Asia-Pacific Dynamism 1550-2000 (Routledge Studies in the Growth Economies of Asia, 27)

Asia Pacific Dynamism 1550–2000

This volume showcases the latest research by a team of international scholars into the economic development of the Asia Pacific region. A geographically and historically diverse range of case studies cover the spectrum of Asian economic activity, from finance to trade and industry, exploring the central theme of the role of the market in intra-Asian economic activity and Asian-Pacific development. Asia Pacific Dynamism 1550–2000 builds on recent breakthroughs in the statistical analysis of Asian economic history, and opens up important new areas for research. Key topics covered include: • a unique standard-of-living comparison, covering the Asian region before 1940 • the role of education in Asian development • technology transfer and economic development • new perspectives on industrialization in Korea and Hong Kong. This book is of fundamental importance to economic historians with a particular interest in the Asia Pacific. It also offers a wealth of original source material, and innovative methodologies, that will be of interest to any economist or historian.

A.J.H.Latham is Senior Lecturer in International Economic History at the University of Wales, Swansea. He has written widely on African and Asian economic history, and is the author of Rice: The Primary Commodity (Routledge 1998). Heita Kawakatsu is Professor of Economic History at the International Research Centre for Japanese Studies, Kyoto. He is the author of Japanese Civilisation and the Modern West, and co-editor of Japanese Industrialization and the Asian Economy.

Routledge Studies in the Growth Economies of Asia 1 The Changing Capital Markets of East Asia Edited by Ky Cao 2 Financial Reform in China Edited by On Kit Tam 3 Women and Industrialization in Asia Edited by Susan Horton 4 Japan’s Trade Policy Action or reaction? Yumiko Mikanagi 5 The Japanese Election System Three analytical perspectives Junichiro Wada 6 The Economics of the Latecomers Catching-up, technology transfer and institutions in Germany, Japan and South Korea Jang-Sup Shin 7 Industrialization in Malaysia Import substitution and infant industry performance Rokiah Alavi 8 Economic Development in Twentieth-Century East Asia The international context Edited by Aiko Ikeo 9 The Politics of Economic Development in Indonesia Contending perspectives Edited by Ian Chalmers and Vedi Hadiz 10 Studies in the Economic History of the Pacific Rim Edited by Sally M.Miller, A.J.H.Latham and Dennis O.Flynn 11 Workers and the State in New Order Indonesia Vedi R.Hadiz 12 The Japanese Foreign Exchange Market Beate Reszat 13 Exchange Rate Policies in Emerging Asian Countries Edited by Stefan Collignon, Jean Pisani-Ferry and Yung Chul Park

15 Japanese Views on Economic Development Diverse paths to the market Kenichi Ohno and Izumi Ohno 16 Technological Capabilities and Export Success in Asia Edited by Dieter Ernst, Tom Ganiatsos and Lynn Mytelka 17 Trade and Investment in China The European experience Edited by Roger Strange, Jim Slater and Limin Wang 18 Technology and Innovation in Japan Policy and management for the 21st century Edited by Martin Hemmert and Christian Oberländer 19 Trade Policy Issues in Asian Development Prema-chandra Athukorala 20 Economic Integration in the Asia Pacific Region Ippei Yamazawa 21 Japan’s War Economy Edited by Erich Pauer 22 Industrial Technology Development in Malaysia Industry and firm studies Edited by K.S.Jomo, Greg Felker and Rajah Rasiah 23 Technology, Competitiveness and the State Malaysia’s industrial technology policies Edited by K.S.Jomo and Greg Felker 24 Corporatism and Korean Capitalism Edited by Dennis L.McNamara 25 Japanese Science Samuel Coleman 26 Capital and Labour in Japan The functions of two factor markets Toshiaki Tachibanaki and Atsuhiro Taki 27 Asia Pacific Dynamism 1550–2000 Edited by A.J.H.Latham and Heita Kawakatsu 28 The Political Economy of Development and Environment in Korea Jae-Yong Chung and Richard J.Kirkby

14 Chinese Firms and Technology in the Reform 29 Japanese Economics and Economists since Era 1945 Yizheng Shi Aiko Ikeo

Asia Pacific Dynamism 1550–2000

Edited by A.J.H.Latham and Heita Kawakatsu

London and New York

First published 2000 by Routledge 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 Routledge is an imprint of the Taylor & Francis Group This edition published in the Taylor & Francis e-Library, 2003. Editorial material and selection © 2000 A.J.H.Latham and Heita Kawakatsu Individual chapters © 2000 the contributors All rights reserved. No part of this book may be reprinted or reproduced or utilized 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 Latham, A.J.H. Asia Pacific Dynamism 1550–2000/A.J.H.Latham and Heita Kawakatsu. p. cm. Includes bibliographical references and index. 1. Asia—Economic conditions—Congresses. 2. Pacific Area—Economic conditions—Congresses. I. Kawakatsu, Heita 1948– II. Title. HC412.I.27 2000 330.95–dc21 99–088193 ISBN 0-203-46580-6 Master e-book ISBN

ISBN 0-203-77404-3 (Adobe eReader Format) ISBN 0-415-22778-X (Print Edition)

Contents

List of figures List of tables List of contributors Preface

vii viii x xiv

Introduction

1

A.J.H.LATHAM AND HEITA KAWAKATSU

PART I 1

11

Globalization, factor prices and living standards in Asia before 1940

13

JEFFREY G.WILLIAMSON

PART II

47

2 Four revolutions in the textile trade of Asia 1814–1994: the impact of Bombay, Osaka, the Little Tigers and China

49

D.A.FARNIE

3 Europe, China and Japan: transfer of silk reeling technology in 1860–95

70

DEBIN MA

4 The colonial origins of Korea’s market economy

86

MYUNG SOO CHA

5 South Korea’s late industrialization in comparative historical perspective

104

JAYMIN LEE

6 Export dynamics in Taiwan and Mainland China 1950–2000: a Schumpeterian approach HANS H.BASS

v

117

vi

Contents

7 Industrialization and institutional change in Hong Kong 1842–1960

149

DAVID W.CLAYTON

8 Education and development: the experience of the Four Little Tigers

169

DEREK H.ALDCROFT

PART III 9 Another monetary economy: the case of traditional China

185 187

AKINOBU KURODA

10 Money and growth without development: the case of Ming China

199

DENNIS O.FLYNN AND ARTURO GIRÁLDEZ

11 California and Nevada minerals in the Pacific Rim 1850–1900

216

DAVID J.ST. CLAIR

PART IV

243

12 The transmission of corporate cultures: international Officers in the HSBC Group

245

FRANK H.H.KING

13 Chaos and instability: the Asia Pacific rice trade in the 1990s A.J.H.LATHAM

265

List of figures

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.1 2.2 2.3 2.4 3.1 3.2 4.1 4.2 4.3 4.4 4.5 6.1 6.2 6.3 6.4 6.5 6.6

International real wage dispersion in the Atlantic economy 1854–1913 Real wage indices for East Asia Real wage indices for Southeast/South Asia Real wages in East Asia relative to Britain Real wages in Southeast/South Asia relative to Britain Trends in wage/GDP per capita ratio—Japan 1870–1938 Trends in wage/GDP per capita ratio—India 1873–1939 Trends in wage/GDP per capita ratio—Indonesia 1870–1940 Trends in wage/GDP per capita ratio—Thailand 1870–1939 Takeo Yamanobe Mill of the Osaka Spinning Company The vital role of labour in the mills of China Head Offices of Peel Holdings PLC Exports of raw silk Number of modern silk reeling basins in China, Japan and Italy 1861–1937 Coefficient of variation: Korean rice prices 1744–1995 Coefficient of variation: unskilled nominal and real wages in southern Korea 1915–90 Coefficient of variation: nominal and real wages of bricklayers in southern Korea 1914–90 Sectoral nominal wage coefficient of variation 1914–90 Real interest rates and return to capital in Korea 1912–91 Taiwan’s export composition 1952–95 China’s export composition 1965–95 Taiwanese direct investment in various countries 1985–96 Chinese SITC-76 import market shares in USA 1988–96 Chinese SITC-75 import market shares in Germany 1988–96 Chinese SITC-85 import market shares in Japan 1988–96

vii

15 26 26 29 29 35 36 36 37 53 54 55 63 72 74 87 90 92 93 95 119 120 123 124 125 126

List of tables

1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.1 3.1 3.2 5.1 5.2 5.3 6.1 A1 A2 A3 A4 A5 A6 A7 A8 A9

The Asian real wage hierarchy near the turn of the century Real wage performance in Asia 1820–1939 Real wage performance in Asia relative to Britain 1830–1939 Real wage performance in Asia by decades relative to the core 1820s–1930s Real wages in Asia relative to Japan 1831–1939 Regional real wage indices for India 1873–1939 Wage/GDP per capita ratio trends 1870–1939 Asian wage/rental ratio trends 1873–1939 Regional real wage indices for Indonesia 1878–1939 World production of, and world trade in, cotton manufactures 1913 Raw silk exports of China and Japan by types Comparative summary of silk reeling technology transfer in China and Japan 1860–95 Patterns of late industrialization: historical comparisons Trend of effective rate of protection Composition of value added in world market prices Advantages and disadvantages of Asian enterprises under different framework conditions Percentage change in merchandise exports, Taiwan and Mainland China 1979–97 Volume and composition of Taiwan’s exports 1952–95 Main export products of Taiwan (representing 75% export value) 1951–69 Main 20 SITC 3-digit level export products of Taiwan 1990–95 Main SITC 4-digit level export products of Taiwan in office machinery and electrical machinery (SITC 75 and 77) 1990–96 Volume and composition of Mainland China’s exports 1965–95 Main export products of Mainland China 1980–95 Economic relevance of small enterprises in Taiwan 1985–95 Decomposition of nominal export growth, various East Asian economies 1983/84–1993/94 viii

20 22 23 24 25 27 34 38 39 50 73 80 111 113 113 131 133 134 135 136 137 137 138 138 139

List of tables

A10 Mainland China’s exports by type of exporting enterprise 1994 A11 Mainland China’s exports by type of producing enterprise 1994 A12 Volume and distribution of Taiwanese investment in Mainland China by branches, Taiwanese data 1991–96 A13 Greater China’s import shares in major markets of industrial economies 8.1 Growth performance by region 1950–92/94 8.2 Total public spending as percentage of GNP 8.3 Primary, secondary and tertiary education ratios 1960–85 8.4 Educational attainment of the working population 11.1 California quicksilver production and exports 1850–90 11.2 Annual quicksilver production from Almaden and Idria 11.3 Distribution of California quicksilver exports to Pacific Rim countries 11.4 Average San Francisco quicksilver prices 1850–90 11.5 California manufacturers 11.6 California manufacturers, excluding gold mining 11.7 Manufacturing per capita in 1870 11.8 California flour mills 11.9 California iron working trades 13.1 World rice exports 1990–98 13.2 World rice imports 1990–98 13.3 China rice exports and imports 1990–98

ix

140 140 140 141 170 172 173 173 220 221 222 226 229 230 230 232 236 266 267 268

List of contributors

Derek H.Aldcroft graduated from the Victoria University of Manchester with degrees of BA (Econ) and PhD 1958 and 1962. He has been research Professor in Economic History at Manchester Metropolitan University since April 1994. He was formerly Professor and Head of Department in Economic History at the Universities of Sydney and Leicester. He has written widely on various aspects of British and international economic history. His latest works include Economic Change in Eastern Europe since 1918 (with Steven Morewood) (Edward Elgar 1996), Studies in the Interwar European Economy (Ashgate Publishing 1997), and Exchange Rate Regimes in the Twentieth Century (with Michael Oliver) (Edward Elgar 1998). Hans H.Bass studied economics, economic history and cultural anthropology at the University of Münster, Germany where he took his PhD. His thesis on nineteenth-century famines was published in 1990. Formerly Senior Research Fellow at the Institute of World Economics and International Management, University of Bremen, Germany since 2000, he is Professor of Economics at Bremen University of Applied Science. He has been guest lecturer at Shanghai Tongji University, Aichi University Toyohashi, and Xi’an Jiaotong University. He is currently working on China in the world economy. His latest works on China include Weltwirtschaftsmacht China (edited with Margot Schüller) (Institut für Asienkunde: Hamburg 1995) and China in der Weltwirtschaft (edited with Karl Wohlmuth) (Institut für Asienkunde: Hamburg 1996). Myung Soo Cha is Associate Professor, teaching economic history and macroeconomics in the Economics Department at Yeungnam University in South Korea. He was educated at Seoul National University (BA and MEcon) and Warwick University (PhD). His research interests include the international business cycle under the classical gold standard, the integration and segmentation of world grain markets since 1870, the impact of the inter-war agricultural depression on the Korean economy, inter-war Korean consumption trends, and long-term change in Korean living standards. He serves as Managing Director of the Korean Economic History Society. David W.Clayton is a Lecturer in the Department of History at the University of York. His PhD thesis, undertaken at the University of Manchester, was published x

List of contributors

xi

as Imperialism Revisited: Political and economic relations between Britain and China, 1950–54 (Macmillan 1997). He is currently working on Hong Kong’s economic development in the 1950s and 1960s, in particular the colonial government’s information infrastructure. D.A.Farnie is a Visiting Professor at The Manchester Metropolitan University. He has published a study of the role of the Suez Canal in world history 1854–1956 (1969). His main research interests remain focused upon the history of the cotton textile industry. He organized a session devoted to the global role of the cotton industry at the Madrid Congress in 1998. He has edited the new comparative study, Region and Strategy in Britain and Japan: Business in Lancashire and Kansai, 1890–1990 (Routledge 2000). Together with David Jeremy he is a founder-editor of the research project on North West Business Leaders 1600– 1990, launched in 1997. Dennis O.Flynn, Alexander R.Heron Professor of Economics at the University of the Pacific, has published over thirty essays on early-modern monetary history since 1978, fifteen of which have been reproduced in World Silver and Monetary History in the Sixteenth and Seventeenth Centuries (Variorum 1996). He has co-edited Studies in the Economic History of the Pacific Rim (Routledge 1998) and Pacific Centuries: Pacific and Pacific Rim history since the sixteenth century (Routledge 1999), and is a General Editor of the series The Pacific World: Lands, peoples, and history of the Pacific, 1500–1900 (Variorum/Ashgate, forthcoming). Arturo Giráldez, Associate Professor of Spanish Literature, University of the Pacific, received a PhD in Spanish Literature at the University of California, Santa Barbara in 1990 and a second PhD in History at the University of Amsterdam in 1999. Author of more than a dozen articles on early-modern monetary history, Giráldez is co-editor of Metals and Monies in an Emerging Global Economy (Variorum 1997) and a General Editor of the Variorum/ Ashgate series, The Pacific World: Lands, peoples, and history of the Pacific, 1500–1900 (forthcoming). Heita Kawakatsu received his DPhil from the University of Oxford, and was Professor of Economics at Waseda University, Tokyo (1990–98). He is now Professor of Economic History at the International Research Centre for Japanese Studies in Kyoto. His main publications in English are ‘International competition in cotton goods in the late nineteenth century’ in W.Fisher et al. (eds) The Emergence of a World Economy, 1500–1914 (1986), and ‘The Lancashire cotton industry and its rivals’ in K.Bruland and P.O’Brien (eds) From Family Firms to Corporate Capitalism (1998). He was co-editor (with A.J.H.Latham) of Japanese Industrialization and the Asian Economy (1994). His publications in Japanese include Japanese Civilization and the Modern West (1991), and he has edited The Asian Trading Sphere and Japan’s Industrialization 1500–1900 (1991) and A New Asian Drama: Five hundred years’ dynamism (1994). Frank H.H.King is Professor Emeritus, University of Hong Kong, and has most recendy been Visiting Professor at the University of Reading, England. A

xii

List of contributors

graduate of Stanford and Oxford Universities, he was an economist with the World Bank and Director of the Centre of Asian Studies, University of Hong Kong. His research focus has been the monetary history of East and Southeast Asia, and he is the author of a four-volume history of the Hongkong and Shanghai Banking Corporation. More recently Professor King has attempted to identify the corporate culture which explains Hongkongbank’s success and to identify the influence of that culture as the bank grew into a multi-corporate structure under HSBC Holdings. Akinobu Kuroda is Associate Professor of Eastern Asian History at the Institute of Oriental Culture, University of Tokyo. He graduated from the Faculty of Letters, Kyoto University, and received his Doctor of Economics from Kyoto University. His major publication is Chuka Teikoku no Kozo to Sekai Keizai (The Structure of the Chinese Empire and the World Economy) (Nagoya, Nagoya UP 1994). His specialties include the economic history of modern China and the comparative study of monetary history and theories of money. By analysing the pattern of traditional China, he is trying to show a concept of political economics which was utterly different from that of Japan and Western Europe. A.J.H.Latham is Senior Lecturer in International Economic History at the University of Wales, Swansea. He was educated at Merton College, Oxford, and the University of Birmingham where he took his PhD at the Centre of West African Studies. This work led to his book Old Calabar 1600–1891: The impact of the international economy upon a traditional society (1973). More recently he has published several books on Asia, and he has a particular interest in the international rice trade. His book Rice: The primary commodity (1998) is a study of the rice trade today, and he is currently preparing a study of the development of the rice trade over the last two centuries. Jaymin Lee is Professor of Economics and currently Chairman of the Department of Economics, Yonsei University, South Korea. He received his BA and MA in economics at Seoul National University and PhD in economics from Harvard University. He was a visiting scholar at the University of Cambridge and Rikkyo University, Japan. He has written mainly on issues related to late industrialization of East Asian NIEs, such as the historical meaning of outward-looking development and the empirical validity of the infant industry argument. His recent work includes editing a special issue on the economic miracle and crisis in East Asia for The Pacific Review. Debin Ma was born in Shanghai. He acquired a PhD in economics from the University of North Carolina at Chapel Hill. His research interest is mainly in the area of economic development and economic history, with a focus on developing an integrated and comparative perspective of economic development of East Asia. He is currently working as a research fellow of the Japan Society for the Promotion of Science for the Asian Historical Statistics Project at the Institute of Economic Research of Hitotsubashi University in Tokyo.

List of contributors

xiii

David J.St. Clair is Professor of Economics at California State University, Hayward. His areas of specialization and interest include California economic history, the history of California quicksilver mining, urban transportation history, and comparative economic systems. He has published in the Journal of Economic History, California History, and is the author of The Motorization of American Cities (Praeger Press 1986). He received his PhD in economics from the University of Utah in 1979 and was the 1979 recipient of the Allan Nevins Prize for the year’s best dissertation in United States economic history. Jeffrey G.Williamson is the Laird Bell Professor and Chairman of Economics at Harvard University. He is also a Faculty Fellow at the Harvard Institute for International Development and Research Associate at the National Bureau of Economic Research. A PhD from Stanford University, Professor Williamson taught at the University of Wisconsin for twenty years before joining the faculty at Harvard in 1983. The author of more than twenty scholarly books and nearly 150 articles on economic history and development, Professor Williamson was recently elected President of the Economic History Association. His most recent books are Industrialization, Inequality and Economic Growth (Elgar 1997), The Age of Mass Migration (Oxford 1998, with T.Hatton), Growth, Inequality, and Globalization (Cambridge 1998, with P.Aghion), Globalization and History (MIT 1999, with K.O’Rourke) and The Mediterranean Response to Globalization before 1950 (forthcoming, Routledge, with S.Pamuk).

Preface

The papers in this volume were presented at the Twelfth International Economic History Congress in Madrid 24–28 August 1998. They can be seen as part of an on-going conversation about the economic activity of Asia and the Pacific region within the world economy which goes back to papers presented by the editors to the Ninth International Economic History Congress in Berne in 1986 (Latham 1986; Kawakatsu 1986). At the Tenth International Economic History Congress in Leuven in 1990 these discussions continued in a session convened by the editors, the papers from this being published as Japanese Industrialization and the Asian Economy (Latham and Kawakatsu 1994a). When the Eleventh International Economic History Congress met in Milan in 1994 the editors chaired yet another session, the papers for which were published in the conference proceedings, as The Evolving Structure of the East Asian Economic System since 1700: A comparative analysis (Latham and Kawakatsu 1994b). A major theme to have emerged during these sessions has been the role of the market in intra-Asian economic activity, and market forces have been seen to have played a crucial role in Asian-Pacific development in many of the numerous diverse papers presented over the years. With time, more and more scholars have been drawn into the colloquy, and this volume brings together the latest contributions. References Wolfram Fischer, R.Marvin McInnis and Jürgen Schneider (1986) The Emergence of a World Economy 1500–1914, Wiesbaden: Franz Steiner Verlag. A.J.H.Latham (1986) ‘The international trade in rice and wheat since 1868: A study in market integration’ in Fischer, McInnis and Schneider (1986), pp. 645–64. Heita Kawakatsu (1986) ‘International competition in cotton goods in the late nineteenth century: Britain versus India and East Asia’ in Fischer, McInnis and Schneider (1986), pp. 619–44. A.J.H.Latham and Heita Kawakatsu (eds) (1994a) Japanese Industrialization and the Asian Economy, London: Routledge. A.J.H.Latham and Heita Kawakatsu (eds) (1994b) The Evolving Structure of the East Asian Economic System since 1700: A comparative analysis, Proceedings of the Eleventh International Economic History Congress, Università Bocconi, Milan. xiv

Introduction A.J.H.Latham and Heita Kawakatsu

Part I The paper which opened the session at Madrid was Jeffrey G.Williamson’s ‘Real wages and relative factor prices in the Third World, 1820–1940: Asia’. Even two decades ago this paper could not have been written. It depends on recent breakthroughs in the historical statistics of the Asian economy. These advances came in the early 1980s and created a revolution in the understanding of Asian development. Williamson’s pathbreaking study carries the work forward and notes that as early as 1914 there were big economic differences between the wealthy industrialized countries of Europe, and the poorer countries of Asia from the Indian subcontinent to Japan. But when did these differences appear? Williamson argues that conventional GDP per capita data is too incomplete and unfocused to provide an answer to this fundamental question. Real wages are a better indicator of the economic well-being of ordinary people than the averages obtained in GDP per capita estimates, as these lump together the rich and the poor and ignore the effect of income distribution. A crucial feature in these real-wage estimates is the price of rice, the key commodity in the market baskets of the Asian urban poor, and the main foodstuff throughout most of the region. Fortunately rice prices are now relatively well recorded as a result of the pioneering work of the 1980s. He therefore suggests a new research agenda for the region, using real wages and wage/rental ratios for thirteen major Asian regions: Burma, China, India (North, South, East and West), Indonesia (Java and Outer Settlements), Japan, Korea, the Philippines, Taiwan and Thailand. Some of these regions have good statistics going back to 1820. When Japanese real wages are examined over time, it appears that although living standards there kept pace with those in Britain from the 1870s to 1914, they did not catch up with them. Elsewhere in Asia living standards were actually falling. Most Asian countries did not make much advantage of the great globalization boom from 1870, and some hardly at all. Williamson closes by setting an agenda for further research. By 1940 there was an enormous gap in living standards between Asia and Europe. This was not due to events in the inter-war years. It was mainly due to Asia’s failure to exploit the economic opportunities of the expanding world market in the preceding two hundred years, particularly from 1870 to 1914. Why did Asia fail in this way? There is as yet no clear answer, and solutions will only be obtained by improving the statistical base for investigation. The innovatory work 1

2

A.J.H.Latham and Heita Kawakatsu

of the last twenty years must be sustained. This is the task for the next generation of scholars working on the development of the Asian economy since 1800. Part II Relating to the globalization issue, D.A.Farnie contributed a paper outlining the sequence of changes in the textile trade of Asia since 1814. First came Britain’s domination of the Indian market, although Farnie is keen to point out that yarn exports were more important than cloth, the yarn being used by India’s expanding army of hand-loom weavers who supplied cloth to local markets. From the 1860s spinning mills were established in Bombay and began to oust Lancashire yarn from India, and then from the China market. But the Bombay mills in turn were hit early in the new century by the rise of the Japanese cotton industry, which thrust Indian yarn out of the China market, and then out of India itself. Japan could buy Indian raw cotton, spin it, weave it, and sell it back to India cheaper than the Indian mills could! By the 1930s India was Japan’s biggest export market, whereas once it had been Britain’s biggest export market. Meanwhile, the Japanese and others had built spinning mills in Shanghai, which enabled Chinese mills to supply their own hand-loom weavers. After the Pacific War of 1941–45 there came a great change in the textile trade, with the worldwide acceptance of denim jeans and factory-made clothing. These factories used unskilled labour, and gave countries with cheap workers like Hong Kong, Taiwan and South Korea a great advantage. They could spin the yarn, weave the cloth, and make the garments for the international market. The new man-made fibres easily fitted into this new structure. But as these countries prospered, another phase began to emerge, with China utilizing her huge reserves of unskilled labour to become the new leader in world markets. Debin Ma was also concerned with textiles, in his case silk. He notes that from 1860 to 1940 Japan enjoyed a spectacular growth in raw silk exports while Chinese silk exports stagnated. Japan’s silk exports only overtook China in the twentieth century, but her success was due to the way she adopted European machine silk reeling technology between 1860 and 1895. China experimented with machine silk reeling, but was less successful with the new technology than Japan. There was admittedly initial resistance to the new technology in both China and Japan, but in Japan government support helped overcome resistance, and Japanese businessmen were successful in adapting the new technology to local conditions. Essential features were retained like the double twisting mechanism and steamheated ovens, but the equipment was simplified, with wooden replacing steel parts, and water replacing steam power. Rural family cooperatives took the place of factories. In China the concept of private enterprise did not exist, so the early reeling factories were in the international enclave of Shanghai, where they were not subject to official interference. But this also meant that the technology was not adapted widely by the local people in China, and remained controlled by foreigners. Cocoons are highly perishable and it proved difficult for the Shanghai mills to procure a regular supply from the countryside because of the poor infrastructure.

