.O., 333 Bofinger, P., 439 Bond, D.L., 439 Borrus, M., 158 Boulton, M., 217 Brabant, V.J., 439 Brady, N., 217 Brander, J.A., 178, 333, 368, 370

441

Branson, W.H., 97 Bruton, H.J., 417 Buffie, E., 417 Burks, A.W., 369 Caves, R., 158, 217, 333 Chamberlin, E., 52 Chenesy, H.B., 417 Clemenz, G., 369 Cline, W., 12, 79, 158, 307 Coates, D., 51, 307 Cooper, R.N., 198 Corden, M.W., 12, 52, 79, 288, 333 Coughlin, C , 333 Cushman, D., 307 De Grauwe, P., 307 de Melo, J., 333, 417 Destler, I.M., 333 Diaz-Alejandro, C , 416 Dixit, A., 178, 369 Dornbusch, R., 242, 333 Driffill, J., 368 Easton, S., 218 Eaton, J., 52, 178 Edgeworth, F.Y., 50 Eichengreen, B., 52 Eitheir, W.J., 333 Elliot, K.A., 334

442

Index of authors

Erb, J., 158 Ergas, H., 368 Ericson, W.A., 369 Feder, G., 417 Feenstra, R., 51, 178 Finger, J.M., 333 Flamm, H., 178 Frame, G.D., 369 Fransman, M., 369 Frenkel, J., 79, 242, 307 Friedman, M., 242, 307 Fung, K.C., 369 Gal-Or, E., 369 Gardiner, W.H., 242 Gerschenkron, A., 369 Giavazzi, F., 242 Giovannini, A., 242 Golstein, M., 79, 307 Gotur, P., 307 Graham, E., 217 Greenwald, B., 242 Grinols, E.L., 333 Grossman, G.M., 52, 178, 288, 369, 417 Hamilton, C.B., 332 Hathaway, D., 12 Havrylyshyn, O., 417 Hazlendine, T., 417 Helleiner, G.K., 418 Helpman, E., 333, 417 Henderson, H.D., 50 Henderson, J.V., 178 Herander, M.J., 333 Hicks, J., 52 Hilton, S., 306 Hirschmann, A., 178 Hong, P., 99 Horn, H., 333 Horstmann, I., 178 Hufbauer, G.C., 158, 334 Hughes, H., 79 Irwin, D., 51, 52, 333 Jacquemin, A., 158 Jackson, J.H., 334 Johnson, H.G., 51, 52, 198, 242, 288, 307, 417

Jones, R., 12, 217, 288, 332 Josling, T.E., 242 Kaldor, N., 52 Kalt, J.P., 158 Katsoulacos, Y., 368 Keesing, D.B., 79 Kelly, M., 79 Kemp, M., 52, 198 Keohane, R.O., 97 Kenen, P.B., 307, 332 Khatkhate, D., 369 Klein, L.R., 127 Koray, F., 307 Kreps, D., 369 Krueger, A.O., 79, 333, 334 Krugman, P.R., 12, 52, 158, 178, 217, 333, 334, 369 Kyle, A., 178 Lai, D., 79 Laird, S., 242 Lastrapes, D., 307 Lawrence, R., 198 Levinshon, J.A., 369 Levy, S., 417 Lewis, T., 51 Li, L., 369 Macdonald, R., 308 Magaziner, I., 158 Magee, S.P., 334 Marks, S.V., 334 Markusen, J., 178 Marshall, A., 49, 50 Masao, W., 369 Maskus, K.E., 308 Mayer, W., 288 McArthur, J., 334 McCulluch, R., 218 McKinnon, R.I., 12, 242, 243, 440 McMillan, J., 198 Meade, J., 308 Meese, R.A., 242 Mesa, D.D., 370 Messerlin, P.A., 334 Michaely, M., 12, 334 Milgrom, P., 370 Mill, J.S., 52 Mino, K., 370

443

Index of authors Mokyr, J., 370 Morita, A., 243 Mueller, D.C., 98 Mundell, R., 265 Munk, B., 288 Murkowski, F., 217 Murphy, K., 178 Mussa, M., 242, 288 Myrdal, G., 178 Narin, F., 369 Nelson, R.R., 158, 368 Nishimuzu, M., 417 Noland, M., 198 Noland, S., 370 Novshek, W., 370 Odell, J.S., 334 Ohlin, B., 48 Ohno, K., 243 Okimoto, D.I., 159 Olson, M., 97, 334 Ono, Y., 370 Pack, H., 417 Palmeter, D.N., 334 Patrick, H., 49, 50, 52, 159, 332 Pauly, P., 127 Pavitt, K., 159 Pearson, G., 334 Perloff, H., 179 Petersen, C.E., 127 Polak, J.J., 79, 440 Porter, M.E., 179, 370 Prestowitz, C.V., 334 Protzman, F., 440 Qian, Y., 308 Ramachandran, R., 336 Reidel, J., 334 Ricardo, D., 17 Richardson, J.D., 334 Roberts, J., 370 Robinson, J., 52 Robinson, S., 417 Robson, J.A., 216 Rodrik, D., 307, 418 Rogoff, K., 242 Rosentein-Rodan, P., 179

