The Arc of Japan’s Economic Development
Interest in Japan has shifted over the years, from a focus on the miracle econ...
63 downloads
645 Views
2MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
The Arc of Japan’s Economic Development
Interest in Japan has shifted over the years, from a focus on the miracle economic growth of the post-war period through the 1970s, to a fascination with the apparent export powerhouse and then to a morbid curiosity into how and why Japan has slowed down. This core textbook provides an overview of the Japanese economy from the period before the Meiji Restoration, and its astonishing growth throughout the twentieth century, before conducting a thorough analysis of the contemporary scene and the implications of the recent financial crisis. The book begins by exploring Japan’s economic development from the 1600s to the end of the American Occupation, focusing on the circumstances behind the government’s preferences for intervention and guidance in economic affairs to achieve its objectives. It then traces the survival of this approach under the Americans. Separate chapters incorporate original research that illustrates the effects and gradual fading of that policy approach in the post-war era. Alexander reviews the foundations of modern Japan’s economic development and shows how the state’s central objective in the 1870s – to develop industrial capabilities to supply a modern military that could protect the country – led to government officials’ distrust of capitalism and a desire to guide economic affairs themselves, especially during war and post-war chaos. The author goes on to highlight how returns on capital declined for 100 years until 2000, how industry growth shifted to a new trajectory in the 1990s and how Japan’s economic characteristics grew to look more like those of the US. The final chapter projects the Japanese economy into the future, emphasizing the rise of diversity across regions, industries, companies and individuals. The research on industrial structure, capital returns and Japan’s similarities and differences from other countries breaks new ground. This book is particularly relevant to all those studying the contemporary economy of Japan, historical and economic development and economic structural change, among others, and will also benefit financial analysts. Arthur Alexander is a Visiting Professor of Economics at Georgetown University and Johns Hopkins School of Advanced International Studies.
The Arc of Japan’s Economic Development
Arthur Alexander
First published 2008 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN Simultaneously published in the USA and Canada by Routledge 270 Madison Ave, New York, NY 10016 This edition published in the Taylor & Francis e-Library, 2007. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Routledge is an imprint of the Taylor & Francis Group, an Informa business © 2008 Arthur Alexander Printed and bound in Great Britain by TJ International Ltd, Padstow, Cornwall All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data Alexander, Arthur J. The arc of Japan’s economic development/Arthur Alexander. p. cm. ISBN 978-0–415-70023-8 (hardcover) – ISBN 978-0-415-70024-5 (softcover) 1. Japan–Economic policy. 2. Japan–Economic conditions. I. Title. HC462.A4597 2007 338.952–dc22 2007015690 ISBN 0-203-79986-0 Master e-book ISBN ISBN10: 0-415-70023-X (hbk) ISBN10: 0-415-70024-8 (pbk) ISBN10: 0-203-79986-0 (ebk) ISBN13: 978-0-415-70023-8 (hbk) ISBN13: 978-0-415-70024-5 (pbk) ISBN13: 978-0-203-79986-4 (ebk)
This book, as with everything else that I have written for more than 40 years, is dedicated to my wife, Elaine, the motivating spirit behind it all
Contents
List of figures List of tables
1 Japan’s place in the contemporary economic world
viii xi 1
2 The nineteenth-century transformation of the Japanese economy 15 3 From Meiji to the Second World War: political and international developments 40 4 Economic developments in the first half of the twentieth century 48 5 Planning supplies for war
58
6 The American Occupation and the post-war economy
68
7 Bank-centred finance and corporate governance
92
8 Capital investment and rates of return
105
9 Structural change in the Japanese economy
122
10 The privatization of Japan’s public corporations
137
11 Is Japan different? Implications for economic growth
172
12 Economic prospects for Japan: moving to a new trajectory
188
207 211 212 221
Appendix Notes Bibliography Index
Figures
1.1 1.2 1.3 1.4
1.5 1.6 2.1 2.2 4.1 4.2 4.3 5.1 6.1 6.2 7.1 7.2 7.3 7.4
Japan’s global ranking by GDP per person, 1950–2000 5 GDP in Japan and the United States, 1886–2005 6 GDP per capita, Japan and the United States, 1886–2005 8 Average annual growth rate of real GDP per capita over the preceding 10 years and GDP per capita, 1975–2005 10 Average annual 10-year growth rate (%) of real GDP per capita for Japan, the United States and Switzerland, 1960–2005 12 Gap between US year and Japanese year in which Japan first attained US level of GDP per capita 13 Elementary school enrolment rates, number of teachers in kindergarten through high school and number of college and university students, 1873–1920 29 Manufacturing and agricultural shares of net national product, 1885–1940 (%) 34 Business failures (suspension of business transactions with banks) and inflation rate (% change of GDP deflator, two-year average at annual rate), 1895–1940 50 Machinery production (000 1934–36 ¥ and number of employees, 1880–1940) 55 Military spending to GDP and military personnel to total population, 1876–1944 (%) 56 Ratio of dividends to current profits, 1923–44 (%) 64 Coal production, 1945–50, million tons, annual rate 79 Inflation rate, wholesale price index, six-month change at annual rate, 1945–51 (%) 85 Sources of corporate external funds, 1935–80 (%) 93 Number of banks, 1920–56 95 Shares owned by type of shareholder, Tokyo Stock Exchange, 1949–2005 (%) 97 Share of total assets accounted for by bank loans, paid-in capital and bonds (large manufacturing firms), 1960–2005 (%) 102
Figures ix 7.5
Mergers and acquisition in Japan, 1986–2005 (number of transactions) 8.1 Real GDP per capita and real capital per worker, Japan, 1965–92, and 59 other countries in 1990 8.2 Ratio of private sector capital stock to private sector GDP (1955–2005, 1990 prices) 8.3 Real returns on total aggregate non-residential capital (Japan: 1910–2004; US: 1914–2005) 8.4 Real returns on private sector non-residential capital (Japan: 1955–2004; US: 1930–2005) 8.5 Ratio of operating profit plus interest to total assets, large corporations, 1960–2006 8.6 Returns on Japan’s direct investment assets abroad and foreign assets in Japan, 1980–2005 8.7 Returns earned on Japan’s direct investment assets abroad and by large, domestic manufacturing companies, 1993–2005 8.8 Returns on US FDI in Japan, all American FDI and Japan’s FDI abroad, 1980–2005 8.9 Ratio of capital payments to private non-residential investment, Japan and US, 1932–2005 8.10 Capital share of national income (Japan, US; 1950–2005, %) 8.11 Returns on Japan’s non-residential capital stock with alternative depreciation and elasticity assumptions, 1910–2004 9.1 Cumulative industry percentage of GDP growth versus cumulative industry percentage of GDP (1970–2005, 42 industries) 9.2 Modified Gini coefficients of industry growth contributions over preceding 10-year period, 1980–2005 9.3 Cumulative industry GDP growth (annual rate) as percentage of base-year GDP versus cumulative industry share of GDP (1970–2005, 42 industries) 9.4 Cumulative industry percentage of total growth versus cumulative industry percentage of output, 1980–90 9.5 Size distribution of 42 actual and simulated industries (2005, billion 1995 ¥, log scale) 9.6 Modified Gini coefficients of industry growth distribution over preceding 10-year period, actual and simulated, 1980–2005 9.7 Standard deviation of 10-year change of detrended industry GDP (trillion ¥, 1980–2005) 9.8 Cumulative real cost reduction as percentage of base-year GDP versus cumulative industry GDP (1970–98, 82 industries, %) 9.9 Cumulative industry percentage of GDP growth versus cumulative industry percentage of GDP (Japan: 1988–98, 42 industries; US: 1985–95, 66 industries)
104 106 107 108 109 111 112 113 114 115 119 120 123 127 129 130 131 131 132 133 135
x Figures 9.10 Cumulative industry percentage of employment growth versus cumulative industry percentage of employment (Japan: 1980–90, 24 industries; US: 1995–2005, 65 industries) 10.1 Domestic passenger yields on Japanese airlines (2000 ¥ revenue per passenger kilometre) 11.1 Real 1980 GDP per capita (2000 $) and 1980 distance from US 11.2 Annual correlations of 1980 distance from US with countries’ real GDP per capita, 1980–2003 11.3 Annual correlations of 1980, 1990 and 2002 distances from US with countries’ real GDP per capita, 1980–2003 11.4 Real GDP per capita growth as function of distance from US and initial income (shown on chart), estimated from growth equation 12.1 Average annual growth rate of real GDP per capita over preceding 10 years and GDP per capita (% and 2000 $ at purchasing power parity) 12.2 Real estate loans and asset prices in Japan, 1980–2007 12.3 Tangible capital stock on the books of large manufacturing and non-manufacturing firms, 1970–2006 (trillion ¥) 12.4 Ratio of interest-bearing debt to GDP, non-financial companies, 1964–2006 12.5 Employment (millions, seasonally adjusted, January 1984 to October 2006) 12.6 Growth rate differential between real GDP and total employment (five-year average, %) 12.7 Coefficient of variation of 10-year GDP change across 42 industries 12.8 Tokyo region urban commercial land price index 12.9 Measures of income inequality: Gini coefficient of household market and disposable income and national income share of top 5% individual taxpayers
135 168 183 184 184 186 190 193 195 196 197 197 201 202 204
Tables
2.1 Gross domestic fixed capital formation, shares of GDP and amounts financed by current account deficits (five-year intervals, 1885–1919, %) 3.1 Japan’s geographic expansion to the 1930s 4.1 Average growth rates of real GDP in selected periods, 1885–1940 7.1 Source of new funds of large Japanese firms, 1911–36 (%) 8.1 Real aggregate rates of return on non-residential fixed capital stock in selected countries, using alternative capital elasticities, 1955–90 (%) 9.1 Industry contributions to GDP growth, 1970–2005 (%) 10.1 Japanese government share sales of privatized corporations 11.1 Countries included in comparisons 11.2 Euclidean distances from the United States, 1980, 1990, 2002 11.3 Correlation coefficients with the United States, 1980, 1990, 2002 11.4 Position, distance and correlation of Japan relative to United States 12.1 Non-performing loans in Japan’s banks, 1999–2006 12.2 Sales rank, general merchandise stores, Japan and US, 1983–98 12.3 Gini coefficients for market and disposable income across population groups, 1985–2000
35 42 49 93 110 124 170 174 178 180 182 194 200 203
1 Japan’s place in the contemporary economic world
Prologue: the arc of the Japanese economy For 150 years, the Japanese economy followed the arc traced by a ballistic trajectory: an economy blasted from the muzzle of a Meiji-era cannon, an arc characterized by increasing, and then falling, economic growth, development and government intervention in the economy. Japan’s approach to economic affairs was conditioned by the circumstances surrounding the origins of the modern Japanese state in the Meiji Restoration of the 1870s. A central objective was to develop industrial capabilities to supply a modern military force that would protect Japan from the European and American colonial powers. The centrality of defence led to the dominance of the military in government and the primacy of military objectives. A crucial institutional feature that influenced events was that the Imperial Army was accountable only to the emperor, bypassing the prime minister and cabinet as well as the Diet, often placing army behaviour beyond the reach of political control. A combination of defence, political and economic motivations created a powerful logic for an expansionist imperialism. This logic was supported by mimicry of western imperialist practice, including the creation of colonies. The Japanese government promoted economic and military expansion in Taiwan and Korea. Japan’s military forces were in Manchuria from the early 1900s, at first to bar Russian expansion and then to develop Manchuria’s resources. To promote industrialization, especially the underpinnings of military industry, the state encouraged the import and then the indigenous development of technology and industry in fields such as steel, machine tools and shipbuilding. Led by military planners in the 1930s, the resources of Manchuria and China were sought to guarantee the availability of industrial inputs and to secure markets for finished goods. This policy became especially prominent when the Great Depression in the early 1930s suppressed Japan’s trade with Europe and North America, which had the side-effect of diminishing the western powers’ ability to influence Japan’s external behaviour. Drawing lessons from the large-scale fighting during the First World War, Japan’s military strategists saw the need for mobilization planning, led by the military itself and by like-minded government bureaucrats. Mobilization staff
2 Japan’s place in the contemporary economic world promoted the ideas of government authority over economic affairs. Manchurian industrial development in the 1930s, coordinated by these mobilization bureaucrats, trained a cadre of government officials to distrust Japan’s capitalists. Motivated by Marxist beliefs and the examples of state planning in the Soviet Union and Germany’s mobilization economy, the economic planners in Manchuria returned to Tokyo to organize the country for total war. Back home, they faced opposition to their policies and were unable to implement them as thoroughly as they desired, but they managed to alter the structure of the economy in important ways, especially in corporate finance and governance. Some scholars refer to these ideas, methods and institutions as the 1935–45 system, which continued to influence public policy and private economic activity for several more decades after the war. Looking backward, there is ample evidence to identify a linear sequence from the Meiji Restoration to the Second World War: defeat, Occupation and Japan’s post-war economic policy. However, it would be wrong to assert that there was an inevitable linear progression. At every step along the way, other paths could have been taken, paths that were vigorously advocated by many politicians and government officials. Until the military takeover of the government in the mid-1930s, Diet members and the electorate sought to rein in the military’s adventures and stop its more dangerous policies. Finance ministers attempted to control the army through the budget. Right up to Japan’s attacks on the United States and British possesions in December 1941 (Australia, Hong Kong, Borneo, etc.), an active political opposition tried to thwart military ambitions. However, when opposition to expansion looked like it might be effective, assassinations and coups ended such endeavours. Although Japan suffered a devastating defeat in 1945 and was ruled for seven years by an occupying army, the native bureaucrats’ methods of economic management that had prevailed for 10 wartime years did not disappear. The Americans chose to work through the existing organs of government. Through 1948 they reinforced Japanese bureaucrats’ inclinations to use the state to monitor the economy and intervene when it was deemed necessary or desirable. Industrial coordination was not a strange idea to American bureaucrats. In both the First World War and the Second World War, the US government created agencies to manage military production, allocate materials, and control prices and wages. The lessons learned from these experiences were that government had an obligation and ability to coordinate vast enterprises in times of crisis. The weight of the evidence is that direct government involvement in the immediate post-war economy, even with the backing of the Americans, most likely did not achieve its main objectives, even in the high-priority production of coal. Nevertheless, the commonly told story is one of successful government intervention. The myth that wise bureaucrats managed the post-war coordination problem became embedded in the official government consciousness and in the mind of the public. This view bolstered officials’ inclinations to guide economic affairs to meet varied objectives, at times strategic, but at other times political or personal. Though often confused, inappropriate and twisted by politicians and indus-
Japan’s place in the contemporary economic world 3 try, an interventionist-minded government presided over the Japanese economic miracle. Battles for dominance within the government weakened any sense of coherent strategy, but bureaucrats’ distrust of markets plus belief in their own vision continued to influence written and unwritten regulation and guidance. Of course, Japan is not unique among developed countries for its faith in bureaucratic judgments and a disinclination to rely on markets to achieve social goals. France and Germany come to mind as other examples of these tendencies. The 1935–45 system, which gradually lost force amidst the conditions of rapid post-war economic growth, eroded even faster in the 1970s and especially since the 1990s, when the economic environment and political institutions changed. In the economy, growth slowed, deregulation occurred in finance, transportation, retail and other areas, and foreign companies took a larger role in the domestic economy. Greater weight was given to shareholders as financial markets regained their importance in corporate funding. In politics, the shift in the 1990s to singlemember parliamentary districts and the increasing centrality of the media in elections altered party power structures and nudged politicians’ incentives towards broader, national interests. By the twenty-first century, Japan’s economy had moved away from its nineteenth-century origins.
The scale of the Japanese economy By any measure, Japan is in the top tier of economic powers. At the beginning of the twenty-first century, Japan’s economy was the third largest in the world, after the United States and China. Fifty years earlier, when Japan was rebuilding from the destruction of the Second World War, its gross domestic product (GDP) ranked number seven. High rates of economic growth following its wartime devastation propelled Japan into third position by 1966, behind the United States and the Soviet Union, where it remained until the dissolution of the USSR in 1990, when it moved up one position. However, China’s rapid growth displaced Japan in the mid-1990s, according to several different estimates. (Such international comparisons are made using purchasing power parities; see Box 1.1.) Not only did Japan’s economy get big, its people also became wealthy – Japan was the first Asian nation to join the European and North American club of rich nations. As measured by GDP per person, it moved from 32nd position in 1950 into the top 10 between 1985 and 1990. Finer distinctions are harder to make because they depend on the other countries included in the comparison and the particulars of how purchasing power parities are constructed. Figure 1.1 shows Japan’s ranking in terms of GDP per capita according to the Penn World Table (PWT 6.1; see Heston et al. 2002). Japan’s economic growth slowed in the 1990s and its relative position deteriorated: from number five in 1994, it slipped to 12th place by the end of the decade. Japan is not the only rich economy to slide down in the rankings. Switzerland, for example, which was in first place on the 1954 list, slipped to number seven by 2000. Japan and Switzerland have not become poorer, but other countries have grown faster and overtaken them; both are still among the world’s most affluent economies.
4 Japan’s place in the contemporary economic world Box 1.1 Purchasing power parity Comparing economic output across countries extends the concept of comparing the same country at different times. The national income framework – the system of national accounts – is the standardized statistical device for describing countries’ macroeconomic affairs. Gross domestic product is the most common measure of aggregate output in the national accounts. However, since prices are used to calculate the values of the millions of goods and services contributing to GDP, price changes from one period to another complicate assessments of changes in real output. Price indices such as the GDP deflator are used to deal with these problems; nominal GDP can be converted into ‘real’ values by dividing the nominal figures by the price index. Economists often make distinctions between nominal and real values. ‘Nominal’ refers to the actual, unadjusted, ‘named’ monetary amount; nominal values and prices are the everyday ones seen in shops, advertisements, annual company reports and interest rates. ‘Real’ values, as the term implies, try to get behind the price changes to measure values as if prices had not changed. To estimate real values, it is necessary to adjust the nominal values to account for price changes that may disguise changes in the output of stuff: of tangible goods and services. Real, price-adjusted values are proxies for the number of items produced or resources used in production; real interest rates, for example, subtract the rate of price inflation from the nominal interest rate to reveal the quantity of real expenditures that can be made at the end of a period with the nominal interest payments. The process of comparing different countries’ outputs begins by valuing each country’s output by its own set of prices, which presents each country’s GDP in its own currency. The next step converts one country’s output into the other country’s currency. The obvious method of conversion is to use the exchange rate, which defines the value of one currency in terms of another. The problem with using the market exchange rate is that it responds to forces that are unrelated to changes in domestic prices, such as international capital flows; therefore, the output of a country estimated with exchange rates would rise and fall with the currency markets, suggesting swings in a country’s output that are not, in fact, occurring. Economists have developed another method for converting one country’s output into the currency of another country by using so-called purchasing power parity (PPP). The ideal method for comparing the outputs of two countries would be to value each country’s production of final goods and services using the same set of prices. In practice, this is impractical because in any modern country there are millions of different goods, services and prices. Instead, a sample of representative prices is used as a substitute for the full array. The PPP exchange rate is computed by calculating the cost of a representative market basket of goods and services in the two countries. The PPP exchange rate between the yen and the dollar, for example, is the cost
Japan’s place in the contemporary economic world 5 in Japanese yen of a basket of goods and services representing total GDP divided by the cost of the same basket in US dollars. PPP, therefore, is the number of units of a country’s currency needed to purchase the same amounts of goods and services as one unit of the currency in the base country. Advanced countries with competent statistical agencies use more than 3,000 prices to generate PPPs. PPPs are usually superior to exchange rates in international comparisons, but there are exceptions. For transactions that cross national borders – international trade and capital flows, for example – exchange rates often give better measures of their economic effects.
Japan’s growth since the nineteenth century Observers of the Japanese economy in the 1960s and 1970s often used the term ‘economic miracle’ to describe the headlong growth of those years. However, a backward look over a longer horizon reveals an experience that is no less miraculous. Today, many countries in all parts of the world have made the transition from backwardness to modernity. Japan’s progress, however, was a first of its kind. In the later part of the 1800s, it deliberately set out to remake itself from a peasant economy into an industrial power, with the primary objective of generating the technology and industry required to produce the armaments to defend itself from the industrialized colonial powers. (This story will be discussed in detail in the next chapter.) The growth of the modern Japanese state is shown in Figure 1.2 along with that of the United States. To make this comparison, Japan’s GDP was converted first to constant 1990 yen and then to dollars using the OECD’s 1990 PPP of 195.3 yen
1
Rank GDP/capita
6 11 16 21 26 31 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000
Figure 1.1 Japan’s global ranking by GDP per person, 1950–2000. Source: Penn World Table 6.1.
6 Japan’s place in the contemporary economic world 10,000
GDP, bn 1990 $
US
1,000
Japan
100
10 1885
1905
1925
1945
1965
1985
2005
Figure 1.2 GDP in Japan and the United States, 1886–2005 (billion 1990 dollars).
per dollar. Note that the vertical axis is plotted on a logarithmic scale; each unit is 10 times the size of the preceding one. On such scales, the slope of a line is a measure of the growth rate; a steeper line means faster growth. Over the 40 years from 1886 to 1926, Japan grew at an annual rate of a bit more than 2.8 per cent. During those same years, American growth averaged 3.5 per cent and the United States overtook the United Kingdom to become the largest and most productive economy. American growth has been remarkably stable, except for the Great Depression of the 1930s and the Second World War, in which the United States was engaged from 1941 to 1945. Japan’s Asian-Pacific war escalated from battles in China and Manchuria in the early 1930s, to widespread warfare in China in the late 1930s, to full-scale warfare in the Pacific in 1941 against the United States and its allies, and defeat in 1945. An explanation for the comparative stability of American growth rates is based on a theory of economic development. One aspect of underdevelopment is that a poor country can grow rapidly by absorbing capital and technology from more advanced economies, if it possesses an appropriate institutional and human infrastructure. In such situations, investment leads to fast growth as the country catches up with its potential. A developing country is aided in this endeavour by the fact that the global economic leaders have already produced and demonstrated productive technologies and methods. For most of the period shown in Figure 1.2, the stock of capital in the United States was appropriate to its stage of development and its place on the technologi-
Japan’s place in the contemporary economic world 7 cal frontier. There was little to be gained from more rapid capital accumulation or from catching up with and imitating the global leader; it was the leader. Growth came from the slow accumulation of capital, the development of technology and the improvement of productivity. Before 1970, Japanese economic growth captured the advantage of catching up by deepening its capital base and absorbing technology from the more advanced countries. The country then reached maturity when productivity rather than capital accumulation should have become the major propellant to growth. However, Japan never fully shifted to this new mode of economic life – a subject that will occupy later chapters. America’s production decline and fitful recovery from the depression is clearly visible in the first half of the 1930s. Japan, in contrast, was much less affected by the global downturn as its attentions turned to wartime production, which drove industrial output for several years before the destruction wrought by aerial bombing and maritime disruption of supplies caused reductions in military output and then collapse in 1945, the year of Japan’s surrender. (Japan’s data for 1945 are largely conjectural; they were estimated by interpolations between adjacent years.) The final attacks on Japan in the Second World War destroyed production capacity, infrastructure and housing; disrupted supplies from former trading partners in Asia; and hastened the return of millions of military personnel and civilians from the war zones, former colonies and occupied territories. Production collapsed, jobs and incomes disappeared and hunger was rampant. By 1955, post-war reconstruction and the reestablishment of business relations at home and abroad enabled the country to embark on a spurt of very rapid growth. Real GDP growth, measured over 10-year periods to average out shortterm fluctuations, jumped above 9 per cent annually in the decade ending in 1960 and surpassed 10 per cent a few years later. A foremost reason for such a spurt is that the nation had preserved its human capital and basic institutions from the pre-war period, but suffered from severe wartime disruptions and capital destruction. Consequently, the economy was in extreme disequilibrium; when investment recovered and relationships re-established, the returns to the new additions of capital were very large. However, by the 1970s, the capital stock approached equilibrium levels, returns declined to values appropriate to the mix of human, physical and institutional capacities, and growth slowed. The 1970s marked a distinct deceleration of the post-war growth miracle. While still outstanding in international comparisons, 5 per cent growth in the 1970s seemed to be a near disaster to a people grown accustomed to doubledigit rates. The second half of the 1980s brought new joy to many Japanese, who believed that the good old days had returned when GDP expanded by 6 per cent annually; this burst of output, however, was not a return of the miracle years, but rather was associated with a bubble of asset prices, which stimulated investment and consumer demand to unsustainable levels. As is evident from Figure 1.2, the 1990s saw a further flattening of the growth curve and deceleration of the economy, which managed to eke out average yearly increases of just over 1 per cent.
8 Japan’s place in the contemporary economic world
Growth of output per capita The size of an economy does not necessarily reveal the economic welfare enjoyed by the nation’s people. In 2000, the PWT ranked China – the world’s second largest economy – only 89th in GDP per capita at 11 per cent of the US value and 15 per cent of Japan’s (PWT 6.1). GDP per capita measures economic wellbeing, although it is an imperfect device. It is a comprehensive account of the current output of goods and services, but does not reveal the distribution of income within the country; it does not show how widely gains are shared or whether economic rewards are reserved for a few favoured beneficiaries. Moreover, per capita GDP does not include pollution, environmental degradation or other ‘bads’ generated by economic activities. Consequently, it does not necessarily reflect lifestyle quality or its distribution. Nevertheless, it is commonly available in standardized definitions and formats, which permits comparisons over time and between countries. Recognizing its limitations, GDP per capita is one of the most commonly cited figures of economic welfare. Figure 1.3 shows GDP per capita for Japan and the United States for 120 years. The picture here resembles the absolute size of GDP pictured in Figure 1.2, with one notable exception. The Japanese curve more closely approaches the American one when we look at per capita GDP than when we focus on size alone. The reason for this difference is a faster rate of population growth in the United States than in Japan. In 1886, Japan’s productivity enabled it to produce only one-quarter of the US output per person. At the beginning of Japan’s military adventures in the early 1930s, it had raised itself to half the American level. With the setbacks of the war plus continued American growth, Japan was not able to reach its 1930 relative
GDP per capita, 1990 $
100,000
US
10,000
Japan
1,000 1885 1895 1905 1915 1925 1935 1945 1955 1965 1975 1985 1995 2005
Figure 1.3 GDP per capita, Japan and the United States, 1886–2005 (1990 dollars).
Japan’s place in the contemporary economic world 9 position until the late 1960s. However, the bonanza of compound growth brought about rapid convergence and Japan reached 80 per cent of US output per person by 1990; its subsequent slowdown ended the race towards the top, which many observers had predicted for Japan when extrapolating 1980s trends.
Slowdown and stagnation Japan’s growth rate of only 1 per cent in the 1990s surprised many people. This stagnation cannot be attributed to a single source; several severe economic problems struck the country at the same time. Some of these were new, but others had their roots in the very policies and procedures that had seemed to be so successful during the high-growth years. The following catalogue gives some idea of the severity and complexity of Japan’s problems as it entered the twenty-first century. • • •
• •
•
• •
Investment with little concern for profitability over the preceding 30 years left the country with too much capital; rates of return fell well below those elsewhere among the advanced economies. The decade-long collapse after 1990 of asset prices, primarily land and company shares, shook the foundations of the nation’s banks, on which most business finance depended. Government-owned financial companies competed with the private sector for deposits, mortgages, life insurance and loans to small businesses, thereby draining whatever profitability might have been available to the crippled private financial sector. The financial authorities had not been strengthened sufficiently to supervise a partially deregulated industry, a recipe in every liberalizing economy for financial promiscuity. Monetary policy delayed reducing interest rates, which eventually were pushed down to zero; when deflation set in owing to excess capacity, the Bank of Japan (BOJ) could not reduce short-term interest rates below its zero floor and it hobbled itself by a stubborn reluctance to pursue other approaches that might have increased the supply of money. The central government, burdened with collapsing tax revenues combined with modest spending increases and tax-cutting fiscal stimulation policies, faced the largest annual deficits and stock of government debt among the advanced countries, which inhibited vigorous fiscal policies. Excessive regulation hindered the creation of new businesses and industries, which slowed the re-employment of people, capital and other resources. Politicians in the Liberal Democratic Party (LDP) depended for support on many of the groups – banks, construction companies, small businesses – that would be hit hardest by policies to strengthen the financial sector, liberalize market regulations and enhance competition; therefore, there was little political enthusiasm for such vigorous policy measures.
10 Japan’s place in the contemporary economic world
2579.61 5.75 2785.60 5.50 10 4712.45 2.32 4978.69 1.45 8 5075.48 0.91 6 5393.97 0.04 5674.95 -0.29 4 5509.99 -0.17 2 5491.39 -0.35 0 5656.94 -0.74 5747.31 -1.44 -2 5893.84 -1.92 -4 5926.92 -1.68 Japan, 1995-2005 5751.38 -1.13 -6 5556.21 -0.92 -8 5417.22 -0.36 5510.13 -0.29 -10 5414.96 0.00 Source: World Bank -12 5301.40 0.12 5249.99 0.56 -14 4972.420 4,0001.67 8,000 12,000 16,000 20,000 24,000 28,000 32,000 4853.81 2.30 Real GDP per capita 5003.28 2.43 Figure 1.4 Average annual 1874.89 -1.24 growth rate of real GDP per capita over the preceding 10 years per capita, 1975–2005 (% and 2000 $ at purchasing power parity). 1873.23 and GDP-1.08 Source: World Bank. 1940.18 -1.57 2257.12 -2.34 2193.70 -1.42 2228.65 -2.14 2117.03 -1.71 2539.86 -2.72 Growth rate real GDP/capita
1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 1995 1996 1997 1998 1999 2000 2001 2002
The dominant reality that magnified the effects of these problems was the maturing of the economy. As growth inevitably slowed, shocks to the economy that at one time could be absorbed with only minor consequences now produced recessions and a cascading flow of new complications; for example, banks that had successfully reduced their burden of bad loans from the collapse of the asset bubble in the early 1990s found themselves encumbered by new rounds of bad loans created by one slowdown after another. The slowing of Japan’s economy was inevitable; when economies catch up with their potential through investment, the returns to that investment – in infra10-yr GDP/cap structure such as roads and dams, in physical plant and equipment, in institutions compound lagged 10 and procedures, in science and technology, in human capital – decline. Japan is no growth rate toyrs exception the law of diminishing returns. The one escape clause to this law is through productivity: as countries mature, an increasing proportion of their growth 3363.29 -2.62 comes from productivity gains rather than from mobilizing more resources. 3349.42 -1.83 Productivity is simply the ratio of outputs to inputs. Productivity rises when 3451.27 -1.17 Figure 1.4: Average Annual Growth greater value can be produced with the same or fewer inputs. The experience of 3332.11 -1.81 rich, mature economies 3204.73 -0.25 is that productivity rarely grows faster than 2 per cent per3453.31 year or falls below 0.05zero when averaged over several years. The relationship between growth and GDP per capita is shown in Figure 1.4, which charts the 3094.54 1.81 experience of 116 non-oil-dependent countries with populations greater than 1 2284.89 5.57 million and GDP per6.37 capita greater than $1,000. The figure plots annualized 102185.86 2352.51 6.111975 to 2005 on the vertical axis and real per capita GDP year growth rates from
Japan’s place in the contemporary economic world 11 at the beginning of the 10-year period on the horizontal axis. The data from the World Bank include countries for which there are at least 10 years of consecutive real GDP per capita observations; the figure plots 2,100 observations. One important point to draw from this chart is that growth rates converge towards 0–2 per cent as we move to the right. The point representing Japan for the 10 years 1995–2005 is shown by the arrow. Despite economic slowing and the burden of serious problems during this so-called lost decade, its experience is not dramatically out of line with that of other rich countries. The American data are at the upper bound and Switzerland includes the points near zero. The country with greater than 2 per cent growth and $24,000–$28,000 income is Norway, which benefited from North Sea oil production. The high flyer at $16,000 GDP per capita is Ireland. To illustrate how hard it is to exceed 2 per cent per capita growth for more than a few years, the United States barely managed to bump through the 2 per cent ceiling during the years that spanned the late 1990s, which included an Internet and telecommunications investment bubble, a booming stock market and historically low unemployment rates. Another point to draw from Figure 1.4 is that the economies represented towards the left of the chart include the very fast-growing ones. The combination of relative backwardness and reasonable policies has the potential to generate truly outstanding growth, reminiscent of Japan’s in previous decades. Many examples of achievements in the 6–10 per cent range can be found among the poorer countries in the chart. However, being poor is no guarantee of growth; that part of the chart also includes many collapsing economies with negative growth over extended periods. Policies matter, as do harder to measure ingredients such as institutions and habits – a subject that we will return in the next chapter when we consider Japan’s transformation in the nineteenth and twentieth centuries. The evidence suggests that even in Japan’s lost decade, its economic performance – while certainly not commendable – is not the worst among rich countries, even very rich ones. Nevertheless, could Japan become like Argentina, with a political system that could not cope with its economic problems and an economy that suffered decline? Switzerland is probably a better model for Japan than is Argentina; Switzerland has been a highly developed country for decades. We do not think of it as a failed state, and neither is Japan. Indeed, Switzerland’s growth rate for the 30 years from 1970 to 2000 was less than Japan’s in the 1990s. Figure 1.5 plots average annual real, per capita GDP growth over 10 years for Japan, Switzerland and the United States. Japan’s growth curve was that of a still-developing country until the end of the 1970s; since then, it has converged with the patterns expected of the developed world. The domestic policies associated with Switzerland’s slow growth provide some revealing comparisons with Japan. The International Monetary Fund’s 1998 evaluation of the Swiss economy notes: A clue to Switzerland’s relatively poorer growth performance may be found in structural rigidities. Switzerland has been frequently criticized as a dual
12 Japan’s place in the contemporary economic world 10 Japan
Growth rate, real GDP/capita
8
6
4
2
0
-2 1960
US
1965
Switzerland
1970
1975
1980
1985
1990
1995
2000
2005
Figure 1.5 Average annual 10-year growth rate (%) of real GDP per capita for Japan, the United States and Switzerland, 1960–2005. Source: PWT 6.1, World Bank.
economy with a highly competitive open sector engaged in external trade and a sheltered sector oriented toward the domestic market. . . . The sheltered sector has been extensively protected. The distribution sector is highly concentrated and, at least until recently, anti-competitive practices such as price-fixing, market-sharing arrangements, pooling of sales, or outright cartel arrangements have been common. . . . Public monopolies in transport, postal services, telecommunications, and electricity resulted in relatively high prices, inefficient operations, and less innovation. In addition, the domestic market led to higher input costs for other sectors and may have hampered the diffusion of the benefits from new technologies. (IMF 1998: 20) As in Japan, Swiss capital productivity had been falling since the mid-1960s. Not surprisingly, funds went abroad looking for higher returns, which drove Switzerland’s current account surplus to 13 per cent of GDP in 2000, compared with Japan’s figure of 2.4 per cent.
Catching up, falling behind The central economic goal of the Meiji leaders in the 1870s was to catch up and overtake the western powers in economic capability. How well did they and their successors accomplish that task? At the time of Commodore Perry’s first uninvited visit in 1853, the industrial technology of the economically advanced nations was
Japan’s place in the contemporary economic world 13 not that many years ahead of Japan. The steam engine, railroad, telegraph and steel production were of relatively recent origin. Only two of Perry’s four vessels were steam driven, and even they carried a full complement of sail – their main mode of propulsion most of the time. In terms of new invention, therefore, Japan was at most 50 years behind the widespread dissemination of the core technologies of the industrial revolution, and in the most advanced areas, by definition, their lag was considerably shorter. The first British non-experimental railroads were just 20 years old, and the first experimental telegraph line had been constructed in the United States less than 10 years before Perry’s gift of a telegraph set to the Japanese rulers. In economic terms, however, the lag was considerably greater. GDP per capita is perhaps the most comprehensive measure of a nation’s overall productivity and mastery of technology. Although specific industries or products may be produced at technological levels considerably greater than the average, GDP per capita measures the overall capabilities of the entire economy. Using this measure, it was not until the 1960s that Japan made significant progress in catching up with the advanced nations. Figure 1.6 shows the gap between Japan’s GDP per capita and that of the United States. It charts the horizontal distance between the two curves of Figure 1.3. The chart indicates, for example, that in 1980, Japan achieved a level of output reached by the United States 16 years earlier in 1964. In the years before the reforms of the Meiji Restoration, Japan gradually fell behind the industrializing west, with an approximate 100-year lag in 1880. Then, until the end of the Second World War, the gap slowly decreased, but not as much as might have been
US-Japan gap in years, real GDP/capita
140 120 100 80 60 40 20 0 1820
1840
1860 1880
1900
1920
1940 1960
1980
2000
Figure 1.6 Gap between US year and Japanese year in which Japan first attained US level of GDP per capita.
14 Japan’s place in the contemporary economic world expected given Japan’s mastery of industrial technology, especially in military industry, which allowed the new Japan to defeat China in 1895 and Russia 10 years later. It was not until the miracle growth years that substantial gains were made. Japan’s best relative performance was achieved in 1992, when it was only 10 years behind the United States. Slippage in the 1990s then caused the output of the average Japanese to fall back to an 18-year lag in 2002. The relative decline in the 1990s should not blind one to the fact that the Japanese economic miracle following the Second World War created one of the world’s largest economies. This remarkable performance, following military defeat, physical destruction and rampant hunger in 1945, gave Japan a new definition of itself. Such success helped to solidify – perhaps, even, petrify – the institutions, habits and political arrangements that were associated with the economic miracle. Many of the practices that had become synonymous with Japan – lifetime employment, long-term business relations, subsidies and protection for disadvantaged industries and groups, political protection for favoured clients – conformed to underlying Japanese values. However, many of these practices had negative side-effects, even during the period of high-speed growth. Their persistence retarded the needed adjustments of a maturing and slower growing economy. The list of well-known problems facing Japan at the turn of the twenty-first century, however, is only one side of the story. In order to develop, Japan had to possess institutions that worked. These institutions do not simply vanish; they continue to give strength to the nation – a market-oriented, capitalist democracy. It responds to shifting pressures, even if only slowly. How it responds to the mix of forces acting on it in the twenty-first century depends on economics and politics. Japan’s economic behaviour is the subject of this book.
2 The nineteenth-century transformation of the Japanese economy
Japan’s economy at the dawn of modernization How did Japan industrialize and become one of the world’s leading economies? Two hundred years ago, Japan was a pre-industrial, agricultural country with technology and living standards similar to those of other peasant economies in Asia or Europe. As one economic historian noted, a Frenchman of 1600 would have readily understood most features of Japanese economic life of 1800 (Crawcour 1989: 569). Within the next 100 years, however, Japan joined the modern world of steel, railroads, telegraph, steamships, banking and armies backed by industrial power. Understanding how an economy can make the transition from a peasant society to a modern, developed one has engaged the energies of economists since Adam Smith. From the end of the Second World War, numerous countries adopted economic development as a central policy goal – an objective supported by international organizations such as the World Bank and analysed by an academic army of economists. Understanding the Japanese experience has been informed by the field of development studies as well as by detailed research on Japan in the premodern period. It is a sad fact that many contemporary nations do not succeed in making the transition to economic growth. According to World Bank data, in 1990 there were approximately 30 countries that had a lower GDP per capita than Japan’s a century earlier. Recent research suggests that it is not only current policies that help determine economic success; the starting point often matters as much. The initial conditions seem to be as important as the programmes put in place by even the most enlightened policymakers. Therefore, Japan’s situation at the beginning of its deliberate attempt to industrialize is the place to start.
Main outlines of the Tokugawa period The story can begin in 1603, when Tokugawa Ieyasu took for himself the ancient title of shogun, or hereditary military ruler. Legally, the shogun was one of many lords under the emperor, who resided in his palace in Kyoto; the shogun’s formal authority was limited to control of his own military forces. However, the increasingly feudal character of Japanese society created a situation in which control of
16 The nineteenth-century transformation military force became equivalent to control of the country, and the emperor served chiefly as a symbol of sovereignty behind the shogun and his government in Edo. (The shogun’s city of Edo was renamed Tokyo, eastern capital, in 1868 at the dissolution of the Tokugawa dynasty.) The Tokugawa reign followed the period of the warring states when increasingly powerful local lords (daimyo, or great names) attempted to enlarge their own areas of control and to prevent others from gaining power. Ieyasu’s immediate predecessors, Oda Nobunaga and his deputy Toyotomi Hideyoshi, who took command when Nobunaga was assassinated, had already begun a period of consolidation by forging coalitions of the most powerful daimyo in the late sixteenth century. When Hideyoshi died in 1598 pursuing an invasion of Korea, the potential for civil war was high. Tokugawa Ieyasu’s subsequent military victory over his main rivals quelled the unrest by further consolidating power over the daimyo.1 Ieyasu adopted both traditional and novel policies to enhance the shogun’s authority over the feudal lords and their domains. Among the actions that would have been recognized by any feudal leader, he dispossessed a large number of daimyo who opposed him, reduced their lands or transferred them to less desirable estates; he then granted confiscated lands to relatives and Tokugawa family retainers to establish them as daimyo and to increase their holdings, or he reserved the lands as Tokugawa house domains. The Tokugawa estates included the most prosperous lands and incorporated about one-fifth of the area of Japan, including the largest cities of Edo, Osaka and Kyoto. One-third of all Japanese people eventually would reside in Tokugawa territories. The chastened daimyo continued to govern their own domains, or han, while a central administration (bakufu, literally tent government) was established in Edo. The early Tokugawa period was feudal in character. Four major social classes existed, which varied somewhat across fiefdoms: samurai, farmers, artisans and merchants. Samurai, which included the daimyo themselves, originally were professional soldiers, but later became administrators in the service of shogun and daimyo. Priests and doctors were not included in the named caste system; the emperor, court nobles and outcasts fell into other categories. Membership in a caste typically was hereditary; each family had its proper place in a hierarchy held together by feudal obligations: protection from the top down and service to superiors in the reverse direction. Having described this structure, it is also necessary to modify it by noting its looseness. Several kinds of samurai, for example, blurred such clear distinctions. Some were hereditary with dedicated stipends; others were granted a lifetime right to samurai status for specific actions in support of a daimyo; still other samurai apparently purchased their positions. This fluidity among all the classes has led historians into questioning the fact of Japanese feudalism. Daimyo granted hereditary samurai fixed annual stipends paid in rice out of the taxes collected from daimyo estates. Samurai constituted roughly 7 per cent of the population. Farmers were by far the largest part of the society, accounting for roughly 80 per cent throughout the Tokugawa period. However, farmers’
The nineteenth-century transformation 17 actual occupation increasingly deviated from their nominal designation as many abandoned the land to take on other jobs, full time or part time, in rural and urban areas. Tokugawa Ieyasu and his successors issued several unusual directives that would influence developments over the next 200 years. The year 1635 saw the introduction of the system of alternate attendance (sankin kotai) by which the lords were required to alternate their residence between Edo and their home domains. The period of attendance in the capital varied from six months for those close to Edo to two years for the more distant han. When the lords returned to their lands, they were obliged to leave their wives and heirs in Edo as virtual hostages. The system ensured the continued surveillance of the daimyo by the shogun and his agents. The annual trips back and forth from the capital were very costly; not only were the processions governed by decree that specified statusoriented particulars, but the daimyo also insisted on making great shows of their cross-country progress, proceeding often with hundreds of retainers, samurai and male family members; since the daimyo had to maintain two residences – the one at Edo typically being designed to reflect status and fashion – the combination of travel and multiple households helped to keep them in permanent debt, which further reduced their threat to the central authorities. The sankin kotai system lasted until 1862. In another attempt to reduce the internal military threat, the shogun prohibited wheeled vehicles because of their possible use in facilitating military opposition. However, the Edo authorities organized a network of roads, post houses and provisioning centres for horses and bearers across the country to cater to the travelling daimyo and their retinues; this network was designed also to provide logistical support to the military in case of domestic uprisings or foreign invasion. In the late 1630s and early 1640s, Iemitsu, the grandson of the first Tokugawa shogun, introduced a policy of national seclusion through a series of decrees. One was the expulsion of Catholics in 1639. The so-called maritime prohibitions controlled trade. Much later, in the 1800s, the bundle of policies was given the designation sakoku (closed country). Earlier versions of these measures had been promulgated against Christians and other Europeans in the late 1500s because of fears that European powers and missionaries were siding with certain daimyo in the internal wars of the time; such outside support included the supply of muskets and cannons, which increased the deadliness of the ongoing civil wars. Although the Tokugawa shoguns desired the gains from trade that contact with the outside world made possible, they also feared that the European powers and proselytizing Christian missionaries would unite with opposing daimyo to attack the bakufu. The bakufu also desired to free Japan from Chinese influence. This history lay behind the seclusion decrees implemented from roughly 1635 to 1644 (Toby 1977: 326). The shogunate prohibited Japanese from making overseas voyages or from returning to Japan from foreign places, with some exceptions. Dutch and Chinese traders were allowed to carry on import and export businesses as before, although this trade was restricted and confined to the island of Dejima and the port of Naga-
18 The nineteenth-century transformation saki and reserved for the bakufu. Two domains were authorized to conduct trade with the Ryukyu Islands and Korea. Japanese agents took up residence in Korea to run warehouses and factories; the number of people at the Korean facilities rose to as many as a thousand. A highly profitable trade transpired in silver coins, from Japanese producers to Chinese buyers through a roundabout route via Tsushima in southern Japan and Korea, despite repeated attempts by the central authorities to limit the outward flow of silver (Tashiro 1982: 291). It would be mistaken to assert that the seclusion policies completely closed off Japan from the rest of the world. The permitted trade with Asian and European partners facilitated some amount of economic and intellectual intercourse. By some accounts, the total volume of trade rose after the seclusion policies were announced compared with the prior period. Nevertheless, the overall effect of sakoku was to insulate most Japanese from foreign influence, especially from industrializing Europe, for more than 200 years. In the period before Tokugawa consolidated power, most taxes were assessed and paid in coins. Under Tokugawa rule, however, the income of the shogun and daimyo depended primarily on rice taxes collected in kind (Yamamura 1988: 3341). Therefore, rice output was of critical importance to the leaders’ prosperity and they made great efforts to enlarge the resources devoted to rice cultivation and to raise its productivity. However, their economic welfare depended on converting the rice tax into money for use in commercial transactions. Since the authorities’ income was based on the collection of rice, but expenditures were in the cash market, civilian demand that might raise prices was unwelcome competition. As commercial business developed in castle towns, partly in response to samurai expenditures, a growing class of merchants and artisans responded to supply the new demand; in turn, these providers of services to the elite generated their own demand for commercial and household consumption. The shogun and daimyo repeatedly tried to stifle this civilian demand while they encouraged rice output. Nevertheless, as opportunities for peasants became more diversified and, as taxes on rice production increased, peasants shifted production to goods that could be sold to the rising urban populations, or they moved off their lands. Another approach to reducing demand was through sumptuary laws, which regulated what people could buy, wear and use according to their place in society. These laws also conformed to certain Confucian values of the roles that people were ordained to play in a properly regulated social order. However, as the incomes of farmers and merchants gradually rose during this period and as peasants found employment in other pursuits, decree after decree unsuccessfully attempted to control consumption. The bakufu and the domains often attributed falling rice harvests and the increased prices of other goods to the decline of peasant morality, leading to the neglect of their proper work in the fields. One bakufu order complained, ‘Peasants are not what they were. Instead of wearing coarse clothing and tying their hair with straw, they were wearing cotton garments and using hair oil and fancy hair ties. And in wet weather, instead of straw capes they were using raincoats and umbrellas’ (Crawcour 1989: 394).
The nineteenth-century transformation 19 The yield from agriculture taxes peaked before 1800. The declining tax flow from rice and the subsequent falling income of the shogun and his daimyo created severe tensions within the society, which contributed eventually to the collapse of the entire system. However, there were good reasons that this system of in-kind tax payments lasted so long. As argued by Yamamura (1988), agriculture, especially rice production, was both the dominant economic activity and a reliable tax base; output could be measured fairly accurately, the tax could be levied and collected with minimum transactions costs, and the product was storable and standardized, which allowed an entire financial system to be developed on a rice base. Measuring, assessing and collecting taxes based on income from commerce, services or the production of other kinds of goods suffered from serious drawbacks. Even modern governments, including Japan, have difficulty collecting taxes from commercial activities (Yamamura 1988: 346).
System evolution during the Tokugawa period Peace reigned for 250 years; domestic stability encouraged investment and improvements in agriculture, which raised living standards and incomes for the average Japanese. The regime’s attempts to maintain rigid social control amidst the nation’s self-imposed insularity reduced the entry of disruptive ideas and methods that had transformed Europe during the Renaissance and industrial revolution. Nevertheless, the economy and social system were far from stagnant. Scholars have described several different sets of forces that gradually changed the nature of the country over the 250 years of Tokugawa rule; indeed, it was the slow accumulation of change that eventually came to undermine the shogun. Change was so slow, though, that often it was difficult to comprehend just how much had shifted. One example is the increase in productivity that came to transform the relative incomes of important classes of society. Living standards, even in a poor district like the Morioka region, examined by University of Washington professor Kozo Yamamura, rose by roughly 0.5 per cent a year from the early 1700s to the end of Tokugawa rule. Nutrition, clothing and housing improved steadily throughout the period (Yamamura 1973: 535). Although almost imperceptible over the lifetime of a single individual, when pursued over several centuries it raised average real incomes by 3.5 times. By the 1850s, peasants – who at the beginning of the period were a short step away from starvation – could afford such luxuries as umbrellas (officially denied them by sumptuary laws) that only the privileged classes had earlier enjoyed. Urbanization One central change that affected economics and politics was urbanization. In the 1500s, improved productivity and new military technology used by the warring daimyo had encouraged the growth of towns and cities. Agricultural productivity gains in that century had allowed the release of resources on an unprecedented scale. With the introduction of firearms and cannon from Europe, defence required
20 The nineteenth-century transformation more strongly fortified castles and larger, trained armies, which in turn demanded the assembly of greater numbers of people to build and defend the expanded, fortified castle towns. The more powerful daimyo required their vassals to abandon their own castles and live in the domain’s castle town, a precursor to the alternate attendance system promulgated in the next century. To supply the food and other needs of the inhabitants, these towns often were sited in the middle of agricultural areas (Nakamura 1981: 268). The flow of people of all classes into cities and the concentration of government in Edo made Japan one of the most urbanized countries of the nineteenth century. It had three of the largest cities in the world. Edo, the administrative centre, had more than 1 million inhabitants. The imperial capital of Kyoto with approximately 300,000 inhabitants was the centre of high culture and superior handicrafts. Osaka, about the same size as Kyoto, was the leading market and financial city. With urbanization came the craft industries and merchant classes that processed the food and other materials of the countryside and catered to society’s elites as well as to the tastes and the incomes of the masses. Sophisticated and popular arts flourished. When Japan finally opened to the rest of the world in the 1850s, the west was astounded by the creativity that had been nurtured there out of sight. Western artists and manufacturers quickly incorporated Japanese ideas into their own artistic creations and products. The larger, concentrated urban markets increased the demand for a wider range of crops and goods, which encouraged farming specialization. Peasants cultivated cash crops in most regions: cotton, indigo, sugar cane, tobacco, silk worms and rapeseed. These products required processing and manufacturing to convert them into usable products: silk-reeling, tool manufacture, weaving of straw mats and sandals, food processing, and the ginning, weaving and dyeing of cotton. Nakamura (1981: 268) argues that the peasants, who at one time barely survived at subsistence levels, increasingly participated in urban markets, where they were subject to impersonal market forces rather than to the directives of their daimyo. Production for market increased the farmer’s range of choices: over the crops to plant, the allocation of time that he and his family devoted to farm labour and other activities, and the purchases that would be made with cash income, including investment in fertilizer and farm implements. A peasant economy that formerly produced only rice for a single patron now had to consider market demand and the capabilities of land and workers. The farm community’s attachment to the cities and towns made farmers and their families aware of urban consumption patterns; trips to the local markets disseminated knowledge of agricultural techniques and introduced the peasant to information flowing into the town from an expanding countryside and from the wider experience gained by those travelling with the local lord on his annual pilgrimage to Edo. In the 1800s, however, the population of the larger cities declined while rural villages near the cities expanded. This return flow was driven mainly by the high cost of doing business in the cities and by the restrictions imposed by city-based
The nineteenth-century transformation 21 guilds. The shifting centre of gravity brought to the countryside many of the urban environment’s attributes, including labour markets, manufacturing, wholesaling, retailing and transportation (Nakamura 1981: 269). Alternate attendance As suggested, the practice of alternate attendance had several effects on economic development. The movement every year of several hundred feudal masters with up to several thousand retainers required roads, means of transportation, post houses, feeding and supplying of people and animals, and considerable planning. The movement of people from their native villages to metropolitan areas across the breadth of Japan educated generations of minor officials and lowly subjects who otherwise would have known little more than the fields around their villages. At the same time that rural folk were learning the ways of the big city, local methods, ideas and arts were disseminated to Edo. This two-way flow fostered the notion of a greater Japan. The movement diffused the modernization process across the 250 castle towns (Vaporis 1997). Daimyo migrations also increased the demand for money and the rise of a sophisticated financial system. Outside a daimyo’s own domain, ordinary barter transactions were inadequate. Rice normally served the purpose of a medium of exchange, but increasing commercialization required a more acceptable and easily portable form of money. Gold, copper and silver coins became popular. To earn this ‘foreign exchange’, local lords had to engage in commercial activities in the major markets of Osaka and Edo. The demand for cash pushed them to find more productive methods for their peasants and forced them to consider other methods for raising revenues; among the alternatives was the licensing of monopoly sellers of commercial products, which further promoted the legitimization of merchants and commerce. Daimyos’ debt forced them to borrow large sums, collateralized by future rice harvests. These rice and credit transactions led to the growth of sophisticated financial practices and markets. Osaka was the main financial centre. A small group of bankers performed many of the functions of a central bank, acting as lenders of last resort, making loans to local governments, controlling the level of bank credits and establishing a market between gold and bank money. As bankers came to realize that most of their transactions in warehouse receipts for stocks of rice did not require their handling of the actual physical rice stocks, they introduced financial instruments based on future deliveries. Thus, the rice market in Osaka initiated such activities as futures trading, which smoothed the price between harvests or when speculators tried to corner the physical market (Takenaka 1969: 144–8). Precursors to growth: education and governance Perhaps the greatest contribution to later growth was the education of the population. Richard Easterlin, a leading economic historian, concluded that the prime
22 The nineteenth-century transformation mover in the drama of economic development is the acceleration in the rate of technological change in a relatively small number of nations. He further concluded that the economic growth differences seen among countries were governed by the extent of the population’s schooling; exposure to formal schooling was positively associated with subsequent economic development. His explanation of this relationship was simple: mastering technological change requires education (Easterlin 1981: 6). Development economist Angus Maddison also asked if there are precursors, or ultimate cause of growth. Although the best single predictor of the growth of an economy is its investment rate, there remain differentials among countries after accounting for the effects of investment. Maddison describes investment as a proximate cause of growth. Acknowledging the difficulties of identifying the ultimate causes, he ascribes the ‘most fundamental’ of these to the recognition of human capacity to transform the forces of nature through rational investigation and experiment. Education is one of the prerequisites for this process to work (Maddison 1995a: 104). Dani Rodrik, a Harvard economist, likewise sought the reasons for growth differences among the East Asian economies in the twentieth century. He noted that investment is the best predictor, but when he sought the drivers of investment he found that institutional quality and the initial conditions of income and education were the best predictors of subsequent growth (Rodrik 1997:13). Education, then, appears to be a leading candidate as a precursor to economic development. By the end of the Tokugawa period, most towns and villages had temple schools (terakoya), which took on the characteristics of modern elementary schools, teaching reading, writing and arithmetic. Two independent estimates of education and literacy in the mid-1800s came to roughly the same conclusion about the spread of reading ability: 40 per cent male literacy and 15 per cent female (Dore 1965: 291–5; Passin 1965: 47–9, 310–12). Education data for 1873, the first year that such data were collected after the Meiji Restoration, show similar rates of primary school attendance. Tests of conscripts in the early Meiji period reveal sizable differences in literacy, with over half of those in the professional classes graduated from higher elementary schools (bankers, government workers, company officials), 40 per cent of artisans and farmers, but only 10–20 per cent of fishermen and labourers. Minimal definitions of male literacy would have included 65 per cent in one small village in the 1850s and 76 per cent in the 1870s; however, a more stringent definition drops the figures to 8 per cent (Rubinger 1990: 608). Japan’s rates of education and literacy in the middle of the nineteenth century were well below those in the economically advanced western nations at the time, but considerably above those in the rest of the still undeveloped world in Europe, Asia or the Americas (Easterlin 1981: Appendix). Rodrik and other development economists also have highlighted the institutional quality of government as a necessary precursor to development. They focus on such areas as competence and corruption to measure the ability of institutions to do their jobs. To gain a sense of Japan’s governmental competence, consider its organization of the movement of hundreds of daimyo to and from Edo. Though
The nineteenth-century transformation 23 routine, it required management on a national scale. Such daily replications of miniature military campaigns occurred in grand style in 1861, when the emperor’s younger sister travelled from Kyoto to Edo to marry the shogun; some 25,000 court retainers accompanied the royal party. Planning took several months and involved thousands of porters, animal tenders and other staff at each of the 69 posts along the Kiso road to feed, house and move the people, luggage and animals. Although it was planned in the capital, local initiative, resources and energy made the whole thing work (Shimazaki 1987: 135–55). Technology, productivity, production With the growth of specialization in production, technology progressed in several fields. Improvements in metal mining, smelting and refinement led to the production of larger volumes of silver and iron; for example, the mechanization of bellows produced higher temperatures in iron smelters that enabled the production of low-carbon steel used in a wide range of tools. Two of the largest silver mines in Japan accounted for one-third of all the silver being produced in the world in the early 1600s, according to one estimate. Technological developments in areas of civil engineering such as water works promoted the expansion of paddy rice cultivation, with the acreage devoted to rice rising an estimated 75 per cent in the 50 years after 1600 (Keiji and Yamamura 1988: 82–3, 103). Such improvements in productivity allowed Japan to feed its 30 million people in the early Tokugawa with a smaller area than that which supported Europe’s 5–10 million population. Although most production was in small craft shops that used little capital, several mining establishments employed more than 1,000 full-time labourers apiece. Spread around the country were 80–90 iron mines, each of which had around 300 workers. The growth of an iron industry promoted the beginnings of a factory system, with the attendant accumulation of capital and organization of paid work (Yamamura 1973: 533). As the Tokugawa period progressed, standards of living improved. Nutrition was more than adequate and most people had adequate shelter and clothing. Even the poor, in the 1850s, on a daily basis were eating sufficient soybean paste (miso) soup and one or more servings of seasonal vegetables, bean curd, dried fish and pickles. Demographer Susan Hanley estimates that the diet was better than that of the working class in the industrial cities of England. Water in Edo was purer than in London; human waste was collected for fertilizer and not allowed to seep into the water supply. Even poor people were accustomed to drinking boiled water. That ordinary Japanese spent much of their marginal income not on food, but on entertainment, travel and investments in land, fertilizer and business, suggests an economy clearly above the subsistence level for much of the eighteenth and nineteenth centuries. Administrative competence, from village to bakufu, enabled the authorities to maintain streets, bridges and water supply systems. Consequently, living conditions were relatively hygienic and healthy (Hanley 1983: 189–91). Life expectancy is one index of living standards. Several studies conclude that life expectancy in mid-1850s Japan was around 42 years, similar to that in much
24 The nineteenth-century transformation of industrializing Europe. Indeed, Japanese were living longer, on the average, in 1840 than people in at least 17 countries 130 years later, according to World Bank data (World Bank 2006). The stage is set Evolving conditions in the nineteenth century set the stage for the transformation that was to come in the later part of the period. However, it is important to emphasize that the late Tokugawa was not an industrializing economy, nor was it likely to become one without a radical transformation of political and economic structures. Certainly, we can point to an accelerating tempo of change. Wage labour was replacing feudal service obligations, commercial farming was supplanting subsistence agriculture, rural industry was expanding, commercial networks increasingly bypassed central and local monopoly channels, several industrial plants in the western style were operating, and tax revenues from traditional sources were unable to maintain the existing political structure. Nevertheless, to take the example of education, radical change was required before the system could support a modern economy. Easterlin estimated that primary school enrolment of more than 400 per 10,000 total population is necessary for rapid economic growth. In 1873, the Japanese figure was only 338. However, the education reforms of the new Meiji government could build on the existing base, and within only 10 years the enrolment rate jumped to 750, and by 1890 to 882 – a level described by Easterlin (1981: 6) as ‘substantial’ exposure to education. The rapid development of a national system of compulsory education by the Meiji authorities in the 1870s was possible only because of an educated class that provided an immediate supply of schoolteachers. The point here is that the existing educational stock established the base for a swift increase, but only after new policies were adopted; bakufu education policy did not envisage such a radical change in the system. Western imperialism reaches Japan Adding to these intensifying disruptions of the accustomed routine was the arrival of the American Commodore Matthew Perry, who sailed into Edo bay in 1853 and, backed by European nations, demanded that Japan open its ports to foreign powers for servicing their merchant fleets and for trade (Crawcour 1974: 115). The treaty port system established in Japan by the United States, Great Britain and other western powers was the mark of imperialism’s extension into Asia. Britain was the most industrialized nation in the mid-nineteenth century and possessed the most powerful navy; its leaders asserted that barriers to trade justified the use of its military power to remove the obstacles. Great Britain fought several wars with China in the 1840s and 1850s to secure such commercial relations. British goals were to gain trade access to China with a minimum of constraints, which required, in British eyes, enforceable protection of traders and merchants. However, experience in the administration of its Indian colony led the colonial
The nineteenth-century transformation 25 office in London to be wary of engaging in territorial control and intrusive administrative practices. Consequently, Britain restricted its activities in China to opening five ports, fixing customs duties, establishing consular offices and courts, and introducing a most-favoured-nation clause in its treaties with China, whereby any further privileges that might be granted subsequently to other countries would be available to Great Britain (Beasley 1987: 16). Eleven western countries eventually signed treaty port agreements with China. The privileges granted by these agreements – described in China and later in Japan as ‘the unequal treaties’ – diminished Chinese sovereignty and did not grant reciprocal rights to Chinese in the western countries. When western gunboats reached Japan in the 1850s to open that country to trade, they had a ready model to apply in the shape of the treaty port system. By 1863, Great Britain, the United States, Holland, Russia and France had forced Japan to sign such treaties. The terms imposed by these treaties were not removed completely until 1911. Japan’s resentment over the absence of mutuality in the treaties, ironically, bequeathed an incentive for Japan years later to seek the same rights in China as those of the western powers as a symbolic element of Tokyo’s climb to great-power status (Beasley 1987: 17). The threat inherent in the treaty port system created what was perhaps the main difference between the incremental change occurring over the 250 years of Tokugawa rule and the transformation put in place by the Meiji rulers: modernization in the first period was a gradual adaptation to slow-moving drift, whereas in the second it was dominated by wilful policy. In addition to the threats from the west and the lessons of industrialization learned from the developed world, Japan also paid close attention to imperialism itself, which seemed as much a part of the modern state as steel mills.