Introduction

3

In Guangdong the situation was better than in Shanghai, and local adaptation proceeded well; but after riots by traditional silk weavers, many of the mills were closed, and those that remained were placed under Government control. There was virtually no connection between the Shanghai and Guangdong millers, and they adopted different technologies. The essential difference between the success of Japan and the failure of China was that in Japan modernization was accompanied by Government support for private enterprise and a Western style market economy. In China private entrepreneurship was ignored or suppressed. China failed because the government tried to manage production units, and many of the government ventures were loss-making and fraud-ridden. Ma suggests that ultimately Japanese success lay in the fact that the Japanese authorities realized that Western economic advance was not just a matter of modern machinery, but involved the adoption of the free market and its legal framework. When this was adopted technical advances would follow naturally. Markets were also an issue for Myung Soo Cha, who dealt with Korea under Japanese rule. He notes that although rudimentary markets had long existed in Korea, South Korea’s modern market economy was actually created during the period of Japanese colonial rule from 1910 to 1945. To demonstrate this he uses price and wage information. Prior to Japanese rule regional differences in rice prices were very much higher than in pre-modern Japan, indicating that rice markets were not as well integrated as in Japan. By contrast, Japanese rice markets appear to have been as well integrated as in modern Korea today. Even in China rice markets were better integrated than in Korea. But when Japan annexed Korea in 1910, market integration was speeded up by railways, telegraphs, telephones, and the adoption of market reforms including property rights and a modern currency system. So Japanese colonialization effectively introduced an integrated national rice market in Korea. What was true for rice was presumably true of other goods. As for labour markets, free and slave labour existed side-by-side before Japanese rule. On the eve of Japanese annexation wage differentials between regions were markedly higher than in Japan. Real wages also showed marked regional differences. But after annexation these differences shrank and from the second decade of Japanese rule a national market for unskilled labour began to emerge. This was similar to what happened in Meiji Japan and, incidentally, Prussia. As for the capital market, at the beginning of colonial rule about 1910, interest rates were 40 to 50 per cent per year in the unregulated moneylending market, and 11 to 14 per cent per year in the regulated modern banking sector. These rates held through the 1920s in both sectors and then decreased during the 1930s. But in the 1960s after the chaos of the Korean War unregulated rates were as high as in the 1920s, and the differential with the regulated sector was similar to those days. The implication is that unlike goods and labour, a modern integrated capital market was not introduced under Japanese rule, and has not been established since. Policies to make South Korea self-reliant after the US withdrawal from Vietnam included financial restrictions which prevented liberalization of the capital market. The post-war industrialization of South Korea was the focus of Jaymin Lee’s paper. He argues that although Gerschenkron’s model of late industrialization has

4

A.J.H.Latham and Heita Kawakatsu

been discredited for countries in nineteenth-century Europe, it may be relevant to South Korea after the Pacific war. Gerschenkron argues that for a late-comer to succeed there must be a sudden spurt of growth in the economy. There must also be rapid and substantial structural change moving resources to dynamic sectors, and government industrial policy must be systematically applied to achieve this. New dynamic infant industries must be created and helped to grow and mature, and they will embody the success of the entire policy. These features do seem to be present in South Korea. After a low growth rate from 1953 to 1962, South Korea’s per capita GNP suddenly jumped to over 6–7 per cent per annum and sustained this rate of growth until the 1990s. This was accompanied by rapid structural transformation brought about by the government’s industrial policy. The government actively promoted infant industries by giving them subsidies and protection. Large government sponsored conglomerates managed investment in major projects and dynamic industries. Heavy and chemical industries were promoted as infant industries, and they flourished and grew. In short, South Korea’s late industrialization does seem to have demonstrated the features outlined in Gerschenkron’s scheme. Indeed, South Korea seems to fit Gerschenkron’s model better than Germany, Russia or Japan. Hans Bass was also concerned with developments since the Pacific War, and examined the export performance of both Taiwan and Mainland China since 1950. His analysis is couched in a Schumpeterian framework, especially Schumpeter’s view that overall growth in an economy is based on a continuing series of innovations. These may be technological innovations or the opening of new markets. In 1950 Taiwan’s foreign trade was made up of agricultural commodities, but these gave way to processed agricultural products, then simple manufactures. Later came technology-based products, which became ever more advanced. China was a much more inward looking economy, and only exported agricultural products and textiles. Petroleum products then became important, but were hit by declining world prices, and were then followed by a range of simple manufactures. Only very recently has China started to produce more advanced technology-based products, and is still a long way behind Taiwan in the sophistication of these goods. So China’s export expansion was less innovation-driven than Taiwan’s. The Taiwan authorities gave duty relief on items which were imported to produce exports, and established export processing zones in which newcomers were encouraged. In China the main stimulus to exporters was to free them from the restrictions of the central plan, although this liberation was only really effective in the 1980s. From this time foreign direct investment was encouraged in export industries. Effectively China was following a similar path to Taiwan, but a step behind. The relative difference in size of the two economies has to be borne in mind. As the export sectors of the two economies grew, economic linkages developed. Taiwanese entrepreneurs made direct investments in China in export-orientated industries, particularly where lower labour and input costs were important. The Taiwan economy was made up of a large number of small independent enterprises, but up to the late 1970s China’s foreign trade was in the hands of a few large government-controlled companies. They were split up into smaller units during the 1980s, as China moved towards a market

Introduction

5

economy. Now direct contact and investment from Taiwan became much easier. So China was able to move towards more sophisticated technology-based exports, although there is still a long way to go. China continues to lack flexibility in its domestic economy and export sector, and some state industries, like the textiles industry, are hopelessly outdated and lacking in investment. Taiwan with its many small family enterprises is much more flexible. But family businesses have limitations in technologically-advanced world markets, so Taiwan may need to move to larger corporations, and China may need to continue splitting its huge government enterprises into smaller, more market-conscious units. Opportunities for more interaction between Taiwan and China may continue in an ever more complex world market situation. David W.Clayton looked at industrialization in yet another Asian country, in this case Hong Kong to 1960. According to Douglass North, institutional evolution is the key to establishing a nation’s comparative advantage. Institutional evolution reduces transaction costs, and so stimulates economic development by giving lower prices in international markets. In Hong Kong there was originally an ethnic divide between Chinese business organizations, and the European dominated Hong Kong Chamber of Commerce founded in 1861. The Chinese organizations were based on the many dialect groups, and the first point of reference to a Chinese businessman was his family and clan. This worked very well when Hong Kong was an entrepôt to China, and also worked well for trade with Southeast Asia where the same dialect groups were well established. Meanwhile the Chamber of Commerce, with its links to other chambers of commerce round the world, was able to regulate trade and provide information for European merchants involved in world markets. Links with the Chinese business community were arranged through compradors who acted as go-betweens. Faced with the problems of reconstruction after 1945, the Hong Kong Government took a more active role in the commerce of the Colony, and established the Department of Trade and Industry. The Chamber of Commerce felt this encroached on their activities, but changed their view after 1951 when the Chinese Communist embargo blocked trade with Hong Kong. Now the Chinese businessmen who had worked the China trade were forced to look for business elsewhere. They were not members of the Chamber, and the Chamber realized there was a need to help and regulate them. Thus over time informal networks merged into formal institutions, then the state intervened to take over the supervision of these activities. This seems to accord with North’s conjecture, but formalization was slow to develop because informal procedures had worked so well. Random shocks like the Pacific War and the Chinese embargo brought about change, rather than straightforward economic evolution. Rapid industrialization in the 1950s also encouraged businessmen to accept state assistance in terms of information and regulation. Hong Kong, South Korea, Taiwan and Singapore were the focus of Derek Aldcroft’s paper. He investigated the role of education in the development process of these countries. They enjoyed the highest growth rates in the world between 1960 and 1985, and living standards more than quadrupled. Income levels in Hong Kong and Singapore moved close to those of Western nations. This spectacular

6

A.J.H.Latham and Heita Kawakatsu

growth was brought about by increased factor inputs, high investment, rising participation rates and inter-sectoral movements of labour, rather than improvements in total factor productivity. Of key importance was education. These countries maintained their faith in the human capital revolution. So was human resource development the key to their prosperity? Aldcroft points out that even in the 1950s they had an advantage that many other Asian countries lacked, a good level of access to elementary education and a high literacy rate. By the 1960s primary school enrolment almost matched those of Western nations, and by the 1970s literacy rates were close to 80 per cent in all four countries. The pattern of educational improvement began with a high priority on basic education, then expansion in secondary and tertiary education. There was also vocational training as the skill requirements of the economy became more demanding. Basic primary education was funded largely by the state in all countries, but private sources made a significant contribution at higher levels. There were, however, differences between the four countries, particularly in the tertiary sector. Technical and vocational education expanded in all countries from the 1960s, especially in South Korea and Taiwan. What is significant is that these countries had a relatively high endowment of human capital prior to the onset of their modern economic growth. Thus it seems that to modernize, countries needed a highly articulate and mobile labour force which was able to develop and utilize new technology and ideas. Tentative estimates suggest that 11–15 per cent of East Asian growth was due to the contribution of education or human capital improvement, and these may be underestimates. Education may also have played an important role in reinforcing national identities and achieving social cohesion in the disarray of the years after the Pacific War. So Aldcroft argues that these four East Asian countries could not have transformed themselves into super-growth states without a concerted effort to raise the quality of their human capital stock. Education was an integral part of their modernization. Part III In this section, papers were given which related to China. Akinobu Kuroda studied the monetized economy of traditional China. He notes that John Hicks believed that money was created by merchants, who needed a convenient means of payment. Coins of a particular intrinsic value could be acceptably exchanged for items of equivalent intrinsic value. But Hicks thought that in China the idea of intrinsic value was bypassed, and from the beginning coins were just tokens to make payments. Both Hicks and Weber believed the Chinese authorities issued copper cash coins of lesser intrinsic value than their nominal value, so as to profit by issuing them. But Kuroda argues that although copper cash coins were issued up to the sixteenth century, the authorities did not profit from them as their intrinsic value was actually greater than their nominal value. For this reason they were often melted down. He believes the intention of the authorities was to unify the country, and they wanted to discourage regional economic autonomy by introducing a nationwide currency system. To drive regional currencies from the market the official coins had to be of high intrinsic value, as low-value coins could be

Introduction

7

counterfeited cheaply or replaced by local alternatives. Copper cash, however, was not good for long distance trade, as it was too heavy to carry in large amounts. So the authorities accepted regional variations in the number of coins strung together in strings of equal nominal value. But despite their intentions, a big increase of forged cash came into circulation from the fifteenth century. People used official coins for saving, and counterfeit cash for everyday transactions. During the sixteenth century silver flowed into the country, and was adopted by the Ming dynasty for taxation. Because it was of higher value in relationship to its weight than copper, it was easier to carry large sums over long distances. Inter-regional payments could now be made more easily, encouraging regional economic integration. But interregional grain movements were also made easier by the new currency, leading to food shortages as grain was traded out of particular regions. A seventeenth-century crisis resulted. The Qing government solved it by issuing more copper cash, and establishing regional granaries which purchased their stockpiles with copper cash. In the eighteenth century no more silver coins were issued, but copper coins continued to be issued. There was no fixed exchange rate between copper and silver coins. Silver became the inter-regional currency and copper cash the regional currency, and the regions maintained their local currency characteristics. Copper cash continued to be used for petty transactions, fulfilling peasant needs for everyday transactions and savings. Silver was used for inter-regional trade. In this way the authorities maintained national economic unity, and discouraged regional separatist tendencies. Silver was also an issue for Dennis O.Flynn and Arturo Giráldez. From the sixteenth century silver from the mines of Peru, Mexico and Japan found its way to China as the price of silver there was higher than anywhere else in the world. The Ming authorities were moving to a silver currency system after the collapse of paper money in the fifteenth century, and their demand for silver for coinage was driving up the price. To secure an inflow of silver, they had to sell silk and other goods abroad in exchange. But did this benefit China? Flynn and Giráldez argue that the silk exported to pay for the silver was a drain on China’s resources. Silk exported could not be consumed in China! Resources put into producing silk were resources which could have been used to produce things needed in China. If paper money had been used instead of silver, those resources converted into goods to be exchanged for silver would have been available for domestic investment and consumption. In China’s case, silver money actually replaced paper money, which had existed since the eleventh century but had failed. It had been debauched when the Ming government overproduced it to finance military expenditure. This is why China moved to a silver system in the sixteenth century, and its use was confirmed when tax payments were stipulated in silver in the 1570s. As a result the price of silver in China jumped, causing the inflow of silver and a doubling of the world silver price. The inflow of silver continued and grew through to the eighteenth century, and ever greater exports of silk and tea had to be made to pay for it. So China’s economy was distorted for three centuries to obtain silver for currency, and resources which could have been used to develop the economy were diverted in this futile pursuit. Silver cannot be consumed like silk and tea! The collapse of

8

A.J.H.Latham and Heita Kawakatsu

the Ming paper currency was a disaster for China, causing permanent long-term damage to her economy. The flow of silver into China was the theme of David St. Clair’s paper. He discusses the flow of silver and gold from California and Nevada in the late nineteenth century. Most of the gold went to eastern United States ports and Britain, and China received very little. But with the Comstock silver boom, huge amounts of silver were produced in California, and most of it went to China from San Francisco. More than half was exported as Mexican silver dollars, used widely in China. This silver continued the flow of silver to China which had been going on for three centuries. As there was an ample flow of California gold to Europe and Western markets, Westerners were happy to see silver go to China. There was enough gold and silver for all! California mercury (quicksilver) was an important part of this story, as it was used to extract both gold and silver from ore. From 1850 to 1900 half the world’s mercury came from California. This was crucial to gold and silver production in California, as mercury amalgamation was the only practical extraction technique until the development of the cyanide process in the 1890s. California quicksilver drove down quicksilver prices, effectively controlled by a Rothschild cartel. The consequences for California of the precious metal boom were enormous. Previously the economy had been rudimentary, with a small population producing hides and tallow. With gold California became a boom area, the mining boom triggering off a boom for suppliers of wheat, cattle and sheep. Manufacturing output also increased to supply the growing population, and to provide mining equipment. But for the Pacific, the crucial effect of the California boom was the flow of silver into China, sustaining a trade pattern vital to Asian and world trade. Also the growth of California stimulated the entire economy of the Pacific Rim, and provided a growing market for western Pacific producers, especially China and Australia. Part IV The last two papers were rather different, and related to the 1990s. Frank King’s paper was about the Hong Kong and Shanghai Banking Corporation (HSBC Holdings), which has a preeminent position in East Asian banking. His paper examines the corporate culture of this financial conglomerate. His focus is on the International Officers who, as executives, function on a Group-wide basis, and maintain the working relationship of the various components of the group. Their origins are directly traceable to the Hongkong Bank founded in 1865, and their role and priorities are to be understood in this historical context. By 1998 the Group staff of 130,000 overall operated 5,500 offices in 79 countries with total assets of £286,391 million. Central to the role of the International Officer is the belief that a generalist banker continues to play a major role in an industry of specialists. The International Officer is a key figure because management traditionally has given priority to a high level of liquidity, rapidity of on-line decision making, and the absence of loan committees. The trust necessary to enable the International Officers to act with such freedom depends on a close personal

Introduction

9

knowledge of colleagues, creating the need for an ‘élite’ corps sufficiently small that such personal knowledge can be maintained. As the bank expanded, the problem of retaining the corporate culture was real. A head office in London is contrary to the traditions of Hongkong Bank. The head office of Hongkong Bank (HKBL) is in Hong Kong. In mid 1998 there were 330 International Officers with some 50 more acting on a temporary basis. They are the direct successors to the 249 executives who managed the Hongkong Bank in 1941, but their role has developed and changed. The International Officer is not designed as a purveyor of old-time Hongkong Bank culture; but from the very nature of his career he plays a role in furthering inter-cultural understanding within the group, and from his background he brings a touch of Hongkong Bank however much transformed in detail. In the mid 1990s 75 per cent of those newly recruited into the International Staff were British citizens, but there were 25 per cent who included Commonwealth citizens, Americans, citizens of South Asian countries, Malaysians and Hong Kong Chinese. They remain generalists but receive continued training, including specialist training and experience. They are assigned to posts of varying character and responsibility, but can be transferred by management without recourse and are available as trouble shooters. So they are on the hypothetical list from which senior managerial officers will usually be selected. Normally an International Officer is assigned to a position fitting his experience. He brings international background and corporate culture from one subsidiary to another or from one region/country to another. For both emergency and routine matters, the posting of this small corps of International Officers is key to the control of the holding company of the HSBC Group. Is a generalist career in banking still viable in today’s fast changing financial world? King argues it is difficult to argue with success. A.J.H.Latham provides an end-note to this collection of papers with a study of the Asia Pacific rice trade in the 1990s. During this decade the world rice trade more than doubled from over 11,000 metric tons in 1990 to over 27,000 in 1998. Thailand was the predominant exporter, followed normally by the USA and Vietnam, which has returned to the international market after obvious difficulties. Other major Asia Pacific exporters are China, which sometimes exports and sometimes imports, and Australia. Uruguay is a major exporter to Peru, Chile, and Brazil, and Argentina has recently become another exporter to these markets. But the disturbing feature of the 1990s with serious implications for the economy of the Asia Pacific region is the instability of the trade. Japan is normally self-sufficient in rice but in 1994 had to import heavily because severe cold wet weather damaged her domestic crop. Indonesia too had to make big imports that year, and in 1995 and 1996, although she also had pursued a selfsufficiency policy. In the early 1990s China exported a growing total of rice amounting to 8 per cent of world exports, but in 1994 was also hit by bad weather and flooding. Because China had already committed herself to exports, she went ahead with these but had to buy rice to make up the shortfall for her home consumers. So in 1995 China made net imports equivalent to 9 per cent of world imports, a switch round of nearly 20 per cent. She continued to import in 1996, and did not return to a net export situation until 1997. In 1998 China’s exports were equivalent

10

A.J.H.Latham and Heita Kawakatsu

to over 12 per cent of world trade, but there were terrible floods in the autumn, the worst since 1954. So looking back, 1995 was a crisis year in the Asian rice trade, with China, Bangladesh and Indonesia being forced to make huge imports which no-one had expected. How could the rice trade cope with this sudden increase in demand, particularly when China, one of the usual large exporters, was now itself actually importing rice? Fortunately India had built up a huge stockpile, and was able to release this onto the world market. India’s ports and roads, however, were hardly able to cope. As has been indicated, 1998 was another crisis year, because El Niño weather conditions struck, causing drought in Indonesia and the Philippines and bringing heavy rain and floods to South America. In the Philippines there was a political crisis, at a time when she needed more rice than her infrastructure could handle. Fortunately, China continued to export, despite her own problems, and India again came into the market. So 1998 was another year of crisis in the rice trade which coincided with the general economic and political crisis in the East. The chaos in the rice trade did not cause the overall crisis, but was an additional destabilizing feature. The origin of the crisis goes back to the autumn of 1994 when the Chinese devalued the yuan. Prices were rising at a rate of 25 per cent a year, and the swap rate for the yuan was falling in response. So the official managed exchange rate was lowered to the level of the swap rate, an effective devaluation of 33 per cent. China was now a very low price exporter and her devaluation hit her Asian competitors. The effects of this devaluation took a couple of years to work through to the rest of Asia, but when Thailand was forced to devalue in July 1997 the Asian economic crisis had arrived, and worsened through 1998. The combination of the devaluation and the rice chaos of 1994 was a disaster for China and the East. By 1998 the general Asian economic crisis was at its height; then to make matters worse, there were more floods on the Yangzi, at a time when China was experiencing massive bankruptcies in her regional development corporations. So rice crises and general economic crises seem to have occurred simultaneously in Asia in the 1990s, with ongoing consequences for the whole Asia Pacific region. Asian Pacific dynamism is likely to be modified by these problems for the foreseeable future.

Part I

1

Globalization, factor prices and living standards in Asia before 1940 Jeffrey G.Williamson

By 1914, there were very big economic gaps between the European industrial core and countries around Asia, from Japan to what we now call Pakistan. When did the gaps appear? Can they be explained by lags in the diffusion of the industrial revolution after 1780, or did the gaps appear much earlier? What about the first great globalization boom after about 1870? Which countries in Asia started catching up, which fell further behind, and which held their own? What role did globalization and demographic forces play? These are the big questions raised by W.Arthur Lewis, but conventional quantitative evidence, like GDP and trade data, is simply not enough to rise to his challenge. In an effort to suggest a new research agenda for Asia, this essay uses a new database on real wages and wage/rental ratios for thirteen major Asian regions—Burma, China, India (North, South, East and West), Indonesia (Java and Outer Settlements), Japan, Korea, the Philippines, Taiwan and Thailand. These thirteen regions, often documented from as early as 1820, form the database for the paper, and they are well balanced among the three major areas consisting of South Asia (four), Southeast Asia (five) and East Asia (four) so that we can say something about ‘north–south gaps’ within the region, as well as between it and the European core. New evidence, controlled conjectures and an agenda Two important features of the world economy after 1970 also characterized the economy after 1870. First, there was rapid globalization a century ago too: capital and labour flowed across national frontiers in unprecedented quantities, and commodity trade boomed as transport costs dropped sharply. Second, the late nineteenth century saw an impressive convergence in living standards, at least within most of what we would now call the OECD club, but what historians call the Atlantic economy. Poor countries around the European periphery tended to grow faster than the rich industrial leaders at the European core, and often even faster than the richer countries overseas in the New World. This club excluded most of what is now called the Third World and Eastern Europe, and even around this limited periphery there were some who failed to catch up. Nonetheless, there was convergence. It was not always that way: unambiguous divergence was the case earlier. The 13

14

J.G.Williamson

Atlantic economy in the first half of the previous century was characterized by high tariffs, modest commodity trade, no mass migrations, and an underdeveloped global capital market. Two profound shocks occurred in this environment still hostile to liberal globalization policy: early industrialization in Britain, which then spread to a few countries on the European continent; and resource ‘discovery’ in the New World, set in motion by sharply declining transport costs linking overseas suppliers to European markets, so much so that real freight rates fell by an enormous 1.5 per cent per annum between 1840 and 1910 (O’Rourke and Williamson 1999, ch. 3). These two shocks triggered a divergence in real wages and living standards across the Atlantic economy that lasted until the middle of the century (Williamson 1996).1 Figure 1.1 shows that the convergence which started in mid-century continued up to 1914: a plot of the dispersion of real wages is given there, documenting what the modern macro-economists call sigma-convergence. The line with the diamonds on the upper left of Figure 1.1 is based on a 13-country Atlantic economy sample including Australia, Belgium, Brazil, France, Germany, Great Britain, Ireland, the Netherlands, Norway, Portugal, Spain, Sweden and the United States. The dashed line in Figure 1.1 documents convergence for an expanded 17-country Atlantic economy sample, now including in addition Argentina, Canada, Denmark and Italy. This measure shows the convergence tide ebbing around 1900. If we exclude Canada and the United States, two ‘exceptional’ rich countries which bucked the convergence tide, convergence continues rapidly up to 1914 (the 15-country sample plotted with the triangles). If we exclude in addition two Mediterranean countries which failed to play the globalization game, Portugal and Spain, convergence up to 1914 is faster still (the 13-country sample plotted with the squares). Note that this convergence took place in spite of a policy backlash starting after the 1870s. That is, with the exception of Britain, Ireland, the Netherlands, and some of Scandinavia, Europe and its overseas settlements began to retreat from liberalism during the 1880s, long before the interwar rush to autarchy. The globalization backlash took the form of rising tariffs and less generous attitudes towards immigrants (O’Rourke and Williamson 1999, ch. 6). While the rise in tariffs was not enough to offset the commodity price convergence induced by transport revolutions, it did mute it somewhat. What about Asia? It is far too easy to assume that enormous distances kept Asia completely out of this first great globalization boom, except for a few expensive spices and raw materials which could bear the freight. True, trade shares were far lower in Asia than in Europe, and a very small share of European financial capital flowed into Asia up to 1914. Yet, transport revolutions created profound price shocks in late nineteenth-century Asia, and a move towards open trade policy (with the help of imperialist gunboats) created even more liberal trade attitudes in the region than was true even of the Atlantic economy after the 1870s, an attitude that seems to have persisted through much of the recent Asian miracle. What happened to Asian real wages and living standards in response to the challenge of both the European industrial revolution and the first great globalization boom? These are the questions that motivate half of this chapter. They are in the tradition

Living standards in Asia before 1940

15

Source: Williamson (1995, Table A2.1; revised in O’Rourke and Williamson 1997) Figure 1.1 International real wage dispersion in the Atlantic economy 1854–1913.

of W.Arthur Lewis who was the first to ask whether the core pulled along the periphery during this first great globalization boom (Lewis 1969, 1978a, 1978b). It was he, together with Alexander Gerschenkron (1952), who first tried to break economic history’s tenacious fixation on the industrial leaders, Lewis focusing on the Third World and Gerschenkron on European late-comers like Italy and eastern Europe. The other half of this chapter asks when the great divide between core and periphery first appeared. By the end of our period, there were huge gaps in living standards and GDP per capita between agrarian Asia and the industrial northwest of Europe. When did the great divide open up? During the late nineteenth-century growth boom? During the early industrial revolutionary decades after 1780? Even before? And what are the explanations? The timing and location of industrial revolutions, population growth, globalization, the rise of efficient political systems, or all of the above? Furthermore, what about within Asia? When Japan started its experiment with industrialization in the 1880s, where did it stand in Asia’s economic pecking order? Were its living standards already far above those in Southeast Asia—in, for example, what we now call Burma, Thailand and the Philippines—and in South Asia—in what we now call Bangladesh, Pakistan and India? Some have recently argued that India probably had living standards comparable with England in the mid-eighteenth century. Others have recently argued the same for China. If so, when did China and India lose this position of alleged parity? And much of South-east Asia was relatively labour scarce and land abundant at the start of the last century. The region absorbed immigrants from the contiguous parts of China and India, and probably

16

J.G.Williamson

generated rapid population growth from its own domestic sources. Did, as a consequence, Malthusian diminishing returns drive down living standards in Southeast Asia, at least relative to the other two wings of Asia? These are not questions that could have been attacked very well even only two decades ago since the data had not been gathered in such a way as to make these comparative judgments possible. Now we have enough data to make some real progress. However, the paper will go light on the detailed analysis and go heavy instead on some new factor price and living standard data (Williamson 1998c, Appendices 1–9), some ‘controlled conjectures’,2 and an Asian agenda which the data suggest. Asia and the world economy: breaking down the tyranny of distance3 In an insightful book entitled The Tyranny of Distance (1966), Geoffrey Blainey showed how distance shaped Australian history. Distance had the same impact on the rest of Asia until late in the nineteenth century, isolating Asia from Europe where, after all, the industrial revolution was unfolding. Late in the nineteenth century, transport innovations started to change all that, although not completely. The appearance of the Suez Canal, cost-reducing innovations on sea-going transport, and railroads penetrating the interior did not completely liberate Asia from the tyranny of distance by 1914. Indeed, economists have shown that growth performance today is still influenced by whether a country is landlocked, by the length of its coastline, and by its distance from Tokyo, New York and Europe (Radelet, Sachs and Lee 1997; Gallup, Sachs and Mellinger 1998). Yet, it was the change in the economic distance to the European core which mattered to late nineteenth-century Asia, even though the levels remained high well in to this century. As we shall see, it is easy to quantify the enormous declines in transport costs and commodity price convergence involving late nineteenth-century Asia. It is far more difficult to find the hard evidence which could be used to reject unambiguously the hypothesis that this commodity price convergence started much earlier. Yet the well documented Southeast Asian Dutch spice trade certainly offers abundant evidence which soundly rejects the hypothesis that commodity price convergence (e.g. trade globalization between the European centre and the Asian periphery) occurred before the mid-late nineteenth century. A recent book by David Bulbeck, Anthony Reid, Lay Cheng Tan and Yiqi Wu (1998) documents the price gap between Southeast Asian ports and Amsterdam for cloves (Malukan and Amsterdam: pp. 58–59), black pepper (Sumatra and Amsterdam: pp. 84 and 103) and coffee (Java plus Sumatra and Amsterdam: p. 175). The ratio of Amsterdam to Southeast Asian clove prices actually rose from 5.6 in the 1580s to double-digit numbers in the seventeenth and eighteenth centuries; it reached a peak of 14.7 in the 1770s; it then plunged to 0.9 in the 1820s and fell to even lower levels by the 1860s and 1870s. The ratio of pepper prices rose from 3 in the 1620s to around 4 in the mid-1700s; it reached a peak of 9.3 in the 1790s; it then dropped back to about 3 by the 1820s, followed by a secular fall to 1.1 in the 1880s. The ratio of coffee prices rose from 2.2 in the 1720s to peaks of 9.5 in the 1790s and 14.7 in the 1800s; it then dropped

Living standards in Asia before 1940

17

to 1.2 by the 1840s, followed by a further secular fall to 0.2 by the 1880s. In short, there is absolutely no evidence of commodity price convergence between the European centre and the Asian periphery from the late sixteenth century to the early nineteenth century, at least based on the Dutch spice and coffee trade. The commodity price convergence we observe in the nineteenth century was the result of what we can rightly call the first great globalization boom. It was a massive regime shift. Transport cost declines from interior to port and from port to Europe ensured that Asian economies became more integrated into world markets. Price gaps between Britain and Asia were driven down by the completion of the Suez Canal, by the switch from sail to steam, and by other productivity advances on long distance sea lanes. The cotton price spread between Liverpool and Bombay fell from 57 per cent in 1873 to 20 per cent in 1913, and the jute price spread between London and Calcutta fell from 35 to 4 per cent (Collins 1996: Table 4). The same events were taking place even farther east, involving Burma and Java. The freight rates on sugar between Java and Amsterdam fell by 50–60 per cent between 1870 and World War I (Yasuba 1978: Graph 2). They fell by about 65 per cent on rice shipments between Burma and Britain (Yasuba 1978: Graph 2). Indeed, the rice price spread between London and Rangoon fell from 93 to 26 per cent in the four decades prior to 1913 (Collins 1996: Table 4). These events had a profound impact on the creation of an Asian market for wheat and rice, and, even more, on the creation of a truly global market for grains (Brandt 1985, 1993; Latham and Neal 1983; Kang and Cha 1996). China and Japan were also involved in these events. The freight rate on coal (relative to its export price) between Nagasaki and Shanghai fell by 76 per cent between 1880 and 1910, and Yasukichi Yasuba (1978: Tables 1 and 5) has estimated that the total factor productivity growth rate on Japan’s tramp freighter routes serving Asia advanced at 2.5 per cent per annum in the thirty years between 1879 and 1909. This commodity price convergence generated an Asian trade boom between 1870 and 1913, just as it did in the Atlantic economy. Export shares in GDP (constant price, Maddison 1995, pp. 190 and 237) almost doubled in India (3 to 5.7 per cent); they more than doubled in Indonesia (1 to 2.2 per cent); and they more than tripled in Thailand (2.1 to 6.7 per cent). But perhaps the greatest nineteenth-century ‘globalization shock’ in Asia did not involve transport revolutions at all. Under the persuasion of American gunboats, Japan switched from virtual autarchy to free trade in 1858. It is hard to imagine a more dramatic switch from closed to open trade policy, even by the standards of the recent Asian miracle. In the fifteen years following 1858, Japan’s foreign trade rose 70 times, from nil to 7 per cent of national income (Huber 1971). The prices of (labour-intensive) exportables soared, rising towards world market levels; the prices of (land and machine-intensive) importables slumped, falling towards world market levels. One researcher estimates that Japan’s terms of trade rose by a factor of 3.5 between 1858 and the early 1870s (Huber 1971); another thinks the rise was even bigger, a factor of 4.9 between 1857 and 1875 (Yasuba 1996, p. 548). Whichever estimate one accepts, this