Rosovsky, H., 159 Rousslang, D.J., 334 Rubio, L., 198 Salvatore, D., 12, 13, 308, 332, 334, 368, 369 Samuelson, P., 52 Sapir, A., 332 Sato, R., 370 Schill, W., 440 Schriesheim, A., 370 Schumpeter, J. A., 370 Schwartz, J. B., 333 Scott, N., 371 Shape, M., 159 Shleifer, A., 178 Smith, A., 53, 179 Smith, D., 218 Smook, N., 50 Sohmen, E., 308 Sonnenshein, H., 370 Spencer, B.J., 178, 333, 368, 370 Srinivasn, T.N., 52, 417 Stegemann, K., 79 Stern, R., 335 Stiglitz, J., 242 Su, V., 127 Sugimoto, M., 370 Syrquin, M., 418 Swain, D.L., 370 Takacs, W.E., 335 Tarr, D., 333, 335 Tassey, G., 370 Taussig, F., 159 Tavlas, G.S., 306 Teitel, S., 418 Thursby, J.G., 308 Thursby, M.C., 308 Torrens, R., 49 Tracy, M., 395 Tsutsui, S., 370 Tybout, J., 418 Tyson, L.D., 158, 159, 333, 335 Ulan, M., 306 Ulph, D., 368 Varangis, P., 306 Venables, A., 179

444

Index of authors

Vernmulst, E.A., 334 Viner, J., 198 Vishny, R., 178 Wade, R., 418 Waelbroek, J., 79 J ' m TI ' Wann, H.Y., 198 Warnecke S.]., 159 Watt, J., 217 Westphal,L.,417,418 Willett, T.D., 334 Williamson, J., 79, 439, 440 Wilson, R., 369

Wolf, B., 53 Wolf, M., 79 Wonnacott, P., 198 Wonnacott, R., 198 ^ KQ Yamamura, K., 159 .. ' ' v Yano, M., 243 Yeats, A., 242 T

Young, L., 334 Zis, G., 308 Zysman, J., 158

Subject index

Adjustment assistance, 313, 314 Agricultural subsidies, 3, 231 Agriculture, protection, 231-4, 385-8, 392, 393

Airbus Industrie, 4, 213, 214 Aircraft industry, 4, 385 Antidumping meeasures, 239, 316, 377—80 Automobile industry, 322, 323, 383, 384 Balance of payments, 245-7 Balance of trade, 74, 222, 244 Barter trade, 252 Beggar-thy-neighbor policies, 2 Bilateral trade, 180-97, 312 Border taxes, 232 Brazil, 9, 66, 67, 131, 187, 190 Bretton Woods system, 7, 231, 292 Cartel (see International cartel) Centrally planned economies, 82, 420—37 CMEA (see Council of Mutual Economic Assistance) Commercial policy, 374-6 Common Agricultural Policy (CAP), 9, 64, 65, 231, 386, 390 Comparative advantage, 5, 41, 148 Computer industry, 4 Concorde, 5 Constant returns to scale, 161, 272 Consumer electronic industry, 384, 385

445

Cost of protectionism, 62 Council of Economic Advisors, 313, 315 Council of Mutual Economic Assistance (CMEA), 100, 420-9 Countervailing duty, 71, 90 Countervailing power, 328 Currency overvaluation, 7, 292 Currency undervaluation, 7, 292 Customs union, 260-5 Deficit in balance of payments, 60-3 Developing countries, 63—7, 207—9, 403-16 Differentiated products, 236 Dillon Round, 2, 376 Direct foreign investments, 200-15 Dollar overvaluation, 59, 327 Dumping, 71, 239, 315, 316 EC (see European Community) Economic development, 56, 57, 406 Economic integration, 208, 419 Economic growth, 56, 57 Economies of scale, 5 Effective tariff protection, 2, 184 Escape clause, 313 European Community (EC), 8-11, 100-27, 231-40, 371-93 European Free Trade Association (EFT A), 100, 185