The Meiji Restoration and subsequent growth Close examination of the period from 1860 to 1880 shows considerable continuity from the Tokugawa regime through the first decade or so following the Meiji Restoration in 1868. The view that we now have, looking backward across almost 150 years, of dramatic rebirth was not immediately apparent to those on the scene. Yet, 1880 was hugely different from 1860. Crawcour’s description of these changes are worth noting: In the context of Japanese history, the changes following the Restoration appear to have been almost instantaneous and the restrictive apparatus of the old order to have been removed, as it were, at a stroke. But in fact these changes took at least ten turbulent years. The Restoration signalled far-reaching changes but had little immediate impact on the economy other than to exacerbate the existing uncertainties and disruption. Nor did it solve any of the existing problems that had beset the bakufu in its last years. (Crawcour 1989: 605)
26 The nineteenth-century transformation The Shogun reacts The last years of the Tokugawa shogunate saw sweeping changes that foretold much of what was to come: centralization, rationalization and bureaucratization (Jansen 1989: 351). These changes in the last two years of the old order were motivated partly by the 1866 bakufu loss to a western-style army created and trained by the Choshu domain and by the death of the youthful shogun Iemochi. Tokugawa Yoshinobu – the last shogun – had been a contender for bakufu leadership in 1858, but the leading daimyo instead chose the infant Iemochi; four years later, Yoshinobu was appointed as guardian to the child shogun, whose death brought Yoshinobu into a position of formal power. Military defeat had demonstrated the inadequacies of the decentralized political system and traditional operating methods. Yoshinobu took radical steps to improve governance. The central authorities regularized foreign relations and established permanent missions in important foreign capitals. The shogun’s brother went to France to represent his country at the Paris Exposition of 1867. A new cohort of younger administrators was brought into the government to plan and implement modernization. Illustrative of Japan’s opening to the world, the entire foreign diplomatic corps residing in Japan was invited to Osaka for an audience with the shogun, where western dress replaced traditional Japanese garb at a dinner prepared by a new French chef. The shogun instituted a form of cabinet system governance, replacing the monthly rotation of generalist advisors. He planned a new army with modern equipment and asked the French for military advice and materiel. The shogun requested his advisors to draw up a new government structure, including a bicameral legislature. The shogun and his bureaucrats were launching a modernization campaign – a ‘Tokugawa restoration’ in one view – that would have emulated many of the programmes of the succeeding regime (Jansen 1989: 352). However, the old regime had lost the support of the regional daimyo and Yoshinobu chose not to resist the final collapse by military means. He agreed to surrender his powers in November 1867, expecting to be the first among equals in any new power structure that emerged. Satsuma and Choshu leaders backed a group of radical samurai who took control of the palace in Kyoto in January 1868 and declared an Imperial restoration. Although Yoshinobu agreed to accept the results of the coup, a short civil war ensued. When the Imperial forces marched on the shogun’s capital at Edo, Yoshinobu finally urged his troops to surrender. Transformation Almost simultaneously with the end of Tokugawa rule, the Emperor Komei died in 1867, and his 16-year-old son, who took the name Meiji, by which the era of his reign is known, became titular ruler in 1868. Within days of the surrender of Edo to the Imperial forces, the young emperor’s advisors presented him with what would be known as the ‘Charter Oath’, five articles issued in the emperor’s name that bridged the transition to the constitutional order of the modern Japanese state. Its provisions included the following points: deliberative assemblies shall be established on an extensive scale, and all governmental matters shall be deter-
The nineteenth-century transformation 27 mined by public discussion; all classes shall be permitted to fulfil their just aspirations; and knowledge shall be sought throughout the world in order to promote the welfare of the empire. This document ended the feudal character of the Japanese state and ‘held out the possibility of changes so basic that it could still be cited as authorization for the democratic institutional changes that followed World War 2’ (Jansen 1989: 359). Government revenues and expenditures had been a growing problem under the shogun. In the confusion surrounding the Restoration, the han governments issued large quantities of paper currency; almost 1,700 different types of daimyo paper money circulated, with their relative values frequently changing (Patrick 1965: 194). In addition, the bakufu had issued gold, silver and copper coins, which circulated together with Mexican silver dollars in the treaty ports – the assigned areas for foreign trade. Coins differed in purity and weight, and many were counterfeit. In 1868, the central government ordered the conversion of all han silver-based currency into newly issued paper notes. Reform of the currency included the creation in 1871 of a new monetary unit, the yen, which replaced the ryo. Since the new regime faced critical problems that demanded even more spending, primarily the need to raise and equip a modern army and navy, it also resorted to issuing its own paper money. Looking for places to reduce expenditures, the 30 per cent of revenues that went to samurai stipends could not escape attention. Starting in 1869, payments to samurai were reduced in several steps; in 1873 and 1874, samurai were encouraged to exchange their annual stipends for lump sum payments or bonds. A military conscription law in 1873 removed the central reason for maintaining these hereditary military retainers, and in 1876 they were required to surrender their entitlements in exchange for a lump sum payment, mainly in government bonds that were redeemable over 30 years beginning in the sixth year after issue. For the samurai, the lump sum payments and bonds were a poor exchange for their guaranteed lifetime incomes. Some former samurai opposed this forceful and unprofitable ending of their customary income; they fomented unrest in the southwest part of the country, resulting in the Satsuma rebellion, the suppression of which required even greater expenditures by the financially stressed government. The new government sought to centralize authority in Tokyo and abolished the domains in 1871. The old domains were converted into civil prefectures, but with many redrawn boundaries and territorial consolidations. In a further move to reform the fiscal system, the authorities abolished the tax system based on rice. However, since agriculture was the predominant form of economic activity, some kind of agriculture tax would be required. Reform, started in 1873, took several years to implement. A new uniform, nationwide property tax was based on the assessed value of the land, and the officially registered taxpayer was given title to his land, the first time that Japan had private land ownership. Land became a capital asset that reflected its productive value and that could be sold. The new tax removed distortionary incentives that had plagued agriculture, mainly because farmers had moved into other crops in order to avoid the rice tax. The formulas used to calculate the new tax were intended to keep the total tax
28 The nineteenth-century transformation burden unchanged from the old system, but because one of the goals of the new monetary tax was to make payments more equitable, some taxpayers who had been favoured under the old system were penalized. Despite the attempt not to raise the average burden, there was widespread opposition to the new tax system, and the rates were reduced in 1877. One feature of the land tax that was to have negative consequences for government revenues over the next several years was that it was fixed in nominal terms. As the economy grew and as inflation emerged, the land tax supplied a diminishing real value of revenues, prompting the financial authorities to search for other taxes, especially on consumption goods such as sake. Infrastructure development dominated investment in these early years of the new regime. By 1877, the government had built 64 miles of railroad lines and 2,800 miles of telegraph lines. In an attempt to head off foreign dominance in shipping, a government shipping company that had been running losses turned over its 30 ships to a private domestic company in 1875. The old post stations – organized to cater to the travelling daimyo crisscrossing the country in their journeys of alternate attendance – were abolished in 1871, as was the official courier system, which was transformed into a modern postal system. Just three years later, more than 3,000 post offices existed; in 1877, Japan joined the Universal Postal Union. Letters could be mailed reliably from Japan to the world, and vice versa. Education In education, 1871 saw Japan’s first Ministry of Education established to develop a comprehensive national system. The following year, the ministry promulgated a plan to offer schooling nationwide. Unlike the class-based schooling that had prevailed, the plan envisaged a unified, egalitarian system of modern national education. However, because of the lack of resources, especially money and teachers, an enrolment rate of only 35 per cent of elementary school-aged children was realized by 1875 (JSA 1999: Table 22.1). To overcome the teacher shortage, the ministry quickly established a system of normal schools – teacher-training institutions – six of which were turning out graduates by 1875. Within a few years of the creation of the national system of education, swelling local complaints about the irrelevance of the centrally designed western curriculum and inadequacy of local control influenced the government in 1879 to replace the original system with a more decentralized approach. The scheme of local control, however, was associated with a decline in standards and falling enrolment. A revised system with national standards, but with administration carried out by the prefectures, replaced the decentralized approach in 1880. New laws and regulations followed in 1886 that completely reorganized the evolving system. In the 1890s, technical and vocational training was given priority to educate the workers and engineers of an industrial nation. Until 1907, four years’ attendance was compulsory, at which time it was lengthened to six years; this basic level of schooling was subsidized by the state.
The nineteenth-century transformation 29 Middle schools and high schools made up the secondary school system. In 1890, the elementary school enrolment rate of 6- to 12-year-olds had reached almost 50 per cent, and by the new century it was more than 80 per cent. Schooling had become almost universal by 1905, less than 25 years after the creation of a modern system of education (see Figure 2.1). The rise in enrolment was made possible by the conversion of temple schools to elementary schools, the opening of new ones and the training of new teacher cohorts. From 1875 to 1895, the number of teachers more than doubled, and it doubled again by 1910. Unlike the swift rise in education at the lower levels, university enrolment grew slowly for the 20 years after 1875. Tokyo University remained the only institution offering training beyond the college level until 1897. However, at the time of the First World War, education reforms encouraged the creation of private universities. By 1926, more than 20 of these had opened. Until this period, most of the growth of enrolment occurred in the many colleges and teacher-training institutions around the country. In 1900, for example, there were 52 colleges and 54 normal schools, but only two universities (JSA 1999: Tables 22.2a, 22.4a). Monetary policy and creation of a modern financial system Japan was one of the first countries to establish a postal savings system, the chief goal of which was to address the absence of local banking establishments. The government used the British system, established in 1861, as a model for its 1875 creation. New Zealand, Canada and Belgium followed in the next few years, the United States in 1910. The stated goals of the Meiji planners were to encourage thrift and to gather small savings to finance industrial development. Although those bureaucrats took credit for teaching thrift to their fellow citizens, the remark200,000
100
160,000 140,000
90 Enrolment rate (right scale)
Number of teachers (left scale)
80 70
120,000
60
100,000
50
80,000
40
60,000
30
40,000 20,000
College and university students (left scale)
20
Elementary enrolment rate
Number of students and teachers
180,000
10
0 0 1870 1875 1880 1885 1890 1895 1900 1905 1910 1915 1920
Figure 2.1 Elementary school enrolment rates, number of teachers in kindergarten through high school and number of college and university students, 1873–1920. Source: JSA (1999: Tables 22.1, 22.3a, 22.4a).
30 The nineteenth-century transformation able growth of postal savings attested to an already widespread habit of savings. The post office began offering money order services in 1875, giro (direct transfer) services in 1906 and life insurance in 1916. Approximately 10,000 post offices provided convenient places to deposit savings in 1900 (Kuwayama 1999: 27). While the new government was refashioning the institutions and legal structures of the state, economic activities moved onto new paths, not all of them smooth. The many changes disrupted traditional relationships and forced rapid adjustments to the new environment. The adaptation process seemed to be going well enough until 1876, when the government resorted to issuing national bonds and printing money to pay for samurai pensions and putting down the Satsuma rebellion. In addition, the government had chartered private banks; only one of these was in business in 1876, but the increased number of interest-bearing bonds and the depreciation of government currency encouraged the creation of many new banks, more than 150 of which had been organized by 1879. These nationally chartered banks issued their own paper currency backed by their holdings of samurai and daimyo pension bonds (Droppers 1898: 158). Inflation ensued, reducing the real value of land taxes – already having been lowered in 1877 following riots – and further weakening government finances. An 1898 article in an American economics journal described the inflation-induced euphoria: These years are still remembered in Japan as a time of great excitement and activity. Farmers received more than double the normal price for rice and other farm products. Their expenditures increased, especially for luxuries. Speculative enterprises of every description were floated – steamship companies, silk mills, and canal companies. What was the precise amount of money in circulation during those years is not known, as the government did not at that time officially state the correct figures. Droppers (1898: 158) Prices rose an estimated 25–50 per cent between 1878 and 1881, depending on the price index used in the comparison (Patrick 1965: 201). The combination of inflation and an exchange rate fixed in terms of silver produced a surge of imports, a trade deficit and an accompanying outflow of metal currency, which tended to undermine confidence in the paper money. The money supply started to contract in 1880, but inflation continued for another year, when Masayoshi Matsukata was appointed finance minister. Matsukata had been one of the authors of the land tax; he also had exposure to international experience, having travelled in Europe in one of the many tours designed to obtain foreign knowledge. By the time he assumed office, the economy had stagnated and the price of rice was falling. His policy was to restore internal and external equilibrium by reducing domestic prices. On the monetary front, Matsukata adopted three goals: gradually reduce the circulation of paper currency and back it up with an increased amount of metallic reserves of gold and silver; reduce the number of private bank notes and eliminate the power of private banks to issue currency; and establish a central bank to assume control of currency for the entire country. The Bank of Japan (BOJ) was duly created in 1882.
The nineteenth-century transformation 31 To reduce the necessity of printing money, the finance ministry raised taxes, sold government-owned enterprises and sought reductions in government expenditures. The resulting budget surplus was used to reduce the outstanding currency; a trade surplus generated an inflow of silver, which was used for the same purpose. As the total amount of money in circulation was brought under control, the discount that the market had assessed on paper money in comparison with silver coins disappeared by the middle of 1885. However, the economy did not escape unharmed; from the euphoria of the late 1870s, a severe contraction marked the years 1882–85. Prices declined 20–30 per cent and national income fell an estimated 21 per cent from 1880 to 1884. The Matsukata deflation policy has been the subject of considerable scholarly analysis. Columbia University economist Hugh Patrick, for example, suggested that a mixed approach consisting of a less drastic monetary contraction and domestic price reductions combined with devaluation of the yen could have achieved the desired results without as much pain. However, he notes that the barrier to a mixed policy was the attachment at the time to unalterably fixed exchange rates in terms of metallic money; the orthodox method of balance of payments adjustment was through the change of relative prices brought about by changing the money supply and the subsequent economic depression that would drive down domestic prices (Patrick 1965: 202). The BOJ gradually took control of the money supply by eliminating the currencies issued by the private banks and by offering to exchange the earlier inconvertible government paper currency into silver coins. The charters of the national banks that had been authorized to issue currency were gradually withdrawn and a new banking system without such rights was authorized. The BOJ realized quite early that its policy of promising to convert currency into silver or gold (specie convertibility) combined with a rule to base the issuing of notes on its specie holdings could restrict the flexibility of monetary policy. The central bank was given the authority in 1888 to issue more notes than the specie reserve requirement; although it generally followed the reserve requirements rules, the BOJ at times of crisis took advantage of its flexibility to increase the money supply. The gradual replacement of gold and silver coins with paper money and the growing confidence of the country in monetary affairs provided the foundation for economic growth in the 1890s. The central bank addressed one other reform in the 1890s – a shift to the gold standard in 1897. Until then, both silver and gold were used as monetary metals, although Japan was on a de facto silver standard; for one thing, the government’s budget was denominated in silver-backed yen. However, most of the country’s trade was with gold-standard countries. Since the price of silver fluctuated sharply with respect to the price of gold, the use of silver introduced uncertainties into trading arrangements. Although silver depreciation stimulated exports, it also drove up the cost of imported military equipment and other capital goods from gold-standard countries. After Japan’s war with China in 1895, military planners anticipated large weapons purchases from abroad; at the same time, victory brought with it a large indemnity from China, already deposited as pounds sterling
32 The nineteenth-century transformation in London, a currency easily converted to gold. If the country was going to a gold standard, this was a good time to make the move. Another factor arguing for going to a gold standard was a changing attitude to foreign investment in Japan. Until the mid-1890s, foreign presence in the form of ownership and investment was feared because of its colonial symbolism. As the capital needs of the economy grew, however, the advantages of foreign financing outweighed the symbolic stain; since most of the potential lenders and investors were on the gold standard, it was thought that Japan’s creditworthiness would be enhanced if it too joined the gold club. Another reason given by many writers of the period was the idea that civilized nations based their money on gold; Japan could join this exclusive fraternity if it adopted the progressive metal (Droppers 1898: 169–85; Patrick 1965: 206–9). The creation of a modern financial system was a critical accompaniment to industrial development. The monetary system stabilized after the Matsukata deflation and the establishment of the BOJ; the chartering of new types of banks that were not authorized to issue banknotes provided the institutional foundation for a private banking system that took deposits and made loans to business borrowers. The postal savings system and the private banks proved to be popular with the nation’s savers; the still largely smallholder agricultural economy set aside substantial amounts of savings as their apparently pre-existing proclivities for thrift now could be implemented more readily. Increased economic diversification, manufacturing Post-Restoration instability created by institutional changes, inflation, deflation and political uncertainty gradually subsided; relative quiet set the stage by the mid-1880s for economic development and modernization. Modern sectors relied on contemporary technology and organization imported from abroad as well as greater capital intensity. Traditional sectors, in contrast, continued to use indigenous technology and relatively low levels of capital. The fact that the Japanese economy in the 1880s was based primarily on traditional industry meant that growth had to be based on that sector. Several propositions that are often applied more generally to economic development are directly applicable to Meiji Japan. First, since foreign earnings from the export of manufactures are usually insufficient to pay for the large amounts of foreign capital required for industrialization, the development of the modern economy depends on the accelerated growth of the traditional economy (Ohkawa and Rosovsky 1973: 12). A second proposition is that the traditional sector was capable of acceleration. Third, the growth potential of the traditional sector is necessarily limited; if it were not ultimately so constrained, development would not have to jump on to the track of industrialization and modernization, but could succeed by staying on older paths. Fourth, when the initial growth phase ended because of the traditional sector’s deceleration, the modern part of the economy had to depend on itself for further growth. Ohkawa and Rosovsky summarize these propositions with the observation that
The nineteenth-century transformation 33 traditional agriculture produced the needed surplus resources for public infrastructure construction, private investment and the foreign exchange needed for importing capital and technology; in addition, increasing agricultural productivity freed up labour that could be employed in modern industries. Various estimates place the growth rate of agriculture output from 1880 to 1900 at around 1.6–1.7 per cent with no increase in capital or labour inputs, a considerable productivity increase from the pre-Meiji experience (Ohkawa and Rosovsky 1973: 10; Akino and Hayami 1974: 476). A different theory about the sources of savings is that the Meiji reforms reallocated income away from the high-consuming daimyo and samurai to landowners, who had higher incentives to save and invest directly in their holdings. The difference in views is whether higher savings were explained by higher agricultural productivity and output or by shifting income to those with a higher propensity to save. In support of the savings propensity argument, it is noted that rural landowners often were entrepreneurs and merchants who established banks and industrial concerns. Research results differ on this issue; one reason is that the basic data are often inaccurate or biased. Both forces appear to have contributed to savings (Kelley and Williamson 1971: 733–5). Private ownership of agricultural land and the new tax system provided the incentives to improve productivity, which was accomplished through more intensive use of fertilizers and improved seed varieties. The changed incentives as well as a unified nation without internal barriers allowed the previous wide disparities of productivity around the country to be exploited as low-productivity regions and individual plots could make use of the better techniques demonstrated in other areas. Moreover, the new Meiji government encouraged the transfer of agricultural knowledge by educating farmers and introducing extension agents who disseminated knowledge of best practice. Indeed, one study on agricultural productivity suggests that farmer education, research and extension services accounted for roughly half the improvement of total increased output between 1880 and 1900 (Akino and Hayami 1974: 476). Figure 2.2 shows the share of agriculture and manufacturing in net national product. Manufacturing was less than 8 per cent of total domestic product in 1867. Its growth, however, was rapid, averaging 6.3 per cent over the 15 years from 1885 to 1900. In 1868, two-thirds of gross manufacturing output was in food processing – a traditional industry. Textiles, dominated by silk production, accounted for about 30 per cent of manufactures. The major export items were tea and raw silk, both products of the traditional economy. By the 1890s, cotton and silk goods were the modern industries. The steady rise in the manufacturing share, which overtook agriculture in the 1920s, epitomized the notion of the modern Japanese state.2 Cotton-spinning mills were the first large industrial factories that were efficient, profitable and oriented towards exports. This industry was the main Japanese manufacturing activity for much of the first three decades of the twentieth century. The story of its productivity growth is similar to the improvement of productivity in agriculture – primarily, the copying of successful practice. The
34 The nineteenth-century transformation
Industry % of net national product
50 45
Agriculture
40 35 30 25 20
Manufacturing
15 10 1885 1890 1895 1900 1905 1910 1915 1920 1925 1930 1935 1940
Figure 2.2 Manufacturing and agricultural shares of net national product, 1885–1940 (%). Source: Ohkawa et al. (1979: Table A10, 273).
dissemination of best practice was facilitated by two institutions: an industry cartel trade association whose publications and services disseminated detailed news and descriptions of production practices, and the participation of one large British company that supplied much of the equipment and technical support (Saxonhouse 1974: 159–62). Whereas the dissemination and use of best practice was similar to agriculture, the critical participation of foreign technology and advice, the heavy dependence on modern equipment and the cartelized business structure were distinctly different. Illustrating the quick adoption of new methods, Saxonhouse (1974) notes that in the early 1880s the Osaka Spinning Company built the first mill in Japan to combine comparatively large size, double shifts under electric lights and steamdriven power. When this combination, using British machinery that was identical to that in comparable plants in Lancashire and Bombay, turned out to be quite profitable, every mill built after 1883 copied this design. This sharing of information depended a good deal on the fact that the British company, Platt Brothers of Oldham, accounted for 87 per cent of the spindles in Japan in 1909. The company’s sales engineers could draw on the whole range of Japanese experience in their marketing and design recommendations. Saxonhouse concludes that the cost of acquiring technological information for any firm was extremely low by international standards. When combined with local tinkering and improvements, the accumulation of small changes had a great impact (Saxonhouse 1974: 164). By the 1920s, cotton yarn and textiles made up 25 per cent of Japan’s exports, and Japanese cotton exports had begun to challenge British dominance. Manufacturing could not have grown without investment, which can be financed
The nineteenth-century transformation 35 from domestic savings or foreign funds. Since the authorities were suspicious of the potential for colonial mischief from foreign ownership, they restricted such investment opportunities. Therefore, domestic savings largely financed capital formation in the early Meiji years. However, financial instability and the absence of a banking system at first hindered the collection of household savings and its channelling to business and government. The share of gross domestic capital formation out of GDP rose from around 12 per cent in 1885 to 16 per cent by 1900. Gross capital formation equals the total savings available to the nation from both domestic and foreign sources; the combination of private savings (households and businesses), government fiscal surpluses and growing trade deficits financed a slowly increasing volume of investment; because of rising productivity in the production of capital goods, their prices relative to the general price level were falling, which had the effect of raising the real volume of investment even more than indicated by the nominal shares. Government investment was larger than the private sector’s until around 1915. According to one set of estimates, it was not until 1917 that private investments in producer durables overtook the government’s share. One reason for the government’s large share of capital formation was that the private sector was still relatively small. The other reason for the government dominance of investment is that it was building a new national infrastructure as well as buying weapons, technology and productive capacity for the military (Ohkawa and Rosovsky 1973: 17). Table 2.1 shows the GDP share of investment for five-year intervals as well as the amount of investment that was financed by the current account deficit – largely the difference between exports and imports. Despite official cautions, Japan tended to import more than it exported, the difference being made up by capital inflows. Trade deficits are associated with net capital imports because when imports are greater than exports, not enough foreign currency is earned to pay for the imports; foreign sources must cover the difference by granting loans to the deficit country. In international trade and finance, capital can flow into and out of a country in many forms; for example, a foreign investor may buy shares in a domestic company, or a foreign bank can issue a Table 2.1 Gross domestic fixed capital formation, shares of GDP and amounts financed by current account deficits (five-year intervals, 1885–1919, %) Interval 1885–89 1890–94 1895–99 1900–04 1905–09 1910–14 1915–19
Capital investment share of GDP 13.5 14.5 18.2 13.9 16.7 17.8 19.7
Source: JSA (1999: Table 13.3).
Share of capital investment financed by current account deficit 1.9 2.8 19.1 11.3 16.7 9.9 –31.1
36 The nineteenth-century transformation loan to a Japanese borrower. At the same time, of course, Japanese are lending and investing in other countries. It is the net flow of capital that matters. When all the flows into and out of the country are added up, if the nation as a whole has a trade deficit, importers must get their hands on foreign currency to pay their suppliers. If there is a market for foreign exchange, those needing such funds can purchase them in a market transaction. When there is no such market, foreign exchange is allocated by rule, law, government agencies, bribes, cronies or the accident of who happens to get their hands on it first. In any event, a trade deficit generally requires that more capital flows into a country than out. By the turn of the new century, net foreign capital inflows financed one-fifth of domestic investment. Loans floated in Europe helped to finance war with Russia in 1904–5. However, the First World War created strong demand for war financing in Europe, which led to higher international interest rates; Japan turned from being a borrower to a supplier of funds to Great Britain, Russia, France and others. The outflow of capital depreciated the value of the yen, which – combined with war-induced demand in Europe and the inability of European exporters to satisfy their Asian customers – led to an export boom and a trade surplus. Military industry The role of government in promoting military capabilities owes its importance to the origins of the modern Japanese state. Although the motivations and rationales of the individuals behind the Restoration changed over time and varied across the actors, an enduring goal was to protect Japan from the colonizing western powers. In order to accomplish this task, advanced technologies, industrial capabilities and modern military proficiency were procured. Economic historian Kozo Yamamura argues that the sustained efforts to build a ‘rich nation, strong army’, a banner motto of the Restoration, contributed in important ways to building the technological foundation for Japan’s subsequent industrialization. This policy of building a modern military capability was stimulated further by wars with China in 1894–95 and Russia in 1904–5 (Yamamura 1977: 113). The strong-army policy created arsenals, shipyards and modern factories, which acted as effective centres for importing, absorbing and disseminating western technologies, including managerial methods. The military industrial build-up plus Japan’s two Meiji-era wars also created a demand for other advanced industrial products, especially steel and machine tools. In the early 1880s, when the textile industry – the other major advanced industry – was still tiny, the arsenals, shipyards and associated plants employed more than 10,000 workers. The machine shops in these enterprises supplied not only military items, but also machine tools for private firms. For example, the Ishikawajima Shipyard, established in 1876 to produce wooden sailing ships, converted to building iron, steam-powered ships in the 1880s; beginning in 1885, it made boilers for textile factories, stone crushers, iron bridges, carpet-weaving machines, printing presses and other machinery for private firms (Yamamura 1977: 118). To meet the exceptional demand for war materiel during the Russian campaign
The nineteenth-century transformation 37 in 1904–5, orders were placed with private firms without previous military production experience. These orders required training workers and management in new skills as well as transferring production equipment and manufacturing methods. An additional catalyst for industrial progress was the desire to produce as much as possible domestically. Two reasons were given for this desire by Masayoshi Matsukata, who had become prime minister in 1891: imported goods were very expensive in terms of scarce foreign exchange; and imports exposed the country to the risk of being cut off in an emergency. These national security rationales were used to justify building a modern steel mill in the face of high costs and delays. Despite persistent losses at the government-financed Yawata steel mill, by 1915 it was competitive with imported steel in simpler product lines; however, it could not turn out more complicated items at a lower cost than European mills, despite the protection of tariffs and lower transportation costs. The military-driven accumulation of industrial capital and the mastery of advanced technical skills that were then disseminated throughout industry undoubtedly produced benefits for the entire economy and for subsequent growth. However, Yamamura (1977: 130–2) notes that other scholars’ macroeconomic analyses of the effects of military spending suggest that such expenditures also may have retarded long-term growth by diverting resources from more productive uses or by inducing inflation that impeded short-term growth. The question about the effects of military spending, in general, has been debated for decades. The indisputable aspect of military spending is that resources are allocated away from other uses in a guns versus butter trade-off. To the extent that these resources would have gone into more productive economic activities, the military will be a drag on the economy, regardless of the technologies or capital acquired through military spending. On the other hand, if defence spending adds to aggregate demand and higher employment, it may enhance growth in a demand-deficient period. In addition, specific types of military procurement of the type described by Yamamura, especially of advanced technology or research, may produce externalities that confer spill-over benefits to civilian industry or that train workers and managers in more productive and advanced methods. If the nation’s security situation demands spending on military activities, such externalities and spill-overs may soften the negative effects of resource diversion. One review of 27 studies on the effects of military spending on economic growth notes ambiguous findings across the research. This literature review divides the models into two basic types: supply models and demand models. Supply models are based on production functions; a typical model includes three sectors – private, non-military public and military – each with its own share of capital and labour. The studies usually allow each sector to influence the others, either positively or negatively. Since there are many possible ways for military spending to influence the economy, there is no theoretical prediction as to the net effect. A good analytical structure allows the data to determine whether there are positive or negative spin-offs, externalities or cross-effects between sectors (Sandler and Hartley 1995: 220). The other approach is to look at the demand side. In these models, output
38 The nineteenth-century transformation is broken down into the standard demand elements of consumption, investment, government spending and net trade balance. Military spending is then added to this standard mix. The authors note that demand models generally bias the results to show a negative impact of military spending since the military sector must compete with, or crowd out, other components of demand. The empirical question is how big this effect is. By carefully analysing the different models and approaches, the authors find a good deal of consistency, once the biases of the underlying approaches are considered. They conclude their review by noting: Models that included demand-side influences, whereby defence can crowd out investment, found that defence had a negative impact on growth. In contrast, almost every supply-side model either found a small positive defence impact or no impact at all. The findings are amazingly consistent despite differences in the sample of countries, the time periods, and econometric estimating procedures. Since we suspect that these supply-side models exclude some negative influences of defence on growth, we must conclude that the net impact of defence on growth is negative, but small. Sandler and Hartley (1995: 220) Although data for Japan are included in several studies, few single out that country for attention. One study that looks at modern Japan attempted to remedy some of the modelling problems identified by Sandler and Hartley. It estimates the growth response to military spending in 80 countries from 1960 to 1991. The evidence for Japan is that the elasticity of GDP to military spending is –0.25; that is, GDP declines by one-quarter of a percentage point for a 1 per cent increase in defence spending (Heo 1998: 654). The problem with this approach is that Japanese spending on defence in the post-Second World War period has been a small percentage of GDP and has been relatively stable; therefore, one may question whether there is sufficient variability in the data to estimate the desired relationships. Moreover, these data do not cover the pre-war period. Another study looked only at Japan and the United States from the 1880s, selecting these two countries because of their dramatically different military spending patterns over time; Japan focused heavily on defence development in the decades before the Second World War and very little thereafter, while the US pattern was the reverse. The authors argue that the shifting patterns of expenditures in these two countries would help to answer the question of the relative effects of crowding out and externalities. The possibly time-varying effects were examined by estimating moving 40-year windows from 1880 to 1990. The revealing finding was that the size effect of military spending was negative throughout the 120 years of their sample; externality effects, however, were positive (Ward et al. 1995: 38). These results are consistent both with the qualitative evidence presented by Yamamura and with those he cited that emphasized the negative macroeconomic effects. The net impact of the size and externality effects is not reported in the
The nineteenth-century transformation 39 study. However, when I apply the reported coefficients to the average value of the variables over the 40-year windows, military spending had a net positive influence on GDP growth throughout the pre-1940 years. According to these estimates, Yamamura was right to focus on the growth effects of military industry. Progress, but . . . Despite Japan’s undeniable industrial progress following the Restoration, especially in cotton production, one could ask why it took so long for Japanese manufacturers to compete successfully with British and American textile manufacturers. From the 1880s, Japanese firms used the most modern machinery with easy access to British assistance and advice; moreover, labour costs were only 10–12 per cent of American and British rates. The quality and training of workers were similar to that found in British and American mills. A landmark study concluded that it was the intensity of labour – how hard and how fast workers did their job – that was the major difference – not only in Japan, but also in other developing countries (Clark 1987: 171). Clark concluded that the failure of poor countries to dominate textile production: proceeds overwhelmingly from inefficient labour rather than from failure to import technology or management skills, or from failure of local capital or input markets, or from scale effects within the textile factories or at the industry level. In cotton textiles all these problems proved minor. Clark (1987: 142) But, the source of such large productivity differences requires further consideration. Why did American workers tend 4.7 times as many spindles and 5.6 as many looms as did Japanese? When asked to do more, workers in the poorer countries objected with walkouts and strikes, even though they could have earned more if they worked more efficiently; nevertheless, within a decade or two, these same factories achieved much higher rates of machinery per worker. Although he does not state it explicitly, Clark implies that social norms on what is appropriate workplace behaviour played an important role in economic development; more over, those norms appear to change with the economy-wide level of productivity – what is acceptable at a later period is unacceptable when average wages are lower. Although individual firms and industries could be more efficient, they are held back by generally accepted ideas of what is proper. Such social force would continue to influence Japanese economic behaviour.
3 From Meiji to the Second World War Political and international developments
Japan’s modern economic development is enmeshed with its experience as the world’s newest colonial empire in the first half of the twentieth century. Its expansionist history depended on the political architecture that unfolded out of the nature of the Meiji Restoration itself. Therefore, an understanding of economic affairs must be accompanied by an excursion through politics and Japan’s colonial experience.
Overview of political developments The new constitution of 1889 provided for a Diet, or parliament, of two houses, the upper one reserved for a European-style peerage composed of former daimyo, government officials and military leaders. Members of the lower house were elected by men over 25 years of age who paid more than a specified amount of direct taxes, which was gradually reduced and then eliminated as a qualification for voting in 1925. The first restrictions limited the electorate to less than 2 per cent of the population; by 1925, this had expanded to 22 per cent. Nominally, the emperor commanded the armies, made war and peace, and had the power to dissolve the lower house. The authority for budget increases, including those for the military, resided in the lower house; if the Diet did not approve a new budget, the one from the previous year was renewed automatically. Despite vigorous and even raucous political life among the ordinary members, Diet leaders held to the principle that the government represented the emperor and therefore should approve government requests and be independent of party politics. This approach lasted until the new century, when the pull of party politics gradually intruded into parliamentary affairs. However, even before 1900, political conflicts among the rising business class, the military, the ageing Meiji-era leaders, government bureaucratic experts and a new class of professional politicians kept political life unsettled. A party leader became prime minister in 1918, the first non-titled person to hold this rank, but policies that he introduced to control the military and bureaucracy were truncated in 1921 by his assassination. In the early 1920s, small political parties consolidated into two centrist parties, which alternated in forming governments until military interests dominated the government in 1932.
From Meiji to the Second World War 41 Attempts by party leaders to form governments or introduce policies that might constrain military affairs were often thwarted by a central feature of the political structure: the army’s chief of staff did not report through the cabinet’s war minister but had direct access to the emperor through his role as commander-inchief. The navy had a less prominent part in shaping policy until it achieved equal status with the army after the war with China in 1895. The absence of a unified authority – whether embodied in the Diet, prime minister or cabinet – held great consequence when government power became fragmented after the war with Russia in 1905 (Beasley 1987: 36). International relations, often determined by the military actions of an uncontrolled army, were out of the hands of the nominal political leaders. The Meiji constitution was so ambiguous over the assignment of executive authority that prime ministers and their cabinets were often forced to compromise or accept outcomes that were inconsistent with the usual notions of parliamentary government. The Taisho era (1912–26), which followed the death of the Meiji emperor, was marked by tensions among the liberalizing tendencies in politics and economics, nationalist movements among junior military officers and right-wing groups, and expansionist views held by the military leadership and like-minded politicians as well as by the public. These tensions continued until Japan’s Second World War defeat in 1945. Military leaders distrusted party politicians, who, they thought, put their own narrow interests ahead of national priorities. These beliefs were reinforced by the government’s acceptance of lower limits on the Japanese navy than on the American and British navies at the Washington naval disarmament conference of 1921–22 and the London follow-up in 1930. Right-wing nationalist organizations increasingly focused on internal purity, external expansion and distrust of business motives and western values. The 1930s Japanese army commanders in Manchuria, in particular, were suspicious of capitalist motives and advocated comprehensive planning rather than markets for the economic administration and industrial development in their area of control. Rising nationalist sentiments were marked by assassinations carried out by disaffected youth, young army officers and nationalist activists; their targets included political leaders who attempted to control the military or who opposed the Asian adventures favoured by the activists. For example, in February 1936, rebel army units killed several leading government officials, just missing the prime minister. The rebels held much of central Tokyo for several days before being arrested. The elimination of leaders who defied the nationalist-military logic achieved what elections and politics could not attain. Prime Minister Tsuyoshi Inukai was assassinated in 1930; his successor Makoto Saito, after leaving the premiership, was killed in 1936 in an abortive military coup; and Keisuke Okada escaped an attempt on his life in 1936. In an effort to form a government that would appease the military and the political parties, Fumimaro Konoe, a popular member of an ancient court family, formed a government in 1937, but one that was heavily tilted towards the military. Nevertheless, popular support for parliamentary government persisted
42 From Meiji to the Second World War throughout the rise of military power, as shown in elections in the spring of 1937, when a general election favoured an anti-military slate of candidates. However, the largely independent army operating in China fomented war there in the summer that pre-empted political decisions from Tokyo. Installation of ineffective cabinets put the military in de facto control of the government.
Japan’s geographical expansion Japan expanded its boundaries almost from the first years of the Restoration. This expansion included the annexation of regional islands, colonial acquisitions on the Asian mainland and imperialist ambitions that brought it to war with the western powers. Most writers on Japan’s imperialism are careful to note that there were many causes behind this expansionist tendency. Strategic aims were always in the minds of the military leaders, especially in the army. Table 3.1 gives a synopsis of Japan’s expansionist activities through the 1930s. In the first phase, the annexation of several island groups in the Japanese cultural sphere established a defence zone for the main islands. In 1872, Japan compelled China to recognize its authority over the Ryukyus, which were annexed in 1879. Russia, in 1875, agreed to transfer its claims over the Kurile Islands north of Hokkaido in exchange for Japan’s recognizing the whole of Sakhalin Island as Russian. Other chains of small islands were annexed over the next 15 years.
Table 3.1 Japan’s geographic expansion to the 1930s Place Ryukyu Islands Kurile Islands
Year in 1872–79 1875
Year out Present 1945
Bonin Islands (Ogasawara) Volcano Islands (Kazanretto, including Iwo Jima) Taiwan Karafuto (Southern Sakhalin) Korea Manchuria Liaotung, China (Kwantung, including Port Arthur) Shantung, China (Kiaochow) Mariana, Palau, Caroline, Marshall Island groups Sakhalin
1876
Present
Circumstances Annexed Exchanged for Sakhalin with Russia Annexed
1891
Present
Annexed
1895 1905
1945 1945
Treaty ending Japan–China war Japan–Russia war settlement
1905–10 1905–32 1905–10
1945 1945 1945
Expanded in stages to colony Expanded in stages to colony Ceded to Japan after Japan–Russia war
1914
1922
1914
1945
1917
1924
Ceded to Japan on entering the First World War Seized German islands in the First World War Seized during Russian revolution
From Meiji to the Second World War 43 Army leaders had their eyes on Korea as early as the first year of the Meiji era. By 1890, with Russia’s Trans-Siberian railway nearing completion, Prime Minister Aritomo Yamagata, who had formerly been army chief of staff, outlined a logic that would serve to justify control of a widening sphere of territory: it was a fundamental principle, he said, to hold a line of advantage to defend the home islands; an essential element of this line was Korea, which would be at risk once Russia completed the railway. With Korea at risk, the whole of East Asia would be threatened. Therefore, it followed, measures to guarantee the independence of Korea were crucial to the defence of the home islands (Beasley 1987: 46). As territories on and around the Asian mainland came under Japanese control, the logic of this argument pushed out the boundaries of what was needed to defend what was held. Thus, Manchuria was needed to protect the Japanese holdings in northwest China, which were required to defend Korea, which was the main line of defence for Japan itself. Anti-foreign insurgency activities in Seoul in 1894 provided the pretext needed by the army staff to send troops to quiet the disturbances. The Japanese foreign minister sought China’s support for joint actions to secure Korean reforms, a proposal that he knew would be unacceptable to China. China’s rejection of these proposals provided diplomatic cover for the initiation of hostilities against China. The ease and rapidity of Japanese victories prompted thoughts of further territorial gains. In the 1895 peace treaty negotiations, Japan sought Taiwan in order to block French or British seizure of the island, which would have threatened southern Japan and the Ryukyu Islands. In addition, Japan received rights over Port Arthur and the Liaotung Peninsula. However, Germany, Russia and France joined in the so-called Triple Intervention to oppose this grant of authority to Japan; one week after signing the peace treaty that incorporated the Port Arthur and Liaotung provisions, the three countries forced Japan to abandon its claims. The European intervenors feared China’s dismemberment and Japan’s expansion into Asia; moreover, Germany desired its own naval base on the China coast, either at Port Arthur (now part of Dalian) or on the Shantung Peninsula across the Yellow Sea. The perceived humiliation implied by the Triple Intervention aroused strong nationalist sentiments and resentments among the Japanese people. Widespread outbursts of popular indignation were backed by press criticism of the government. The army and navy took advantage of public support to launch a military build-up. For government officials, from the prime minister on down, the insult from the western powers loomed large in their strategic thinking and supported their determination to learn from this diplomatic defeat, mainly by reinforcing the nation’s military strength (Ikle 1967: 129). Subsequently, in 1898, Germany obtained rights to the Shantung Peninsula and Russia to Port Arthur and Liaotung. Russia extended a spur of the Trans-Siberian railway to Port Arthur plus an east–west line across central Manchuria linking Vladivostok with the Trans-Siberian line in a more direct route than around the Chinese borders.
44 From Meiji to the Second World War In 1900, Russia sent in large numbers of troops to protect its rail line in Manchuria following attacks on Russian railroad installations by Chinese nationalists associated with the Boxer rebellion. Japanese leaders, led by Yamagata, who had returned to lead his second government a few months earlier, were convinced that Russia’s advances would not stop in Manchuria, but would inevitably extend into Korea and threaten Japan itself. In order to defend against this possibility, the army staff sought to confront Russia militarily, but the country’s diplomats, remembering the Triple Intervention, countered that they required at least the acquiescence of one major European power. Nevertheless, Japan sent in the largest of several foreign military contingents to suppress the Boxer rebellion. Subsequently, Japan established an alliance with Great Britain in 1902. Having failed to obtain Russian agreement to its demands (especially removal of Russian troops from Manchuria), Japan decided on war in late 1903 and launched attacks on Russia early the following year. After Japan sank two-thirds of the formidable Russian fleet in the Tsushima Straits in May 1904 and captured the Russian-held city of Port Arthur, the US president, Theodore Roosevelt, mediated a peace treaty between the two sides. Roosevelt was concerned with both Russian expansionism and rising Japanese power. In the peace conference held at Portsmouth, New Hampshire, in August 1905, Japan gained control of the Liaotung Peninsula, including Port Arthur, and the South Manchurian railroad, as well as half of Sakhalin Island. Russia agreed to evacuate southern Manchuria, which was restored nominally to China. Russia also accepted Japan’s control of Korea, which Japan formally annexed in 1910. Military influence in the government increased as Aritomo Yamagata moved from the premiership to other high-level councils, including chief of the army staff and member of the genro (an advisory council to the emperor composed of the ageing Meiji leaders). From his position in the political shadows and backed by the military and the government bureaucracy, Yamagata influenced the selection of prime ministers and the shape of foreign relations until his death in 1921. Following Yamagata’s logic as well as popular demands, many in the elite circles in and out of government anticipated an independent Manchuria under Japan’s sponsorship, perhaps in cooperation with Russia. The overthrow of China’s Manchu dynasty in the winter of 1911–12 altered Japan’s strategic thinking. Although the cabinet turned down a proposal for a Japan-controlled Manchuria in 1912 as too risky, the goal did not vanish. Tellingly, the cautious views of the foreign minister on this policy led to his assassination in 1913. The idea of an independent Manchuria recurred regularly in army plans over the next two decades until it was finally achieved in 1932. The ageing and deaths of the original Meiji leaders, and then the death of the Meiji emperor himself in 1912, left the government in an increasingly fragmented state. The historian W.G. Beasley described Japan’s expansionism under these conditions as taking place under ‘a diversity of sub-imperialisms, rarely checked and never coordinated by the central government’ (Beasley 1987: 104). With the outbreak of war in Europe in 1914, Great Britain, Germany and Russia were absorbed by their war efforts and paid less attention to activities in Asia,
From Meiji to the Second World War 45 leaving Japan’s ambitions almost unchecked. In late August 1914, in support of its British alliance, Japan declared war on Germany and moved troops into Germany’s leased territory on the Shantung Peninsula. It also occupied German-held islands in the western Pacific: the Mariana, Palau, Caroline and Marshall Island groups. When China sought the return of Shantung, Japan reacted by presenting its own ‘Twenty-One Demands’; China accepted most of them in 1915. The result was a considerable expansion of Japanese interests in China, especially in Manchuria. Elements in Japan’s Kwantung (Liaotung) Army in Manchuria precipitated incidents beginning in 1928 to force Tokyo into accepting the de facto Occupation of all of Manchuria. Although there are some questions about its origins, elements of the Kwantung Army probably designed the so-called Manchurian incident of 18 September 1931 to draw Japan into expanding warfare in China and a fullscale takeover of Manchuria. Japanese troops used the pretext of a bomb (planted by Japanese soldiers) along the Japanese-controlled South Manchurian Railway to occupy Mukden. With reinforcements from Korea, the Japanese army began to expand throughout northern Manchuria. Neither the army’s high command in Tokyo nor the prime minister was able to restrain the Kwantung Army; within three months, Japanese troops had spread throughout Manchuria. Throughout the 1930s, field commanders took initiatives, sometimes – but not always – endorsed by the high command in Tokyo, and reluctantly accepted by the cabinet. Tokyo officials feared, with a good deal of evidence to support their anxieties, that the Kwantung Army was not far from becoming a separate political entity. International economics also helped shape Japanese views towards China and Manchuria. The collapse of world trade in 1929 in the wake of the Great Depression saw the disappearance of Japan’s customary export markets, especially for some agricultural commodities, silk and manufactures; large numbers of unemployed workers and devastated farmers faced starvation. Economic policy shifted from an emphasis on trade to autonomy and self-reliance. The collapse of trading relations had the side-effect that possible economic threats from the great powers, which had acted as a deterrent to bolder Japanese actions, lost much of their restraining influence. In particular, diminished western influence removed any remaining constraints on Japan’s policies towards the mainland. ‘Co-prosperity’ with China and closer economic relations with an independent Manchuria under Japanese control became Japan’s goals, driven by the new combination of economic and political circumstances. With Japan’s traditional trading partners out of the picture, journalists, politicians and army officers looked to Manchuria as Japan’s economic lifeline, including as a place to establish many of Japan’s dispossessed farmers on land of their own. The loss of export markets during the depression was only part of Japan’s economic dilemma. Exports were the necessary cost of imports. The 1929–30 slump threatened the supply of raw materials and other goods on which the economy depended, most especially the heavy industries on which military planners had built their plans. Japan was deficient in some grades of coal and was almost
46 From Meiji to the Second World War wholly dependent on imports for iron. American scrap iron was one source, but the nearest ores were in north China and Manchuria, conveniently also a source of coal. Malaya and the Philippines were other suppliers of iron. In addition, Malaya was a principal producer of tin and rubber. Petroleum, too, was a concern, especially for the navy. More than half of Japan’s supply came from the United States, but 20–25 per cent was from the Netherlands East Indies (Indonesia) to the south, and more could be obtained there if the North American supplies were unavailable because of economic or other reasons. Meanwhile, in China itself, the warring Communists and Kuomintang Nationalists under Chiang Kai-shek came to view Japanese military actions as an attempt to dismember the country. Both sides in the civil war negotiated a limited truce in 1936 in order to form a united front against the Japanese forces in the country. On 7 July 1937, a small Japanese force engaged a resisting Chinese garrison near the Marco Polo Bridge in a village near Peking. Neither side made concessions and the engagement broadened into full-scale hostilities between China and Japan; within months of the ‘China Incident’, Japan’s newly created North China Army had 200,000 men in the field. In the fighting that followed, Japan seized areas in the northeast that included Inner Mongolia, Peking, Nanking and Shanghai; in southern China, it took control of Hong Kong and Canton (Guangzhou) as well as several coastal areas facing the East China Sea. Chiang Kai-shek’s military forces withdrew up the Yangtze River to Chungking in Szechwan province and the Communists organized guerrilla actions against the Japanese forces throughout the country. Although the Japanese army held the principal cities, ports and means of transportation, fighting continued until 1945. The main economic motives for Japan’s moves into Korea, north China and Manchuria – access to raw materials and export markets to generate the currencies to pay for imports – also applied to Southeast Asia as well. However, the European powers had greater concerns for that region than for the northern areas. As the idea of economic autarchy developed, it included an element referred to as ‘advance to the south’, which had a defensive military rationale as well as an economic motivation. Foreign Ministry officials were aware that any such moves would excite Great Britain, the Netherlands and the United States. While some military leaders were looking forward to war with these nations, others urged that caution accompany ambition. A 1936 cabinet policy document stated that expansion to the south was to be primarily economic and was to be achieved ‘by gradual and peaceful means’, avoiding provocation of other countries as much as possible (Beaseley 1987: 224). The causes and history of Japan’s Pacific war are not our main concern here, but we will record the key steps along the way. Anti-Japanese sentiment was rising in the United States as Japan fought Chiang Kai-shek’s Nationalist army, which had wide American popular as well as government support; the sinking of an American gunboat in the Yangtze in 1937 added to the emotions, so there was little surprise when the United States renounced its 1911 treaty of commerce with Japan in 1939.
From Meiji to the Second World War 47 Germany’s 1940 invasion of the Netherlands, Belgium and France, and Britain’s almost single-handed opposition to Germany, allowed both the civilian and military leaders in Japan to embrace more belligerent policies, with less concern over European opposition. The army and navy accepted the formula of ‘defence in the north, advance to the south’. When the second Konoe government was formed in July 1940, the cabinet approved a policy anticipating a resort to force in Southeast Asia, ‘if circumstances at home and abroad make that advantageous’ (Beasley 1987: 226). In 1940, Japan invaded northern Indochina, primarily as an attempt to block supplies to the Chinese Nationalists, but also as part of a broader move to the south. The United States reacted in July 1941 by freezing Japanese assets and embargoing oil, followed within days by Britain’s abrogation of its commercial treaties and Dutch restrictions on oil and bauxite exports from the Netherlands East Indies. These moves by the western powers pushed the hitherto more moderate navy into collusion with the more extremist views of the army. Prime Minister Konoe – faced, on one side, by the United States and, on the other, by his own military leaders, who wanted to seize the sources of oil production in the Dutch East Indies – told US diplomats that he wanted to negotiate an Indochina withdrawal directly with President Roosevelt, which would strengthen his hand in dealing with his own military. However, he was headed off by the State Department, which wanted prior concessions before any meeting. Feeling that negotiations had collapsed, Konoe resigned in October 1941, replaced by his war minister General Hideki Tojo, who continued the talks but was met again with the American line of agreement first, presidential meeting second, which the new Japanese government would not accept. In the meantime, the army’s plans for seizing oil fields and other resources to the south were already at an advanced stage, as was the navy’s preparations for a strike against the American naval fleet at Pearl Harbor, the only force capable of slowing Japan’s plans – given the inability of the other colonial powers to respond because of their desperate situations in Europe. Economic sanctions against Japan, particularly of oil and steel, forced the economic planners to face the consequences of diminishing stockpiles, which, the military authorities reminded them, required taking over Asian supplies. As became evident to most parties, failure to make progress in the Japan–American talks was equivalent to a march towards war. In December 1941 and the first months of 1942, Japan attacked American forces in Hawaii and seized Burma, Thailand, the rest of French Indochina, British Malaya, the Netherlands East Indies and the Philippine Islands. Occupying and administering this vast territory, as well as organizing the home front for total war, required new forms of organization and control. Although Japan emerged from its 1945 defeat with the military’s reputation in ruins and enthusiastic popular support for a pacifist constitution imposed by the Americans, the bureaucracy and its methods of wartime economic organization lived on in the post-war period.
4 Economic developments in the first half of the twentieth century
An economy on the march The years following the Meiji Restoration were ones of institution building, often in a trial and error process. However, an enduring influence on economic policy for many decades was the fear of colonial domination by the western powers. The vivid examples of the treaty port system imposed by the industrial military powers of Great Britain, the United States, France and Germany on Vietnam, Korea, Siam and China, as well as the six ports opened in Japan itself, were constant reminders of the need to industrialize quickly. The positive effects of state building and industrialization were readily absorbed by Japanese planners: at the time of the Chinese revolution that overthrew the Ching dynasty in 1911, 50 Chinese cities were subject to extraterritoriality; Japan, however, abolished these foreign rights in 1899. Japan was able to negotiate the end of foreign intrusion largely because it had built the institutions of laws, courts and enforcement that satisfied western qualms on the quality of domestic justice (Kayaoglu 2006: 20). Economic growth and development responded fitfully and episodically to Japan’s wars, the First World War in Europe, global depression in the 1930s, and the drive to develop industrial and military power. Military spending for the wars with China and Russia stimulated aggregate demand in 1894 and 1904. Disruptions caused by the European war, starting in 1914, shifted much Asian demand from the warring nations to the newly industrializing Japan. Economic growth, which had averaged a bit more than 2.5 per cent from 1885 to 1913, jumped to 6.9 per cent during the European war years of 1914–19. As Europe recovered after the war, Japan’s boom turned into a slump in the same industries that had benefited just a few years earlier. Growth rates over selected periods are shown in Table 4.1. The decade following the post-First World War collapse was not a good one for Japan. Business expansion during the war turned into excess capacity as demand fell when European producers regained their old markets; falling profits and undercapitalized banks led to several banking panics. In 1920, a stock market crash was accompanied by collapsing prices of commodities such as rice, raw silk and cotton yarn; several trading companies, banks and manufacturing businesses went into bankruptcy. As this crisis was being resolved, smaller panics
Economic developments in the first half of the twentieth century 49 Table 4.1 Average growth rates of real GDP in selected periods, 1885–1940 Period 1885–1913 1914–19 1920–31 1932–40
Growth rate (% per year) 2.5 6.9 1.9 5.1
Source: JSA (1999: Table 13.3).
were touched off in early 1922 when a speculator went bankrupt, and later that same year when several banks in western Japan failed. The Tokyo–Yokohama earthquake of 1 September 1923, caused massive physical destruction and considerable financial disruption. More than 100,000 people lost their lives, a half-million lost their homes and one-quarter million lost their jobs. The government and the BOJ quickly introduced policies to avert some of the avoidable financial consequences; for example, they proclaimed a payments moratorium in the hard-hit areas and bought commercial bills from businesses that could not collect payments because of the disruptions (Nakamura 1988: 456). Seeing an opportunity, banks sold many formerly uncollectible loans to the central bank, which found subsequently that almost half the earthquake bills were worthless. Relief expenditures supporting earthquake victims used up Japan’s foreign reserves, depreciating the yen against the dollar and gold. Yen depreciation made it harder for borrowers of foreign currencies to repay their loans, which contributed to another financial panic in early 1927, when a Kobe trading company was unable to meet its obligations. At the same time, the Diet held that the uncollectible earthquake bills were the responsibility of the originating banks rather than the BOJ, which caused the failure of the semiofficial Bank of Taiwan. In the wake of that failure, it was revealed that many other banks suffered from bad loans and mismanagement; bank runs occurred at several banks as depositors, fearing for the safety of their money, rushed to withdraw their deposits. The government’s inability to deal with this panic caused the cabinet to collapse. A new government closed banks throughout the country for several days and announced a 20-day payments moratorium to give time for the authorities to reorganize failing banks and to provide new capital. Nevertheless, 32 banks suspended operations. General economic weakness in the 1920s forced many companies to reduce the scale of their operations or to go out of business. The most common path to business failure was for a bank to initiate the action by imposing a freeze on transactions of a company that failed to pay its bills or had twice bounced cheques within six months. For the affected business, such a credit freeze – called suspension of business transactions with banks – was equivalent to a death sentence. In 1920, the number of such failures doubled from the year before, and it doubled again by 1926, peaking in 1930 at more than 11,000 (see Figure 4.1.). Other companies simply disappeared without formal notification. Many displaced workers found refuge by establishing their own businesses, usually small manufacturing operations. A typical factory might be an owner–operator of a single machine tool such as a lathe, plus one helper.
50 Economic developments in the first half of the twentieth century
12,000
40
10,000
30
8,000
Inflation rate (right-hand scale)
20
6,000
10
4,000
0
2,000
Business failures (lefthand scale)
Inflation rate (%)
Business failures
One source of the banking panics and business failures was the steady rate of deflation from 1920 to 1932, which followed rampant inflation during the First World War. As shown in Figure 4.1, the number of failures was negatively related to the inflation rate. Throughout the 1920s, the Japanese government had planned to return to the gold standard, which it had abandoned along with most other countries in 1916 and 1917. The gold standard had stabilized exchange rates, enlarged world trade and helped to maintain discipline on domestic monetary policy and price levels. As argued by Robert Mundell in a lecture delivered on his receiving the Nobel Prize in economics, ‘The system gave the world a high degree of monetary integration and stability.’ (Mundell 2000: 328). The beneficial effects ended when widespread government deficits caused by paying for the European war forced governments to abandon the links between gold, domestic money creation and international exchange rates. Mundell’s explanation of the gold standard’s beneficial results was that the price of gold was relatively stable over the long run, which lent stability to domestic prices and international exchange rates. In the short run, however, the random nature of gold discoveries on the supply side, and the new industrial uses of gold or political crises on the demand side, made gold prices, exchange rates and domestic prices more volatile than they would be under the post-Second World War order. After the First World War, events conspired to render the gold standard a less desirable exchange rate regime than it had been before the war. War-induced deficit spending and inflation made relative prices across countries different from those that had prevailed when the gold standard was abandoned. Additionally, the
-10
0 -20 1895 1900 1905 1910 1915 1920 1925 1930 1935 1940
Figure 4.1 Business failures (suspension of business transactions with banks) and inflation rate (% change of GDP deflator, two-year average at annual rate), 1895–1940. Sources: JSA (1999: Table 11-21); Ohkawa et al. (1979: Table A50, 387).
Economic developments in the first half of the twentieth century 51 1913 creation of the Federal Reserve System in the United States suddenly gave the world a central banker that was considerably larger than any other – one with the power to act as an oligopolist, setting the price of gold rather than being a price taker in a decentralized, competitive market. According to Mundell’s analysis, the Fed first forced the gold price to decline by half, which brought about global inflationary pressures; then, it engineered a price increase that sent economies into a decade-long deflation. A consensus for re-establishing the gold standard emerged at a 1922 international monetary conference in Genoa, Italy. Despite the American intervention to raise the price of gold, in 1921 it was still 40 per cent below the pre-war equilibrium value. Great Britain insisted on restoring the pound sterling to the pre-war value; the United States had already returned to gold at the pre-war parity in 1919. Germany, France and Italy revalued their currencies in line with the different inflation rates they had experienced. Japanese officials had participated in the Genoa conference and intended to implement a return at the pre-war parity. Japanese financial journalists at the conference sent home stories noting that prominent economists, chief among them John Maynard Keynes, were publicly urging their finance officials not to resurrect the gold standard; if they insisted on doing so, to do it at new exchange rates that reflected relative price changes since 1916; if they were so stubborn as to restore the old rates, then to try to bring down domestic prices and wages such that the external rate would be consistent with the domestic price level (Nakamura 1988: 464). Japanese officials postponed their move for several years because of the instability caused by the several banking panics, business failures and the Tokyo earthquake. In January 1930, the government lifted its embargo on trade in gold and returned to the gold standard at the pre-war parity. The 1916 exchange rate had been ¥100 to $50; the yen had depreciated over the intervening years and slipped as low as $43.75 for ¥100 in 1929. Restoration of the old parity required deflationary policies to drive down domestic prices and wages to a level compatible with foreign prices. Deflationary effects from restoring the pre-war parities were predicted by leading economists. When the British government returned to the gold standard in 1925 at the old rate, it ignored Keynes’ arguments, which – among other things – predicted that an inappropriate exchange rate would force coal exports to fall with accompanying labour unrest. Indeed, Great Britain was able to maintain labour peace and a continuing supply of coal only with a subsidy over the following winter; but then the country suffered a strike in the coalfields and an economywide general strike the next year. Seven years later, Great Britain abandoned the gold standard (Harrod 1951: 358–61). The same predictable disaster that befell Great Britain struck the Japanese economy. The finance minister put forth two rationales for the return to gold at the old rate: national prestige and the necessity of ‘consolidating the business world’, by which he meant eliminating non-competitive firms through the pressure of an appreciated currency (Nakamura 1988: 464). Indeed, Keynes’ logic on the depressing effects of returning to gold at the old parities was understood and
52 Economic developments in the first half of the twentieth century accepted by the Japanese authorities. Following the return to gold, cheaper coal imported from Manchurian mines forced the dismissal of 40 per cent of the coal industry workforce. Attempts to push down wages in other industries provoked strikes and general labour unrest. Efforts to maintain prices in the face of foreign competition led firms to organize price cartels. In turn, the government enacted the Important Industries Control Law, which allowed it to force companies to adhere to cartel agreements. General deflation in 1930 and 1931 averaged 10 per cent per year. As austerity policies pushed prices down and unemployment up, and as the government attempted to rein in military budgets in the tight fiscal situation, an assassination attempt in April 1930 was made on the prime minister, who died of his wounds several months later. A few days after the April 1931 Manchurian incident, Great Britain went off the gold standard. Japan was drawn into outright warfare at the same time that currencies became unstable; speculators quickly saw that the yen probably would be forced to sever its links to gold. The central bank raised interest rates to protect the currency and impose losses on speculators. As the end-of-year contract dates for gold approached, the cabinet collapsed and a new prime minister and finance minister took office in mid-December. Two days later, the finance minister, Korekiyo Takahashi (who would hold the job until assassinated in 1936), took the country off the gold standard; he also lowered the central bank’s discount rate, relaxed monetary conditions and let the yen fall to a new equilibrium of around ¥100 to $31 over the following year – a real depreciation in terms of the relative prices of Japan and its major trading partners. He increased fiscal spending, mainly on rural support and military expenditures, each of which increased by about 2 per cent of GDP. Rather than financing the spending through taxes, he issued bonds accompanied by expansionist monetary policies, which acted to transfer the funding to the central bank. Takahashi explicitly argued that government had a role in promoting aggregate demand. As shown in Table 4.1 and Figure 4.1, the economy responded to this medicine; deflation ended, the number of business failures plunged and growth resumed at almost the same pace as during the First World War boom. However, the success of Japan’s exports arising from real yen depreciation encouraged boycotts of Japanese goods in Europe and the United States, which further shifted Japanese opinion against international integration.