18

J.G.Williamson

combination of declining transport costs and the dramatic switch to free trade unleashed powerful globalization forces in Japan. Other Asian nations followed this liberal path, most forced to do so by colonial dominance or gunboat diplomacy. Thus, China signed a treaty in 1842 opening her ports to trade and adopting a 5 per cent ad valorem tariff limit. Siam adopted a 3 per cent tariff limit in 1855. Korea emerged from its autarchic ‘Hermit Kingdom’ about the same time, undergoing market integration with Japan long before colonial status became formalized in 1910. India went the way of British free trade in 1846, and Indonesia mimicked Dutch liberalism. In short, by the 1860s commodity price convergence was driven entirely by the sharply declining transport costs in Asia without much change in tariffs one way or the other. Asia’s commitment to globalization started more than a century ago. With Asian globalization forces now on the agenda, let us turn now to the growth of nations in the region. The growth of what? Most economists who have written about the comparative growth of nations have used GDP per capita or per worker to measure catching up and convergence, or falling behind and divergence. This and other essays of mine favour instead real wage rates (purchasing-power-parity adjusted, and typically for urban unskilled workers). I can think of at least four good reasons why it is a mistake for the convergence debate to have ignored wages and other factor prices, especially for the nineteenth century and earlier. First, the pre-1940 real wage data—especially for Asia—are of far better quality than the GDP data, and they are certainly available for a wider sample. Indeed, while Angus Maddison (1995) is able to document real GDP per capita for a surprisingly large part of early Asia, he can still only record observations for the following: Burma, the Philippines, Korea and Taiwan start only with the turn of the twentieth century and offer nothing for the previous one; Thailand starts with 1870, and repeats only every twenty years until 1913; Bangladesh and Pakistan start in 1820 but then leap over eighty years to 1900; China and Japan start in 1820 but then leap over fifty years to 1870; and India and Indonesia start in 1820, leap to 1850 and then report observations only for every twenty years up to 1913. While impressive, such GDP per capita data are usually not enough to deal adequately with the questions raised in the introduction. Real wages can be documented for the following: Burma from 1873; China from 1902; India from 1873; Indonesia from 1820; Japan from 1831; Korea from 1907; the Philippines from 1899; Taiwan from 1897; and Siam4 from 1820. Furthermore, we can begin making statements about PPP-adjusted (purchasing-power-parity adjusted) real wages relative to the European core from each of those dates. In addition, these real wage time series are typically available annually, so that epochs and major turning points can be identified with much greater clarity than is true for the GDP data, which is typically reported for every two decades or even longer. Second, income distribution matters, and wage rates (especially when combined with other factor prices) offer a window through which to look in on distribution

Living standards in Asia before 1940

19

issues. Real people earn wages or skill premiums or profits or rents, not that statistical artefact known as GDP per capita. GDP per worker hour may sound like a good measure of aggregate productivity, but surely the living standards of ordinary workers as captured by real wages are a better indicator of the economic wellbeing of the vast majority in any society. By averaging all incomes, macroeconomists (and economic historians who mimic them) throw away valuable information. Third, factor price movements help us understand the growth of nations. For example, productivity catch-up in a poor country is more likely to increase all factor prices equally than is mass emigration (easing population pressure on the land) or an export boom for agricultural products (increasing the demand for land). The open economy forces which may have been important in driving late nineteenthcentury economic change in Asia—trade, migration and capital flows—operated directly on factor prices, and thus only indirectly on GDP per capita.5 An exclusive focus on GDP per capita misses most of the story. Fourth, economic change nearly always involves winners and losers, a fact which is crucial in accounting for the evolution of policy and the survival of empires, perhaps more so in politically independent societies like China, Japan and Siam than in dependent colonial societies like Indonesia and India. Still, changes that would increase GDP per capita but would also cause losses to some politically powerful group are often successfully resisted even in colonial economies, and examining the behaviour of factor prices is a good way to start the search for the sources of such political resistance. The Asian real wage hierarchy around the turn of the twentieth century Table 1.1 pulls together estimates of the real wage hierarchy around Asia and between it and the European industrial leader, Britain. The assessment is made around the turn of the twentieth century. All of the estimates in Table 1.1 calculate urban unskilled wages from various parts of Asia relative to urban unskilled wages in Britain.6 More importantly, none of the wage relatives in Table 1.1 are calculated at the prevailing exchange rate. It is well known that the use of exchange rates, dominated by tradable goods, is inferior to the use of purchasing-power-parity, the latter constructed from workers’ market baskets. However, trying to construct PPPadjusted real wages based on common market baskets and region-specific relative prices would entail another research project.7 Table 1.1 uses a shortcut: we take Maddison’s 1913 PPP-adjusted GDP per capita estimates as our benchmark and project our real wage series forward and backward from that point. The Asian real wage hierarchy around the turn of the twentieth century is clear enough, and while some of the estimates in Table 1.1 seem to be consistent with other qualitative and quantitative accounts, some are quite surprising. Burma was at the bottom of the Asian hierarchy in 1909–13, a real wage level less than oneeighth that of Britain. In Latin America, only the previously slave-based Brazilian northeast had levels of living that low (Williamson 1999a), and there were no countries in the Mediterranean that had levels of living that low (Williamson 1999b).

20

J.G.Williamson

Table 1.1 The Asian real wage hierarchy near the turn of the century

Sources and notes: The Asian data are taken from Williamson (1998c, Table 1). The British data are taken from my revised Atlantic economy database (Williamson 1995; O’Rourke and Williamson 1997), extended to 1820 by relying on Lindert and Williamson (1983, Table 3; 1985, Table 1). The 1913 benchmark is based on the Maddison (1995) GDP per capita estimates, but the 1909–13 averages exploit our real wage data as well.

China and India were only slightly above the Burmese level, and the figures here, 14.1 and 14.6 per cent of Britain, were about the same as Egypt, 13.8 per cent (Williamson 1999b: Table 1). Given the fragile character of these estimates, however, it would be wiser to say simply that Burma, China and India were all packed close together at the bottom of the Asian hierarchy just prior to World War I. After all, Burma was just 10 per cent below the average for this group of three, and India just 7 per cent above. Furthermore, all three (and others in Asia) exhibited enormous instability in the real wage and living standards since so much depended on the relative price of rice, the key foodstuff throughout most of the region, and the key commodity in the market basket of the urban poor. Thus, these three often changed places in the ranking from one half-decade to another. Thailand, Taiwan and Indonesia were in the middle of the hierarchy, 15.1, 16.4 and 18.3 per cent of Britain. Korea, the Philippines and Japan were the high income regions of Asia, respectively 25.2, 27.2 and 27.4 per cent of Britain. Thus, these three high income Asian regions had living standards that were a little more than a quarter of Britain, levels which appear to have exceeded those of the Italian Mezzogiorno, Portugal, Serbia and Turkey in the Mediterranean (Williamson 1999b), but they still fell behind Latin American regions like Argentina, southeast Brazil, Colombia, Uruguay and even Mexico (Williamson 1999a). These real wage estimates for 1909–13 offer an amazing confirmation of the historical persistence of the wealth of nations. This ranking changed very little in the 80 years between World War I and the end of the recent Asian economic miracle. History and initial conditions matter. What has changed, however, is the size of the north–south gap in Asia. The 1992 ratio between the two at the top, Japan and Korea, and the two at the bottom, China and India, was about 50 (IBRD 1993, pp. 18–19).8 The 1909–13 ratio based on Table 1.1 was less than 2. This was not solely an East Asian economic miracle at work since the economic distance between the middle and the bottom also surged

Living standards in Asia before 1940

21

over this century. The 1992 ratio between the two in the middle, Indonesia and Thailand, and the two at the bottom was 3.6, while the 1909–13 ratio was 1.2. However, at the beginning of the twenty-first century, the great gaps in Asia are primarily due to East Asian development and to events after World War II, not before. The ratios of the same top and middle two to the same bottom two in 1925– 29 were 2 and 1.03. These small gaps were even smaller at the end of the Great Depression 1935–39. Clearly, the twentieth century was one of spectacular divergence in Asia, but most of it has been a post-World War II phenomenon and most of it has been driven by East Asian success. Nonetheless, the big north–south gaps we see in Asia today are based on an economic hierarchy which was already well-established at least a century ago. The most surprising results coming from Table 1.1, however, relate to the evolution of Asian real wages from the 1870s to just before World War I. In spite of impressive industrialization efforts, there is no evidence that living standards in Japan were catching up on Britain, even during a period economic historians label ‘British failure’. Elsewhere in Asia, there was a spectacular collapse in real wages relative to Britain, most of which took place in the two decades between 1873–83 and 1899–1903. In the 1870s, real wages for the unskilled in Bangkok were twofifths those in urban Britain, and in India’s cities they were a third. By the turn of the century, the figures had dropped to something like one-seventh and one-sixth. Just before World War I, India’s position had fallen still further. So, while unskilled urban workers in Japan and the Philippines managed to keep up with Britain’s real wage growth, the rest of Asia fell further behind. Why? What happened during the first great globalization boom? When did the core-periphery gap open up? The first great globalization boom When did these gaps open up? It appears that some of the gaps opened up during the globalization boom after 1870, most exploiting the boom badly, and some exploiting it not at all. Perhaps the comparison with Britain is unfair. Table 1.4 offers alternatives: comparisons with the Netherlands, which makes things a little worse, and comparisons with the average of Britain, France and Germany, which makes things a little better. But using alternative definitions of the European core does not change the nineteenth-century story. Perhaps the comparison with the European core is unfair: trend acceleration is a common attribute of modern economic growth, so late-comers have a hard time catching up with the early industrializers until they have had a chance to gain some momentum. What follows, therefore, will rely on both relatives and absolutes. Table 1.2 documents real wage performance. Tables 1.3 and 1.4 offer comparisons with the European core using various definitions of the latter. Table 1.5 limits the comparison to the growth performance within Asia, each region gauged relative to the ultimate leader of the pack, Japan.

22

J.G.Williamson

Table 1.2 Real wage performance in Asia 1820–1939 (1913=100)

Sources and notes: Williamson (1998c, Table 2). The base year is 1913=100.

Let us begin with the extreme version of the labour surplus model and the nineteenth-century real wage trends in Table 1.2 and Figures 1.2 and 1.3. The extreme version of W.Arthur Lewis’s (1954) labour surplus model predicted a constant real wage, as did the classical model developed by British economists who had not yet appreciated when they were writing how the industrial revolution was making a break with the past. Empiricists using twentieth-century data dealt the labour surplus model severe blows in the 1960s, so severe in fact that it is no longer the dominant paradigm it was 40 years ago. But perhaps the model might do better in pre-industrial nineteenth-century Asia?9 It does not. Prior to 1914, real wages in Asia underwent enormous short-run and long-run variation. They collapsed by 42 per cent in India between the early 1870s and World War I. They more than doubled in Indonesia between the early 1820s and 1910–14. In Siam, they surged from the early 1820s to the early 1880s, then lost all of those gains by World War I. In Japan, real wages showed no long-run trend at all until the 1880s, after which they started their steady rise which persisted for a century. Theory tells us that even in steady state every country can reach different equilibrium living standards (Barro and Sala-i-Matin 1995), but there is no evidence that nineteenth-century Asia was in steady state. Consider now Burma, Indonesia and Thailand, the southeast ‘frontier’ of Asia.

Living standards in Asia before 1940

23

Table 1.3 Real wage performance in Asia relative to Britain 1830–1939 (%)

Sources: Williamson (1998c, Table 3), and see notes to Table 1.1, for Britain.

The available time series for Burma is short, but what we do have suggests a living standards peak 1900–04 to 1915–19. Indonesia and Thailand peaked earlier, in the 1880s. The enormous rise in living standards in these two parts of the southeast Asian frontier are consistent with globalization, rising export prices, and settlement on an extensive margin. The collapse after the 1880s must be explained by other factors, since the positive globalization forces were still at work until primary product prices collapsed after World War I. Presumably, the extensive margin disappeared, and continued immigration into the region began to press downwards the living standard which was still relatively high compared to emigrating areas. These are all simply assertions which will be tested in future research, but any explanation will have to deal with the exceptional success of the Philippines since living standards doubled under US occupation. Korean and Taiwanese experience seems at first sight to be consistent with the nationalist critique that Japanese imperialism eroded workers’ living standards (Kimura 1995; Kang and Cha 1996). Certainly living standards there fell from pre-to post-occupation. But correlation is not necessarily causation. How much of these trends can be attributed to Japanese imperial policy? If the answer is most of it, then exactly how did Japanese policy in Korea and Taiwan differ from US policy in the Philippines, and can the policy differences explain the strikingly different experience with living standards? On the other hand, how much of the decline of

Sources: Williamson (1998c, Table 4). The ‘core’ countries are from my revised Atlantic economy database, described in notes to Table 1.1. The average wage was taken for the period 1909–13, and treated as the wage in 1911.

Table 1.4 Real wage performance in Asia by decades relative to the core 1820s–1930s

Living standards in Asia before 1940

25

Table 1.5 Real wages in Asia relative to Japan 1831–1939 (%)

Sources: Williamson (1998c, Table 5).

living standards in Korea and Taiwan might be explained by the collapse in primary product prices in world markets facing these relatively small economies, which were heavily dependent on trade? If so, why were living standards not sagging elsewhere in Asia from the turn of the century to the late 1920s, especially in the Philippines? We add these questions to the growing agenda, but some of them will be confronted again when we look at wage–rental ratio trends. China’s time series is too short and too weak in quality to say much at all, except that unskilled urban real wages in the major cities changed hardly at all between 1910–14 and 1935–39. We conclude with a comment about India’s experience. Real wages there collapsed sharply between the late 1880s and World War I, and Table 1.6 shows that it was happening everywhere on the subcontinent. India seemed to share the same dismal post-1880s experience that was true of Southeast Asia. Why? The next section will ask whether there is any evidence that India (and China) underwent a fall in living standards from the mid-eighteenth century to World War I, and whether the post-1880s collapse is simply part of that longer trend. If so, it would be very different from the Southeast Asian experience. Table 1.3 reports how Asian living standard growth measured up to that of Britain, while Table 1.4 expands the comparison to include other parts of the European core. The stories are similar, so I will focus on Britain and Table 1.3. Japan’s real wage was 33 per cent of Britain’s in the early 1830s, and it was 32 per cent of

26

J.G.Williamson

Figure 1.2 Real wage indices for East Asia (1913=100).

Figure 1.3 Real wage indices for Southeast/South Asia (1913=100).

Living standards in Asia before 1940

27

Table 1.6 Regional real wage indices for India 1873–1939 (All-India 1900=100)

Source and notes: Williamson (1998c, Table 8). All-India 1900=100. The four regions are taken as a percentage of All-India.

Britain’s in the late 1920s. There was no catching up here, but at least Japan was able to hold her own, first by switching in the 1850s from autarchy to an open trade policy (with some help from American gunboats), and second by mounting a very successful Meiji industrialization program after the 1880s. While the former was not effective (real wage gains up to the late 1850s and early 1860s disappeared by the 1870s), the latter certainly was. The other two success stories in nineteenth-century Asia were brief: during their real wage surge from the 1820s to the 1880s, Indonesia and Thailand were actually catching up on Britain, which was impressive since Britain was undergoing unusually fast real wage growth during that period.

28

J.G.Williamson

With the exception of the Philippines, the rest of Asia was falling behind the European core. The core-periphery gap was already wide by the end of the first third of the nineteenth century, but it became even wider during the rest of the nineteenth century, especially after the 1880s. When did the core-periphery gap open up? Looking at the more distant past Conjectures on the origins of the core-periphery gap Later in this section we will review the recent revisionist arguments that living standards in China and India were probably on a par with those in Britain around the mid-eighteenth century, that is, immediately prior to the industrial revolution. But assume for the moment that they were on a par around 1700 or 1750. Tables 1.1 and 1.3 suggest that by the 1870s Indian living standards had fallen from the assumed parity to a third of Britain. By 1910–15, the figure had fallen to 15 per cent of Britain, where the Anglo-Indian gap had reached its maximum. Over two centuries, living standards in India relative to Britain dropped from 100 to 15 per cent, a fall of 85 percentage points. That fall does not have its source in the interwar period, since the real wage in India (relative to Britain) was slightly higher in the 1930s than around World War I. About 18 percentage points of that fall took place during the first great globalization boom after the 1870s. It appears that Indian workers did not benefit as much from globalization as British workers did from it and from the second industrial revolution. About 67 percentage points were lost between the mid-eighteenth century and the 1870s. One supposes this measures the disadvantage of failing to industrialize. Now, can we confirm the assumption of parity in the eighteenth century? And does the story apply to China as well? Were Indian living standards on a par with Britain in the eighteenth century? Prasannan Parthasarathi (1998) has recently published a paper which argues that living standards in South India were at least equal to those of Britain in the eighteenth century. According to Table 1.6, South India was roughly on a par with all-India in the 1870s and 1880s, so what was true of Madras may well have been true of some India average, at least in the late nineteenth century. What kind of evidence does Parthasarathi marshal? Much of it is quantitative, and all of it is careful in trying to make the comparison as comparable as possible in terms of the workers, time worked, and the goods and services consumed. This is not the place to survey in detail all the evidence that Parthasarathi brings to bear on the issue. Some highlights will have to serve to state the case. The evidence for India is mainly based on spinners and weavers in the South, and on the grain content (rice) of their wages, compared with the grain content (wheat) of rural and small-town spinners and weavers in Britain. So measured, weekly wages for men and women in textiles were roughly comparable in the

Living standards in Asia before 1940

Figure 1.4 Real wages in East Asia relative to Britain (%).

Figure 1.5 Real wages in Southeast/South Asia relative to Britain (%).

29

30

J.G.Williamson

eighteenth century. Weekly wages in agriculture were also comparable. Furthermore, agricultural productivity was about the same, or perhaps even higher in India.10 According to this evidence, eighteenth-century workers’ living standards in India could not have been very different from those in England. Yet the evidence does not deal with non-grain consumption, it is not confirmed with mortality data, nor is it confirmed by travellers’ accounts or official reports on the quality of homes and furnishings. And are we really sure that the South was like India as a whole in the eighteenth century? These issues may matter, as we shall see in a moment. Were Chinese living standards on a par with Britain in the eighteenth century? Kenneth Pomeranz (1997) has recently published a paper which argues that eighteenth-century living standards on the lower Yangzi and Lingnan—China’s two most advanced regions—were at least as high as they were in Britain and the Low countries—Europe’s two most advanced regions. Parthasarathi and Pomeranz seem to be unaware of each other’s work, but they are clearly saying the same thing. As with Parthasarathi for India, Pomeranz tries to make the comparison as comparable as possible. The evidence for China takes the following form: compared to England, equally high calorific intake based on landless agricultural labourers’ seventeenth-century diets and eighteenth-century per capita grain consumption countrywide; even higher eighteenth-century rural Chinese life expectancies than Europe; the same share of expenditures on basic foodstuffs by farm labourers in seventeenth- and early nineteenth-century China as by the rural poor in England in the 1790s; late eighteenth-century European travellers noted that the average Chinese smoked more than did the average English, and that China’s successful farmers in the interior had the same quality furnishings and house construction as did provincial English farmers; per capita tea, sugar and silk consumption were higher in China about 1800; and furniture inventories in early twentieth-century Chinese homes were about the same as that in eighteenth-century Netherlands. According to this evidence, eighteenth-century living standards in Yangzi and Lingnan could not have been very different from those in England and the Low countries. But note that the timing in these comparisons is often a bit vague. Are we talking about the seventeenth-, eighteenth- or early nineteenth-century China? In a moment, we will argue that it may matter. Were Thai living standards on a par with Britain in the nineteenth century? If the case for eighteenth-century parity looks plausible for ‘labour-surplus’ India in South Asia and China in East Asia, surely it ought to be even more plausible for labour-scarce Southeast Asia. Some time ago, James Ingram (1964) offered some rough evidence that suggested it might be so even as late as the 1880s. In 1882, the

Living standards in Asia before 1940

31

average weekly wage for weavers in Lancaster was 15 shillings and wheat cost 45 shillings per quarter (Chapman 1904, p. 67). Ingram (1964, p. 114) concludes that the wheat equivalent of a week’s wage in Lancaster must have been 160 lb. When Ingram makes the same calculation for unskilled rice-mill workers in Bangkok, he concludes that a week’s wage bought 200 lb of rice in the 1880s. This may be weak evidence upon which to build a solid case for living standard parity between Britain and Thailand, but it is certainly suggestive and well worth augmenting. Has the case been overstated? Pre-industrial sources of core-periphery gaps A recent paper by Robert Allen (1998) suggests that the arguments by Pomeranz and Parthasarathi are probably overstated.11 Allen has exploited wage and price data which scholars first started collecting for Strasbourg and southern England. In 1929, the International Scientific Committee commissioned similar studies covering all of England, Germany, France, Austria, Poland and Spain (Cole and Crandall 1964). Belgium and Italy were added later. Allen was able to find sufficient material to reconstruct real wages over the four centuries 1500–1913 for London and fifteen other European cities, five of which were in the Mediterranean and many of the remainder in central and eastern Europe. Like our Asian data, the wages are for urban labourers and craftsmen, and the rates of pay are typically daily. Allen then uses the prices and fixed market basket weights to construct true PPP-adjusted wages across these cities. This is an extraordinarily high quality database, and Allen uses it to say something concrete about living standard convergence and divergence over the very long run in Europe. That is, it speaks to the question: when did the core-periphery gap appear in Europe? When did an economic gap emerge between the rich northwest of Europe—led by England and the Lowlands—and what eventually became the poor south, central and eastern parts of Europe? The central message of both Allen’s (1998) paper and my own on the Mediterranean (Williamson 1999b) is that the core-periphery gap in Europe was not the product of the late nineteenth-century globalization surge, although in Asia a small part of it was. Instead, the core-periphery gap was the product of two events: seventeenth- and eighteenth-century pre-industrial economic success in the European core, while the periphery stagnated; and industrial revolutionary gains in the European core up to the mid-nineteenth century which the periphery was slow to copy. In 1500, the cities in Europe were pretty much on par; by 1850, living standards around the periphery were only 30–40 per cent of those in London. Starting at parity early in the sixteenth century, Florence and Milan were only about 59 per cent of London early in the eighteenth century, while Valencia was only 54 per cent; by World War I, the figure for Florence and Milan was 36 per cent, while it was 48 per cent in Madrid. In short, a good share of the core-periphery gap in Europe was more than simply the product of some uneven timing of industrial revolutions or of some inability to exploit the first great globalization boom. Rather, pre-industrial events between

32

J.G.Williamson

1500 and 1800 mattered just as much. Pomeranz’s (1997) work on China and Parthasarathi’s (1998) on India suggests that Asia was different, and that all of the core-periphery gap between Europe and Asia was the product of the timing and location of industrial revolutions, and, we now see, the inability of Asia to exploit the globalization boom very well after the mid-nineteenth century. Can the experience with the emergence of a core-periphery gap within Europe be so different from the experience with the emergence of a core-periphery gap between Europe and Asia? Are we sure that the gap between the European core and both China and India was due solely to the presence of an industrial revolution in the former and its absence in the latter? If so, we have to explain why about half of the core-periphery gap within Europe emerged between 1500 and 1800, well before any industrial revolution took hold in the European core, and why that did not happen in Asia. The Hindu equilibrium, classical steady states and constant pre-industrial living standards In the 1950s, W.Arthur Lewis (1954) got us thinking again about labour surplus and stable real wages. I say ‘again’ because his brilliant model had its ancestral roots in classical thinking by pessimistic British economists who may have been writing during the first industrial revolution, but who failed to see it as a permanent change in economic possibilities (Williamson 1985). Thus, while the classical British economist was obsessed with pre-industrial steady states (e.g. constant real wages and living standards), Lewis used it to explore real wage turning points under modern industrialization. As mentioned already, the empirical evidence which accumulated in the 1960s and 1970s was not very kind to the labour-surplus model, but it might still speak to the centuries of pre-industrial living standard experience in Asia. It is still too early to give firm answers to this central question, but some good data has been lying in the archives for some time, waiting to speak to it. For example, Robert Allen and I hope to resolve it by comparing econometrically Abul Fazl’s (1871) household data from about 1595, in what is now the Punjab, Uttar Pradesh and Haryana (Akbar’s empire), with similar evidence from the nineteenth and twentieth century. This project will also make it possible to say much more about what Deepak Lal (1988) asserts was a Hindu equilibrium in pre-independence India. For labour-scarce and resource-abundant Southeast Asia, Ingram (1964, Table III, p. 115) offers evidence that there was absolutely no change whatsoever in the daily wage in rice for Bangkok’s unskilled coolie labour between 1688 and 1889, two centuries of long-run real wage stability! If more and better evidence can confirm constant real wages in pre-industrial Asia—but at different levels in different places as modern growth theory predicts (Barro and Sala-i-Matin 1995)—then the next step is to explore the impact of exogenous mortality-induced demographic shocks (Lavely and Wong 1998) and exogenous globalization-induced price shocks (see above) on ‘transitory’ departures from these long-run pre-industrial steady states. We have already seen plenty of

Living standards in Asia before 1940

33

evidence of those ‘transitory’ departures between 1870 and 1940. They need an explanation (e.g. how much is due to favourable globalization price shocks) and we need to know more about the process whereby such transitory gains were erased and low-level living standard equilibria were reattained. Hints and hunches about inequality trends in Asia Eli Heckscher and Bertil Ohlin argued that the integration of global commodity markets would lead to convergence of international factor prices, as countries everywhere expanded the production and export of commodities which used their abundant (and cheap) factor intensively. As Table 1.8 confirms, limited historical evidence from Asia seems to be consistent with Heckscher and Ohlin: the trade boom between the 1870s and the 1920s led to rising wage–rental ratios in relatively labour-abundant East Asia, and to falling wage–rental ratios in the relatively landabundant Punjab, and probably in Southeast Asia where we do not have wagerental ratio data but where real wages were stable or falling. As a consequence, conditions probably improved for the poor unskilled worker relative to the rich landlord in East Asia, while the opposite was probably true of Indonesia, Siam and Burma. All of this borders on speculation, of course, since, so far, it is guided only by limited information on wage-rental ratio trends. So much for globalization. What about labour surplus? In his famous model of the labour surplus economy, W.Arthur Lewis (1954) showed how early industrialization could create inequality (and also a rising surplus to finance domestic-savings-constrained accumulation). Stable real wages implied rising profit shares economy-wide. According to his model, the worker fails to share in GDP per capita growth since elastic labour supplies keep wages and living standards stable. The labour surplus model could also be used to predict stable real wages in Southeast Asia, since the migration of surplus labour from India and China might have served to create an elastic labour supply in, for example, Burma, Siam and Indonesia.12 The Lewis model is quiet about what happens to land rents, but the classical model from which it was derived clearly predicted a rise in rents. It follows that the globalization and the Lewis model both predict falling wage–rental ratios and rising inequality in Southeast Asia, far more so than in labour-abundant East and South Asia. Yet, discriminating empirically between the competing Lewis and Heckscher–Ohlin views will prove difficult. Complete income distributions at various benchmarks between the mid-nineteenth century and World War II are unavailable for any Asian country, including Japan. But even if such data were available, it is not obvious that they would be the best way to search for the underlying causes of changing inequality. Our interest here is factor prices: wages, rents and the structure of pay. How did the typical unskilled worker near the bottom of the distribution fare relative to the typical landowner or capitalist near the top, or even relative to the typical skilled blue-collar worker or educated white-collar employee near the middle? The modern debate over OECD (Organization for Economic Cooperation and Development) inequality has a fixation on wages, but since land and landed interests were far more

34

J.G.Williamson

Table 1.7 Wage/GDP per capita ratio trends 1870–1939 (1913=1.00)

Sources and notes: Williamson (1998c, Table 7). The real GDP per capita data are taken from Maddison (1995). The base year is 1913=1.00. Table 1.8 Asian wage/rental ratio trends 1873–1939

Sources and notes: Williamson (1998c, Table 6). The base year is 1913=1.0.

important to late nineteenth-century inequality trends—especially in more agrarian Asia—we need to add them to any distribution inquiry. In any case, we have two kinds of evidence available to document inequality trends in Asia prior to 1940: trends in the wage–rental ratio, which we have already explored, and which are limited to Japan, Korea, the Punjab and Taiwan; and trends in the ratio of the unskilled wage to GDP per capita, which we have not yet explored, and are available for the full Asian sample between 1870 and 1940. Table 1.7 reports trends in the ratio of the unskilled worker’s wage (w) to the returns on all factors per person as measured by Angus Maddison’s (1995) estimates of GDP per capita (y). True, the ratio could be influenced by changes in the labour participation rate alone. If there was a sharp increase in population from, say, a rise in fertility and thus no increase in workers of adult age, w/y would (spuriously) rise

Living standards in Asia before 1940

35

Figure 1.6 Trends in wage/GDP per capita ratio—Japan 1870–1938.

as y fell. In contrast, if there was a sharp increase in population from the immigration of adult labour, w/y would seem to be more stable. While this was not a period of dramatic demographic transition in Asia (Bloom and Williamson 1998), the immigration into Southeast Asia probably tends to make w/y trends overstate rising inequality there. True, some of the observed changes in w/y could also be driven by the performance of the price of wage goods (in the cost-of-living index underlying the real wage, dominated by the price of rice and other grains) relative to the GDP deflator (underlying Maddison’s real GDP per capita estimates). But such relative price movements have clear distributional implications on the expenditure side, since the poor are more dependent on rice and other foodstuffs (as a share of their budgets) than are the rich. So, apart from the demographic issues, trends in w/y should approximate changes in the economic distance between the working poor near the bottom of the distribution and the average citizen in the middle of the distribution.13 Table 1.7 shows that any successful explanation of changes in w/y in Asia between 1870 and 1940 will have to be complex: the Heckscher–Ohlin trade model (Flam and Flanders 1991) and the Lewis labour surplus model (Lewis 1954) will not, by themselves, account for the variety. Japan, India, Indonesia and Thailand document the longest time series, and Figures 1.6 to 1.9 show that they share the same trends. They all underwent a long sharp decline in w/y before flattening out or even rising after World War I. The turning point for Indonesia seems to be the late 1920s, but for the other three it is

36

J.G.Williamson

Figure 1.7 Trends in wage/GDP per capita ratio—India 1873–1939.