446

Subject index

European monetary system, (EMS), 87, 231 Export pessimism, 10, 66 External economies, 166—76 Fair trade, 17—41 Fixed exchange-rate system, 57-60, 86, 349 Flexible exchange-rate system, 59, 60, 87, 221-41, 248, 291-305 Foreign exchange rate, 59, 60 Free trade, 18-36, 42-8 Full reciprocity, 25, 312 GATT (General Agreement on Tariffs and Trade), 1, 3, 77, 150, 151, 180-6, 396-406 Generalized System of Preferences (GSP), 9, 76-8, 376 Heckscher-Ohlin (H-O) theory, 246 High-tech industries, 4 Hong Kong, 9, 70 IMF (see International Monetary Fund) Import restrictions, 88 Import substitution industry, 66, 415 Import tariff, 250-2 Increasing returns to scale, 167—9 Industrial policy, 135-49, 160-77 Industrial targeting, 176, 177 Infant industry argument, 32—4, 414, 415 International cartel, 85, 239 International debt problem, 63, 64 International macroeconomic policy coordination, 2 International Monetary Fund (IMF), 12, 58, 397 International monetary system, 86 International Trade Commission, 313—7 International trading system, 3 Intra-industry trade, 195—7

LINK system, 99-103 Macroeconomic policy coordination, 2 Malaysia, 65 Mature industries, 3, 414 Mercantilism, 1 Mexico, 187-90 Most-favored-nation principle, 64, 76, 398 Multifiber Agreement, 64, 321 Multilateral trade, 180, 183-5, 403-6 Multilateral trade negotiations, 2, 3, 11, 80, 371, 376, 392-4 Dillon Round, 2, 376 Kennedy Round, 2, 376 Tokyo Round, 2, 3, 376 Uruguay Round, 3, 11, 80, 371, 376, 392-4 Multilateral trading system, 4, 5, 184, 185 Multinational corporation, 5 New protectionism (see non-tariff barriers) Newly industrializing countries (NICs), 8, 9 Nominal tariff, 2 Non-tariff Barriers (NTBs), 1-10, 319-25, 329-31, 396-406 North American Free Trade Area (NAFTA), 8, 331 OECD (see Organization for Economic Cooperation and Development) Oligopolistic markets, 23, 24, 83, 235, 236 Omnibus Trade and Competitive Act of 1988, 21, 313, 315 OPEC (see Organization of Petroleum Exporting Countries) Orderly marketing arrangements, 77, 312 Organization for Economic Cooperation and Development, (OECD), 63-7, 229-30 Organization of Petroleum Exporting Countries (OPEC), 85, 87

Japan, 4-11, 71-6, 135-41, 193-8, 336-66 Kennedy Round, 2, 376 Korea, 46, 57, 64, 131, 194 Laissez-faire, 35 Law of comparative advantage, 5, 72 Level-playing field, 8, 18, 153, 154

Preferential trade arrangements, 183, 374, 376, 397 Pressure groups, 328 Product differentiation (see differentiated products) Purchasing-power-parity (PPP) theory, 223, 291

Subject index Quantitative restrictions (see quotas) Quotas, 57, 224-6 R & D (see research and development) Ratchet effect, 59, 60 Reciprocity, 20-5, 33, 34, 36-48 Research and development (R & D), 139, 346-50 Ricardian trade model, 5, 17 Second best, theory of, 11, 19 Semiconductors industry, 85, 139, 176 Shipbuilding industry, 87, 138 Singapore, 9, 204 Smoot-Hawley tariff, 325 Soviet bloc countries, 422 Soviet Union, 10, 82, 422-4, 427 State trading, 375, 380 Steel industry, 322, 382 Strategic trade policy, 3-6, 163-6, 329-31, 343-65 Subsidy of exports, 87, 88 Synfuel, 5

447 Taiwan, 320 Tariff, 57-9, 244-65 Technical, administrative and other regulations, 1, 2 Technological progress, 338—43, 350—7 Telecommunications industry, 140 Terms of trade, 258-60 Textile and apparel industry, 321, 380—2 Tokyo Round, 2, 3, 376 Trade and Tariff Act of 1984, 90, 315 Trade liberalization, 419-37 Trade Reform Act of 1974, 80, 89 Trigger-price mechanism, 7, 90 United States, 1-11, 54-63, 81-4, 131-57, 186-93, 311-32 Uruguay Round, 3, 11, 80, 371, 376, 392-4 Volatility of exchange rates, 6, 59, 60, 223, 294-304 Voluntary export restraints, 55, 77, 238, 239