Growth of the zaibatsu The great financial and industrial combines known as zaibatsu rose in size and power after the Meiji Restoration. (The word zaibatsu comes from zai wealth and batsu, variously translated as clique, powerful person or family, or person of influence; the term was coined in the 1930s to refer to financial cliques, often tinged with the meaning of robber baron.) The four main zaibatsu – Mitsui, Mitsubishi, Yasuda and Sumitomo – had their origins in pre-Meiji times. Mitsubishi began as a Tosa domain trading company with shipping interests; privatized after the Restoration, it was purchased from the state by its former manager. Domination
Economic developments in the first half of the twentieth century 53 of shipping led to expansion in coal mining and numerous other industries. Mitsui and Sumitomo both originated as trading houses in the seventeenth century. Mitsui specialized in dry goods and currency exchange, including trading in government notes based on Osaka rice warehouse receipts. Sumitomo traded in copper, a business line that extended to recent times; in 1996, a Sumitomo copper trader in New York was responsible for a $2.6 billion copper trading fraud emanating from the escalation of a series of loss-making transactions. Zenjiro Yasuda, the founder of the Yasuda zaibatsu, ran away from home in the 1850s to become a shop assistant in Edo, then a money changer and broker, a major lender to the new Meiji government, the founder of a giant banking empire (including Fuji Bank – one of the largest Japanese banks until its merger with two others in 1999) and the buyer of industrial enterprises, railways and shipping companies. A distinguishing characteristic of these four and other smaller zaibatsu was their close family control through interlocking shares and holding companies in a bewildering array of financial, industrial and other businesses. Another feature that prolonged their influence was their tendency to rely on competent managers rather than insisting on direct family management. The major zaibatsu also controlled banks and other financial institutions; Mitsui and Yasuda incorporated financial capabilities early in their development and the others picked up banking, brokerage and insurance companies as part of their diversification. However, it has not been established that the financial institutions controlled by zaibatsu offered any special advantage to group members (Miwa and Ramseyer 2002: 155). The production and trade boom during the First World War greatly benefited the leading zaibatsu, which emerged from the war as the dominant companies across a wide range of industries. In the post-war panics, smaller zaibatsu suffered from the failures of their own speculative enterprises and from their customers’ collapse; however, since the four main zaibatsu, with their conservative and competent management, suffered few losses in the panics, their dominant positions increased (Nakamura 1988: 455). In the panic of 1927, the big zaibatsu, with their stocks of cash, extended their ownership by acquiring firms in the newer industrial products or by starting their own in fields such as rayon and chemicals. One economic historian wrote that the zaibatsu probably hit their peak of power and economic influence at this time (Nakamura 1988: 459). In the 1920s and 1930s, another class of companies emerged, often called the new zaibatsu. Nissan, operating under the holding company Nippon Sangyo (Japan Industrial Corp.), was one of the few to survive the Second World War. Another was the aircraft producer Nakajima, which became Fuji Heavy Industries after 1945. A common feature of the new zaibatsu was that their founders tended to be technical specialists with military connections. Since they had weak links with the financial community and the older zaibatsu, they frequently depended on government contracts and capital. An anti-capitalist ideology of the Japanese army leaders in Manchuria led them to distrust the old zaibatsu, which, they claimed, were interested in profits above national interests. In fact, a slogan of the Japanese-established Manchurian state
54 Economic developments in the first half of the twentieth century of Manchukuo was ‘denounce capitalism, keep out the zaibatsu.’ When war with China intensified in 1937, the government gave top priority to expanding Manchurian and Chinese coal and steel production. The Kwantung Army first gave the job of Manchurian economic development to the South Manchurian Railway Company (Mantetsu). When the job turned out to be beyond the capability of the railroad company, Nissan’s capital, technology and management were brought in under the umbrella of the Manchurian Industrial Development Company to oversee the heavy and chemical industries. Nissan’s founder moved the company’s headquarters to Manchuria to oversee these efforts (Beaseley 1987: 216). The family-owned zaibatsu did not survive after 1945. The American Occupation authorities dismantled them because of their presumed roles as war financiers and producers. The authorities divested the interlocking shares, abolished the central holding companies and dispersed the shares owned by the core families. The new zaibatsu, having been especially associated with the military, were also dissolved. Nakajima Aircraft was split into more than 15 companies. Each factory began making products using its aircraft technology, such as scooters using bomber tail wheels. Five of the companies later joined forces under the slogan ‘Aircraft again!’ and merged into Fuji Heavy Industries, the maker of Subaru automobiles; the company also produced jet-engine trainer aircraft for the Air Self Defence Force and sections of passenger aircraft for Boeing Corp.
Industrial development and the military Despite the export experience gained during the First World War, the increased participation in international markets made Japanese producers acutely aware of the relatively backward nature of their capabilities. Mitsubishi Heavy Electrical Machinery Co., for example, failed at producing transformers, electric motors and generators in the early 1920s; even such light electrical products as fans flopped because of weak design and deficient technological competence (Yamamura 1986: 74). As a direct result of these problems, Mitsubishi signed a contract with the American electrical machinery producer Westinghouse in 1924 under which the American company provided production and management techniques. One of the most important lessons that Mitsubishi engineers learned from their Westinghouse contacts was how to design new products for which they had no previous experience. The other heavy electrical products companies also turned to foreign manufacturers for technology and production skills. The results of these technology transfer activities became apparent in the 1930s. For example, in 1922–27, fully 98 per cent of Japan’s new steam turbines were imported; by 1933–37, domestically produced items had climbed to 77 per cent of the total (Yamamura 1986: 78). The same process occurred in machine tools, the quintessential industrial product of the machine age. Machine tools are the devices that cut, shape, squeeze and otherwise form steel and other materials into the parts that are assembled into finished products. The surge of machinery orders during the First World War quickly vanished when the European producers re-entered the market. A 1921
exhibition judged the domestic industry harshly: ‘In all categories of machine tools, the quality of Japanese products was no more than an inferior imitation of imported machine tools.’ (Yamamura 1986: 79). As in electrical machinery, linkups with foreign firms provided licences and technical expertise; however, the main avenue of advancement was through copying and imitation, with or without 822271licensing agreements. 932914 The military was intensely interested in the progress of the machine tool indus566894try, which supplied the capital equipment required to turn out the guns, ammuni997945tion, vehicles, aircraft and warships sought by the military commanders. Army 1097561officials took the initiative in advancing the capabilities of domestic companies. 971106In 1928, for example, the Nagoya army ordinance factory bought a large number 986998of milling machines from an American producer and loaned one to each of four 1193384 large Japanese machine tool makers, which were ordered to produce 10 machines 1715337 as nearly identical as possible to the original model. The best imitators earned 2480670 2585949top place in army contract competitions (Yamamura 1986: 82). By the 1930s, 2495193the combination of cheaper electricity distributed by an efficient network and a 3412893capable, modern machine tool industry enabled almost all industries to produce 4942041products that were more advanced than before. 5580526 Electricity and machine tools were used in a very wide range of unrelated products and industries such as bicycles, sewing machines, automobiles, aircraft and guns; indeed, it is difficult to find a product that did not benefit. Convergence in the application of electricity and metal forming also produced feedback from user to producer that carried benefits for all other users. Solving a problem fashioning high-strength steel tubing for bicycles enabled aircraft designers to use similar materials in wing structures. Figure 4.2 shows the real value of machinery production as well as the number of employees in the industry. Because of the very rapid rise in output and workers, the scale is logarithmic, with each division 10 times the preceding one. The speed 10,000,000 000 ¥; number of employees
7 2 1 6 2 2 7 5 8 1 1 3 4 5 3 7 1
Economic developments in the first half of the twentieth century 55
Production value 1,000,000
100,000 Employees 10,000
1,000 1890 1895 1900 1905 1910 1915 1920 1925 1930 1935 1940
Figure 4.2 Machinery production (000 1934–36 ¥ and number of employees, 1880–1940). Source: JSA (1999: Tables 6-06A, S6-1).
56 Economic developments in the first half of the twentieth century 10.0
10
Personnel (right scale)
1.0
Spending (left scale)
1 1875
Military personnel/population (%)
Military spending/GDP (%)
100
0.1 1885
1895
1905
1915
1925
1935
Figure 4.3 Military spending to GDP and military personnel to total population, 1876– 1944 (%). Source: JSA (1999: Tables 26-01A, 26-03A).
of growth can be judged by the fact that fewer than 2,000 people were turning out machinery in 1886, whereas 1938 saw almost 1 million, which doubled to 2 million in 1942. From 1885 to 1915, the value of output increased by more than 25 times, an 11.6 per cent annual growth rate. Production plateaued for more than a decade following the First World War, before taking off again in the 1930s, driven by the substitution of domestic products for imported ones in the early part of the decade, by the resurrection of growth when Japan abandoned the gold standard and by military demands after 1937. The direct impact of the military on the economy can be seen in the value of military expenditures as a proportion of total output and the number of people in the military services relative to the population. Figure 4.3 shows both of these ratios, again using logarithmic scales to accommodate the large range of values. The 1877 Satsuma revolt, the last major attempt of old-guard feudal elements to protest the ending of samurai status and stipends, demonstrated the value of a trained conscript army, which the national government had difficulty in fielding. The army was reorganized and expanded in 1882–84; the 1873 conscription law was revised in 1883 with the aim of placing the army in readiness for crisis or contingency, at home or abroad. The increase in military personnel at that time is evident in Figure 4.3. Before the Sino-Japanese war of 1894–95, military spending was only about 2 per cent of GDP. The war, however, quickly drove that figure up to 10 per cent; it fell back to about half that level until the Russian war, when it climbed to more than 20 per cent. Again, it declined after the war, but not to the old level; it stayed in the range of 4–5 per cent until 1925–56, when it fell below 3 per cent of total
Economic developments in the first half of the twentieth century 57 output as a result of the heavy expenditures occasioned by the Tokyo earthquake and tight budgets resulting from the weak economy. The explicitly expansionary fiscal policies implemented after Japan left the gold standard in 1930 raised the military allocation. War in China in 1937 started the hyper-expansion of military spending, which hit 14 per cent of GDP that year; by 1940, more than one-fifth of all output was diverted to military spending, a ratio that climbed to one-third in 1942 and one-half the following year. The share of the total population in the military services (uniformed and civilian) was well below 1 per cent until the expansion of conflict in China. As the Pacific war unfolded, the total number of people in the war and navy ministries exceeded 10 per cent of the population – an indicator of a state involved in total war.
5 Planning supplies for war
Overview Japan’s war in China after July 1937 expanded unexpectedly in its intensity, duration and geographic range. Raw material shortages quickly appeared, which made the development of Manchuria’s resource base a critical issue. Japan organized its government to plan and administer a war effort that involved mobilization on a scale unseen since the First World War in Europe. Shortages of munitions in Japan’s war with Russia in 1904–5 reinforced the lessons that the European general staffs learned in 1918: total mobilization would be essential if Japan were to be involved in a large-scale war. The government chose to impose economic controls to allocate resources rather than to rely on markets to do the job. After Japan attacked the Pacific territories of the United States, Great Britain, the Netherlands and others in December 1941, counter-attacks on Japanese transportation networks and industrial facilities intensified supply shortages and increased the premium for effective planning. Military planners and government administrators turned to an increasingly controlled and planned economy, taking advantage of the experience gained just a few years earlier in Manchuria’s government-led industrialization. However, despite the passage of the necessary legislation by a militarydominated government and the creation of an implementing bureaucracy within the government, Japan was never able to achieve the coordination of economic activities urged by civilian and military officials. The reasons for the inadequate mobilization effort included slowness in recognizing the scale of the demands, wild disagreements between the army and navy, disputes between business on the one side and military and government planners on the other, differences within the government about the wisdom of planning, and Diet opposition to the government’s militaristic policies.
Planning and administering war production Mobilization planning got its start in 1918 with the passage of the munitions industries mobilization law – the first basic law related to wartime industrial con-
Planning supplies for war 59 trol. It was not applied until 1937, and was replaced by a general mobilization law a year later. However, policies after 1918 included the creation of a succession of bureaus that drew up mobilization plans based on industrial capabilities. In 1927, the Resources Bureau of the Ministry of Commerce and Industry (MCI: the home of the mobilization function) made first use in Japan of the methods then being pioneered in the Soviet Union’s five-year plans. Military officers were assigned to the MCI, which gained power over the finance and foreign affairs ministries when they resisted the growing power of the military. Within MCI, the transferred officers used their bureaucratic and political authority to promote career bureaucrats they considered to be ‘reformists’ and to block the advancement of others. Reformists in the military saw their likeminded colleagues at MCI as people they could work with, as opposed to political party leaders who were viewed as obstacles to the military’s nationalist goals. The ‘reform’ or ‘new bureaucrats’ also were variously described as nationalistic, promilitary, anti-liberal, pro-fascist, anti-party and Nazi (Johnson 1982: 124). The reform cliques in the military and civilian bureaucracies pushed for unprecedented state control over the economy and weakening of political party influence. Many of the individuals were strongly influenced by Nazi and fascist models of statist programmes in Europe. However, the constitution established several co-equal organs – House of Peers, Privy Council, military and judiciary – all of which could and did undercut or otherwise ignore cabinet decisions. Therefore, despite the reformists’ preferences for state control, divisions among themselves and the fragmentation of authority in the Meiji constitution prevented them from establishing either a fully functioning economic planning mechanism or a fascist regime. Moreover, equally passionate conservatives in the political parties, the business community, local elites and the traditional right wing countered the reformists in their endeavours. Nevertheless, the reformists reworked the organizational structures and transformed the state’s relation with civil society (Garon 1990: 345–6). In the face of perceived crises in the economy and the challenge of supplying military needs, economists associated with mobilization planning in the Resources Bureau saw a need for economic controls and the ability to guide specific industries; they sought laws to administer specific strategic sectors. The first of these was the petroleum industries law of 1934. Automobile manufacturing came under government control in 1936; the manufacturers of cars and trucks required government licences, only two of which were granted – to Toyota and Nissan. By 1939, the foreign car manufacturers in Japan (Ford and General Motors) were out of business, replaced by the chosen domestic companies (Johnson 1982: 132). Artificial petroleum derived from coal liquefaction came under government control in 1937, machine tools and aircraft in 1938, shipbuilding in 1939 and ‘important machines’ in 1941. An electric power law was passed in 1938, but only after two years of bitter debate in and out of the Diet. When finally implemented in 1941, the electricity industry was consolidated from 33 generating companies and 70 distribution utilities into nine public utilities, which continued to dominate the Japanese electricity market into the twenty-first century (Johnson 1982: 126).
60 Planning supplies for war Despite the authority granted to the government by these industry laws, they represented compromises between the state and the companies, which were able to retain private ownership and a large degree of self-management. Industrialists struggled against the demands of the reform bureaucrats until the end of the war in 1945; as a result, attempts by the state to assume more complete authority over industry were only partly successful. An important 1937 administrative action was the consolidation of the military and government mobilization and planning offices into a new, more powerful agency, the Cabinet Planning Board, which was responsible for drafting the mobilization law that replaced the old one from 1918. Implementation was to be by imperial ordinances, which the government could issue on its own without reference to the Diet – a process that permitted the government to do almost anything it wanted. This law was fiercely debated and argued by the Diet and other affected parties, primarily business leaders. The Cabinet Planning Board became the home of the reform bureaucrats. It issued its first materials plan at the end of 1937 as a brief report on foreign currency reserves. The following year, the exercise was expanded to 96 commodities that were authorized for import. By 1939, the plan covered 400 commodities with eight priority levels, broken down into sub-priorities. As with most such plans, events quickly made the allocations obsolete. One of the more important sources of disruption was the bitter fighting between army and navy personnel over allocations. In addition, after the outbreak of war in Europe, Great Britain embargoed exports from India, Canada and Australia, which destroyed the import section of the plans (Johnson 1982: 140–1). Foreign trade was even more tightly controlled than was domestic industry. When the finance minister Korekiyo Takahashi took Japan off the gold standard at the end of 1931, an implementing foreign exchange control law that came into force the following year gave the finance ministry authority to approve and license all foreign transactions, a control that would last until 1964. Even before the expansion of fighting following the July 1937 China incident, the government had to find the means to accommodate planned increases in military demand and growing domestic economic imbalances. As early as 1935, inflation loomed as demand outstripped supply. The finance minister tried to rein in military spending, but he was assassinated when groups within the army reacted against what they felt to be undue civilian interference in military affairs. His successor allowed the military budget to expand, which produced a balance of payments crisis and further stimulated inflation. By early 1937, the inflation rate exceeded 20 per cent and the current account deficit reached 2.5 per cent of GDP. Real wages were falling, labour disputes were rising and higher taxes were in the works to pay for the military build-up. However, since the military had ruled out budgetary solutions to this fiscal dilemma, economic controls and rationing seemed to be the only available approach. In mid-1937, the government announced a three-pronged strategy to achieve macroeconomic balance that made use of existing controls, but that also foretold the expansion of government authority over the economy. The new plan sought
Planning supplies for war 61 three objectives: (1) equilibrium in the international balance of payments; (2) expansion of production capacity; and (3) control over the supply and demand of goods. The first objective restricted imports to those products that would support military industry; the second was key to raising military output; and the third was intended to avoid the politically unacceptable effects of income distortions caused by inflation and increased taxes. As implied by the third principle, planning rather than market forces was to be the preferred method for meeting the objectives (Okazaki and Okuno-Fujiwara 1999: 17–18). General wage and price controls came into effect in September 1939 and workers in non-military industries, mainly services and commerce, were required to transfer to military production. Fearing soaring prices from poor rice harvests, the government instituted rationing of essential consumer items in 1940. As often occurs when demand and supply are equilibrated not by prices but by administrative means, black markets quickly developed, which, in turn, triggered additional controls and the establishment of an economics police to crack down on these activities. Meanwhile, in Manchuria, the Kwantung Army and the South Manchuria Railroad – the designated leader of industrial development – were discovering that army officers and railroad managers lacked the skills to organize the chemical and heavy industries in the region. The army in Manchuria distrusted markets; junior and senior army officers alike believed that industry’s competition-driven profit orientation diminished the role of nationalist ideology. Also, there was a strong Marxist strain in both army and civilian thinking among many of those responsible for Manchurian development. The vivid examples of Stalinist planning in the Soviet Union as well as the centralization of economic control in Hitler’s Germany provided accessible blueprints for how to organize a war economy. The army turned to the reform bureaucrats in Tokyo, the best of which transferred to Manchuria. The bureaucrats, for their part, turned to one of their favourite new zaibatsu, Nissan. In one of history’s ironies, Nissan had planned to raise funds in the American capital markets to finance the large investments planned for Manchuria; however, the American government’s condemnation of Japan’s war in China shut out that source of finance. With the intensification of the war in China, the government introduced three new laws in 1938. One measure legalized enforcement of the old 1918 mobilization measures. The second attempted to direct the flow of investment to highpriority military industry; the BOJ screened large investments, whereas financial industry self-regulation groups reviewed smaller amounts. In addition, dividend payments by corporations were restricted. The third law authorized the government ‘to restrict or prohibit the import or export of any commodity and to control the manufacture, distribution, transfer, and consumption of all imported raw materials’ (Johnson 1982: 136). This law granted MCI the discretion to control almost everything. Although revised in 1949, the statute formed the legal basis for post-war industrial policy. Together with the foreign exchange controls and the financial system controls, this system survived into the post-war era, when it served as the legal infrastructure of economic policy (Okazaki and Okuno-Fujiwara 1999: 21).
62 Planning supplies for war Following munitions shortages in 1939 and 1940, army officers and bureaucrats who had been sent to organize industry in Manchuria were brought home to lead a renewed industrial expansion. As former Manchurian hands took over influential offices in Tokyo, they brought with them the best control-oriented civil servants they had identified in their former positions; these bureaucrats, who had gained actual management experience in industrial development, influenced the policy orientation of a generation of government managers. As military production grew to record levels, coordination of industry began to break down because of confused authority and bureaucratic rivalries, but especially because of the almost implacable competition and impenetrable boundaries between the Imperial Japanese Army and Navy. The differences between the services derived from distinct clan origins of officers, from competition for budgets and real resources, and from different styles of professionalism (the army emphasizing fighting spirit and the navy technical competence). The army considered Russia and China with their great land armies on the landmass of Asia as its traditional foes, whereas the navy looked at the Pacific fleets of the United States and Great Britain. The Japanese armed forces were no better integrated between 1941 and 1945 than they had been in the 1930s, despite the increased severity of their situation: ‘They visualized different enemies and fought different wars’ (Coox 1988: 23). Succeeding governments tended to temporize rather than coordinate the policies of the armed services. A reluctance to tamper with the concept of an autonomous service responsible only to the emperor resulted in not having a permanent mechanism to integrate military affairs. Institutional barriers extended to the services’ suppliers; separate facilities and even separate companies often were selected as contractors for the army and navy for the same product. The government attempted to bring some order to industry and procurement in 1943. It consolidated government procurement authority in a single Ministry of Munitions – the antecedent of the post-war Ministry of International Trade and Industry (MITI) – and tried to enforce corporate consolidations: ‘Neither measure worked as intended, for neither service would accept civilian control.’ (Samuels 1994: 105). Johnson argues that integrated state control over industry was achieved only during the American Occupation in 1945, when the hold of the military and the opposition of large business were broken only by the authority of a victorious army (Johnson 1988: 154–6). However, as described in the next chapter, this integration was often an illusion held by the planning bureaucrats.
Restructuring the economic system The imposition of economic controls had several side-effects that intensified debate on the nature of the Japanese economy. For example, profits suffered when the prices of imported goods jumped upwards at the same time that price controls limited what Japanese companies could charge for their products. In the wake of falling profits, the incentive to produce also declined. The consequences for the
Planning supplies for war 63 military were severe. In 1938, before the imposition of general price controls, more than half of 17 industrial categories exceeded the planned production quantities; after price controls came into effect in 1939, only a single class of goods exceeded its target that year. The situation did not improve much the next year, when only two products out-performed the plan (Nakamura 1988: 489). Policy debate centred on solving the shortfall problem. One view was to allow prices to rise in order to make higher rates of production more profitable; those, mainly in the Cabinet Planning Board, who questioned whether profits should continue to be the basis for business organization opposed this course of action. To meet its objectives in a way that was consistent with its values, the Cabinet Planning Board in 1940 and 1941 devised a three-level structure called the ‘new economic system’. The lowest level of this system comprised individual firms whose objective was to achieve quantitative planning targets rather than to maximize profits. Industrial associations in the middle were to be agents of the government; they collected information from the firms on their capabilities, assigned priorities, established targets and allocated resources to firms. The government, at the top level, established the basic plans and goals (Okazaki and Okuno-Fujiwara 1999: 24). In a process that resembled the evolving planning system in the Soviet Union, the industry control associations coordinated plans with the Cabinet Planning Board in an iterative process and linked the plans to the capabilities of individual firms and factories. At about the same time that the new economic system was being formulated, the government published a directive that selected the best smaller firms as designated sub-contractors to specified major producers. The prime contractors were obligated to write long-term contracts with their suppliers and to provide guidance, financing, raw materials and a steady flow of orders. The private sector fought the rigid and exclusive ties envisaged in this system; subsequently, the policy was revised to allow greater scope for voluntary association (Okazaki and Okuno-Fujiwara 1999: 28–30). Production, however, continued to fall short of the targets. In 1942, only two industries exceeded the plan. Internal debate within the government pitted those officials in MCI, who tended to be close to business, against the reform bureaucrats at the Cabinet Planning Board. MCI staff noted that profits still motivated companies; private firms, they said, did not always accept the planning board’s notions of the public interest based on idealized production targets. They argued that profits could be made to serve the public interest by allowing prices to rise, a policy that was implemented in early 1943. Profits jumped that year and output surpassed plans in 7 out of 16 sectors. Re-examination of the corporate system led to the munitions corporation law of 1943, which made major alterations to corporate governance. The law changed the crucial requirement of the commercial code that corporate executives be the agents of the shareholders; instead, executives were responsible for achieving production targets based on directives from industrial associations or government. The selection of directors was to be made independently of the shareholders; instead, management experience in the firm would be the chief criterion for appointment.
64 Planning supplies for war Profits were no longer to belong solely to the shareholders, but should be allocated to workers, shareholders and directors. By March 1945, almost 700 firms were designated as munitions companies (Okazaki 1999: 118–19). In a companion regulation to the munitions law, the finance ministry designated one financial institution to fund each munitions corporation, considering previous business relations in the assignment. The financial institutions were expected to provide funds in a timely and straightforward manner. From July 1944, such lending was covered by guarantees issued by a government bank. Although banks were encouraged to monitor the munitions corporations, the combination of guarantees and profits from higher prices weakened the incentive to monitor and gave the managers considerable executive freedom (Okazaki 1999: 119). The controls and restrictions changed economic behaviour in fundamental ways. One example illustrates the transformation of corporate governance; dividend payments (shown in Figure 5.1), which had averaged 60–80 per cent of profits until 1937, fell to 30 per cent in 1944. The downward trend continued in the post-war years, when the average payout fell below 10 per cent. Nevertheless, despite the transformation of the Japanese economy wrought by the reform bureaucrats and their military colleagues, and despite their design for a new economy, several scholars note that they were not successful in fully integrating and coordinating Japanese industry during the war years. The appearance of comprehensive plans and the laws authorizing control of almost everything could not overcome the conflicts from scattered and competing government agencies and the military services, as well as from the business community, operating under contradictory incentives and pressures. Nakamura notes that Japan’s early military successes, first in China and later against the western allies, exceeded expectations and encouraged the government leaders to overlook the necessity of all-out, coordinated mobilization (Nakamura 1988: 488). In addition, despite the
Ratio dividends to profits (%)
80 75 70 65 60 55 50 45 40 35 30 1923
1926
1929
1932
1935
1938
1941
1944
Figure 5.1 Ratio of dividends to current profits, 1923–44 (%). Source: JSA (1999: Table 15.06).
Planning supplies for war 65 ascendancy of politics over economics, the cabinet was unable to bring sufficient political power to bear on the warring factions within the government, particularly the military services; the result was inconsistent plans and unattainable goals.
Supplying the war economy Areas directly under Japan’s authority in the 1930s, including the home islands, Korea, Taiwan, Manchuria and parts of China bordering Manchuria, made up a yen trading bloc. Certain key raw materials such as coal, iron ore, steel and salt could be supplied from within this area; but oil, bauxite, scrap iron, alloy metals such as cobalt and nickel, crude rubber, raw cotton and wool came from American, British and other foreign sources (Nakamura 1988: 486). Yen bloc supplies did not require foreign exchange, but imports from other areas needed it. Therefore, Japan had to export to earn the currencies needed to purchase war materiel and inputs from outside the yen area. However, recession in the United States and the outbreak of war in Europe in September 1939 conspired to weaken export markets. Japanese planners had forecast shortages of essential materials resulting from the European war, either because of a general breakdown of trade or because of blockades and sanctions directed against Japan. Accordingly, Tokyo mobilized its foreign exchange reserves and gold to stockpile strategic goods. In January 1940, the United States, having abrogated its treaty of commerce and navigation with Japan, was free to impose economic restrictions on trade. With Great Britain struggling for its existence and the Netherlands and France invaded by German forces, the oil, rubber and tin to Japan’s south in the European colonies looked more inviting than ever before. In September 1940, Japanese forces moved into northern French Indochina, provoking the United States to tighten its export restrictions. Negotiations to ease tensions with the United States started in early 1941, but in July, Japanese forces advanced into southern French Indochina; the United States froze Japanese assets and imposed a total embargo on petroleum. Military planners estimated that they held enough oil reserves for up to two years of fighting the United States and Great Britain. Others were less sure of the ability of Japan to wage an extended war. Civilian experts at the Cabinet Planning Board, for example, warned the cabinet that the nation did not have the supplies and materiel to fight the industrial countries (Johnson 1982: 152). The military, though, contended that what might not be possible under a market system could be feasible under the conditions of controls and planning that were being demonstrated in the Soviet Union and were about to be introduced into Japan (Nakamura 1988: 480). Because planning priorities favoured military production, supplies available for civilian consumption declined. Food production in 1941 was 22 per cent below the 1937 level; textiles were 40 per cent lower. Both products continued to slide over the next four years (Nakamura 1988: 489). In early 1942, within months after the outbreak of general war in the Pacific, it looked to many Japanese planners and industrial leaders that they had achieved their goal of resource independence. The Greater East Asia Co-Prosperity Sphere,
66 Planning supplies for war an idea that had been a rallying slogan, now seemed a reality. Japan occupied a vast area running from the Netherlands East Indies in the south, west to Malaya and Burma up to the border with India, and including most of Southeast Asia, French Indochina, major Chinese coastal cities and ports, Inner Mongolia and Manchuria. Exports to non-yen areas were hardly needed because most industrial supplies could be obtained from controlled regions. Firms quickly drew up plans to invest in resource development in the occupied lands. Allied attacks on Japanese shipping, however, demonstrated the fragility of these hopes. The total ship tonnage available for transporting supplies peaked in October 1942. Despite the high priority attached to shipbuilding, losses outpaced new construction (Nakamura 1988: 487). American air and sea attacks, eventually reaching the domestic shipbuilding industry itself, weakened the supply effort. In 1942, the tonnage of new ship construction was less than 30 per cent of losses. Although production increased rapidly the next year, losses doubled; new capacity replaced only some 40 per cent of the destroyed ships. Shipping production more than doubled in 1944 from the preceding year, but destruction of shipping capacity was even greater; new output replaced 80 per cent of the losses – an improvement, but still not enough to make up for the destruction of transportation capacity. The loss of ocean-going transport was part of a worsening problem; with the fall of the Marianas and the subsequent blocking of the sea lanes to Southeast Asia in July 1944, it became impossible to obtain supplies from anywhere but Manchuria, China and the home islands. The centrality of Manchuria to the war effort, especially food for the population, was more critical at this time than it had been earlier. By 1945, food shortages at home made the transportation of grain from the mainland the central priority, but even that was halted by air attacks (Nakamura 1988: 488). The concentration on weapons production pushed up the output of aircraft and ships throughout the war; aircraft output climbed from 6,200 in 1941 to 26,500 in 1944 while the shipyards’ production of warships doubled over those years. At the same time, agriculture and rice output fell by 40 per cent. Textile machinery was scrapped and melted down for munitions. Attempts to economize on shipping led the authorities to establish iron and steel plants in China and Manchuria so that only finished materials rather than bulky ores needed to be carried to Japan; in addition, production of final products was initiated near the new steel plants. However, these steel mills did not, in general, live up to expectations and the authorities at home resorted to finding scrap metal in streetcar rails and other lower priority uses. Food shortages for the urban population became severe in 1943. By early 1944, the daily caloric intake was an estimated 1,400 calories, down from 2,200 calories in 1941 (Nakamura 1988: 491). Despite rationing and price controls, the official consumer price index more than doubled from 1937 to the beginning of 1945; including black market prices, the estimated price level climbed by almost 400 per cent. Little of what occurred previously, however, prepared the government or the people for the destruction brought by the allied firebombing of more than 60 Japa-
Planning supplies for war 67 nese cities, and then the atomic bombing of Hiroshima and Nagasaki. Estimates made after the war concluded that one-quarter to one-third of the nation’s wealth had been destroyed, including 80 per cent of all ships, one-third of all industrial machine tools and 25 per cent of railroad rolling stock and motor vehicles (Dower 1999: 45). The human loss was as staggering. Out of a total population of 74 million, at least 2.7 million military personnel and civilians died as a direct result of the war. Millions more were injured, sick or malnourished. Of the returning military personnel, some 4.5 million were identified as being wounded or ill. In Tokyo, 65 per cent of the residences were destroyed; Osaka and Nagoya suffered equally as badly, with 57 per cent and 89 per cent of their housing uninhabitable (Dower 1999: 45–6). With Japan’s surrender to the Allied powers in August 1945, the widespread destruction and chaos confronted the occupying army with an economic challenge that far surpassed the mobilization problems faced by the wartime planners. However, it was this same group of people who would be given the job of peacetime reconstruction.
6 The American Occupation and the post-war economy
The American Occupation of Japan from 1945 to 1952 either altered the life of the country profoundly or accomplished almost nothing. Assessments vary by writer and subject (Passin 1990: 119). For those who claim that the American Occupation had little lasting effect, it would be obstinate to ignore seven and a half years of massive penetration and infusion of foreign influence and foreign ideas. The occupying power introduced an impressive list of changes, including a shift in the locus of sovereignty from the person of the emperor to the people, the emperor’s renunciation of divinity, extension of suffrage to women, revision of the legal codes, renunciation of war in the constitution and a reformed education system. In economic affairs, the assertion of little effect has more credibility, but must be balanced by noting significant changes wrought by the occupiers. The Americans arrived in Tokyo in September 1945 following Japan’s surrender with several economic objectives. At the top of the list was the dissolution of Japan’s military capabilities, including its war industry. Other priority items were eliminating the zaibatsu, reducing economic concentration and encouraging labour unions. Occupation staff in Tokyo implemented extensive land reform, an item that had been on and off the list during pre-surrender planning. However, a major and lasting effect of the Occupation on the Japanese economy was largely inadvertent. A pre-surrender decision by the wartime allies left the existing Japanese government in place to implement the policies of the conquering nations, predominantly the United States. In particular, American economic directives to continue wartime wage and price controls and to create an economic planning agency to direct economic reconstruction reinforced Japanese bureaucrats’ own inclination to arrange economic affairs as well as their disinclination to trust markets. The occupying powers left in place many of the economic institutions and structures erected during the country’s march to war. Most important for the course of future economic developments were the preservation of bank-centred finance and the legal structure of corporate governance, which will be discussed in the next chapter. In combination, these institutions dethroned both shareholders and profitability from their premier positions influencing corporate behaviour.
The American Occupation and the post-war economy 69
War’s residue Japan’s heritage of war was economic chaos and personal misery. At least 2.7 million military personnel and civilians died as a result of the war out of a population of 74 million. According to estimates cited by historian John Dower, millions more were injured or fell sick. One-quarter of the nation’s wealth was destroyed, including 80 per cent of shipping, one-third of industry’s machine tools and almost one-quarter of railroad rolling stock and motor vehicles (Dower 1999: 45). In addition to the atomic bombing of Hiroshima and Nagasaki, more than 60 cities had been heavily bombed, rendering 30 per cent of the population homeless. Tokyo lost two-thirds of its housing, Nagoya almost 90 per cent. At the end of the war, some 6.5 million Japanese were stranded across the former empire and war zones. By the end of 1946, 5 million of them had returned to the home islands from former military and colonial outposts to face a bewildering landscape of destruction, homelessness, hunger and unemployment. Another million trickled home over the following year; others died, or just disappeared. Allied attacks on Japan’s sea and rail transportation system had strangled the nation of most foreign supplies. The end of hostilities brought the complete cessation of imports until an operating government could devise the regulations and procedures permitting needed trade (Cohen 1949: 418). Japan’s economic policies during the war had focused on maximizing current military production. Maintenance was delayed or forgotten; producers of military products converted and cannibalized consumer goods factories to support higher priority output, and scrapped civilian production facilities. When peace returned, the years of neglect retarded the resumption of civilian production. Adding to the undersupply, forced labour from Korea and China returned home or simply wandered off, leaving coalmines, in particular, short-handed.
Planning for the post-war Occupation American government planning for a post-war Japan had started even before the United States joined the fighting. Following the outbreak of war in Europe in 1939 and the breakdown of relations with Japan, the State Department, with support from President Franklin Roosevelt, created several secret advisory groups that sought the views of outside scholars on post-war policy issues such as armaments reductions, economics and security. State Department officials had two goals, one strategic and the other bureaucratic: to avoid the mistakes after the First World War and to keep control of post-war policy within the department. In February 1941, it set up its first full-time unit, which took on a more comprehensive role in December after the United States entered the war (Mayo 1984: 6–7). Planning for Japan roughly paralleled that for Germany. Staffed at first mainly by a small cadre of Japan specialists, ‘old Japan hands’, the initial focus was on military and political objectives. By early 1942, they arrived at fundamental principles that they believed should govern a post-war settlement: reduction of the Japanese military establishment, dismemberment of its empire, reconstruction
70 The American Occupation and the post-war economy of the economy and creation of a regional security system. Several participants spoke of the desirability of a strong economy to support a post-war East Asia. They considered a viable Japanese economy to be a prerequisite for lasting peace in the Pacific (Mayo 1984: 15, 21). By late 1943, many planners noted the limited number of trained Americans who would be available to administer a defeated Japan. If the United States aimed for many changes in the country, it seemed necessary to rely on a domestic Japanese mechanism to assume the basic responsibility, led by the assumed existence of moderate and liberal personnel. To encourage the ascent of such people, and to assist Japan’s political parties and government bureaucrats to establish control, planning staff recommended formal revisions of the constitution to destroy the privileges of the military and to strengthen the Diet (Mayo 1984: 18–20). The Japan hands preferred maintaining the country’s economic capacity for the post-war support of Asia and the world economy. To that end, they debated such policies as zaibatsu dissolution, on the one hand seeing these business groups as key elements in Japan’s economic growth and industrial structure, but on the other identifying the concentration of wealth as counter to democratization goals. On the whole, though, the Japan hands were not particularly disturbed by the degree of monopoly capitalism in Japan that attracted the attention of others (Mayo 1984: 33). In early 1944, the planning mechanism was reorganized and augmented with economists, political scientists and lawyers. A serious split emerged between the Japan hands and the economists. As a result, the final policy statements and military directives owed much to the thinking of the original Japan hands, but the economic and financial sections were shaped by others at the State Department and elsewhere in the government (Mayo 1984: 29). The issue of workers’ organizations actively entered the discussion in mid1944. By April 1945, the approach to labour unions had changed from permission to encouragement. Land reform was another issue that arose at this time, but only in a vague way. One official, though, focused his attention on the subject. Robert Fearey, private secretary to Joseph Grew, US ambassador to Japan in the period leading up to the war, returned to Washington to work on post-surrender plans at the State Department. He took a personal interest in agrarian reform and studied it with Department of Agriculture specialist Wolf Ladejinsky. Whereas Fearey and Ladejinsky advocated a major land reform, other Japan specialists opposed the idea because of its possible negative effect on productivity and food supplies. The issue remained quiescent until after the war. Professional economists brought into the planning favoured strong anti-trust laws and the break-up of the zaibatsu and other large companies, based on the economic rationale of promoting competition and efficiency (Mayo 1980: 205–28). Edward Mason, formerly chief economist at the Office of Strategic Services (OSS), moved to a new economics position at the State Department. Mason had pioneered the field of industrial organization at Harvard University and was a leading scholar of monopoly and industrial concentration. He recruited an OSS colleague, Edwin Martin, to help on the Japan planning process. Before joining
The American Occupation and the post-war economy 71 the OSS, Martin had served at the Office of Production Management, Roosevelt’s creation for organizing the nation’s industry for war. Both men favoured breaking up business concentration, a view based on economic argument rather than detailed knowledge of Japan. Martin’s draft plans called for extensive military government controls over Japanese currency, production and distribution. Although not wishing to impair the Japanese economy through extensive reparations, he advised that no steps be taken that would provide a living standard for Japanese out of line with that of neighbouring countries. This last sentiment paraphrased planning documents written for Germany (Mayo 1984: 39). Since planning for a defeated Germany was progressing ahead of the work on Japan, State and War Department officials urged that the German model simply be applied to Japan. The post-war planners for Germany identified the industrial combines and trusts as active collaborators with the German military and government in pursuing the war. Demilitarization of Germany, therefore, required their dissolution. They believed that the zaibatsu must also have been responsible for Japan’s militarism and sought to apply the German plans to Japan, just switching names on the documents, as one official suggested (Mayo 1980: 215). Secretary of the Treasury, Henry Morgenthau, who convinced President Roosevelt that Germany should be punished severely by reducing its economy to pre-industrial levels, reinforced the element of revenge that ran as a subtext to these deliberations. Roosevelt and the British prime minister, Winston Churchill, initialled a report issued at the Quebec conference in September 1944 that called for Germany to be converted into a country that would be primarily agricultural and pastoral in character. The next month, the president confided that Morgenthau ‘had pulled a boner’, and dismissed that approach; nevertheless, higher level decision-makers preserved an inclination to apply this kind of harsh treatment to Japan, along with the other policies produced for Germany (McCreedy 2001: 723). There was another source of the idea of eliminating both German and Japanese industrial capabilities. US business interests advising the planners wanted to reduce German and Japanese economic strength in order to minimize American companies’ post-war competition. By 1945, the views of the economists on economic issues dominated those of the Japan hands, who were thought to be overly influenced by their former acquaintances and too soft on Japan. With the defeat of Germany in May 1945, efforts accelerated to complete the final Japan plans, especially those dealing with the economy, which were running late because of internal disagreements. As the end of hostilities with Japan grew nearer, however, planning remained incomplete, except for deconcentration. On 15 August 1945, Japan accepted the Potsdam Declaration, announced barely three weeks earlier by the Allied powers ranged against it – an acceptance that was equivalent to unconditional surrender. From 2 September 1945 until 28 April 1952 the country came under the dominance of military power commanded nominally by the victorious allies; however, the Americans ran the show. The Potsdam Declaration incorporated the terms of the Cairo Declaration of
72 The American Occupation and the post-war economy November 1943, which called for stripping Japan of all territories it had occupied since 1914, as well as Korea, Taiwan and the Pescadores Islands; the February 1945 Yalta Agreement added the Kurile Islands to the list. Thus, at war’s end, Japan embraced just the home islands: Honshu, Hokkaido, Kyushu and Shikoku. The mainly uninhabited Ogasawara (Bonin) island group returned to Japanese control in 1968; Okinawa and the Ryukyus returned to Japanese sovereignty in 1972. Since Japan had surrendered with its government intact, the emperor cooperative and a new cabinet taking shape, the planners made provision for an indirect Occupation, with the American military taking an overall supervisory role. With this decision finalized, the draft plans had to be rewritten to reflect this fundamental decision (Mayo 1984: 45). The economic plans were thoroughly revised. The redrafted version called for economic demilitarization and wide distribution of wealth and corporate ownership, the last item to include a programme for the dissolution of the large industrial and banking combinations. The supreme command to be set up in Tokyo was directed to enact an extensive purge of political, educational, police and economic personnel. The passage on dispersion of ownership was closely modelled on the directive for the American zone in Germany. A phrase in earlier drafts that mentioned rural land tenure was dropped, but the Tokyo-headquartered Supreme Commander for the Allied Powers (SCAP) was authorized to remove all legal obstacles to labour activity, preventing strikes only if they threatened military security (Mayo 1984: 47–9).
The Occupation’s initial approach to economic policy The document outlining the initial post-surrender policy for the supreme commander, issued by the State–War–Navy Coordinating Committee, was formally dated 6 September 1945, although drafts had been disseminated for comment and review since June (SWNCC 1945). General Douglas MacArthur had a copy of this document on his flight to Japan for the surrender ceremonies. A second document, produced by the Joint Chiefs of Staff, was a military directive giving the commander detailed guidance on policy objectives (JCS 1945). Both policy statements included the same basic approach to economic affairs and governed SCAP’s early policies. American policy defined the chief objective as demilitarization of Japan, which included eliminating industry’s war potential. The initial guidance was clear: no responsibility for economic affairs. Washington directed that the task of economic reconstruction be left in the hands of the Japanese people and their government. SCAP would not assume any responsibility for the economic rehabilitation of Japan or the strengthening of the Japanese economy. You [SCAP] will make it clear to the Japanese people that you assume no obligations to maintain, or have maintained, any particular standard of living in Japan. (JCS 1945)
The American Occupation and the post-war economy 73 ‘The plight of Japan is the direct outcome of its own behaviour and the Allies will not undertake the burden of repairing the damage. . . . The Supreme Commander will exercise his authority through Japanese governmental machinery and agencies’ (SWNCC 1945). International trade matters were the sole economic area reserved for the Occupation authorities. The thrust of American policy in the months following victory was punitive. A reparations mission headed by oil industry executive Edwin Pauley proposed eliminating Japan’s war potential by destroying heavy industry and transferring machinery to Japan’s wartime victims. To that end, the mission called for reparations of 600,000 machine tools, leaving 175,000 in Japanese factories. The underlying aim of the commission was to make Japan a pre-industrial economy. It should concentrate on silk, tea and lumber, and focus its exports on handicrafts, porcelain and toys (Cohen 1949: 419–26). Pauley previously headed a reparations mission to Germany, which had made similarly harsh recommendations. The consequences of the Pauley approach would have so retarded Japan’s recovery that General MacArthur stalled implementation. Two succeeding commissions considerably reduced the scope of the proposed reparations in a deliberate attempt to moderate the negative impact on industry. In the end, some 19,000 machine tools were sent to other countries. American government bureaucrats assigned to SCAP were suspicious of Japan’s big business conglomerates as a general proposition, but especially because of corporate links to conservative politicians. They personally favoured the idea of breaking up the zaibatsu and other concentrations of business power (Schonberger 1989: 90–110). Many of those advocating zaibatsu break-up and strong anti-monopoly policies were labelled as ‘liberals’ or ‘New Dealers’ because of their presumed antibusiness inclinations. SCAP officials responsible for putting these policies into effect received the same appellation, appropriate in many cases because they often were alumni of New Deal and other Washington regulatory and war-planning agencies. However, before putting too much stress on the so-called New Deal orientation of Occupation personnel in the first post-war years, it should be noted that assorted official and personal objectives occupied the minds of American administrators. There was no single ‘Occupation line’. Preferences and policies varied across individuals and agencies, and over time. As described by one historian, the SCAP supreme commander, General Douglas MacArthur, believed in promoting Christian ideals and American imperialism; Charles Kades, deputy chief of the government section and chair of the committee to write a constitution, spoke of his commitment to a social-democratic model for Japan; the Canadian diplomat– historian and MacArthur advisor E.H. Norman was a Communist party member (Moore 1979: 724); intelligence chief General Charles Willoughby professed a staunchly anti-Communist view and was described by MacArthur as ‘my favorite fascist’. These personal and policy differences were intensified by interdepartmental rivalries, which led, for example, to the deliberate spread of stories that the government section was a haven for ‘New Dealers, pinkos, and other radicals’ (Baerwald 2003).
74 The American Occupation and the post-war economy Among those Americans assigned to Tokyo, few had the professional competence and knowledge about Japan needed for the technical tasks assigned to them. Furthermore, their appointments often were brief. Their on-the-job training ended before they could gain competence. Both the American occupiers and their Japanese counterparts knew that the Occupation was temporary, which encouraged the Japanese to out-wait the Americans, while the Americans saw little sense in imposing changes that would be abolished as soon as they departed. Moreover, just as there was no single voice expressing the American view, the Japanese side was divided; the chief gap was between the government bureaucrats and the conservative politicians. The bureaucrats believed in the efficacy of government planning and control whereas politicians like Shigeru Yoshida distrusted these methods during the war and fought their extension afterward.
The Americans look to state economic planning For various reasons discussed in the previous chapter, Japan’s wartime military and civilian planners distrusted markets and capitalist motives. They preferred state control of the economy to achieve national goals together with state planning and coordination. Although SCAP implemented purges to remove high-level officials in government and industry, they did not go very deep; for example, the Ministry of Finance lost only nine relatively junior bureaucrats (Calder 1993: 44). As Dower put it, ‘When the Americans departed, the native mandarins carried on, stronger than they had been even during the war’ (Dower 1999: 26–7). On the last day of the war, a group of young bureaucrats and seasoned academics organized by a young foreign ministry bureaucrat, Saburo Okita (foreign minister in 1979), assembled to formulate an approach to economic recovery. The members, especially those from the universities, consisted of many individuals who had been critical of the military government, and who, in turn, had been marginalized because of their socialist leanings. The group produced a report in March 1946 that recommended state economic planning at the national level as crucial to Japan’s survival. They looked to a future in which the state would control basic industry and banking, but they were wary of two dangers: the capture of a technocratic government elite by an autocratic state; and the likelihood that private firms would make decisions affecting the nation on the basis of private, inherently selfish, profit motives (Hein 1990: 114). In a revealing economic and logical non sequitur, they believed that the US ideal of laissez faire economics had to be rejected as an inappropriate model for Japan because of the country’s dependence on foreign trade (Hein 1990: 116). Similar views surfaced on the American side as soon as the Occupation settled down to work. Because of the destruction of the physical infrastructure and the disruption of customary relationships, economic affairs were chaotic. The situation seemed to call for some kind of centralized coordination. Many of the wartime organs, though, had been disbanded. The Cabinet Planning Board, established in 1937 as warfare expanded in China, had brought together personnel from across government agencies and military
The American Occupation and the post-war economy 75 services and was responsible for mobilization, industrial planning and macro economic policies; this wartime agency was disbanded in 1945. Also, as part of its policy of economic democratization, SCAP abolished many of the industry control associations, which had the nominal task of allocating scarce resources among producers, but actually performed this role rather ineffectively. American administrators viewed the control associations as holdovers from the war and mechanisms that fostered cartelization. However, scrapping them in the face of almost total supply disruptions and the breakdown of normal market mechanisms left the economy without either self-guidance or centralized oversight. Wartime industrial coordination was not a strange idea to American bureaucrats. During both the First World War and the Second World War, the US government created agencies to manage military production, allocate materials and control prices and wages. The lesson learned from this experience by many Washington agency directors and bureaucrats, as well as by political thinkers more generally, was that the government had the obligation and capacity to coordinate vast enterprises in times of crisis. A broader lesson, drawn from the conspicuous breakdown of the economic order in the Great Depression, was that the government provided an alternative to markets, which many considered to be undependable guarantors of full employment and equitable income distribution. These views dominated economic policy in the first Roosevelt administration when corporatism – a partnership between business and government – provided the intellectual source of several key programmes; these same views continued to be held by many politicians and bureaucrats over the succeeding Roosevelt terms, even though the policies that they generated fell out of favour (Schlesinger 1959: 87–98). In early 1946, many (but not all) SCAP officials were convinced that a powerful agency, like the US War Production Board in the recent war, was needed to coordinate economic activities in Japan. Officials in the price control division of SCAP, for example, thought that the Japanese government was not organized for effective coordination because each ministry ran its own domain. Because the situation was so serious, American officials thought that something had to be done, but the Americans were uncertain and divided on specific measures, especially since SCAP’s directives had said that the Americans were not to assume responsibility for economic affairs. Amidst these doubts, the 1945 rice harvest was the lowest in over 20 years, raising the fear of mass starvation. According to the unpublished memoirs of a member of the economics section, W.S. Egekvist, the situation was becoming so serious that something had to be done. ‘I felt SCAP could help the Japanese government improve its ability to handle its own mounting economic problems by promoting the establishment of a civilian Economic Stabilization Board, similar to the one established in the US during the war.’ (Ishikawa 1995). With approval from the chief of the economics section, preparations for a planning and coordination organization went forward and were presented to the Japanese government in the spring of 1946. The government approved the Economic Stabilization Board (ESB) in May; the Reconstruction Finance Bank (RFB) followed in January 1947 as the ESB’s
76 The American Occupation and the post-war economy financial arm. When the American authorities directed the Japanese government to establish the ESB, ‘SCAP expected it to be used as an instrument for central planning and control’ (Hollerman 1979: 708). Prime Minister Yoshida’s first choice to head the ESB was Tokyo University economist Hiromi Arisawa, a leading member of the foreign ministry’s post-war study group. Yoshida and Arisawa, along with several other Marxist economists, had become friendly through family and personal connections (Gao 1994: 134). Combining a Marxist perspective on planning with German notions of organizing for total war, Arisawa’s work had formed a leading intellectual support for mobilization planning. Unlike many of his Japanese academic colleagues who were wed to Marxist doctrine, Arisawa was described as a flexible, pragmatic scholar who was as much at home with western, modern economics as with Marxist theory (Gao 1994: 117, 123). The government had jailed Arisawa in 1938 for his Marxist views, but he was acquitted after 18 months in jail. He lost his university position, but a few months after his release from jail, the army asked him to participate in a study on whether Japan could sustain a two-front war. Using the recently published input–output mathematical technique pioneered by the Soviet economist Wassily Leontieff, Arisawa concluded that Japan could not win a long war against the Atlantic powers because the 10-times difference in economic capacity would swell to a 20-times difference after a single year; he advised against war. The army minister rejected his position paper on the grounds that, although the economic data and reasoning were impeccable, the conclusion ‘was incompatible with national policy’, and he ordered all copies of the report burned (Gao 1994: 132). Arisawa turned down the ESB offer but agreed to act as an economic advisor to Yoshida. This match seemed a strange one to many observers because Yoshida disliked the wartime controlled economy and the planning bureaucrats who had cooperated with the military (Johnson 1982: 177). One explanation for the relationship, in addition to their previous personal connections, was that Yoshida co-opted these planning technocrats, making use of their expertise as he saw fit and keeping them under surveillance rather than allowing them to work for the benefit of others that Yoshida distrusted. A feature of the pre-war Japanese government that had attracted Arisawa and other economists to the idea of coordinated economic planning – an ‘economic general staff’ was a frequently used term to describe their proposals – was the fragmented nature of the state imparted by the Meiji constitution. Under the Occupation, they thought, SCAP’s authority would overcome that fragmentation, convincing the planners that, finally, coordination would be possible. However, the private business sector, political parties, and the Diet did not always grant the government or SCAP automatic approval of their proposed schemes. The ESB mandate was broad. As the central economic control agency and presumed heir to the Cabinet Planning Board (including inheriting many staff members of the wartime agency), it took over the allocation and control association functions. At first, it supervised price controls, transportation, finance and public works budgets. As the process of eliminating control associations proceeded, the
The American Occupation and the post-war economy 77 ESB established public corporations to take over their functions; by mid-1947, it had assumed consolidated authority over all economic planning with a staff of 2,000. However, conservative politicians in the Japanese cabinet, in contrast to Japanese government officials, ‘failed to share SCAP’s enthusiasm for controls’, in the words of an Occupation economist (Hollerman 1979: 708). A former US Navy intelligence officer, economist Jerome Cohen, concluded: ‘Thus the wartime control system with its vestiges of cartel domination was abolished, but the substitute allocation system had a very similar appearance’ (Cohen 1949: 431). One former SCAP economist, writing as an academic in an American economics journal, described the ESB line as closer to the British Labour Party than to the American New Deal. Its approach to economic policy was ‘close to being the official position of all wings of Japan’s divided Socialist party’ (Bronfenbrenner 1950: 286).
The coal crisis as an example The case of coal was illustrative of the American Occupation’s and Japanese government’s use of their powers to influence economic events. Coal and steel were industry’s essential inputs. The collapse of coal production in mid-1945 confronted SCAP with an obstacle to economic recovery. Imports of high-quality anthracite and coking coal from Manchuria, Sakhalin and China had virtually disappeared, which curtailed steel production. Despite the repeatedly stated policy that economic welfare was not a SCAP objective, one of the Occupation’s first directives was a 15 September 1945 order to maximize coal output, followed by cabinet adoption of emergency measures, including directives to raise wages and commodity rations for miners and allocate transportation to haul coal. These measures, however, were not fully implemented. SCAP followed in mid-December with direct orders to speed up production and scolded the Japanese government for a response described as ‘inadequate and unsatisfactory’ (Hein 1990: 67–8). However, years of mining infrastructure neglect, declining productivity of obsolete equipment and the wartime failure to develop new reserves hindered the revival of production (Cohen 1949: 166–7). These problems were exacerbated by the departure of more than 200,000 miners from the coalfields – including more than a hundred thousand Koreans and numerous Chinese forced labourers as well as women, children and the elderly who had been conscripted into the mines. Figure 6.1 shows monthly coal output. In order to rebuild the mining facilities, steel was required. However, coal was needed to produce steel, and both coal and steel were needed to run the railroads that would deliver coal to the blast furnaces and steel to the mines. The dilemma was a familiar one to wartime planners; it was a situation that seemed to cry out for the kind of coordination familiar to them. Arisawa organized discussions among several economists in the summer of 1946, seeking a way out of the dilemma caused by the mutual needs of coal, steel, transportation and industrial production. By November, these discussions
78 The American Occupation and the post-war economy had attained enough coherence that, amidst predictions of imminent economic collapse, the prime minister established a personal advisory group called the Coal Committee, chaired by Arisawa. The group put forward the idea of priority production, focusing first on coal, with coal producers assigned top priority for government loans and subsidies. The ESB made priority production its central objective (Johnson 1982: 181–2). Arisawa summoned mining directors and chief engineers to provide information. Based on the technical characteristics of production, he calculated the supply capacity. On the demand side, he estimated the possible coal use by SCAP, power companies, railroads and industry. Such technical planning methods were abetted by a promotional campaign. The daily output of coal was posted in the streets of large cities. The evening radio programmes sent words of thanks to hard-working coal miners all over Japan. To deal with the labour shortage, emergency measures authorized the priority distribution of clothing, food and shelter to coal workers in those mines promising to meet production goals (Miwa and Ramseyer 2005: 20). However, the supplies for workers and other allocated materials were not linked to actual output, so had little effect on incentives. In 1947, the ESB added fertilizers to the list of priority sectors and later that year it included electricity after the generating companies vigorously lobbied Diet members for priority status. A contemporary estimate indicated that the coal industry received 80 per cent of its material requests whereas the other sectors obtained only 30 per cent (Hein 1990: 120). By transferring miners from metal mines to coal, by providing food and other special incentives and by recruiting out-of-work labourers, the number of miners reached wartime levels by the end of 1947, but the workforce was inexperienced. The wartime coal industry association, disbanded in May 1946, was almost immediately reconstituted in a somewhat modified form along ‘more democratic lines’ designed to be acceptable to Occupation officials. In April 1947, a public corporation was formed to monopolize the purchase, sale and distribution of coal at fixed prices. Since coal was sold at a lower price than it cost to produce, the difference was made up by subsidies. The mines received payment on a cost-plus basis, which created an incentive to ‘embellish’ their costs, especially since the authorized payments averaged about 82 per cent of the actual cost of production. However, the subsidies payments were delayed. To make up for the delayed payments, the government loaned mines the missing amounts until the subsidies arrived. In its first year of operation, one-third of RFB financial assistance was for the coal industry. By 1948, the RFB accounted for 98 per cent of mine financing (Samuels 1987: 92–4). Subsidies to the coal industry absorbed 25–30 per cent of the government’s general budget in 1947 and 1948 (Hein 1990: 122). With the rampant inflation, however, the subsidies continually lagged behind the difference between the controlled selling price and production cost; since payments were late and costs were rising, delay was equivalent to discounted payment (Miwa and Ramseyer 2005: 20).
The American Occupation and the post-war economy 79 The subsidies were computed on the basis of average mining costs, which favoured the most productive mines. Moreover, the mandated price differences between different grades of coal were small; since the more productive mines were prohibited from selling their best coal at a market price, they preferred to sell poorer grades from less efficient seams, saving their better reserves for a later time when price controls might disappear. The coal producers openly acknowledged that they had no incentive to rationalize production or to lower their costs. The pricing scheme had the greatest incentive effect on many marginal producers, who added little to the total supply (Hein 1990: 122–3). General MacArthur continued to prod the government to take more energetic steps to increase coal output. A Socialist-led cabinet that assumed power in May 1947 had as its central plank the nationalization of the coal industry, following the lead of its British counterpart. Despite SCAP’s weak support of the nationalization law, vigorous opposition from the conservative political parties in the Diet as well as from mine owners forced considerable weakening of the proposed law. A watered-down version finally passed the Diet in December 1947 and became effective the following April. Implementation occurred only in 1949, by which time the officials involved with managing the mines had become discouraged. In fact, the law gave the government little control; moreover, the managing agency was required to draw many of its personnel from the mining industry itself, which further weakened state authority over the industry (Samuels 1987: 103). With coal nationalization under consideration, the cabinet focused on the priority production scheme. Despite this attention, coal output stagnated throughout 1947, reaching the target level of 30 million tons only at year-end. Figure 6.1 shows monthly coal output. The Socialist government fell in February 1948, and was followed by a coalition led by Democratic Party leader Hitoshi Ashida, who continued some of the existing control mechanisms. In July 1948, Yoshida took his Liberal Party out of the coalition, forming a new government in October with an explicitly non-interventionist platform.
Million tons, annual rate
60 50 40 30 20 10 0 1945
1946
1947
1948
1949
1950
Figure 6.1 Coal production, 1945–50, million tons, annual rate. Sources: Hein (1990: 66); Miwa and Ramseyer (2005: 47).
80 The American Occupation and the post-war economy Yoshida initiated a decontrol policy immediately and accelerated it when he won a landslide election in January 1949; he ended the priority production system later that year in September (Miwa and Ramseyer 2005: 22). In the two years following the formation of the ESB in May 1946, coal output increased by 10 million tons (annual rate). In the seven months after Yoshida ended the coalition, from August 1948 to March 1949, coal output soared by 12 million tons, or 40 per cent. The schemes to revive the coal industry reminded one American economist of the Munitions Company Act of 1944, which was intended to grant the newly established Munitions Ministry supervisory power over the heretofore-fragmented military industries. It is worthwhile spending a moment on the 1944 venture. So important was the 1944 ministerial reorganization that Prime Minister Hideki Tojo made himself munitions minister and appointed his former Manchurian economic czar, Nobusuke Kishi, vice minister for day-to-day operations. Kishi would himself become prime minister in 1957. An evaluation of that earlier attempt to consolidate authority over critical industries during the war was harsh: ‘In practice, it can be categorically stated that the desired unification was never achieved’. Problems were numerous: the military services continued to operate independently of each other and of government coordinators; industry control associations administered allocations and priorities without reference to higher level directives; orders and supplies were never balanced; production was consistently overestimated; and special priorities were never observed (Cohen 1949: 76). Most of the same problems bedevilled the post-war attempt to bring coordinated management to coal supplies, even with the acknowledged power of SCAP. In April 1948, the leading business newspaper, Nihon Keizai Shimbun, gave its analysis of the situation: ‘The Emergency State Coal Control Law is a mere halfway measure, which, however, does not excuse the incompetence of its administration’ (Cohen: 1949: 474). Perhaps the most important impediment to raising coal output was the price controls imposed by the priority production system. Prices set below costs exacerbated the mismatch between supply and demand. A black market developed to allocate the difference between production and demand by establishing a price that brought some balance to the market. Cement producers acquired 10–20 per cent of their coal on the black market, pulp and paper firms 11 per cent, and bicycle manufacturers 45 per cent. Approximately 10 per cent of coal shipments disappeared en route to approved customers into the black market or other unauthorized uses (Miwa and Ramseyer 2005: 22). Samuels notes that the Occupation’s authority enhanced state controls over the economy in comparison with wartime confusion. In designated industrial sectors, he writes, the ability of the government to intervene in the economy reached a high point as SCAP’s military government backed the preferences of the native bureaucrats. Nevertheless, he advises that it would be incorrect to attribute coal industry recovery solely – or at all – to government efforts: Indeed, success, at what were enormously high real and nominal costs, was
The American Occupation and the post-war economy 81 achieved in spite of government programmes. To be more precise, if we are to credit the government programmes with smashing the bottlenecks in coal and steel production, we should share that credit with the private firms that reinterpreted, thwarted, and transformed government programmes every step of the way. Samuels (1987: 94)
Business deconcentration As directed by Washington, SCAP emphasized economic democratization. The Holding Company Liquidation Commission (HCLC), established to accomplish the deconcentration objectives, issued the first dissolution order in September 1946; it included the core holding companies and main subsidiaries of the big five zaibatsu (Mitsui, Mitsubishi, Sumitomo, Yasuda and Fuji). Over the next several years, 83 companies were dissolved or reorganized (Bisson 1954: 108–14, 256–61). The first group clearly fell within the definition of a major zaibatsu. However, the rationales and criteria for break-up lost clarity as the goal of strengthening competition mixed with the original objective of eliminating the financial power of the zaibatsu families and holding companies. The next policy phase focused explicitly on concentrated industrial power. Under pressure from SCAP, the Diet passed a law in December 1947 that empowered the HCLC to identify and eliminate ‘excessive concentrations of economic power’. The HCLC quickly designated 325 companies for break-up in February 1948. However, as the commission began work on its new mandate, vigorous discussion and criticism of the policy erupted in the United States in business magazines, Congress and the War Department. In response to criticism and backed by further detailed reviews, the HCLC removed scores of companies from the list. By the end of July, only 100 companies remained scheduled for dissolution, with banks excluded from further consideration. A few months later, a review board appointed by Washington issued four principles for implementing the deconcentration law, among which was that no order should be issued unless a case could be made that a company restricted competition or impaired entry into a business. This criterion effectively removed many of the remaining companies designated for reorganization (Bisson 1954: 142–5). In the end, 18 corporations reorganized under the deconcentration law, although many others had voluntarily done so in anticipation of HCLC orders (Bisson 1954: 148–9, 291–2). Although corporate reorganization under the auspices of the American authorities would continue for another two years, 1948 marked the end of the active phase of deconcentration. As part of the so-called reverse course described below, pressures from Washington and elsewhere dictated that Japan’s economic revival was the new policy objective and that deconcentration interfered with this goal. Several scholars have tried to determine if corporate break-up and reorganization had a lasting and significant effect on the structure of Japanese industry. One study of the post-war democratization policy concluded that, by the 1960s, ‘In
82 The American Occupation and the post-war economy prewar terms, zaibatsu no longer exist in Japan. . . . The absolute control of large firms once enjoyed by former zaibatsu families cannot be found.’ The Iwasaki family owned the Mitsubishi Bank before 1945; by 1948, approximately 30,000 shareholders were the new owners (Yamamura 1967: 33, 126). However, soon after the Occupation ended, many of the zaibatsu-affiliated firms recombined into looser confederations called keiretsu, or families of firms. Other data indicate that the deconcentration programme did not radically transform the structure of Japanese industry. For one thing, contrary to the beliefs of those involved in the pre-surrender planning and post-surrender policy implementation, Japanese industry was not very concentrated to begin with, at least compared with American industry. Although comparability issues cloud interpretation of the figures, a comprehensive assessment concluded that Japanese manufacturing in the 1930s was probably no more concentrated than American (Rotwein 1964: 275–6). In addition, the dissolution of the zaibatsu and the reorganization measures were not the only forces affecting industry; other changes came from the entry of large companies into heavy industry during the war, wartime destruction and the post-war entry of war industry firms into civilian production. The net result of all these influences was a small measurable reduction in overall industry concentration. The Japan Fair Trade Commission published 1937 and 1949 concentration data for 45 industries (reproduced in Hadley 1970: 322–3). The change in concentration ratios at the one-, three- and five-firm levels was small. For example, the correlation of the 1949 values against 1937 for the shares of the largest three firms is 0.87. The average experience was a small reduction in concentration. Another view, though, considers concentration ratios to be uninformative about the oligopolistic nature of Japan’s economy. The zaibatsu and other large companies tended to be conglomerates, with holdings across diverse industries and products, but not necessarily with excessive domination in any single one. ‘With positions varying among themselves in the different markets from strength to weakness, Oligopolist A did not challenge Oligopolist B in markets of A’s strength because it faced B in markets where B was stronger’ (Hadley 1970: 318). Nor did the published data consider cartels and other collusive arrangements. In addition, focusing on the year 1937 does not recognize the fact that considerable numbers of mergers and consolidations occurred under government direction between then and 1945. According to this view, the Occupation’s break-up of the large companies did much to increase competitive pressures, even though concentration ratios may not have changed much (Hadley 1970: 331–2). Even if there remains disagreement over the net impact on industrial structure of the American policy, it is hard to argue with the statement that, ‘owing to the disruptions caused by the war (and probably to a lesser degree, the deconcentration programme of the American Occupation), the positions of the largest leading Japanese firms were weakened’ (Rotwein 1976: 61).