Figure 1.8 Trends in wage/GDP per capita ratio—Indonesia 1870–1940.

Living standards in Asia before 1940

37

Figure 1.9 Trends in wage/GDP per capita ratio—Thailand 1870–1939.

1915–19. (The Burmese w/y time series obeys the same Asian laws of motion, even though its time series is much shorter.) Why did the real wage lag behind GDP per capita in these four Asian countries during the first great globalization boom? Is this evidence of some weaker version of the Lewis model, no constant wage but rather sluggish growth and modest trickling down? If so, why the common turning point for four economies with such different attributes? Since it seems unlikely that such dissimilar economies could share the same Lewis turning point, perhaps a more likely explanation lies with world markets. These four countries were more likely to have shared similar price shocks which produced the same trends in w/y. The Philippine time series is much shorter, but what we have is consistent with the trends seen in Figures 1.6 to 1.9 for the other four (and for Burma). The Philippine turning point is 1910–14, and it shares the steep decline up to that point and the sharp reversal thereafter. The experience of Taiwan and Korea is similar, with their turning points in 1915–19, although the Korean time series is a little short for us to be completely confident about long-run turning points. As an aside, this evidence does not offer much support for the nationalist critique of Japanese imperialist policies. If they really did tend to exploit the ordinary workers in occupied Korea and Taiwan, while favouring Japanese landlords in those two regions and Japanese consumers at home, why do we not see that redistribution in the form of falling w/y after the 1910–19 decade? Why do they rise instead? The answer may lie with world markets rather than imperial policy. We have found an important Asian stylized fact. Real wages lagged behind

38

J.G.Williamson

GDP per capita growth everywhere in Asia up to the World War I decade (with the exception of China). Real wages outstripped GDP per capita growth thereafter. We interpret these trends as a proxy for rising inequality during the first great globalization boom and falling inequality during the interwar years. What accounts for this stylized fact?14 This chapter will avoid this question, but it is added to that lengthening agenda. Within versus between regional performance in Asia: India and Indonesia 1870s–1930s In the 1950s and 1960s, a literature emerged on regional inequality within countries, or what became known as North-South dualism. Some argued that the natural forces were towards divergence (Myrdal 1957; Hirschman 1958), while conventional theory argued for convergence. The evidence seemed to support a life cycle of regional inequality for industrializing countries, first rising and then falling (Williamson 1965). Four classic examples of regional dualism often cited were the Italian Mezzogiorno, the American South, the Brazilian Northeast and the east wing of Pakistan (now called Bangladesh, and what in Table 1.6 is labelled ‘East’). The literature on regional inequality and North-South dualism died down in the 1970s, only to awake again over the last decade in two different guises. First, Paul Krugman created the new economic geography which gave elegant theoretical argument to the rise of regional dualism in an industrializing country (Krugman 1991, 1998; Krugman and Venables 1990, 1995). Second, empirical studies using the new growth theory showed that there were forces of convergence embodied in the regional growth experience of Japan and the United States (Barro and Sala-iMatin 1991). What do we find in Asia? While the pre-1940 period is hardly one of vigorous industrialization in Asia, with the exception of Japan, it was one of globalization as well as regional integration through railroad development and internal migration from labour-abundant to labour-scarce areas. What influence did these forces have on regional ‘dualism’? William Collins (1998), using the same 1873–1906 data underlying the longer time series in Table 1.6, but at the district level, has shown recently that late nineteenth-century India was not characterized by unconditional or even conditional convergence. Despite the period’s transport revolution with the advent of the railroads, large scale labour mobility did not take place, and globalization forces—favouring some regions at the expense of others—weakly offset any regional convergence trends. Table 1.6 seems to confirm Collins’ findings, and for a longer period. There is no evidence of unconditional convergence between India’s four major regions between 1873 and 1939. On the contrary, real wages in the East started 30 per cent below the all-India average and ended up 36 per cent below; real wages in the West started 17 per cent above average and ended up 44 per cent above. The South collapsed to 55 per cent below average. These are clear forces of divergence, and we need to know whether they confirm Krugman–Venables industrial concentration effects,

Living standards in Asia before 1940

39

Table 1.9 Regional real wage indices for Indonesia 1878–1939 (Java 1912=100)

Source and notes: Williamson (1998c, Table 9). Java 1912=100. The Outer Provinces are taken as a percentage of Java.

40

J.G.Williamson

or whether they reflect uneven world price shocks associated with globalization, favouring some regions at the expense of others, or whether it is other forces entirely. The Indonesian experience is consistent with that of India (Table 1.9). There is no evidence of real wage convergence between the high-wage outer provinces and low-wage Java between 1878 and 1939. On the contrary, the wage gaps were even bigger in the 1930s than in the 1870s. This problem has persisted into the 1990s (Saldanha 1997), and any explanation of the present should also explain the past. An agenda for comparative Asian economic history There is so much left to be done. Other scholars may have their own agenda, but what follows is based on the real wage and relative factor price data base recently constructed and used to carry the discussion in the text. The database can, of course, be improved. First, I have been able so far to secure time series information on farm rents or land values only for the Punjab, Japan, Korea and Taiwan, and the latter two only cover this century. We need this information to help discriminate between explanations of growth in more labour-scarce Southeast Asia and the more labour-abundant South Asia. The search continues, therefore, for this kind of information for Indonesia, the Philippines and Thailand, as well as for other parts of India, although I have recently found it for Burma. Second, four of our regions have various real wage problems. While I have now been able to push Burmese wage and price observations up to 1940, I have made little progress for China. The Chinese real wage data even for big coastal cities like Shanghai are very poor: they cannot be trusted prior to about 1908 or after 1936. Surely, we can do better. Real wages for Japan prior to the 1880s are fraught with problems, so much so that inferences about the switch from total autarchy to free trade in 1858 are very different if one uses Huber’s (1971) data or those reported here. The issue is much too important to be left in this ambiguous empirical state. And real wages for the Philippines in the decade prior to 1910 must be improved and also extended backwards before 1899. The exceptional real wage experience of this Southeast Asian country is sufficiently distinctive to warrant much more archival attention. Third, I have been unable to find anything useful for either Ceylon or Vietnam. Ceylon is a special disappointment since the impact of its absorption of heavy immigration from the subcontinent is of real interest, especially compared with Burma or Thailand. Still, the database we do have is much more than I thought was possible when I started this project four years ago, and it has not been used at all by scholars interested in long-run Asian growth. Consider now what the real wage and relative factor price data tell us. To begin with, there is absolutely no evidence to support the view that Asia was in some steady state, let alone that there was some Lewis-like constant real wage that characterized any part of Asia. There was enormous variance over time and across regions in Asia between 1820 and 1940, so that economists do not have to look only to the 1990s to find it. The challenge is to offer explanations for this evidence

Living standards in Asia before 1940

41

and then to use it to help guide future policy debate for the next century. This paper points to the following agenda: 1 What were the size of the price shocks which various Asian regions absorbed as the globalizing effects of the Suez Canal, the Panama Canal, railroads to the interior and productivity advance on ocean transport took place? Were they bigger than those which characterized the Atlantic economy after 1870? After all, the political reaction almost everywhere in the Atlantic economy was to mute the impact of those globalizing forces with tariffs (O’Rourke and Williamson 1999, ch. 6). The opposite was true of Asia. 2 By 1940, there were already very big real wage and living standard gaps between rich and poor regions in Asia. Furthermore, there was hardly any change in the economic hierarchy in Asia between the 1890s and the 1930s. History mattered, and there was strong persistence. Why? 3 By 1940, there was an enormous living standard gap between Asia and Europe. When did it emerge? None of it emerged during the interwar years. Some of it emerged during the first great globalization boom after 1870, pointing to the inability of Asia to exploit globalization prior to World War I, an interesting finding given the huge literature that points out how East Asia exploited globalization so well after World War II! Why did Asia exploit globalization so badly between 1870 and 1914? Much of the gap appeared over the century 1770–1870 when Asia failed to mimic the European industrial revolutions, but what about the impact of pre-industrial progress in Northwest Europe between 1500–1770? 4 What would happen if we applied new growth convergence analysis to this real wage database? How much of the variance in real wage growth might be explained by initial living standards, globalization-induced price shocks, demographic events, geographic disadvantages and so on? 5 Wage-rental and wage-GDP per capita ratios trace out a distinctive Asian stylized fact: they fall to about the World War I decade, then rise thereafter. Why? It should be easy to initiate an econometric attack on this problem much like that which has been done already for the Atlantic economy (O’Rourke, Taylor and Williamson 1996; Williamson 1997), where price shocks, factor endowments and factor-saving productivity advance are all allowed to have their say. 6 Why the spectacular economic demise of Southeast Asia after the 1870s or 1880s? Does this fact offer more evidence supporting the ‘curse of resources’ (Sachs and Warner 1995) and diminishing returns to land? Indeed, why do real wages collapse almost everywhere in late nineteenth-century Asia? 7 Imperialism seems to have played a minor role in Asia, even in interwar Korea and Taiwan. Will this view be supported when explanations of real wage performance are ‘conditioned’ by other variables which must have mattered? 8 What was the spatial impact of globalization on regions within British India, within Indonesia and within the Philippines? Many Asian countries have serious spatial or regional inequality problems today that were bequeathed to the present by the past. What accounts for that past?

42

J.G.Williamson

These are just some of issues which this database will let us pursue. There is no longer any reason for historians to fail to take up the challenge that economists have thrown down by exploring the comparative performance of Asia since 1960 or 1970. Economists will continue to use only that narrow empirical window to advise Asian policy debate in the coming years as long as historians fail to supply a wider empirical window that lets us look at two centuries of economic performance rather than simply the recent quarter-century. Endnote I would like to thank Greg Clark, Bill Collins, Pierre van der Eng, Dave Feeny, Ben Legarda, Peter Lindert, Susan Woolcott and Yasukichi Yasuba for their help in guiding me to the data underlying the analysis in this paper. In addition to those scholars, conversations on the topic with Bob Allen, Kevin O’Rourke and Ron Findlay have also been very useful. It is a pleasure to acknowledge the excellent research assistance of Chenghuan Chu, Ximena Clark, Kai Lee, Anupam Mishra and Matt Weinzierl, all of whom have been involved at various points since 1995 when this project started. One of the goals of the project has always been to make the database available and it can be found in the appendices to HIER Discussion Paper 1844, ‘Real wages and relative factor prices in the Third World 1820–1940: Asia’ (Harvard Institute of Economic Research, Department of Economics, Harvard University (August 1998)) which is cited in this essay as Williamson (1998c). In addition, those data (with sources and methods) are reported at my website http:// www.economics.harvard.edu/~jwilliam. However, the Asia database is being repaired and augmented (spring/summer 2000), so users may prefer to make the request for a diskette from me at [email protected]. The Asia HIER working paper has two companions: ‘Real wages and relative factor prices in the Third World 1820–1940: The Mediterranean Basin’ (HIER Discussion Paper 1842: July 1998) and ‘Real wages and relative factor prices in the Third World 1820– 1940: Latin America’ (HIER Discussion Paper 1853: November 1998) which are also available upon request. The data from Latin America and the Mediterranean are in those papers and are reported at my website, but, since they have also undergone repairs and extensions, I will make them available on diskette too. Notes 1 Robert Allen (1998) argues that the divergence within Europe started much earlier. This evidence will be discussed below. 2 I borrow this term from a paper written by Paul David (1967) more than thirty years ago on United States growth performance prior to the 1840s. 3 This section draws heavily on O’Rourke and Williamson (1999, ch. 3). 4 Throughout this paper, the labels Siam and Thailand are used interchangeably. 5 For a summary, see Williamson (1996), and O’Rourke and Williamson (1999). 6 True, one of the Asia regions, Indonesia, could not yield truly urban wages. But even in this case, all we need assume is that the wage gap between city and countryside was

Living standards in Asia before 1940

7 8 9 10 11 12 13 14

43

relatively stable over time to assert that the Indonesian time series is a perfectly adequate proxy for changes in the real wages of urban workers. This is, in fact, included in a research proposal of mine pending at the National Science Foundation. All the calculations which follow in this paragraph use unweighted averages. The odds are not good even here, given that the classical model did not even explain the years between 1780 and 1820, the period for which it was originally constructed (Williamson 1985). Of course, theory predicts that yields would have been higher in India if labour was cheaper and thus used more intensively. However, Parthasarathi appears to be talking about labour productivity in agriculture. See also Williamson (1999b), where Allen’s argument is applied to the Mediterranean. As long as there is no comparable elastic supply of land on these frontiers. For this endogenous-land argument, see Myint (1958) and Findlay (1995, ch. 5). Hla Myint’s theoretical work was motivated by Burmese experience. It turns out that this statistic is highly correlated with more comprehensive inequality measures in the few cases where both are available in the Atlantic economy. See Williamson (1998b, Table 5). Presumably, the explanation emerges by applying the same econometric analysis that we propose to implement for the wage-rental ratio trends.

References Allen, R. (1998) ‘The Great Divergence: Wages and prices from the Middle Ages to the First World War’, paper presented to the Conference on Regions, Queens University, Kingston, Canada (March 6–7). Barro, R.J. and X.Sala-i-Matin (1991) ‘Convergence across states and regions’, Brookings Papers on Economic Activity 1, pp. 107–82. —(1995) Economic Growth, New York: McGraw-Hill. Blainey, G. (1966) The Tyranny of Distance: How Distance Shaped Australia’s History, Melbourne: Macmillan, revised 1982 ed. Bloom, D. and J.G.Williamson (1998) ‘Demographic transitions and economic miracles in emerging Asia’, World Bank Economic Review 12, pp. 419–55. Brandt, L. (1985) ‘Chinese agriculture and the international economy 1870–1913: A reassessment’, Explorations in Economic History 22, pp. 168–80. —(1993) ‘Interwar Japanese agriculture: Revisionist views on the impact of the colonial rice policy and labor-surplus hypothesis’, Explorations in Economic History 30, pp. 259–93. Bulbeck, D., A.Reid, L.C.Tan and Y.Yu (1998) Southeast Asian Exports Since the 14th Century: Cloves, Pepper, Coffee, and Sugar, Leiden, The Netherlands: KITLV Press. Chapman, S.J. (1904) The Lancashire Cotton Industry, Manchester. Cole, A. and R.Crandall (1964) ‘The International Scientific Committee on price history’, Journal of Economic History, 24, pp. 381–88. Collins, W.J. (1996) ‘Regional labor markets in British India’, mimeo., Department of Economics, Harvard University (November). —(1998) ‘Labor mobility, globalization, and wage convergence in late 19th Century India’, mimeo., Department of Economics, Harvard University (February 28). David, P.A. (1967) ‘The growth of real product in the United States before 1840: New evidence, controlled conjectures’, Journal of Economic History 27, pp. 151–98. Fazl, A. (1871) Ain-e-Akbari, tr. H.Blochmann, Calcutta. Findlay, R.F. (1995) Factor Proportions, Trade, and Growth, Cambridge, Mass.: MIT.

44

J.G.Williamson

Flam, H. and M.J.Flanders (1991) Heckscher-Ohlin Trade Theory, Cambridge, Mass.: MIT Press. Gallup, J.L., J.D.Sachs and A.D.Mellinger (1998) ‘Geography and Economic Development’, Annual World Bank Conference on Development Economics 1998, Washington, D.C.: The World Bank, pp. 127–70. Gerschenkron, A. (1952) ‘Economic backwardness in historical perspective’, in B.Hoselitz (ed.), The Progress of Underdeveloped Areas, Chicago: University of Chicago Press. Hirschman, A.O. (1958) The Strategy of Economic Development, New Haven, CT: Yale University Press. Huber, J.R. (1971) ‘Effect on prices of Japan’s entry into world commerce after 1858’, Journal of Political Economy 79, pp. 614–28. IBRD (1993) The World Bank Atlas 1994, Washington, D.C.: The World Bank. Ingram, J.C. (1964) ‘Thailand’s rice trade and the allocation of resources’, in C.D.Cowan (ed.), The Economic Development of South-East Asia, London: Allen and Unwin. Kang, K.H. and M.S.Cha (1996) ‘Imperial policy or world price shocks? Explaining interwar Korean living standards’, paper presented to the Conference on East and Southeast Asian Economic Change in the Long Run, Honolulu (April 11). Kimura, M. (1995) ‘Standards of living in colonial Korea: Did the masses become worse off or better off under Japanese rule?’, Journal of Economic History 53, pp. 629–52. Krugman, P.R. (1991) Geography and Trade, Cambridge, Mass.: MIT Press. —(1998) ‘The role of geography in development’, Annual World Bank Conference on Development Economics 1998, Washington, D.C.: The World Bank, pp. 89–107. Krugman, P.R. and A.Venables (1990) ‘Integration and the competitiveness of peripheral industry’, in Bliss, C. and J.Braga de Macedo (eds), Unity with Diversity in the European Community, Cambridge: Cambridge University Press, pp. 56–77. —(1995) ‘Globalization and the Inequality of Nations’, NBER Working Paper 5098, National Bureau of Economic Research, Cambridge, Mass. (April). Lal, D. (1988) Cultural Stability and Economic Stagnation: India 1500BC-AD1980, Oxford: Oxford University Press. Latham, A.J.H. and L.Neal (1983) ‘The international market in rice and wheat 1868– 1914’, Economic History Review 36, pp. 260–75. Lavely, W. and R.B.Wong (1998) ‘Revising the Malthusian narrative: The comparative study of population dynamics in late Imperial China’, Journal of Asian Studies 57, pp. 714–48. Lewis, W.A. (1954) ‘Economic development with unlimited supplies of labour’, Manchester School of Economic and Social Studies 22, pp. 139–91. —(1969) Aspects of Tropical Development, Uppsala: Wiksell. —(1978a) The Evolution of the International Economic Order, Princeton, N.J.: Princeton University Press. —(1978b) Growth and Fluctuations 1870–1913, Cambridge: Allen and Unwin. Lindert, P.H. and J.G.Williamson (1983) ‘English workers’ living standards during the Industrial Revolution: A new look’, Economic History Review, Second Series, 36, pp. 1–25. —(1985) ‘English workers’ real wages: Reply to Crafts’, Journal of Economic History 45, pp. 145–53. Maddison, A. (1991) Dynamic Forces in Capitalist Development, Oxford: Oxford University Press. —(1995) Monitoring the World Economy 1820–1992, Paris: OECD Development Centre Studies. Myint, H. (1958) ‘The classical theory of international trade and the underdeveloped countries’, Economic Journal 68, pp. 317–37. Myrdal, G. (1957) Economic Theory and Underdeveloped Regions, London: Duckworth.

Living standards in Asia before 1940

45

O’Rourke, K.H. and J.G.Williamson (1997) ‘Around the European periphery 1870–1913: Globalization, schooling and growth’, European Review of Economic History 1, pp. 153–90. —(1999) Globalization and History, Cambridge, Mass.: MIT Press. O’Rourke, K.H., A.M.Taylor and J.G.Williamson (1996) ‘Factor price convergence in the late 19th century’, International Economic Review 37, pp. 499–530. Parthasarathi, P. (1998) ‘Rethinking wages and competitiveness in the eighteenth century: Britain and South India’, Past and Present 158, pp. 79–109. Pomeranz, K. (1997) ‘Re-thinking 18th century China: A high standard of living and its implications’, paper presented to the All-UC Group in Economic History Conference on Rethinking the History of Wages, Prices and Living Standards, Davis, California (November 14–16). Radelet, S., J.Sachs and J.-W.Lee (1997) ‘Economic growth in Asia’, ch. 2 Emerging Asia, Manila: Asian Development Bank. Sachs, J.D. and A.Warner (1995) ‘National resource abundance and economic growth’, NBER Working Paper No. 5398, National Bureau of Economic Research, Cambridge, Mass. (December). Saldanha, J.M. (1997) ‘Growth and convergence in Indonesia’, mimeo., Department of Economics, Harvard University (June). Williamson, J.G. (1965) ‘Regional inequality and the process of national development’, Economic Development and Cultural Change 13, 4, pt. II (July), Supplement. —(1985) ‘The historical content of the classical labor surplus model’, Population and Development Review 11, pp. 171–91. —(1995) ‘The evolution of global labor markets since 1830: Background evidence and hypotheses’, Explorations in Economic History 32, pp. 141–96. —(1996) ‘Globalization, convergence and history’, Journal of Economic History 56, pp. 277–306. —(1997) ‘Globalization and inequality, past and present’, World Bank Research Observer 12, pp. 117–35. —(1998a) ‘Real wages and relative factor prices in the world economy 1820–1940’, paper presented to the Conference on 20th Century Growth, Valencia, December 13–14. —(1998b) ‘Growth, distribution and demography: Some lessons from history’, Explorations in Economic History 35, pp. 241–71. —(1998c) ‘Real wages and relative factor prices in the Third World 1820–1940: Asia’ HIER Discussion Paper 1844, Harvard Institute of Economic Research, Department of Economics, Harvard University (August). —(1999a) ‘Real wages, inequality and globalization in Latin America before 1940’, Revista de Historia Economica 17, special number, p. 101–42. —(1999b) ‘Real wages and relative factor prices around the Mediterranean 1500–1940’ ch. 3 in S.Pamuk and J.G.Williamson (eds) The Mediterranean Response to Globalization Before 1950, London: Routledge, forthcoming. Yasuba, Y. (1978) ‘Freight rates and productivity in ocean transportation for Japan, 1875– 1943’, Explorations in Economic History 15, pp. 11–39. —(1996) ‘Did Japan ever suffer from a shortage of natural resources before World War II?’, Journal of Economic History 56, pp. 543–60.

Part II

2

Four revolutions in the textile trade of Asia 1814–1994 The impact of Bombay, Osaka, the Little Tigers and China D.A.Farnie

The first shipment of Lancashire calico made from Liverpool to India in 1814 proved to be a defining event in world history, because it set in motion the process of industrialization in Asia. In that process the cotton textile industry played a preeminent role. What remained essentially a nomadic industry first migrated from east to west during the eighteenth century and then returned, in a new factory incarnation, to its ancestral home. The modern era in its history began in the 1780s when Lancashire assumed supremacy in the manufacture and export of cotton yarn and cloth. The era of the cotton industry as a standard bearer of the British export economy (1788–1949) largely coincides with the era of British primacy in world trade (1787–1939). Three later turning-points in the global history of the trade may be distinguished, occurring respectively in the years 1871, 1913 and 1955. Down to 1871 the share of cotton goods in world trade increased steadily as the largest flow of textile exports ever known in the history of the world gathered momentum. One item in that trade, cotton yarn, proved to be even more important than calico itself. World exports of yarn expanded throughout the century to reach a peak in 1899, in harmony with the continuing increase in the number of hand looms. That traffic supplied the primary stimulus from 1854 to the establishment of spinning mills in the ports of Asia, a process which was designed to replace the import of yarn by a local product. From 1872 to 1984 the share of textiles in world trade underwent a progressive decline as the cotton industry developed in markets formerly supplied by alien imports and as domestic production increased faster than the proportion entering into world trade. In 1913 the reign of King Cotton reached its very zenith, as the volume of world trade in both cotton textiles and textile machinery reached an all-time peak,1 together with the average world per capita consumption of cotton (Table 2.1). The advent of man-made fibre undermined the traditional textile industries after the climacteric of 19262 and transformed production from the 1960s into a multifibre, multi-process mode of operation. Markets were simultaneously affected by a profound change in patterns of consumption. From the 1950s jeans became fashionable wear, first among the ‘baby boomers’ of the postwar generation of the rising middle class. So began a process of the global homogenization of clothing, eroding immemorial national and social differences in the style of garments. The jeans revolution transformed denim from a prosaic and functional element of clothing 49

50

D.A.Farnie

Table 2.1 World production of, and world trade in, cotton manufactures 1913

Source: My own estimates, based upon the statistics assembled by A.Kertesz, Die Textilindustrie sämtlicher Staaten. Entwicklung, Erzeugung, Absatzverhältnisse (Braunschweig, Vieweg 1917).

into a universal and classless symbol of liberation and rebellion. The manufacture of such uniform clothing remained paradoxically more dependent upon unskilled labour than any other industry and thus favoured those suppliers with access to cheap labour. Spurred on by an exploding demand, exports of clothing began from the mid-1950s to expand more rapidly than exports of textiles proper and increased (1958–84) nearly seven times as fast as exports of textiles.3 Those exports first surpassed exports of textiles from 1987 in a historic transformation of the patterns of global commerce. Exports of clothing from Asia had however surpassed its exports of textiles from 1984. Those shipments expanded (1980–94) at a rate 35 per cent faster than those from the rest of the world. By 1994 Asia’s exports of textiles and clothing had become twice as important in its total exports as they were in world trade in general. The successive transformations of the textile trade were powered by the emergence of a sequence of new suppliers of the world market. Bombay and the China market 1873–1914 India had been the birthplace of the cotton industry and the world’s leading manufacturer and exporter of calico. It became, during the 1820s, a net importer of cotton manufactures and remained such for 120 years. It also became, from 1843, Lancashire’s largest single export market and remained such until 1939, in a transformation representing a major landmark in world history. Lancashire never became, however, the ‘clothier of Hindostan’,4 whose markets remained largely the preserve of native hand loom weavers.5 A marked revival of domestic weaving began in the 1820s, fostered by the import of cheap Lancashire millspun yarn6 and reinforced from the 1860s by the advent of superior mill-spun yarn from the new mills of Bombay. Those mills consumed local cotton and found their primary and exclusive market amongst native weavers. The expanding rail network of India was designed less for the export of primary produce than for the carriage of mill-spun yarn in bulk to the domestic weavers of the subcontinent, as Heita Kawakatsu has cogently argued. A secondary market was opened up from 1873 through the export of yarn to the hand loom weavers of

Four revolutions in the textile trade

51

China. During the boom of the 1870s the population of Bombay expanded sixfold as fast as that of Calcutta, which it actually surpassed in size by 1880. By 1876 Bombay had been recognized as the most dangerous single competitor of Lancashire and a greater threat than either the USA or Europe.7 From the 1880s internal demand was further enlarged by the establishment of hand loom workshops, or karkhanas. British cotton manufactures attained their greatest relative importance in the foreign trade of India during the year 1871, when they accounted for 73 per cent of the value of British exports to India and for 57 per cent of the value of merchandise imports into India.8 Thereafter those proportions declined as the mills of Bombay expanded production and reduced exports of raw cotton from India from the peak volume of 1889–90. The continuing growth in population nevertheless enabled Lancashire to increase its exports to India for another 40 years. Lancashire engineers also found in India their largest single foreign market for textile machinery from 1891 until 1969. From 1882–83 India became a net exporter of cotton yarn, some 60 years before it became a net exporter of piece goods. From 1884–85 its exports of yarn surpassed in value its exports of cloth. The China market remained for 30 years a major stimulus to production by the mill owners of Bombay. Their warp yarn proved most suitable to the hand looms of China. Bombay thus tapped levels of demand which had been closed to Lancashire and thus created the incentive to establish spinning mills in Shanghai, a project first mooted in 1879 by two native merchants. Indian competition proved effective enough to reduce the exports of Lancashire yarn to China from their peak value of 1880 and their peak volume of 1885. Indian shipments of yarn to China surpassed those from Britain first in volume from 1883 and then in value from 1892. Thus Lancashire suffered its first great defeat in the world market as its exports of yarn to Asia declined from their peak volume of 1884. China remained the world’s largest importer of cotton yarn from 1885 until 1918 while India became, in succession to Britain, the world’s leading exporter of cotton yarn from 1899 until 1906. Bombay became far more dependent than Manchester upon a single market in so far as China became from the 1890s the exclusive market for its exports. Thus its mills remained highly vulnerable to foreign competition. Their owners had recognized Japan from 1896 as a formidable rival in the Shanghai market, because of the greater uniformity of its yarn and the greater volume of yarn per bale.9 The exportorientation of the Bombay mills reached its highest point in the year 1901, when yarn exports increased to form 61 per cent of total production. Competition from Japan reduced Indian exports of yarn from their peak level of 1905 but Japan only surpassed India in its exports during 1914. Bombay apparently offered little resistance to the new competitor. It had lost its original advantage in cheap freight rates as Indian cotton came to be carried to Japan more cheaply than Bombay yarn was carried to China.10 Thus Japanese competition effectively transformed India from an exporter of a semi-finished manufacture into an exporter of raw cotton to the mills of Osaka as Indian exports to Japan first surpassed in value from 1913–14 its exports to China. The loss of trade proved to be rapid, total and irrecoverable as yarn exports slumped by 95 per cent (1905–24). Little immediate

52

D.A.Farnie

compensation was apparently supplied by the launch of the Swadeshi (‘own country products’) movement in Bengal in 1905. The mills of Bombay sharply reduced after 1905 the rate of expansion of their spindleage. They were also encouraged to install power looms in order to weave up surplus stocks of yarn. Thus they entered into competition with the domestic weavers of India and, from 1909–10, surpassed those hand looms in their volume of production of piecegoods.11 Within six more years those mills had profited by wartime curbs upon exports from Lancashire and had expanded their cloth production to surpass from 1915–16 the volume of net imports of grey cloth. Within two more years their production surpassed from 1917–18 the total volume of net imports of cotton piece goods. The import trade rather than the products of the native hand looms bore the full brunt of the increased output by the mills. Those measures proved inadequate to avert an absolute reduction in mill spindleage in 1916–18, the first in the history of the Bombay industry. That reduction occurred as Japanese imports replaced from 1916 imports from Lancashire, because Japanese goods competed directly with Indian goods as Lancashire goods did not. Japan and the world market 1885–1935 The irruption of Japan into the world market proved much more disruptive than the growth in yarn exports from Bombay had been, because its exports extended over the whole range of textiles and expanded at an unprecedented pace. In Japan the cotton industry fulfilled an even more important role than it did in Britain, although it never dominated the export trade. The Japanese industry differed profoundly from that of Lancashire in its total lack of tariff protection, in the speed with which the Japanese mastered the techniques of machine-production and in their capacity to reduce costs to the lowest level in the world.12 Between 1885 and 1919 Japan expanded its capacity and its production at double the rate of Lancashire in the years 1760–1812. It increased exports of yarn (1890–1915) at fivefold the rate of Lancashire in 1800–30 and exports of piece-goods at double the rate. Its advantages were inevitably misrepresented by hostile propaganda. They resided primarily in better commercial organization and only secondarily in superior industrial organization.13 They were however minimized by a whole series of eminent soothsayers,14 seeking to calm fears that in the end proved impossible to allay. Japan competed with India first in its own domestic market from 1888 and then from 1890 in the export market of China, the jewel in the crown of the Asian trade. China remained the largest potential market in the world, the land of 400 million customers,15 but also home to the world’s largest stock of hand looms. From 1890 it was stimulated to build its own cotton mills and, as a late starter, made exclusive use of the high-speed technology of ring spinning. Its spindleage expanded faster than anywhere else in the world.16 The advent of modern factory industry helped Shanghai to raise its population by 1901 above that of Peking itself. The hypnotic lure of the China market encouraged Japan from 1902 to establish spinning mills in China, so as to steal a march upon the rival mill owners of Bombay. Japan’s unique

Four revolutions in the textile trade

53

Figure 2.1 Takeo Yamanobe (1851–1920) in 1914. He learned the techniques of machine-spinning in Blackburn and Oldham in 1879 and became chief engineer of the Osaka Spinning Company. That company was established in 1880, inaugurated its mill in 1883 and supplied the model for imitation by other Japanese companies. Yamanobe became ‘the father of Japanese industry’ and the Arkwright of the modern Orient. Source: Yonekichi Uno Yamanobe Takeo Kun Sho Den, Osaka 1918.

achievement became the creation by 1930 of an extensive Japanese industrial enclave upon the soil of mainland China. It first captured from Lancashire the peripheral markets of Korea in 1908 and Manchuria in 1909. Then it invaded the Indian yarn market from 1910–11 and became ‘the most important and the most dangerous competitor to Lancashire’.17 Finally, it became the world’s leading exporter of yarn in 1914 and, above all, captured the import trade in piece-goods to China during 1917. In 1926 the firm of Toyo Menka bought a mill in Bombay as a prelude to expansion into the domestic market of India upon the pattern followed in China. That project was however doomed to failure by the lack of the crucial privilege of extra-territoriality as conceded in the treaty ports of China. As Japan lost control over the China trade after 1928 it increased its shipments to India, which replaced China in 1932 as Japan’s largest foreign market. Its achievement in the Indian market during the 1920s remains well-nigh incredible. Japanese spinners bought raw cotton in India, transported it over 5,000 miles, spun and wove it in their own

54

D.A.Farnie

Figure 2.2 Mill of the Osaka Spinning Company in 1888, modelled upon the contemporary mills of Oldham and equipped with machinery made by Platts of Oldham. Source: Platt Brothers & Co. Ltd. Catalogue of Cotton Spinning and Weaving Machinery 1925, p. 353.

mills in Japan and then shipped the yarn and cloth over another 5,000 miles back to India, there to compete successfully with Indian mills in their very own market.18 The conclusion is irresistible, that Japan had developed an industrial capacity wholly different from that of any other state. During the world economic depression it became the only country to expand its production and trade. It doubled its share of the world market for piece-goods, from 18 per cent in 1928 to 32 per cent in 1932. That export-drive was inspired by the vital need for foreign exchange in the wake of the collapse of its export trade in raw silk to the USA after 1929.19 Japan crowned all of its previous achievements by becoming in 1933 the world’s leading exporter of piece-goods and by reversing in 1934 the whole balance of the East-West trade in piece-goods. Its competition wreaked havoc throughout the commercial world because nothing comparable to Japanese prices, either in scale or in range, had been known since the irruption of Lancashire goods on to world markets in the 1780s. The ensuing trade war between Japan and both Britain and India in 1933– 34 clearly revealed that the Indian government was more determined than the British government to protect its cotton industry. The cost-cutting capacity of the Japanese shattered the morale of Lancashire manufacturers so that it never fully recovered. Baldwin’s proposal for a world cartel of 2,200 firms to divide up the global market was fatuous in the extreme. The capture of Shanghai by Japanese

Four revolutions in the textile trade

55

Figure 2.3 The vital role of labour in the mills of China. Source: Far Eastern Review, May 1928, p. 227.

troops in 1937 moved the conflict out of the commercial sphere and added China to the captive imperial markets of Taiwan and Korea. The Indian cotton industry enjoyed a range of advantages highly conducive to its development. The agriculture of India, like that of China, produced all of the necessary raw material. Its mills employed much cheaper labour than that of Japan and remained free until 1882 from the constraints of factory legislation. Its home market generated an immense demand for piece-goods and ranked second in size in 1913 only to that of the USA. The industry expanded under the aegis of Indian management and, between 1882 and 1922, under a regime of free trade. It never however developed the competitive capacity characteristic of Japan, but succumbed to the restrictive influence exerted by the most conservative culture in the world. It was also constrained by the slow growth in per capita income and by the enervating influence after 1922 of tariff protection. Its management remained poor and corrupt,20 strongly inclined towards nepotism and pocketing princely commissions but failing to raise either productivity or profitability. Commissions were determined in bizarre fashion, being calculated upon production rather than upon profit, and were paid irrespective of any profit generated or any loss suffered.21 ‘The Japanese millowner runs his mill on a commercial basis, the Indian millowner runs it for his own aggrandisement.’22 The turnover of firms remained high, in contrast to the position in Japan. Indian mills resembled those of the American South in so far as

56

D.A.Farnie

they fulfilled a social function even more than an economic function, employing a high proportion of labour and developing few productive linkages with the national economy. Thus they imported all of their machinery and even their mill stores from Lancashire, so limiting the degree of control by managers over their costs of operation.23 No local improvements were made to the imported machines, in contrast to the pattern in Japan. Repeated attempts from 1888 to establish a native machinemaking industry failed. A Textile Machine Makers’ Association was established only in 1959. The greatest achievement of those mills was to recapture the domestic market for domestic producers during the 1930s. Thus they created employment in a labour-intensive industry, an achievement reinforced by the survival of hand spinning as well as hand weaving, under the influence of the Mahatma.24 They proved unable however to create a permanent factory labour-force and they failed to raise labour-productivity to the levels reached in Japan. In a striking exception to practise elsewhere they employed male rather than female labour. They adopted ring spinning even more slowly than Lancashire mills. They absorbed the swelling harvest of Indian cotton and made their country for the first time a net importer of raw material in 1942. After the establishment of independence strict controls were imposed for almost forty years (1947–85) upon the mill sector of the industry. A series of national economic plans from 1950 onwards suffered continuing frustration. The Hindu rate of economic growth not only remained low but was exalted by the economist Raj Krishna as the rate most appropriate to India. Hand looms enjoyed however from the 1950s a new golden age,25 which continued long after they were outlawed in China in 1958. The emergence of the Little Tigers (Hong Kong, Taiwan and Korea) After World War II international trade expanded at an unprecedented rate, averaging (1950–90) treble the rate achieved between 1800 and 1913. New exporters appeared in the most unexpected of places.26 The causes of that modern apocalypse must be reserved for study by Asian scholars. Suffice it to suggest here that it occurred without the benefit of professional advice and that its outcome confounded all the prophecies made by development economists. Thus Hong Kong became within 30 years (1950–80) the world’s leading exporter of clothing.27 It was the first of three small states to make textiles into a leading sector of its economy. Those three states in East Asia shared certain common characteristics. They had few natural resources but ample reserves of cheap labour, ideal for deployment into export-led textile production. All three faced a continuing threat to their survival as separate states and all therefore sought to create a firm economic base for their independence. Each had experienced Japanese administration and aspired to emulate and even to compete with Japan. Their success in that competition had no precedent in the annals of world economic history and transformed them, in the eyes of outside observers, into veritable tiger economies. Together the three states increased their GDP per employee at an average annual rate of 5.2 per cent (1960–88) and their merchandise exports at an even more impressive rate of 19.9 per cent (1964–

Four revolutions in the textile trade

57

93), or almost double the global average. All three became leading exporters of clothing and so helped to transform patterns of commerce, first in Asia and then in the world. All three expanded their clothing manufacture and exports even more rapidly than their textile exports. They more than quadrupled their share of Asia’s exports of textiles, raising it from 9 per cent in 1962 to 44 per cent in 1991. They also increased their share of Asia’s exports of clothing from 52 per cent in 1962 to 67 per cent in 1980. Hong Kong’s own achievement was truly historic. Its products met those of Japan in the markets of East Asia and entered into successful competition with them. Its exports of clothing surpassed those of Japan from 1963, its exports of yarn followed suit from 1965. Its success exerted a hypnotic fascination upon Japan and inspired the other states to follow its example. Taiwan and Korea developed a much larger textile industry than Hong Kong and became much more specialized in textile exports than Hong Kong. Taiwan overtook Japan in two waves of expansion, first in exports of cotton yarn in 1969 and in exports of clothing in 1973 and then in exports of cotton cloth in 1986, in production of man-made fibre in 1989 and in exports of textiles in 1990. Korea followed a similar pattern, overhauling Japan in two distinct phases. First it became the main source of supply to Japan as it became a net importer of cotton yarn from 1967, supplemented from 1973 by its shipments of cotton cloth. Then it surpassed Japan in exports of clothing from 1972 and in exports of textiles from 1990. Hong Kong The textile industry of Hong Kong was a new creation by refugees, exiled by the revolution of 1949 from mainland China. Its primary advantages comprised security under the law, unrestricted freedom to trade, abundant supplies of cheap labour and, above all, an all-pervasive entrepreneurial culture wherein private enterprise was firmly supported by a government policy of ‘positive non-intervention’.28 The small-scale family firms of the colony moved beyond the frontier of comparative advantage in order to develop their competitive advantage to the very highest degree. Local entrepreneurs remained unique in their ability to anticipate market trends and to adapt faster than competitors to any new trend. Thus they expanded employment in the garment industry (1950–59) at a rate sevenfold as fast as the rest of manufacturing industry. They secured financial support from the Hong Kong and Shanghai Banking Corporation as well as from the overseas Chinese community. Their imports of British textile machinery reached a peak value during 1961. Their efficiency in tapping the British home market was such as to evoke growing British protests from 1954 onwards. Finally, Hong Kong was forced in December 1958 to accept voluntary export restraints on its shipments to the UK. Local manufacturers immediately adapted operations to that new constraint. They shifted more resources from textiles into the unrestricted sphere of garment-making whilst maintaining the pattern of intensive production by means of three daily shifts. They also invested in mills in Singapore, which remained free from restrictions in the British market until 1965 and in the US market until 1966. Thus

58

D.A.Farnie

textiles provided a peak share of 24.5 per cent of total manufacturing employment in 1959 while textiles and clothing together attained in 1960 a peak share of 54.5 per cent of exports. In 1970 Hong Kong increased its cotton spindleage to an absolute peak and the share of manufacturing in GDP to a peak of 30.9 per cent, thereafter diversifying its economic activity into the service sector. During the climactic year of 1973 textiles and clothing attained a peak share of 51 per cent in both manufacturing output and exports, compared to proportions of 31.5 per cent in the exports of Taiwan and 39.9 per cent in the exports of South Korea. Their joint share in GDP rose by 1975 to 16 per cent, compared to a proportion of 5.9 per cent for Taiwan, 4.7 per cent for Korea and 1.9 per cent for the UK Hong Kong suffered from the repercussions of the oil-price shocks of the 1970s. Its exports of textiles sank from 1975 below the level of those of Taiwan and Korea but its exports of clothing continued to expand until 1982. After the key reforms of 1979, Hong Kong embarked upon a most productive partnership with China. It became once more an entrepôt, for the re-export of manufactures from the mainland. From 1980 it became in succession to Italy the largest exporter of clothing in the world. From 1981 its re-exports of clothing surpassed its own domestic exports. Its business people enlarged the focus of their attention to include the immediate hinterland as local wage rates rose to tenfold their level on the mainland. They began to invest heavily in China, especially in the region of the Pearl River Delta of Guangdong. That bold strategy conferred a triple benefit upon its practitioners. It transplanted manufacturing operations to low-cost sites and so preserved a competitive advantage in labour-intensive processes of production. It also enabled manufacturers to borrow some of China’s quota under the Multi-Fibre Arrangement of 1981 in the US market. Indirectly it contributed to an impressive increase in labour-productivity in Hong Kong itself as employment in manufacturing reached its all-time peak level during 1984 and unemployment began to increase from 1985. Hong Kong attained a climacteric in the textile trades in the years 1987–89, by which time the share of the textile industry in total manufacturing employment had declined from 24.5 per cent in 1959 to less than 9 per cent in 1988. Exports of cotton cloth reached a peak in 1987. Consumption of raw cotton reached an absolute peak in 1988–89. Production of cotton yarn reached its peak volume in 1989. The index of clothing production reached a peak in 1989, that of textiles in 1992. From 1989 Hong Kong overtook Japan as China’s largest trading partner. From 1988 its exports were first surpassed by those of China. From 1992 Hong Kong became the world’s leading importer of textiles, in succession to Germany, and so made re-exports once more into a staple trade. The island colony differed from the other tigers in its compact structure as a city-state, in its role as an uncoached pioneer of a new economic order and in its superior economic performance, reflected in the achievement of a higher rate of value-added per employee in textiles (1985–94) than either Taiwan or Korea. Hong Kong became a positive capitalist paradise, ‘the world’s most aggressively pro-business economy’, and ‘the epitome of laissez-faire capitalism’. 29 It achieved unprecedented prosperity without any input whatsoever by graduates in business

Four revolutions in the textile trade

59

or in economics. In fact it became itself the most successful business school in the world because it was also the most practical. Its commerce created wealth in such abundance as to become from 1987 China’s largest single source of foreign exchange and to generate by 1990 a per capita GDP surpassing that of the UK itself. Hong Kong supplied the most impressive example in world history of self-sustained growth. Its history demonstrated conclusively that economic growth did not depend upon natural resources, upon formal education, upon any reduction in population-growth or upon central planning but upon unrelenting business enterprise. Its history proved the sheer irrelevance to economic development of the concepts of an Asiatic mode of production or of an Asian system of values. It also demonstrated the fundamental irrelevance of dependency theory, whose exponents had solemnly proclaimed the impossibility of any developing economy industrializing under the capitalist system.

Taiwan Taiwan resembled Hong Kong in its lack of natural resources, in the predominant role assumed by small family businesses and in the stimulus supplied by refugees from the mainland, especially from Shanghai. It differed from Hong Kong however in many respects: in the greater influence exerted by both Japan and the USA, in the role of US aid (1951–65), in its function as a potential model economy, on the Rostovian pattern, for emulation in the Third World, in the prestige conferred by government service and by association with the Generalissimo Chiang Kai-Shek (1887–1975), in the larger role played by government policy during the heyday of dirigisme in the 1950s, especially in the resort to tariff protection from 1949, in the entrenched hostility to foreign capital as well as to multinational enterprises and in the constant threat emanating from Red China, 100 miles distant. Taiwan had benefited from 50 years of Japanese administration (1895–1945) which laid the basis for its industrial renaissance, created the necessary infra-structure for economic development and developed textile manufacture from 1920, as well as the cultivation of fibre crops during the 1930s.30 A series of plans beginning with the first FourYear Plan of 1953 ranked textiles and clothing as spearhead industries. They were buttressed by the introduction in 1961 of export incentives and by the establishment in 1966 of export processing zones, which were specifically designed to attract Japanese capital. The rate of expansion of GDP per employee surpassed that of Korea until 1980 and even that of Japan itself in 1973–80. The number of cotton spinning spindles expanded at an average annual rate of 15 per cent (1957–75) or 28 per cent faster than in Korea (1961–79) and 18 per cent faster than in Hong Kong (1950–70): they surpassed those of Korea from 1966 and those of Hong Kong from 1968, reaching a level by 1987 sixteen-fold that of Hong Kong. Those machines were worked by three daily shifts, or for longer hours than anywhere else in the world. Production of cotton yarn and cloth increased 80 per cent faster than in Hong Kong (1954–87). Exports of clothing first surpassed those of textiles from 1968. Taiwan surpassed Hong Kong from 1980 in the production of both

60

D.A.Farnie

cotton yarn and cotton fabric. The total value of exports suffered little check from the two oil-price shocks of 1973 and 1979, save in the year 1977, and surpassed those of Korea from 1979. Taiwan achieved a much higher ratio of exports to GDP than Korea, i.e. 53.6 per cent compared to 28.1 per cent in 1978–80, a ratio elevenfold that of India.31 During the 40 years after 1950 the island was transformed from an agricultural backwater into an economic colossus. The textile firms remained a myriad of small family businesses and provided the new industrial groups with a solid core. The loss of US patronage in 1979 proved a major challenge and stimulated the formulation in 1980 of the Ten-Year Textile Industry Revitalization Plan. During the 1980s the average annual rate of growth of GDP per employee slumped below the level attained by Hong Kong, Korea and even China. The consumption of raw cotton reached a peak level in 1986–87, one year earlier than either Korea or Hong Kong. During 1987 peak levels were registered of spindleage, of the production of textiles and clothing and of exports of clothing, followed by peak exports of cotton cloth in 1990. From 1989 Taiwan nevertheless surpassed Japan in the production of man-made fibre yarn and became the world’s leading producer of polyester fibre, surpassing even the USA in the 1990s. The economic renaissance of China encouraged the island to expand trade with Hong Kong, to which port its exports first surpassed those to Japan from 1990. Its limited reliance upon foreign capital spared it from the repercussions of the Asian crisis of 1997–98 which devastated the economy of South Korea. Korea* Korea began to develop its resources almost a decade later than Taiwan, from 1961, because of the pressing need for post-war reconstruction after the Korean War. It differed from Taiwan in its homogeneous stock of non-Chinese, in the relative unimportance of small-scale enterprise and in its less creative response to Japanese administration (1910–45). Japan exerted more influence in Korea than in any other state as Koreans watched their neighbours like hawks and imitated every successful move they made. Korea benefited from a similar array of advantages to Taiwan, including a work ethic even more effectively internalized than in Japan, US aid (1953–65) and the introduction from 1965 of export incentives. It became the most neo-mercantilist of all the tiger economies, maintaining close links between the government and the giant business conglomerates. Economic plans were implemented from 1962, upon the pattern of Taiwan and exports of clothing developed at similar annual rates of 30 per cent (1965–89) while other exports increased even faster. It diversified however into the production of man-made fibre from the 1960s or much earlier than Taiwan, and undertook joint ventures with the Japanese firm of Toray from 1972. Its exports of textiles and clothing enjoyed two full decades of uninterrupted expansion from 1962 to 1981. Korea first exported more clothing than textiles from 1967, one year ahead of Taiwan. It raised the average annual rate of growth of GDP to 9.7 per cent in 1972–76 *‘Korea’ refers throughout to the Republic of South Korea.

Four revolutions in the textile trade

61

during a turbulent decade32 and trebled its share of world trade between 1975 and 1988. It achieved a higher rate of growth of GDP per employee in 1980–88 than Hong Kong, Taiwan or Japan. Like Taiwan it experienced the irresistible impact of competition from China. It raised its spindleage to a peak in 1988, its consumption of raw cotton to a peak in 1988–89 and its exports of clothing to a peak in 1989. Its exports of clothing declined thereafter more than those of Taiwan, but its exports of textiles first surpassed those of Taiwan in 1992, an achievement repeated from 1994 in respect of exports of manufactures and of total merchandise exports. The renaissance of the textile industry in China 1979–94 The re-emergence of China as a textile superpower took place in two main phases, in 1969–71 and in 1987–88. China had been a net exporter of cotton cloth since 1947, ending a century of limited dependence upon alien imports. It made textiles into its principal export from 1960 and became from 1969 the world’s largest exporter of cotton cloth in succession to Japan, as well as a net exporter of textile machinery. It acquired the largest loomage in the world from 1970 and the largest spindleage from 1971. The progress made by China after the reforms of 1979 remains incredible33 and its official statistics have been the subject of some controversy.34 China became the world’s largest producer of raw cotton from 1981– 82, the world’s largest exporter of cotton yarn (1981–86) and the world’s largest producer of woollen blankets from 1985 as well as of woollen yarn from 1993. It surpassed Italy in the value of its textile exports in 1984, Japan in 1985, and Germany in 1991, to rank first in the world. Above all, it expanded exports of clothing (1979– 93) twice as fast as exports of textiles, making clothing in 1986–87 the fastest growing product group in world trade. It rose rapidly in status from the eighth largest exporter of clothing in 1980 to the first in 1988, surpassing Taiwan in 1981 and both Italy and Korea in 1988. From 1988 it raised its exports of clothing above its exports of textiles. It also became the largest producer of man-made fibre in Asia from 1992, surpassing Japan in the volume of its output. At home China ended the rationing of clothing in 1983 after 30 years. Abroad it doubled the share of textiles and clothing in its export trade to 40 per cent (1978–93). Thus its economic renaissance brought to an end the era of Japanese supremacy in the global textile trade. It also terminated the golden age of expansion enjoyed by Taiwan and Korea but did not affect Hong Kong adversely before 1997. Its shipments also served to revive world trade in textiles, which almost doubled their share in world merchandise exports (1984–94). The Middle Kingdom had become a true Dragon Economy and the core of a new world system. No state in Asia could withstand the influence exerted by China. Even Japan began from 1983 to import more textiles from China than from Korea and became from 1993 China’s biggest trading partner. Japan had lost in 1945 its three captive markets in Taiwan, Korea and China but reconstructed its textile industry after the war with incredible rapidity, aided by the enterprise of its general trading companies. In 1950 it re-enacted its triumph of 1934 by reversing the eastward flow of textiles, evoking immediately a hostile reaction abroad. Then in the late 1960s it began to

62

D.A.Farnie

face competition from its imitators in the tiger economies. Alone in Asia Japan followed the example of Western states in making the great transition to net import, of cotton yarn from 1967, of clothing from 1973 and of textiles from 1987. It had remained the second largest exporter of textiles from 1973 to 1984 as well as the second largest exporter of textile machinery. Its great textile firms preserved their identity and their record of profitability by diversifying their production and the location of their plants. The protectionist reaction in the West From 1934 and again from 1950 the balance of East-West trade in textiles swung decisively in favour of Asia, under the impetus supplied by Japan. Western states began to build protective dykes around their own clothing and textile industries, which proved the most susceptible of all to import-penetration. Tariffs upon textiles and clothing were already the highest levied upon any product group. An additional non-tariff barrier came to be raised, in the form of rigid discriminatory quotas. The driving force behind that reaction was the USA, whose textile industry enjoyed the highest tariff protection in the world and had secured in 1961 additional recognition as one vital to national security. Under American pressure a short-term multilateral agreement was reached in 1961–62 upon the most acceptable mode of protection against the competition of Japanese cotton goods. Then followed a long-term arrangement in 1962 and the Multi-Fibre Arrangement (MFA) in 1974. Thus the industry which in England had been the pioneer of the policy of free trade became from 1961 the pioneer of a new mercantilism, limiting the global movement towards freer trade which had begun with the conclusion of GATT in 1947. The MFA remained unique in function and in status, becoming the most comprehensive system of protection in the whole history of the world economy. Indeed it became increasingly rather than less restrictive. It imposed enormous costs upon the consumers of the West in order to protect domestic clothing industries. It successfully concealed the true level of those costs, however, by employing opaque quotas rather than transparent tariffs. Paradoxically it encouraged the progressive up-grading of exports in the developing states rather than in the developed states, where in accordance with the logic of comparative costs the process should have taken place. In the developed states the increase in Asian competition undoubtedly encouraged from 1963 a notable rise in productivity. Britain experienced a total transformation in the pattern of its foreign trade as the export of textiles gave way to their import. It became a net importer in succession of cotton manufactures and clothing from 1960, of textiles from 1978 and of manufactures in general from 1983. Its balance of trade in textiles and clothing was converted into a multi-billion pound deficit, increasing sixty-fold (1973–95) to a level of £5,000 millions. For Lancashire the revolution in East-West commerce entailed not only the loss of its historic staple trade but also the virtual destruction of a legendary industrial civilization. Total employment in manufacturing in the north-west region sank by at least 772,000 or by 31 per cent (1911–95).35 The proportion of the regional labour-force employed in industry declined from 53 per

Four revolutions in the textile trade

63

Figure 2.4 The Head Offices of Peel Holdings PLC in Manchester 1993. Peel Spinning and Manufacturing Co. was established in Bury in 1885 and became Peel Holdings in 1981, changing in function from a manufacturing firm into a property company. It became the vehicle for one of the region’s leading entrepreneurs, John Whittaker, who bought the Manchester Ship Canal Co. in 1987 and inaugurated the Trafford Centre in 1998 as a shopping city. Its history reflects the transition made by Lancashire from a manufacturing to a service economy. Courtesy J.G.Farnie.