The American Occupation and the post-war economy 83
Labour and land reforms Labour unions Independent labour unions had been abolished in 1940, partly to ward off labour disruptions, but also to suppress political discourse. As early as 1900, police regulations outlawed strikes; although they did occur, strictly speaking, they were illegal until a new law in 1926 authorized police conciliation of labour disputes. The government treated union leaders as socialist political agitators, regularly imprisoning them. Union membership peaked in 1936 at 420,000, just a few per cent of all workers. The Americans sought to alter this situation with the enactment as early as December 1945 of a law that recognized unions as the collective bargaining agent of their members. Two other laws, passed over the next two years, established methods to handle labour disputes and set standards for employment, including such items as workplace safety and minimum wages. The laws prohibited strikes by public safety workers and government bureaucrats. Unionization of the workforce quickly increased, rising to more than 6.5 million, or half of all company employees in 1948. That share then declined to around one-third, where it remained for the next 20 years. Amidst rampant inflation and food shortages in 1947, non-Communist national labour associations and public employees unions joined a confederation of unions with strong links to the Communist Party to organize a scheduled general strike. This planned action, combining labour grievances with political statement, confronted the Americans with a test of their New Deal convictions about labour rights. Barely a day before the strike was scheduled to begin, General MacArthur intervened to announce that he would not permit ‘the use of so deadly a social weapon’ (Dower 1999: 269). The more radical unions lost much of their power and influence after this confrontation. Although unions combined into confederations, largely for political purposes, their operational locus was mainly within a firm. This orientation descended directly from wartime practice, when company labour organizations were intended to substitute for independent unions while still promoting patriotism and mobilization. The labour confederations formed the foundation for post-war organization by the political left. The long-term consequence was that Japan’s labour movement did not encompass industry-wide organizations, but acted mainly at the company level. Unions tended to identify with the fortunes of the company, which some have argued made them more flexible than unions in other industrial economies. Land reform Although land reform was not mentioned in the directives that governed initial policy, planning staff in the State and Agriculture Departments had worked on land and agriculture issues for a year or so before surrender. Both Robert Fearey
84 The American Occupation and the post-war economy and Wolf Ladejinsky, who had been working on the issue in Washington, were assigned to the SCAP staff as soon as the war ended. General MacArthur apparently already had a keen interest in the subject (Sackton 1980: 128). Shortly after arriving, Fearey prepared a memo for the general outlining the salient features of the proposed land reform scheme. MacArthur immediately seized the subject and had his staff work out the details (Fearey 1980: 152–2). Since half of all workers were employed in agriculture, problems in that sector would be significant, if only because of their scale. Two main issues confronted the agriculture specialists. Half of all farmers owned no land at all and depended on rented land as tenant farmers; another 20 per cent were both renters and owners. The second issue was that plots were uneconomically small, with two-thirds of all farmers cultivating less than 2.5 acres. Even in years of good yields, high rents kept farm incomes low. The potential for agrarian discontent and political instability had been a concern to political leaders and government officials for decades, but not enough to compel them to deal with this touchy subject. A motive for the Occupation’s attention to land issues was a belief that rural discontent had played a part in directing Japan onto an expansionist path to open foreign markets and provide new lands to colonize and farm. Also, there was a view that owner–tenant relationships helped to preserve feudal tendencies that were inimical to democracy. Another motivation to pay attention to the countryside was a fear that the Communist Party might take advantage of restlessness there (Dore 1958: 183). MacArthur addressed a memorandum to the Japanese government in December 1945 urging attention to a broad agriculture reform effort, including land reform, to bring more equal opportunity to Japan’s farmers. The Agriculture Ministry, which had been pushing reform for many years, convinced politicians on all sides that, if they did not act, the Occupation authorities were likely to take more drastic action than the politicians desired. The Diet passed a bill by the end of December that made some changes in ownership, but the Americans considered the timing to be too slow and the redistribution too partial. While the cabinet was reconsidering SCAP’s negative assessment of its effort, elections were held and Yoshida became prime minister. Mr Yoshida and his agriculture minister accepted the Americans’ land reform philosophy; new legislation proceeded rapidly (Sackton 1980: 130–1). The second land reform law, pushed by the Americans and passed in March 1946, prohibited absentee agricultural land ownership, restricted resident owners to at most 2.5–10 acres (the higher figure was for Hokkaido) and disallowed resident cultivation of more than 7.5–30 acres (again, depending on area). The government purchased land at the controlled price that had prevailed since 1939. Because of the rampant inflation, this policy was equivalent to outright confiscation. The land was then resold at the same price to tenants. By 1950, there were no registered owners of more than the permitted acreage; attempts to bypass the regulations through such devices as multiple ownership by family members were rigidly policed. Organized landlord opposition to the reforms was weak and diminished when
The American Occupation and the post-war economy 85 MacArthur in February 1948 instructed the government to take steps to prevent organized resistance. The goal of distributing land to working farmers was realized; in 1955, tenants worked only one-tenth of agricultural land (Dore 1958: 184). The land reform policy achieved its goal of redistributing land to tenants. It brought a more equal distribution of assets and income. The class structure based on land holding disappeared. Small tenant farmers became small owner cultivators, without any apparent change of farm size. The production structure of traditional agriculture remained. The evidence indicates that redistribution had no effect on productivity. For example, several studies find that productivity increased at almost the same pace among both tenants and owners (Kaneda 1980: 141). The rapid productivity growth that occurred in the post-war period probably resulted from the technological potential accumulated since the 1930s. After the war, labour, fertilizers, seeds and equipment that had been restricted became available. The land reform programme had little to do with these changes.
Reverse course and the Dodge Line By 1947, it was becoming evident that the economy was in trouble. Inflation was one of the most visible signs. Within a half-year of the end of hostilities, prices were increasing at an annual rate above 1,000 per cent, seemingly unconstrained by the imposition of price and wage controls. Surges of renewed bouts of inflation up to the end of 1948 severely disrupted business recovery (Figure 6.2). The first inflationary wave began in the months immediately following surrender, when the Japanese government hastened to fulfil its outstanding obligations by printing money before the occupying forces could intervene. The second surge
1,600 1,400
Inflation rate (%)
1,200 1,000 800 600 400 200 0 1945
1946
1947
1948
1949
1950
1951
Figure 6.2 Inflation rate, wholesale price index, six-month change at annual rate, 1945–51 (%). Source: (JSA 1999: Table 17.1).
86 The American Occupation and the post-war economy came from the establishment of the RFB, which began operating in early 1947. The RFB provided subsidized loans to private companies from funds that it borrowed from the BOJ. Even when these business loans were repaid, the interest rate of a few per cent were so far below the inflation rate that the loans were even better than free money; the RFB essentially was paying companies to take its funds. Most of the RFB loans were issued as part of the ESB’s priority production scheme, and amounted to little more than a transfer from the government, via the BOJ, to the companies. The government also directed private banks to allocate half of their loans to the priority industries and companies designated by the ESB. Finance minister Tanzan Ishibashi’s supply-side approach to inflation was based on the presumption that increased credit would stimulate production more than it pushed up the money supply, so that higher output would slow inflation. This theory was inconsistent with the unfolding evidence. One problem was that government allocation of commodities, plus price and wage controls, restricted market adjustments to the pressures created by subsidized loans and directed lending, thereby exaggerating resource misallocation and stimulating even greater price increases. By mid-1947, inflation had once again surged to more than a 500 per cent rate. Supplies of industrial materials continued to be uncertain despite the coordination mandate given the ESB. Production in 1948 of key products such as iron and steel, textiles and yarn, machine tools, and cement was still less than half 1930–34 levels. The priority items – coal, electricity and fertilizer – had attained their 1930–34 average outputs, but remained 20–40 per cent below peak-year quantities (Cohen 1949: 470). To many observers, the policy of breaking up the zaibatsu and other large companies was destroying the foundation of Japan’s industrial and trading capabilities. American foreign aid, running at $400 million annually by 1947, helped plug the gap between Japan’s imports of food and raw materials and what could be sold abroad, but politicians and others in the United States criticized these expenditures. In addition to weak economic performance and the increasing burden of US aid, the communist takeover of China and the new American policy of containing the Soviet Union added to the pressure to make Japan an economic success and a capitalist leader in the Pacific. In the face of these developments, policymakers in Washington and Tokyo contemplated a turnaround in the initial approach of economic democratization, central government economic planning, control over prices and wages, and materials allocations. A May 1947 declaration by Assistant Secretary of State Dean Acheson that the United States was prepared to undertake the reconstruction of Germany and Japan as the workshops of Europe and Asia presaged this shift. Army Under Secretary William Draper, a former Wall Street banker and wartime general involved in the economic reconstruction of Germany, visited Japan in the summer of 1947 and concluded that, among other things, the reparations policy and zaibatsu breakup seriously disrupted economic recovery. Returning to Washington, he set to
The American Occupation and the post-war economy 87 work to construct a new set of guidelines to make SCAP policy consistent with economic revival, an effort informally called ‘crank-up’ by those working on it (Schonberger 1989: 163). The US government publicly announced this ‘reverse course’ in January 1948 (Schonberger 1989: 175–6). The policy redirection prompted Washington to send several emissaries and study groups to Tokyo to explain the change and to gather facts that would support specific recommendations. The first visit was by George Kennan in early March, sent by Secretary of State George Marshall to convince General MacArthur to accept the decision to soften, if not abandon, economic deconcentration, a policy change that MacArthur had been ignoring. In April 1948, an economic mission comprising business and senior Army representatives departed for Japan to survey the spectrum of economic problems; usually referred to as the Draper mission, Percy Johnston, chairman of Chemical Bank and Trust Company, nominally headed the group. The underlying aim of the mission was to buttress the case for the economic reverse course. Its recommendations focused on balancing the government’s budget, mainly to allow a stable exchange rate. An interdepartmental US government mission headed by Federal Reserve economist Ralph Young in June recommended a fixed, unified foreign exchange rate and noted that price stabilization underwritten by a balanced budget was a prerequisite to that goal. The Young mission put together a nine-point stabilization programme, which became the basis for American government policy. To put a face on this policy initiative and to oversee Japanese government implementation, Washington officials as well as General MacArthur and members of his economics section requested the appointment of a strong, knowledgeable, assertive individual. Draper had raised the name of Detroit banker Joseph Dodge as early as his first meetings with MacArthur in April 1948. Dodge had already performed a similar service in Germany. Returning to Washington, Draper tried to recruit Dodge to lead a direct American intervention in Japanese policymaking. It required President Harry Truman to persuade Dodge to take an appointment as SCAP advisor on economic affairs (Schonberger 1989: 188, 202). Dodge arrived in Tokyo in February 1949 to plan, persuade and cajole the cabinet and Diet into implementing a new course of economic austerity based on the Young report’s nine points, commonly referred to as the nine commandments. The programme that he implemented took on the name of the ‘Dodge Line’, although its general outline preceded his arrival. Dodge’s major objective was to link Japan to the world economy on a competitive basis. He worked under a mandate from Washington to establish a stable exchange rate in three months. To accomplish this goal, Dodge emphasized balancing the government budget through controlling expenditures and expanding revenues; using American aid more effectively, mainly by eliminating RFB liabilities and phasing out RFB lending; establishing a single exchange rate to replace the multiplicity of commodity export and import rates; and decreasing the role of government in the economy. Most of Dodge’s energies focused on balancing the budget (Schonberger 1989: 206). Within two months, Dodge presented to the Japanese government a balanced,
88 The American Occupation and the post-war economy consolidated 1949 budget. This document included the general and special accounts, the railroad, communications and banking companies, and all other activities financed by the government. The new budget was considerably higher in real terms than the spending of the previous year, but represented appreciable cuts from the totals that would have been produced by existing methods; it also included increased taxes to pay for the expenditures (Schonberger 1989: 208). A year earlier the 1948 consolidated deficit had been estimated at ¥160 billion (6 per cent of GDP). In 1949, Dodge turned that into a surplus of ¥260 billion (7.7 per cent of GDP). Members of the ruling Liberal Party feared the political repercussions implied by the huge fiscal shift from deficit to surplus. In particular, it would have required abandoning tax cut promises made in the recent election campaign. However, Prime Minister Yoshida personally had decided to accept the new budget. Yoshida saw Dodge’s policies as consistent with his own desire to end the planning role of the ESB and to lift controls on the economy. Dodge viewed the ESB as one of the greatest obstacles to the success of his programme (Schonberger 1989: 216). Dodge’s status gave Yoshida the strength to resist SCAP’s economics section’s counter-demands to retain controls. By September 1949, the prime minister had eliminated the priority production system and terminated RFB lending. As the Diet continued to demand changes to soften the presumed effects of what they described as an austerity budget, Dodge delivered what some observers labelled the toughest speech given by a high-ranking American official, saying that Japan was living beyond its means and failed to comprehend the real and serious situation of the economy. That same day, the Diet passed the 1949 budget (Schonberger 1989: 215). In the area of foreign exchange, Dodge confronted a fractured currency with a system of multiple rates. American government agencies conducted all trade. A SCAP organization purchased goods from a Japanese government agency that acquired them from producers at a yen price that included costs and profit; SCAP sold the goods on world markets at the going price in dollars. Each transaction generated its own implicit exchange rate. The dollars were used to purchase essential materials in world markets, subsequently resold by the Japanese agency for yen. The yen gap between the cost of export goods and the revenue generated by imports was covered by budget allocations; the dollar deficiency between the revenues from exports and the cost of imports was covered by American aid. Japanese producers were insulated from export competition as costs were subsidized by the trading scheme. Price subsidies had become the largest single item in the Japanese budget by 1949 (Cohen 1949: 497). While the Dodge mission worked on the budget, it tasked SCAP’s economists and Japanese officials to estimate the effects of a unitary exchange rate set at various hypothetical levels, generally within the range of 270 to 400 yen to the dollar. By March 1949, they had settled on a rate of 330; however, they were urged by Washington to adopt a rate of 360 yen per dollar. One thing that may have stimulated the demand for a cheaper yen was that Washington officials were informed
The American Occupation and the post-war economy 89 that Great Britain planned to devalue sterling later that year, which would have left the yen over-valued. Another rationale for a 360 rate was that prices had continued to rise in the period after the 330 figure had been calculated. A third reason offered for the cheaper yen was that it would enable more exporters to sell on world markets, which would help soften the effects of the stabilization policies. On 25 April 1949, as soon as the Diet passed the budget, the new exchange rate of 360 yen per dollar became effective; the yen would stay at that level for the next 22 years. On the issue of reducing the role of government in the economy, the results were mixed. Dodge shut down the RFB and eliminated its lending, but subsidies continued at the former level. Price controls on some items were removed quickly, but general controls were phased out over several years. Trade was restored to private parties – exports freed in principle in December 1949 and imports a month later. However, the December 1949 foreign exchange control law prohibited Japanese from holding foreign currency, except with the permission of the Ministry of International Trade and Industry (MITI), newly created from the Ministry of Commerce and Industry. This law gave MITI influence over imports until the 1960s, when the demand for foreign exchange relative to its availability became less intense. One prominent study concluded that Dodge set the direction for dismantling the regime of economic controls established during the war and Occupation, but accomplished only part of this objective (Nanto 1980: 51). Dodge left Japan after three months, but closely followed affairs from Detroit and fought to keep his programme on track as Japanese politicians and government officials tried to derail it. At the same time, he worked with the US government and Congress to keep aid flowing to Japan during the period of stabilization. The American government sent him back to Japan frequently to review developments and to shore up implementation (Schonberger 1989: 217–18). The impact of the Dodge Line on the economy has been hotly debated, often with disregard of the facts. The economist Dick Nanto has produced one of the better summaries. He notes that 1949 inflation-adjusted budget expenditures were 24 per cent greater than the previous year’s – one of the largest increases in Japanese history. Of course, to balance spending with revenues, taxes had to increase, resulting in a surplus after decades of deficits. A budget surplus at less than full employment imposes a drag on the economy, as this one did. Dodge was said to have had little sympathy for the Japanese taxpayers, who, he believed, were experiencing a higher standard of living than citizens of the European countries on the winning side of the war (Nanto 1980: 50). However, the policies did not drive Japan into a depression. The quarterly index of industrial output declined in only one quarter of the entire period of the Dodge Line. Prices increases were decelerating in early 1949, even before Dodge’s arrival, and continued to slow through the first half of 1950. Severe deflation was the chief worry among most Japanese and quite a few Americans; consumer prices fell 14 per cent from a May 1949 peak to the trough 13 months later when the bout of deflation ended. Industrial production increased 20–30 per cent annually from 1947 to 1950. However, growth before the Dodge policies was based partly
90 The American Occupation and the post-war economy on free or subsidized imports; Dodge argued that any slowdown was from an artificially high and unsustainable level. As the new policies took effect, there were several months of stagnation in the second half of 1949 until growth resumed in 1950. The Korean War, starting in June 1950, brought orders for the American war effort to Japanese producers later that year, but the economy had been expanding well before these orders arrived. Total employment grew more than 4 per cent in 1949; in 1950, as the Dodge Line had its full impact, it fell by 0.9 per cent (JSA 1999: 6–11, 3-03). Although the economy as a whole was improving, the introduction of market prices in domestic and international commerce induced considerable reallocation of demand and output, imposing gross job losses, business bankruptcies, social disruption and forced adaptation. It also meant new opportunities for business and workers. Net job loss was temporary, production continued to expand except for a few months, and business as a whole was growing. Considerable churning occurred as people lost and found jobs. A State Department secret report stated in May 1950: ‘Every major power element in the Japanese body politic considers itself injured and its interests jeopardized’ (Schonberger 1989: 222). SCAP and State Department officials repeatedly warned Dodge of massive social unrest and dangerous political implications of strict adherence to his stabilization plan. Nevertheless, he refused to deviate from his goals of balancing the budget, controlling inflation and establishing a fixed exchange rate, which arguably set the stage for Japan’s growth over the next three decades, without the predicted upheavals.
Implications Ironies abound in the economic story of the American Occupation. Japanese government officials expected the occupiers to dismantle the web of wartime economic controls; in the first weeks after the end of fighting, they had begun this task. The economic bureaucrats were surprised when the first Americans on the scene told them to maintain price and wage controls and, a few months later, to establish a planning agency. Perhaps the most surprised were Japanese businessmen, who welcomed the invaders as like-minded capitalists. They were shocked to be confronted with New Deal philosophy, corporate dissolution and extensions of the wartime controls. Another irony is that the objectives and policies advocated by the Japan hands in pre-surrender planning, which had been overtaken by the concepts of retribution and deconcentration, were adopted after 1948 when the reverse course came to grips with economic failure and the demands of America’s anti-communist, containment policy. However, although the main organs of state planning and intervention lost much of their power and authority, their heritage persisted. Despite evidence that the priority production programme most likely did not achieve its main objectives, especially as it applied to coal, the commonly told story is one of successful government intervention in the economy. The myth of wise bureaucrats acting with their industry counterparts to solve the critical postwar coordination problem became embedded in the popular consciousness and
The American Occupation and the post-war economy 91 was repeated by scholars long after the event. For example, Chalmers Johnson asserted, ‘[t]here can be no question that priority production achieved results’. Based on this belief, he notes, ‘[p]riority production had an important effect on later bureaucratic attitudes as a precedent for bolder rather than more cautious, fiscally responsible courses of action’ (Johnson 1982: 185–6). This view bolstered public officials’ inclinations to guide economic affairs to meet assorted objectives, be they strategic, political or personal. One reason that belief in the success of the priority production programme was so pervasive was that production did indeed recover in the designated areas. However, detailed examination suggests, according to Samuels and others, that output growth occurred despite the policy rather than because of it. A powerful and common logical fallacy was at work in propagating belief in the policy’s success: post hoc ergo propter hoc; if one event occurs after another, then the first is considered to be the cause of the second. The fallacy lies in coming to a conclusion based only on the order of events. If coal supplies increased after the priority policy, then growth could be attributed to the policy. Another reason that the policy took on a heroic coloration is that the bureaucrats involved in it struggled under challenging circumstances to resolve a national crisis. The investment of personal energy into a venture that brought apparent victory took on a legendary character in later retelling. Although the bureaucrats would have two powerful weapons of persuasion and control – command over foreign exchange and finance – their efforts turned out to be effective only so long as there were private shortages of these two commodities. Foreign exchange control was critical for enforcing MITI’s industrial policy, particularly for those industries such as oil refining that depended on foreign exchange for its raw materials. This tool remained effective until 1964, when MITI’s authority to allocate foreign exchange ended (Calder 1993: 34). The finance ministry’s authority to exercise administrative guidance came from the 1930s mobilization laws and their continuation after the war. These measures established the foundation for the exercise of administrative guidance in the post-war years. Under this regime, the Ministry of Finance (MOF) required that financial institutions obtain its permission to make long-term loans or underwrite large securities issues. By one estimate, the financial situation loosened sufficiently such that, by 1955, the financial lever had lost much of its potency (Calder 1993: 15). As in many other countries, government planners and regulators became the pawns of the industries and companies they were supervising and of politicians seeking electoral advantage. Internal battles for dominance within the government apparatus further weakened any sense of coordinated influence. Nevertheless, despite losing its coherence, bureaucrats’ inclinations to distrust markets and profit motives continued to influence written and unwritten regulation and guidance until the 1990s.
7 Bank-centred finance and corporate governance
The canonical post-war economic system The American Occupation of Japan preserved certain key elements fashioned during Japan’s war years, roughly 1935–45. Among the wartime systems that would be central to the performance of the Japanese economy for the next 50 years were bank-centred finance and corporate governance. These two institutions formed crucial parts of what may be called the canonical Japanese post-war economic system. By canonical, I mean the standard, orthodox or commonly accepted notion of the main strands of the economy that, as a whole, set it apart from other developed nations. In a comprehensive summary, Lincoln describes these elements as an interlocked system with mutually reinforcing parts, given additional strength by their consistency with cultural norms (Lincoln 2001: 16–55). The system consisted of the following main elements: bank-centred finance; corporate governance with weak shareholders, managerial control, and oversight by main banks; keiretsu: networks of businesses centred on a financial institution, trading company or large manufacturer and its suppliers, often owning stakes in one another as a means of finance and mutual security; reduced price competition, including cartels, both legal and informal; internal labour markets with a commitment to so-called lifetime employment; and tight regulation of key sectors along with industrial policy administered by government agencies. This chapter deals with the first two elements.
Bank-centred finance Until the late 1930s, large Japanese companies raised most of their funds by retaining earnings, by selling shares in the company (equity finance) and by issuing bonds. Small firms were more dependent on loans from banks and wealthy individuals. For large corporations, bank loans accounted for no more than 15 per cent of total funding from 1911 to 1936 (see Table 7.1). Neither did banks supply funds by buying the shares and bonds of large firms. In 1919, for example, banks held only 3.2 per cent of the stock of 511 large companies; individuals, with 76.2 per cent, held the bulk of the shares (of these, zaibatsu families accounted for
Bank-centred finance and corporate governance 93 Table 7.1 Source of new funds of large Japanese firms, 1911–36 (%) Equity finance Retained earnings Bonds Bank loans Other
1911–19 34.4 33.4 4.5 4.6 23.1
1919–26 48.8 4.8 26.4 6.4 13.6
1926–31 39.4 –2.0 44.1 14.6 3.9
1931–36 53.6 28.6 9.1 –1.7 10.4
Number of firms
123
111
134
155
Source: Miwa and Ramseyer (2002: 139).
only 2 per cent); non-bank businesses held the rest (cited by Miwa and Ramseyer 2002: 137). This picture changed in the mid-1930s, particularly among firms supplying the military. Figure 7.1 shows that the source of corporate external financing shifted dramatically from 1935 to 1945. According to BOJ figures for 1935, stocks and bonds represented 70 per cent of the share of funds that companies raised from external sources, with stocks accounting for three-quarters of that amount; private financial institutions, mainly banks, supplied 30 per cent. Within 10 years, the ratios were reversed, with banks providing the bulk of companies’ financing needs. In the post-war period, these relative positions changed little. For several decades, private banks supplied more than three-quarters of the funds that firms did not generate internally through retained earnings. The shift in business financing developed in several stages. A significant step was the 1928 banking law, which required licensing of banks and minimum capi100 Private banks 90 80 70
%
60 50 40 30 20
Stocks
10 0 1935
1940 1945
1950
1955 1960
1965
1970 1975
1980
Figure 7.1 Sources of corporate external funds, 1935–80 (%). Source: JSA (1999: Table 11.18).
94 Bank-centred finance and corporate governance tal requirements; the law gave the finance ministry considerable leeway to direct private financial affairs by issuing administrative guidance on issues not spelled out in the law itself. The 1932 foreign exchange control law built on this example. In addition to giving the government an entry point into business affairs, the foreign exchange law created the precedent of leaving its detailed application to designated government officials (Ueda 1999: 42). The largest private banks, in 1933, established a committee to regulate bond issues for the stated purpose of restoring stability and soundness to the bond market. However, since bonds competed with bank loans, the bank-dominated committee discouraged bond flotations. It established the principle that bonds would not be issued without collateral, usually real estate or government bonds. With the support of the finance ministry, the bank cartel operated under regulations that gave it the sole right to manage the required collateral, for a fee. Securities companies could participate in underwriting bond issues, but only the banks could set the fee and collect it. The fee acted as price of entry to the bond market; it was set at a high level to discourage the use of bond financing, but would earn the banks a compensatory amount if firms chose to issue bonds. After the war, bond underwriting was restricted to securities companies, but banks retained the right to set and collect trustee fees for handling collateral. According to early 1980s estimates by the bond underwriters association, at a time when the closely controlled system was beginning to unravel, the fee on bonds issued in Japan was 15 times greater than on unregulated Euroyen bond issues (Weinstein and Yafeh 1998: 637). Euroyen are deposits denominated in yen at banks outside Japan, and thus not under the jurisdiction of the Japanese authorities. Legislation in 1937 and 1940 authorized government intervention in private funding for equipment investment and operating capital. Among other provisions, firms had to apply for ministerial approval for equipment funds above a minimum value. The government imposed strict controls on bond flotations and bank lending (Ueda 1999: 43–4). As the war intensified, the government increasingly pushed banks to provide long-term funds to munitions firms. In 1941, the finance minister assembled the leading private financiers to discuss the government’s new corporate finance programme. As he described it, the government would require banks to supply longterm funds for expansion of productive capacity. The minister noted that Japanese banks generally had maintained lending practices directed towards short-term needs, but they would no longer be free to do so (Miwa and Ramseyer 2002: 138). In 1944, the ministry directed individual banks to take responsibility for specified munitions firms. Another trend that contributed to a changed role for banks was a steep decline in their number. The Kanto earthquake of 1923 and the 1927 banking panic had drastically reduced the number of Japanese banks; many were forced into insolvency directly by the two events, while others closed when depositors fled the smaller ones, fearing their safety. The BOJ encouraged the trend to fewer, larger banks in 1927 by providing emergency funds only to the large ones; the 1928 banking law encouraged further consolidation.
Bank-centred finance and corporate governance 95 During the war, the finance ministry promoted radical consolidation, particularly the concentration of banking activities in the top five zaibatsu banks; the share of all bank loans accounted for by these five rose from under 20 per cent in 1925 to 57 per cent in 1940 (Calder 1993: 28). Figure 7.2 portrays the shrinking number of Japanese banks. In 1920, there were more than 2,000 banks competing for business; not only was competition intense, but the large number made government oversight, control and intervention difficult. By 1945, the number had been reduced to 69, including a dozen or so large, nationwide banks. Influence by government officials, including the ability to form and manage cartel-like behaviour, became both easier and more likely. At the end of the war, most banks were insolvent. The Occupation authorities prohibited payment of amounts owed by the government to companies and banks for war-related expenses. Also, inflation reduced the real value of banks’ outstanding loans. Zaibatsu dissolution left member banks without the support of group companies that they would have expected in the past. Companies were forced to turn to the government-owned RFB for working capital and funds to cover the gap between costs and controlled prices. Within SCAP headquarters, disputes arose concerning bank policy. The antitrust division wanted to break up the banks in much the same way it was dissolving other large corporations. The finance division opposed this approach because it considered an intact banking structure essential for rebuilding the economy. When the general policy towards deconcentration was relaxed in 1948, it was decided to preserve the existing banking structure (Beplat 1980: 236–9). SCAP also wanted to amend the banking act of 1927 to reduce the discretion of the Ministry of Finance (MOF) and set clear standards for banking operations, accounting, and regulation. Ministry officials preferred its system of vague regulations and administrative guidance. After considerable discussion, the law 2,100 1,800 1,500 1,200 900 600 300 0 1920 1924 1928 1932 1936 1940 1944 1948 1952 1956
Figure 7.2 Number of banks, 1920–56. Source: JSA (1999: Table 11.9); includes national banks, private and ordinary banks, savings banks, and special banks.
96 Bank-centred finance and corporate governance remained unchanged and the ministry was able to continue its customary practices (Ueda 1999: 55). When the Dodge plan closed the RFB, the commercial banks were the only source of business finance. However, the banks had lost most of their capital; with a still-recovering economy, deposits were not flowing into the banking system in the volume needed to refinance industry. The BOJ decided to step into this breach. However, its ability to inject funds into the market through open market operations by purchasing government bonds was crippled. Inflation had wiped out the value of existing government bonds and the government budget was in surplus under the Dodge plan; the value of government bonds held by the banks or the general public was too small to allow effective open market operations. The central bank relied on one remaining instrument: lending to the banks. The BOJ made liberal loans in 1949 to prevent a possible hard landing from the government budget surplus and RFB closure. Private bank borrowing from the BOJ peaked in 1951, and then declined as the economy and banking system became self-sustaining. However, the BOJ would continue to use such lending, ‘window guidance’ as it was called, as a central tool of monetary policy for the next 25 years. This tool, operating directly through the banking system, also served as a means to induce banks to steer in directions favoured by the government. With the departure of the Americans, the old zaibatsu tended to reform themselves as keiretsu; the main difference was that the holding company structure was destroyed, as were the investments of the former zaibatsu families. Crossshareholding among former zaibatsu members, which had been prohibited by the Occupation, was allowed after the Americans left. Lending by the central keiretsu bank, often the same one to which companies had been tied during the war, became the dominant financial source of funds for member companies. Non-keiretsu firms formed similar relations with their own banks. The so-called main bank was usually a company’s largest single lender. Although the Occupation had limited banks’ holdings of shares of their customers to 5 per cent of a firm’s outstanding shares, this limit was raised to 10 per cent in 1953 (reduced again to 5 per cent in 1977). Within these limits, the main bank was not only the largest supplier of loans to a company, it was often also its principal shareholder. With this financial exposure to the firm and access to company information, the main bank had the motive and means to monitor firm performance and organize assistance in times of trouble or crisis. The main banks dispatched staff to their clients to oversee projects and reconstruction plans, and replaced management when it seemed necessary (Okazaki 1999: 133). The centrality of bank financing raises the question of why other sources did not develop. It was noted above that stringent requirements and collateral fees were strong disincentives to issuing bonds. The stock market was another potential source of funds. Stock markets were closed at the end of the war and did not reopen until May 1949. But laws and regulations from the wartime period combined with norms of corporate governance that made shareholding riskier than it had been previously. As described in Chapter 5, the munitions corporation law
Bank-centred finance and corporate governance 97 of 1943 transformed corporate governance by restricting the rights of shareholders. It changed the commercial code requirement that corporate executives be the agents of shareholders; instead, executives would be responsible for production, based on directives from industrial associations or government. Under the munitions law, shareholders were no longer the residual owners of profits, which were to be allocated to workers, shareholders and directors. When the stock market reopened in 1949, the Occupation authorities expected it to supply long-term capital while banks covered short-term needs. The breakup of the zaibatsu and other deconcentration measures had placed most shares in the hands of individuals (Figure 7.3). These holdings were small; most minor investors had little incentive to monitor company performance. When it came to investing in new shares, the weakness of shareholder rights and the poor profit situation kept investors of all types away from such opportunities. Since individuals were reluctant to place their savings in company stock, and banks remained poorly funded, the central bank’s supply of funds to commercial banks became the main source of bank lending to companies, although banks also acquired corporate securities. By 1970, banks were the predominant owner of corporate shares as well as the main provider of funds through loans.
Corporate governance In 1881, the Japanese government invited a German scholar to produce a draft commercial code, which was implemented in 1899. The code was based largely on German practice; its corporate governance portions included an anti-shareholder stance inherited from its origins. Although the code was amended several 70 60
Individuals Financial institutions
50 40 30 20 Business 10
Foreign 0 1949 1954 1959 1964 1969 1974 1979 1984 1989 1994 1999 2004
Figure 7.3 Shares owned by type of shareholder, Tokyo Stock Exchange, 1949–2005 (%). Source: TSE (2006).
98 Bank-centred finance and corporate governance times, the legal structure put in place in 1899 remained largely intact until 1945 (West 2000: 11). The American occupiers had different ideas for Japan’s post-war company law. Lawyers, judges and scholars, with University of Illinois training and Chicagobased experience, dominated the legal section of SCAP headquarters. In rewriting the Japanese commercial code, they relied heavily on Illinois law. Their 1949 corporate governance draft emphasized shareholder rights. When it was submitted to a Japanese review committee, one of the leading Japanese scholars commented, ‘The request from General Headquarters for revision came just as we expected, but it included provisions on strengthening the role of and elevating shareholders beyond anything we had imagined. If these requests became reality, wouldn’t they simply be used by troublemakers?’ (West 2000: 13). In fact, Illinois business law was one of the most advanced in the United States and served as a basis for the American model corporation act. After more than 50 meetings with a review committee, the draft law was circulated to legal scholars and the business community. Both groups criticized the proposed drastic increase in shareholders’ rights. Although the US side compromised, its position ultimately prevailed on most issues. The Diet passed the final draft in May 1950; it dramatically increased shareholder rights, established boards of directors, curtailed the pre-war power of statutory auditors, fostered individual share ownership and introduced concepts of director accountability – all largely along the lines of the Illinois Business Corporation Act (West 2000: 18). This transplant of foreign corporate law into an existing Japanese framework produced an odd assemblage; elements of the older German approach continued in effect while some parts of the Illinois system were not included in the Japanese package. After the Diet approved the legislation, several Japanese organizations pressed for a provision that would limit frivolous shareholder suits against a company. The Americans accepted this amendment, which called for a bond to be posted by plaintiffs; subsequently, the value of the required bond was set at such a high level that it created a substantial barrier to shareholder complaints. After passage, business groups continued to complain about the shareholder provisions of the American-drafted code. In 1954, many of these objections were incorporated in revisions that weakened shareholder rights. However, the greatest effect of the widespread dislike of the American-imposed norms appeared in the wide gap between law and practice. For example, the board of directors nominally monitored corporate behaviour and possessed the authority to make important strategic decisions. However, since the boards typically comprised long-serving managers, they rarely took actions against themselves. As noted by Milhaupt, The common practice of having a very large board composed almost exclusively of senior managers is consistent with the view of board seats as incentive devices for loyal employees. It is harder to square with the legal conception of the board as the locus of corporate monitoring and strategic decision making. (Milhaupt 2001: 2092)
Bank-centred finance and corporate governance 99 Individual shareholders were also unlikely candidates to monitor firm behaviour. Zaibatsu break-up and other Occupation policies were effective in distributing shares widely among the population. Although individuals held more than 70 per cent of all shares in 1949, banks and companies had accumulated two-thirds by the 1970s, most of which were designed to be ‘stable’. That is, the holdings were meant to cement long-term business relations and contribute to management stability; outsiders, whether domestic or foreign, faced solid defences against uninvited takeover attempts because of the shares held by friendly owners. Several hostile bids in the 1950s impressed on managers the need for such defences; cross-shareholding emerged as a response, and then increased rapidly as the Japanese economy started to open up to foreign capital in the 1960s and 1970s. By their very nature, stable holdings had to be less profitable than unconstrained portfolios oriented towards profit maximization; only by accident would defensive shareholdings also be the most profitable use of a firm’s money. The corporate governance system in which these holdings were embedded was fashioned to achieve objectives other than profits.
The role of norms Many writers have observed that the canonical elements of Japan’s post-war economy formed a mutually reinforcing, integrated system (Johnson 1982: 12; Lincoln 2001: 19). Milhaupt (2001: 2086) explains the presence and survival of this linked complex in terms of social norms: rules that are not promulgated by an official source nor enforced by threat of legal sanctions, yet are regularly complied with. His system includes several elements: the main bank, which performs the key monitoring role over client firms; the absence of an external market for corporate control; management-dominated boards that focus on operations rather than monitoring; and the lifetime employment system, in which certain employees enjoy job security and rise to the top management positions. Milhaupt argues that the Japanese banking system up to the 1990s could be viewed as a regulatory cartel in which both the regulator and regulated cooperated to enforce market segmentation, control entry, regulate output and allocate the gains among cartel members. Control of financial products and entry of new members were managed by the MOF and private groups such as the Federation of Bankers Association. This system was supported by a cluster of norms: survival of the weakest member bank with no failures (the convoy system); support of weak members by keiretsu or stronger banks; and an implicit government assurance that the system would work as expected (Milhaupt 2001: 2089). The government guarantee that no bank would fail buttressed the entire scheme. It also implied that no major borrower would fail or leave large amounts of unrecoverable debt that could threaten a major lender. Government backing encouraged depositors to trust their money in the banks, which then lent to expanding companies in Japan’s miracle growth years. The assurance of safety, however, introduced a noxious element into the economy – moral hazard. Moral hazard exists because the insured take greater risks than they would without insurance,
100 Bank-centred finance and corporate governance relying on the protection gained through the insurance; when a financial system is insured, the result tends to be greater lending and investment with lower returns than otherwise. Moral hazard combined with corporate governance based on main bank monitoring would turn out to produce future economic difficulties in the form of very low returns on investment. As will be seen in the next chapter, this problem did not begin with the asset bubble and its collapse in the late 1980s, but started to appear as soon as the post-war growth spurt ended in the early 1970s. Linked to the main bank system was the absence of a market for corporate control. Weak or failing firms were the responsibility of the main bank. A shared understanding was that Japanese managers did not sell their companies, particularly to an uninvited bidder. Until the late 1990s, there were almost no hostile takeover attempts. In any event, stable cross-shareholding among Japanese companies and banks as well as several legal obstacles made hostile takeovers unlikely to succeed. Since the almost total absence of management mobility between firms made a takeover risky for entrenched managers, the anti-takeover norm certainly acted in the interest of those very managers who had been keen promoters of the idea that hostile takeovers border on the immoral. The norms provided private benefits to favoured groups, specifically MOF bureaucrats, major commercial banks and corporate managers. The anti-takeover norm protected managers from loss of their jobs and also justified their managerial power. The corporate board, with a self-serving emphasis on preserving employee job security and loyalty plus the exclusion of outsiders, served to insulate themselves from ouster for poor performance. The main bank system provided banks with stable, long-term customers and the possibility of extracting economic rents from their clients. MOF regulators benefited from public prestige and enhanced discretion in their official functions; they also found post-retirement employment at the regulated firms (Milhaupt 2001: 2102). Milhaupt’s central conclusion is that the main bank and corporate governance system was a largely self-enforcing, informal response to the post-war legal environment. Because government bureaucrats also received private benefits, there was a built-in coordination mechanism. Since the practices ‘continued to generate significant private benefits to powerful groups, they were relatively impervious to change despite their eventual inefficiency for society as a whole’ (Milhaupt 2001: 2102).
End of an era Japan’s classical post-war economic system changed in a gradual fashion from the 1970s. The first moves occurred in the tightly bound financial system. When the government incurred large fiscal deficits in the 1970s, it was forced to relax its interest rate controls in order to make government debt attractive to financial institutions. At around the same time, Japanese companies started to venture abroad. They found that they could raise funds in London more cheaply than in Tokyo; additionally, the range of financial products was much broader in the foreign centres. In the early 1980s, the Tokyo manager of a large American financial
Bank-centred finance and corporate governance 101 company told me that he sold 100 products in New York, but only two in Tokyo – stocks and bonds. The rise of the offshore Euroyen market, beyond Tokyo’s regulatory range, further encouraged liberalization at home. Japan joined the International Monetary Fund and the Organization for Economic Cooperation and Development in 1964, both of which stipulated that members not restrict their international current account transactions. The further opening of international capital markets in the 1970s and early 1980s following the collapse of the fixed exchange rate system exposed Japanese companies and financial institutions to new threats and opportunities. The 1980 amendment of the foreign exchange control law permitted crossborder capital transactions with only token official notice. Within three years, the value of funds raised abroad surpassed domestic bond issues. As a result, the domestic bond cartel was forced to revise its criteria for issuing bonds. In 1979, only two companies met the criteria; by 1989, 300 firms could issue straight bonds. Financial market innovations, deregulation and the substantially increased scale of international financial flows in the 1980s turned London, New York, Hong Kong and Singapore into global financial centres. Tokyo lagged far behind these developments, largely because of the still remaining controls and regulations. Businesses sought more diverse and less expensive financial services; foreign demands for greater openness, as well as Tokyo’s loss of prestige among financial centres, led Prime Minister Ryutaro Hashimoto in 1996 to announce a ‘big bang’ scheme of financial market deregulation, phased in over several years through 2001. A new Financial Services Agency was split from the MOF to supervise the liberated financial sector. These changes helped end the system of bank-centred corporate finance, especially for large firms. Figure 7.4 shows the share of all assets financed by banks, stocks (paid-in capital) and bonds for large manufacturing firms. One major source of funds, retained earnings, is omitted for purposes of clarity because it overlaps the curve of paid-in capital. As the role of bank lending fell in the 1980s, that of bonds and stocks increased. In the 1990s, other factors also played a part in bringing an end to the old financial regime. Banks’ non-performing loans and dangerously low capital highlighted the need for greater flexibility in developing new revenue-producing financial products; relatively unfettered foreign financial institutions aggressively competed with Japanese banks at home and abroad; Japanese companies’ desire for more sophisticated financial services pushed the case for more permissive regulation; scandals among finance ministry officials cast doubt on the ministry’s ability to regulate the industry; pressure from the United States to open pension markets and trust banking reinforced domestic demands for higher returns; and the apparent failure of the old system to monitor banks and other financial intermediaries – demonstrated by the outright failure of several major financial companies in 1997 and the suspected insolvency of many others – legitimated proposals for a new oversight regime. The rise of market-based finance displaced the locus of interests in monitoring
102 Bank-centred finance and corporate governance 45 40 35 Bank loans 30 25 20
Paid-in capital
15 10 5 Bonds 0 1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
Figure 7.4 Share of total assets accounted for by bank loans, paid-in capital and bonds (large manufacturing firms), 1960–2005 (%). Source: Ministry of Finance, Financial Statements Statistics of Corporations.
corporate behaviour. Banks, fearful of causing the collapse of their customers, had been unwilling to identify weak firms. However, by the late 1990s, tighter financial regulations restricted the banks’ ability to bail out troubled borrowers, thereby reducing the value of the main bank relationship to businesses. As business ties to banks weakened, shareholder monitoring became more important. The shift of corporate funding from banks to capital markets helped drive corporate governance reform. More corporate law revisions occurred in the 1990s than at any other time in Japanese history. Significantly, the push for reforming corporate governance regulations came from the leading business groups working with political leaders, and supported by METI. This alliance dethroned the slow-motion process controlled by the Ministry of Justice, which customarily required decades to introduce a change in company law. Oversight and monitoring were strengthened in 1993 under pressure from the United States in trade negotiations by reducing the costs to initiate shareholder suits and allowing collection of damages. From 1950 to 1990, shareholders in Japan filed fewer than 20 derivative suits against directors. (A derivative suit is an action brought by a shareholder, not on its own behalf, but on behalf of the corporation, on the grounds that the corporation is not pursuing its duties owed to the company and its shareholders.) By 1999, there were 286 such suits before the courts, 95 filed in 1999 alone (West 2001: 352). The possibility of successful major suits against company directors came to the fore in 2000 in the Daiwa Bank case. A shareholder suit alleged that Daiwa had allowed almost $1.5 billion of losses and fines in New York arising from unau-
Bank-centred finance and corporate governance 103 thorized trading by a company employee, which the company compounded by filing misleading information with the US Federal Reserve; the Fed fined Daiwa $340 million and withdrew its business licence. The Osaka court found the directors liable for breach of duty and fined them about $775 million in damages. As quoted by Milhaupt (2001: 2083), the court found that the directors ‘had persisted in following informal local rules that only apply in Japan, despite the fact that Daiwa’s operations had expanded on a global scale’. After appeal, a 2001 agreement had 49 executives pay the bank a total of ¥250 million (about $2 million) in restitution. Because of the increased risk to company directors of unlimited penalties, the Diet approved changes to the commercial code in 2002 to limit the amount of damages shareholders could seek from directors. The maximum was set at six years’ compensation. Other changes reduced the shareholding threshold to demand the inspection of records, allowed stock options, simplified merger procedures, allowed stock swaps to underwrite mergers, eliminated the ban on holding companies, facilitated the ability to spin off companies, removed prohibitions on treasury stock and authorized companies to adopt an American-style board of directors. These many changes resulted in a thorough revision of the commercial code, culminating in a new corporation law that took effect in May 2006. The new law embraced a reversal of the regulatory philosophy. It shifted from a stance that everything that is not specifically allowed is prohibited to one where everything that is not specifically prohibited is allowed. Under this greater flexibility, it would be up to the courts to rule on problematic issues as they arose. One scholar interprets these changes by noting that mangers, under the old system, had limited flexibility but could not be held responsible as long as the company acted within the law; under the new system, management is free to act as it sees fit, but the law greatly expands the rights of shareholders, the need for disclosure and the accountability of managers (Schaede 2006: 9–10). The changes in laws paralleled changes in norms. Norms, though, were resistant to quick transformations. Because of their informality, they could not be amended as a regulation or law might be. Perhaps even more important, norms are entwined with special interests. These, however, were also changing. For example, the removal of banking regulation from the MOF to the independent Financial Services Agency contributed to a breakdown of the personal interests of MOF bureaucrats in protecting and preserving the regulatory banking cartel. Legal changes that made shareholder suits more likely forced corporate officers to consider seriously bids for mergers or acquisition; failure to accept clearly profitable offers could be interpreted as acting against shareholder interests. The number of mergers and acquisitions subsequently increased to levels that would have been unthinkable as recently as the mid-1990s. Moreover, the number of foreign companies taking over Japanese, counted in single digits in the 1980s, has been over 100 per year since 1999 (see Figure 7.5). Although successful hostile takeovers are still infrequent, they are increasing at a fast rate, from only a single recorded case in 1995 to 53 in 2005 (Schaede 2006: 40).
104 Bank-centred finance and corporate governance 2,400
2,000
150
Foreign with domestic firms (right scale)
125
1,600
100
1,200
75
800
50
400
25 Domestic with domestic firms (left scale) 0 1995 2000 2005
0 1985
1990
Figure 7.5 Mergers and acquisition in Japan, 1986–2005 (number of transactions). Source: Nomura Securities (2006).
The evidence seems clear that the 1935–45 system slowly dissolved, with acceleration in the 1990s. Bank-centred finance is no longer the norm for large companies; corporate governance is returning to the hands of the shareholder; at the end of the 2005 fiscal year, more than one-third of listed firms had appointed outside directors. Nevertheless, banks remain central to smaller firms; lifetime employment persists, although the legal support and nominal allegiance to it is fading; and even though managers are responding to shareholder interests with greater alacrity, many companies are openly hostile to demands to foster the financial interests of their owners. As Milhaupt (2001: 2126–7) notes, ‘it seems unlikely that countries can embrace the shareholder supremacy norm for corporate activity without wholesale revision of deeply entrenched views and practices in other areas of society’. Nevertheless, Japan has moved a great distance in this direction.
8 Capital investment and rates of return
In a capitalist economy, the payment to capital is one of the chief means for determining how well the economy is doing, especially in comparison with other countries. The rate of return on the accumulated stock of past investments is a convenient index of capital’s performance. Investment is the one indispensable ingredient for modern economic development and growth. Japan is no exception. Investment increased its productive capacity, stimulated growth by contributing to aggregate demand and advanced productivity through the introduction of improved technologies and methods. An economy can also grow if other inputs expand or if efficiency increases. For economies far from the technological frontier, as Japan was for most of its modern history, investment is the primary source of growth. Investing in industrial plant and equipment, infrastructure such as roads and telecommunications, research and development, and in human potential through education, training and improved health requires that current output be diverted from current consumption. A portion of national income must be saved and devoted to uses that will have a future payoff. The Japanese economy put more than one-third of its total output into gross capital investment in the peak years of its post-1945 economic miracle. Indeed, Japan has invested in new capital with an enthusiasm that at times has appeared to be excessive. Japan’s growth experience fits the standard theory. Figure 8.1 plots a 1990 cross-section for 59 countries of GDP per capita against the amount of capital invested per worker. Also plotted are the time-series data for Japan from 1965 to 1992. Japan’s experience fits comfortably within the broader record. Although some countries seem to get more GDP per capita from the same amount of capital as employed by Japan, others got less. It is not possible from only these data to decide how well Japan is using its resources, although it may be hard to argue against any process that enlarged real output per person by four times in less than 30 years. As capital accumulates, the returns from additional investment tend to decline. Mature economies must look towards increased productivity as the main source of improved economic welfare. Unfortunately for Japan, not only did the marginal returns from its investment decline as the stock of capital expanded, capital
106 Capital investment and rates of return 20000
GDP per capita
16000
12000 Japan
8000
4000
0 0
15,000
30,000
45,000
60,000
75,000
Capital per worker
Figure 8.1 Real GDP per capita and real capital per worker, Japan, 1965–92, and 59 other countries in 1990 (1985 dollars at purchasing power parity). Source: Summers et al. (1994).
productivity was below that achieved by other countries at similar levels of development. The alleged qualities of discipline and patience attributed to Japan’s corporate investors became synonymous with low returns and inefficient capital. The problem and paradox of Japanese investment is illustrated in Figure 8.2, which shows the private sector’s ratio of capital stock to GDP for Japan and the United States since 1955. (See Appendix to this chapter for a more detailed explanation of methods and data.) American output growth required little additional capital deepening, whereas Japan’s capital, after recovering from wartime destruction, continued to climb. According to these estimates, Japan’s private sector by 2004 required 50 per cent more capital per unit of output than did the American. Why did Japan’s business sector continue to invest so prodigiously and build up such a large capital base? Did high returns to investment justify this accumulation? This chapter argues that low productivity reduced returns below the potential rates demonstrated in other countries. To put it bluntly, Japanese investors were wasting resources. I allege that the chief source of this behaviour was weak corporate governance, based on the bank-centred financial system.
Analysing rates of return Before proceeding, it might be useful to define the meaning of capital as used here and in most other studies. Some writers define capital as those assets that meet three criteria: means of production, produced means of production and durable. This definition rules out housing, consumer durables, human capital, non-produced assets, such as natural resources, and such things as social capital or institu-
0.77 0.76 0.79 0.77 0.78 0.77 0.79 0.89 0.96 0.97 0.99 1.10 1.16 1.18 1.20 1.20 1.18 1.20 1.24 1.29 1.33 1.35 1.35 1.40 1.42 1.42 1.45 1.49 1.55 1.63 1.71 1.75 1.75 1.76 1.76 1.82 1.84 1.83 1.86 1.88 1.87 1.87
Capital investment and rates of return 107 2.0 1.8
Japan
1.6 1.4 1.2 1.0
US
0.8 0.6 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Figure 8.2 Ratio of private sector capital stock to private sector GDP (1955–2005, 1990 prices). Note: Penn World Table depreciation method; see Appendix for details.
tional capital (as useful as these concepts may be for other purposes). Assets that meet the three criteria include such durable goods as non-residential structures, machinery and equipment. The chief means for assessing the payoff from investment is the rate of return on capital. Rate of return can be defined as the income derived from additions to the capital stock divided by the value of that increment, or the marginal product of a unit of capital holding other inputs constant. Rates of return below the cost of capital indicate that resources are being wasted. Returns below those available elsewhere suggest deficient management of the investment process or weak incentives influencing business decisions; persistently low returns relative to comparably risky alternatives often lead to a flow of capital from low-return projects to those with higher return. For a company, returns below the cost of capital can contribute to insolvency. Most economic growth studies examine productivity rather than rates of return. Productivity can be defined simply as the ratio of outputs to inputs; total factor productivity includes all inputs in the calculation. Although productivity and rate of return share several attributes, they do not always tell the same story. For example, countries can have very different average productivity levels and yet companies in these countries can generate high payoffs to capital and compete fiercely with each other across national borders; the balancing factor is exchange rates, which can convert a company in a low-productivity country into an international competitor. Even within a country, industries and products with fast-growing productivity are not necessarily profitable; one need look only at the airline industry in the United States or at semiconductor memory production almost anywhere to see two examples of fast-growing productivity and low profitability. However, profitability of companies within an industry in a specific country will tend to be
1 8 9 0 0 7 4 0 0 9 2 4 2 0 1 5 1 3 9 0 1 2 6 2 6 6 7 8 7 7 4 6 5 5 2 0
108 Capital investment and rates of return related to productivity. In the US airline industry, Southwest Airlines is the most profitable and the most productive. The important point is that companies do not take productivity to the bank; they take their returns. This chapter will consider aggregate total returns in the Japanese economy, aggregate private returns, returns in corporations and returns on Japan’s foreign assets as well as the earnings of foreigners in Japan. For comparison, returns on comparable American assets will be examined.
Economy-wide returns
Japan As might be expected from a capital-poor, but rapidly industrializing nation, returns on PWT K, 0.35 returns to new investment in Japan 100 years ago were very high, around 50 per cent according to alternative methods of measuring the capital stock. elasticity 63.93 59.93 57.31 55.59 54.05 56.26 59.41 61.13 61.24 60.10 55.38 56.25 53.12 50.16 54.46 50.90 49.03 47.94 48.89 46.88 45.51 44.75 45.60 48.56 50.75 50.54 48.89 51.16 49.90 47.48 48.28 41.22 39.25 37.99 34.57 19.07 22.93 24.07 26.52 26.34
Figure 8.3 shows returns for both the United States and Japan from the early part of the twentieth century. In capital-poor Japan, the gains from new investment started at very high levels and then fell gradually as investment expanded the capital stock. High returns from exports during the First World War (1915–18) and then when Japan left the gold standard and stimulated the economy with Keynes-like policies (1932–38) show up clearly in the figure. Brisk infusions of capital into the war economy in 1940 combined with widespread destruction and business disruptions in the following years pushed down returns. Following the wartime collapse, returns quickly rose during the rebuilding effort, reached a peak in 1960 and then fell for the next 45 years. Excepting Japan’s experience during the war and its immediate aftermath, roughly 1940 to 1955, Japan’s rate-of-return trajectory follows a remarkably stable downward course for a century. The American experience offers a sharp contrast to Japan’s. Although already a developed economy by 1900, the US economy still benefited from capital deep70 Japan
60 50 40 30 20 10 0 1910
US 1920
1930
1940
1950
1960
1970
1980
1990
2000
Figure 8.3 Real returns on total aggregate non-residential capital (Japan: 1910–2004; US: 1914–2005). Note: Penn World Table depreciation method; elasticity of capital 0.35; see Appendix for details.
Capital investment and rates of return 109 ening; returns rose along with high investment rates until the late 1920s and the Great Depression. Recovery and post-war growth restored the American trajectory to a slowly rising trend that approached an asymptote. The capital stock on which the returns in Figure 8.3 are based includes public as well as private investment. Although the provision of infrastructure in an undeveloped economy is expected to have a high payoff, the Japanese government’s 30-year, politically motivated support of economically dubious public investment projects from the 1970s probably drove down aggregate returns. Moreover, if the gradual accumulation of economic reform has actually affected performance, it would show up in the private economy rather than in government activities. Both Japan and the United States publish information on GDP originating in the private sector and on private investment. The capital data are represented above in Figure 8.2 and are the basis for estimating returns in the business part of the economy (Figure 8.4). The qualitative story about the private sector is similar to that for the total economy; Japan’s business returns recovered from wartime losses and kept improving until 1960, levelled off for about eight years, and then declined steadily and substantially. The American curve for the private sector closely tracks the aggregate curve, but is several percentage points higher, remarkably holding to a relatively stable long-term plateau. The large and steady fall of returns in Japan arises from one main cause: continuous expansion of the capital stock relative to output. High rates of investment continued in Japan well after the economy decelerated in the 1970s. Although the slowing economy had less need for new capital, the share of investment in GDP barely declined. Consequently, the ratio of capital stock to output rose, and continued to rise even as returns fell. The American capital–output ratio remained
50 45 Japan
40 35
US
30 25 20 15 10 5 0 1930
1940
1950
1960
1970
1980
1990
2000
Figure 8.4 Real returns on private sector non-residential capital (Japan: 1955–2004; US: 1930–2005). Note: Penn World Table depreciation method; elasticity of capital 0.35; see Appendix for details.
110 Capital investment and rates of return relatively constant for almost 50 years whereas Japan’s seemed to grow without bound, even as the economy decelerated to under 1 per cent growth in the 1990s. The relative decline of Japan’s returns is not unique to comparisons with the United States, but also appears alongside other advanced economies. Making use of Maddison’s capital stock estimates, Table 8.1 shows the rates of return of Japan and four other countries. Japan and the European countries had very high returns in the wake of wartime destruction. Returns then fell in all countries, but Japan’s returns fell below those elsewhere. This result does not depend on the technical details of the estimation method. The table shows two possibilities, one with variable elasticities of output with respect to capital as estimated by Kim and Lau (1994), and the other with constant elasticities. Both results are similar. Due to Japan’s relentless capital expansion, returns fell below those in the United States and Western Europe around the late 1970s, and kept falling.
Corporate returns We can narrow the analysis of the private economy by focusing on non-financial corporations. The MOF publishes data on manufacturing and non-manufacturing corporations in three size groups according to their capital base, the largest with capital greater than 1 billion yen. To calculate returns, I have used operating profit plus interest received, after depreciation, but before payments of interest and dividends. A central assumption about an ongoing establishment is that profit represents a payment to the capital used to produce revenues and income. If depreciation were not deducted from the income stream, it would be equivalent to generating income by selling the capital stock. Operating profit is recorded before the payment of interest; since we are interested in the income generated by the capital stock, the cost of a large part of that capital should not be deducted prematurely in the analysis. Interest and dividend receipts are included in income Table 8.1 Real aggregate rates of return on non-residential fixed capital stock in selected countries, using alternative capital elasticities, 1955–90 (%) Year Japan Kim and Lau variable elasticities 1955 40.3 1960 34.1 1970 18.9 1980 8.0 1990 3.9
United States United Kingdom Germany
France
12.5 12.1 10.7 7.9 5.3
– 23.4 15.2 10.5 8.1
– 23.8 13.9 8.8 5.3
– 26.5 18.8 10.6 6.0
Constant elasticity 0.35 1955 23.6 1960 26.2 1970 20.5 1980 12.9 1990 10.8
14.5 15.2 16.4 15.6 14.8
– 24.1 18.6 15.5 14.7
– 22.5 17.5 14.7 13.5
– 23.7 21.5 16.6 14.8
Sources: Maddison (1995b); Kim and Lau (1994).
Capital investment and rates of return 111 because they are part of the flow generated by assets. (The qualitative results do not change if interest and dividend receipts are omitted.) The most inclusive measure of capital is total assets, which includes financial assets, inventories and fixed assets such as plant and equipment. Since all of these assets assist in the production of goods and services, and since payments to these assets must be generated by profits, I use total assets as the denominator of the rate of return. In order to clarify the situation among Japanese companies, I focus on large manufacturing and non-manufacturing firms; Figure 8.5 shows returns in large companies. The patterns in the other size classes are similar. Several points can be made: manufacturing produced higher returns than non-manufacturing; large companies tend to earn higher returns than small; returns fell from 1960 to 2000, a downward trend especially apparent among large firms and those in manufacturing; returns for all groups have turned up since around 2000. Rising corporate returns since 2000 is consequential; although increases in the past were followed by even larger declines, rates of return now appear to have experienced a turning point. Anecdotal evidence indicates that companies are paying more attention to their bottom line than at any time since the 1930s. For example, stacks of books next to cash registers in Tokyo book stores with titles that include words like ‘understanding return on equity’ are indicative of real change in business thinking. A possible explanation for why corporate data show an upturn in business returns when similar growth does not appear in the GDP calculations is that the estimating method of the aggregate capital stock does not permit or recognize premature write-offs or scrapping, whereas businesses have been engaging in such practices since the late 1990s. In addition, aggregate capital adjusts for price changes in annual investment figures. The MOF corporate survey, in contrast, 12 11 10
Manufacturing
9 8 7 6 5 4 3 2 1960
Non-manufacturing
1965
1970
1975
1980
1985
1990
1995
2000
2005
Figure 8.5 Ratio of operating profit plus interest to total assets, large corporations, 1960–2006. Source: Ministry of Finance, Financial Statements Statistics of Corporations.
112 Capital investment and rates of return reports only nominal figures. The investment deflator has reflected faster productivity improvements than the GDP deflator; the result is that real aggregate investment and capital have risen faster than real output because of different rates of price changes, thereby further reducing the aggregate rate of return compared with nominal calculations.
Foreign returns As of June 2006, Japan owned more than $4,400 billion in foreign assets, of which $326 billion was foreign direct investment (FDI). The largest portion of Japan’s international assets is placed primarily in portfolio investments: stocks and bonds. I focus here on FDI because it represents investments by companies in business activities. When deciding to commit a company’s funds to FDI, corporate managers, presumably, compare the expected returns with alternative uses of the firm’s money. The returns earned in foreign investments should be informative of the company’s overall profitability. The BOJ adjusts original historic costs or book values for price and exchange rate changes. Although these estimates only go back to 1996, the IMF reports FDI stock from 1980; the IMF numbers were spliced to the BOJ data. The return on FDI assets is calculated as the income earned on the assets divided by the capital stock. In addition to Japan’s earnings on its foreign assets, I also calculate the returns earned by foreigners in Japan by dividing income payments by the stock of foreign-owned assets (liabilities, in Japan’s accounts). Figure 8.6 shows the returns on Japanese companies’ FDI assets and for for-
24 21 Foreign assets in Japan 18 15 12 9 6 3 Japan's assets abroad 0 1980
1985
1990
1995
2000
2005
Figure 8.6 Returns on Japan’s direct investment assets abroad and foreign assets in Japan, 1980–2005. Source: Bank of Japan.
Capital investment and rates of return 113 eign companies operating in Japan. Foreign companies consistently earned more in Japan than domestic firms did away from home. From around 6 per cent in 1980, Japan’s FDI profitability fell to below 1.5 per cent in 1990, and then started to rise. By 2005, FDI returns had reached record highs of almost 8 per cent and were converging on foreign experience in Japan. One factor that may have driven down Japanese companies’ profits in the decade before 1990 was the surge of overseas investments in those years. The experience in the United States is that it takes many years to overcome the costs and attendant problems of a new venture. US multinationals got a head start on their foreign counterparts in the post-war period after 1945. Internal Revenue Service data for large American corporations operating abroad show that returns on assets rise steadily with experience – from 2.1 per cent for companies that are less than a year old to 16.1 per cent for concerns that have been in business for more than 29 years (Latzy and Miller 1992: 86). The falling returns on Japan’s FDI after 1980 are consistent with many new investments, but as foreign operations matured, returns remained significantly below what foreigners were earning in Japan. The prevalence of low returns in Japan appears in the macroeconomic data, corporate financial accounts and the FDI estimates. Are the low returns a function of being in Japan, or – alternatively – being a Japanese company? One way to attack this question is to compare Japan’s FDI earnings abroad with those of large domestic manufacturing companies at home. These firms are the most profitable group in the MOF corporate survey and those most likely to be investing in foreign operations. This comparison will indicate whether geographic place of business makes a difference to Japanese companies’ profitability. Figure 8.7 shows that the FDI data are remarkably congruent with manufacturers’ returns; Japan’s FDI returns are in the same range as domestic operations in the years after the initial FDI surge. However, it must be noted that this comparison is based on
8 FDI assets abroad 7 6 5 4 3 2 1993
Large domestic manufacturing companies 1995
1997
1999
2001
2003
2005
Figure 8.7 Returns earned on Japan’s direct investment assets abroad and by large, domestic manufacturing companies, 1993–2005.
114 Capital investment and rates of return different kinds and sources of data, and the results should be considered as no more than suggestive. To extend the comparison, I include, from US data, American corporations’ FDI performance in Japan and around the world together with Japanese companies’ global FDI returns (Figure 8.8). Returns on American-owned direct investment in Japan produce the same returns, on the average, as American investments in the rest of the world. This performance was consistently above Japan’s foreign operations. Recalling the similarity in results of Japanese companies’ foreign and domestic activities (Figure 8.7), it seems clear that companies earn low returns, not countries. Japanese firms have done badly wherever they are.
Dynamic efficiency Japan’s returns on its aggregate capital stock have fallen for the better part of a century. However, they have not fallen below zero. There always have been some positive net gains to investment, in contrast to the last days of the Soviet Union, when marginal returns were estimated to have been negative. Another way to look at returns is through the concept of dynamic efficiency. This idea is derived from models of economic growth, so-called Golden Rules, that define an economy as dynamically efficient if the income attributable to capital is greater than investment; a competitive economy is dynamically efficient if the cash flow generated by production after the payment of wages is greater than the level of investment (Abel et al. 1989: 2). Another way to think about this concept is that capital should yield more in potential consumption than it drains from the economy. 16 US FDI in Japan
12 All-US FDI
8
4
Japan's FDI abroad 0 1980
1985
1990
1995
2000
2005
Figure 8.8 Returns on US FDI in Japan, all American FDI and Japan’s FDI abroad, 1980– 2005.
Capital investment and rates of return 115 Dynamic efficiency is a fairly crude test; it says nothing about profit maximization or optimization; all it requires is that the output generated by capital should be greater than its cost. Several studies have found that for Japan, the United States and other advanced economies, the conditions for dynamic efficiency are satisfied. Figure 8.9 shows Japan’s and America’s annual payments to capital as a ratio to private non-residential investment. If that ratio is greater than 1, the economy is dynamically efficient. For much of the period after 1945, the ratio was around 2.0 for Japan and 3.0 in the United States. Note that before the Second World War, both countries had greater cash flows from their investments than after the war. By 1950, both countries settled into a long-term stability, with the Japanese ratio roughly one-third less than the American. Since the early 1990s, Japan’s payments to capital relative to investment have started to climb. Indeed, during the rush in the United States to invest in information technology at the end of the 1990s, Japanese investors received higher incomes from their commitments than did their American counterparts.
A new chapter? The story told by dynamic efficiency is consistent with the narratives related by the other techniques used to examine returns: high payments from capital in the early 1930s, declining after that, and upturns in the 1990s. The most recent data suggest that convergence between Japan and the United States may be occurring. In a 1967 evaluation of economic policy exercised by the American Occupation authorities and later by the Japanese government, University of Washington economics professor Kozo Yamamura wrote:
10
8
6 US 4
2 Japan 0 1932 1938 1944 1950 1956 1962 1968 1974 1980 1986 1992 1998 2004
Figure 8.9 Ratio of capital payments to private non-residential investment, Japan and US, 1932–2005.
116 Capital investment and rates of return In sum, one could perhaps conclude that the growth-oriented policy has been to encourage over-investment and to protect the resulting excessive competition. . . . Economic growth is also subject to a decreasing return and an increasing cost. What we argue here is that the Japanese economic policy dedicated to growth has passed the break-even point. Yamamura (1967: 200) The fact that foreigners earn more in Japan and globally than do Japanese companies suggests that we look at the factors governing company performance for answers to why returns in Japan have fallen so low for so long, and why there may have been an upturn. I argue that the downward course of Japanese rates of return resulted from corporate governance that did not emphasize profitability, a banking system that did not adequately monitor borrowers, government policy that protected insolvent firms from dissolution, and politicians that liked it that way. The evidence reviewed in this chapter suggests that the conditions that influenced Japanese corporate behaviour for so long may be changing. Although not yet seen at the broadest macroeconomic level, we may finally be starting to witness the awaited change of business incentives and a corresponding upturn of rates of return. As many people in Japan and elsewhere wonder how a nation that is ageing will support itself in the future, higher returns will provide one part of the solution. The main regret is that it did not happen decades earlier.
Appendix Aggregate returns The method used to estimate economy-wide rates of return makes use of production functions, which describe outputs as a function of inputs. My particular objective is to obtain the change of output related to changes in capital, holding other inputs constant. This quantity can be interpreted as the macroeconomic real return to capital. A simplified representation of a production function shows output (Q) as related to the flow of input factors: capital services obtained from the capital stock (K), labour services (L) and possibly other inputs. If the flow of capital services is a fixed proportion of the capital stock, then K can be represented by the capital stock itself. Productivity is frequently included in the function, either as a factor multiplying the production function or as a factor augmenting one or both of the inputs. Focusing on just capital and labour, the production function is: Q = f(K, L). The change of output related to the change of capital, holding other inputs constant, is the partial differential of Q with respect to K, or ∂Q/∂K. Elasticity of output with respect to capital If production exhibits constant returns to scale and if the elasticity of substitution between capital and labour equals 1.0, the production function can be written in
Capital investment and rates of return 117 the form pioneered in 1928 by Charles Cobb and Paul Douglas (later a US senator) (1928: 152). Their equation was: Q = ALbK(1 – b); the exponents on the input factors, labour and capital, sum to 1.0, which generates constant returns to scale. The elasticity of substitution between capital and labour is an important parameter of production functions. It was designed as a measure of the ease with which one factor can be substituted for another. When the elasticity is zero, there is no substitutability; factors must be used in strict proportions (for example, one worker for each sewing machine). For large values of the elasticity, one factor can be substituted easily for another without limit or loss of productivity (highway construction, for example, where labour can be supplemented by capital from simple picks and shovels to large earth-moving and paving machines). Another parameter of production functions that is particularly useful for estimating returns to capital is the elasticity (e) of output with respect to capital, e = (∂Q/Q)/(∂K/K), defined as the ratio of the incremental percentage change of output with respect to an incremental percentage change of capital. In the Cobb–Douglas formulation, the exponent on the capital term is equal to this elasticity. If we multiply it by the ratio of output to capital, the result is the desired partial differential of output with respect to capital: e(Q/K) = [(∂Q/Q)/(∂K/K)]Q/ K = ∂Q/∂K. The task of finding appropriate values of the elasticity of output with respect to capital might appear to be easy because scores of studies have used production functions, often to generate estimates of productivity growth. However, not many studies actually estimate this elasticity in fully articulated equations because productivity estimates are often bound up with the particular functional forms assumed for the production function. To get around the circular problem that productivity estimates are tied up inextricably with assumptions about the form of the production function, economists often simply assume values of key parameters based on their reading of the empirical literature. Two questions, in particular, have exercised the imaginations of theorists: are scale returns constant and is the elasticity of substitution between capital and labour equal to 1.0? Another issue is whether productivity change can be written as a simple multiplying factor of the production function, or whether it influences one or both of the inputs in a more complex manner. For many of the more complicated production functions that have been proposed since 1928, it has not been possible to disentangle factor-augmenting technological change from other production function parameters. Assumptions about the production function’s key parameters hinge on the apparent fact that the share of national income going to labour is relatively constant over extended periods for many countries, and that the share does not seem to vary with an economy’s ratio of capital to labour; because of this regularity, the Cobb–Douglas equation is often chosen as the applicable form. Professors Cobb and Douglas came up with this formulation precisely because it is consistent with the presumed stability of the labour share of income. Functions in which the elasticity of substitution is 1.0 automatically generate this outcome; furthermore, if inputs are paid their marginal product (an assumption about the competitiveness of factor markets), the labour share will be the exponent on the labour variable.
118 Capital investment and rates of return The theoretical life is not this easy, however. Constant shares could also be generated if the elasticity of substitution were less than 1.0 and technological change enhanced labour productivity. An intriguing finding in studies looking into the nature of production is that the elasticity of substitution between capital and labour seems to increase with economic development. One representative study divided a sample of 82 countries into four groups, based on the ratio of capital to labour. For the richest group with the most capital, the elasticity was above 1.0 by a statistically significant amount, whereas for the poorest group, it was significantly below 1.0. However, the estimated elasticity for the highcapital countries was 1.08, economically close to the Cobb–Douglas formulation (Duffy and Papageorgiou 2000: 109). Another study, using the same data, tested whether the elasticity of substitution varied in a continuous manner with capital intensity by estimating a function with a variable elasticity of substitution, allowing the elasticity to change with the capital–labour ratio. The elasticity, again, was statistically different from the Cobb–Douglas assumption of 1.0, but the estimated values were very nearly unity; Japan’s ranged from 1.006 to 1.024 and the American values were 1.008 to 1.031, depending on the specification. Also in this study, estimates of returns to scale fell in the range of 0.97 to 0.998. Such values suggest that the Cobb–Douglas assumptions are close enough to reality for many purposes, especially for richer countries (Karagiannis et al. 2004: 9). Representative of a simplified approach is Collins and Bosworth (1996: 154– 5), who assume that capital’s share of GDP is 0.35 for all 88 countries in their study of productivity. Their review of the literature and in a 2003 update suggests that the share tends towards constancy somewhere between 0.3 and 0.4, which warrants use of a Cobb–Douglas framework. Another prominent study came to similar conclusions: The Cobb–Douglas assumption of stable income shares is a good one: first, we find no systematic tendency for country labour shares to vary with real GDP per capita or the capital–labour ratio. Indeed, most estimated labour shares lie between 0.6 and 0.8, and the average value of the labour share is 0.65, similar to that observed in the United States and other industrialized countries. Second, the time series of labour shares by country tend to be quite stable, with no systematic tendency to rise or fall over time. Bernanke and Gurkaynak (2001: 25–6) The capital shares of Japan and the United States, calculated from each country’s national accounts, are shown in Figure 8.10. Neither shows a time trend from 1950 to 2005, although both have cyclical swings responding to business conditions. In fact, the American data back to 1929 show no trend. The Japanese and American capital shares of GDP for 1950–2005 were 0.36 and 0.32 respectively. These figures were computed as 1.0 minus the labour share, with adjustment made for the share of labour in private unincorporated enterprise. The analytical issue of unincorporated enterprises is to separate out the labour portion of their income. I assumed that the share of labour income in entrepreneurial income was the same as in the corporate sector (Gollin 2002: 472).
Capital investment and rates of return 119 45 Japan Capital share (%)
40
35 US 30
25 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005
Figure 8.10 Capital share of national income (Japan, US; 1950–2005, %).
In contrast to the studies that assume Cobb–Douglas functions, Stanford economists Jong-Il Kim and Lawrence Lau formally test the usual simplifying production function assumptions for a group of advanced and developing Asian countries. According to their results, returns to scale are not constant, markets are not homogeneous, and the elasticity of output with respect to capital declines with the capital intensity of production (Kim and Lau: 1994: 249). Below, I extend Kim and Lau’s elasticities estimated for 1990, the last year of their sample, to 2005. In principle, therefore, we have at least two approaches to obtaining the elasticity of output with respect to capital: capital share of national income, generally derived as the remainder after the labour share has been estimated; and values obtained from full production function estimations. Capital stock Two approaches are commonly used to estimate the national stock of capital: surveys of firms and perpetual inventory methods that add each year’s investment to the depreciated stock of capital. The survey method can produce inconsistent estimates because of incomplete coverage and problems encountered in answering complex questionnaires. For the perpetual inventory approach, two depreciation methods are common: one sums up past investment while continually depreciating each vintage of capital; a second method postulates fixed asset lives with constant productivity throughout the life of the item. The Penn World Table (PWT) uses the perpetual inventory method with annual depreciation (Summers et al. 1994). The PWT compilers use annual depreciation rates for structures, machinery and transportation equipment of 3.5 per cent, 15 per cent and 24 per cent respectively. Angus Maddison (1995b) has favoured the second method in his attempts to create standardized capital estimates for several countries. Maddison’s standardized estimates assume asset lives that approximate
120 Capital investment and rates of return as closely as possible those in the United States: structures are given 39-year lifetimes, and equipment 14 years. To test the sensitivity of the rate of return figures to different capital assumptions, I use both the Maddison and the PWT assumptions below. The Japanese national accounts publish nominal stock figures and the American accounts estimate real as well as nominal quantities; they both use a perpetual inventory method based on modified PWT methods. However, a deficiency of all perpetual inventory methods is that premature scrapping or writing off of capacity is not recognized. Therefore, these methods will tend to show higher than actual stocks of capital in times of unusually active reductions of productive capacity. It is particularly easy to calculate the capital stock according to the Maddison scheme: simply add up the real investment for the number of years corresponding to the lives of the different asset classes. The stock of capital is the cumulative value of past investment still in existence. The problem with this approach is that it requires a stream of data at least as lengthy as the longest-lived asset. The other problem is that if the productivity does decline over time, this method does not take the deterioration into account; this deficiency is not an issue if investment is stable, but if it grows rapidly – as is common in fast developing economies – the capital stock figure may be distorted. The PWT-based capital stock estimates are 50–60 per cent of the value compiled according to the Maddison method from the same data. Consequently, rates of return computed from the PWT capital figures will be roughly twice as high as the Maddison figures.
70 60
PWT; 0.35 elasticity
50 PWT; Kim & Lau elasticity 40 30 20
Maddison; 0.35 elasticity
10 Maddison; Kim & Lau elasticity 0 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Figure 8.11 Returns on Japan’s non-residential capital stock with alternative depreciation and elasticity assumptions, 1910–2004.
Capital investment and rates of return 121 Alternative estimates of returns In Figure 8.11, I plot four different rates of return using two elasticity estimates and two capital series based on the Maddison and PWT depreciation assumptions. The elasticities include the variable estimates of Kim and Lau to the constant figures used by Collins and Bosworth (1996). Several points can be drawn from these alternative renderings of Japan’s real rate of return on aggregate capital. First, and most obvious, returns have fallen sharply over the years as capital has accumulated. Second, as expected, the PWT depreciation assumptions lead to lower levels of capital and higher returns than do the Maddison fully productive life assumption. Third, the rate of decline gradually decelerated. Fourth, the Kim and Lau elasticities produced the highest estimates earlier and lower returns later because their procedure allowed the elasticity to vary over time. With these considerations in mind, in most estimates above I use the PWT capital depreciation assumptions because they allow longer time series than do the Maddison assumptions; constant elasticities were chosen because they seemed to be consistent with the underlying data. As seen in Figure 8.11, the qualitative results are almost identical regardless of the specific assumptions.
9 Structural change in the Japanese economy
How evenly are growth and productivity gains shared across Japanese industries, and how have the distributions of growth and productivity changed over time? Which industries have contributed to efficiency gains? Were the 1990s, the decade of so-called stagnation, different in kind from earlier economic eras? How does change in Japan compare with the United States? These and other questions will be addressed in this chapter.3 One strong finding emerges from analysing these questions: economic development used to be broadly shared across sectors, but gradually became more dispersed; after the 1990s, only a handful of industries contributed to the growth of the entire economy, whereas a significant number declined.
GDP growth across industries The main question is not which industries grew fastest or slowest, but which ones contributed the most to national growth or to higher productivity. To describe the evolution of Japan’s output, I have adapted an approach developed by University of Chicago’s Arnold Harberger, who analysed productivity growth (Harberger 1998: 6). Harberger’s main analytical device is a graph that plots the cumulative percentage of growth contributions, ranked from largest contributor to smallest, on the vertical axis, and the cumulative share of GDP accounted for by each industry on the horizontal axis. An industry’s contribution to GDP growth, git, is defined as: git = 100*((Xit – Xi (t – 1))/(Yt – Y(t – 1))) where Xit is GDP originating in industry i in period t and Yt is total GDP at time t. The plot ends at 100 per cent on both axes; that is, 100 per cent of cumulative GDP by industry accounts for 100 per cent of observed growth. The resulting graph is similar to the Lorenz curve used to measure income inequality across population groups (Lorenz curves, however, plot the cumulative percentage of income by group, listed from lowest income to highest, the reverse of the ranking used here.) In such plots, the 45-degree line represents equality: each group’s income share is equal to its population share.
Structural change in the Japanese economy 123 Japan’s national accounts include a 42-industry breakdown of gross domestic product4 (ESRI 2006). Figure 9.1 plots the cumulative percentage of each industry’s contribution to 10-year GDP growth against the cumulative percentage of GDP by time period. Growth in the 1970s was the most equally distributed, demonstrated by its proximity to the 45-degree line of perfect equality. The 1980s decade was only slightly less equal. The 1990s started to see differentiation across the economy marked by a new phenomenon, negative growth among a significant number of industries. Until the 1990s, only a few industries actually declined over a 10-year period. This changed as growth slowed. In 1990–2000, real output declined in 21 industries, accounting for 40 per cent of GDP; 2000–2005 saw a slight moderation of hard times, with 18 experiencing falling output, representing one-third of total output. On the growth side of the ledger, two industries, electrical machinery and business services, accounted for almost two-thirds of the aggregate increase between 2000 and 2005. Table 9.1 shows the industry contributions to aggregate GDP growth. To clarify what is being measured here, a very fast-growing industry, such as precision instruments, with 16 per cent annual growth in the 1970s, made only a small contribution to aggregate growth because its share of the total was a tiny 0.4 per cent; wholesale trade, in contrast, grew at a slower 10 per cent rate, but accounted for a larger 3.4 per cent of 1970 GDP; consequently, it was a greater factor in overall economic expansion. A commonly used method to measure the inequality of a distribution is the Gini coefficient. As conventionally measured, it is the area between the 45-degree line and the distribution curve divided by the area of the triangle formed by the 45-degree line and axis. If all members of a sample receive equal shares, the Gini 140
Cumulative % of GDP Growth
1990–2000 120 2000–05 100 80 45 deg. line
60 1970–80
40
1980–90
20 0 0
10
20
30
40
50
60
70
80
90
100
Cumulative % of GDP
Figure 9.1 Cumulative industry percentage of GDP growth versus cumulative industry percentage of GDP (1970–2005, 42 industries).
9.7
Growth contribution 10.9
Growth contribution 9.4
Machinery Retail trade Chemicals
Transport equipment Public administration Other manufacturing
3.7 3.3 3.2
2.8
Transport equipment Government. services
Business services 5.0
5.2
2.6
2.8
5.4
Electrical machinery Transport
6.7
Community services Finance and insurance Public administration Other real estate Personal services Food products, beverages Machinery
2.0
2.6
2.7
4.3 4.2 4.0
5.1
7.5
8.2
7.6
Construction
Construction
8.0
Renting of 9.3 dwellings Finance and 9.2 insurance Personal services 8.4
Industry Wholesale trade
1980–90
Retail trade
Business services 8.2
Industry Renting of dwellings Wholesale trade
1970–80
Table 9.1 Industry contributions to GDP growth, 1970–2005 (%)
19.0
23.2
Transport equipment
Govt. services
Govt. utilities
Community services Public administration Finance and insurance Retail trade Electricity supply Chemicals
1.6
1.7
2.2
4.1 3.2 2.3
6.4
6.7
8.8
Renting of 18.2 dwellings Communications 12.5
Electrical machinery Wholesale trade
Growth Industry contribution Business services 24.3
1990–2000
1.9
4.5 4.1 1.9
4.9
5.1
6.5
10.5
10.6
Other household 1.6 services Other 1.2 manufacturing
Govt. utilities
Community services Renting of dwellings Public administration Finance and insurance Transport equipment Electricity supply Wholesale trade Machinery
Communications 12.2
Growth Industry contribution Electrical 39.8 machinery Business services 25.0
2000–2005
2.5
1.5
1.6
1.8
Pulp, paper products Precision instruments Goverment. utilities Fabricated metal products
Precision 0.6 instruments Rubber products 0.4
Wearing apparel 0.3
0.8
0.7
0.7
0.8
0.7
0.8
0.8
0.9
0.9
0.9
1.2
Government utilities Non-metallic mineral products Other household services Private education
Government services Pulp, paper products Agriculture
Community 1.6 services Publishing, 1.4 printing Other real estate 1.2
Fabricated metal 2.0 products Communications 1.9 Electricity supply 1.7
0.8
0.9
Non-ferrous 0.8 metals Private education 0.8
Other household services Wearing apparel
Other 1.2 manufacturing Communications 1.1
Gas and water supply Electrical machinery Chemicals
Transport 2.3 Electricity supply 1.9
Iron and steel
Wood, wooden products
Non-metallic mineral products Precision instruments Pulp, paper products Textiles
Non-ferrous metals Gas and water supply Other manufacturing Petroleum, coal products Leather, fur products Fabricated metal products Rubber products
Other household services Private education Food products, beverages Forestry
–0.7
–0.6
–0.5
–0.5
–0.5
–0.4
–0.4
–0.3
–0.1
0.0
0.2
0.3
0.6
1.0 0.9
1.4
1.1
0.2
0.2
0.4
–0.3
–0.3
–0.2
–0.1
0.0
0.0
Pulp, paper –0.3 products Non-metallic –0.6 mineral products
Fishing
Leather, fur products Iron and steel
Forestry
Precision instruments Mining
Rubber products 0.1
Gas and water supply Non-ferrous metals Agriculture
Private education 0.7
Govt. services 0.9 Personal services 0.9
Chemicals
0.5 0.5
Fishing Furniture
Iron and steel Fishing
Mining
Growth of all industries
–0.3 –0.3
–1.8
4.6
–0.2
–0.1
0.2
Gas and water supply Textiles Non-ferrous metals Furniture Food products, beverages Wood, wooden products Leather, fur products Petroleum, coal products Forestry
Industry
1980–90
4.0
–0.2
–0.1 –0.1
–0.1
–0.1
0.0
–2.4
–1.8
–1.5
Growth of all industries
Construction
1.4
–16.0
Wearing apparel –3.3 Other real estate –3.5
Machinery
Publishing, printing Fishing
–1.3
Furniture
0.0
–0.9 –1.0
–0.7
Growth contribution
Personal services –1.1 Agriculture –1.1
Iron and steel Transport
Mining
Industry
1990–2000
0.2 0.1
0.3 0.3
0.3
Growth contribution
Growth contribution
Growth of all industries
Construction
Petroleum, coal products Fabricated metal products Wearing apparel Retail trade
Furniture Wood, wooden products Textiles Publishing, printing Food products, beverages Transport
1.4
–11.6
–2.6 –5.6
–2.6
–2.3
–1.8
–1.3
–1.0 –1.3
–0.8 –0.9
Other real estate –0.7
Industry
2000–2005
Note: Industry growth contributions defined as: 100*((Xit – Xi (t – 1))/(Yt – Y(t – 1)))) where Xit is GDP originating in industry i in period t and Yt is total GDP at time t.
Agriculture Publishing, printing Petroleum, coal products Growth of all industries
Wood, wooden products Leather, fur products Forestry
Mining 0.3 Non-metallic 0.3 mineral products Textiles 0.2
0.5
Growth contribution
Rubber products
Industry
1970–80
Table 9.1 Continued.
Structural change in the Japanese economy 127 coefficient is zero because the distribution curve is identical to the 45-degree line. At the other extreme, if a single member received the entire amount, the Gini would be 1.0 because the one recipient would have 100 per cent and the distribution curve would comprise the triangle between the origin, vertical axis, and the horizontal line from the 100 per cent mark. For the samples described in Figure 9.1, the standard Gini coefficient interpretations must be modified because of the existence of negative values, which are not allowed in the standard Gini approach. However, a modified Gini coefficient can be calculated with much the same interpretation as the standard version. As with the standard Gini coefficient, the modified version calculates the area between the curve and the 45-degree line; again, as with the standard approach, that area is divided by the total area above the 45-degree line and the line representing 100 per cent. The difference between the modified Gini and the standard one is that the ratio can go above 1.0 because the curves may range above 100 per cent because of negative observations. Still, the interpretation remains the same. Higher coefficients represent greater dispersion or inequality among the sample observations. I calculated modified Gini coefficients for industry growth contributions over successive 10-year periods, 1970–80 to 1995–2005. The results are shown in Figure 9.2. Inequality rose after 1990, doubling between 1990 and 1995; it then shot up over the next 10 years, reflecting the bulging distribution curves in Figure 9.1. Since 1970, the industries that have done the most for the economy have not been those that might come to mind when considering the Japanese economic
1.6 1.4
Gini coefficient
1.2 1.0 0.8 0.6 0.4 0.2 0.0 1980
1985
1990
1995
2000
2005
Figure 9.2 Modified Gini coefficients of industry growth contributions over preceding 10year period, 1980–2005.
128 Structural change in the Japanese economy miracle. At most, only two manufacturing industries appear in the top-ten list in any decade. Note that electrical machinery, which made its way to the top of the list with annual growth of 18 per cent in the 1980s followed by 9 per cent in the next 10 years, includes semiconductors and computers. Manufacturing made up a much larger share of the losers; four or five manufacturing industries brought up the bottom of the rankings in all periods. The declining industries include the main representatives of the old economy: iron and steel, textiles, wearing apparel and machinery. Clearly, the decline of manufacturing has been a central theme of the Japanese economy for the past 30 years. Harberger (1998: 6) noted that productivity growth in the American economy was highly concentrated in a few industries and that these industries were very different from decade to decade. The relative contributions of those industries with the greatest influence can be gauged by the steepness of the Lorenz curve at the origin. When a few industries contribute more to growth than implied by their relative size, the curve rises quickly from the origin. As seen in Figure 9.1, the concentration of GDP growth in a small number of industries in Japan began only in the 1990s. A possible reason for the increasingly skewed character of Japan’s growth is that the growth rate itself has slowed. If the overall trend slows, any given increase of sectoral output will be a higher percentage of a smaller total aggregate growth; at the extreme, if aggregate growth were zero, positive increases in any single industry would be an infinite percentage of the total. One method for dealing with this issue is to consider an industry’s growth as a fraction of base-year GDP rather than as a percentage of the overall growth rate. This approach normalizes industry change with respect to the scale of the economy. This measure is defined as: 100*((Xit – Xi (t – 1))/Y (t – 1))/n where Xit is GDP originating in industry i in period t, Yt is total GDP at time t and n is the number of years in the period. (Arithmetic rather than compound rates are used here because the sum of industry compound growth rates do not add up to aggregate compound growth.) Figure 9.3 shows cumulative industry growth as a percentage of GDP against the cumulative GDP share. The curves terminate at each period’s aggregate growth rate on the vertical axis. A measure of inequality on this type of chart is the area between the plotted curve and an imaginary line drawn from the origin to the right-hand end of each curve. In general, the more curved or convex a curve, the greater the amount of inequality it represents. The 1970s and 1980s curves are almost straight lines; growth was evenly spread throughout the economy and most sectors participated in economic expansion to the same degree. As in Figure 9.1, the most convex curves are those of the 1990s and later. What is notable about the more recent curves is the presence of significant numbers of losers; if there were no declining industries in 1990–2000, for example, aggregate annual growth would have been 2.1 per cent rather than 1.4 per cent.
Structural change in the Japanese economy 129 Cumulative Industry GDP Growth (%)
6 1970-80 5 1980-90
4 3
2000-05 2 1990-2000 1 0 0
10
20
30
40
50
60
70
80
90
100
Cumulative % of GDP
Figure 9.3 Cumulative industry GDP growth (annual rate) as percentage of base-year GDP versus cumulative industry share of GDP (1970–2005, 42 industries).
Analytical concerns Might differentiation be a function of the number of industries into which the economy is divided? If the economy were cut into finer bits, there would be a greater probability of more extreme outcomes. On the other hand, each of the smaller bits would make a smaller contribution to the total. Figure 9.3 examines these possibilities. The Japan industrial productivity (JIP) database breaks down GDP into 84 industries from 1970 to 1998 (Fukao et al. 2003). In addition to the JIP’s 84 industries covering all of GDP, the Ministry of Economy, Trade and Industry publishes production indices for up to 105 manufacturing industries. Since manufacturing is 20–25 per cent of total GDP, dividing it so finely increases the possibility for even greater variability in growth rates. Figure 9.4 shows the growth contribution curves for each of the three databases over the same period, 1980–90. The curve of 42 industries included six with negative growth, accounting for 9 per cent of GDP; the 84-industries sample includes 16 with negative growth, covering 12 per cent of the economy; and the 106 manufacturing industries included 23 declining ones, representing 14 per cent of the manufacturing sector. Finer breakdowns yield smoother curves with greater concavity, a conclusion that must be noted when comparing samples of different sizes Another concern is that the apparent change of outcomes in Japan after 1990 may be an artefact of slowing growth, as suggested earlier. The analytical question is whether there is greater variability across industries in more recent periods than earlier, after taking account of economic deceleration. To test whether the increased variability of growth across industries may be related to the simple arithmetic of slower growth rather than to any change in
130 Structural change in the Japanese economy Cumulative % of output growth
120 106 manufacturing industries
100 80 60
84 GDP sectors 40 20 42 GDP sectores 0 0
10
20
30
40
50
60
70
80
90
100
Cumulative % of output
Figure 9.4 Cumulative industry percentage of total growth versus cumulative industry percentage of output, 1980–90.
behaviour, I constructed a simulated sample of Japan’s experience from 1970 to 2005. I started with a random sample of 42 industries based on a lognormal distribution with the same mean and variance of GDP as the actual 1970 sample. Each simulated industry then grew according to a linearly declining trend growth rate similar to the average deceleration of GDP, declining from 6 per cent in 1970 to just under 1 per cent in 2005; a random element was added to the trend rate for each industry-year observation with a mean of zero and a variance equal to the total sample one-year growth rate variance. The structure of the simulated economy ended up remarkably like the real one. The distribution of simulated industry size at the end of the 35-year run is shown in Figure 9.5, along with the actual distribution. Although the simulation mimicked the change in industry structure, real industries demonstrated longer periods of extended growth and decline than did random ones. For example, the maximum 35-year annual average growth rate of an industry in the simulation was 4.0 per cent versus the 8.0 per cent rate for the very real electrical machinery industry. Gini coefficients for the real and simulated samples are shown in Figure 9.6. As was considered likely, the Gini numbers do increase with declining growth rates in the simulation, doubling from 0.3 at the beginning to 0.6 at the end of the simulation. The actual Ginis follow the simulated trend until the 1990s, and then diverge. Some of this increase must be caused by the arithmetic of slower growth, but a major part seems to be attributable to the emergence of industries exhibiting longer and stronger than average growth, and others in substantial decline. Despite these results, there remains a nagging suspicion that slower growth is dictating the effects. A different test of this possibility removes the general influence of aggregate change while preserving the remaining variability at the industry level. If each industry’s growth is made up of two parts – the industry’s
Structural change in the Japanese economy 131
Industry GDP
100,000
10,000 Simulated
1,000 Actual
100 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 Industry rank by size
Figure 9.5 Size distribution of 42 actual and simulated industries (2005, billion 1995 ¥, log scale).
1.6 1.4 1.2 Gini coefficient
Actual 1.0 0.8 0.6 0.4 Simulated
0.2 0.0 1980
1985
1990
1995
2000
2005
Figure 9.6 Modified Gini coefficients of industry growth distribution over preceding 10year period, actual and simulated, 1980–2005.
share of aggregate growth plus an independent component – the relevant question is whether the industry components are becoming increasingly varied or unequal over time.
132 Structural change in the Japanese economy This model can be stated as: ∆Xit = (Xi (t – 1)/Y(t – 1)) * ∆Y + eit where, as above, Xit is GDP originating in industry i in period t and Yt is total GDP at time t; eit is the variable component at the industry level. This formulation says that an industry’s share of aggregate growth equals its base-year share of total output. It is then a simple matter to estimate the industry-specific component by calculating the detrended change in industry output: eit = ∆Xit – (Xi(t – 1)/Y(t – 1)) * ∆Y
2.17 2.17 2.02 2.28 2.35 3.13 2.91 2.94 3.12 2.92 3.31 3.41 3.03 2.82 2.42 3.96 3.82 3.18 3.09 2.68 2.77 2.40 3.76 4.35 3.49 4.74
The standard deviation of eit, actual minus expected industry growth, is a gauge of industry volatility after removing the influence of aggregate growth. The results of this calculation are shown in Figure 9.7, which plots the standard deviation of industry-specific variability (eit, the detrended change of industry GDP) over successive 10-year periods. The figure shows both the actual and simulated experience. The variability of industry growth was fairly stable until the mid-1990s, when a distinct upturn occurred. The standard deviation of simulated industry growth increased slowly over the entire period, perhaps responding to the decelerating aggregate trend. The finding that change became more varied over time and that growth became increasingly heterogeneous is not an artefact of Japan’s inevitable slowdown, but a real change in the Japanese economy.
6.0 5.5
Actual
5.0 Standard deviation
std dev
4.5 4.0 3.5 3.0 2.5 Simulated 2.0 1980
1983
1986
1989
1992
1995
1998
2001
2004
Figure 9.7 Standard deviation of 10-year change of detrended industry GDP (trillion ¥, 1980–2005).
Structural change in the Japanese economy 133
Total factor productivity In his study, Harberger looked at total factor productivity (TFP) change across industries. Through an insightful shift of emphasis, he gave a new label to TFP, calling it ‘real cost reduction’. This simple recasting transforms our appreciation of what is going on. Instead of focusing on technology or technical change or R&D, real cost reduction ‘makes one think like an entrepreneur or a CEO or a production manager. . . . Labels do not change the underlying reality, but they may change the way we look at it’ (Harberger 1998: 3). Recasting TFP as real cost reduction points the way to calculating it for each industry in an additive manner. The calculation of real cost reduction from the TFP numbers is straightforward. Real cost reduction for industry i between time (t – 1) and t is: RCRit = – ((1/TFPit)/(1/TFPi(t – 1)) – 1)* Xi (t – 1)/Y(t – 1)
Cumulative % of Real Cost Reduction (TFP) Growth
where RCRit is real cost reduction for industry i at time t as a percentage of baseyear GDP, TFP is total factor productivity, X is industry value-added and Y is aggregate GDP. Since TFP is defined as rising when the cost of output falls, the cost reduction factor is the inverse of TFP. If we consider cost reduction as a positive number, then it is necessary to put a minus sign in front of the above definition. A joint Japanese government–university research group subdivided the national accounts to analyse productivity change. The JIP database includes TFP estimates from 1970 to 1998 for 83 industries (the catchall ‘not elsewhere classified’ was not included in the productivity calculations) (Fukao et al. 2003). I dropped the housing sector, which the JIP compilers identified with imputed house rent. 0.6 1990-98 0.4 1980-90
1970-80
0.2 0.0 0
10
20
30
40
50
60
70
80
90
100
-0.2 -0.4 -0.6 -0.8 Cumulative % of Real Cost Reduction (TFP)
Figure 9.8 Cumulative real cost reduction as percentage of base-year GDP versus cumulative industry GDP (1970–98, 82 industries, %).
134 Structural change in the Japanese economy Figure 9.8 plots the cumulative value of real cost reduction by industry, from largest to smallest reduction, as a percentage of cumulative industry GDP. Cost reduction is more concentrated in fewer industries than is GDP growth; rising efficiency occurred in industries representing just one-third to one-half of total output. Moreover, the weighted sum of cost reductions was negative in the 1970s and 1980s, despite strong aggregate growth in those years; productivity gains turned positive only in 1990–98, when growth slowed. If readers think that there is something strange about the Japanese economy where the majority of industries exhibit deteriorating efficiency, they should consider Harberger’s description of similar outcomes as characteristic of American manufacturing from 1948 to 1985 (Harberger 1988: 9). Harberger considers two kinds of growth processes that he likens to mushrooms and yeast. Yeast causes bread to expand evenly, like a balloon filling with air; mushrooms pop up in scattered locations. The transition in Japan to differentiated GDP growth as well as American and Japanese TFP growth across industries fit the mushroom analogy; a small number of industries pop up that exhibit very fast productivity change and output growth. Critics of American productivity trends note that most recent growth has been concentrated in computers and information technology; in criticizing the narrow distribution of TFP gains, they ignore the historical pattern of development that typically has been dominated by a select vanguard of leading sectors whose members change over time. Harberger drew a number of conclusions from his analysis of American productivity data: a small fraction of industries can account for all aggregate TFP growth; the complementary fraction of industries contains winners and losers, which cancel each other; the losers are a very important part of the picture most of the time and contribute greatly to the variability of aggregate performance; and there is little evidence of persistence of the leaders (Harberger 1988: 10). Harberger’s conclusions concerning the United States can be applied without revision to Japan.
Comparisons with the United States The US Department of Commerce publishes real GDP originating in 66 industries from 1977. The American Lorenz-type growth curves are very similar to Japan’s until the 1990s, when Japanese growth became much more highly skewed. To examine this further, I chose two decades with the same growth rates (2.2 per cent) in both countries – 1988–98 for Japan and 1985–95 for the United States. Figure 9.9 shows the growth distribution curves for Japan and the United States. Japan’s industry growth contributions during the decade of the 1990s stand out for being more heterogeneous than America’s even though the larger number of American industries would tend to produce greater inequality. The greater growth disparities among Japanese industries visible in the figure are borne out by the modified Gini coefficients: 0.66 for Japan and 0.44 for the United States. To account for the effect of growth on the shape of employment curves, I chose two 10-year periods in each country with about the same employment growth rate
Structural change in the Japanese economy 135 120 Cumulative % of GDP growth
0.00 12.50 16.85 22.72 33.92 39.00 42.16 47.01 48.31 51.53 56.37 58.07 59.93 60.47 61.31 65.69 66.70 68.00 69.24 70.93 71.72 72.74 73.28 73.72 74.07 74.81 75.08 76.06 76.53 76.81 76.93 77.27 78.28 78.51 78.79 78.84 79.59 79.63 80.03 80.43 80.90 80.95 81.06 81.18 81.37 81.57 82.06 82.63 83.21 84.61 85.90
Ind % of Growth, US, 1.2% growth, Gini: .47
100
Japan 0.00 14.51 1 80 27.47 2 US 38.85 3 60 49.69 4 58.61 5 66.50 6 40 73.40 7 79.35 8 20 85.06 9 90.33 10 0 95.10 11 0 12 10 20 30 40 50 60 70 80 90 100 98.64 Cumulative % of GDP 101.87 13 Figure 9.9 Cumulative 104.44 14 industry percentage of GDP growth versus cumulative industry percentage 106.78 15 of GDP (Japan: 1988–98, 42 industries; US: 1985–95, 66 industries). 108.47 16 109.79 17 110.99 160 18 112.10 19 113.17 140 20 114.11 21 115.01 120 22 115.89 23 Japan 116.75 100 24 117.55 25 80 118.21 26 118.81 27 60 US 119.37 28 119.81 29 40 120.24 30 20 120.68 31 121.10 32 0 121.41 33 0 10 20 30 40 50 60 70 80 90 100 121.61 34 Cumulative % of employment 121.73 35 121.79 36 industry percentage of employment growth versus cumulative Figure 9.10 Cumulative 121.83 37percentage of employment (Japan: 1980–90, 24 industries; US: industry 121.86 38 65 industries). 1995–2005, 121.85 39 121.83 40 (1.2 per cent): 1980–90 in Japan and 1995–2005 for the United States. Japan’s 121.71 41 national accounts 42 publish employment data for 26 industries; in similar fashion 121.55 121.38 43 to the GDP breakdown, 65 American industries have employment data. The com121.16 44 parison shown in Figure 9.10 is revealing, given the alleged stability of employ120.94 45 120.69 46 120.43 47 120.16 48 119.81 49 119.28 50 Cumulative % of employment growth
of US yees, 05, 65 ries
136 Structural change in the Japanese economy ment in Japanese companies. We might expect even less variation over time of workers across industries than we found for GDP; however, employment change is more highly skewed than GDP. The Japanese curve lies outside the American at every point, despite the bias towards heterogeneity from the greater number of American industries. The Gini coefficients calculated from these samples confirm the visual inference: the Japanese coefficient was 1.06 compared with the American value of 0.47. Moreover, Japanese employment growth continued to become more unequal throughout the 1990s. A good deal of the Japanese employment story can be told by just two sectors – agriculture and services; the decline of the former and rapid growth of the latter account for much of the dispersion from 1970 to 2000. Construction and trade (wholesale and retail) were the other main movers on the plus side, while textiles lost jobs over much of the period.
Conclusions The Japanese economy became more heterogeneous in the last decade or so of the twentieth century. The 1990s are a convenient marker, although some changes began earlier and others took shape later in the decade. The lingering rigidities imposed by the post-war model of economic development finally gave way to the cumulative pressures of slower growth, deregulation and foreign entry. In short, one could argue that pent-up change finally broke loose from the frictions of earlier policies and practices, forced by dislocations in the underlying economic foundations. These changes are evident in the increased amount of variations across industry.
10 The privatization of Japan’s public corporations
Privatization of large government undertakings has been a central theme in the deregulation and liberalization of the Japanese economy. Privatization of the government’s highway corporation and postal system were at the top of Prime Minister Junichiro Koizumi’s agenda in his attempt to redesign Japanese politics, the LDP and the economy. By most measures, the actual privatization measures were drastically weakened versions of original plans. Given the disappointment with the progress made in this campaign, it may be worthwhile to examine privatization in the 1980s under the aegis of Prime Minister Yasuhiro Nakasone. The results of the earlier effort were mixed. On the plus side, the former government-owned enterprises were transformed into more efficient, profitable and competitive organizations than they were in 1980. However, only a handful of the companies were fully privatized, and the enormous debt of the nationalized railroads was not reduced. The same kind of political barriers that prevented Mr Koizumi from fully realizing his goals also placed constraints on the previous attempts. Ministries, agencies and bureaucrats often barred the way toward fuller privatization, chiefly because they did not want to lose control over the industries and companies that they supervised. The concept of industrial planning under government management that had survived from the 1930s was a continuing factor in the extended and partial process of privatization.
Japan’s public economic enterprises Historical background of privatization Many of Japan’s state-owned industries had military or strategic economic origins. Government ownership of railroads and telecommunications was asserted to be necessary for military coordination as well as for industrial development. The post-Second World War development of aviation carried these same rationales, although the military reason was muted in Japan compared with its use in other countries’ justification of state-owned airlines, including the United States. Japan’s salt and tobacco monopoly is another matter; those products were the traditional targets of state control, mainly for revenue purposes, going back to
138 The privatization of Japan’s public corporations the Roman Empire’s monopoly of salt. Tobacco, from an early stage of its introduction into Europe, also fell under state monopolization in the 1670s. Salt and tobacco shared similar characteristics: they were widely used and absorbed small shares of personal income; in addition, the quantity consumed was insensitive to price, especially if its use was addictive. Under these conditions, large profits could be earned by raising prices well beyond the marginal cost of production, but only if competitors could be barred from entering the market. Thus, if a government could establish the monopoly for itself, it could raise prices and enjoy the revenues. Government ownership of railroads, telecommunications, airlines and the salt and tobacco monopoly came under intense scrutiny in the early 1980s because of debt, inefficiency and declining competitiveness, especially compared with performance in private companies in Japan and other countries. A drive to privatize these industries was pushed by business interests and popular sentiment, but was opposed by those who had profited from state ownership or who would be penalized by privatization and market liberalization. The political power of small groups and their supporters among Diet members as well as government agencies that would lose authority over their supervisory charges blocked the complete sale of government corporations to the public. After 25 years, only three railroads and the government-owned airline were sold off completely. The government still owns substantial shares of the telecommunications company and the tobacco company. No shares in three non-privatized passenger rail lines and a freight rail company have been sold to the public. The public corporations included the largest employers in Japan: Japanese National Railways (JNR) and Nippon Telegraph and Telephone (NTT). NTT’s privatization transactions constituted the largest initial and secondary offerings ever up to that time. In addition to these two companies, the Japan Monopoly Corporation – the only authorized buyer and seller of tobacco, tobacco products and salt – was converted to a private company (Japan Tobacco) and partially sold off in the 1990s. Japan Airlines Co., Ltd. was founded as a private company in 1951; under a 1953 law allowing government ownership, the company was reorganized as a public corporation, with the government holding 49 per cent of the shares (later diluted to 34.5 per cent). In November 1987, the government divested all its shares as part of the broader drive to privatize public corporations. Although these conversions were large-scale events and required major political efforts to accomplish, Japan’s nationalized production capacity was not as large a part of its economy as that of many other advanced or developing countries. In comparison with the United Kingdom, France and Germany, for example, there was remarkably little public enterprise in Japan in the late 1970s. Whereas the European countries had nationalized all or parts of their postal, electricity, gas, oil, coal, steel, shipbuilding and automobile industries – as well as railroads, telecommunication and airlines – Japan’s only government-owned production companies were in the last three sectors and the post office. Among the advanced economies, only the United States had a smaller share of nationalized industries than Japan (Calder 1990: 164).
The privatization of Japan’s public corporations 139 Japan actually had gone through several phases of nationalization and privatization in the years following the Meiji Restoration of 1868. In the 1870s, the government inaugurated a programme of state industries to serve several national objectives: to teach production techniques to private firms; to stimulate import substitution; and to foster industries important for military purposes. However, the government factories were a severe drain on government finances. Their poor profitability, the rise of an entrepreneur class and the government’s need for cash to finance military activities led to privatization starting in 1880. Military demands stimulated the next phase of government industrialization after the Sino-Japanese war of 1894–95. The following year the government created the Yawata Iron and Steel Works. Following the Russo-Japanese war of 1905, more than 90 per cent of the railway system was nationalized and the government embarked on a major investment in new lines and rolling stock. However, the large conglomerates (zaibatsu) strongly opposed government intervention in the economy; subsequently, most of the expansion of government-owned facilities occurred in Japan’s East Asian colonies, with private companies involved profitably as builders and managers. As war in Asia and the Pacific expanded in the 1930s, military industry again became the focus for nationalization. Yawata Steel combined with several smaller, private firms to become Japan Steel in 1934. By 1939, the MOF was a major shareholder in a wide range of strategic manufacturing firms and trading companies. As the war progressed, the government extended its reach to include the electricity generation business. In the immediate post-war years, price controls and other government interventions in the economy set the stage for still further takeovers of formerly private businesses – a process that consolidated wartime policies. This process reached its height when the Socialist Party briefly gained control of the government in 1947–48. The Socialists attempted to nationalize the coal industry following the practices of its European counterparts. (See Chapter 6 for details.) However, vigorous opposition from the more conservative parties in the Diet as well as mine owners forced considerable weakening of the proposed law. The Dodge Line, designed to bring discipline to the fiscal and monetary situation, first weakened and then in September 1949 ended this attempt at nationalization (Samuels 1987: 103). A conservative government, which owed no allegiance to the labour unions, took power at the same time that the post-war Occupation authorities implemented a new economic policy that included a balanced government budget for the first time in decades. The mounting losses of the state-owned companies were affected by the fiscal belt tightening. From 1948 to 1949, the number of government employees fell by more than 100,000. By 1951, most of the nationalized firms had been privatized. NTT and JNR, which had been bureaus in their respective ministries, were converted into independent public corporations, wholly owned by the government. However, as the conversion of state-owned production enterprises into private establishments was occurring, a new effort was under way to create another class
140 The privatization of Japan’s public corporations of government organizations with business functions – banking and financial organizations. These new ventures were financed by off-budget funds, unchecked by Mr Dodge and his Occupation staff; the financing was drawn from the postal savings system via the Fiscal Investment and Loan Program (FILP). Japan’s rapid growth in subsequent decades swelled postal savings and gave the government a growing trove to finance activities without the necessity of going before the Diet for budget approval. Borrowing an organizational form common in Japanese colonial administration in Manchuria, the government created scores of new so-called parastatal organizations. These were used to manage the enormous expansion of infrastructure in the 1960s and 1970s. They also provided a convenient vehicle to reward the friends of the LDP with construction contracts and to serve as retirement homes for senior civil servants. By 2001, there were 157 governmentaffiliated corporations and other special government institutions, which became the targets for a second round of privatization under Prime Minister Junichiro Koizumi. Rationale of the 1980s privatization drive Japan’s double-digit growth rate in the 1950s and 1960s led to even higher rates of tax revenue growth, which permitted rising expenditures without the fear of deficit spending. Additionally, money flowing into the FILP from household deposits in the postal system provided another source of revenues to back economic and political objectives. Post-war government planners, pushed at first by American Occupation officials, took steps to eliminate the perceived dangers of unbalanced public finances. The government had the authority to use bonds to finance investment, reflecting the theory that long-lasting construction projects that would yield benefits over many years could be financed by debt whereas current revenues should pay for current transactions. Until 1974, the bonds issued by Tokyo in any year did not exceed the value of public works expenditures. By 1975, however, in the aftermath of the first oil crisis and slowing growth, a recession-induced drop in revenues threatened to leave the government with insufficient funds to meet budgeted expenditures. To make it possible to issue debt to finance current outlays, the Diet passed the Fiscal Exception Law, which authorized, for one year only, deficit-covering bonds. That constraint, though, was elastic. The ratio of outstanding central government debt to GDP, which was 10 per cent in 1975, rose to 30 per cent in 1980 and 40 per cent in 1985. The regulated rigidities of the financial system made it difficult to fund the increasing bond issues. The MOF in 1980 warned that Japanese public finances were facing an emergency. Although it announced at every opportunity its intention of returning to sound fiscal practices, deficit-financing bonds were issued every year until 1989. One obvious way to deal with the mounting deficits was to raise taxes. However, the business community, under the leadership of the Japan Federation of Economic Organizations (Keidanren), fought both the introduction of new taxes
The privatization of Japan’s public corporations 141 such as a value-added tax or consumption tax and the raising of existing ones (especially corporate taxes). The business community argued that the first task should be to cut the fat in the government’s own administrative structure, and then, if necessary, consider tax increases. The business sector observed that, unlike its own cost-cutting approach to dealing with economic slowdown, the government – especially public corporations – had accumulated ever-increasing deficits, liabilities, personnel, excessive wages and profligate work practices (Choi 1991: 112). In 1980, the business sector adopted the slogan, ‘Administrative Reform without Tax Increase’. Despite the sloganeering, it was clear to analysts at the time that no amount of fat trimming and subsidy pruning could bring the budget into balance without reform of the tax structure. Nevertheless, administrative reform became the principal symbol for addressing the deficits (Choi 1991: 118). In March 1981, Prime Minister Zenko Suzuki established the second Provisional Commission on Administrative Reform (PCAR), modelled after a 1961 attempt to rationalize government. It was situated in the Administrative Management Agency. He appointed Keidanren leader, 84-year-old Toshio Doko, to chair the panel of independent experts consituting the PCAR. One of the forces preparing the way for administrative reform in Japan was the rhetoric and examples of deregulation and privatization in the United States and the United Kingdom espoused by Ronald Reagan and Margaret Thatcher. These conservative politicians and their free-market ideologies resonated with their Japanese counterparts. In addition, the country had been running deficits for five years by the time the second panel was appointed. The personality of the players also played a role in legitimizing reform. As head of Keidanren, Mr Doko brought along business support for his efforts. Just as important was his own public personality, that of an effective manager and entrepreneur who had successfully restructured the troubled Ishikawajima Heavy Industries. A lover of Chinese classical poetry, he had the image of someone who was personally austere; it was often said, for example, that he washed out his own clothes on business trips. Another important personality was Yasuhiro Nakasone. Mr Suzuki assigned the younger Nakasone, a potential rival, to head the low-status Administrative Management Agency. Nakasone seized on administrative reform as the vehicle for his own rise in the public’s eye and in intraparty manoeuvring. He took on the task with great energy and soon the prime minister joined him as a strong supporter of reform (Vogel 1996: 57). The PCAR chair and its eight commission members included three from business, two labour leaders, two former civil servants, one journalist and an academic. Most of the chiefs or deputies of the various subcommittees and working groups were Nakasone friends, acquaintances or colleagues. Before accepting the chairman’s post, Mr Doko sent a note to the prime minister requesting a document of agreement to four conditions: (1) the prime minister should actively promote recommended reforms; (2) financial reconstruction should not include increased taxes; (3) administrative reform should occur in both
142 The privatization of Japan’s public corporations the central and local governments; and (4) the recommendations should include solving the privatization of public enterprises and the deficits in the ‘three Ks’: the rice subsidy (komei), the national health insurance system (kokumin kenko hoken) and the Japanese National Railways (kokutetsu). A week later Mr Suzuki replied that he would stake his political future on realization of the reforms. The focus of administrative reform soon settled on the three Ks with JNR as the biggest single target. The railroad had become a symbol of government waste; it had been running enormous deficits for years, was clearly overmanned, and LDP leaders saw JNR reform as a way to break public sector unions as well as the Japan Socialist Party (JSP) (Choi 1991: 394). In 1980, subsidies to JNR came to around ¥676 billion ($5.2 billion) and the rail operations employed roughly 200,000 redundant workers, about half the total, using the performance of Japan’s private railways as a basis for comparison. (Dollar values are converted from yen at exchange rates of the period.) This situation had not arisen by accident, but was the result of implicit agreements between the LDP and JSP in the 1960s. To obtain JSP acquiescence on a wide range of issues, the LDP agreed to wage increases for public employees, which had an especially big impact on JNR because of its large workforce. For its part, the socialist party agreed to the creation of the Japan Railroad Construction Corp., which became the source of LDP patronage in routing and construction decisions. Throughout the 1960s and 1970s, JNR increased salary and benefit packages and expanded the network of unprofitable local lines, all of which led to the accumulation of deficits that the government automatically covered (Fukui 1992: 17). NTT, in contrast to JNR, was profitable, but nonetheless was overmanned and its employees overpaid. Costs were high and the business community as well as government leaders were concerned about the new combinations of computers and electronics that were being installed in the American telecommunications system and that were likely to bypass Japan with its highly regulated, government-owned monopoly. In addition, the fact that both NTT and the tobacco company were profitable was attributed to their monopoly status rather than to their managerial effectiveness. Therefore, profitability itself was not a bar to privatization. The second PCAR believed that the implementation failures of its predecessor 20 years earlier were based on the inability to gain acceptance by the government bureaucracy. The new commission, therefore, worked assiduously to win over the Finance Ministry. According to Steven Vogel (1996: 59), the MOF gave the commission its full support in exchange for a pledge not to threaten financial regulation. In order to assure implementation, the commission and the government organized a public relations campaign to educate and mobilize public opinion. At first, the topic of administrative reform was a barely audible theme in public discourse. Over the next few years, though, a flood of books, pamphlets, surveys and articles hammered home the message of the scale of the three Ks problem and the necessity for reform. After Mr Nakasone became prime minister in November 1982, he staged televised performances in which he could promote the reform idea to the urban middle class. By the end of 1982, budget discussions proceeded
The privatization of Japan’s public corporations 143 in an atmosphere of impending crisis. The campaign paid off with the general acceptance of major changes to reduce the deficit. In addition, the public propaganda pointed out the implications of administrative dysfunction and focused on the demoralizing effect of dependence on the state. It pointed to the ‘English’, ‘Dutch’ and ‘European’ diseases of economic malaise, slowing growth and rising unemployment. Members of the commission called for a Japanese way of handling the problem (Choi 1991: 132). The privatization of the three large public corporations and the deregulation of telecommunications became the core proposals of the commission. In particular, the reform of JNR was considered to be at the heart of the recommendations because it was the symbol of the administrative and financial ills of Japan. Thus, what had begun as a search for ways to deal with a growing fiscal deficit was transformed into a commitment for thorough administrative reform, and ended with proposals to privatize the big three public corporations. Privatization and abolition of other government enterprises, particularly in the financial sector, were taken off the table. Prime Minister Koizumi would raise the problem of the other government enterprises 20 years later.
Japan National Railways privatization Early JNR developments In 1872, the Japanese government built the trunk line between Tokyo and Yokohama, the country’s first railway. Private companies were permitted to build only regional lines. However, the government could not fully finance the investment required for its planned system and invited private companies to participate in construction and ownership of some segments. The authorities offered incentives for private investment in return for government supervision, cooperation with the police and military, and acceptance of a government option to purchase the lines in the future. By 1889, the length of private lines exceeded government-owned tracks (Fukui 1992: 4). The two wars that Japan fought in the following years, with China in 1895 and Russia in 1905, convinced the military that an integrated transportation system was essential for military industry and war fighting. Private industry fought against this intervention into economic affairs. In fact, the foreign minister, who was the son-in-law of the Mitsubishi company’s leader, resigned from the cabinet in protest. Nevertheless, the military lobbied strenuously for nationalization. The government approved the Railways Nationalization Law in 1906, under which it purchased private lines from 17 companies that could be integrated into a national network (Ishikawa and Imashiro 1998:10). The Railway Agency (converted to the Ministry of Railways in 1920) managed the Imperial Government Railway under an accounting system that was separate from the general budget. The basic structure of what would later become the JNR was thus established in the early part of the century. Since the highway system was primitive, the railroads had a virtual monopoly
144 The privatization of Japan’s public corporations over the shipment of passengers and goods. Opponents of nationalization maintained that lack of competition would result in the stagnation of improvements. They also argued that government management would lead to inefficiency, that excessive regulation would interfere with expeditious operations, that investment would increase the state’s financial burdens and that the large scale of the enterprise would be too complex for a government bureau to manage. All of these prophecies were fulfilled (Ishikawa and Imashiro 1998:10). Forty years later, in the immediate aftermath of the Second World War, the government requested the national railway to absorb a quarter-million repatriates from Japan’s colonies, including former workers of the South Manchurian Railways. Despite the departure of female employees who had joined the company during the war, the workforce increased from 537,000 to a peak of 610,500 between 1946 and 1948 (Ishikawa and Imashiro 1998:10). The Occupation authorities created public corporations for the railroads, NTT and the tobacco monopoly with the aim of increasing efficiency by separating operations from the government bureaucracy. The JNR public corporation was created in 1949. Its employees were given the right of collective bargaining, but were not allowed to strike. The latter restriction was deemed to be warranted by the centrality of railroads in national transportation. Highways remained inadequate and the war had destroyed much of the country’s shipping capacity. Despite the goal of an efficient, independent corporation, the JNR was given little discretion in managerial matters. Most people, including its employees and unions, continued to regard it as part of the government. A 30-year JNR veteran later wrote: ‘In reality, the public corporation in Japan was created so as to preserve the Japanese bureaucracy that existed before the war. Government structures were maintained and little effort made to change the old bureaucracy.’ (Ishikawa and Imashiro 1998:12). For the 20 years following the formation of the JNR public corporation, business grew steadily: passenger-kilometres went up 5 per cent annually and freight increased at a 3.6 per cent rate. On average, JNR operated profitably as it rebuilt its infrastructure and introduced a network of bullet train (Shinkansen) lines. Growth and profitability, though, masked lurking problems. Rationalization efforts dictated by the Occupation authorities had reduced employment to 442,000 by 1952. However, considerable overstaffing persisted, in large part due to the absorption of the repatriated workers, which also distorted the age structure of the workforce. The burden of retirement payments would appear in the 1970s and early 1980s when more than 20,000 workers would retire every year. By 1980, personnel costs amounted to 78 per cent of JNR’s operating revenues compared with 40 per cent for private railways (Fukui 1992: 15). As the nation’s road system improved and economic growth increased personal income, trucking and private automobile ownership soared. In addition, the shipping industry fully recovered and went on to become globally dominant. Aviation, too, developed rapidly; the introduction of the high-capacity Boeing 747 on the dense Tokyo–Osaka route brought competition to the Shinkansen lines. National economic growth decelerated from double-digit rates in the 1960s to
The privatization of Japan’s public corporations 145 just 3 per cent by the end of the 1970s. The JNR’s share of passenger traffic fell from 55 per cent in 1955 to 30 per cent in 1975 and close to 20 per cent in 1985. Automobiles, which carried less than 20 per cent of all passengers in 1955, were responsible for two-thirds of the total in 1985. The absolute number of passengers carried by the JNR peaked in 1974. The change in transportation modes hit freight even harder. The JNR took slightly more than half of the country’s freight transport in 1955. Freight volume peaked in 1970, and by 1985 the JNR’s share had slipped to under 5 per cent, displaced mainly by trucks and ships (Fukui 1992: 22). The LDP’s use of the JNR to promote political interests is reflected in rail line construction. Despite the dramatic declines in volumes and shares, between 1965 and 1980, the JNR added more than 600 kilometres of lines whereas the private railroads reduced their track by 1,815 kilometres. Most of the JNR’s new lines were to remote areas, which accounted for more than 40 per cent of operated lines but only 5 per cent of volume. Although the company reduced by half the number of passenger cars and the kilometres run by freight trains over this period, the JNR incurred its first full-year deficit (¥230 billion) in 1964 and would never be in the black until its privatization more than 20 years later. Most of the losses were generated by local operations. The main lines represented about half of the whole rail network, but 90 per cent of the total volume of rail transport. After deducting two special accounts – pension funds for repatriated employees and Shinkansen capital costs to remote areas – the main lines tended to operate profitably (Fukui 1992: 15). Borrowing, mainly from the MOF’s Trust Fund Bureau, covered the growing deficits and kept the deficits out of the regular government budget. In contrast, other countries with nationalized railroads, such as France, the United Kingdom and Germany, funded their transportation deficits through subsidies from the national budget. Japan’s approach to funding the rail line’s deficits produced ballooning debt, soaring interest payments and further losses. By 1986, long-term liabilities reached ¥19.7 trillion ($117 billion). On a per capita basis, this debt came to ¥162,000. The 1985 loss of ¥1,850 billion represented an outflow of ¥5 billion per day, or ¥15,300 per person for the year. Such figures were clearly unsustainable. By the 1980s, subsidies had climbed to around ¥650 billion annually (about $5 billion), including wartime repatriates’ retirement payments, Shinkansen construction costs and other special accounts (Fukui 1992: 17). The company and government formulated at least five reconstruction plans, beginning in 1964. The first three made the same assumptions: transport volume would continue to rise; capital investment would therefore be required; and fares should be raised every three years. All of these plans were abandoned within two or three years after their adoption. The fifth plan was intended to be the final one; the LDP issued a party resolution calling it ‘the plan with no successor’. Adopted in 1981, it too was abandoned.