64

D.A.Farnie

cent to 20 per cent. A whole way of life centred around the mill, the chapel, the Sunday school and the co-op suffered massive erosion. Such an outcome would have been wholly inconceivable to the most enlightened observers during the nineteenth century, when Manchester had seemed to be ‘the metropolis of the manufacturing world’.36 Conclusion The emergence of a whole complex of new industrial exporters within ‘the voiceless Far East’ will always remain a subject of compelling interest to historians. That economic renaissance was, however, an unexpected phenomenon, which confounded the prophecies made by development economists. Those scholars had in the 1960s predicted that outstanding rates of economic growth would be achieved by India, Pakistan and Sri Lanka as well as by Argentina, Brazil and Colombia. They had not expected that the ‘Little Tigers’ of East Asia would undergo any comparable development to that attained in South Asia or in South America.37 In the event the economic achievements of the states of East Asia far surpassed those of the other states. Those achievements originated primarily in the ceaseless private endeavours of innumerable small business people. Those entrepreneurs undoubtedly benefited from the political conjuncture of the age as their host-states were enlisted at the onset of the Cold War as anti-communist satellites in the Western bid to stem the onward march of global revolution. The clothing industry became the most dynamic sector within the developing Asian economy. Because it remained more dependent upon unskilled labour than any other industry it proved especially suited to the resource-endowment of overpopulated societies. It also generated higher returns upon the capital invested than did textile manufacture. Thus the garment industry expanded production and exports to a greater extent than did the textile industry and proved less susceptible to recession than did the trade in textiles. It made a larger contribution to intercontinental trade than to intra-Asian trade. From the 1950s Asia supplied a larger proportion of world trade in clothing than of world trade in textiles. Since the 1870s textiles had provided a declining share of world trade. Clothing however from the 1950s consistently increased its share of world trade in manufactures. Its exports generated from 1982 a long-term recovery in the joint share of textiles and clothing in world merchandise trade. Those exports proved to be less affected by the competition of China than were textile exports. By 1984 Asia supplied 58 per cent of world exports of clothing and, exceptionally, 64 per cent of world exports of textiles. The example set by the ‘Little Tigers’ stimulated emulation throughout the ‘Golden Crescent’ which extended southwards in the 1980s from Korea to Malaysia. Of Asia’s fifteen leading export-economies, thirteen built up by 1990 thriving export-oriented textile and clothing industries, often with the aid of Hong Kong entrepreneurs. Thus Thailand became the fastest-growing economy in the world (1985–94), the eleventh largest exporter of clothing from 1987 and the eighth largest in 1991. Indonesia became the fifteenth largest exporter from 1991 and the

Four revolutions in the textile trade

65

eleventh largest from 1993. India signally failed however to profit from the favourable conjuncture of the age. India’s golden age had occurred during the 1940s after the outbreak of the Sino-Japanese War removed the two leading textile producers of Asia from competition in the world market. India remained a rejoicing neutral power and enjoyed a prolonged wave of industrial expansion. From 1941– 42 it became a net exporter of cotton cloth for the first time since the 1820’s. During the three years 1942, 1943 and 1950 it even became the world’s principal exporter of cotton piece-goods. It surpassed the UK from 1954 as an exporter of cotton piece-goods and remained the chief source of supply of such goods to the British market until 1970, its shipments thereto reaching peak volumes during the years 1960 and 1968. Its eminent position in world trade in textiles was however undermined by the post-war recovery of Japan, by the expansion of domestic demand from 1951 and by the foreign-exchange crises of 1956 and 1966. In the world league of exporters India ranked second to Japan for twelve years (1955– 66) and then third to Japan and Hong Kong from 1967. In the value of its textile exports India declined from ninth in the world in 1973 to fifteenth in 1985. It was surpassed in rapid succession by Taiwan in 1975, by South Korea in 1976, by Hong Kong in 1979 and finally by Pakistan in 1983–93. The rate of growth in the value of its textile exports sank to little more than one-third of that for the rest of Asia during the years 1962–90 (5 per cent: 13.75 per cent). Its share in total world trade had peaked at 8 per cent in 1865 under the influence of the Cotton Famine but declined to 1 per cent in 1964 and further to 0.54 per cent in 1990, as its population grew more rapidly than ever before. In the export of clothing India tapped however a valuable niche market, expanding its shipments in 1962–96 half again as fast as the rest of Asia (29.5 per cent: 19 per cent) and faster than Taiwan itself (23.4 per cent). The value of its clothing exports surpassed that of its textile exports from 1986, raising its global ranking therein from twenty-fourth in 1973 to tenth in 1996. Pakistan benefited from its supply of cheap labour and cheap cotton to become an important producer of textiles. It became the world’s leading exporter of cotton yarn from 1967. In the volume of exports of cotton piece-goods it surpassed first the UK from 1968 and then India from 1978. It rose in status from the world’s fifteenth largest exporter of textiles in 1984 to the twelfth largest in 1991, becoming more dependent than any other state upon such exports. For its part Bangladesh achieved, with the help of India, its ‘second independence’ in 1971. It became, together with Mauritius, more dependent upon the export of clothing than any other state. From 1993 it began to catch up with Pakistan in the export of textiles. The history of both Pakistan and Bangladesh illustrates the continuing relevance of the staple theory of economic development, first formulated in 1923 by W.A. Mackintosh. Intra-Asian trade expanded within the sphere of textiles even more than in that of clothing. Singapore followed the example of Hong Kong in developing an entrepôt trade in textiles and in clothing, which from 1993 surpassed its domestic exports of clothing. The prolonged boom generated linkages which further reinforced the process of economic expansion. A widening range of

66

D.A.Farnie

demands was generated for fibre, for chemicals and for machinery. China remained the world’s largest producer of cotton, even after per capita consumption reached a peak during 1987. Japan derived major benefits from the demand for machinery and increased its exports of textile machinery to surpass those from Britain in 1973. One social effect was to accelerate the global homogenization of clothing, as the influence of fashion prevailed over that of tradition, in Asia as in the West. The process of economic growth finally became one of sustained economic development as the tiger economies diversified their operations into the field of hi-tech industry. The industrialization of Asia and the final decline of the British cotton industry coincided with the formulation in Cambridge in 1962 of a bold revision of economic history.38 That iconoclastic re-interpretation of the past denied that the cotton industry had served as a leading sector during the Industrial Revolution in Britain. It was however seriously flawed insofar as it undertook no comparative analysis of the experience of Britain with that of other states. Such a comparative analysis clearly reveals the close association between the expansion of the cotton industry and the process of economic development. That association may be measured by the industry’s notable contribution to GDP in the history of Japan and Hong Kong, as well as of England. Such a superfluous revision found no favour within the world of Asia. The historic experience of the continent serves indeed as a permanent refutation of the whole thesis. The cotton industry retains its unique position within the pantheon of economic history for two good reasons. First, it has served as a prime mover in the process of global economic development. Second, it has continued to exert a hypnotic fascination upon every new generation of students. It seems to be a truth, universally acknowledged, that every economic historian is, by virtue of his profession, an authority upon the history of the cotton industry. Notes The place of publication of all the sources cited below is London, unless otherwise indicated. 1 A.Kertesz, Die Textilindustrie sämtlicher Staaten. Entwicklung, Erzeugung, Absatzverhältnisse, Braunschweig: Vieweg, 1917. 2 H.Frankfurther, ‘Hochkonjunktur in der Textilindustrie?’, Wirtschaftsdienst, 29, 22 Juli 1927, pp. 1084–5. H.Grünbaum, ‘Zur Struktur der Welttextilindustrie’, Vierteljahresheft für Konjunktur-forschung, 24, 1931, pp. 6–45. A.J.H.Latham, The Depression and the Developing World, 1914–1939, Croom Helm, 1981, pp. 175–8, emphasizing the depression in world agriculture which set in from 1924–5. 3 G.A.T.T., International Trade. Trends and Statistics, 32 vols, 1961–95. W.T.O. Annual Report, 1996, 2 vols. 4 R.A.Arnold, The History of the Cotton Famine, Saunders, 1964, p. 17. 5 J.Forbes Watson, The Textile Manufactures and the Costumes of the People of India, Eyre and Spottiswood, 1866. Textile Manufacturer, 3, July 1877, p. 212, Sorabjee S.Bengalee, ‘Is there any competition between Lancashire and Bombay Mills?’. W.W.Hunter, The Imperial Gazetteer of India, Trübner, 1886, 2nd edn., VI, p. 600. 6 P.Dicken, Global Shift (Transforming the World Economy), Paul Chapman, 1986,

Four revolutions in the textile trade

7 8 9 10 11 12

13

14

67

3rd revised edition 1998, p. 283, passes, for a Manchester geographer, a thoroughly misleading judgment on the British textile industry of the nineteenth century as one which ‘effectively strangled the development of an indigenous textiles industry in the major colonies, especially India’. Textile Manufacturer, June 1876, p. 162; September 1876, p. 258; June 1877, p. 171. Manchester Chamber of Commerce, Bombay and Lancashire Cotton Spinning Inquiry. Minutes of Evidence and Reports, Manchester, Ireland, 1888. D.A.Farnie, The English Cotton Industry and the World Market 1815–1896, Oxford: Clarendon, 1979, pp. 117, 119. Textile Manufacturer, 24, 15 August 1898, p. 282, ‘The future of the Indian cotton industry’. Textile Recorder, 16, 15 August 1898, p. 155, ‘The Bombay yarn market’. Textile Recorder, 25, 15 October 1907, p. 188, ‘The Bombay Mill Industry v. The Japanese Mill Industry’. Peter Harnetty, ‘“De-industrialization” revisited: The hand loom weavers of the Central Provinces of India, c.1800–1947’, Modern Asian Studies, 25, July 1991, pp. 455– 510. Freda Utley, ‘Cotton trade costs of production in Japan I—Spinning’, Manchester Guardian Commercial, 25 April 1929, pp. 489–90; ‘Cotton trade costs of production in Japan II—Weaving’, Manchester Guardian Commercial, 2 May 1929, p. 517–18. Freda Utley, Lancashire and the Far East, Allen & Unwin, 1931. G.B.Rea, ‘The secret of Japan’s trading success’, Far Eastern Review, December 1929, pp. 553–6. B.Ellinger, ‘Japanese competition in the cotton trade’, Journal of the Royal Statistical Society, 93:ii, 1930, pp. 135–232; Textile Weekly, 24 January 1930, p. 595. B.Bowker, ‘Lancashire at the cross roads… Do the salesmen know their ABC?’, The Observer, 9 February 1930, pp. 17–18; ‘Lancashire and Japan’, Spectator, 15 December 1933, pp. 888–9. G.W.Daniels, ‘The cotton trade’, Textile Recorder, 38, April 1921, p. 45. E.F.Crowe, ‘Japanese competition’, Manchester Chamber of Commerce, Monthly Record, July 1925, p. 219. H.Kendrick, ‘The Lancashire cotton trade’, The Nation, 40:7, 20 November 1926, pp. 264–5. W.B.Cunningham, Report on the Cotton Spinning and Weaving Industry in Japan, summarized in Manchester Chamber of Commerce, Monthly Record, March 1927, pp. 80–1. S.S.Hammersley, ‘Lancashire cotton trade and foreign competition, Textile Recorder, 45, August 1927, pp. 84–5. B.Ellinger and J.H.Grey, ‘Lancashire’s declining trade with China’, Transactions of the Manchester Statistical Society, 9 November 1927, pp. 1–55. H.Clay, ‘A major and a minor problem for Lancashire spinners’, Manchester Guardian Commercial, 12 July 1928, pp. 45– 6. W.S.Ascoli, ‘Problems of the Lancashire cotton trade’, Manchester Guardian, 12 December 1928, p. 18iv. W.Abbott, ‘Foreign competition in the cotton industry “Lancashire not finished yet”’, Textile Mercury, 22 December 1928, p. 478; 29 December 1928, p. 495. G.W.Armitage, ‘What’s wrong with cotton?’, Manchester Guardian Commercial, 24 January 1929, pp. 109–11, reprinted from the Manchester Guardian, 18–24 January 1929. A.Crickmore, ‘Foreign competition at home and abroad’, Textile Weekly, 17 January 1930, p. 552. B. and H.Ellinger, ‘Japanese competition in the cotton trade’, Journal of the Royal Statistical Society, 93:ii, 21 January 1930, pp. 135–232. J.Haslam, ‘The vitality of the Lancashire cotton trade’, Textile Weekly, 14 February 1930, pp. 663–4. J.Kerfoot, ‘Can Lancashire compete with Japan in low-grade cloths?’, Textile Weekly, 6 June 1930, pp. 299, 305. G.Wallace, ‘The Japanese Bogy’, Daily Dispatch, 29 March 1930, p. 6vi. B.Ellinger, ‘Lancashire’s bulk trade. Can it be regained?’, Textile Weekly, 9 October 1931, pp. 140–3. G.C.Allen, ‘The last decade in Japan’, Economic History (Economic Journal Supplement), January 1933, p. 643. B.Ellinger, ‘Japan: Her commerce and finance. II—Textile prices, wages, organisation’, Manchester Guardian Commercial, 1 April 1933, p. 248. B.Ellinger,

68

15 16 17 18 19 20 21 22 23 24

25

26 27

28 29

D.A.Farnie

‘Can cotton agreement with Japan be reached?’, Manchester Guardian Commercial, 13 May 1933, p. 363. G.B.Sansom and H.A.Macrae, Economic Conditions in Japan, Department of Overseas Trade, 1935. N.Skene Smith, ‘Japanese competition and international trade theory’, Economic Journal, September 1936, pp. 424–30. Carl Crow, Four Hundred Million Customers, Hamish Hamilton, 1937. H.D.Fong, Cotton Industry and Trade in China, Tientsin: Nankai University, 1932, 2 vols. Kang Chao, The Development of Cotton Textile Production in China, Cambridge, Mass.: Harvard University Press, 1977. Manchester Guardian, 6 June 1911, p. 10v; 9 June, p. 12iv; 13 June, p. 13iv; 28 June, p. 14v, A.S.Lewis. Far Eastern Review, 22 September 1926, pp. 400–3, ‘Why Bombay mills cannot compete with Japanese mills’. Kazushi Ohkawa et al. (eds), Estimates of Long-term Economic Statistics of Japan since 1868. Vol. 11 Textiles by Shozaburo Fujino et al., Tokyo: Toyo Keizai Shinposha, 1979, p. 309. Textile Manufacturer, 13, February 1887, p. 62, ‘Bribery amongst employees in Indian cotton mills’. Textile Recorder, 16, 15 July 1898, p. 112, W.C.Wood, ‘Cotton Industry of China and India’. Textile Recorder, 25, 15 October 1907, p. 188, ‘The Bombay mill industry v. The Japanese mill industry’. Textile Manufacturer, 14, April 1888, p. 149, ‘Growing Competition’; 25 October 1897, p. 362, ‘The Making of Machinery in India’. Textile Recorder, 46, 14 July 1928, p. 42, ‘Cotton Spinning in India’. M.K.Gandhi, The Wheel of Fortune, Madras: Ganesh, 1922, (bound in khadi rather than in machine-woven cloth). Textile Recorder, 42, November 1924, p. 55, W.W.Taggart, ‘Hand spinning with modern mechanism’, on a 12-spindle ring frame devised for domestic use. J.E.Orchard, Japan’s Economic Position. The Progress of Industrialisation, New York, 1930, p. 488. Yukihiko Kiyokawa, ‘Technical adaptations and managerial resources in India: a study of the experience of the cotton textile industry from a comparative viewpoint’, The Developing Economies, Tokyo, XXI, 1983, pp. 197–233. Tirthankar Roy, Artisans and Industrialization: Indian Weaving in the Twentieth Century, Delhi: Oxford University Press, 1993. Textile Recorder, 37, 15 May 1919, pp. 3–4; 39, 14 May 1921, p. 51; 15 March 1922, p. 39; 40, 15 March 1923, p. 64; 41, 15 June 1924, p. 109; 44, 15 May 1926, pp. 157–8, ‘Hand loom weaving in India’. M.P.Gandhi, The Indian Cotton Textile Industry, Calcutta: Mitra, 1930, pp. 82–94. D.R.Gadgil, The Industrial Evolution of India in Recent Times, Calcutta: Oxford University Press, 1924, 4th edition 1942, p. 290 (a reference I owe to John Latham). Ezra Vogel, The Four Little Dragons: the Spread of Industrialization in East Asia, Cambridge, Mass.: Harvard University Press, 1991. Asia and World Textiles. Proceedings from the 74th World Conference, Manchester, Textile Institute, 1994. P.T.Bauer, ‘The lesson of Hong Kong’, in Equality, the Third World and Economic Delusion, Weidenfeld, 1981, pp. 185–90, reprinted in revised form from The Spectator, 224, 19 April 1980. Ken Davies, Hong Kong after 1997, Economist Intelligence Unit, Research Report, 1996. Tony Fu-Lai Yu, Entrepreneurship and Economic Development in Hong Kong, Routledge, 1997, p. 10. Jon Woronoff, Hong Kong: Capitalist Paradise (Hong Kong: Heinemann Asia, 1980. Fortune, 14 November 1994, p. 61, Miron Mushkat of Lehman Bros. C.Patten, ‘Manchester and Hong Kong, A tale of two cities’, Manchester Memoirs, Vol. 133,

Four revolutions in the textile trade

30 31 32 33

34

35 36 37 38

69

1994–5, Manchester Literary and Philosophical Society, 1996, p. 22. Fortune, 26 June 1995, pp. 40–59, Louis Kraar, ‘The Death of Hong Kong’. Samuel P.S.Ho, Economic Development of Taiwan, 1860–1970, New Haven, Yale University Press, 1978. Lloyd G.Reynolds, Economic Growth in the Third World, 1850–1980, New Haven, Yale University Press, 1985, p. 410. Robert Summers and Alan Heston, ‘The Penn World Table (Mark 5): An expanded set of international comparisons, 1950–1988’, Quarterly Journal of Economics, 106, May 1991, pp. 357–8. Yanrui Wu, Productive Performance in Chinese Enterprises. An Empirical Study, Macmillan, 1996, pp. 119–36, ‘The Textile Sector’. Economist Intelligence Unit/ UNIDO, China Industrial Development Review, 1996, pp. 123–33, ‘Textiles and Garments’. OECD, China in the 21st Century: Long-Term Global Implications, Paris, OECD, 1996. Financial Times, 7 November 1994, China Supplement, iv, Martin Wolf, ‘Baffling questions for China-watchers’. Paul Krugman, ‘The Myth of Asia’s Miracle’, Foreign Affairs, 73:6, November 1994, pp. 62–78; 74:2, March 1995, pp. 170–7. Paul Krugman, ‘Whatever Happened to the Asian Miracle?’, Fortune, 18 May 1997, pp. 8–10. In order to measure the full extent of the trade of China, re-exports from Hong Kong have been added to exports from China for the years 1980–94. The statistics published annually by GATT are made more complex by regular later revisions. C.H.Lee, British Regional Employment Statistics 1841–1971, Cambridge: Cambridge University Press, 1979, 1911 Census, Series B. Regional Trends, 31, 1996, Central Statistical Office. P.Gaskell, The Manufacturing Population of England, Baldwin, 1833, p. 229. T.Fu-Lai Yu, op. cit. 1997, pp. 6–9. D.A.Farnie, op. cit. 1979, pp. 18–36.

3

Europe, China and Japan Transfer of silk reeling technology in 1860–95* Debin Ma

The arrival of British and American gunboats in East Asian waters around the midnineteenth century meant that Western imperialism was about to engulf the world’s last frontier: the Far East. The recognition of Western technological and military supremacy provoked shocked responses that would become the defining theme of modern East Asia. In China, the Self-strengthening movement started in the 1860s and in Japan, the Meiji-Restoration took place in 1868. Within three decades of these initial responses, China and Japan were at war with each other, not with the Western imperialists. The Sino-Japanese naval war in 1894–5 was the test of these East Asian responses to the Western challenge. China’s defeat by Japan was to fundamentally alter the East Asian power structure which had been dominated by China for centuries. This paper is part of a study that asks if there is a link between this reversal of political and economic ordering and the performance of specific economic sectors in China and Japan. The raw silk sector is examined here within the broad framework of East Asian modernization in the context of Western imperialism. This study argues that in Japan, economic reforms and industrialization created the condition for spectacular growth of raw silk exports 1860–1940 in contrast to China where raw silk exports stagnated. This paper concentrates on European silk reeling technology and compares its adaptation in Japan and China between 1860 and 1895.1 Although Japanese raw silk exports only overtook China and dominated the global market in the twentieth century, this initial period was crucial in setting the long-run trajectory and developmental pattern of the raw silk sectors and in explaining their performance for the entire period. The paper is divided into three parts. The first section gives an overview of the general political and social conditions in China and Japan, as well as the economic and geographical environment of the major raw silk-exporting regions. It outlines *

This paper is a revised version of a chapter of my dissertation. I wish to thank my dissertation committee members at UNC, Chapel Hill, especially Robert Gallman, Paul Rhode, Miles Fletcher. I also want to thank the participants of the Asian Pacific Dynamism session at the 12th International Economic History Congress in Madrid, Spain in August 1998 and particularly A.J.H.Latham. My thanks also go to the Japan Society for the Promotion of Science for supporting my research stay in Japan. I am solely responsible for all the errors.

70

Transfer of silk reeling technology

71

the central question of this paper with time series data. The second section provides a detailed description of the initial pattern of technology transfer to Japan and China. The final section provides a summary of the economic reforms which determined why these two countries adapted modern silk reeling technology in different ways. Silks of East Asia: two countries and three regions The mid-nineteenth century was not a good time for Qing China and Tokugawa Japan. China, the long self-regarded ‘Middle Kingdom’, saw defeat and humiliation at the hands of a few British ships, which sailed half way round the world to win the so-called Opium War. As for Japan, a country then largely unknown to the rest of the Western world, no shots were even fired. The mere arrival of American Commodore Matthew C.Perry’s battle fleet in Edo Bay was enough to dispel any thoughts of resistance. After a series of diplomatic manoeuvres from the West, China and Japan surrendered their tariff autonomy and opened up designated treaty ports for free trade.2 Of the East Asian products destined for the European market, raw silk became one of the most important. With the opening of the Suez Canal, and the completion of intercontinental railroads in the USA after 1869, Chinese and Japanese raw silk exporters found themselves part of a well integrated global trading network. They were joined by Italy, a late industrializer in Western Europe. Their combined raw silk exports exceeded more than three-quarters of the total global trade. The force behind East Asian raw silk exports was comparative advantage. Silkworm rearing was an extremely labour- and skill-intensive activity, with little room for drastic labour-saving innovations. It was usually carried on in densely populated regions with dispersed small dwellings where there were few opportunities for non-agricultural work. This was certainly true of Japan and the two leading raw silk export-oriented regions of China, Shanghai and Guangdong.3 In the late nineteenth century, population densities in Japan and the Shanghai and Guangdong areas were roughly comparable.4 Rice paddy fields with small individual family land holdings or tenancies and intricate pre-modern transportation systems formed the common socio-economic background. The economic advantages seemed to extend beyond mere factor endowment. The use of silk originated in China several millennia ago. Even though industrialization in Europe had undoubtedly left East Asia economically far behind, the Chinese might have maintained technological superiority in both sericulture and silk reeling well into the modern age. Claudio Zanier, an Italian economic historian, painstakingly established that it was probably only in the late seventeenth and early eighteenth centuries that Italian Piedmontese silk reeling technology surpassed that of China. This came with the invention of a silk thread crossing system and rigid axle transmission mechanism.5 He also pointed out that as late as the mid-nineteenth century, Europe still believed it had much to learn from East Asia in sericultural technology. This is indicated by the publication of two translated sericultural texts from China and Japan during that period (Zanier 1994, p. 77).

72

Debin Ma

Figure 3.1 Exports of raw silk. Source: Chinese export data from Xu et al. 1990, Appendix. Japanese export data from Fujino et al. 1979, Vol. 11, Table 63.