146 The privatization of Japan’s public corporations JNR privatization and break-up PCAR deliberations focused on the three big public corporations, especially the JNR. As an executive member of the commission staff later observed, ‘[w]ithout the successful reform of the JNR, administrative reform under the PCAR could not be considered a success’ (Choi 1991: 407). The PCAR issued a sequence of reports during its deliberations so that the government could act on them immediately without having to wait two years for a final report. At first, privatization and break-up were not considered. The commission’s first report of July 1981 merely suggested that drastic measure – possibly including privatization – be taken for certain deficit local lines. However, a series of briefings to the commission recommended the dual proposition of privatization and break-up. The JNR management fought these recommendations, a stance that supported the commission’s views that management was not up to the task of reforming itself. The commission gradually accepted the more radical approach as the only one that made sense. By November 1981, this policy had gained the support of most of the commission members; in particular, it was pushed vigorously by a former vice minister of the Ministry of Transport (MOT), who berated the JNR managers at commission meetings. As the views coalesced, the JNR lashed back in public and enlisted the support of LDP political leaders. The JNR president, for example, complained in the press of PCAR’s ignorance of the JNR situation, claiming, ‘[t]hey focus only on the deficit problems because they do not fully understand at all what the JNR problem is and how it has to be solved. Nevertheless, they do not even try to study it’ (Choi 1991: 423). Meeting several times weekly, the subcommittee charged with public corporation reform sought to lay out the details of the issues and the policy responses in order to counter such arguments. In classic bureaucratic–political terms, they were building the record. By the end of January 1982, the subcommittee acknowledged privatization and split-up as the official line. In the first week of February, it presented 10 emergency reform measures and urged the government to adopt them immediately. The proposals included immediate freezes on hiring, Shinkansen construction, and other new investment. In April, it submitted a report recommending, among other things, the creation of a JNR Restructuring Supervisory Commission to develop detailed plans that would implement the main proposals. Several troubling issues remained to be decided, including what to do about extending Shinkansen lines, the enormous debt accumulated by the various JNR organizations, and the timing of privatization. A major political–bureaucratic battle then erupted over the form of the proposed supervisory commission. The PCAR had recommended that it be established with ministerial status, which would have made it equivalent in rank to the MOT. Many others wanted it to be an ordinary advisory committee under a ministry. Administrative Management Agency director Yasuhiro Nakasone intervened in the matter; he decided it would be structured as an ordinary advisory group under the prime minister’s office. However, he gave it ministerial-level authority
The privatization of Japan’s public corporations 147 to order ministries to provide desired information (the equivalent of subpoena power). Because of continued political sensitivities, the term ‘privatization and break-up’ did not appear in the enabling legislation. The law included such code phrases as, ‘to strive for the establishment of an effective type of management’ (Choi 1991: 435–8). In August 1982, the government submitted the law to the Diet and established a provisional office for the new commission. However, Prime Minister Suzuki resigned in October, Nakasone took over in November and the Diet was wrapped up with a bribery scandal involving the American Lockheed Corp. and former Prime Minister Kakuei Tanaka. The law establishing the supervisory commission for the JNR reconstruction law finally passed in May 1983. The supervisory commission expected to deliver its final report in July 1985. It was troubled by the continuing disagreement with JNR management, which, the commission members believed, would cripple any reform plans. In February, Nakasone publicly urged JNR officials who opposed reform to resign. The JNR executives continued to insist that only a unified system could serve the public interest. In June, the prime minister fired the president and vice president and named a cooperative MOT vice minister to the top post. Within days, all the top leadership positions were filled with supporters of the reform policy (Choi 1991: 449). The supervisory commission duly published its report on schedule and the Diet passed the government’s bills on 28 September 1985. The new law called for JNR privatization and break-up on 1 April 1987. The company would be split into six passenger operating companies, divided by geographic region. Earlier studies had shown that 95–99 per cent of all passenger trips took place within regions, so that this kind of split had some rationality. Separate companies were formed for Hokkaido, Kyushu and Shikoku. The main island of Honshu was divided into three zones with an independent rail company in each zone: east, west and central (Tokai region). A Shinkansen holding company would own the assets of the bullet train lines and lease them to the operating companies. A separate freight company would lease lines from the six new companies. The most troubling issue was the disposition of the debt, which was estimated at more than ¥37 trillion ($287 billion). The three main passenger companies and the Shinkansen holding company took on some of it. The remainder (¥25.9 trillion) was parked with the JNR Settlement Corporation. The three passenger companies on Japan’s smaller main islands (Hokkaido, Shikoku and Kyushu) were spared this burden because it was thought that their profitability prospects were too bleak to shoulder any debt. In fact, a special fund was established to subsidize their operations (Fukui 1992: 43). The reform plan assumed that shares in the six passenger lines and the freight company would be sold to the public when the companies had established a viable financial prospect. The three main companies – JR East, JR West and JR Central – met the Tokyo Stock Exchange (TSE) listing criteria by the end of fiscal 1991 (Fukui 1992: 82). The JNR Settlement intended to liquidate its liabilities by selling its assets: the
148 The privatization of Japan’s public corporations shares of the passenger and freight lines and the excess land that had been transferred to it. The original plan called for a series of open bids on the land. However, in a political decision based on weird economic logic, the cabinet in October 1987 postponed the sale of real estate in areas where land prices were sharply rising in the belief that releasing more land on the market would only stimulate higher prices. Consequently, the JNR Settlement missed the opportunity of unloading its assets when prices were high. Two years later, when prices started to fall, it withheld property because of the realistic belief that sales would drive down the market. Consequently, the net value of its liabilities continued to increase. Selling JR shares The MOT abandoned its plans to sell the railway companies’ shares in 1991 because of a collapsing stock market. Although an advisory committee to the ministry recommended selling each company’s stake in a lump sum, it acknowledged that instalments might be required to avoid overloading the market. The companies were anxious to get the sale under way because they hoped to escape from MOT control even if the public held just a fraction of their shares. Under the law, ministry officials had the final say on such subjects as board appointments, board decisions, financing arrangements and diversification plans (Kyodo, 29 November 1991). After deciding to skip a 1992 flotation, the MOT allowed JR East to apply for listing after the 1992 fiscal year’s books closed (Kyodo, 15 March 1993). Among those opposing a share distribution was the Tokyo Stock Exchange president, concerned that such a large sale would have adverse effects on the market. What created such great anxiety was the collapse of NTT share prices following a spectacular initial public offering and rapid run-up in price (see below). Most analysts mentioned the NTT example as a cautionary experience. JR East acted on schedule and applied in April 1993 for a listing. On listing day, 26 October, following two auctions, pandemonium ruled. A deluge of buy orders hit the exchange, but hardly anyone wanted to sell. No transactions cleared during the exchange’s normal business hours. TSE officials worked to complete at least one transaction so that a price could be listed. In after-hours dealings, enough shares were rounded up to manage a single consolidated transaction. Things did not improve on the second day of trading as more than 100,000 orders clogged the TSE computer system. Trading was suspended after 90 minutes. Exchange and finance ministry officials prevailed upon the JNR Settlement to release additional shares to the market. Late in the day, it offered around 570,000 shares at the ¥600,000 price, thereby diluting the original issue by some 29 per cent (Wall Street Journal, 28 October 1993). In the days following the listing, Tokyo prices collapsed, with the JR listing getting the blame. Responsibility for the market decline as well as for the original fiasco was well distributed. TSE officials said that the MOF refused to take its advice and was concerned only with maximizing revenues to pay off the JR debt and make up for general revenue shortfalls. Finance officials said that any
The privatization of Japan’s public corporations 149 problems were the exchange’s, and – from their point of view – the sale was a tremendous success. Brokers complained that the listing was over-hyped and that investors liquidated their other holdings to participate in the JR East sale, which drove down the rest of the market. Table 10.1 (at the end of the chapter) summarizes all the privatization transactions. The next planned listing was JR West, to be followed by JR Central, both tentatively scheduled for 1995. However, the Kobe earthquake of January 1995 forced a delay because of extensive damage to both companies’ tracks. A subsequent stock market plunge reinforced the inclination to delay. Moreover, the cost of rebuilding earthquake-damaged lines and a slowing economy caused profits to fall below listing requirements. At the beginning of the new fiscal year, JR West announced plans to apply for a listing later in 1996. Based on his own privatization experience, the president of JR East urged the other rail lines and the MOT to sell all their shares at one go. One reason for his recommendation was to get out from under the control of the government. This objective was becoming especially important because growing debt at the JNR Settlement was causing the MOF and other government bureaus to consider the passenger rail companies as a source of funds to help liquidate the debt. If the companies were wholly in private hands, the reasoning went, the government could not impose a levy on them (Kyodo, 13 May 1996). In late August 1996, 750,000 shares were offered at auction, but only 675,000 were taken up. Shareholders got a sweetener in the form of coupons for 20 per cent ticket discounts based on the number of shares that were held. Although analysts considered the price to be fair, apparently investors were looking for a bigger bargain (Asian Wall Street Journal, 30 August 1996). One year later it was the turn of the Central Japan Railway Co. (JR Central or JR Tokai), which offered 1.5 million shares (of the total 2.24 million) in October 1997. The issue opened smoothly with the closing price on the first day up 7.2 per cent from the auction price. Apparently, the authorities were learning how to run an initial public offering (IPO). Meanwhile, according to the original privatization legislation, the JNR Settlement Corp. was slated to go out of business in September 1998. Its pension liabilities, real estate assets and company shares were transferred to an existing government corporation, the Japan Railway Construction Public Corp. (renamed the Japan Railway Construction, Transport and Technology Agency in 2003). In 1999, a second offering of JR East went on the blocks. At first, the government considered unloading all its remaining 1.5 million shares, but ultimately the fear of market disruption led it to hold back a half-million shares. There were also those in the LDP and the MOF who did not want to give up fully their control over the company. LDP Diet members, in particular, feared that a completely privatized railroad would shut down many unprofitable local lines. The agency enlisted a foreign brokerage house to participate in the offering; Morgan Stanley Dean Witter co-managed the JR East offering with Nomura Securities. Shares were released on 2 August 1999 at ¥652,000, with 20 per cent sold abroad.
150 The privatization of Japan’s public corporations The Diet passed legislation in 2001 authorizing full railroad company privatization, including management decision-making without government approval. At the end of the year, the government still owned 12.5 per cent of JR East, 31.7 per cent of JR West and 39.6 per cent of JR Central, and proclaimed its intention to sell its remaining shares as soon as possible. In June 2002, the government announced pending sales of its stake in JR East, JR West, Japan Tobacco and NTT. The 500,000 shares of JR East still owned by the government were put on the market that month. However, plans for the other companies dragged on, and their sale during fiscal 2002 was finally cancelled fearing destabilization of a weak stock market (Nikkei, 18 February 2003). The finance ministry viewed share sales as a revenue-raising measure, whereas stock market officials worried about the price-reducing effect of releasing so many shares. The holding company also wanted to obtain the best price that it could in order to offset the large amount of debt that remained on its books. In February 2004, the ritual began again, with the announcement of the imminent sale of the residual JR West shares in government hands. This disposition occurred in April with the government ridding itself of all its JR West holdings. Early in 2005, the Transport Ministry announced its intention of selling most of its JR Central holdings that year. The railroad holding agency sold 600,000 of its shares in July 2005; JR Central directly purchased the remainder of the holdings in April 2006 in an off-hours stock exchange transaction, which kept the shares off the market and thus did not affect overall prices. As of early 2007, JR East, JR West and JR Central had been completely sold by the government. None of the other government-owned railroads have been sold. More than ¥4.0 trillion in proceeds were realized in the sales ($36 billion at contemporary exchange rates). Dealing with labour Public sector unions were militant and industrial relations conflictual. Some unions representing public employees were openly Marxist and supported the Communist Party. Others were the backbone of the JSP. The limited autonomy enjoyed by public sector management turned wage setting and labour relations into political affairs. Union militancy caused some public employees to split off into unions that were more cooperative. At the JNR, a breakaway faction established a more cooperative splinter group from the dominant and militant union. In the wake of a victory over a 1971 management reform plan, the militant union and the even more radical locomotive drivers union exerted their influence throughout the company. Labour discipline began to decline and the practice of falsified allowances, unauthorized time off and idle work proliferated. The JNR management desperately tried to avoid labour conflict (Mochizuki 1993: 185). When the PCAR looked at the JNR issue, it saw a labour discipline problem as well as a financial problem. After the commission published its basic JNR recommendations in July 1982 intimating privatization and break-up, the militant unions offered to participate in management’s reform plan, which included job
The privatization of Japan’s public corporations 151 reductions, but not break-up. The unions seemed to have believed that LDP and JNR opposition would block splitting the company. In 1986, when the JNR reform plan appeared to be inevitable, the locomotive engineers under its charismatic leader made a complete switch and joined the cooperative unions to sign a labour management document that laid out reform guidelines for labour. Following this dramatic move by his fellow unionist, the militant president pushed a more cooperative line, but it was defeated on the shop floor. The failure of its approach led to disastrous consequences: its membership dropped from 187,000 in June 1985 to 110,000 in November 1986. By April 1988, it represented only 17 per cent of the employees of the JNR successor companies (Mochizuki 1993: 186–7). Well before privatization occurred in 1987, the JNR had implemented parts of its own reform plans as well as the emergency recommendations of the commission. It had reduced recruitment in 1977 and nearly stopped it after 1982. Attrition allowed the workforce to diminish by 95,000 between 1980 and 1985 (22.6 per cent). As privatization drew near, the JNR, with government support, revised its pension scheme to make early retirement more advantageous if taken sooner rather than later. In addition to the usual severance allowance, the JNR offered a premium, equivalent to 10 months of standard wages, to those who offered to resign by 1986. Funds to cover the extra retirements were raised through the issuance of government-guaranteed bonds. In 1985 and 1986, 101,000 left the company compared with the average attrition rate of 54,000 (Watanabe 1994: 98). Post-reform labour reductions and efficiency gains Privatization plans estimated the optimum size of the new JR companies at 183,000 employees, assuming the same level of labour productivity as on the private railroads, adjusted for operational differences. Since the JNR would have an estimated 276,000 employees on the eve of reform, redundancies came to 61,000. Planners figured that 20,000 would accept the early retirement package and that the remaining 41,000 would be transferred to the JNR Settlement Corp., which would help them find new jobs by 1 April, 1990, the mandatory ending date of the placement programme (Watanabe: 1994: 97). As it turned out, attrition and the sweetened retirement package considerably reduced the need for transfers; the JNR Settlement placed only 7,630. At the conclusion of the placement period, 1,047 were dismissed after refusing to accept the new jobs offered them (Watanabe 1994: 99). From 1987 on, the railroad system (including JNR Settlement) made a net contribution to government revenues instead of being a drain. Of course, 70 per cent of the old debt had been transferred to the JNR Settlement. Passenger and freight volume reversed their steady decline and grew while fares held steady. Debt repayment proceeded steadily until 1991, when the three main passenger companies bought the Shinkansen system from the Shinkansen Holding Corp., which was dissolved. The one major disappointment was the debt of the JNR Settlement Corp., which continued to grow because of its failure to sell land and its delay in selling the shares of the passenger companies. By the time the company
152 The privatization of Japan’s public corporations was dissolved, the original ¥25.9 trillion debt had grown to more than ¥28 trillion.
Nippon Telegraph and Telephone Public Corporation privatization Events leading to privatization The Japanese government operated telecommunications beginning with the first telegraph lines in 1868, extending the Communications Ministry’s monopoly to the telephone service in 1889 on the rationale that it could best expand service to rural areas and preserve security. It managed the transition from manual to automatic exchanges after the 1923 earthquake that destroyed much of the existing network. The Second World War would destroy the system even more thoroughly (Takano 1992: 3). Under the American Occupation, the Communications Ministry was split into two parts: one for postal services and the other for telecommunications. Prime Minister Shigeru Yoshida wanted to privatize telecommunications, but the communications minister and others opposed this move. They reached a compromise in 1952 whereby the Ministry of Telecommunications became a public corporation, renamed Nippon Telegraph and Telephone Public Corporation (NTT), with a monopoly over domestic telecommunications. The Ministry of Posts became the Ministry of Posts and Telecommunications (MPT) with supervisory responsibility over NTT and a newly created public corporation for international telecommunications, Kokusai Denshin Denwa (KDD) (Vogel 1996: 139). NTT set itself two goals after the war: fully meeting the demand for telephone service and building a national direct-dial network. Barred from increasing rates, the carrier raised more than half its investment funds by requiring subscribers to buy telephone bonds for ¥100,000 ($278). NTT achieved its two goals by 1978, when it cleared the backlog of orders and completed its national network. However, the manpower required for creating a nationwide infrastructure was no longer necessary; moreover, the automatic systems installed as part of achieving the NTT goal of national direct dialling created large-scale staff redundancies. Nevertheless, the payroll continued to grow, peaking at 329,000 in 1979. NTT virtually ran itself. MPT had a small supervisory office, but one of its two chiefs was reserved for an NTT official. In 1971, a group of young MPT bureaucrats created several study groups to consider reforms in telecommunications. Their report recommended liberalization of so-called value-added networks in data transmissions and questioned the basic NTT monopoly. The Ministry of International Trade and Industry also actively sought the liberalization of telecommunications, mainly to promote the interests of the electronics and computer industries that it oversaw. In addition, at least five government commissions established after NTT’s creation recommended greater flexibility and liberalization, with a 1975 report even mentioning privatization and break-up (Vogel 1996: 142–3).
The privatization of Japan’s public corporations 153 Three separate technological developments contributed to liberalization pressures: the development and spread of advanced computer-linked terminals; the creation of alternative transmission methods via microwave and satellites; and the growing importance of sophisticated value-added networks combining data processing with communications. The new technologies challenged the economies-of-scale monopoly rationale and demonstrated that a unified system might not be able to keep up with blossoming technology-based demands. Meanwhile, in the late 1970s, US trade negotiators insisted that Japan open its telecommunications equipment market to foreign suppliers, which meant that the cosy relations within the ‘NTT family’ of monopoly buyer and favoured suppliers would be disrupted. NTT managers desired greater liberalization in their own realm of operations to meet the new challenges, but also wanted to maintain their monopolies over a narrower band of services. Liberalization ideas were brewing in the MOF as well. In the late 1970s, a group of officials in the budget bureau, pondering ways to reduce lingering deficits, contemplated privatizing the public corporations. They were ambivalent about Japan Tobacco because it was a big generator of tax revenues and wary of JNR because of its huge debt and militant unions. At the top of their privatization list was NTT because of the potential size of the sale, reinforced by the other reasons for privatization already made by others (Vogel 1996: 151). Despite a wide range of growing opinion that something should be done about the NTT monopoly, it was not until the 1980s, when the United States began the process of deregulating the telephone service, including breaking up American Telephone & Telegraph, that reform in Japan became a serious policy option. Just before the formation of the second PCAR, two separate reports recommended liberalizing value-added networks and competition in basic telephone services. What was significant is that one of these reports came from within the MPT, which suggested reforming NTT management and privatization. One more event was propitious for reform. In January 1981, just before the official formation of the PCAR, the government dismissed the NTT president. Among the reasons was his involvement in a scandal concerning payments for phoney overtime work – a method to reward employees which seemed to get around the law that required salaries to be equal to those at JNR. The new president, Hisashi Shinto, was the first outsider to take command of the company. He had a reputation as a rationalizer and cost-cutter as head of Ishikawajima Harima Heavy Industries, where he had succeeded Toshio Doko, who had gone on to chair the PCAR. It was said that Doko personally selected Mr Shinto for the NTT job (Vogel 1996: 152). Shinto quickly saw that the NTT required drastic overhaul; he appointed a group of mid-level managers to review the possibilities. They came up with three alternatives: transforming NTT into a completely privatized company; becoming a special private company similar to Japan Airlines with both private and government ownership; and remaining a public corporation, but with greater autonomy and flexibility in structure and business operations (Takano 1992: 6).
154 The privatization of Japan’s public corporations NTT suppliers vigorously fought privatization. In addition to their self-serving interest in preserving a guaranteed market, they also valued the collection of technical talent, network coherence, focused resources and regulatory knowledge. Since the suppliers were staffed at high levels with former NTT officials, the feeling was even stronger that past structures should not be altered. However, they were sensitive to the charge that American negotiators could use active campaigning against reform as evidence of barring foreign manufacturers from the Japanese market. The reform commission reached a consensus on NTT by May 1982. As background to privatization, it noted the necessity of restoring government finances and pointed to innovations in telecommunications technology. It concluded that privatization was necessary, competition should be established to eliminate the negative effects of monopoly, and that the limits of managing such a huge organization should be taken into account in rationalizing the corporation’s scale. The last observation pointed to reorganizing NTT into geographical entities, perhaps linked through a central organization (Takano 1992: 6). With the reform commission report in hand, LDP Diet members worked to find a way to implement the suggestions in a way that would satisfy the bulk of their constituents. The LDP, NTT suppliers, some parts of NTT management and the labour unions agreed to privatization, but break-up continued to be unacceptable. Unions at NTT were more cohesive than at JNR; a single union represented all employees. In the early 1960s, it had purged Communists from its ranks. NTT executives stressed good labour relations because of the need for flexibility to respond to the rapid pace of technological change in the industry. Unlike their counterparts at the railroads, they did not engage in union-busting activities. The union, in exchange for its cooperation, persuaded management not to dismiss employees or reduce the workforce. By the 1980s, the telecommunications workers had the shortest working week in the public sector at 37 hours. Although the company’s monopoly status allowed it to be profitable, an independent audit revealed 100,000 redundant employees. Reviewing these figures, the administrative reform commission was concerned that NTT would not be able to compete against foreign companies as the sector became more internationalized (Mochizuki 1993: 192). The union’s chief concern was that break-up implied more competition, which would impose job loss, keener competition among suppliers (possibly from foreigner companies) and the loss of political influence. The union was concerned also that break-up would fragment it into separate enterprise unions. Although the union adamantly opposed change, soon after the release of the PCAR’s July 1982 report recommending both privatization and break-up, the union shifted gears to embrace privatization. It hoped to escape the wage restrictions that kept it locked into the same pay scales as the railroad workers. The union leader decided to cooperate with the reform movement to secure the best possible outcome for his workers. Abandoning its ties with the JSP because of its ineffectiveness in the Diet, the union approached the LDP. The union was
The privatization of Japan’s public corporations 155 concerned that privatization would remove many of its protections written into law, such as proscriptions against layoffs and wage cuts. Since the break-up was opposed by most of the interested parties, the union was able to attain most of its objectives (Mochizuki 1993: 194–5). Within the ruling political party, the tactical job of dealing with the LDP conservative wing that opposed liberalization and break-up went to a senior member of the main LDP faction, Shin Kanemaru. In July 1983, he proposed a solution put forth by NTT that called for minor changes to its status as a public corporation. The conservatives readily accepted his proposal; however, the finance ministry and LDP leadership roundly condemned it, an outcome that did not surprise its author because the proposal was designed to fail. Kanemaru then went back to the NTT supporters in the LDP and told them that he had done his best and they had best throw their support behind a more radical approach. NTT head Shinto and reform commission members also helped to lobby the recalcitrant LDP members. MOF officials, anxious to get their hands on the privatization money, joined in. Prime Minister Nakasone next recruited Ryutaro Hashimoto, a prominent LDP politician, to find consensus in August 1983. Hashimoto submitted a report that endorsed the reform commission on privatization and competition, but that did not clearly call for break-up, stating such a possibility should be reconsidered in 10 years (Takano 1992: 10). The core struggle then settled on the different policy goals of NTT along with its main union and the MPT. The telephone company understood that a basic shift was under way where it required greater business flexibility to play in a more competitive and fluid environment. The MPT, however, viewed the situation as a continuing political struggle with MITI and NTT, and therefore sought greater authority over its industry (Vogel 1996:156). The Diet passed three telecommunications laws in December 1984; the MPT achieved many of its goals as the compromises worked out in the legislative process tended to swing in its favour. One point that it lost concerned the revenues from privatization, which it wanted devoted to research and development. The MOF agreed to share one-third of the revenues with MPT and MITI for a new R&D facility. The remainder would be reserved for deficit reduction. Selling and restructuring NTT Mr Hashimoto’s negotiations with the various interests indicated that legislation could not be passed if break-up was written into the law; instead, it called for a review of the issue in five years. On 1 April 1985, the new corporation was created with the issue of 15.6 million shares. The NTT privatization law required that the government retain at least one-third of the total shares, a fraction that would give it blocking authority in corporate decisions. The MOF and MPT agreed that government-owned shares should be higher than the minimum – above 50 per cent, at least within the five-year period of legislatively mandated stability. Therefore, the maximum number of shares that would be sold by 1990 would be 7.8 million.
156 The privatization of Japan’s public corporations Implementation of this plan called for selling 1.95 million shares in four lots in each fiscal year starting in 1986 (Takano 1992: 27). The Finance Ministry decided to establish a market price through competitive public bidding. Since a political goal was widespread ownership, the auction was combined with a public offering. Before the NTT listing, the largest equity issue ever placed in Japan was for ¥144 billion by Tokyo Electric Power Co. in 1981. The MOF analysts projected ¥416 billion from the NTT sale at an estimated price of ¥213,210, almost three times the previous largest sale; some private estimates projected the price at closer to a million yen. The first auction, of October 1986, put the weighted average price, and the one set for the public offering, at ¥1,197,000. The lowest bid was five times the MOF estimate and the public offering was overbid by nine times. The total came to a staggering ¥2.4 trillion ($15.3 billion). The price of NTT took off after the listing, caught up in the frenzy that was gripping Japanese stocks as a whole. The price hit a high of ¥3.18 million in April 1987. With this response to motivate them, the authorities scheduled the next tranche for November 1987. The selling price of ¥2.55 million was considerably below the peak price, but substantially above that of the first sale. The MOF realized almost ¥5 trillion. Although the stock market bubble would continue for another year, NTT was already on its way down at the time of the second release. For the third offering in 1988, the government reduced the number of shares to 1.5 million because of fears about the market’s absorptive capacity. The planned fourth placement was postponed. The government continued to hold almost two-thirds of the company going into the 1990s (Takano 1992: 33–4). With privatization in process, NTT’s no-layoff agreement with the unions meant that reducing the number of redundant workers had to take place by attrition. NTT had almost 304,000 workers at the time of privatization in 1985, which dropped to 266,000 five years later. The five-year postponement of a decision on NTT’s structure came up for decision, but in 1990 the future of the company was deferred for another five years. The finance minister announced in August 1992 that there would be no further sales until after March 1994, although he said that he hoped to resume sales as soon as market conditions improved (Wall Street Journal, 25 August 1992). In 1995, two advisory commissions recommended breaking up the company into a long-distance firm and two regional ones. By this time, Mr Hashimoto had become prime minister. All the parties who opposed break-up in 1985 continued their opposition, and it was taken off the table. MPT – perhaps to counter the continuing looming presence of an unreconstructed NTT – accelerated telecommunications deregulation measures. Before the government offered additional shares of NTT to the market, the company decided to list one of its subsidiaries, NTT Data Communications Systems Corp., to raise funds for investment. NTT Data listed on the second section of the TSE in April 1995 as NTT offered 47,000 shares of its holdings of 200,000. NTT, not the government, raised ¥49 billion ($575 million).
The privatization of Japan’s public corporations 157 In the 1990s, evidence was growing that Japan was falling further behind other advanced countries in telecommunications services. For example, foreign companies were making important inroads in Internet services in Japan. Moreover, the American government continued to press Tokyo to free up its telecommunications market and to increase NTT purchases from US suppliers. Despite the international pressure, it was still a surprise when NTT and MPT announced plans in December 1996 to restructure the firm into four parts – two regional carriers, a long-distance company and a holding company of the three operating companies. Owners of NTT shares would exchange them for shares in the holding company. At the same time, KDD would be permitted to enter the domestic market and NTT could venture into international business. The immediate stimulus for NTT’s agreement to this arrangement after more than 10 years of obstruction was the wave of global mergers and partnerships occurring in telecommunications; the giant new foreign firms created in this process were entering the Japanese market. The Diet passed amendments to the telecommunications laws in June 1997 enabling the agreement to be implemented. The company set up the new structure on 1 July 1999. In the midst of the restructuring debate, the stock market had improved sufficiently that the MOF announced plans to sell 500,000 NTT shares in the 1997 fiscal year. The price of NTT shares, which had fallen as low as ¥450,000 in 1992, had doubled by May 1997. The imminent passage by the Diet of the NTT restructuring legislation cleared the outlook for the firm and removed uncertainty from that front. However, towards the end of the fiscal year, the sale was abandoned to avoid overloading the market. The MOF budget planners suggested that a million shares might be put on the market in the next fiscal year because the government’s budgeting depended on proceeds from NTT share sales (Kyodo, 21 March 1998). Despite MOF hesitancy, NTT announced plans less than a month later to sell additional shares in NTT Data in what was being touted as the largest offering of new shares since 1993. The company raised ¥150 billion on the issue date of 11 May 1998; the equity sale allowed NTT to avoid increasing its interest-bearing debt, to which it was becoming more sensitive. The MOF screwed up its nerve the following October to go forward with the fourth tranche of NTT’s share sale. However, ministry officials still hedged their bets by saying that they would determine the timing of the offering only after studying the market situation. Several foreign firms joined the team to manage the offering (Wall Street Journal, 2 October 1998). Again, though, NTT beat the ministry to the punch with the largest IPO in Japan’s history. At the time that the MOF announced its tentative plans to float more of NTT, the firm was in the process of selling its mobile phone subsidiary NTT DoCoMo, with a listing in October 1998. Goldman Sachs International and Nikko Securities were the global coordinators for the 545,000-share sale, some 18 per cent of which the underwriters allocated to the foreign market. The offer brought NTT’s stake down to 67 per cent from 95 per cent. The ¥3.9 million price raised ¥2,126 billion, which although smaller than the second and third NTT
158 The privatization of Japan’s public corporations tranches met expectations as the largest IPO. Within days of the listing, NTT DoCoMo surged 20 per cent higher (Wall Street Journal, 12 October 1998). Buoyed by the success of the DoCoMo launch, the MOF proceeded with its fourth NTT sale on 15 December 1998, ending the 10-year drought of NTT shares. The sale brought the government’s holdings down to 59 per cent. A new twist was that about 40 per cent of the shares had been set aside for foreign distribution, a change from the past when foreigners were barred from holding NTT shares. The World Trade Organization’s telecommunications trade pact that had taken effect the previous February forced Japan to end foreign ownership restrictions on common carriers or so-called Type I firms – businesses that use their own circuits to provide services. (However, the 20 per cent cap on foreign ownership of NTT and KDD remained in place for the time being.) As the subscription unfolded, demand from individual domestic bidders poured in and the government dictated that individuals should receive at least one share, which pushed the foreign allocation down to 30 per cent of the offering. In the year following the fourth tranche, NTT shares almost doubled in price. When the MOF decided the following October to set the next sale, the price was over ¥1.3 million, 50 per cent higher than the previous December. The prospectus listed 952,000 shares for sale, which would bring the government’s stake in the company to about 53 per cent. The offering was one of the five largest of the 1990s and among the top 10 of all time (Financial Times, 9 November 1999). In the midst of a June 2000 political campaign, Prime Minister Yoshiro Mori indicated that he thought that NTT should be completely privatized. Government ownership and control kept the company from making full use of its resources, the prime minister said. His comments suggested that the Japanese government was considering revising the law to allow the sale of its entire shareholding in NTT. Other voices suggested that the industry was changing so fast that the company’s structure and continued regulation no longer made much sense (Financial Times, 12 June 2000). A few months later, NTT managers petitioned the government to liberalize the rules under which it operated. In particular, it sought relaxation of the government’s one-third stake requirement, restrictions on foreign ownership and the issue of new shares that hindered its expansion. It suggested that it might want to issue new shares in its own name to raise funds for acquisitions and other purposes. Although the telephone company retained a credit rating higher than the government’s, it preferred equity financing to borrowing to keep its balance sheet under control. Issuing new shares, of course, would compete with MOF sales in the market (Financial Times, 29 September 2000). While NTT was petitioning the government, the MOF was in the process of organizing the next sale. It sold 1.2 million shares in October 2000. Two weeks later it placed another 100,000 shares entirely to foreign buyers that had been set aside to fill any over-allotment. As the Japanese government’s debt continued to grow, the MOF had already included the proceeds from the next projected sale in its fiscal 2001 budget. However, LDP Diet members began to question the wisdom of further sales in a weak
The privatization of Japan’s public corporations 159 market, especially one in which telecommunications stocks were hard hit around the world. A planned release of DoCoMo added further caution to MOF plans. Specialists at the MOF were pondering other means to complete the privatization without unduly upsetting markets. They sought greater flexibility in the timing and volume of sales. For example, they were reviewing the possibility of using exchangeable or convertible bonds, which could later be turned in for equity. Another idea was to dribble out shares in small amounts; however, this scheme might depress prices because investors would not be sure about pending releases that could affect future prices. Some industry analysts complained that the ministry was not willing to consider the key to greater investor enthusiasm: breaking up the company. Analysts noted that the returns generated by NTT were less than the company’s cost of capital. Until this situation was improved, investors were unlikely to be wild about new shares (Financial Times, 23 January 2001). Despite annual announcements of million-share sales, the government subsequently conducted no further transactions of NTT on the open market. However, the company – concerned about the low returns experienced by its investors – used its large cash reserves to buy back shares held by the public – including 92,000 shares held by the Finance Ministry in 2002 and 85,000 the following year. This method of reducing the government’s holdings was repeated in 2004, when the company bought 800,000 shares in an off-market transaction. These sales reduced the government’s holdings to 40.8 per cent of outstanding shares. NTT purchased the final 1.12 million available shares in September 2005, bringing the government’s holdings down to the legally mandated one-third of outstanding shares. Any further privatization would require change in the law, In fact, an April 2006 study group of the Internal Affairs and Communications Ministry, the successor agency to the Ministry of Posts and Telecommunications, urged scrapping the requirement for government ownership. The minister, Heizo Takenaka, had been appointed to that post by Prime Minister Koizumi to implement post office privatization. As a former economics professor with no interest group attachments, he had already dealt with banking as director of the Financial Services Agency. The LDP’s subcommittee on telecommunications urged that any changes to the NTT law be delayed for study until around 2010. Thus, it appears that the battle over the future of government involvement in NTT will continue for many years.
Japan Tobacco Tobacco and salt monopoly France and Austria established tobacco monopolies in the 1670s to raise revenues. Japan and Sweden did so in the twentieth century. Tobacco was not subject to taxation in Japan until 1876, when the government imposed a general business tax. It remained in operation until 1898, at a time when Japan was burdened with the costs of the 1894–95 war with China. The business tax was abandoned and the government established a state monopoly for the purchase and resale of Japanese
160 The privatization of Japan’s public corporations tobacco leaf. The monopoly was extended to foreign leaf in 1900. The final step in the direction of complete state monopoly over purchase, manufacture and sale was taken in the law passed in April 1904, when the need for increased revenues had again become imperative because of the Russian war. Bureaucrats knowledgeable about French achievements established the Japanese tobacco manufacturing monopoly. The Monopoly Bureau of the MOF recruited engineering and law graduates from the top Imperial universities; the head of the bureau in 1912, Osachi Hamaguchi, was a graduate of Tokyo University and a future prime minister of Japan. Within several years, the bureau had substantially raised prices and revenues, concentrated its output in large factories and nearly doubled the productivity level that had been achieved under private management. By 1912, productivity in Japan’s tobacco industry was estimated at half the American level, whereas average productivity in Japan was only 10 per cent of the US (Hannah 2005: 31–3). The 80-year-old Japan Tobacco and Salt Public Corporation (formerly known as the Japan Monopoly Corp.) was privatized as the Japan Tobacco Industry Co. (JT) on 1 April 1985. The privatization law required that the government hold at least half of all shares, but additional clauses raised that limit to two-thirds; there were no immediate plans in 1985 to sell the permissible one-third government stake. The law was considered a defeat for the cause of administrative reform. Political pressures were able to divert the reform movement and produce a major victory for the agricultural interests of the LDP and the 90,000 or so tobacco growers concentrated in the southern island of Kyushu, who vigorously opposed privatization. The farmers were joined in opposition by elements in the MOF who expected to find jobs in the tobacco company after retirement from their government careers (Australian Financial Review, 1986). As the monopoly producer and seller of tobacco products and salt in Japan, the company had been obliged to buy all leaf tobacco produced in the country at fixed prices, typically about three to four times world levels. Under the new arrangements, this scheme was retained. Moreover, the company’s monopoly over tobacco product manufacturing continued after privatization. The old regime limited foreign sales to 10,000 retail outlets out of the 260,000 licensed to sell tobacco products. In addition, importers’ advertisements were restricted to English-language media; the foreign companies were required to pay a 15 per cent excise tax within 30 days of sales on top of a 24 per cent tariff while the JT had a six-month tax payment grace period. Foreign products held a scant 2 per cent market share in 1984 (Financial Times, 4 April 1984). Negotiations with the US government had already expanded the number of outlets to 70,000 in 1984. The privatization law allowed foreign brands in all licensed outlets. Immediately after the law passed the Diet with its promise of competition from foreigners, the newly minted company announced cuts in the number of offices and staff (Financial Times, 24 August 1984). In 1986, after further negotiations, the Japanese government agreed to eliminate the 24 per cent tariff on tobacco products in exchange for the United States
The privatization of Japan’s public corporations 161 dropping charges of unfair trade practices. By mid-1987, foreign brands were selling at the same price as domestic ones and the foreign market share had jumped to 8 per cent. Selling the government’s shares Towards the end of 1991, the MOF announced plans to sell one-third of its JT shares in fiscal 1992. Apparently, ministry bureaucrats had been planning such a move for some time, but hesitated because of the collapsing stock market. Falling tax revenues, however, shifted the balance of considerations. Although the sale was expected to encounter little difficulty, MOF officials noted strong opposition from tobacco growers, who were concerned about the influx of private capital into tobacco operations. The sale never occurred and little more was heard until May 1993, when a MOF advisory panel met on the subject for the first time in a year. An official confirmed that it was considering a sale ‘some time this year’ while monitoring the market reaction to a sale of JR East. As autumn approached, the date was set back to early 1994. By this time, foreigners controlled 20 per cent of the market. JT had reduced the number of its wholesale outlets by 36 per cent to 174 and had cut staff levels by 20 per cent. Although bank debt, at less than 1 per cent of total assets, was lower than in the railroad companies, the JT management wanted to diversify in the face of a declining overall market, growing foreign competition and its own high costs dictated by the requirement to buy domestic output at artificially high prices. The company wanted to be able to raise capital to fund diversification (Financial Times, 10 September 1993). Early 1994 came and went without further mention of a sale. Finally, an announcement came in July of a two-stage offering: 230,000 shares would be sold at auction in August and the remaining 436,660 shares would be distributed through a lottery when the shares were listed in October 1994 at the average weighted auction price. Anticipating the cost-cutting demands of future shareholders, the company had closed six of its 32 factories since 1985, and by 1994 had further trimmed staff by 32 per cent of the 1985 level. These cost reductions enabled JT to remain marginally profitable (Far Eastern Economic Review, 4 August 1994). Recalling the wild ups and downs of NTT share prices in the months after it was first listed, JT officials and the Japan Securities Dealers Association recommended a price limit on the first trading day to hold down soaring prices ‘as a result of excessively heated demand’ (Kyodo 20 April 1994). In contrast, analysts at non-Japanese brokerage houses were less enthusiastic, mentioning that the company was losing its share in a declining market, JT’s diversification ventures were all losing money, exports were minimal and high production costs were mandated by law (Asian Wall Street Journal, 12 July 1994). As the August auction progressed, the average auction price came in at ¥1.4 million. Individuals accounted for an estimated 80 per cent of the bids, with some going over ¥2 million. By most measures, the JT was more overvalued than
162 The privatization of Japan’s public corporations previously listed government issues, which made the feared initial price surge improbable (Asian Wall Street Journal, 1 September 1994). The denouement occurred in mid-October when the MOF announced that twothirds of the lottery winners had withdrawn their earlier application. Even those who agreed to accept their lottery shares were not a happy crowd. On listing day, the market opening of the company was delayed until the afternoon as 30,000 sell orders overwhelmed 5,000 buy orders. JT closed the day at a 24 per cent discount from the auction price. The company’s market price did not approach the auction day level for another five years. MOF officials said that the unsold shares would probably be offered in the next fiscal year, but this was delayed until 1996. The authorities tried to learn some lessons from the 1994 fiasco for the next sale. They invited foreign firms to apply as underwriters and reserved some shares for foreign investors. They also used a book-building method to determine an appropriate offering price. Nomura Securities and the London arm of Goldman Sachs were chosen to lead the underwriting for the June 1996 transaction. Because of the legal requirement that the government hold two-thirds of JT shares, the company could not issue new shares on its own or make use of equity swaps in its diversification goal. In order to finance transactions, the company was forced to accumulate cash reserves, borrow the funds or sell bonds in the capital market. It did all three, but not without complaints from stockholders. In 1997, the company held ¥650 billion in cash and short-term instruments. Shareholders called for distributing it in dividends or stock buybacks. The financing requirements of acquisitions came to the fore in 1999 when JT bought the international tobacco business of RJR Nabisco Holdings for $7.8 billion. The company arranged the largest ever syndicated loan in Asia, a $5 billion, 364-day facility. The company planned to pay back the loan within six months through a capital market issue. In the months following the deal, the price of the company’s shares hit an all-time high, which eased raising capital through dollar and yen bonds as well as loans. This experience convinced management that the company needed greater financial flexibility to implement the diversification plan required to cope with a declining market in tobacco products. The only way to do this was to reduce or eliminate the government’s mandated stake in the company. The same idea occurred to government officials and politicians, but for different reasons. They saw the government’s shares in JT and the other partially privatized companies as potential revenue sources, especially as deficits widened in the 1990s. A bill was introduced in the Diet in December 2001 that would allow the MOF to divest itself of its entire JT holdings and allow management to issue new shares, but limit the maximum additional shares to one million. At the same time, JT would continue to exercise its monopoly rights over domestic manufacturing as well as its responsibilities to take up the domestic tobacco crop. Former MOF personnel dominated JT’s top management. Although they recognized certain market imperatives, they also sought to protect the traditional interests of farmers, retailers and their former ministry colleagues expecting postretirement jobs. It was only in 2000 that a CEO took over who had risen through
The privatization of Japan’s public corporations 163 the ranks and was not a MOF bureaucrat. He reiterated the company’s preference for full privatization to a MOF advisory group in 2001. A MOF official said in reply that three issues – the requirement to purchase all domestically produced tobacco leaves, JT’s manufacturing monopoly and shareholding rules – were closely related (Nikkei, 5 March 2001). Despite JT’s preference for full privatization, a view that was shared by a few Diet members and government staff, the Finance Ministry’s advisory council recommended only that the government’s share be allowed to fall to 50 per cent. If the company issued new shares, the government’s holding could then be decreased to a minimum one-third of all issued shares. The Diet passed a law with these limits, which became effective in April 2002 (Kyodo, 7 December 2001). Eight years after the previous sale, the Japanese government released 14 per cent of its JT stake in June 2004, generating ¥241 billion for the national treasury. This reduction of government holdings brought its share down to the legal minimum of 50 per cent of originally issued shares. Post-privatization performance The principal resistance to further privatization came from the tobacco farmers and their LDP supporters. Under the law, and especially as long as the government holds a controlling interest in the company, it can compel its purchase of the total Japanese tobacco crop. Privatization had been orchestrated to allow this practice to continue. However, this constituency is in secular decline. From about 100,000 tobacco farmers in the early 1980s, the number declined steadily to 18,000 in 2005. Moreover, in order to reduce the cost of its required purchases, JT in August 2004 offered farm households a payment to quit farming. JT paid ¥4.2 billion to 4,000 tobacco growers; the number of families involved in tobacco growing is expected to fall to 14,000 by 2008. Production fell by two-thirds from the 1980s to 2005. In 1999, the area devoted to tobacco fell below 25,000 hectares for the first time. A gentlemen’s agreement between JT and the farmers had committed the company to set prices that would keep this area under cultivation. However, the cultivated area fell to 21,000 hectares by 2005. The age distribution of the farmers is growing older at a fast pace, with the majority of them over 65 years and few new ones taking the place of those who leave agriculture. By 2003, almost half of unmanufactured tobacco was imported and 27 per cent of the cigarettes sold in the country came from abroad (US Department of Agriculture 2004: 7, 12). JT managed to remain profitable in the face of declining demand and growing foreign competition, except for 2004, when restructuring costs helped produce a loss. With the cigarette market in long-term decline, JT announced a major restructuring plan in 1999 to eliminate 4,500 jobs (14 per cent of the payroll) and reduce its 20-plus processing plants to 10 within five years. It emphasized diversification into a broad range of new businesses, particularly tobacco machinery, food processing and biotech. By 2005, all segments earned positive profits.
164 The privatization of Japan’s public corporations Imports have been the driving force for improved efficiency. Competition forced attention to costs, which promoted a search for greater efficiency within JT and convinced farmers that both foreign leaf and imported manufactured cigarettes were threats. Remarkably, in 2005 the company’s return on its assets (operating income plus financial income to total assets) was a healthy 9.3 per cent (Japan Tobacco 2005: 99).
Japan Airlines privatization and deregulation Post-war law and regulation The basic framework for Japan’s aviation industry was established by the 1952 Civil Aeronautics Law, which gave MOT the mandate to set routes and fares and approve operators. The Japan Airlines Co., Ltd (JAL) had been formed the year before. In 1953, the government became the airline’s controlling owner by putting up half its capital. JAL entered international markets in 1954 as Japan’s ‘flag carrier’, the term used at that time to denote a country’s designated international airline. Although JAL had a monopoly on overseas routes, several companies competed in the small and chaotic domestic market of the early post-war period. Two of them merged in 1957 to form All Nippon Airways Co., Ltd (ANA). A 1970 cabinet resolution brought the industry under the heavy regulatory hand of the MOT, which implemented the policy in 1972. The so-called 1970–72 system, also referred to as the ‘aviation constitution’, created a segmented industrial structure built around three companies – JAL, ANA and Toa Domestic Airline Co. Ltd (TDA) – the last formed under official guidance through the merger of two smaller carriers. The government assigned JAL responsibility for international service and a few domestic trunk routes – connecting Tokyo, Osaka, Sapporo and Fukuoka. ANA was given domestic trunk routes, local routes and some short-haul international charter flights. The TDA, which merged with several other small carriers to form Japan Air System Co., Ltd in the 1980s, received local routes and a portion of the domestic trunks. Competition among the three carriers was almost non-existent. Near monopolies on most routes as well as regulated fares allowed the MOT to assign unprofitable flights serving outlying islands or small towns to ANA and JAS. The earnings from high-density routes subsidized this far-flung network, parts of which had been established under pressure from influential Diet members looking after their constituents’ interests. The 1970–72 system was internally consistent. It allowed the three airlines to earn sufficient profits from the larger domestic and international centres to crosssubsidize other locales. The absence of competition on prices and routes plus official barriers to entry of new carriers guaranteed a relatively quiet life for the companies. It also brought high fares based on costs that grew increasingly out of line with those of deregulated airlines in the United States and elsewhere.
The privatization of Japan’s public corporations 165 Competition stirs up domestic and international markets Following the late 1970s’ deregulation of airline operations in the United States, the American government turned its attention to the international system’s cartelized market structure and government price supports. In negotiations over bilateral pacts, the mechanism that governed international aviation relations among most countries, Washington sought more liberal entry and pricing arrangements. Its eventual aim was deregulated international ‘open skies’. Japan was an early target of American negotiators because its 6 per cent shares of total US international passenger traffic seemed small relative to the potential market that might arise under liberalized fares and routes. Japan’s aviation policymakers hoped to introduce such reforms as might be desirable at a pace that would allow domestic participants time to adjust to new economic forces created under a freer regulatory regime. Ministry planners expected implementation to last decades. Before this policy could be articulated very clearly, however, a sharp increase in the demand for international cargo flights and the introduction of Boeing Co.’s wide-body 747 aircraft made lowercost freight traffic technically and economically feasible. The MOT opened a small crack in Japan’s segmented market structure in 1983 when it granted a licence to Nippon Cargo Airlines Co. Ltd (NCA), a venture formed by ANA and six Japanese shipping companies. The appearance of NCA as a competitor to JAL on international cargo routes was the first challenge to the old system. The crack widened further in 1985 when the United States and Japan began a new round of negotiations over transpacific rights. The agreement that emerged was based on the principle of ‘balanced expansion’. It allowed three new carriers from each country to launch services across the Pacific. To forgo this opportunity would have left Japan faced with a dominant American presence in the all-important US–Japan market. Moreover, by this time, Japan had signed aviation agreements with Australia, France, Great Britain and other countries that also allowed more than one Japanese carrier to serve those markets. The logic of the 1970–72 system made it impractical for Tokyo to open the international market to new entrants and still keep the domestic market tightly regulated. JAL would face stepped-up competition and probably declining revenues and profits without the opportunity to make up the difference elsewhere. Moreover, ANA was losing money on some 70 per cent of its domestic flights. New competition on its major, profitable routes would jeopardize the carrier’s financial survival. Unsure of how to handle these issues, the MOT asked a transportation advisory council in September 1985 to review three issues: allowing domestic carriers onto international routes, privatizing JAL and introducing competition in aviation. In October, the aviation policy committee of the LDP recommended that domestic carriers be given international rights and that JAL be privatized. The MOT council recommended multiple carriers on international routes, new entrants on routes serving domestic city pairs (as warranted by traffic volume) and the privatization of JAL.
166 The privatization of Japan’s public corporations The advisory council advocated a framework of ‘double and triple tracking’, two or three carriers servicing a particular route based on passenger volume. The MOT implemented this suggestion by declaring that two airlines would be allowed on routes with at least 700,000 passengers a year and three carriers would service routes with more than 1 million passengers annually. For the first time, JAL, ANA and JAS faced competition on their most important routes. In return, however, they were permitted into market segments formerly denied them. In its measured way of promoting a more competitive industry, the MOT lowered the hurdles for double and triple tracking in 1992 to 400,000 and 700,000 passengers a year respectively. In 1996, they were cut further to 200,000 and 350,000. Taking advantage of the new opportunities, by 1990 ANA controlled 10 per cent of Japan’s international passenger traffic, doubling that share to 20 per cent in 1997. JAL did the same in the domestic market, quickly capturing 20 per cent of in-country traffic Selling JAL’s shares As a capital-intensive company, JAL required financing for aircraft acquisition. In the 1970s, it raised capital through bond sales in Japan, the United States, and Europe. Dollar borrowing was used to pay for the aircraft it bought from the American manufacturer, Boeing. The company indirectly increased its capital base in 1979 by issuing convertible bonds, which could be converted to shares over the next several years. Since the share price was rising over this period, most of these bonds were profitably converted. By 1980, the government’s original 50 per cent share of JAL had been diluted to 41 per cent through conversion of the convertible bonds. Several months before the Diet’s December 1980 authorization of the second PCAR, Administrative Management Agency director Nakasone threw some thunderbolts at JAL and the transportation ministry by proposing that the government sell its remaining holdings in JAL and the Electric Power Development Corp. His rationale was twofold: to use the income from the sales to reduce new government bond issues, and to bring more profit-oriented management to the public corporations (Nikkei, 23 September 1980). It was estimated that the sale of JAL shares could realize ¥25.5 billion ($120 million). The MOT held serious reservations about the idea of selling JAL shares. If its holdings fell below one-third, it would lose control of the still-regulated company. Furthermore, the MOT wanted to be guaranteed priority rights to subscribe to any new shares that the company subsequently might offer (Nikkei, 16 December 1980). After cabinet discussion, the government agreed to sell 5 per cent of its stake in the company (2 per cent of all shares). The MOT agreed to this sale since it would preserve its controlling interest in JAL. At the same time, JAL embarked on an ambitious equipment programme. Planning to spend $2.4 billion over the next four years, the company projected the need for at least 1 billion dollars of financing. It had already gone to the Eurobond market seven times and was hoping to attract equity investors. However, for that
The privatization of Japan’s public corporations 167 to be a feasible option, the government share in the company would have to come down. The tripling of its share price over the past year gave a strong motivation for the government to raise cash from share sales (Euromoney, September 1985). While deregulation discussions were proceeding, JAL experienced the worst air disaster in history when a Boeing 747 crashed soon after takeoff on 12 August 1985 with the loss of 520 lives. When top JAL executives resigned, Prime Minister Nakasone personally appointed a new president and chairman. Observers likened this move to the sacking of the president of JNR when he objected to the pace and scale of privatization earlier that year. The new JAL president had worked on privatization issues in the Administrative Management Agency under Nakasone. He noted that the company’s vitality had been sapped by government intervention. He also suggested that government backing had led to militancy among JAL’s non-unionized pilots and four rival employee unions. The government was always there, he said, to bail out the company if overstaffing, excessive wages or rigid work rules increased costs (Far Eastern Economic Review, 16 January 1986). With new leadership at the airline and with the ministry’s advisory council’s report in hand, the MOT formally adopted the council’s recommendations in May 1986, planning to sell its portion of the company gradually over several years. However, by early 1987, the Finance Ministry was suggesting that JAL shares should be sold by the autumn of that year, if not earlier. Enthusiastic investor support of the first NTT sale, plus JAL’s zooming stock price, seemed to warrant a quick disposal. In anticipation of privatization, the airline was forced to consider its efficiency: 1,100 ground personnel were reassigned to subsidiaries and the company proposed a reduction in crew size to increase efficiency. For example, the carrier flew the Los Angeles–Tokyo route with 23 cockpit and cabin crew compared with 16 on Northwest Airlines or 17 on United Airlines’ longer Tokyo–New York flight (Business Week, 31 August 1987). In September 1987, the Diet passed the JAL privatization law requiring that all the government’s shares be sold by the end of the year. The Finance Ministry set the sale for 15 December. About half of the government’s 34.5 per cent stake in the company was already allocated to stable shareholders such as JAL affiliates and subsidiaries, its banks and institutional investors (Kyodo News, 24 November 1987). Life after privatization and deregulation Perhaps the best indicator of the effect of JAL privatization and industry deregulation on performance is the behaviour of prices and costs. Figure 10.1 shows domestic yields (yen revenue per passenger kilometre) adjusted for price changes by the consumer price index, for both ANA and JAL. The trend is downwards for almost the entire period since 1980. The jump in JAL yields after 2001 was caused by its takeover of Japan Air System Co., whose shorter routes imposed higher costs. Although the introduction of greater domestic competition did not
168 The privatization of Japan’s public corporations 28 26 ANA 24 22 20 18 JAL 16 14 1980
1984
1988
1992
1996
2000
2004
Figure 10.1 Domestic passenger yields on Japanese airlines (2000 ¥ revenue per passenger kilometre). Source: company annual reports.
noticeably accelerate the reduction in yields, the steady improvement over 25 years may not have been so persistent without the introduction of several carriers on most routes. On international routes, the Japanese airlines have to compete with some of the most efficient in the world. Both Japanese companies struggled to bring their costs and prices down to meet the challenge of international competition. By 1998, yields for the Japanese companies had finally converged to those of the American carriers. The evidence suggests that privatization, and especially deregulation and competition, have had the desired effects. Efficiency has risen and prices have fallen. JAL management is acutely sensitive to profitability considerations. New entrants have driven down prices on the most popular trunk routes. Although constrained by geography and airport limitations, Japanese aviation is approaching international standards. Still, it has taken more than 20 years for the Nakasone vision to be realized in this very Japanese example of privatization and deregulation.
Conclusions: privatization with Japanese characteristics More than a quarter-century after the privatization process of Japan’s state-owned companies began, the government had disposed completely of four: JAL, JR East, JR West and JR Central. It still owned 33 per cent of NTT and half of JT. In addition, the government continued to hold all of the shares of the three smaller island railroads and a freight line. Table 10.1 gives the full sequence of share sales. The sales generated proceeds of ¥20.6 trillion ($167.5 billion) for the government. The Finance Ministry estimated total financial gains at ¥31 trillion for the 10 years from 1985 to 2004, an amount that included sales proceeds, dividends and taxes paid by the privatized companies (Nikkei, 3 September 2005).
The privatization of Japan’s public corporations 169 Perhaps the main conclusion of this review is how difficult it is to accomplish even such broadly accepted and justifiable policies. It required the abundant political skills of Mr Nakasone, both when he was an agency director and as prime minister, to proceed as fast and as far as he did. Opposition was fierce from many sources: company executives, ministerial guardians, employee unions, suppliers and Diet members who depended on support from these various constituencies. Despite Nakasone’s political skills, plus broad business and public support, he was forced to accept watered-down versions of the PCAR’s recommendations and to postpone for many years final solutions, some of which ultimately were never attained. This experience illustrates the concept of reform with Japanese characteristics. Such change includes manifest fairness to incumbents, time to adapt, reluctance to engage in confrontation and difficulty in abandoning employees and others in long-term relations. However, managing change in such a drawn-out way, and incompletely at that, leaves the privatized entities and industries at risk from a world that is changing even faster. As Japanese politicians and bureaucrats manage change at their preferred pace, they may be left even further behind than when they started, despite the unambiguous result of real, positive progress.
Date 1 Apr 1985 1 Apr 1985 1 Dec 1985 1 Oct 1986 1 Jan 1987 9 Feb 1987 1 Apr 1987 1 Apr 1987 1 Apr 1987 12 Nov 1987 15 Dec 1987 21 Oct 1988 10 Oct 1993 10 Oct 1993 27 Oct 1993 15 Aug 1994 27 Oct 1994 17 Jun 1996 8 Oct 1996
Action JT privatization NTT privatization JAL offering NTT first tranche auction NTT first tranche lottery NTT first tranche cooling off release JR Central privatization JR West privatization JR East privatization NTT second tranche JAL final sale NTT third tranche JR East lottery JR East auction JR East release JT auction JT lottery JT offering JR West offer
Shares 2,000,000 15,600,000 2,531,000 ,200,000 1,650,000 100,000 2,240,000 2,000,000 4,000,000 1,950,000 48,099,600 1,500,000 1,330,000 600,000 570,000 230,000 164,270 272,390 695,000
Table 10.1 Japanese government share sales of privatized corporations Proceeds (bn ¥) 5.8 239.4 1975.1 160.0
4972.5 645.2 2,850.0 505.4 228.0 342.0 330.7 236.2 222.0 248.1
Price (¥) 2,270 1,197,000 1,197,000 1,600,000
2,550,000 13,414 1,900,000 380,000 380,000 600,000 1,438,000 1,438,000 815,000 357,000
36.7 5.2 22.1 4.7 2.1 3.2 3.3 2.4 2.0 2.2
0.03 1.5 12.8 1.0
Proceeds (bn $)
JR West auction JR Central offer JR Central auction NTT fourth tranche JR East release NTT fifth tranche NTT sixth tranche NTT sixth tranche set aside JR East final sale NTT purchase from MOF NTT purchase from MOF JR West final sale JT offering NTT purchase from MOF NTT purchase from MOF JR Central sale JR Central purchase from Railway Agency
675,000 660,000 700,000 1,000,000 1,000,000 952,000 1,200,000 100,000 500,000 91,800 85,157 634,344 289,334 800,145 1,123,043 600,000 286,071
240.0 236.9 251.3 855.0 652.0 1,586.0 1,138.8 94.9 266.0 39.6 45.9 260.7 241.3 366.5 542.4 477.0 328.9 167.5
357,000 359,000 359,000 855,000 652,000 1,666,000 949,000 949,000 532,000 431,000 539,000 411,000 843,000 458,000 483,000 795,000 1,150,000 20,583.6
2.1 2.0 2.1 6.9 5.5 15.2 10.5 0.9 2.3 0.4 0.4 2.4 2.2 3.5 4.8 4.3 2.8
Note: Dates may be approximate as many sales occurred over a period of days or weeks. Dollars converted at average exchange rate during month of transactions.
Total government proceeds
8 Oct 1996 7 Aug 1997 7 Aug 1997 15 Dec 1998 13 Jul 1999 9 Nov 1999 23 Oct 2000 6 Nov 2000 21 Jul 2002 8 Oct 2002 15 Oct 2003 12 Mar 2004 11 Jun 2004 26 Nov 2004 6 Sept 2005 29 Jul 2005 5 April 2006
11 Is Japan different? Implications for economic growth
Introduction The assertion of differentness The Japanese economy is different. This assertion, as often as it is made, is meaningless as it stands. Analytical questions intrude from the start: How is differentness measured? Compared with whom? Along what dimensions? All nations, societies and economies differ from one another. It requires little thought to note, for example, that American states or Japanese prefectures differ among themselves. To bring content to the assertion of differentness, at a minimum, it is necessary to name comparison economies, enumerate the qualities being compared and define a metric for appraising differentness. Previous research and its critique Several years ago, I attempted to address this issue (Alexander 2002: 255–72). The sample of countries included as many as could be found with comparable data in the realm of economic structure and government–economy relations – those attributes thought to make Japan unique. Almost 50 variables extracted from 11 studies formed the core of that analysis. One measure used to assess the difference or similarity between countries was each one’s Euclidean distance from the United States. Euclidean distance is based on the familiar formula from geometry: the square root of the sum of the squares of the differences between each country’s variables and the American values. A second measure was the correlation of each country’s variables with those of the United States. A key finding was that Japan fits squarely in the group of Anglo-American economies, including the United States, United Kingdom, Canada and Australia. The conclusion that Japan did not differ in significant ways from the presumed paragons of competitive capitalism generated criticism. Of course, critics noted, it is clearly apparent that Japan and the United States are different from Rwanda or Haiti or many of the other 100-plus countries in my sample that were shown by the data to be very different from all developed countries; the inclusion of such
Is Japan different? 173 obviously different economies might have made the separation between Japan and the United states appear to be relatively smaller. Although those asserting differentness had not drawn up a list for comparison, it was felt that the inclusion of such obviously disparate economies must have skewed the results. Other critics noted that the period of the observations was blurred; data were collected opportunistically from any available source and spanned many years; since Japan was changing over these years, it was not clear as to just when the comparisons were being made. As Japan deregulated, liberalized and opened its economy to foreign investment and trade, we might expect that recent years were perhaps more like the United States than earlier ones. One general relationship stood out in the earlier study: the more similar an economy was to the United States, the higher was its gross domestic product per capita. This linkage implied the hypothesis that if a country wanted to become richer, it should emulate the habits, institutions and structure of the United States. This hypothesis was inferred from the data, but remained untested in the previous research. New research This chapter seeks to advance the earlier study in several ways. First, the sample of economies excludes those that are poor or tiny. The inclusion rules require real, 2002 per capita GDP (in 2000 purchasing power parity dollars) greater than $5,000 and a population greater than 4 million; countries that are either former republics of the Soviet Union or former communist regimes are excluded, as are those for which there are insufficient data to perform statistical analysis. Altogether, 38 countries met these criteria, shown in Table 11.1. The second departure in this analysis is that three specific periods are examined: 1980, 1990 and 2002. If a particular variable was not published for those exact years, data were selected within a year or so of the specified times. Third, with several periods, it becomes possible to examine explicitly the relationship between income and similarity to the United States. Main results Japan was more similar to the United States in 1980 than in any of the later periods, echoing the findings from the previous research. However, by most measures, its distance from the United States increased in 1990 and 2002. The new data show that the link between American traits and real GDP per capita grows stronger with time; that is, the effects of possessing the institutions and methods of a rich country like the US appear to build up slowly over many years rather than appearing in full force instantly. By implication, the more dissimilar a country is to the United States, the lower its income, and the more its performance deteriorates as time passes. The relationship seen earlier between similarity to the United States and income holds even more strongly over longer intervals.