So, in the area of silkworm rearing and silk reeling, the relative degree of backwardness between East Asia and Europe was slight and the required technological catch-up minimal. Favourable factor endowments and a strong technological tradition could explain the rise and dominance of East Asia in the world raw silk market, but not the relative positions of Chinese and Japanese raw silk exports. The data presented in Figure 3.1 show that in 1873 China exported three times as much raw silk as Japan, but around 1905 Japanese raw silk exports overtook China, and on the eve of the collapse of the world commodities market in 1930, Japanese raw silk exports were three times more than those of China. Before setting out to analyse the causes of the contrasting performance of Japanese and Chinese exports, it is important to distinguish two main types of raw silk exported. Hand-reeled (or re-reeled) denotes raw silk reeled using traditional methods. Machinereeled was produced in East Asia by machines imported or adapted from Europe, usually driven by an inanimate power source. Table 3.1 presents a breakdown of these two types of raw silk for the three regions. It shows that on the scale of machine-reeled silk production, Japan ranked highest and Shanghai lowest, with Guangdong taking the intermediate position. The order is reversed when handreeled silk is examined. Shanghai always exported more than Japan, which in turn exported more than Canton. But machine-reeled raw silk exports did not immediately displace traditionally hand-reeled raw silk in all three regions. In

Transfer of silk reeling technology

73

Table 3.1 Raw silk exports of China and Japan by types (in metric tons)

Source: Chinese raw silk data from Chen 1989, p. 170, Xu et al. 1990, pp. 688–707, Guangdong hand-reeled silk from Wong 1995, pp. 90–1. Amount for machine-reeled raw from Shanghai before 1894 were insignificant and not separately reported in the Maritime Customs Reports. Japanese raw silk data from Fujino et al. 1979, Vol. 11, Tables 4–4 and 63, Hand-reeled silk 1879–96 based on Ikawa 1992, p. 252; and 1860–67 based on Ishii 1972, p. 41, Table 4.

time the exports of hand-reeled silk in all three regions declined and became insignificant as a share of exports. So while recognizing the importance of handreeled silk in the earlier period, it is safe to claim that the key factor accounting for the eventual output of raw silk exports lies in the regions’ (productive capacity) for machine-reeled raw silk. This, as seen in Table 3.1, is exactly where Japan differed from the areas in China. Between 1890 and 1900, Japan exported exactly four times as much machine-reeled raw silk as Shanghai, and nearly twice as much as Guangdong. This export ratio increased to more than ten times as much between Japan and Shanghai and more than eight times as much between Japan and Guangdong in 1920 and 1930. Figure 3.2 shows the number of modern machine-reeling basins in Shanghai, Guangdong, Japan and Italy.6 Several important patterns emerged. First, Italy had most reeling basins in the 1860s and 70s, and continued to lead up to the 1890s. Afterwards, the silk reeling output in Italy was maintained at a high level but no longer grew. This reflects the simple fact that for East Asia, Italy (as well as France) was the source of machine-reeling technology before the 1890s and the country to be caught up with. Second, Shanghai pioneered the introduction of machine-reeling technology, but failed to sustain expansion before 1895. Exports of machine-reeled silk were so small that before 1894, the Shanghai Customs records did not indicate exports of machine-reeled silk separately from hand-reeled (see Table 3.1). By comparison, the silk reeling industry in Guangdong started ten years later, but quickly took the lead. The number of modern reeling basins hardly increased in Shanghai before 1896 and there were fewer than five thousand within the metropolis (and practically none in the neighbouring rural areas). But the number of modern reeling basins in Guangdong grew from 2,400 to 26,356 between 1882 and 1894.7 However, the growth of the Guangdong silk reeling industry peaked around the early 1920s, much earlier than all the other reeling industries. The most dramatic finding is the performance of the Japanese silk reeling sector. As will be seen later, the modern Japanese machine-reeling industry started around

74

Debin Ma

Figure 3.2 Number of modern silk reeling basins in China, Japan and Italy 1861–1937. Source: Data for China from Xu et al. 1990, pp. 611–14. Data for Japan is from Fujino et al. 1979, Vol. 11, p. 304, Table 60, Col. 4. Data for Italy is from Federicó 1996, Table AXVIII, pp. 222–3.

1870, almost ten years later than in Shanghai, but it grew fastest. Modern reeling basins in Japan increased from 16,856 in 1879 to 85,988 in 1895 and continued to grow at a high rate throughout the twentieth century. In the twentieth century, there were roughly four times as many silk reeling basins in Japan as in Italy and Guangdong, and ten times as many as in Shanghai. With this analysis, it is possible to isolate the reasons for the contrasting performance of Chinese and Japanese raw silk exports. Why did the Chinese machine-reeling silk industry, which started at the same time as the Japanese or even earlier, turn out to be less successful? The following section examines the patterns of technology transfer in the three regions between 1860 and 1895. Patterns of technology transfer and export performance European reeling machines in Shanghai and Japan To the political regimes of China and Japan in the mid-nineteenth century, powerful guns and ships may have symbolized the Western challenge. But to the vast and scattered traditional silk-producing and trading regions of East Asia, the challenge came by way of new machines and institutions. These were European silk reeling machines with an additional twisting mechanism and rigid axle mechanism, operated in modern factories using an inanimate power source. The superiority of European silk reeling equipment dated back to the second half of the seventeenth century, when Italian machines began to adopt the rigid axle mechanism to transmit power to move the reel, in place of the driving-belt system it had borrowed from

Transfer of silk reeling technology

75

China. Chinese (and Japanese) silk reeling machines seemed to have relied on the driving-belt well into the nineteenth century. But the most crucial European innovation was the additional twisting mechanism (Chambon style in France and Tavelle in Italy).8 This additional process of spinning several silk threads together dry greatly enhances the cohesiveness, evenness and uniformity of the silk thread, features that are absolutely essential for high quality raw silk. These innovations along with the adoption of steam to heat the reeling basin and drive the reel, which came in the first half of the nineteenth century, constituted the basic European silk reeling system, a system China and Japan were to emulate. European-style silk reeling factories were established in the city of Shanghai in 1861 and in Japan in 1870. It is interesting to note that the initial process of bringing European reeling technology to East Asia linked Shanghai with Japan. The first modern silk reeling establishment in East Asia, the Shanghai Ewo Filature, was set up in 1861 by the most powerful European trading company in the Far East, Jardine, Matheson & Co. It closed down after ten years of unprofitable operation. In 1869, when Jardine began thinking about closing their loss-making Ewo Filature in Shanghai, their Yokohama office was visited by Maebashi han officials from Gunma prefecture in Japan, who expressed interest in a modern reeling mill. Negotiations between Jardine’s Yokohama office and the Maebashi officials to transfer the equipment of Ewo Filature to Japan eventually failed, and Maebashi later bought equipment directly from a Swiss firm and founded Japan’s first modern reeling factory. This was one of many instances where Japan and Shanghai found common ground in acquiring European reeling technology. The best example of linkages between Shanghai and Japan in technological transfer is found with Paul Brunat, a French silk reeling expert who played a large role in transferring European silk reeling technology to East Asia. He and some French assistants worked for the Japanese government in 1872 to set up a model silk reeling mill in Tomioka, Gunma prefecture. The Tomioka silk mill, with 300 silk reeling basins, is said to have been a copy of one of the best French filatures of the time. It represented Meiji Japan’s ambitious effort to introduce the most up-todate Western silk reeling technology. Unfortunately, the European style factory soon proved to be too advanced for the pre-industrial economy of Japan to follow, and consistently produced financial losses.9 The Tomioka silk mill, along with other loss-making government enterprises, were put up for sale in the 1880s and 1890s by the government at a price lower than the original investment. Brunat returned to France in 1875 after the expiration of his contract with Tomioka, but returned to East Asia again in 1878. This time he was employed by an American trading company Russell & Co. to establish a silk reeling factory in Shanghai. This later turned out to be modestly successful and expanded in the 1880s.10 These anecdotes show that the silk reeling industries in Japan and Shanghai shared an almost identical technological source, and each had its own lessons in absorbing European technology and institutions in an East Asian context. It is therefore surprising that the development of modern silk reeling in Japan and Shanghai during the 1880s and 1890s turned out differently. While the modern Shanghai reeling industry began almost ten years before that in Japan, its growth

76

Debin Ma

spurt came in the later 1890s, almost two decades later than its Japanese counterpart. There seem to be two inter-related causes. The first concerns the legal environment for private enterprises in Japan and China. In Japan, central and local government gave strong administrative leadership in protecting and encouraging the modern silk reeling industry, which partly substituted for a lack of formal legal protection for private ventures. The administrative leadership was extremely important in overcoming hesitation and resistance to Western technology and institutions in a society which had experienced a long period of comparative seclusion.11 There was enthusiastic experimentation during the first ten years of the Meiji period with the joint-stock corporation system newly introduced from the West. The Commercial Code was drafted between 1881– 84 and partially adopted in 1893. With subsequent revisions of the Commercial Code, legal protection for private enterprise found a solid base in Japan (Nishikawa et al. 1990, ch. 5). More significantly, in 1881, as part of the Constitutional movement, economic policy in Meiji Japan saw a major shift when it was decided to give full scale support for the private sector and all government-owned and operated factories were sold to the private sector. The sale included the Tomioka Mill which was bought by Mitsui at a third of the price of its initial investment (Smith 1965). The second cause is related to the way knowledge was transferred and technology adapted. Again the Japanese government played a pivotal role. Take the case of the government-run reeling enterprise, the Tomioka Mill. During its years of operation, the Tomioka Mill served as a model and knowledge base for teaching Western technology and management methods to private entrepreneurs. It drew a large number of visiting entrepreneurs, and employed labourers on a rotating system, who were expected to take with them the skills they had learned at Tomioka when they returned to their home villages. Quantitative studies by Kiyokawa confirm that while the Tomioka mill had little direct influence on the adoption of French filature technology, the factory made a significant impact on the diffusion of largescale filatures (more than 100 basins) in Japan.12 The Tomioka factory was also actively involved in sericultural and silk reeling technology research and produced useful results.13 In this sense, Tomioka failed in a financial sense but succeeded as a means of diffusing modern reeling technology and the factory system in Japan. The financial failure of Tomioka also taught Japanese entrepreneurs the importance of adapting technology to local conditions. They recognized that European technology, though superior, was too expensive, too capital-intensive, and too large-scale for the East Asian economy where there was little capital accumulation and labour was cheap. A ‘technological compromise’ resulted—a silk reeling machine that substituted cheap labour and local materials for capital. The key feature of European silk reeling technology remained—the twisting mechanism and the steam-heated oven. But the machine was greatly simplified, wooden parts replaced those of steel, and a porcelain basin replaced a metal one. Re-reeling, better fit for the relatively high humidity in Japan, replaced direct reeling. Finally, labourers or water, provided motive power instead of steam. Large-scale factories were replaced by small-scale family rural cooperatives or cottage

Transfer of silk reeling technology

77

industries.14 These adaptations led to the spread of modern silk reeling factories to the rural sericultural regions, with the highest concentration in the Suwa district of Nagano prefecture. In China, private enterprise had no legal premise or sanction in the nineteenth century. This explains why all the early modern reeling factories were in Shanghai, to take advantage of the city’s extraterritorial status. Most Chinese-owned and operated reeling factories registered themselves as foreign enterprises to avoid any legal ambiguities and extortion from local governments.15 Yet this protection was tenuous, as the initial treaty port system only granted Westerners the right to trade, not the right to establish and operate factories. So the legality of foreign-owned silk reeling ventures was often called into question.16 European reeling technology and the factory system based on Western capital was established in Shanghai, a colonial enclave with a legal and institutional environment distinct from the rest of China. Foreign direct investment brought to China the whole spectrum of up-to-date Western technology and management, but it also bore the stigma of Western economic exploitation and encountered strong local resistance.17 So European-style reeling factories made little headway outside the treaty port. Confirming the modern reeling industry within a colonial enclave, away from the traditional raw silk production centres, also meant that modern Western technology was not easily accessible to the most competent local entrepreneurs, the traditional silk reelers. Almost all the early silk reeling factories were operated and managed by foreigners or comparadores, Chinese merchants who acted as go-betweens for Western business in China. They had little interest in carrying out technological adaptation suited to local conditions. So there was almost a complete lack of technological localization in Shanghai. The Shanghai factories, featuring all-steel equipment and fully steam-powered, were imposing even by European standards and usually larger in scale than the average European silkreeling establishments. The extremely large-scale and capital intensive nature of the Shanghai reeling factory stood in the way of its rapid diffusion and expansion in a financially strapped economy.18 If the modern reeling industry in Shanghai had trouble establishing itself beyond the city borders, it also encountered difficulty in bringing supplies of cocoons from the rural areas. Since finished raw silk was much cheaper and easier to ship than cocoons, silk reeling had always been integrated with sericulture in farm household production. Even with the rise of mechanization, modern reeling industries around the world, including Japan, were usually rural or township-based. Shanghai was not far away from the main sericultural regions of Jiangsu and Zhejiang, all roughly within a distance of two hundred miles. But political instability, primitive infrastructure and underdeveloped banking facilities in the sericultural regions increased the difficulties and costs of the reeling industry in Shanghai. Transport was by slow-moving, man-powered boats on rivers and canals.19 This brought extra risk to the silk reelers as fresh cocoons were highly perishable.20 Furthermore, the existence of numerous currencies and woefully inadequate banking facilities in the sericultural areas meant that armed boats were used to ship silver to pay farmers for cocoons. The costs of introducing pupae-killing and cocoon-drying

78

Debin Ma

equipment, and of building cocoon markets and maintaining market order posed additional burdens that seriously restricted the growth of the Shanghai reeling industry. The problem of cocoon procurement was to haunt the Shanghai reeling industry for many years. The case of the Guangdong silk reeling industry Bearing in mind the problems of the Shanghai industry, it is easier to understand the success of the Guangdong modern reeling industry. Then there was technological indigenization and geographical integration of sericulture and silk reeling. Like Japan (or unlike Shanghai), the Guangdong silk reeling industry was centred in the heart of the sericultural region. The sector’s rise owed much to the energetic efforts of a local entrepreneur by the name of Chen Qi-Yuan. Chen, an expert in Chinese traditional reeling technology, witnessed the working of the French Chambon style silk reeling machine during his business travels in Vietnam. In the early 1870s, he experimented and succeeded in making a reeling machine, which, like the Japanese reelers, enabled labour to be substituted for capital. He kept the crucial feature of the double twisting mechanism and steamheated oven, but manpower replaced steam power, wood and bamboo took the place of steel structures and, to add a Chinese touch, all-purpose Chinese chopsticks were used to pick out cocoons from the boiling basins. The modified machines turned out to be much cheaper, a blessing to the capital-scarce rural economy of Guangdong. The modern silk reeling establishments in Guangdong required only little more capital than the traditional hand reeling plants and only one-fifth of that of the modern Shanghai silk reeling factories employing an identical number of workers.21 In 1873 he opened a modern silk reeling enterprise in his hometown, Nanhai county in Guangdong province, which is now widely regarded as China’s first modern private enterprise. Later, Chen described the design of his silk reeling machine in a book, bearing a title highly reminiscent of a classical Chinese agronomic treatise.22 In this sense, Chen’s endeavour was a creative adoption of Western technology to the classical Chinese agronomic tradition. This reeling mill was an almost instant success and expanded quickly. It gave rise to a modern reeling industry, owned and operated wholly by local private initiatives in Guangdong. Modern silk reeling factories using his modified machines were quickly established in the major sericultural regions. Table 3.1 shows that, prior to the twentieth century, Guangdong’s exports of machine-reeled silk were earlier and greater than Shanghai’s, and not much below Japan’s. This was remarkable as China’s southern coast, through its early confrontation with Western imperialism, harboured the strongest anti-Western sentiments.23 These sentiments did not blind the local entrepreneurs to the opportunities for profit by introducing Western technology. Indigenous development of locally appropriate technology proved effective in bringing modern technology and the factory system to Guangdong as it had done in Japan. While Guangdong succeeded in adapting technology to local conditions, it could not escape the issue that beset private enterprises in the rest of China, the issue of

Transfer of silk reeling technology

79

legality. The problem was serious in the case of the Guangdong silk industry, because, as a rural and township-based industry, it lacked the extraterritorial protection of the Shanghai industry. Aware of the risk, the founder of the Guangdong modern reeling industry, Chen Qi-Yuan, avoided running modern factories in the large port of Canton, as their conspicuous presence would make them the easy target of governmental antagonism and local resentment. So he chose the rural Nanhai county, where he had powerful connections. This seemed to be a wise move until a vicious riot by local traditional silk weavers in 1881 destroyed several modern reeling mills and equipment. The militia put down the riot, but the local county magistrate ordered the closing of the reeling mills and the sale of the equipment. Any future operation had to come under government supervision, a major setback for the nascent Guangdong reeling industry. The reeling industry did reestablish itself several years later in nearby Shunde county.24 Although the Guangdong reeling industry, being rural-based, had avoided initially the Shanghai reelers’ problem of transporting bulky and perishable cocoons, it suffered from social chaos and local hostility. Finally, one difference set apart the Guangdong reeling industry from its counterparts in Shanghai and Japan. Western reeling technology came to Guangdong through the learning experience of an individual merchant’s trip to Southeast Asia. The French Chambon style machine was the only one Chen saw and that was the only one to which Guangdong was exposed. Guangdong merchants never saw the full range of Western technology and management under the direct guidance of European experts. The Shanghai reelers gained them through direct foreign investment, and the Japanese through governmental efforts such as the setting up of Tomioka. Both Japan and Shanghai, after a period of learning, switched to the more productive Italian Tavelle system in the late nineteenth century. Yet the French Chambon style technology persisted in Guangdong into the 1930s, even when the advantages of the Italian Tavelle system became known in the twentieth century. The irony here is that although Shanghai had an intimate technological linkage with Japan, it had had almost no technological dialogue with Guangdong, another region of China. This is a poignant reminder of the disintegrated nature of premodern China. Summary and discussion The main characteristics of the transfer of reeling technology in Japan and China are summarized in Table 3.2. Of the three regions, Japan set the standard as the best performer in all the five categories, namely, private enterprise, technological transfer, access to technology, technical localization and geographic location. Shanghai and Guangdong matched the standard in some categories but not in others. We can also consider the growth trajectory of the modern silk reeling industry in the three regions before the twentieth century by recalling those periods when critical events occurred. First, 1861, the year the Shanghai Ewo Filature was established in Shanghai, marked the beginning of modern European silk reeling

80

Debin Ma

Table 3.2 Comparative summary of silk reeling technology transfer in China and Japan 1860–95

technology in East Asia. Nine years later in 1870, that technology found its way to Japan. About the same time entrepreneurs in Guangdong introduced French silk reeling technology through the intermediary of the French colonial presence in Annam, Vietnam. Then in 1881, the year the modern silk reeling factories in Guangdong had to shut down due to the riots by local silk weavers, the Meiji Japanese government decided to switch to full scale support for private enterprise and sell all governmentowned enterprises including the Tomioka silk mills. At the same time, a new commercial code was being drafted in Japan. In 1894–5, after almost three decades of introducing modern science and technology from the West, Japan and China fought each other with newly equipped Western gunboats. The peace treaty signed by a humiliated China granted foreign enterprise legal status in the treaty ports. Ironically, this paved the way for the rise of a dynamic modern private sector in China. In 1904, the Qing government promulgated its first commercial code largely modelled on the Japanese version (Zhu 1996, p. 202). The Shanghai and Guangdong silk reeling industries experienced sustained growth during the first decades of the twentieth century as can be seen in Table 3.1 and Figure 3.2. But by then, they were competing in the world market with a fullyfledged and well-capitalized Japanese silk reeling industry with giant enterprises such as Katakura. But the growth of the Shanghai and Guangdong reeling industries did not eliminate the flaws left from the early days of technology transfer. The large-scale, capital intensive Shanghai reeling industry found that its inability to procure cocoons of consistent quality crippled its growth in the nineteenth century and impeded its expansion in the twentieth. The reeling industry in the Shanghai region did not make marked progress in technology and institutions until the late 1920s and 1930s. This was when it migrated out of Shanghai to the neighbouring provinces. By then, several decades had already been lost and world raw silk demand was already declining. Much the same could be said of the persistence of the less efficient French Chambon twisting methods in the Guangdong reeling industry in the twentieth century.25 It is clear that in China, entrepreneurship existed, technology transfer occurred, capital investment took place and initial conditions were actually more favourable than in Japan. None of these things could explain the contrasting performance of

Transfer of silk reeling technology

81

the Japanese and Chinese silk reeling industries. What really differed and mattered was the reform policies of the two governments. How and why did the roles of the Japanese and Chinese governments become so different? This is a question that is too large to be fully explored in this paper. There were major differences in the economic reforms carried out by the Meiji reformers in Japan and the leaders of the Self-strengthening movement in China in response to the Western threat. The Meiji reformers undertook a fundamental overhaul of the traditional system to build institutional and infrastructural foundations essential for a Western style market economy based on private entrepreneurship. This included property rights protection, public education, a modern communication and transportation system, technological innovation and diffusion, and a modern monetary and financial system. In contrast, the pillar of the Self-strengthening movement was in the Chinese government’s financial and managerial commitment to capital-intensive projects establishing Western style defence, mining, shipping, heavy and later some light industries. Private entrepreneurship was ignored or suppressed. The new ventures turned out to be loss-making and fraud-ridden. The Self-strengthening movement did not provide essential conditions for modern economic growth such as property rights protection, modern public education and a national network for diffusing technology. But why did reformers in these countries adopt different policies when confronted with a common challenge? One line of thought points to a different understanding and interpretation of Western supremacy. Was the Western challenge to East Asia based on military supremacy and machinery as the Chinese leaders of the Self-strengthening movement thought? Or was it derived from the broader framework of market institutions, and legal and political systems as the Meiji reformers saw it, after experiencing a decade of learning and overseas travel? The question of how to explain the strength of the West that so vexed East Asia in the nineteenth century will continue to perplex many developing nations today.

82

Debin Ma

Notes 1 The making of raw silk involved two main stages of production. Sericulture denotes the process of planting mulberry trees, feeding silkworms, and subsequently collecting cocoons spun by the silkworms. The continuous threads are then reeled off and wound on to bobbins to form the so-called raw silk. This paper focuses on silk reeling technology transfer only. 2 The treaty port clauses stipulated that all goods entering the treaty port were exempt from or subject to low and fixed import and export duties. 3 To put China and Japan on the same geographic plane is to commit the obvious fault of comparing a large country with a small one. Chinese sericultural regions, according to the exhaustive study done by the Japanese scholar Uehara in the late 1920s, could be roughly divided into four regions, each of which rivalled the whole of Japan in size and potential for raw silk production. So, this paper mainly compares Japan with China’s two most actively export-oriented regions, the rural areas surrounding Shanghai (which includes the provinces of Jiangsu, Zhejiang and Anhui) and the Guangdong province. Raw silk exports of these three regions were separately serviced by the three major treaty ports of Yokohama, Shanghai and Guangzhou (previously known as Canton). 4 Population data for China and Japan can be found in Yan et al. 1955, pp. 370–4 and Japan Statistical Association 1987, Vol. 1, pp. 48–9. 5 For silk reeling technology, see Zanier 1994, pp. 38–52 and section 2 of this paper. 6 Reeling basin is used for boiling cocoons to be reeled into raw silk. It is commonly used as a measurement for reeling machines. 7 As we will see later, Guangdong silk reeling basins were much simpler and had lower productivity per basin. So the number of basins somewhat exaggerated the growth of Guangdong silk reeling industry relative to that in Shanghai. 8 The French Chambon system is sometimes also called the double-twisting system while the Tavelle is referred to as self-twisting. The French Chambon system was limited to the use of only two reels per silk reeling basin. The Tavelle system had no such restrictions. It was more productive than the Chambon system as more reels could be added on to improve labour productivity. 9 The scale was too large and the machines too costly for the average Japanese entrepreneur. Spare parts could not be procured locally and had to be imported. Most Japanese cocoons barely met the quality standard required by the Tomioka factory. 10 The above narrative was based on Furuta 1988, chap. 2, Kiyokawa 1987, Smith 1965, pp. 54–60, Xu et al. 1990, pp. 127–31. 11 There were widespread ignorance and resistance to Westerners and Western technology. The often-cited example of farmers spreading rumours that red wine drunk by French experts at Tomioka was actually blood from female Japanese reelers was just one case in point. See Tsurumi 1990, p. 29. For local farmers’ resistance to European reeling technology, see ibid., p. 31. 12 Kiyokawa 1987, pp. 34–6. Tsurumi 1990, chap. 2 also contains a good description of the labour system at Tomioka, which acted as an effective means of technological diffusion. 13 For example, Otaka, the manager of the Tomioka silk mill pioneered the research on the raising of summer-autumn crop and also invented a new pupae stifling apparatus. See Ogura 1970, p. 8. 14 Kiyokawa 1995, ch. 6. 15 Xu et al. 1990, p. 188. 16 Some local authorities wanted these foreign factories closed down and brought this issue to Li Hong-Zhang in the 1880s, the most important government official of the Self-strengthening movement. Li bowed to foreign pressure and allowed these factories

Transfer of silk reeling technology

17 18

19 20

21 22 23

24 25

83

continued operation. The justification for his decision was that these factories were already in existence. Suzuki 1992, pp. 330–1. For the local Chinese resistance to modern silk reeling factory, see Brown 1979. The Shanghai reeling industry used more labour per reeling basin than the Italian. The lack of technological adaptation in Shanghai made an interesting contrast to the rural based traditional hand-reeling industry in the Shanghai region, which adopted technological and institutional innovations somewhat similar to those of the Japanese traditional silk reelers. In the late nineteenth century, the diverse, small-scale producers of hand-reeled raw silk in both the Shanghai region (mainly in Zhejiang province) and Japan (mainly in Gunma prefecture) found it difficult to fulfil orders placed by the increasingly mechanized European and American silk weavers. Raw silk produced by individual family producers came in small quantities and of varying sizes and types. In response, producers in both regions added an extra process, re-reeling—raw silk reeled by individual producers was gathered, sorted according to type and size and then rereeled. The re-reeled raw silk greatly enhanced the uniformity of the traditional hand-reeled raw silk, although it was still inferior to machine-reeled silk. In Japan, the re-reeling process was done through the formation of ‘joint re-reeling cooperatives’, whereas in China, it was carried out through the extension of the traditional putting out system operated by the rural silk merchants. Both systems proved viable as exports of rereeled traditional raw silk were important even in the twentieth century. But the more important point is that Chinese entrepreneurs were capable of technological adaptation under proper circumstances. See Furuta 1988, ch. 4. Steamboats were largely adopted after the 1890s. See Suzuki 1992, p. 347. The problem was partially eased by the adoption of cocoon drying facilities and the appearance of cocoon marketing centers in the main sericultural regions in the mid1870s. See Xu et al. 1990, p. 129. Chen, Yao-Min 1985. Xu et al. 1990, pp. 112–27. For the diffusion network of Chinese agricultural best practice based on the dissemination of agronomic treatise, see Gang 1993. The ambush of British soldiers in the San-yuan-li district in Guangdong province led directly to the Opium War in the 1840s and marked the beginning of a nationalist movement in China. So 1986, p. 60–1. Ibid., pp. 116–27. By emphasizing the importance of this initial pattern of technology transfer, the impression must not be left that this initial pattern was the only explanation for the differential export performance for the entire period. There were also other major developments in reeling technology and sericulture in Japan and China in the twentieth century.

References Brown, Shannon R. (1979) ‘The Ewo Filature: a study in the transfer of technology to China in the nineteenth century.’ Technology and Culture, July 1979, Vol. 20, No. 3, pp. 550–68. Federico, Giovanni (1997) An Economic History of the Silk Industry, 1830–1930, Cambridge: Cambridge University Press. Furuta, Kazuko (1988) ‘Technology transfer and local adaptation: the case of silk-reeling in modern East Asia’, a dissertation submitted to Princeton University. Gang, Deng (1993) Development vs. Stagnation, Westport, Connecticut: Greenwood Press. Japan Statistical Association Historical Statistics of Japan, Vols. 1 and 3, Tokyo: Japan Statistical Association, 1987.

84

Debin Ma

Kiyokawa, Yukihiko (1987) ‘Transplantation of the European factory system and adaptations in Japan: the experience of the Tomioka Model Filature’, Hitotsubashi Journal of Economics 28, pp. 27–39. Ma, Debin (1998), ‘The great silk exchange, globalization of trade technological diffusion in historical perspective’, PhD dissertation submitted to Department of Economics, University of North Carolina at Chapel Hill. Ogura, Takekazu (ed.) (1970) Agricultural Development in Modern Japan, Tokyo: Fuji Publishing Co. Smith, T.C. (1965) Political Change and Industrial Development in Japan: Government Enter-prise 1868–1880, Stanford: Stanford University Press. So, Alvin Y. (1986) The South China Silk District: Local Historical Transformation and World-System Theory, Albany: State University of New York Press. Tsurumi, E.Patricia (1990) Factory Girls: Women in the Thread Mills of Meiji Japan, Princeton, New Jersey: Princeton University Press. Twitchett, Denis and Fairbank, John K. (1980) The Cambridge History of China, Cambridge: Cambridge University Press. Wong, Chor Yee (1995) ‘Proto-industrialization and the silk industry of the Canton Delta, 1662–1934’, unpublished dissertation at the University of Wisconsin, Madison, 1995. Zanier, Claudio (1994) Where the Roads Meet, East and West in the Silk Production Processes (seventeenth to nineteenth century), Kyoto: Institute Italiano di Cultura, Scuola di Studi sull’ Asia Orientale.