174 Is Japan different? Table 11.1 Countries included in comparisons Country Argentina Australia Austria Belgium Brazil Canada Chile Colombia Costa Rica Denmark Dominican Republic Finland France Germany Greece Hong Kong Ireland Israel Italy Japan Malaysia Mexico Netherlands New Zealand Norway Portugal Singapore South Africa South Korea Spain Sweden Switzerland Thailand Tunisia Turkey United Kingdom United States Venezuela
2002 GDP/capita (2000 $) 10,664 27,256 28,223 26,526 7,480 28,728 9,432 6,139 8,505 29,730 6,414 25,568 26,090 26,141 18,052 25,541 35,352 19,055 25,453 25,788 8,811 8,662 27,932 20,959 35,734 17,463 23,099 9,750 16,570 20,777 25,028 29,205 6,740 6,508 6,145 25,139 34,557 5,182
Population (million) 36.5 19.7 8.0 10.3 174.0 31.4 15.6 43.7 4.0 5.4 8.6 5.2 59.5 82.5 10.6 6.8 3.9 6.6 57.7 127.0 24.3 101.0 16.1 4.0 4.5 10.2 4.2 45.3 47.6 40.9 8.9 7.3 61.6 9.8 69.6 59.2 288.0 25.1
Source: World Bank (2006).
Data I sought databases that included comparable information on economic structure, behaviour and government–economy relations for as many countries over as many years as possible. These data refer to economies as a whole and not to industries, firms or micro-behaviour. A central source was the Fraser Institute’s Economic Freedom of the World, which had online data from 1970 to 2002 for 123 countries and 38 variables, organized within five sections: size of government (expendi-
Is Japan different? 175 tures, taxes and government enterprises); legal structure and security of property rights; access to sound money; freedom to trade internationally; and regulation of credit, labour and business (Gwartney and Lawson 2004). A second important source was the World Bank’s database on financial development and structure, with 192 countries and 22 variables (17 used in this study), covering the years from 1960 to 2003 (Beck et al. 1999). Another World Bank compilation on governance has country data on voice and accountability, political stability, government effectiveness, regulatory quality, rule of law and control of corruption. The information covers 209 countries from 1996 to 2004 (Kaufmann et al. 2005). Similar information on earlier years came from other published studies. Fewer data were available further back in time. Altogether, 66 variables were used for 2002, 46 for 1990 and 40 for 1980. The variables and their sources are listed in the Appendix.
Analysis The first task in analysing the data was to standardize each period’s observations by subtracting the variable’s mean and dividing by the standard deviation. This operation generated new variables with a mean of zero and standard deviation of 1.0. The motivation for doing this is that the scale of the original variables varied widely; without some kind of normalization, a variable whose normal range was in the thousands would overwhelm one with a range of single digits or fractions. Similarity metrics The principal method used to produce summary measures of comparisons was a slightly modified version of the so-called Euclidean distance between each country and the United States. The Euclidean distance makes use of the familiar theorem that the distance between two points is the square root of the sum of the squares of the distances measured along each of the separate dimensions between two points; if variable k in country i is Xik and variable k in country j is Xjk, the distance between countries i and j, Dij, is defined as: Dij = (∑k (Xik – Xjk)2)1/2 According to this formula, distance is always positive, with a minimum of zero. However, there is a problem with this measure; its size depends on the number of variables in the comparison, which, because of data availability, differs over time and across countries. Suppose, for example, that there is only a single variable describing the economy and that the difference between two countries with respect to the value of this variable is 1.0; using the Euclidean distance equation, the distance between them would be 1.0. If, in contrast, two variables were used, also with differences equal to 1.0, the distance between the countries would be 1.414 (the square root of 2.0). Three such variables would produce a distance of 1.732 (square root of 3.0), and so on. Although the two points lie one unit apart in
176 Is Japan different? each dimension, adding dimensions increases the measured distance. In order to remove possible biases from missing data or variability in the number of variables at different times, it is necessary to divide the Euclidean distance by the square root of the number of variables. Since the variables have been standardized with means of zero and standard deviations of 1.0, the modified Euclidean distances can be interpreted as the number of standard deviations that one country is from the other. Table 11.2 shows the distance measures of all countries from the United States for the three periods. Japan’s position is shown in bold type. One of the first things that jumps out of the table is that countries with British heritage tend to cluster near the United States. In 2002, for example, the countries nearest the United States were the United Kingdom, Canada, Australia and New Zealand. Because of the presumed a priori similarities among these countries, their clustering produces some comfort that the Euclidean distance is doing the job assigned to it. The notable feature about Japan is that its position falls from one period to another, in both rank and distance. In 1980, Japan’s position was number 10 with a distance less than 1.0; by 2002, Japan had fallen to number 24 at a distance of 1.29. In 1980, Japan was clearly in the group of rich Anglo-American countries; however, by 2002, its neighbours were more heterogeneous. A second method of analysis is based on correlations of the variables between countries. Correlations vary from 1.0 (perfectly correlated) to –1.0 (each variable in one country is equal to the same variable in the other country, but with the opposite sign. A correlation of zero implies no relationship between the two sets of observations. Table 11.3 shows the correlations between the United States and the other countries for the three periods, ranked by similarity to the United States. The same patterns emerge as with the distance measure. The Anglo-American countries cluster together and Japan moves further away from the United States over time, from a statistically significant correlation in fourth place in 1980 to a practically zero correlation in the 13th position in 2002. Amendments and additions Several improvements can be made to this analytical approach. First, many of the variables are highly correlated among themselves; if they happen to have a large effect, they would bias the results by strengthening them without adding additional information. To deal with this problem, I first examined the correlation matrix among all the variables for each period to identify those that are closely related to each other. I then performed a factor analysis of the highly correlated variables; this statistical process combines the variables into a smaller set based on the linear relationships among the original set. Two or three factors were formed from 10–14 variables in the three periods. Factors are calculated to be statistically independent of each other. These factors were substituted for the correlated variable (shown in the Appendix); distances and correlations were recalculated with the remaining variables and factors. Another possible problem arises from the fact that different variables were used
Is Japan different? 177 in the three periods because of limited data availability. To correct possible biases from this source, a common set of variables was used for all three periods. Finally, some variables may be outliers that are unrepresentative of the economy as a whole. To test if a few variables may be driving the results, the three variables in each period that made the largest contribution to the distance between Japan and the United States were removed from the calculations. The results of these estimations are shown in Table 11.4. The first entries for ‘Distance’ and ‘Correlation’ present the same information as in Tables 11.2 and 11.3. The second line of Table 11.4 shows the result from excluding the three most extreme of Japan’s variables from the calculations; as would be expected, Japan’s position with respect to the United States moves up from 10 to 6, and the distance falls from 0.89 to 0.74. However, the trend of increasing dissimilarity over time does not change. Substituting independent factors for closely related variables also draws Japan closer to the United States, but does not change the tendency to become farther apart from 1980 to 2003. By several measures, Japan was quite similar to the United States in 1980, but drew further away as time progressed. Indeed, every single approach shows increasing distance, lower correlation and more remote position for Japan. Examining the variables for 1980 and 2002 shows that Japan moved away from the United States across most of them; however, three variables brought the countries closer together, including fewer restrictions in Japan on foreign direct investment into the country and reduced exchange rate controls. The variables showing the greatest increased distance included actual versus expected size of trade (Japan’s trade was smaller than expected). Two variables that contributed the most to increased differences over time were associated with the stock market; American markets had vastly increased amounts of trading relative to GDP and to the number of outstanding shares. Nevertheless, when the three variables with the largest contributions to Japan–US differences were removed from the estimation of the metrics, the qualitative trend did not change.
Income and similarity to the United States My previous research found that real GDP per capita was related to the similarity of a country to the United States: the more like the United States, the richer the country. This relationship is illustrated in Figure 11.1, which plots 1980 values of income per capita against 1980 distance (with factors substituted for correlated variables). The correlation coefficient of the points in the figure is statistically significant at –0.71. However, one set of observations for a single period does not allow going beyond the reasonable inference that similarity to the United States is related to income. At issue is disentangling cause and effect. It could be that countries adopt American-style institutions as they grow richer rather than becoming rich because they adopt these institutions. Now, with three distance measures at different times and several decades of income data, we can check whether this hypothesis holds up to more intense scrutiny, primarily by testing the time phasing of the relationship.
Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1980 Country United States Canada Netherlands Ireland Australia Germany Finland Belgium Denmark Japan Austria United Kingdom France New Zealand Sweden Norway Spain Hong Kong Italy Malaysia
Distance 0.00 0.39 0.68 0.71 0.77 0.77 0.82 0.89 0.89 0.89 0.91 0.92 0.93 0.94 0.94 0.96 1.02 1.09 1.13 1.15
1990 Country United States Canada United Kingdom Australia New Zealand Ireland France Germany Norway Austria Netherlands Switzerland Denmark Finland Sweden Spain Costa Rica Belgium Singapore Japan
Table 11.2 Euclidean distances from the United States, 1980, 1990, 2002 Distance 0.00 0.68 0.70 0.73 0.79 0.86 0.95 0.96 0.97 0.99 1.00 1.02 1.02 1.09 1.11 1.13 1.15 1.17 1.17 1.19
2002 Country United States United Kingdom Canada Australia New Zealand Switzerland Spain Netherlands Ireland France Denmark Israel Sweden South Korea Italy Chile Singapore Finland Austria Costa Rica
Distance 0.00 0.75 0.84 0.86 0.91 0.93 1.05 1.07 1.09 1.10 1.14 1.17 1.18 1.19 1.19 1.21 1.21 1.21 1.21 1.23
21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
Greece Portugal Switzerland Singapore South Africa Costa Rica South Korea Venezuela Chile Thailand Dominican Republic Colombia Israel Brazil Mexico Tunisia Turkey Argentina
1.30 1.32 1.32 1.43 1.46 1.47 1.50 1.54 1.55 1.60 1.61 1.65 1.68 1.79 1.80 1.86 2.02 2.12
Italy Israel Chile Malaysia Portugal Korea, Republic of Hong Kong Greece South Africa Thailand Mexico Turkey Tunisia Venezuela Colombia Argentina Dominican Rep. Brazil
1.22 1.24 1.25 1.27 1.31 1.33 1.35 1.46 1.50 1.51 1.60 1.69 1.70 1.71 1.71 2.00 2.05 2.07
Portugal Germany Norway Japan South Africa Belgium Greece Hong Kong Thailand Malaysia Tunisia Dominican Rep. Mexico Brazil Colombia Turkey Argentina Venezuela
1.23 1.25 1.28 1.29 1.34 1.34 1.36 1.38 1.51 1.52 1.54 1.61 1.67 1.81 1.83 2.05 2.05 2.30
Rank 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
1980 Country United States Canada Netherlands Japan Hong Kong Ireland Germany Australia Belgium United Kingdom Switzerland Finland Denmark New Zealand Austria Norway France Sweden
Correlation 1.00 0.87 0.65 0.60 0.56 0.53 0.51 0.51 0.46 0.44 0.40 0.39 0.37 0.32 0.27 0.26 0.24 0.24
1990 Country United States United Kingdom Canada Australia Switzerland New Zealand Ireland Hong Kong Japan Germany Denmark Singapore Netherlands France Norway Austria Costa Rica Belgium
Table 11.3 Correlation coefficients with the United States, 1980, 1990, 2002 Correlation 1.00 0.59 0.58 0.44 0.41 0.40 0.37 0.32 0.29 0.28 0.28 0.20 0.19 0.14 0.13 0.13 0.12 0.11
2002 Country United States United Kingdom Switzerland Canada Australia New Zealand South Korea Hong Kong Denmark Singapore Spain Netherlands Japan Finland Malaysia Ireland Chile France
Correlation 1.00 0.54 0.40 0.35 0.30 0.30 0.25 0.16 0.14 0.12 0.09 0.05 0.04 0.01 0.00 –0.03 –0.04 –0.05
19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38
Spain Singapore Italy Portugal Costa Rica Malaysia Venezuela Greece Dominican Republic Israel South Korea Chile South Africa Thailand Mexico Colombia Argentina Turkey Brazil Tunisia
0.15 0.11 –0.03 –0.11 –0.16 –0.16 –0.18 –0.20 –0.26 –0.36 –0.39 –0.46 –0.46 –0.47 –0.51 –0.58 –0.59 –0.62 –0.70 –0.78
Sweden Finland Italy Spain Malaysia Korea, Republic of Chile Israel South Africa Portugal Argentina Dominican Rep. Venezuela Thailand Greece Turkey Brazil Mexico Colombia Tunisia
0.08 0.04 –0.07 –0.08 –0.13 –0.14 –0.15 –0.16 –0.23 –0.26 –0.31 –0.32 –0.34 –0.34 –0.37 –0.38 –0.40 –0.41 –0.42 –0.61
Israel Sweden Italy Costa Rica Dominican Rep. Brazil Thailand Germany South Africa Tunisia Colombia Argentina Turkey Norway Mexico Venezuela Austria Belgium Greece Portugal
–0.07 –0.07 –0.08 –0.09 –0.09 –0.11 –0.12 –0.14 –0.15 –0.17 –0.18 –0.18 –0.18 –0.19 –0.20 –0.22 –0.24 –0.24 –0.27 –0.35
20 11 16 20 9 10
10 6 3 3 4 3 0.89 0.74 0.69 0.79 0.60 0.68
Distance from US Distance Distance excluding three outliers Distance (with factor substitution) Distance, same variables all periods (with factor substitution)
Correlation with US Correlation Correlation (with factor substitution)
0.29 0.25
1.19 1.03 1.16 1.08
1990
1980
Metric Rank relative to US when measured by Distance Distance excluding three outliers Distance (with factor substitution) Distance, same variables all periods (with factor substitution) Correlation Correlation (with factor substitution)
Table 11.4 Position, distance and correlation of Japan relative to United States
0.04 0.08
1.29 1.13 1.36 1.39
24 15 23 17 13 13
2002
Is Japan different? 183
Real GDP per capita, 2000 $
30,000 25,000 20,000 15,000 10,000 5,000 0 0.0
0.5
1.0
1.5
2.0
2.5
Distance from US
Figure 11.1 Real 1980 GDP per capita (2000 $) and 1980 distance from US. Income source: World Bank (2006).
If 1980 distance measures were related to income in later years, it would support the causal link from similarity to income. In fact, the correlation between later income and 1980 distance from the United States becomes even stronger over time. Figure 11.2 plots the correlation between 1980 distance and future income. The chart shows that income levels in 2003 are even more closely linked to 1980 distances than were 1980 incomes. The same kind of relationship holds for distances calculated for 1990 and 2002. We can modify Figure 11.2 by plotting the correlation of each of the three sets of distances with income occurring a given number of years before or after the year of the distance measure. For example, instead of plotting the correlation of 1980 distance with income in 1980, 1981 and so on, we can plot it for one year, two years and up to 23 years in the future; the 2002 distance measure would be plotted against one future year and 22 past years. This set of correlations is shown in Figure 11.3. The upper-right curve for 1980 in Figure 11.3 is the same as was plotted in Figure 11.2. The correlations indicate that the similarity of a country to the United States is better at predicting future than past income, independently of the date associated with the distance measure; this relationship is prima facie evidence that structure causes income. These observations recall the discussion among scholars studying economic development. A common finding is that countries do not experience unconditional income convergence; poor countries do not have an automatic advantage in catching up with richer ones. The ranks of poor countries do include all of the fastest growing economies, but backwardness is no guarantee of rapid growth; all the collapsing economies also are among the poorest. However, conditional convergence does seem to hold; if countries get their policies right, low income provides a growth advantage.
Correlation between distance from US and GDP/capita
Correlation between distance from US and GDP/capita
-13.00 -0.69 -12.00 -0.68 -11.00 -0.67 -10.00 -0.678 -0.67 184 Is Japan different? -9.00 -0.695 -0.68 -8.000 -0.687 -0.682 -0.85 -7.000 -0.694 -0.685 -6.000 -0.705 -0.685 -5.000 -0.709 -0.681 -4.000 -0.703 -0.688 -3.000 -0.708 -0.704 -0.80 -2.000 -0.713 -0.695 -1.000 -0.719 -0.697 0.00 -0.72 -0.717 -0.70 1.00 -0.72 -0.706 -0.71 2.00 -0.72 -0.706 3.00 -0.74 -0.714 -0.75 4.00 -0.75 -0.719 5.00 -0.76 -0.721 6.00 -0.75 -0.722 7.00 -0.76 -0.725 8.00 -0.77 -0.727 -0.70 9.00 -0.78 -0.739 1980 10.00 1984 1988 1992 2000 -0.78 -0.730 1996 11.00 -0.77 -0.732 Figure 11.2 Annual correlations of 1980 distance from US with countries’ real GDP per 12.00 -0.77 -0.733 capita, 1980–2003. 13.00 -0.78 -0.738 14.00 -0.78 -0.84 15.00 -0.79 16.00 -0.79 1980 distances 17.00 -0.79 -0.80 18.00 -0.80 19.00 -0.81 20.00 -0.80 -0.76 21.00 -0.81 22.00 -0.82 -0.82 -0.72 23.00 1990 distances -0.68 2002 distances -0.64 -24 -20 -16 -12
-8
-4
0
4
8
12
16
20
24
Years from date of distance measure
Figure 11.3 Annual correlations of 1980, 1990 and 2002 distances from US with countries’ real GDP per capita, 1980–2003.
The sample of countries analysed in this chapter, with a lower income cutoff of $5,000 GDP per capita, excludes the poorest in the world; nevertheless, if growth depended on initial income plus the adoption of good policies, proxied by the variables used to calculate distance from the United States, we would be observing a form of conditional convergence.
Is Japan different? 185 This idea can be tested by a regression equation with dependent variable of real GDP per capita growth and independent variables of initial income and distance from the United States. Since countries with at least $5,000 income have achieved a degree of economic success, we might expect that they have already put some effective policies into place. Two equations are shown below, one with initial income only, the second adding similarity to the US; specifically, the dependent variable is the compound growth rate of real per capita GDP from 1980 to 2003 and the independent variables are the log of income in 1980 and the squared 1980 distance from the US. Growth rate (1980–2003) = 8.3 – 0.7 log (1980 income); adjusted R2 = 0.06 (2.4) (1.8) Growth rate (1980–2003) = 22.4 – 2.0 log (1980 income) – 1.0 (5.6) (5.0) (4.8) (1980 distance)2; adjusted R2 = 0.42 n = 38; t statistics in parentheses The results of this equation mimic the usual finding that growth rates are negatively related to initial income; unconditional convergence in this selected group of countries is present, but weak compared with the effect when distance is added to the equation; that is, poorer countries tend to grow faster than richer ones when their performance is conditioned on similarity to the United States. The second equation shows that growth falls by one percentage point for countries one standard deviation away from the United States, and that the penalty rises with the square of the distance. This equation is plotted in Figure 11.4. This figure suggests that poor countries have a high payoff to imitating advanced country standards; the further they are from these norms, the more they suffer. Eventually, bad behaviour dominates the potentially positive effects of low income; negative growth is predicted for economies with $5,000 GDP per capita that are more than two standard deviations away from American behaviour. The fact that Japan’s distance from the United Sates has been increasing suggests that its income growth is likely to slow over the next decade or so, as it has already done since the 1990s. This analysis provides some backing for the idea that further deregulation, opening and competition will benefit Japan in the long term, at least with respect to income per person.
Conclusions A major surprise from this analysis is that Japan and the United States are drawing apart rather than coming together. The ending of the post-Second World War economic system in Japan had led me to expect that economic liberalization, market opening, greater foreign participation in the economy and more competition would have made Japan more like the United States. However, Japan is chasing
186 Is Japan different? 6
Growth rate of real GDP per capita
$5,000 4 2
$15,000
0 $25,000
-2 -4
$35,000 -6 -8 0.0
0.5
1.0
1.5
2.0
2.5
3.0
Distance from US
Figure 11.4 Real GDP per capita growth as function of distance from US and initial income (shown on chart), estimated from growth equation.
a moving target. The United States also has become more liberalized, more open, more subject to competition and more influenced by foreign investors and trade. Change at a pace that is comfortable to most Japanese may be insufficient to keep up with the more dynamic economies in the twenty-first century. Although I have used the United States as a convenient benchmark, other countries are changing in similar ways. Japan may find that even as it struggles to keep up, it is falling further behind. On a broader front, the results reported here are consistent with the findings of economists studying the growth process, for example Harvard’s Dani Rodrik (2004: 1): ‘Institutional quality holds the key to prevailing patterns of prosperity around the world’. Among the institutions listed by Rodrik are rule of law, democracy and openness to foreign trade and investment – all variables used in this chapter. He concludes that large-scale institutional transformation is hardly ever a prerequisite to getting things going. Nevertheless, in an assessment central to the findings reported above, ‘sustained economic convergence requires acquiring high-quality institutions. We need to distinguish between stimulating economic growth and sustaining it. Solid institutions are much more important for the latter than for the former’ (Rodrik 2004: 10). In a similar vein, 10 year earlier, Sachs and Warner emphasized the role of institutions in economic growth: Economic growth, and therefore economic convergence, requires reasonably efficient economic institutions. Poorly managed economies – such as those with the absence of secure property rights, autarkic trade policies, inconvert-
Is Japan different? 187 ible currencies, and so forth – are unlikely to experience convergence no matter what the underlying production technology or initial level of human capital. Sachs and Warner (1995: 5) The institutions and variables noted by these authors are those used in the present study. My search that began by attempting to determine if Japan is different from other countries has ended by featuring the kinds of behaviour that seem to be essential for continuing growth. We can go further; Japan did not arrive at its position among the largest, richest and most productive economies by accident. The virtues that brought it this far also establish the foundation for the future. Among those virtues is the ability to adapt; whether mutations to the system will occur fast enough for Japan to maintain its position among the world’s leading economies will be revealed over the coming years.
12 Economic prospects for Japan Moving to a new trajectory
Growth prospects The new trajectory Japan’s long-term economic prospects are good. One reason for optimism is that the economy is shifting its trajectory from the arc of the last 75 years. This is not to say that Japan has not changed before; it has always been evolving, despite the view of some observers that the shifting images are merely a mirage hiding a static reality. Nevertheless, there are many additional elements being stirred in the twenty-first century. • • • • • • •
Japan’s businesses and government have rid themselves of the illusion that renewed high-speed growth is just around the corner, maintained since the deceleration of growth in the 1970s and abandoned only in the 1990s. Business has shed most of the excesses that were accumulated before and during the asset bubble and collapse of the late 1980s and 1990s. Changes in the structure of industry, as shown in Chapter 9, accelerated in the 1990s, perhaps reflecting the overcoming of rigidities built up over the post-1945 period. Government bureaucrats’ predilections for market interventions, and the instruments to implement these desires, are waning. New corporate governance rules have created pressures for profitability and productivity. Deregulation of such important industries as finance, telecommunications, retailing and transportation have encouraged greater competition, broader ranges of products and services, and efficiency gains. The increase in foreign portfolio investment as well as direct investment in company ownership have added to the pressures to attend to the bottom line.
As a result of these many alterations to economic life, the nation is becoming more differentiated into winners and losers: across industries, regions, companies
Economic prospects for Japan 189 and people. Success and failure will be more clearly visible, which will have political implications for how these are dealt with. The various changes are not independent of each other. For example, the easing of entry restrictions into retailing enabled large foreign firms as well as domestic companies to enter the industry, invigorating competition while introducing capital and expertise to revive failing Japanese giants. At the same time, traditional small shops have disappeared by the thousands. I noted in Chapter 2 that the view we now have of dramatic rebirth during the period surrounding the Meiji Restoration was not directly apparent to those on the scene. Yet, 1880 was vastly different from 1860. Much the same can be said about the years from 1985 to 2005. Taken as a whole, the economic scene in Japan altered dramatically over these two decades. The fact that Japan has adapted to prolonged and multiple domestic and external challenges should come as no surprise to anyone who has observed this country, or, indeed, any rich nation. With many of the problems that retarded economic growth since the 1980s reduced or eliminated, Japan can now return to the growth path of most rich, mature economies: around 2 per cent annual growth rates of real GDP per capita. Long-term growth Optimism is warranted for Japan’s long-term growth prospects, a sentiment that arises neither from deep theoretical foundations nor from elaborate econometric estimations, but rather from crude pattern recognition. The pattern on which I focus is the growth experience of rich countries over the past 40 years. Figure 12.1 zooms in on a chart introduced in the first chapter. It shows the growth rate of real GDP per capita of non-oil-dependent countries with GDP per capita greater than $12,000. The figure plots annualized 10-year growth rates against the real per capita GDP at the beginning of the 10-year period. For present purposes, the interesting part of the figure is to the right, which portrays the fortunes of the world’s richer countries, including Japan. The relevant detail pertaining to Japan’s future is that annual growth of rich economies clusters in the 0–2 per cent range. Japan’s deceleration and decline in the 1990s brought its growth below America’s progress of 2 per cent since the 1970s. For reasons outlined below, one could reasonably expect that Japan’s future per capita growth will converge with the American experience. This projection is made despite Japan’s ageing population and slowly declining labour force. A reason for optimism is that the country possesses opportunities in its very shortcomings. For example, women make up a smaller share of the labour force than in many other advanced economies. At 60 per cent participation, it is more than 10 percentage points below countries such as Sweden, United States and United Kingdom. Moreover, women are used in inefficient ways; they are more likely than are men to be part-timers and they find it even more difficult than men to re-enter the labour force if they temporarily withdraw, as they are likely to do when they have children (Ono and Rebick 2003: 26).
190 Economic prospects for Japan 9
Growth of real GDP/capita (%)
8 7 6 5 4 3 2 1 0 -1 12,000
16,000
20,000
24,000
28,000
32,000
Real GDP per capita ($)
Figure 12.1 Average annual growth rate of real GDP per capita over preceding 10 years and GDP per capita (% and 2000 $ at purchasing power parity). Source: World Bank (2006).
The ratio of female to male wages in the late 1990s was 0.64, the lowest among 13 countries studied. Moreover, the proportion of females in private sector managerial positions rose from only 5 per cent in 1990 to 8.3 per cent in 2001; the entry of women into higher paying jobs compares poorly with other countries such as the United States, where women hold some 45 per cent of managerial positions (cited by Higuchi and Hashimoto 2005: 372). Personal observation suggests that there may be more competent, intelligent and hard-working Japanese women working at higher level jobs in New York or Washington than in Tokyo. It is no accident that such competent people leave their country to find opportunities elsewhere. When half the labour force is underutilized, the potential for greater productivity gains is obvious. Another disadvantage that can benefit Japan is that its productivity lags the global leaders. Various estimates put the country’s manufacturing productivity at roughly 70–80 per cent of the most efficient economies, varying from industry to industry. According to one study, for example, in 2000, value-added per hour worked in all of Japanese manufacturing stood at 72 per cent of the US level after peaking at 79 per cent in 1991. However, considerable variability existed across industries. Productivity in food products was only 30 per cent of the American level, whereas Japan’s metal products producers were 60 per cent more productive and transportation equipment was better by 26 per cent (Inklaar et al. 2003: 9–11). Manufacturing is among the most efficient sectors in Japan, but accounted for only one-quarter of total output in the 1990s. Across the economy as a whole, productivity is generally figured to be lower. A study that ranked countries against the technological frontier placed Japan’s relative efficiency at 60 per cent of the
Economic prospects for Japan 191 global standard in 1965 and 61 per cent in 1990 (Kumar and Russell 2002: 532). This result is fairly typical of such research, which shows Japan’s economy-wide productivity at two-thirds the American value. Lower productivity can be an advantage because better ways of doing things have already been demonstrated elsewhere. Japanese companies would not have to experiment with uncertain methods or technologies to achieve higher levels of efficiency, but have models on which to base their plans. Having such models, though, does not guarantee their speedy or successful adoption. It took a quarter of a century for American automobile companies to adopt the Toyota system of manufacturing, which was demonstrably more efficient than the way that Detroit had made things. And it required a similar length of time for Japanese airlines to reach American efficiency levels after deregulation and competition were introduced into that industry. Nevertheless, once pressure to change impinges on an industry, having others to follow can speed up the desired adjustments. Jumping into and crawling out of a banking crisis From the mid-1970s, Japan’s business and government failed to recognize and adjust to decelerating growth. Businesses indulged in over-investment, hired too many employees and borrowed too heavily. Banks lent too freely to overleveraged customers and acquired large amounts of loans that could not be repaid, especially after the collapse of a land and shares asset price bubble in the early 1990s. The financial authorities were complicit in allowing banks to cover up their non-performing loans, which were eroding bank capital. As the central government incurred ever-growing deficits, largely because of falling revenues from a weak economy, the authorities were reluctant to use aggressive stimulatory fiscal measures. The BOJ, confronted by deflationary pressures caused by excess supply relative to weak demand, was late in recognizing the problem and reluctant to use what it regarded as unorthodox monetary policies. The above list of problems confronting Japan’s business executives, political leaders and government bureaucrats would have challenged decision-makers in any country. In combination, in a country not known for its lightning speed of reaction to adversity, the Japanese response was particularly slow in coming. Nevertheless, almost all the problems had been reduced or eliminated by 2005. The banking crisis took just a year or so to develop, festered for a decade and required three years for a cure, once the political leadership decided to confront the problem. Its story can be retold in a few paragraphs. To begin, what happened in Japan occurred in many other countries. A count by World Bank economists in 2003 found 117 systemic banking crises (defined as exhaustion of almost all bank capital) occurring in 93 countries since the late 1970s, including in many of the most advanced economies (Caprio and Klingebiel 1996, 2003). Japan’s banking crisis followed a familiar path. As summarized by Posen (2000: 4–6), banking crises usually begin when a highly regulated and protected banking system undergoes gradual deregulation. Until the 1980s, Japan’s main bank system of finance and the convoy system of regulation continued to hold
192 Economic prospects for Japan sway. Under that regime, banks grew in size and complacency. However, with the onset of rapidly changing financial technology and gradual deregulation, banks found themselves in long-term decline, squeezed at both ends – by borrowers seeking better terms in deregulated credit markets and by savers looking for higher returns. In response, banks sought new customers, often lower quality borrowers, relying even more heavily than before on collateral, especially real estate and liquid assets such as company shares. Reinforcing this risky behaviour, partial deregulation restricted banks from entering potentially profitable new businesses such as selling investments to retail depositors. While this sequence of events was transpiring, the Finance Ministry’s regulatory bureaus did not recognize that more, not less, oversight was needed in the liberalized system. The banks’ balance sheets grew particularly dodgy because of lax accounting practices applied to the increasingly risky loan portfolios and the greater potential for borrowers to default on their loans. Lending based on real estate and shares promoted a price bubble in these assets until the central bank started raising interest rates in 1989. As interest costs went up, the economy slowed, and in 1990–92 borrowers could not repay their loans; the loans became non-performing. Regulators failed to force banks to comply with published standards as nonperforming loans mounted. Ministries and officials had few incentives to deliver bad news. In addition, the practice of bureaucrats finding post-retirement jobs in the financial sector inhibited regulatory discipline. Finally, all participants – regulators, bankers, politicians and investors – had little experience in dealing with the new phenomenon of risky and non-performing loan portfolios. They all had incentives to wait for the economy to improve and for asset prices to rise. Thus, they collectively engaged in a policy of ‘crossed fingers’, wishing for the best and following a policy of forbearance. When property prices did not rise, but continued a decline that would last for 15 more years, a new and dangerous stage of the crisis unfolded: evergreening. Banks lent additional amounts to borrowers to cover the interest and repayments that could not be repaid. Outstanding loans to real estate and construction industry borrowers continued to climb until 1998. Figure 12.2 illustrates the sequence of events. Real estate-based lending soared in the mid-1980s, reaching more than three times the 1980 level in 10 years. Share prices took off at an even faster rate, climbing almost six times between 1980 and 1990. Land prices took off more slowly than did the stock market and topped out a year later at a lower relative peak, but when property started to tumble, it continued longer and fell further than share prices. The telling feature of the figure is the performance of loans based on property, which continued to grow for another seven years after asset prices started their plunge. As non-performing loans piled up, the financial as well as political cost of any future rescue effort rose accordingly. The financial authorities’ recovery hopes were not helped by the fiscal and monetary decision-makers, who were reluctant to engage in stimulatory policies; more than that, a rise in the consumption tax in 1997 helped send the economy into recession. The hoped-for recovery never came.
Economic prospects for Japan 193 600 Share prices (Nikkei) 550 500
Property-based loans
Index, 1980=100
450 400 350 300 250 200
GDP
150 100 1980
Land prices 1984
1988
1992
1996
2000
2004
Figure 12.2 Real estate loans and asset prices in Japan, 1980–2007 (1980 = 100). Sources: loans: Bank of Japan; land prices: Japan Real Estate Institute. Note: Propertybased loans are made up of loans to real estate companies and construction firms. Land prices are the average for Japan’s six largest urban areas.
Two events occurred that finally forced the political leaders to pursue solutions. At the end of 1997, several major financial institutions went into bankruptcy, despite attempts by MOF officials to rescue them. According to interviews with bureaucrats and politicians, these events shocked them into a realization that the entire financial system was in danger. Two banking laws in 1998 under Prime Minister Keizo Obuchi played a major part in resolving the immediate problem of banks running out of capital, which had been exacerbated by poor information on the banks’ health; the legislation recapitalized potentially healthy banks and closed the unsalvageable ones. The second event was the selection of Junichiro Koizumi as prime minister in 2001. Breaking with LDP methods in general, Mr Koizumi appointed as his economics minister an economic professor, Heizo Takenaka. At the end of September 2002, Koizumi appointed Takenaka to head the Financial Services Agency, the banking watchdog split off from the MOF. Takenaka directed major banks to reduce their non-performing loans. As shown in Table 12.1, large banks drew down their bad loans from more than 8 per cent of outstanding lending in 2002 to 1.5 per cent in September 2006 by writing them off their books and restructuring viable borrowers (earlier data are unreliable.) The banks were forced to pay for these measures by reducing bank capital, which put many banks at or below required minimum levels. A rising stock market and recovering economy beginning within months of Takenaka’s hard-line policies finally brought some longed-for help to the banking sector. Regional banks, however, were not pushed as fast or as hard by the Financial Services Agency to clean up their books. They
194 Economic prospects for Japan Table 12.1 Non-performing loans in Japan’s banks, 1999–2006 End of period March 1999 September 1999 March 2000 September 2000 March 2001 September 2001 March 2002 September 2002 March 2003 September 2003 March 2004 September 2004 March 3005 September 2005 March 2006
Large banks (% of all loans) 6.1 5.7 5.4 5.1 5.3 6.2 8.4 8.1 7.2 6.5 5.2 4.7 2.9 2.4 1.8
Regional banks (% of all loans) 6.2 6.2 6.2 7.1 7.3 7.7 8.0 8.3 7.8 7.5 6.9 6.3 5.5 5.2 4.5
Total value (trillion ¥) 33.9 31.3 31.8 32.9 33.6 36.8 43.2 40.1 35.3 31.6 26.6 23.8 17.9 15.9 13.4
Source: Financial Services Agency.
improved, but as of early 2006, were running several years behind their big-city brothers. Delay in addressing the banking problem was costly to the government and to the nation. One estimate put the fiscal outlays required to pay insured depositors and rescue failing banks at 24 per cent of GDP, compared with an average across crisis countries of 15 per cent. The bank-associated slowdown in growth chopped an estimated 48 per cent from national output, accumulated over 12 years, considerably greater than the average 12.5 per cent (Claessens et al. 2004: 37–8). Gaining corporate health During the post-war years of high-speed growth, companies scrambled to keep up with demand by building new factories and getting them running, hiring and training new workers, and rounding up the funds to finance their expansion. The talents rewarded in these efforts and the routines learned by managers and their bankers became embedded in the norms and expectations of business and finance. As growth slowed in the 1970s, those experiences had diminishing and then negative value. Another 20 years were to pass before behaviour would adapt to new circumstances. The natural inclination to continue with successful methods was reinforced by the highly regulated banking community that made its money from the spread between deposits and lending rates – a spread guaranteed to be profitable for even the weakest banks by the MOF as it administered a convoy system of bank regulation in which all banks sailed together, charging the same rates to customers on similar products and paying the same interest to depositors. It was a system with few financial products and little innovation in which no bank failed. This environment produced bankers who did not know how to evaluate business plans
Economic prospects for Japan 195 or assess the risk of new projects. The consequence was over-investment and low returns. Amplifying these outcomes was a system of corporate governance that paid scant attention to shareholders, plus the norm of lifetime employment, which constrained the adjustment to slower growth. The results were predictable: overleverage, over-investment and over-employment. They were intensified by the asset bubble, but would have emerged as pressing problems eventually, with or without the bubble. While these problems were building, the regulatory world of banking and the legal structure of corporate governance had begun to change, as described in Chapter 7. The cumulative impact of such mutations, when coupled to strengthened financial markets, greater foreign participation in the economy and slower growth, finally forced companies to look after their health. The outcome was a Japanese business sector that was considerably stronger in 2006 than it was even a few years earlier. Companies wrote off and scrapped unproductive investment, used their cash flow to pay off debt and bring leverage (the ratio of debt to equity) down to levels not seen since the 1960s, and reduced employment through attrition, hiring freezes, retirement and outright layoffs. With costs and investment under control, business rates of return turned up for the first time in almost half a century. Figure 12.3 shows the value of tangible capital stock on the books of large Japanese manufacturing and non-manufacturing companies, unadjusted for price changes. Although it required several decades after growth slowed in the 1970s, Japanese corporations finally slowed their net investments and then reduced their capital stock as managers scrapped and wrote off capital, or allowed depreciation to reduce its value. On company balance sheets, the counterpart of assets such as investments is 105 Non-manufacturing
90
Trillion ¥
75 60 45 Manufacturing 30 15 0 1970
1974
1978
1982
1986
1990
1994
1998
2002
2006
Figure 12.3 Tangible capital stock on the books of large manufacturing and nonmanufacturing firms, 1970–2006 (trillion ¥). Source: Ministry of Finance, Financial Statements Statistics of Corporations.
196 Economic prospects for Japan liabilities. Japanese companies had borrowed too much in the form of interestbearing debt – loans and bonds – to withstand a slowdown of growth, revenues and profits. Figure 12.4 shows one representation of this excess: private non-financial companies’ loans and bonds as a ratio to GDP. This ratio fluctuated between 0.8 and 0.9 from 1965 to 1985. Indebtedness relative to total national output then jumped 50 per cent over the next 12 years before companies started to reduce their borrowing in 1997–98. By devoting cash flow to debt payback, the debt–GDP ratio fell back to 1960s levels within a few years. Investments and loans were not the only things that Japanese companies acquired too much of. Despite economic slowdown in the 1990s, companies continued to hire staff at a high rate (Figure 12.5). Job growth plateaued in 1992, as companies grew uncertain of their prospects. It was not until 1997, though, that they realized that projected sales did not warrant the number of employees that they had on staff. Once the process of labour force reduction began, it took another five years until companies reached an equilibrium level of desired staff. Only in 2002 did employment start to rise. Another way of describing the labour force build-up is that labour productivity declined drastically. This decline can be illustrated by examining the difference between the growth rates of GDP and the labour force. During the 1970s, output grew 8–9 percentage points faster than labour inputs (Figure 12.6). When deceleration of output growth occurred in the 1970s, the rate of increase of the labour force did not slow as fast; consequently, the difference between output and employment growth dipped to 2–3 percentage points. In the 1990s, further slippage occurred. A turnaround then came about in the early 2000s as companies slowly returned to a desired level of jobs, given expected future growth prospects. The extended adjustment to a more efficient and flexible labour force included dimensions other than just reducing the number of workers. The Organization for 1.2
Debt/GDP
1.1
1.0
0.9
0.8
0.7 1964 1968 1972 1976 1980 1984 1988 1992 1996 2000 2004
Figure 12.4 Ratio of interest-bearing debt to GDP, non-financial companies, 1964–2006. Source: Bank of Japan, Flow of Funds. Note: Interest-bearing debt equals loans plus bonds.
Economic prospects for Japan 197 66 Employment, millions
65 64 63 62 61 60 59 58 57 1983
1986
1989
1992
1995
1998
2001
2004
2007
Figure 12.5 Employment (millions, seasonally adjusted, January 1984 to October 2006). Source: Ministry of Internal Affairs and Communications, Statistics Bureau, Labour Force Survey. 10
Growth difference (%)
9 8 7 6 5 4 3 2 1 0 1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
Figure 12.6 Growth rate differential between real GDP and total employment (five-year average, %). Source: Ministry of Internal Affairs and Communications, Statistics Bureau, Labour Force Survey.
Economic Cooperation and Development (OECD) reckoned that by some measures, the Japanese labour market was beginning to look more like the European model with increasing dualism between so-called regular employees and nonregular employees, including part-timers and short-term contract workers (OECD 2005: 166). From 1990 to 2003, the proportion of regular workers fell from close to threequarters of non-management employees to two-thirds, whereas non-regulars jumped from 18.8 to 28.1 per cent. Although the rise of the non-regular labour force gives much needed flexibility to firms, the increasing dualism introduced threats to long-term growth. Non-regular employees were paid only 40 per cent
198 Economic prospects for Japan of the wages of their counterparts and were granted few benefits; they received little in-house training and had no sustained commitment to the firm. The lack of movement between regular and non-regular workers also created equity concerns and possible lifelong wage gaps. The danger to the economy is that such employees obtain little of the experience and human capital needed to expand the productivity required to counter the effects of a shrinking labour pool (OECD 2005: 173). As in many European economies, the chief reason given for the rise of nonregular employment is the difficulty of reducing the number of regular workers in the face of variable demand; the harder it is to dismiss employees, the more reluctant firms are to hire them in the first place. Although Japanese law seemingly gives companies the right to dismiss workers, courts have set severe conditions on the practice. As a result, new labour market entrants have borne much of the burden of employment adjustment. The OECD (2005: 180) recommends explicit changes to the law that increase the flexibility of hiring and dismissing regular employees.
End of the 1930–45 economic system The policies noted above are part of a larger movement in Japan: disappearance of the 1930–45 system of wartime economic management. Legitimate arguments persist on either side of the question of whether the planning and guidance exercised by the Japanese government in the years immediately after the war contributed positively or negatively to post-war reconstruction and growth. If these policies and approaches held back the economy, the net impact could not have been very large, considering the rapid pace of reconstruction and subsequent expansion. However, most observers tend to agree that the contributions since the 1970s have been largely negative. Therefore, the demise of the wartime legacy should have positive effects on the economy. Several forces are ending the influence of the wartime system. Gradual deregulation of the financial system and greater participation of foreign capital have reduced the willingness of companies to accept government guidance. Legal changes, including the ability of shareholders to bring suits against managers who act contrary to shareholder interests, further constrain firm behaviour. Changes in the electoral law in 1994 that eliminated multimember Diet districts reduced the influence of narrow interests on politicians, who now must appeal to broader constituencies. The attack by Prime Minister Junichiro Koizumi on the financial and organizational bases of the LDP factions and the strengthening of the prime minister’s office have reduced the special-interest clientelism practised by politicians through government ministries involved in public construction, post offices and a multitude of other sectors. A law that came into force in 2006 allows the Fair Trade Commission to grant exemption from penalties to the first whistleblower to report bid-rigging or cartel violations. If the experiences in the United States or Italy are valid, this law should make a great change in the way that certain industries in
Economic prospects for Japan 199 Japan do business, particularly construction and many industrial materials; prices should fall and the incentives for greater productivity should increase. One prominent structural heritage of the wartime system was bank-centred finance. Even here, we are witnessing changes in the way that banks operate as they become more oriented to the business plans and profitability of their borrowers; they are relearning the basic role of bankers in evaluating their customers. In addition, banks are trying to move away from their traditional profit base of interest rate spreads between deposits and loans and are targeting services other than the provision of loans to large borrowers. Nevertheless, Japan continues to be over-banked, a structural feature that is likely to remain for some time. The many changes noted above give me confidence that the Japanese economy has escaped from burdens accumulated over the past 50 years, and especially the residue of the asset bubble collapse of the early 1990s. Just as important, the drag on productivity that accompanied government regulation and oversight of large sectors of the economy are being removed or weakened. Greater openness and competition should bring higher growth and faster productivity improvements to the Japanese people. However, accompanying these changes, individuals will face more unpredictability and greater differentiations into winners and losers, a challenge that government and private institutions will have to confront.
Emergence of winners and losers An email from a former Japan Economic Institute colleague, Douglas Ostrom, living in Tokyo in 2006, gives a ground view to one of the major changes occurring in Japan – differentiation. A slightly edited version of his comments follows: Like many infrequent visitors to Japan, I was initially surprised at how prosperous the country seemed, despite 15 years of dreary news reports. Japan remains a rich country and in many respects it seems to be better off than when we lived here 10 years ago. What caught my eye in a recent New York Times article, however, was the statement that ‘restaurants are packed.’ They are not! We have had very much the opposite experience. In restaurant after restaurant we go into of all types, we are surprised at the lack of other customers. Many restaurants are only 10 per cent full at times when one would expect them to be busy. Their low prices have surprised us, in part because the yen is now so weak. Nor are stores very crowded in general. I have also been surprised at the number of empty lots, apparently abandoned houses, and nearly empty apartment units. Next to our house in affluent Kunitachi, for example, is an empty lot and next to that is an old apartment that seems to be mostly empty. I have also been surprised at the number of boarded-up businesses in several different places I have visited. The one exception to this pattern seems to be downtown Tokyo and glitzy shopping areas in a few other places. For example, one day we did venture into Mitsukoshi of Ginza and could barely make our way through the store for the crowds. I couldn’t believe it was not some incredibly special event,
200 Economic prospects for Japan but I saw no evidence of that. I am wondering whether the ‘packed restaurant’ the reporter saw was in Ginza. I am wondering if what I am seeing is some combination of restructuring and the lack thereof. As regulations have loosened, companies have become freer to open new, typically large-format, stores and restaurants. There are many of these now. They have driven older competitors out of business (the empty storefronts) or to the wall (uncrowded, cheap restaurants). But the new places do not seem too busy either and they, too, are cheap. So the restructuring has been real but it is also incomplete. Change hits retailing Since it was first published in 2000, I have used a chart produced by the McKinsey Global Institute, the research arm of the business management consultants, to illustrate the absence of dynamic change in Japanese business. Table 12.2 reproduces their data on the sales ranks of the 10 largest general merchandise stores over a 15-year period in Japan and 10 years in the United States. The remarkable feature of the Japanese rankings was their constancy over the 15 years. Among the top five companies only mild reshuffling occurred. The American companies, over a shorter span, saw two companies in the top 10 that had not existed a decade earlier, and the number one company in 1993, Wal-Mart, had jumped from the 17th position. This kind of stability among companies is found not just in retailing, but is common across Japanese industries. Therefore, it is informative to revisit the McKinsey list. Daiei and Mycal went into bankruptcy; Marubeni and an equity partner acquired Daiei while Mycal was bought by Aeon (a renamed Jusco, converted into a holding company of a broad range of retailers). Seiyu and Seibu also flirted with bankruptcy; Seiyu came under the control and management of WalMart and Ito-Yokado took over Seibu. These changes in ownership and control were accompanied by large-scale store closures. On the American side, there was the usual amount of churning produced by consolidation in the retail industry. Among other changes, Sears Roebuck bought K-Mart, which slipped to number Table 12.2 Sales rank, general merchandise stores, Japan and US, 1983–98
Daiei Ito–Yokado Jusco Mycal Takashimaya Seiyu Uny Mitsukoshi Seibu Marui
Japan 1983 1 2 4 5 7 3 10 6 9 13
1998 1 2 3 4 5 6 7 8 9 10
Source: McKinsey Global Institute (2000: 30).
Walmart Stores Sears Roebuck K-Mart Dayton Hudson J. C. Penney Home Depot Kroger Safeway Costco American Stores
US 1983 17 1 2 12 5 – 4 3 – 9
1993 1 2 3 4 5 6 7 8 9 10
Economic prospects for Japan 201 nine in 2005 despite the takeover. Albertsons bought American Stores and jumped four positions while Home Depot took over the second slot. What is occurring in Japanese retailing is beginning to look more like the American model. Growth differences across industries In addition to greater turmoil among the relative fortunes of firms, instability appeared also at the broader level of entire industries. As described in Chapter 9 on structural change, industry performance became more differentiated in the 1990s. One way to measure such variability is with the coefficient of variation: here, the standard deviation of 10-year industry change in GDP divided by average GDP change. This variability measure across 42 industries (charted in Figure 12.7) shows how the distribution changed over 25 years. The normalized standard deviation was fairly stable until the 1990s and then climbed rapidly. The sharp rise was not merely a matter of slower overall growth, which reduced the size of the denominator. Growth, after all, had been slowing since the 1970s with no impact on the normalized dispersion of outcomes; the average and standard deviation of industry growth had climbed in tandem. This linked pattern dissolved in the 1990s when the variability across industries continued to grow while the average change slowed and then declined. To clarify what is being demonstrated in the industry data, it is important to note what is not changing. Critically, the distribution of industry GDP shares was stable from 1970 to 2004. Stable output shares together with rising growth variability would be consistent with the following scenario. Suppose that the steel industry accounted for 30 per cent of total output for the 20 years from 1970 to 1990, and that electronics produced 10 per cent. If, in the 1990s, electronics jumped from 10 per cent to 30 per cent and steel fell from 30 per cent to 10 per cent, the distribution of industry shares would remain constant. In contrast, the variability of changes in output would produce big jumps in the later period. In 4.0
Coefficient of variation
3.5 3.0 2.5 2.0 1.5 1.0 1980
1984
1988
1992
1996
2000
2004
Figure 12.7 Coefficient of variation of 10-year GDP change across 42 industries (standard deviation/mean).
other words, the distribution of large and small industries has remained constant, but there is much more turmoil; as in the retail sector, the relative rankings of industry size are now shifting more than previously. Land price variability
Kanagawa prefecture
100 90.3 85.7 81.4 77.3 73.5 70 67.6 65.5 63.9 62.8 62.2 62.1
Another example in which greater variability is observed is land prices, which declined for 14 years after 1991. This decline was widespread, occurring in every prefecture and for every type of land. However, despite the ubiquity of falling prices, it was not uniform. To consider just the Tokyo region, large differences appeared since 2000. A construction boom in central Tokyo attracted commercial and residential customers from less desirable outlying areas. As a result, prices of Saitama Chibamore diverse. land have become Pref.AccordingPrefecture to data published by the Japan Real Estate Institute, land prices rose in the 24 metropolitan wards of Tokyo in 2005, but not elsewhere in the region. Figure 12.8 shows commercial land prices in Tokyo’s 24 metropolitan wards and the surrounding prefectures of Saitama, Chiba and Kanagawa. Land prices in Chiba, to the east of Tokyo Bay, fell by 45 per cent from 2000 to 2006, 100 100 whereas central Tokyo was down a net 7.5 per cent, increasing by 14 per cent 91 84.8 from the 2004 trough. The 87 78.5 other neighbouring prefectures lie between these two extremes. 83.6 73 80.4 77.5 75.3 73.5 72 70.8 100 70 69.7 70.4
67.9 63.3 60 57.7 55.4 54.1 53.7 53.8 54.4
90 Index, 2000=100
0 1 4 1 8 7 3 4 3 7 3 6 5
202 Economic prospects for Japan
Tokyo
80 Saitama 70
60
Kanagawa Chiba
50 2000
2001
2002
2003
2004
2005
2006
2007
Figure 12.8 Tokyo region urban commercial land price index (2000 = 100). Source: Japan Real Estate Institute, Urban Land Price Index.
Economic prospects for Japan 203 Rising personal income inequality Considering the increased differentiations across firms, industries and regions, it should not be surprising to find similarly rising diversity across individual incomes. Surveys that assess income distributions in Japan agree that inequality has been increasing since the 1980s. In contrast, Japan’s reputation in the 1970s was that of possessing the most equal income distribution of all OECD members. More recent OECD comparisons found that Japan ranked number 14. Only the United States, Mexico and Turkey had more unequally distributed income. Several forces drove change in Japan. Chief among them was the ageing of the population. However, even among the working-age population, inequality rose from the 1980s. A larger proportion of people older than the customary retirement age of 65 affects income distribution in three separate ways. First, older people’s incomes are lower than those of the working-age population; therefore, as the share of the older population grew from 10 per cent of the total in the mid-1980s to 17 per cent in 2000, income inequality increased because of larger between-group differentials. Second, the within-group inequality of the over-65s is greater than among working-age people, largely reflecting fewer numbers of working elderly. Third, even among those working, wider income disparities arose in the older group of workers (OECD 2006: 98). The other major influence on inequality was the rise of non-regular workers in the labour force. They received lower wages and fringe benefits, and worked fewer hours than so-called regular workers. From 19 per cent of employed persons in the early 1990s, they grew to account for more than 30 per cent in 2004 (OECD 2006: 99). As described in Chapter 9, the Gini coefficient summarizes the inequality present in distributions. It is based on the Lorenz curve, which plots cumulative shares of the population, from poorest to richest, against the cumulative share of income that they receive. The Gini coefficient ranges from 0 in the case of perfect equality to 1.0 in the case of perfect inequality. Table 12.3 shows OECD estimates of Gini coefficients for Japan. OECD statisticians have selected national surveys that maximize the comparability of inequality across countries and over time. The Japanese data come from the Comprehensive Survey of Living Conditions conducted by the Ministry of Health, Labour and Welfare. The data, based on household income, are adjusted to account for Table 12.3 Gini coefficients for market and disposable income across population groups, 1985–2000 Market income 1985 1995 0.317 0.369 0.309 0.338
2000 0.410 0.362
0.575
0.629
Total population Working-age population Elderly population 0.473
Source: OECD (2006: 99, 104).
Disposable income 1985 1995 0.278 0.295 0.276 0.290
2000 0.314 0.310
204 Economic prospects for Japan differences in household size, variations of which can distort the interpretation of inequality measures. The table reports coefficients for market income and disposable income. Since disposable income is measured after taxes and transfer payments, both of which tend to reduce income disparities, inequality should be lower than for market income. According to these results, inequality increased for both kinds of income as well as for working-age and elderly persons. Even after accounting for the redistributive effects of taxes and transfer payments in the working-age population, inequality rose over the 15-year period. Although the impact of tax and social spending reduced inequality between the mid-1980s and 2000, the growing disparities of market income more than offset the effects of these policies (OECD 2006: 103). Figure 12.9 includes a longer time series for disposable income than was used by the OECD. Also shown for comparison is the OECD’s market income measure, adjusted for family size. The figure also brings in another measure of inequality for comparison: the share of national income reported by the top 5 per cent of taxpayers. The taxpayer data are for individuals rather than households and reflect total income, including all income items shown on tax returns and before all deductions: salaries and wages, small business and farm income, partnership and fiduciary income, dividends, interest, rents and royalties, but excluding capital gains. The denominator is total personal income as reported in the national accounts (Moriguchi and Saez 2005: 7). The income accounted for by the top 5 per cent of taxpayers rose sharply after 1980, swelling by 5 percentage points to almost a quarter of all income. This 0.420
25 Gini, disposable income
0.370
24
Gini, market income
23
0.345
22
0.320
21
0.295
20
Income share (%)
Gini coefficient
0.395
Top 5% income share (right scale) 0.270 19 1962 1967 1972 1977 1982 1987 1992 1997 2002
Figure 12.9 Measures of income inequality: Gini coefficient of household market and disposable income (left scale); and national income share of top 5% individual taxpayers (right scale). Sources: MHLW; OECD 2006; Moriguchi and Saez (2005: Table 3).
Economic prospects for Japan 205 ascent more or less paralleled the rise in the Gini coefficient. The story in Japan, though, is not of the super rich at the very top of the income pyramid, but rather of those just below the top; the gains in income shares appear only in the 95–99 percentiles. That is, the very top tiers of Japanese taxpayers did not increase their piece of the economic pie. This result stands in contrast to incomes of top earners in the United States. The American growth in inequality came in the ranks above the top 1 per cent; for example, the top 0.01 per cent of American taxpayers claimed 0.5 per cent of personal income in 1980 compared with 3 per cent in 2000, whereas the same group in Japan remained at 0.5 per cent over the same period (Moriguchi and Saez 2005, Figure 4). The main story told by Moriguchi and Saez is not the shifts after 1980, but rather the very large decline in inequality from 1935 to 1945. The top 1 per cent accounted for one-fifth of all personal income in 1938, but this figure fell abruptly and precipitously to under 8 per cent within 10 years. The authors attribute this dramatic fall mainly to the collapse of capital income caused by wartime taxation, destruction and hyperinflation. Labour income concentration also declined, probably because of wartime wage controls, in less dramatic fashion. The top share remained low in the post-war years because of the break-up of zaibatsu holdings and the introduction of progressive income and inheritance taxes. Confining their attention to only labour income, these two authors find that Japanese top wage income shares, in contrast to the United States, remained relatively stable and low after 1970. Because wages became the major component of income after the war, the fall in wage income inequality contributed to the permanent decline in income concentration. They suggest that institutional factors such as corporate governance and internal labour markets were important determinants of Japan’s wage equality (Moriguchi and Saez 205: 3). Now, however, corporate governance is shifting to a more shareholder-based, capital market-oriented model in which returns to investors are regaining a hold. At the same time, inter-firm hiring is acquiring acceptance, especially for higher managerial and technical skills. The forces that helped to cap incomes in the postwar years may be evaporating, a phenomenon that seems to be reflected in the higher relative earnings at the top of the income structure.
Implications of growth with differences Greater variability of economic fortunes may be becoming the new norm. It is yet unclear whether this phenomenon is a temporary adjustment of the economy to more liberalized constraints, or whether it represents a permanent feature of a less fettered economy. A future with greater variance will require policies that allow and encourage greater mobility of people, capital and other resources so that new opportunities can be exploited and unprofitable ventures can be abandoned. Such policies would include not discouraging the exit of insolvent firms and declining industries, loosened employment practices, more portable pensions and more flexible housing markets. It would also require the construction of needed infrastructure in new areas and abandonment in declining ones. It means that public
206 Economic prospects for Japan policy to assist individuals suffering from misfortune must be sharpened, so that the affected people can be helped without creating incentives for the more fortunate to take advantage of public assistance. The instinct of government policymakers and politicians is to preserve old arrangements, to subsidize declining industries and to underwrite regions with few prospects. That approach to dealing with differentiation will become even more costly in the future than it has been in the past because there will be fewer resources to distribute as well as more cases of decline, even as new possibilities for growth arise. The tension will be between assisting individuals versus preserving larger collectives such as industries and regions. Policies that enhance productivity growth will make life riskier for individuals. Programmes that attempt to preserve predictability will retard productivity and growth, and reduce the resources that can be delivered to unlucky individuals. Not only is the economy of Japan changing, but the pressures on politicians to adapt to this more differentiated world will also be keenly felt. The arc describing Japan’s economy and economic policymaking that has characterized so much of the past may be more like a roller coaster ride in the future.
Appendix Variables and their sources used in Chapter 11 statistical analyses, years of use, and use in factor analysis
Variable 1A General government consumption as share of total consumption 1B Transfers and subsidies as a share of GDP 1C Government enterprise, investment as share of gross investment 1Di Top marginal income tax rate 1Dii Top payroll tax rate 2 L egal system and property rights 2A Judiciary independence 2B Impartial courts 2C Protection of intellectual property 2D Military in politics 2E Law and order 3A Average growth of money minus growth of real GDP 3B Standard deviation of annual inflation 3C Most recent inflation 3D Freedom to own foreign currency bank accounts 4Ai International trade tax revenues (% of total trade) 4Aii Mean tariff rate 4Aiii Standard deviation of tariff rates 4Bi Hidden import barriers 4Bii Costs of importing 4C Actual vs. expected size of trade sector 4D Difference between official and black mkt exchange rates 4Ei Access to capital markets; FDI restrictions 4Eii Index of capital controls 5Ai Ownership of banks x x x x x x x x x x x
x x x x x x
x x x x
x
x x x
x
Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004
x x
x x
Gwartney and Lawrence 2004 Gwartney and Lawrence 2004
1990 x
Used in 1980 x
Source Gwartney and Lawrence 2004
x x x x x x x x x x x x x x x x x x x
x x
x x
2002 x
x x x
Factors
5Aii Competition in domestic banking 5Aiii Extension of credit 5Aiv Interest rate regulations 5Av Interest rate controls 5Bi Impact of minimum wage 5Bii Hiring and firing practices 5Biii Wages set by centralized collective bargaining 5Biv Unemployment insurance 5Bv Use of conscripts 5Ci Price controls 5Cii Administrative conditions on new business entry 5Ciii Time with government bureaucracy 5Civ Starting a new business 5Cv Irregular payments Deposit bank assets to central bank + deposit bank assets Liquid liabilities to GDP Central bank assets to GDP Deposit money bank assets to GDP Private credit by deposit money banks to GDP Credit by deposit money banks and other financial institutions to GDP Bank deposits to GDP Financial system deposits to GDP Concentration of three largest banks Overhead costs to total assets Net interest margin Life insurance premiums to GDP Non-life insurance premiums to GDP Stock market capitalization to GDP Stock market total value traded to GDP Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999
Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Gwartney and Lawrence 2004 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Beck et al. 1999
x x x x
x x x
x x
x x x x x x
x x x x x x x x
x x x x x
x
x
x x x
x x
x x
x x x
x
x
Continued on next page.
x x x x x x x x x
x x x x
x x x x x x x x x x x x x x x
Source Beck et al. 1999 Beck et al. 1999 Beck et al. 1999 Kaufmann et al. 2005 Kaufmann et al. 2005 Kaufmann et al. 2005 Kaufmann et al. 2005 Kaufmann et al. 2005 Kaufmann et al. 2005 Transparency Intl. 2004 Deininger and Squire 2004 Freedom House 2004 Freedom House 2004 Freedom House 2004 Mauro 1995 Mauro 1995 Mauro 1995 Knack and Keefer 1995
Variable
Stockmarket turnover ratio Private bond market capitalization to GDP Public bond market capitalization to GDP Voice and accountability Political stability Government effectiveness Regulatory quality Rule of law Control of corruption Corruption perceptions index Gini coefficient of income inequality Political rights Civil rights Press freedom Index of low corruption Index of low red tape) Index of legal system efficiency Corruption index x x x x
x x x x x x x
x x x
x x x
1990
x
Used in 1980 x x x x x x x x x x x x x x
2002
x x x
x x x x x
Factors
Notes
1 Japanese names customarily place the family name first. That practice is followed for this page only because of the traditional method of naming the shoguns. All other references in this book use the western tradition of family name last. 2 Comprehensive, unified and revised estimates of economic data were gathered by a group of economists at Hitotsubashi University starting in 1965. This collection was published over several years in a series of volumes known as the Estimates of Long-term Economic Statistics of Japan since 1868, commonly referred to as LTES. The most important series in a national accounting framework were presented in one book with 145 pages of tables, including revised data and some not yet published in the multi-volume series (Ohkawa et al. 1979). 3 This chapter has benefited from comments made on an earlier draft at the Japan Economic Seminar at Columbia University, September 2004, especially by Lee Branstetter, Edward Lincoln and David Weinstein. 4 Generating a consistent GDP time series from 1970 to 2005 required stitching together data with different base years. The published real 1970–79 GDP figures for electrical machinery were bizarre and inconsistent with other information. To produce more reliable data for this one industry, I linked the published 1980 value to the growth rates given in the Japan industrial productivity database (Fukao: 2003).
Bibliography
Abel A., Mankiw, G., Summers, L. and Zeckhauser, R. (1989) ‘Assessing Dynamic Efficiency: Theory and Evidence’, The Review of Economic Studies, 56: 1–19. Alexander, A. (2002) In the Shadow of the Miracle, Lanham, MD: Lexington Books. Akino, M. and Hayami, Y. (1974) ‘Sources of Agricultural Growth in Japan, 1880–1965’, The Quarterly Journal of Economics, 88: 454–79. Asian Wall Street Journal (12 July 1994) ‘Japan Tobacco Reception is Tepid as Analysts Await Privatization’. —— (1 September 1994) ‘Japan Tobacco Pricey? It May Not Be for Long’. —— (30 August 1996) ‘Price of JR West is Lower than Forecast’. Australian Financial Review (27 October 1986) ‘Fast-Track for Privatization Program, Japan’. Baerwald, H. (2003) The Occupation of Japan as an Exercise in ‘Regime Change’: Reflections after Fifty Years by a Participant, Japan Policy Research Institute, JPRI Occasional Paper 29, February 2003. Bank of Japan, Loans and Discounts Outstanding by Sector. —— Flow of Funds, Financial Assets and Liabilities, Private Nonfinancial Corporations. Beasley, W.G. (1987) Japanese Imperialism: 1894–1945, Oxford: Oxford University Press. Beck, T., Demirgüç-Kunt, A. and Levine, R. (1999) A New Database on Financial Development and Structure. Policy Research Working Paper 2146. Washington, DC: World Bank. Beplat, T.E. (1980) ‘Comments’, in L.H. Redford (ed.) The Occupation of Japan: Economic Policy and Reform, Norfolk, VA: The MacArthur Memorial. Bernanke, B.S. and Gurkaynak, R. (2001) Is Growth Exogenous? Taking Mankiw, Romer, and Weil Seriously, Cambridge, MA: National Bureau of Economic Research, Working Paper 8365, July 2001 Bisson, T.A. (1954) Zaibatsu Dissolution in Japan, Berkeley, CA: University of California Press. Bronfenbrenner, M. (1950) ‘Four Positions on Japanese Finance’, Journal of Political Economy, 58: 281–8. Business Week (31 August 1987) ‘Taking a Flier on JAL’. Calder, K.E. (1990) ‘Public Corporations and Privatization in Modern Japan’, in E.N. Suleiman and J. Waterbury (eds) The Political Economy of Public Sector Reform and Privatization, Boulder, CO: Westview Press.
Bibliography 213 —— (1993) Strategic Capitalism: Private Business and Public Purpose in Japanese Industrial Finance, Princeton, NJ: Princeton University Press. Caprio, G. and Klingebiel, D. (1996) Bank Insolvencies: Cross-Country Experience. Policy Research Working Paper 1620. Washington, DC: World Bank. Choi, E. (1991) ‘The Breakup and Privatization Policy of the Japan National Railways, 1980–87: A Case Study of Japanese Public Policymaking Structure and Process’, unpublished PhD dissertation, Ohio State University. Claessens, S., Klingebiel, D., and Laeven, L. (2004) Resolving Systemic Financial Crises: Policies and Institutions. Policy Research Working Paper 3377. Washington, DC: World Bank. Clark, G. (1987) ‘Why Isn’t the Whole World Developed? Lessons from the Cotton Mills’, Journal of Economic History, 47: 141–73. Cobb, C. and Douglas, P. (1928) ‘A Theory of Production’, American Economic Review, 18 (supplement): 139–65. Cohen, J.B. (1949) Japan’s Economy in War and Reconstruction, Minneapolis, MN: University of Minnesota Press. Collins, S.M. and Bosworth, B.P. (1996) ‘Economic Growth in East Asia: Accumulation versus Assimilation’, Brookings Papers on Economic Activity, 1996: 135–91. Coox, A.D. (1988) ‘The Effectiveness of the Japanese Military Establishment in the Second World War’, in A.R. Millett and M.W. Murray (eds) Military Effectiveness. Vol. 3. The Second World War, Boston, MA: Unwin Hyman. Crawcour, S. (1974) ‘The Tokugawa Period and Japan’s Preparation for Modern Economic Growth’, Journal of Japanese Studies, 1: 113–25. —— (1989) ‘Economic Change in the Nineteenth Century’, in M. Jansen (ed.) The Cambridge History of Japan, Vol. 5. The Nineteenth Century, Cambridge: Cambridge University Press. Deininger, K. and Squire, L. (2004) Measuring Income Inequality: A New Database, Washington, DC: World Bank. Dore, R. (1958) ‘The Japanese Land Reform in Retrospect’, Far Eastern Survey, 27: 183–8. —— (1965) Education in Tokugawa Japan, Berkeley, CA: University of California Press. Dower, J. (1999) Embracing Defeat: Japan in the Wake of World War II, New York: W.W. Norton & Co. Droppers, G. (1898) ‘Monetary Changes in Japan’, Quarterly Journal of Economics, 12: 153–85. Duffy, J. and Papageorgiou, C. (2000) ‘A Cross-Country Empirical Investigation of the Aggregate Production Function Specification’, Journal of Economic Growth, 5: 87–120. Easterlin, R.A. (1981) ‘Why Isn’t the Whole World Developed?’ Journal of Economic History, 41: 1–19. Economic and Social Research Institute (ESRI) (2006) Gross Domestic Product Classified by Economic Activities. Annual Report on National Accounts 2006 on CD-ROM. Tokyo: Cabinet Office. Euromoney (September 1985) ‘JAL’s Growing Thirst for Funds’. Far Eastern Economic Review (16 January 1986) ‘An Overdue End in Sight to the Monopoly Game’. —— (4 August 1994) ‘Thank You for Not Smoking’. Fearey, R. (1980), ‘Comments’, in L.H. Redford (ed.) The Occupation of Japan: Economic Policy and Reform, Norfolk, VA: The MacArthur Memorial. Financial Services Agency, Status of Nonperforming Loans.
214 Bibliography Financial Times (4 April 1984) ‘Reformists Lose Out to Tobacco Lobby’. —— (24 August 1984) ‘Japanese Monopoly Prepares for Tobacco War’. —— (10 September 1993) ‘Brand New Direction Heralds Flotation of Japan Tobacco: The Company’s Plans’. —— (9 November 1999) ‘NTT Offer to Net Treasury ¥1,560 bn’. —— (12 June 2000) ‘Japanese Premier Points to Full Privatization of NTT’. —— (29 September 2000) ‘NTT Petitions Tokyo to Relax Rules on Ownership’. —— (23 January 2001) ‘Tokyo Toys with Unusual Plans’. Freedom House (2004), Freedom in the World 2004: The Annual Survey of Political Rights and Civil Liberties, Washington, DC. Fukao, K., Inui, T., Kawai, H., and Miyagawa, T. (2003) Sectoral Productivity and Economic Growth in Japan, 1970–98: An Empirical Analysis Based on the JIP Database, ESRI Discussion Paper Series 67, October 2003. Fukui, K. (1992) Japanese National Railways Privatization Study. Discussion Paper 172, Washington, DC: World Bank. Gao, B. (1994) ‘Arisawa Hiromi and His Theory for a Managed Economy’, Journal of Japanese Studies, 20: 115–53. Garon, S. (1990) ‘Toward a History of Twentieth Century Japan’, Monumenta Nipponica, 45: 339–52. Gollin, D. (2002) ‘Getting Income Shares Right’, Journal of Political Economy, 110: 458–74. Gwartney, J.and Lawson, R. (2004) Economic Freedom of the World: 2004 Annual Report, Vancouver: The Fraser Institute. Hadley, E.M. (1970) Antitrust in Japan, Princeton, NJ: Princeton University Press. Hanley, S. (1983) ‘A High Standard of Living in Nineteenth-Century Japan: Fact or Fancy?’ Journal of Economic History, 43: 183–92. Hannah, L. (2005) The Whig Fable of American Tobacco, 1895–1913, Tokyo: University of Tokyo, Center for International Research on the Japanese Economy, discussion paper 2005-CF-352. Harberger, A. (1998) ‘A Vision of the Growth Process’, American Economic Review, 88: 1–32. Harrod, R.F. (1951) The Life of John Maynard Keynes, New York: Harcourt, Brace and Co. Hein, L.E. (1990) Fueling Growth: The Energy Revolution and Economic Policy in Postwar Japan, Cambridge, MA: Harvard University Press. Heo, U. (1998) ‘Modeling the Defense–Growth Relationship around the Globe’, Journal of Conflict Resolution, 42: 637–57. Heston, A., Summers, R. and Aten, B. (2002) Penn World Table Version 6.1, Philadelphia, PA: Center for International Comparisons at the University of Pennsylvania. Higuchi, I. and Hashimoto, M. (2005) ‘Issues Facing the Japanese Labor Market’, in T. Ito, H. Patrick, and D.E. Weinstein (eds) Reviving Japan’s Economy, Cambridge, MA: MIT Press. Hollerman, L. (1979) ‘International Economic Controls in Occupied Japan’, Journal of Asian Studies, 38: 707–19. Ikle, F.W. (1967) ‘The Triple Intervention: Japan’s Lesson in the Diplomacy of Imperialism’, Monumenta Nipponica, 22: 122–30. Inklaar, R., Wu, H. and van Ark, B. (2003) Losing Ground: Japanese Labour Productivity and Unit Labour Cost in Manufacturing in Comparison to the US, Groningen,
Bibliography 215 Netherlands: Groningen Growth and Development Centre, Groningen University, Research Memorandum GD-64. International Monetary Fund (IMF) (1998) Switzerland: Selected Issues and Statistical Appendix. IMF Staff Country Report 98/43. Washington, DC: International Monetary Fund. Ishikawa, K. (1995) ‘Postwar Economic Reconstruction’, NIRA Review, Summer 1995, Tokyo: National Institute for Research Advancement. Ishikawa, T. and Imashiro, M. (1998) The Privatization of Japanese National Railways, London: Athlone Press. Jansen, M. (ed.) (1989) The Cambridge History of Japan, Vol. 5. The Nineteenth Century, Cambridge: Cambridge University Press. Jansen, M. (1989) ‘The Meiji Restoration’, in M. Jansen (ed.) The Cambridge History of Japan, Vol 5. The Nineteenth Century, Cambridge: Cambridge University Press. Japan Real Estate Institute, Urban Land Price Index, http://www.reinet.or.jp/e/index.htm (accessed 20 March 2007). Japan Statistical Association (JSA) (1999) Historical Statistics of Japan on CD-ROM, Tokyo. Japan Tobacco (2005) Annual Report. Johnson, C. (1982) MITI and the Japanese Miracle, Stanford, CA: Stanford University Press. Joint Chiefs of Staff (JCS) (1945) Basic Directive for Post-Surrender Military Government in Japan Proper, JCS 1380/15. Kaneda, H. (1980) ‘Structural Change and Policy Response in Japanese Agriculture after the Land Reform’, in L.H. Redford (ed.) The Occupation of Japan: Economic Policy and Reform, Norfolk, VA: The MacArthur Memorial. Karagiannis, G., Palivos, T. and Papageorgiou, C. (2004) Variable Elasticity of Substitution and Economic Growth: Theory and Evidence, Louisiana State University. Kaufmann, D., Kraay, A. and Mastruzzi, M. (2005) Governance Matters IV: Governance Indicators for 1996–2004, Washington, DC: World Bank. Kayaoglu, T. (2006) The Extension of Westphalian Sovereignty: State-Building and the Abolition of Extraterritoriality, Tacoma, WA: University of Washington. Keiji, N. and Yamamura, K. (1988) ‘Shaping the Process of Unification: Technological Progress in Sixteenth- and Seventeenth-Century Japan’, Journal of Japanese Studies, 14: 77–109. Kelley, A.C. and Williamson, J.G. (1971) ‘Writing History Backwards: Meiji Japan Revisited’, Journal of Economic History, 31: 729–76. Kim, J.-I. and Lau, L. (1994) ‘The Sources of Economic Growth of the East Asian Newly Industrialized Countries’, Journal of the Japanese and International Economies, 8: 235–71. Knack, S. and Keefer, P. (1995) ‘Institutions and Economic Performance: Cross-Country Tests Using Alternative Institutional Measures’, Economics and Politics, 7: 207–27. Kumar, S. and Russell, R.R. (2002) ‘Technological Change, Technological Catch-up, and Capital Deepening: Relative Contributions to Growth and Convergence’, American Economic Review, 92: 527–47. Kuwayama, P.H. (1999) Postal Banking in the United States and Japan, A Comparative Analysis. Discussion Paper, Tokyo: Bank of Japan, Institute for Monetary and Economic Studies. Kyodo (24 November 1987) ‘MOF to Sell 48.1 Million JAL Shares on December 15’. —— (29 November 1991) ‘Three JRs’ Concerns Shifting to Real Privatization’.
216 Bibliography —— (15 March 1993) ‘Government to OK Listing of JR East Shares in September’. —— (20 April 1994) ‘Price Limit on Planned JT Stock Listing Mulled’. —— (13 May 1996) ‘JR East Head Urges JR West to List All Stock at Once’. —— (21 March 1998) ‘MOF Scraps Planned Sales of NTT Shares in FY ‘97’. —— (7 December 2001) ‘MOF proposes cutting government’s JT stake to 50% from 67%’. Latzy, J. and Miller, R. (1992) ‘Controlled Foreign Corporations, 1988’, Statistics of Income Bulletin, 12: 119–22. Lincoln, E. (2001) Arthritic Japan: The Slow Pace of Economic Reform, Washington, DC: Brookings Institution Press. McCreedy, K.O. (2001) ‘Planning the Peace: Operation Eclipse and the Occupation of Germany’, Journal of Military History, 65: 713–39. Maddison, A. (1995a) ‘Explaining the Economic Performance of Nations, 1820–1989’, in A. Maddison (ed.) Explaining the Economic Performance of Nations: Essays in Time and Space, Brookfield, VT: Edward Elgar. —— (1995b) ‘Standardised Estimates of Fixed Capital Stock: A Six-Country Comparison’, in A. Maddison (ed.) Explaining the Economic Performance of Nations: Essays in Time and Space, Brookfield, VT: Edward Elgar. Mauro, P. (1995) ‘Corruption & Growth’, Quarterly Journal of Economics, 110: 681– 712. Mayo, M.J. (1980) ‘American Economic Planning for Occupied Japan: The Issue of Zaibatsu Dissolution, 1942–1945’, in L.H. Redford (ed.) The Occupation of Japan: Economic Policy and Reform, Norfolk, VA: The MacArthur Memorial. —— (1984) ‘American Wartime Planning for Occupied Japan’, in R. Wolfe (ed.) Americans as Proconsuls: United States Military Government in Germany and Japan, 1944–1952, Carbondale, IL: Southern Illinois University Press. Milhaupt, C.J. (2001) ‘Creative Norm Destruction: The Evolution of Nonlegal Rules in Japanese Corporate Governance’, University of Pennsylvania Law Review, 149: 2083–129. Ministry of Finance, Financial Statements Statistics of Corporations, Tokyo. Ministry of Health, Labour and Welfare (MHLW), Comprehensive Survey of Living Conditions of the People on Health and Welfare, Tokyo. Ministry of Internal Affairs and Communications, Statistics Bureau, Labor Force Survey, Tokyo. Miwa, Y. and Ramseyer, J.M. (2002) ‘Banks and Economic Growth: Implications from Japanese History’, Journal of Law and Economics, 45: 127–64. —— (2005) The Good Occupation. Discussion Paper 514. Harvard University, Cambridge, MA: John M. Olin Center for Law, Economics, and Business. Mochizuki, M. (1993) ‘Public Sector Labor and the Privatization Challenge: The Railway and Telecommunications Unions’, in G. Allinson and Y. Sone (eds) Political Dynamics in Contemporary Japan, Ithaca, NY: Cornell University Press. Moore, R.A. (1979) ‘Reflections on the Occupation of Japan’, Journal of Asian Studies, 38: 721–34. Moriguchi, C. and Saez, E. (2005): The Evolution of Income Concentration in Japan, 1885–2002: Evidence from Income Tax Statistics, Berkeley, CA: University of California. Mundell, R.A. (2000) ‘A Reconsideration of the Twentieth Century’, American Economic Review, 90: 327–40.
Bibliography 217 Nakamura, J. (1981) ‘Human Capital Accumulation in Premodern Rural Japan’, Journal of Economic History, 41: 263–81. Nakamura, T. (1988) ‘Depression, Recovery, and War, 1920–1945’, in P. Duus (ed.) The Cambridge History of Japan Vol. 6. The Twentieth Century, Cambridge: Cambridge University Press. Nanto, D. (1980) ‘The Dodge Line: A Reevaluation’, in L.H. Redford (ed.) The Occupation of Japan: Economic Policy and Reform, Norfolk, VA: The MacArthur Memorial. Nikkei (23 September 1980) ‘Sale of Government Holdings in Special Corporations Urged’. —— (16 December 1980) ‘Transport Ministry Agrees to Sell a Portion of JAL Shares’. —— (5 March 2001) ‘Government Considers Cut in Tobacco Company Holdings’. —— (18 February 2003) ‘Government To Cancel Sale Of NTT, JR West, JT Shares For FY02’. —— (3 September 2005) ‘Govt Got Y31 trillion In Privatization Proceeds Over 2 Decades’. Nomura Securities Co. Ltd. (2006) Trends in Japanese Companies Related M&As in 2005. OECD (2005) OECD Economic Surveys Japan, Paris: Organization for Economic Cooperation and Development. —— (2006) OECD Economic Surveys Japan, Paris: Organization for Economic Cooperation and Development. Ohkawa, K. and Rosovsky, H. (1973) Japanese Economic Growth: Trend Acceleration in the Twentieth Century, Stanford, CA: Stanford University Press. Ohkawa, K., Shinohara, M., and Meissner, L. (1979) Patterns of Japanese Economic Development: A Quantitative Appraisal, New Haven, CT: Yale University Press. Okazaki, T. (1999) ‘Corporate Governance’, in T. Okazaki and M. Okuno-Fujiwara (eds), The Japanese Economic System and Its Historical Origins, Oxford: Oxford University Press. Okazaki, T. and Okuno-Fujiwara, M. (1999) ‘Japan’s Present-Day Economic System and Its Historical Origins’, in T. Okazaki and M. Okuno-Fujiwara (eds) The Japanese Economic System and Its Historical Origins, Oxford: Oxford University Press. Ono, H. and Rebick, M. (2003) Constraints on the Level and Efficient Use of Labor in Japan. Working Paper 9484. Cambridge, MA: National Bureau of Economic Research, February 2003. Passin, H. (1965) Society and Education in Japan, New York: Teachers College, Columbia University. Passin, H. (1990) ‘The Occupation: Some Reflections’, Daedalus, 119: 107–29. Patrick, H.T. (1965) ‘External Equilibrium and Internal Convertibility: Financial Policy in Meiji Japan’, Journal of Economic History, 25: 187–213. Posen, A. (2000) ‘Introduction: Financial Similarities and Monetary Differences’, in A. Posen and R. Mikitani (eds), Japan’s Financial Crisis and its Parallels to U.S. Experience. Special Report 13. Washington, DC: Institute for International Economics. Redford, L.H. (ed.) (1980) The Occupation of Japan: Economic Policy and Reform, Norfolk, VA: The MacArthur Memorial. Rodrik, D. (1997) TFPG Controversies: Institutions and Economic Performance in East Asia. Working Paper 5914. Cambridge, MA: National Bureau of Economic Research. February 1997. —— (2004) Getting Institutions Right, Cambridge, MA: Harvard University.
218 Bibliography Rotwein, E. (1964) ‘Economic Concentration and Monopoly in Japan’, Journal of Political Economy, 72: 262–77. —— (1976) ‘Economic Concentration in Japan – A Second View’, Journal of Asian Studies, 36: 57–77. Rubinger, R. (1990) ‘From “Dark Corners” into “The Light”: Literacy Studies in Modern Japan’, History of Education Quarterly, 30: 601–12. Sachs, J. and Warner, A. (1995) Economic Convergence and Economic Policies, Warsaw: Center for Social and Economic Research. Sackton, F.J. (1980) ‘The Transfer of Land Ownership to the Peasants: The Priceless Economic Reform’, in L.H. Redford (ed.) The Occupation of Japan: Economic Policy and Reform, Norfolk, VA: The MacArthur Memorial. Samuels, R. (1987) The Business of the Japanese State, Energy Markets in Comparative and Historical Perspective, Ithaca, NY: Cornell University Press. —— (1994) ‘Rich Nation, Strong Army’: National Security and the Technological Transformation of Japan, Ithaca, NY: Cornell University Press. Sandler, T. and Hartley, K. (1995) The Economics of Defense, Cambridge: Cambridge University Press. Saxonhouse, G. (1974) ‘A Tale of Japanese Technology Diffusion in the Meiji Period’, Journal of Economic History, 34: 149–65. Schaede, U. (2006) Competition for Corporate Control: Institutional Investors, Investment Funds, and Hostile Takeovers in Japan, New York: Columbia University, Center on Japanese Economy and Business, Working Paper 248. Schlesinger, A.M. Jr (1959) The Coming of the New Deal, Boston, MA: Houghton Mifflin. Schonberger, H.B. (1989) Aftermath of War: Americans and the Remaking of Japan, 1945–1952, Kent, OH: Kent State University Press. Shimazaki, T. (1987) Before the Dawn (Yo-ake Mae), trans. William Naff, Honolulu, HI: University of Hawaii Press. State-War-Navy Coordinating Committee (SWNCC) (1945) Politico-Military Problems in the Far East: United States Initial Post-Defeat Policy Relating to Japan, SWNCC 150/4. Summers, R., Heston, A., Nuxoll, D.A. and Aten, B. (1994) Penn World Table, version 5.6, Cambridge, MA: National Bureau of Economic Research. Takano, Y. (1992) Nippon Telegraph and Telephone Privatization Study, Washington, DC: World Bank Discussion Paper 179. Takenaka, Y. (1969) ‘Endogenous Formation and Development of Capitalism in Japan’, Journal of Economic History, 29: 141–62. Tashiro, K. (1982) ‘Foreign Relations during the Edo Period: Sakoku Reexamined’, Journal of Japanese Studies, 8: 283–306. Toby, R.P. (1977) ‘Reopening the Question of Sakoku: Diplomacy in the Legitimation of the Tokugawa Bakufu’, Journal of Japanese Studies, 3: 323–63. Tokyo Stock Exchange (TSE) (2006) 2005 Share Ownership Survey, Tokyo. Transparency International (2004) Corruption Perceptions Index. Ueda, K. (1999) ‘The Financial System and Its Regulations’, in T. Okazaki and M. OkunoFujiwara (eds) The Japanese Economic System and Its Historical Origins, Oxford: Oxford University Press. US Department of Agriculture (2004) Japan Tobacco and Products Annual 2004, Washington, DC.
Bibliography 219 Vaporis, C. (1997) ‘To Edo and Back: Alternate Attendance and Japanese Culture in the Early Modern Period’, Journal of Japanese Studies, 23: 25–67. Vogel, S. (1996) Freer Markets, More Rules: Regulatory Reform in Advanced Industrial Countries, Ithaca, NY: Cornell University Press. Wall Street Journal (25 August 1992) ‘Japan Government Won’t Sell NTT Shares Before March 1994’. —— (28 October 1993) ‘JR East Issue Derails in Tokyo Trading’. —— (2 October 1998) ‘Japan Picks Daiwa, Goldman, LTCB Warburg to Lead NTT Share Sale’. —— (12 October 1998) ‘Pricing of NTT DoCoMo’s IPO at $33,425 a Share Viewed as Cheap’. Ward, M.D., Davis, D.R. and Lofdahl, C.L. (1995) ‘A Century of Tradeoffs: Defense and Growth in Japan and the United States’, International Studies Quarterly, 39: 27–50. Watanabe, S. (1994) ‘Restructuring of the Japanese National Railways: Implications for Labour’, International Labour Review, 133: 89–111. Weinstein, D. and Yafeh, Y. (1998) ‘On the Costs of a Bank-Centered Financial System: Evidence from the Changing Main Bank Relations in Japan’, Journal of Finance, 53: 635–72. West, M.D. (2000) ‘The Puzzling Divergence of Corporate Law: Evidence and Explanations from Japan and the United States’, Michigan Law and Economics Research Paper 00–011. —— (2001) ‘Why Shareholders Sue: The Evidence from Japan’, Journal of Legal Studies, 30: 351–82. World Bank (2006) World Development Indicators – 2006, Washington, DC: World Bank. Yamamura, K. (1967) Economic Policy in Postwar Japan: Growth versus Economic Democracy, Berkeley, CA: University of California Press. —— (1973) ‘Toward a Reexamination of the Economic History of Tokugawa Japan, 1600–1867’, Journal of Economic History, 33: 509–46. —— (1977) ‘Success Illgotten? The Role of Meiji Militarism in Japan’s Technological Progress’, Journal of Economic History, 37: 113–35. —— (1986) ‘Japan’s Deus ex Machina: Western Technology in the 1920s’, Journal of Japanese Studies, 12: 65–94. —— (1988) ‘From Coins to Rice: Hypotheses on the Kandaka and Kokudaka Systems’, Journal of Japanese Studies, 14: 341–67.
Index
Abel, A. et al. 114 Acheson, Secretary Dean 86 acquisitions and mergers 103–4 administration of war production 58–62 ‘Administrative Reform without Tax Increase’ 141 Aeon 200 aggregate returns on capital 116 agriculture: agricultural economy 15; development after Meiji Restoration 33; tax yields in Tokugawa period 19 Akino, M. and Hayami, Y. 33 Albertsons 201 Alexander, A. 172 All Nippon Airways (ANA) 164, 166, 167 alternate attendance (sankin kotai), Tokugawa system of 17 American Occupation and post-war economy 2–3, 68–91; aid burden on US 86–7; anti-trust legislation 70–1; austerity and Dodge Line 87–8; business deconcentration 81–2; Cabinet Planning Board 74–5; coal crisis 77–81; collusion 82; corporate reorganization 81–2; demilitarization 72–3; Dodge Line 85–90; economic planning 70–2; ESB (Economic Stabilization Board) 75–7, 78, 80; fiscal shift 88; foreign exchange 87, 88–9; government role, reduction in 89; HCLC (Holding Company Liquidization Commission) 81; implications of 90–1; industrial materials, supplies of 86; inflation, supply-side approach to 86; inflationary waves 85–6; initial approach to economic policy 72–4; Japan Fair
Trade Commission 82; JCS (Joint Chiefs of Staff) 72–3; labour unions 83; land reform 83–5; market prices, introduction of 89–90; materials shortages 86; MITI (Ministry of International Trade and Industry) 62, 89, 91; MOF (Ministry of Finance) 91; New Deal orientation 73, 77, 90; oligopolies 82; planning for Occupation 69–72; planning mechanism 70; Potsdam Declaration (1945) 71–2; price controls 80; price stabilization 87; priority production system 80; professional competence 74; punitive policy 73; residue of war 69; ‘reverse course’ 87; RFB (Reconstruction Finance Bank) 75–6, 78, 86–9, 95, 96; rice harvest 75; SCAP (Supreme Commander, Allied Powers) 72–4, 75–7, 77–80, 81, 84, 87–8, 90, 95, 98; state controls 80–1; state economic planning 74–7; SWNCC (State–War– Navy Coordinating Committee) 72–3; wartime industrial coordination 75; Yalta Agreement (1945) 72; zaibatsu, breaking up of 86 American Stores 201 anti-trust legislation 70–1 Argentina 11, 174, 179, 181 Arisawa, Hiromi 76, 77–8 Ashida, Prime Minister Hitoshi 79 Asian Wall Street Journal 149, 161–2 asset price collapse 9 atomic bombs 67, 69 austerity, Dodge Line and 87–9 Australia 2, 60, 165, 172, 174, 176, 178, 180
222 Index Australian Financial Review 160 Austria 174, 178, 180, 181 Baerwald, H. 73 balance sheets 195–6 ballistic arc of economy 1–3 bank-centred finance 92–7, 101, 104; bonds 94, 96; business financing, stages of shift in 93–4; government intervention in private funding 94; insolvency 95; keiretsu 96, 99; longtern funds 94; MOF (Ministry of Finance) 95, 99, 100; private banks 94; reduction in number of banks 94–5; SCAP and 95; securities companies 94; shareholding regulations 96–7; stock market 96–7 Bank of Japan (BOJ) 112; bank rationalization, encouragement for 94; creation and money supply control 30–1; creation of 30; foreign returns 112; gold standard, shift to 31–2; investment screening 61; monetary system stabilization 32; money supply control 31; money supply problems 9; refinancing, liberal loans for 96; RFB and 86; stimulatory fiscal measures, reluctance to use 191; stocks and bonds (1935) 93; Tokyo–Yokohama earthquake (1923), financial consequences 49; ‘window guidance’ 96 banking crisis 191–4 bankruptcy of financial institutions 193 Beasley, W.G. 25, 41, 43, 44, 46, 47, 54 Beck, T. et al. 175, 209, 210 Belgium 47, 174, 178, 180, 181 Beplat, T.E. 95 Bernanke, B.S. and Gurkaynak, R. 118 Bisson, T.A. 81 Boeing Corporation 54, 165 bonds 94, 96; flotations, controls over 94; interest-bearing bonds 30 Brazil 174, 179, 181 Britain see United Kingdom Bronfenbrenner, M. 77 Burma (Myanmar) 47 business deconcentration 81–2 business financing, stages of shift in 93–4 Business Week 167 cabinet governance in Meiji period 26 Cabinet Planning Board: American Occupation 74–5; post-war economy 60, 63, 65
Calder, K.E. 74, 91, 95, 138 Canada 174, 178, 180 canonical post-war system 92, 99, 100 capital accumulation 105–6; in Meiji period 35 capital investment and return rates 105–21; aggregate returns 116; analysing rates of return 106–8; capital accumulation 105–6; capital stock 119–21; capital to labour, ratio of 117–18; Cobb–Douglas formulation, production function 116–19; company performance, factors governing 116; corporate behaviour, influences on 116; corporate returns 110–12; dynamic efficiency 114–15; economy-wide returns 108–10; fall of returns, large and steady 109–10; FDI (foreign direct investment) 112–14; foreign returns 112–14; GDP per capita 105; growth, returns and costs of 116; MOF (Ministry of Finance) 110, 113; operating profits 110–11; output, elasticity with respect to capital 116–19; payoff from investment 107; private sector capital stock, ratio to GDP 106, 107; production function, Cobb–Douglas formulation 116–19; productivity, profit and 107–8; profits 110–11; returns on capital 119–21; returns on capital, US vs Japan 108; scale returns 117; substitution between capital and labour, elasticity of 117–18; total assets 111 capital productivity 11–12 capital stock 195; investment and return rates 119–21 capital to labour, ratio of 117–18 Caprio, G. and Klingebiel, D. 191 cartels, organization of 52 central administration (bakufu) in Tokugawa period 16, 17, 18, 26, 27 centralization of authority in Meiji Tokyo 27 Chiang Kai-shek 46 Chile 174, 179, 180, 181 China 1, 3, 14, 31, 36, 69, 143; ‘China Incident’ 46, 60; communist takeover in 86; economic power of 8; Kuomintang 46; Manchu dynasty in 44; War in (1937) 57, 58, 61 Choi, E. 141, 142, 143, 146–7 Churchill, Prime Minister Winston S. 71 Civil Aeronautics Law (1952) 164 civil bureaucracies, reform cliques in 59 Claessens, S. et al. 194
Index 223 Clark, G. 39 closed country (sakoku) 17 coal crisis 77–81; parliamentary activity during 78, 79 Cobb, Charles 117 Cobb–Douglas formulation, production function 116–19 Cohen, Jerome B. 69, 73, 77, 80, 86, 88 Collins, S.M. and Bosworth, B.P. 118, 121 collusion during American Occupation 82 Colombia 174, 179, 181 colonial domination, fear of 48 commercial codes: corporate governance 97–8; revision of 103 company behaviour, influences on 116 company directors, penalties for 102–3 company law, post-war arrangements 98 company performance, factors governing 116 comparisons: with advanced nations 12–14; with United States 134–6 constitutional provision for Diet (Parliament) 40–1 controls: currency controls 71; dismantling of economic controls 89, 90; economic behaviour, effects on 64; enthusiasm of SCAP for 77; exchange controls 61; financial system controls 61; imposition of economic controls 58, 62; interest rate controls 100; money supply control, Bank of Japan (BOJ) and 30–1; over bond flotations 94; planning and 65; price controls 61, 62, 63, 66, 68, 76, 79, 80, 89, 139; wage controls 85, 86, 90, 205 Coox, A.D. 62 corporate governance 97–9; commercial codes 97–8; corporate system, re-examination of 63–4; crossshareholding 99; Diet (Parliament) 98, 103; foreign corporate law, transplant of 98; Illinois Business Corporation Act 98; post-war company law 98; reforms of 102; shareholder provisions 98–9 corporate health, gain of 194–8 corporate reorganization during American Occupation 81–2 corporate returns on capital investment 110–12 Costa Rica 174, 178, 179, 181 cotton-spinning mills 33–4 Crawcour, S. 15, 18, 24, 25 cross-shareholding 99
currency 31, 32, 35, 51, 71, 88; controls over 71; depreciation in Meiji period 30; foreign currency 36, 60, 89; instability of 50–2; protection of 52–3; purchasing power parity (PPP) 4–5; reform in Meiji period 27 Daiei 200 daimyo (great names), power of 16–17, 21, 26 Daiwa bank case 102–3 deceleration of post-war growth 7 deflation (1920–32) 50 deflation (Meiji Restoration) 31 Deininger, K. and Squire, L. 210 demilitarization 72–3 Denmark 174, 178, 180 Diet (Parliament) 1, 2, 49, 58, 70, 76; austerity, Dodge Line and 87–9; business deconcentration 81; coal crisis 78, 79; constitutional provision for 40–1; corporate governance 98, 103; land reform 84; parliamentary government, popular support for 41–2; party leadership 40–1; privatization 138–9, 140, 147, 149–50, 154–5, 157–9, 160, 162–3, 164, 166–7, 169; war production planning 59, 60 differences: differentiation 199–200; growth differences across industries 201–2; implications of growth with 205–6; see also Japanese economic difference diminishing returns 10, 105–6, 108, 109–10, 113 disposable income 204–5 DoCoMo 157, 158. 159 Dodge, Joseph 87, 88, 89–90, 140 Dodge Line 85, 87, 89, 90, 139 Doko, Toshio 141, 153 Dominican Republic 174, 179, 181 Dore, R. 22, 84, 85 Douglas, Paul 117 Dower, John 67, 69, 74, 83 Draper, William 86, 87 Droppers, G. 30, 32 Duffy, J. and Papageorgiou, C. 118 dynamic efficiency 114–15 Easterlin, Richard 21–2, 24 Economic and Social Research Institute (ESRI) 123 economic diversification in Meiji period 32–6
224 Index Economic Freedom of the World (Fraser Institute) 174 ‘economic miracle’ 5, 14 economic planning (American Occupation) 70–2 economy: acquisitions and mergers 103–4; administration of war production 58–62; agricultural economy 15; asset price collapse 9; atomic bombs 67, 69; balance sheets 195–6; ballistic arc 1–3; banking crisis 191–4; bankruptcy of financial institutions 193; business deconcentration 81; Cabinet Planning Board 60, 63, 65; canonical post-war system 92, 99, 100; capital productivity 11–12; capital stock 195; cartels, organization of 52; civil bureaucracies, reform cliques in 59; collapse after First World War 48, 50–1; colonial domination, fear of 48; commercial code revision 103; company directors, penalties for 102–3; comparisons with advanced nations 12–14; controls, imposition of 58; corporate governance reforms 102; corporate health, gain of 194–8; corporate system, re-examination of 63–4; currencies, instability of 50–2; Daiwa bank case 102–3; deceleration of post-war growth 7; deficits, large annual 9; deflation (1920–32) 50; differences, implications of growth with 205–6; differentiation 199–200; diminishing returns 10; disposable income 204–5; ‘economic miracle’ 5, 14; economic weakness (1920s) 49; end of 1930–45 system 3, 198–9; financial market innovations 101; financial panic (1927) 49; firebombing, destruction by 66–7; food shortages 66; foreign exchange control 101; global depression (1930s) 1, 6, 45, 48, 52, 75, 109; gold standard and restoration of gold parity 50–2; government debt 9; Greater East Asia Co-Prosperity Sphere 65–6; growth differences across industries 201–2; growth during lost (1990s) decade 11; growth long-term 189–91; growth prospects 188–98; growth since nineteenth century 5–7; heterogeneous 136; implications of growth with differences 205–6; Important Industries Control Law 52;
industrial associations 63; industrial development and military 54–7; industry laws for military production 59–60; inequalities 203–5; inflation 60; initial conditions, importance of 15; institution building 48; institutional barriers against military suppliers 62; investment 9; labour, intensity of 39; labour flexibility 196–8; labour force build-up 196; labour productivity 196; labour unrest 52; land price variability 202; loan performance 192–4; longterm prospects 188–98; macroeconomic balance, strategy for 60–1; Manchurian incident (1931) 52; market-based finance, rise of 101–2; maturation 10; MCI (Ministry of Commerce and Industry) 59, 61; mergers 103–4; military bureaucracies, reform cliques in 59; MITI (Ministry of International Trade and Industry) 62, 89, 91; mobilization planning 58–9; MOF (Ministry of Finance) 103; monetary policy 9; monitoring 102; munitions, shortages of 62; non-performing loans 192–4; norms, changes in 103; norms, role of 99–100; output per capita, growth of 8–9; oversight 102; personal income inequality, rise in 203–5; planning supplies for war 58–67; political weaknesses 9; postwar system, change and end of era 100–4; price controls 61; production incentives 62–3; productivity 10, 39, 191; profitability of private financial sector 9; property prices 192; prospects, moving to new trajectory 188–206; raw materials, shortages of 58; regulatory excess 9; retailing, change in 200–1; revenue-producing financial products 101; scale 3; severity and complexity of problems at millennium 9; shareholders 195; shareholding threshold reduction 103; shipping 66; slowdown and stagnation 9–12; staffing 196; stock market crash (1920) 48–9; sub-contractors, designation of 63; supervision, deficiencies in 9; system restructure for war production 62–5; Tokyo–Yokohama earthquake (1923), effects of 49; trajectory shift from arc of past 75 years 188–9; transition from peasant society 15;
Index 225 wage controls 61; war economy, supply arrangements 65–7; wars with China and Russia, effects of 48; wartime economic destruction 67, 69; winners and losers, emergence of 199–205; zaibatsu, growth of 52–4; see also American Occupation and post-war economy; bank-centred finance; capital investment and return rates; corporate governance; Japanese economic difference; Meiji Restoration; privatization; structural change; Tokugawa period economy-wide returns 108–10 education: comprehensive national system 28–9; precursor of growth 21–2 Egekvist, W.S. 75 electorate 40 Electric Power Development Corp 166 emperor, powers of 40 ESB (Economic Stabilization Board) 75–7, 78, 80 Euromoney 167 exchange controls 61 export markets 45–6 Far Eastern Economic Review 161, 167 FDI (foreign direct investment) 112–14 Fearey, Robert 70, 83–4 Federation of Economic Organizations (Keidanren) 140–1 feudalism 16 financial market innovations 101 financial panic (1927) 49 financial system: controls on 61; creation of modernity in 29–32 Financial Times 158, 159, 160, 161 Finland 174, 178, 180 firebombing, destruction by 66–7 First World War 1–2, 36, 58, 75; military production during 53, 54–5 Fiscal Investment and Loan Program (FILP) 140 fiscal shift (American Occupation) 88 fiscal system reform 27–8 food shortages 66 foreign capital inflows (Meiji period) 35–6 foreign corporate law, transplant of 98 foreign exchange 87, 88–9, 101 foreign returns 112–14 France 43, 47, 65, 138, 174, 178, 180 Fraser Institute 174 Freedom House 210
Fuji Bank 53 Fuji Corporation 81 Fuji Heavy Industries 53, 54 Fukao, K. et al. 129, 133 Fukui, K. 142, 143, 144, 145, 147 Gao, B. 76 Garon, S. 59 GDP (gross domestic product): change in 201; growth across industries 122–9; private sector capital stock, ratio to 106, 107 GDP per capita: capital investment and return rates 105; comparison with US 5, 8, 13; global ranking 3, 5; limitations as measure of economic well-being 7; real growth in 7, 10–11 geographical expansion 42–7 Germany 2, 43, 69, 86, 138, 174, 178, 180, 181; invasion in Europe by (1940) 47 Gini coefficient of inequality 123, 127, 130, 131, 134–5, 136, 203, 204–5, 210 global depression (1930s) 1, 6, 45, 48, 52, 75, 109 gold standard: and restoration of gold parity 50–2; shift to 31–2 Goldman Sachs International 157, 162 Gollin, D. 118 governance: government formation 40–1; intervention in private funding 94; ownership of industry, concerns about 138; precursor of growth 22–3; radical improvement in (Meiji period) 26; role of, reduction in 89; share sales of privatized corporations 170–1 government debt 9 Greater East Asia Co-Prosperity Sphere 65–6 Greece 1 81, 174, 179 Grew, Ambassador Joseph 70 growth: differences across industries 201–2; long-term 189–91; during lost (1990s) decade 11; prospects for 188–98; returns and costs of 116; since nineteenth century 5–7; subsequent to Meiji Restoration 25–39 Gwartney, J. and Lawson, R. 175, 208, 209 Hadley, E.M. 82 Hamaguchi, Osachi 160 Hanley, Susan 23 Hannah, L. 160 Harberger, Arnold 122, 128, 133, 134
226 Index Harrod, R.F. 51 Hashimoto, Prime Minister Ryutaro 101, 155–6 HCLC (Holding Company Liquidization Commission) 81 Hein, L.E. 74, 77, 78, 79 Heo, U, 38 Heston, A. et al. 3 heterogeneous economy 136 Hideyoshi, Toyotomi 16 Higuchi, I. and Hashimoto, M. 190 Hitler, Adolf 61 Holland see Netherlands Hollerman, L. 76, 77 Home Depot 201 Hong Kong 174, 178, 179, 180 Ikle, F.W. 43 Illinois Business Corporation Act 98 import trade 17–18 Important Industries Control Law 52 Indochina 47, 65, 66 industrial associations 63 industrial development and military economy 54–7 industrial materials, supplies of 86 industrial progress, Meiji limits on 39 industrialization 32–3 industry laws for military production 59–60 inequalities 203–5 inflation 60; inflationary waves during American Occupation 85–6; Meiji Restoration inflationary pressures following 30; supply-side approach to 86 infrastructure development 28 Inklaar, R. et al. 190 insolvency 95 institution building 48 institutional barriers against military suppliers 62 institutional quality of government 22–3 interest-bearing bonds 30 international aviation agreements 165 international developments (1870–1940) 42–7; Burma (Myanmar) 47; ‘China Incident’ 46, 60; Chinese Kuomintang 46; export markets 45–6; geographical expansion 42–7; Indochina 47; international economics and 45; Korea, economic motives for more into 46; Korea, strategic importance of 43;
Liaotung Peninsula 43, 45; Malaysia 47; Netherlands East Indies 47; Pearl Harbor 47; Philippines 47; Port Arthur (Dalian) 43, 44; Russia and TransSiberian Railway 43–4; Second World War 46–7, 58; Shantung Peninsula 43, 45; Taiwan 43; Thailand 47; Trans-Siberian Railway 43–4; Triple Intervention 43; see also Manchuria Inukai, Prime Minister Tsuyoshi 41 investment 9; in manufacturing 34–5; screening by Bank of Japan 61 Ireland 11, 174, 178, 180 Ishibashi, Minister Tanzan 86 Ishikawa, K. 75 Ishikawa, T. and Imashiro, M. 143–4 Ishikawajima Shipyard 36 Israel 174, 179, 181 Italy 174, 178, 179, 181 Ito-Yokado 200 Jansen, M. 26, 27 Japan Airlines (JAL): competition, domestic and international 165–6; deregulation and privatization 164–8; post-privatization and deregulation 167–8; post-war law and regulation 164; privatization 138, 153, 164–8; selling JAL shares 166–7 Japan Economic Institute 199 Japan Fair Trade Commission 82 Japan Industrial Corp 53 Japan Monopoly Corporation 138 Japan National Railways (JNR) 138, 139, 142, 143–52, 168; break-up and privatization 146–8; early developments 143–5; funding rail deficits 145; labour, dealing with 150–1; post-reform labour reductions and efficiency gains 151–2; Railways Nationalization Law (1906) 143; repatriated workers, absorption of 144; selling shares 148–50; transportation modes, changes in 144–5 Japan Real Estate Institute 202 Japan Securities Dealers Association 161 Japan Statistical Association (JSA) 28, 29, 35, 49, 50, 55–6, 85, 90, 93, 95 Japan Tobacco (JT): competition and profitability 163–4; post-privatization performance 163–4; privatization 138, 150, 153, 159–64, 168; production 163; selling the government’s shares 161–3; tobacco and salt monopoly 138, 159–61
Index 227 Japanese economic difference: additions and amendments 176–7; analysis 175–7; assertion of 172; comparison countries 174; conclusions 185–7; correlation coefficients with US 180–1; data 174–5; Euclidian distances from US 178–9; growth, implications for 172–87; income, real GDP per capita 177, 183–5; main results of research 173–4; measurement of 172–87; new research 173; position, distance and correlation of Japan relative to US 182; previous research and critique of 172–3; similarity metrics 175–6; similarity to US 177–85 JCS (Joint Chiefs of Staff): American Occupation and post-war economy 72–3 Johnson, C. 59, 60, 61, 62, 65, 76, 78, 91, 99 Johnston, Percy 87 K-Mart 200 Kades, Charles 73 Kaneda, H. 85 Kanemaru, Shin 155 Karagiannis, G. et al. 118 Kaufmann, D. et al. 175, 210 Kayaoglu, T. 48 Keiji, N. and Yakamura, K. 23 keiretsu 96, 99 Kelley, A.C. and Richardson, J.G. 33 Kennan, George 87 Keynes, John Maynard 51 Kim, J.-I. and Lau, L. 110, 119, 121 Kishi, Prime Minister Nobusuke 80 Knack, S. and Keefer, P. 210 Koizumi, Prime Minister Junichiro 137, 140, 143, 159, 193, 198 Kokusai Denshin Denwa (KDD) 152, 157, 158 Komei, Emperor of Japan 26–7 Konoe, Prime Minister Fumimaro 41, 47 Korea 1, 69, 72, 174, 179, 180, 181; economic motives for more into 46; strategic importance of 43 Kumar, S. and Russell, R.R. 191 Kurile Islands 72 Kuwayama, P.H. 30 Kyodo 148, 149, 157, 161, 163, 167 labour: flexibility of 196–8; force build-up 196; intensity of 39; productivity of 196; staffing 196; unrest of 52
labour unions: during American Occupation 83; privatization and 139, 154–5 Ladejinsky, Wolf 70, 84 land price variability 202 land reform 83–5 Latzy, J. and Miller, R. 113 LDP (Liberal Democratic Party) 9, 193, 198; privatizations and 137, 140, 142, 145–6, 149, 151, 154–5, 158–60, 163, 165 legal structures, refashioning of 30 Leontieff, Wassily 76 Liaotung Peninsula 43, 45 liberalization ideas 153 life expectancy 23–4 Lincoln, E. 92, 99 Living Conditions, Comprehensive Survey of 203 loan performance 192–4 long-term: funds 94; prospects for economy 188–98 Lorenz curves 122, 128, 134, 203 MacArthur, General Douglas 72, 73, 79, 83, 84, 85, 87 McCreedy, K.O. 71 McDiarmid, R. 88 McKinsey Global Institute 200 macroeconomic balance, strategy for 60–1 Maddison, Angus 22, 119–21 Malaysia 47, 174, 178, 179, 180, 181 Manchu dynasty in China 44 Manchuria 1, 2, 6, 41, 53–4, 58, 61–2, 65, 66, 77, 140; Japanese expansion into 42–7; Manchurian Incident (1931) 52; South Manchurian Railroad 44, 45, 54, 144 manufacturing development in Meiji period 32–6 market-based finance, rise of 101–2 market prices, introduction of 89–90 Marshall, Secretary George 87 Martin, Edwin 70–1 Marubeni 200 Mason, Edward 70 materials shortages 86 Matsukata, Masayoshi 30, 31, 37 maturation of economy 10 Mauro, P. 210 Mayo, M.J. 69–70, 71, 72 MCI (Ministry of Commerce and Industry) 59, 61
228 Index Meiji Restoration 1, 2, 13, 22, 40, 43, 48, 52, 139, 189; agricultural development 33; cabinet governance 26; capital formation 35; centralization of authority in Tokyo 27; changes following Restoration 25; cottonspinning mills 33–4; currency depreciation 30; currency reform 27; deflation 31; economic diversification 32–6; education, comprehensive national system 28–9; expenditure reductions 27; financial system, creation of modernity in 29–32; fiscal system reform 27–8; foreign capital inflows 35–6; gold standard 31–2; governance, radical improvements in 26; growth subsequent to 25–39; industrial progress, limits on 39; industrialization 32–3; inflationary pressures 30; infrastructure development 28; interest-bearing bonds, increase in numbers 30; investment in manufacturing 34–5; legal structures, refashioning of 30; manufacturing development 32–6; military industry 36–9; monetary policy 29–32; money supply contraction 30; postal savings system 29–30; productivity 33; Shogun’s powers, surrender of 26; stagnation 30; state institutions, refashioning of 30; strong-army policy 36; tax reform, and opposition to 27–8; taxation 31; textiles 33–4; trade deficits 35–6; transformation, implementation of 26–8; see also Tokugawa period mergers and acquisitions 103–4 Mexico 174, 179, 181, 203 Milhaupt, C.J. 98, 99, 100, 103, 104 military bureaucracies, reform cliques in 59 military industrial demands 139 military industry 36–9 military leaders and party politicians 41 MITI (Ministry of International Trade and Industry) 62, 89, 91 Mitsubishi Heavy Electrical Machinery Co. 54–5, 81 Mitsui 53, 81 Miwa, Y. and Ramseyer, J.M. 53, 78, 80, 93, 94 mobilization planning 58–9 Mochizuki, M. 150–1, 154–5 modified Gini coefficients 127, 131, 134
MOF (Ministry of Finance) 103, 194; during American Occupation 91; bankcentred finance 95, 99, 100; capital investment and return rates 110, 113; privatization 139, 140, 142, 145, 149, 153, 155–6, 157, 158–9, 161–3 monetary policy 9, 29–32; system stabilization 32 money supply: contraction of 30; control of 31; problems of 9 Moore, R.A. 73 Morgan Stanley Dean Witter 149 Morgenthau, Secretary Henry 71 Mori, Prime Minister Yoshiro 158 Moriguchi, C. and Saez, E. 204–5 MOT (Ministry of Transport) 146, 147, 148, 149, 164, 166–7 MPT (Ministry of Posts and Telecommunications) 152, 155 Mundell, Robert 50–1 munitions, shortages of 62 Mycal 200 Nakajima Aircraft 53, 54 Nakamura, J. 20–1 Nakamura, T. 49, 51, 53, 63, 64, 65, 66 Nakasone, Prime Minister Yasuhiro 137, 141, 142, 146, 147, 155, 166, 167, 168, 169 Nanto, Dick 89 nationalism 41 nationalization 139 Netherlands 46, 47, 65, 174, 178, 180 Netherlands East Indies 47 New Deal orientation 73, 77, 90 New York Times 199 New Zealand 174, 178, 180 Nihon Keizai Shimbun 80 Nikkei 150, 163, 166, 168 Nippon Cargo Airlines Co. Ltd. (NCA) 165 Nippon Telegraph and Telephone (NTT): events leading to privatization 152–5; KDD (Kokusai Denshin Denwa) 152, 158; liberalization ideas 153; Ministry of Posts and Telecommunications (MPT) 152, 155; privatization 138, 139, 142, 150, 152–9, 168; selling and restructure 155–9; technological developments 153 Nissan 54, 61 Nobunaga, Oda 16 Nomura Securities 149, 162 non-performing loans 192–4
Index 229 Norman, E.H. 73 norms: changes in 103; role of 99–100 Norway 11, 174, 178, 180, 181 Obuchi, Prime Minister Keizo 193 Ohkawa, K. and Rosovsky, H. 32–3, 35 Ohkawa, K. et al. 34, 50, 219n2 Okada, Prime Minister Keisuke 41 Okazaki, T. 64, 96 Okazaki, T. and Okuno-Fujiwara, M. 61, 63 Okita, Minister Saburo 74 oligopolies 82 Ono, H. and Rebick, M. 189 operating profits 110–11 Organization for Economic Cooperation and Development (OECD) 5, 196–7, 198, 203–4 Osaka Spinning Company 34 Ostrom, Douglas 199–200 output: elasticity with respect to capital 116–19; per capita, growth of 8–9 overmanning 142 overseas voyages, prohibition of 17–18 Paris Exhibition (1867) 26 parliamentary government, popular support for 41–2 Passin, H. 22, 68 Patrick, Hugh 27, 30, 31, 32 Pauley, Edwin 73 payoff from investment 107 PCAR (Provisional Commission on Administrative Reform) 141, 142, 146, 150, 153, 166, 169 Pearl Harbor 47 Penn World Table (PWT) 3, 5, 8, 12, 107, 108, 109, 119–21 Perry, Commodore Matthew Galbraith 12–13, 24 personal income inequality, rise in 203–5 Pescadores Islands 72 Philippines 47 planning 69–72; mechanism for 70; supplies for war 58–67 Platt Brothers 34 political developments (1870–1940) 40–2; electorate 40; emperor, powers of 40; government formation 40–1; military leaders and party politicians 41; nationalism 41; parliamentary government, popular support for 41–2; party leadership 40–1; Taisho era 41 political weaknesses 9
Port Arthur (Dalian) 43, 44 Portugal 174, 179, 181 Posen, A. 191–2 post-war system, change and end of era 100–4 postal savings system 29–30 Potsdam Declaration (1945) 71–2 price controls 61, 62, 63, 66, 68, 76, 79, 80, 89, 139 price stabilization 87 priority production system 80 private banks 94 private sector capital stock, ratio to GDP 106, 107 privatization 137–69; ‘Administrative Reform without Tax Increase’ 141; All Nippon Airways (ANA) 164, 166, 167; characteristically Japanese 168–9; Civil Aeronautics Law (1952) 164; Diet (Parliament) 138–9, 140, 147, 149–50, 154–5, 157–9, 160, 162–3, 164, 166–7, 169; drive towards 138; Electric Power Development Corp. 166; Federation of Economic Organizations (Keidanren) 140–1; Fiscal Investment and Loan Program (FILP) 140; government ownership, concerns about 138; government share sales of privatized corporations 170–1; international aviation agreements 165; Japan Airlines (JAL) 138, 153, 164–8; Japan Monopoly Corporation 138; Japan Securities Dealers Association 161; Japan Tobacco (JT) 138, 150, 153, 159–64, 168; Japanese National Railways (JNR) 138, 139, 142, 143–52, 168; labour unions 139, 154–5; liberalization ideas 153; military industrial demands 139; MOF (Ministry of Finance) 139, 140, 142, 145, 149, 153, 155–6, 157, 158–9, 161–3; MOT (Ministry of Transport) 146, 147, 148, 149, 164, 166–7; MPT (Ministry of Posts and Telecommunications) 152, 155; nationalization and, phases of 139; Nippon Cargo Airlines Co. Ltd. (NCA) 165; Nippon Telegraph and Telephone (NTT) 138, 139, 142, 150, 152–9, 168; overmanning 142; price controls 139; Provisional Commission on Administrative Reform (PCAR) 141, 142, 146, 150, 153, 166, 169; public expenditure 138; public opinion, mobilization of 142–3; rationale
230 Index for 1980s privatizations 140–3; subsidies to JNR 142; technological developments 153; Toa Domestic Airline Co. (TDA) 164; Tokyo Electric Power Co. 156; Tokyo Stock Exchange (TSE) 147, 148–9 production function, Cobb–Douglas formulation 116–19 production incentives 62–3 productivity 10, 39, 191; capital productivity 11–12; Meiji Restoration 33; profit and 107–8; Tokugawa period 23 profits: capital investment and return rates 110–11; of private financial sector 9 property prices 192 public expenditure 138 public opinion, mobilization of 142–3 purchasing power parity (PPP) 3, 4, 10, 106, 173, 190 rates of return on capital 106–8 raw materials, shortages of 58 Reagan, President Ronald 141 refinancing, liberal loans for 96 regulatory excess 9 Renaissance 19 residue of war 69 retailing 200–1 returns on capital 119–21; US vs Japan 108 revenue-producing financial products 101 ‘reverse course’ of American Occupation 87 RFB (Reconstruction Finance Bank) 75–6, 78, 86–9, 95, 96; Bank of Japan and 86 rice harvest 75 RJR Nabisco Holdings 162 Rodrik, Dani 22, 186 Roosevelt, President Franklin D. 69, 71 Roosevelt, President Theodore 44, 47 Rotwein, E. 82 Rubinger, R. 22 Russia 36, 44, 143; and Trans-Siberian Railway 43–4; Triple Intervention against Japanese expansion 43; War with (1904–5) 58 Sachs, J. and Warner, A. 186–7 Sackton, F.J. 84 Saito, Prime Minister Makoto 41 Samuels, R. 62, 78, 79, 80–1, 91 Sandler, T. and Hartley, K. 37–8 Saxonhouse, G. 34
scale of economy 3 scale returns 117 SCAP (Supreme Commander, Allied Powers) 72–4, 75–7, 77–80, 81, 84, 87–8, 90, 95, 98 Schaede, U. 103 Schlesinger, A.M. Jr. 75 Schonberger, H.B. 73, 87–8, 89, 90 Sears Roebuck 200 seclusion policies 17–18 Second World War 2, 7, 75; international developments (1870–1940) 46–7, 58 securities companies 94 Seibu 200 Seiyu 200 Shantung Peninsula 43, 45 shareholders 195; provisions for 98–9; shareholding regulations 96–7; threshold reduction for 103 Shimazaki, T. 23 Shinto, Hisashi 153, 155 shipping 66 Shogun’s powers, surrender of 26 Singapore 101, 174, 178, 179, 180, 181 Sino-Japanese War (1894–95) 56–7 slowdown and stagnation 9–12 Smith, Adam 15 social classes 16 South Africa 174, 179, 181 Soviet Union 2, 86; Stalinist planning in 61 Spain 174, 178, 180 stagnation 30 state controls 80–1 state economic planning 74–7 state institutions, refashioning of 30 stimulatory fiscal measures, reluctance to use 191 stock market 96–7; crash (1920) 48–9; stocks and bonds (1935) 93; see also Tokyo Stock Exchange (TSE) strong-army policy 36 structural change 122–36; analytical concerns 129–32; comparisons with US 134–6; GDP growth across industries 122–9; Gini coefficient of inequality 123, 127, 130, 131, 134–5, 136, 203, 204–5, 210; modified Gini coefficients 127, 131, 134; outcomes, apparent change of 129–30; simulation of experience (1970–2005) 130–2; TFP (total factor productivity) 133–4 sub-contractors, designation of 63 subsidies to Japan national railways (JNR)
Index 231 142 substitution between capital and labour, elasticity of 117–18 Sumitomo 53, 81 Summers, R. et al. 106, 119 sumptuary laws 18 Suzuki, Prime Minister Zenko 141–2, 147 Sweden 174, 178, 181 Switzerland 3, 11–12, 174, 178, 179, 180 SWNCC (State–War–Navy Coordinating Committee) 72–3 system: evolution during 19–25; restructure for war production 62–5 Taisho era 41 Taiwan 1, 43, 72 Takahashi, Minister Korekiyo 52, 60 Takano, Y. 152, 153, 154, 155–6 Takenaka, Minister Heizo 159, 193 Takenaka, Y. 21 Tanaka, Prime Minister Kakuei 147 Tashiro, K. 18 tax reform 27–8, 31 technological developments 23, 153 textiles 33–4 TFP (total factor productivity) 133–4 Thailand 47, 174, 179, 181 Thatcher, Prime Minister Margaret 141 Toa Domestic Airline Co. (TDA) 164 Toby, R.P. 17 Tojo, General Hideki 47 Tojo, Prime Minister Hideki 80 Tokugawa, Iemitsu 17 Tokugawa, Iemochi 26 Tokugawa, Ieyasu 15–18 Tokugawa, Yoshinobu 26 Tokugawa period: agricultural tax yields 19; alternate attendance, practice of 21; alternate attendance (sankin kotai), system of 17; authority of shogun 16; central administration (bakufu) 16, 17, 18, 26, 27; closed country (sakoku) 17; daimyo (great names), power of 16–17, 21, 26; education, precursor of growth 21–2; feudal early period 16; governance, precursor of growth 22–3; import trade, authorization of 17–18; income and expenditure 18; institutional quality of government 22–3; internal military threat 17; life expectancy 23–4; main outlines 15–19; overseas voyages, prohibition of 17–18; power, consolidation of 18; productivity 23; seclusion policies
17–18; social classes 16; specialization in production, growth of 23; sumptuary laws 18; system evolution during 19–25; technology 23; transformation, stage set for 24; urbanization in 19–21; warring states 16; Western imperialism 24–5; wheeled vehicles, prohibition of 17; see also Meiji Restoration Tokyo Electric Power Co. 156 Tokyo Stock Exchange (TSE) 147, 148–9 Tokyo–Yokohama earthquake (1923), effects of 49 total assets 111 trade deficits 35–6 Trans-Siberian Railway 43–4 Transparency International 210 Triple Intervention 43 Truman, President Harry S. 87 Tunisia 174, 179, 181 Turkey 174, 179, 181, 203 Ueda, K. 94, 96 United Kingdom 6, 46, 65, 138, 174, 178, 180; British Labour Party 77; gold standard 51–2; opposition to Germany 47 United States 3, 5, 11, 46, 138, 174, 178, 180, 203; aid burden on 86–7; capital– output ratio 109–10; capital stock 6–7; Dept. of Agriculture 163; Federal Reserve 103; Federal Reserve system 51; Pacific War 46–7, 58; stability of growth in 6–7; treaty abrogation by 65 urbanization 19–21 Vaporis, C. 21 Venezuela 174, 179, 181 Vogel, Steven 141, 142, 152, 153, 155 wage controls 61, 85, 86, 90, 205 Wal-Mart 200 Wall Street Journal 148, 156, 157–8 Ward, M.D. et al. 38 wartime: economic destruction 67, 69; economy, supply arrangements 65–7; industrial coordination 75; production planning 59, 60; wars with China and Russia, effects of 48 Watanabe, S. 151 Weinstein, D. and Yafeh, Y. 94 West, M.D. 97–8, 102 Western imperialism 24–5 wheeled vehicles, prohibition of 17 Willoughby, General Charles 73
232 Index ‘window guidance’ (BOJ) 96 World Bank 11, 15, 24 Yalta Agreement (1945) 72 Yamagata, Prime Minister Aritomo 43, 44 Yamamura, Kozo 18, 19, 23, 36–7, 38–9, 54–5, 82, 115–16 Yasuda, Zenjiro 53, 81
Yawata Steel Mills 37 Yoshida, Prime Minister Shigeru 76, 79, 80, 84, 88, 152 Young, Ralph 87 zaibatsu: breaking up of 86; growth of 52–4