Literature in Japanese Fujino, Shozaburo, Fujino, Shiro, Ono, Akira (1979) Estimates of Long-term Economic Statistics of Japan Since 1868, Vol. 11, Textiles. Tokyo: Toyo Keizai Shinposha. Ikawa, Katsuhiko (1992) ‘Seishi-gyo to America Sijo’ (Raw Silk Sector and the American Market) in Naosuke Takamura (ed.) Kigyo Bokko: Nihon Shihonshugi No Keisei (The Rise of Firms: the Formation of Japanese Capitalism), Kyoto: Mineruva Shobo. Ishii, Kanji (1972) Nihon Sanshi-gyo Shi Bunseki (Analysis of Japanese Silk Industry), Tokyo: Tokyo Daigaku Shuppankai. Kiyokawa, Yukihiko (1995) Nihon No Keizai Hatten To Gijutsu Fukyu (Japanese Economic Development and Technological Diffusion), Tokyo: Toyo Keizai Shinposha. Nishikawa, Shunsaku, Abe, Takeshi (eds) (1990) Sangyoka no Jidai (1) (Times of Industrialization, Vol. 1), Tokyo: Iwanami Shoten. Suzuki, Tomoo (1992) Yomu Undo No Kenkyu (A Study of the Westernization Movement in China in the Latter Half of the Nineteenth Century), Tokyo: Kyuko Shoin. Uehara, Shigemi (1929) Shina Sanshi-gyo Taikan (Overview of the Chinese Silk Sector), Tokyo: Okata Nichiei Dou.

Literature in Chinese Chen, Zi-Yu (1989) Jindai Zhongguo de Jijie Saosi Gongye, 1860–1945 (The Silk Industry of Modern China, 1860–1945), Taipei, Taiwan: Institute of Modern History, Academia Sinica. Chen, Yao-Min (1985) ‘Qing mu shunde jiqi cao si ye de cai sheng, fa zhan jiqi ying xiang’ (The beginning, development and significance of machine-reeling industry in late-Qing Shunde) in Ming qing guangdong shehui chinji xingtai yanjiu (Studies on

Transfer of silk reeling technology

85

the Socio-economic Conditions of Ming and Qing Guangdong), edited by the Historical Association of Guangdong, Guangdong: Guangdong People’s Publishing House. Xu, Xin-Wu (ed.) (1990) Zhongguo Jindai Saosi Gongyeshi (Modern History of Chinese Silk-reeling Industry), Shanghai: People’s Publishing House, 1990. Yan, Zhong-Ping et al. (eds) (1955) Zhongguo jin dai jing ji shi tong ji zi liao xuan ji (Selected Statistics on Modern Chinese Economic History), Shanghai: Science Publishing Co. Zhu, Ying (1996) Wan qing jing ji zheng ce yu gai ge cuo shi (Late-Qing Economic Policy and Reform Measures), Central China Normal University Press.

4

The colonial origins of Korea’s market economy* Myung Soo Cha

Rudimentary markets ‘embedded’ in social relations have long existed in Korea, but it has been only during the last 100 years that markets came to the fore to encompass social relations. How did this ‘great transformation’ come about?1 This paper argues that South Korea’s contemporary market economy has its origin in the colonial period (1910–45). Not only did goods and labour markets become integrated in this period, but also a repressed and dual capital market first appeared in the final years of Japanese rule. Market development entails increasingly efficient arbitrage, which raises the degree of parallelism in the variation of prices observed in different regions and lowers inter-regional price differentials. This paper uses inter-regional and intersectoral price and wage differential evidence to outline the course of the evolution of national markets for goods and labour in Korea. Price and wage differentials, as measured by coefficient of variation, are preferred to price correlation as an index of market integration, because the latter may reflect the incidence and intensity of aggregate shocks as well as the degree of market integration (Cha 1997). To chart the evolution of the capital market, this paper relies on the idea that improving financial intermediation lowers interest rates, as the coming of railways reduces goods prices by shifting the supply schedule to the right. Goods, as represented by the rice market, are dealt with in the first section. The second section turns to labour market integration, and the third section discusses the evolution of the capital market in Korea. Each of these three sections includes some international comparisons. The final section summarizes and concludes.

Goods market: rice Rice has long been one of the most important food items in Korea, and it was also widely used (together with cotton textile) as a means of payment before the modern *

This project was supported by the Institute of Korean Culture, Yeungnam University. I am grateful to Sungho Chun, Bill Collins, Insong Gill, Sooyeol Huh, Jongil Kim, A.J.H. Latham, Hunchang Lee, Younghoon Lee, and the participants at the Economic History section of 41st Korean Historical Society Meeting and at section C56 of the 12th International Economic History Congress.

86

Colonial origins of Korea’s market economy

87

Figure 4.1 Coefficient of variation: Korean rice prices 1744–1995. Source: 1744–61 and 1779–1804: Chun (1996a, 1996b, 1997); 1890–1907: Kimura and Uranagase (1987); 1914–38: Chosen Sotokufu, various years; 1964–95: Economic Planning Board of Korea, various years.

currency system was introduced with Japanese colonial rule in the 1900s. The rice market is likely to have emerged earlier than markets for other commodities, a conjecture supported by the fact that rice prices provide the longest available time series data in Korea. Figure 4.1 shows the coefficient of variation calculated using rice prices observed in two places, one located in the southwestern (Cholla) and the other in the southeastern (Kyongsang) region of the Korean peninsula from the mid-eighteenth century to the present. This long period consists of five shorter disconnected periods: 1744–61, 1779–1804, 1890–1907, 1914–38, 1964–95. Places vary from one sub-period to another, but the distance between the two places remains around 200 km (125 miles) throughout. The only exception is from 1890–1907, when rice prices in the central–eastern region were substituted for those in southwestern regions, for which data is not available. Another difficulty with this sub-period is that the prices used are those observed in the treaty ports. Such prices are likely to overstate the degree of market integration. Because these prices are likely to have been influenced by rice prices in Japan (the predominant rice market for Korea in this period), coefficients of variation of treaty port prices would be lower than those calculated using rice prices observed in different inland regions. Figure 4.1 also shows trends of the coefficient of variation by sub-period, fitted by ordinary least squares regression. Although the slope of the time trend is not significantly different from zero at 5% level, the direction of the slope in the first three sub-periods suggests several episodes in pre-modern Korean monetary history.

88

Myung Soo Cha

In the mid-eighteenth century Korea was frequently buffeted by adverse regional climatic shocks. As a means of relieving famine, central government granted provincial governments the privilege of minting bronze coins to pay for grain imports from other regions. Hence regional productivity shocks led to regional monetary shocks, which widened price differentials between regions. Weather became more favourable in the late eighteenth and early nineteenth century, so there were less frequent and weaker regional monetary shocks (Chun 1996a and 1997; Palais 1996, ch. 25). The price differential decreased in 1779–1804. Finally, during the third sub-period (1890–1907) the coefficient of variation first rose and then fell, showing the impact of fiscal and monetary reforms imposed by Japan in the early 1900s. The reforms put an end to fiscal deficits by separating royal household and government budgets and introducing new taxes. This fiscal rectitude enabled the Japanese-controlled government to issue new coins in exchange for debased coins. The level of coefficient of variation in dynastic Korea was about 0.2. How does this compare with those in other countries? Surprisingly, as early as the mid-seventeenth century, Tokyo and Nagoya rice markets (about 200 km apart) appeared better integrated than in pre-modern Korea and as well integrated as in modern Korea, with the coefficient of variation fluctuating around 0.05. On the other hand, the coefficient of variation of rice prices in Tokyo and Aizu (again about 200 km apart) was about 0.3, which was much higher than between Tokyo and Nagoya. This is in line with Miyamoto’s (1981) conclusion based upon price correlation evidence that Tokugawa rice markets remained regionally segmented. Regions along the bakufu-controlled Tokyo-Osaka trade route seemed more closely integrated between themselves than with the rest of Tokugawa Japan. Coefficients of variation for the two pairs of cities display trends moving in opposite directions in the following two centuries, rising slowly in Tokyo-Nagoya and falling in Tokyo-Aizu. By the early 1840s, coefficients of variation for the two pairs of cities nearly converged at 0.2, indicating the emergence of a national market as domains (hans) began to infringe upon the bakufu trade monopoly (Crawcour 1989, pp. 587–600). The coefficient of variation rose sharply for both pairs in the 1840s and 1850s, only to decline sharply below 0.05 after the arrival of Admiral Perry.2 The coefficient of variation of rice prices observed in two cities in the Yangzi delta, Suzhou and Hangzhou (again about 200 km apart), ranges from 0.1–0.2 in the early eighteenth century (Chuan and Kraus 1975, ch. 3). This is lower than in Korea, but similar to that between Tokyo and Nagoya in the early eighteenth century. Grain markets remained far less developed in Gansu, a landlocked frontier region in the far west. The coefficient of variation for Gansu millet prices was as large as 0.5 in the mid-eighteenth century, which with the end of famine and military campaigns fell rapidly to 0.2 by the time of the Opium War (Perdue 1992, p. 124). As in both Korea and Japan, the opening of the ports following the war provided a powerful stimulus to market integration, while the Taiping Rebellion in the following decade dealt an enormous blow to market activities. In Indian rice and wheat prices, the coefficient of variation was as high as 0.4–

Colonial origins of Korea’s market economy

89

0.5 until the mid-nineteenth century, then fell to 0.1–0.2 half a century later as a result of the building of the railways (Hurd 1975). The coefficient of variation of wheat prices in Eton College and Exeter (again about 200 km apart) in the seventeenth century tended to fluctuate within 0.1–0.2 (Mitchell 1988). The coefficient of variation fell rapidly during the first half of the eighteenth century to reach 0.05 on the eve of the Industrial Revolution. The time trends in Figure 4.1 (dotted lines) convey an impression that price differentials in Korea did not fall smoothly during the past two and half centuries, but in two downward shifts, the first one sometime in the nineteenth century and the second one in the early twentieth century. The first shift may seem apparent rather than real, as most of the nineteenth century is not represented in Figure 4.1. But it probably did occur, as the nineteenth century is known for recurrent peasant uprisings and fiscal crises. They led to debasement, inflation, and fragmented currency areas, where different types of debased coins circulated. This political and economic chaos may have discouraged market activities and widened price differentials. What caused the first sharp contraction of the price differential was probably the port-opening in 1876 imposed by Japan, which allowed foreign merchants to travel around the country and introduced modern communication and transportation technology to Korea. Japan annexed Korea in 1910, after defeating her two main competitors in the Sino-Japanese and Russo-Japanese wars. Integration into the Japanese empire gave a further stimulus to market evolution by expanding railways, telegraphs and telephone lines, and establishing market institutions including land property rights and a modern currency system. In short, Japanese colonization was of critical importance in the evolution of internal rice markets. Japan’s invasion (1592–98) and the wars with Qing China (1627 and 1636) had also expanded markets dramatically in dynastic Korea. These wars caused an ‘emergency conversion’ from a tribute to a market economy by destroying the bureaucratic machinery of the command system. This is shown by the fact that the government began to mint and circulate bronze coins from 1678.3 Labour market In dynastic Korea (i.e. Korea before 1910) both free and slave labour existed side by side. Around one-third of the rural population could be bought, sold or inherited, which led Palais (1996) to describe dynastic Korea as a ‘slave society’. Free peasants, although poor, were not tied to the soil by the landowning élite as in medieval Europe. Given paucity of quantitative information, it is difficult to know how efficiently slave markets functioned or how well regional labour markets were integrated before 1910. Regional wage data collected at the beginning of Japanese rule provide a window on traditional Korean labour markets. Figure 4.2 shows that in the 1910s the coefficients of variation of nominal wages of unskilled workers in different parts of the southern half of the Korean peninsula were about 0.4. This implies an inter-regional nominal wage gap substantially larger than in Meiji Japan. In the 1880s, the coefficient of

90

Myung Soo Cha

Figure 4.2 Coefficient of variation: unskilled nominal and real wages in southern Korea 1915–90. Source: 1914–38: Chosen Sotokufu, various years; 1954–63: Bank of Korea, various years; 1985–91: Korea Institute of Labor Research (1996). Notes: solid and dashed lines represent, respectively, the coefficient of variation of nominal and real wages; regional real wages were calculated by deflating regional nominal wages with regional rice prices; regions include Seoul, Taegu, Pusan, Mokpo until 1962 and Seoul, Taegu, Pusan Kwangju in 1984–90; Kwangju and Mokpo are about 70 km apart; unskilled labour is represented by male servants until 1937 and by coolies after 1954.

variation of unskilled wage in Tokyo, Osaka, and Kyoto remained between 0.1– 0.2 (Nihon Teikoku Tokei Nekan, various years).4 The nominal wage gap in Korea in the 1910s was also greater than in pre-World War I colonial India. The average coefficient of variation of nominal wages for agricultural workers in six regions of Bombay district was 0.25 from 1873–1906.5 In the mid-nineteenth century, the coefficient of variation of unskilled wages in different regions of both France and Sweden was slightly larger than 0.2 (Boyer and Hatton 1994). The coefficient of variation of wages received by both artisan and unskilled workers between four regions in antebellum USA fluctuated around 0.1 and 0.15, respectively.6 What about the real wage gap? The real wage gap may be greater or smaller than the nominal wage gap depending on how the nominal wage and the cost of living index are correlated across regions. When the two nominal variables are positively correlated, the coefficient of variation of nominal wages will be larger than that of real wages. When the variables are negatively correlated, the coefficient of variation of nominal wages will be smaller. When nominal wage and price levels vary independently of each other across regions, the two coefficients of variation are not likely to differ significantly. Cross-sectional positive correlation will result only if all regional shocks generating price and wage differentials come from the demand side of the economy. Negative correlation will result if regional shocks

Colonial origins of Korea’s market economy

91

are supply shocks.7 When shocks are equally likely to be supply or demand shocks, nominal wage and price levels vary independently of each other across regions, and the coefficient of variation of nominal and real wage should be roughly comparable. For the lack of a better regional cost of living index, regional real wages are derived by deflating regional nominal wages with regional rice prices. The coefficient of variation of real wages obtained is shown in Figure 4.2 alongside the coefficient of variation of nominal wages. The proximity of the two series indicates that regional shocks were equally likely to be generated by the demand and supply sides. Assuming regional shocks in other countries in the pre-industrial stage were also not predominantly from either supply or demand side, the larger coefficient of variation of nominal wages in 1910s Korea would imply a higher real wage gap in Korea.8 Although labour was sufficiently mobile for the terms of trade shock during the treaty-port period to cause both specialization of agriculture towards export commodities (including rice and beans) and the decline of handicraft industry (Kajimura 1968; Miyajima 1974), labour markets in early twentieth-century Korea remained characterized by a high degree of regional segmentation. Figure 4.2 shows that this unusually large inter-regional wage dispersion in Korea shrank rapidly during the colonial period. Only from the second decade of Japanese colonial rule did a national market for unskilled labour begin to emerge. The steady decline in the unskilled wage gap among different regions also suggests that the Lewisian model of unlimited labour supply is not applicable to colonial Korea.9 Figure 4.3 shows the coefficients of variation of nominal and real wages of bricklayers, chosen to represent skilled workers. As with unskilled labour, coefficients of variation of nominal and real wages remain close to each other throughout the period. The coefficient of variation of skilled wages, which was as high as that of unskilled wages in the early 1910s, fell rapidly a decade later to a low of about 0.1, a level maintained in the following seven decades. This is in contrast to unskilled labour markets, which became integrated in a more gradual manner during the colonial period. Skilled labour markets also emerged before unskilled labour markets in both nineteenth-century England and Meiji and Taisho Japan (Hunt 1981, p. 155; Boyer and Hatton 1994, p. 94, Nihon Teikoku Tokei Nenkan, various years). Being both better educated and better informed, skilled workers tend to be more mobile than unskilled workers. What is unusual about Korea is the compression of the unskilled wage gap in the early stage of economic development (Figure 4.2). In the late nineteenth and early twentieth century inter-regional nominal wage gaps did not fall in Bombay district. Although grain markets were being rapidly integrated due to the building of the railways (Hurd 1975), this did not lead to real wage convergence (Collins 1997). Regional unskilled wages in France and Sweden did not converge from 1860– 1910 (Boyer and Hatton 1994). In the USA, substantial wage gaps existed and were sustained between regions and even among different areas in the same region until the late nineteenth century.10 In Britain, nominal wages of farm labourers in different regions did not converge during the Industrial Revolution and the following half

92

Myung Soo Cha

Figure 4.3 Coefficient of variation: nominal and real wages of bricklayers in southern Korea 1914– 90. Source: 1914–38: Chosen Sotokufu, various years; 1954–63: Bank of Korea; 1985–91: Korea Institute of Labor Research (1996). Notes: solid and dashed lines represent, respectively, the coefficient of variation of nominal and real wages; regional real wages were calculated by deflating regional nominal wages with regional rice prices; regions include Seoul and Taegu.

century: the standard deviation of farm labourers’ wages county to county did not change from 1767–1870 (Hunt 1986, p. 951, Table 1). The coefficient of variation of wages between regions remained in the range of 0.11–0.14 during 1860–1910.11 Outside Korea the convergence of unskilled nominal wages occurred during the transition to modern economic growth only in Prussia and Meiji Japan. The coefficient of variation of unskilled wages fell from 0.23 in 1870 to 0.17 in 1910 in Prussia, and in Japan from around 0.2 in the 1880s to 0.1 in the early 1920s (Boyer and Hatton 1994; Taira 1970, Nihon Teikoku Tokei Nenkan, various years). Japan, Korea, and Prussia were later-industrializers undergoing relatively faster growth and structural change, driven by productivity advances, which were unbalanced probably to a greater extent than in earlier industrial revolutions. The larger difference in sectoral rates of technological progress led to wider inter-regional wage gaps, providing stronger incentives for workers to move. Moreover, coefficients of variation cited above seem to indicate that pre-industrial labour markets in both Britain and USA were already better integrated than in the lateindustrializers. Song (1991) has argued that the ‘authoritarian’ state in post-colonial South Korea developed a labour market functioning like a commodity market by repressing the growth of trade unions.12 Similarly, autocratic political systems in Prussia, Meiji Japan, and colonial Korea aided labour market integration by preventing workers from acting together.

Colonial origins of Korea’s market economy

93

Figure 4.4 Sectoral nominal wage coefficient of variation 1914–90. Source: 1914–40: database compiled and maintained by Institute of Economic Research of Hitotsubashi University; 1970–90: Bank of Korea, various years.

As labour markets mature, regional and sectoral wages will equalize. Figure 4.4 shows the coefficient of variation of wages in different nonagricultural sectors.13 During the colonial period, wage convergence is observed only among industries drawing largely upon unskilled labour, including spinning, food processing, and construction. It is not seen among industries where skilled workers predominate, such as metal, machinery, transportation, and woodworking. Non-convergence among skill-intensive industries may be explained by industry-specific skills, which impede intersectoral labour mobility. The coefficient of variation of nominal wages in all sectors (not shown in Figure 4.4) tends to rise during the colonial period, much like the coefficient of variation calculated among the skilled labour-intensive sector. For the post-1970 decades, the coefficient of variation over all nonagricultural sectors is shown in Figure 4.4, because the industry classification for this period does not allow one to separate skill-intensive sectors from those using mainly unskilled labour. In contrast to the colonial era, nominal wages in different industries were equalized in the 1970s and 1980s. This is probably another evidence of the fluidity of the post-colonial South Korean labour market, where Virtually all of manufacturing’s growth was accompanied by hiring ever higher numbers of new entrants to the labour force’ (Kim and Topel 1995, p. 229).

94

Myung Soo Cha

Capital market In pre-modern economies, high nominal rates of interest ranging from 30–40 per cent per year normally prevail in rural areas, while the rate of return to riskless capital is about 10 per cent (Perkins 1969, p. 92; McCloskey and Nash 1984, p. 179; Clark 1989). The rate of return to capital began to fall from this high level first in the Netherlands after 1600 and then in England after 1700, declines which were delayed in other parts of Europe until after 1800 (Pressnell 1960; de Vries 1976, pp. 211–12; Clark 1988; Hart 1997). What caused the rate of interest to fall in early modern northwestern Europe— rightward shifts in the savings supply schedule or leftward shifts in the investment demand schedule or both? While demand shifts represent changing productivity of capital, resulting from both labour force growth and technological progress, prominent causes of supply shifts include changes in the dependency ratio and financial development.14 To simplify, possible causes of the falling interest rate are demographic, technology shocks, and financial development. Of these three forces one may reasonably discount the role of productivity shocks on investment demand in early modern England and Holland. Population increased in both seventeenth-century Holland and eighteenth-century Britain. This would suggest that the rate of interest declined because the population growth expanded savings supply. It thus lowered the dependency ratio faster than it boosted investment demand by labour force growth. One objection to this line of explanation is that population grew during the eighteenth century in Britain and in many other parts of Europe, while the interest rate fell only in Britain. It appears more likely that the Dutch and British interest rates fell because savings supply increased following ‘financial revolutions’, which was a consequence of enhanced government credit and political stability after civil revolutions (Clark 1996; Hart 1997; Hicks 1969, p. 94; North 1990, p. 43; North and Weingast 1989, pp. 819–28). The improvement in financial intermediation lowered interest rates in other parts of the world in different periods by shifting savings supply schedule to the right. Davis and Gallman (1994) attribute falling rates of interest in antebellum USA to financial development. Shimbo (1978, pp. 233–43) presents evidence of falling interest rate in late eighteenth-century Japan, which may well have been a consequence of the development of the credit system as documented by Crawcour (1961) and Ohkura and Shimbo (1978). In the Korean countryside, the nominal rate of interest hovered around an average of 36 per cent per year from the mid-eighteenth to the mid-nineteenth century, and rose to as high as 60 per cent during the latter half of the nineteenth century (Chun 1998). This rise probably reflected expectations of inflation generated by the large-scale debasement, which was introduced to deal with deepening fiscal crises during the mid-nineteenth century. Despite fiscal and monetary reform programs carried out under Japanese control in the 1900s, moneylenders were still charging interests at rates as high as 40–50 per cent per year at the beginning of colonial rule (Chosen Sotokufu, various years). Such high pre-

Colonial origins of Korea’s market economy

95

Figure 4.5 Real interest rates and return to capital in Korea 1912–1991. Source: Chosen Sotokufu, various years; Bank of Korea, various years; Pyo (1998). Notes: solid line: real interest rate (black market); dotted line: real interest rate (bank loan); solid line with triangles: return to capital.

modern rates of interest co-existed in the 1910s with much lower rates of interest (11–14 per cent) on bank loans. The financial system remained divided into unregulated (traditional or informal) and regulated (modern or formal) sectors, largely because newly established modern financial institutions were not sufficiently well developed to penetrate traditional markets where small- and medium-sized firms and households predominated.15 Figure 4.5 shows that ex post real interest rates in the regulated and unregulated financial sectors remained in the range of 10–20 per cent and 30–40 per cent per year in the 1910s and 1920s respectively, and that both declined in the 1930s. The decline of interest rates in the 1930s reflected increased savings supply rather than slack investment demand: 1930s Korea saw an explosion of investment activities in nonagriculture stimulated by rising Japanese militarism and recovery from the Great Depression. The expansion of savings supply did not seem to have much to do with either improvement in financial intermediation or to the falling dependency ratio. The financial intermediation ratio, after rising up to 1932, fell during the following six years.16 Also, the inter-regional moneylender interest rate differential did not close in inter-war Korea (Chosen Sotokufu, various years), and the falling mortality and rapid population growth raised the dependency ratio slightly.17 What probably caused the interest rate to fall was a surge of capital inflows from Japan to finance a politically motivated industrialization drive in the northern part of the colony. The ratio of current account deficits to gross domestic expenditure doubled from 3 per cent in 1918–31 to 6 per cent in 1932–38.

96

Myung Soo Cha

The division of Korea into industrial North and agricultural South after World War II and the destruction during the Korean War (1950–53) caused the South Korean capital/labour ratio to fall sharply, raising the real interest rate back to a level comparable to that in the 1910s. In the regulated sector, the real interest rate (as represented by the bank loan rate) fluctuated around zero, a consequence of low interest rate policy aimed at stimulating investment and growth. The financial repression generated excess demand for credit, which had to be met in the unregulated sector. This caused the real interest rate in the unregulated sector (represented by the black market rate) to shoot up as high as 40–50 per cent per year in the early 1960s. The interest rate differential between the unregulated and regulated sectors came to exceed that in the 1910s, when a more liberal financial regime prevailed. It is difficult to calculate the level of the competitive general equilibrium rate of interest in a dualistic financial system. Surely, the hypothetical rate should lie somewhere between bank loan and moneylender interest rates, and its movement is better reflected in the black market, rather than in the bank loan rate controlled by the government. The real black market rate was on the decline in the post-1960 decades (Figure 4.5). The rate of return to capital, as shown in the figure, also fell following closely the black market rate in 1970–91, suggesting that the latter— although most likely an overestimate—may be used as a proxy for the hypothetical equilibrium rate of interest. The declining real interest rate in the post-1960 decades was a net effect of various exogenous shocks shifting investment demand and savings supply schedules. These included changing dependency ratios, labour input growth, total factor productivity growth, foreign savings inflow, and the development of financial intermediation. To assess the impact of these changes on the interest rate, an equation is estimated, where the real interest rate (proxied by the real black market rate) is the dependent variable and shifters in the savings-investment market are explanatory variables. The regressors include the foreign savings inflow, labour input growth, the dependency ratio, and the financial intermediation ratio, but exclude total factor productivity growth, which invariably turned out to be insignificant in different regressions using different estimates of the variable.18 The result using ordinary least squares method for the period from 1965–90 is: RRt=21.22-0.84 FSt-0.59 LFt+0.74 DRt-7.45 FIRt (0.68) (2.29)

(1.38)

(1.13)

(2.94)

adjusted R2=0.77; DW=1.14 Notes: RR—real black market interest rate; FS—current account deficits divided by GNP; LF— labour force growth rate; DR—dependency ratio; FIR—financial intermediation ratio. Parenthetically shown are t-ratios.

All estimated coefficients are of correct signs except that associated with labour force growth. The only significant variables are the foreign savings inflow and the financial intermediation ratio. The latter drove most of the decline in the real interest rate. Other variables remaining constant, post-1960 financial development as

Colonial origins of Korea’s market economy

97

reflected in the financial intermediation ratio would have accounted for nearly fourfifths of the estimated fall in the real interest rate.19 After increasing in the late 1960s and the early 1970s, the financial intermediation ratio declined slightly from 1972–78 and then started to increase again (Patrick 1994, p. 326, Fig. 8.1). This indicates that financial development appeared to suffer a setback in the 1970s, which is corroborated by Cho’s (1988) finding that variance of costs of capital facing different types of borrowers started to fall only after 1980. The turning points in the financial intermediation ratio fit nicely with wellknown regime shifts in recent South Korean financial history. Interest rate reforms in 1965 raised ceiling rates for time deposits and bank loans. There was a return to a low interest rate policy and government control of credit allocation from the early 1970s. Finally there was financial deregulation from the early 1980s onwards (Cho 1990; Cole and Park 1983 and Park 1994). Government policy towards financial markets appeared to have an important impact upon the development of capital market in the post-colonial period. South Korea’s government oscillated between financial repression and liberalization, as its development policy swung between import substitution and a more market-conforming strategy. The country was not alone in starting the postwar period with import substitution, using financial repression as a central policy instrument. Low interest rate policies and intervention in credit allocation were introduced in Latin American countries as well (Cardoso and Helwege 1995). It was widely believed that channelling of low interest rate loans to strategic sectors would foster growth, a view which seemed justified by the Great Depression, Keynesian economics, and Soviet experiment (Krueger 1990).20 Apart from the impact of post-war interventionist ideology, financial repression in South Korea had its roots in the command economy placed over the Japanese empire during World War II. An important building block of the system included interest rate control and credit rationing, introduced after the Manchurian incident in 1937. Although the war ended and the Japanese left Korea in 1945, colonial mandarins remained much in control, surviving the post-colonial campaign to purge collaborators. They would not let the system of command go, because it could conveniently be recycled for the cause of import substitution industrialization.21 When the import substitution went to the wall and the ensuing political crisis in the early 1960s enforced stabilization, the pendulum swung back towards a more market-oriented regime, which included financial liberalization in 1965 (Haggard, Kim and Moon 1991). Another war intervened to help to put an end to the liberal regime. As the USA withdrew from the Vietnam War in the early 1970s, Park Chung Hee’s confidence in USA security assurances declined. His response was to ‘become self-reliant’ by creating heavy and chemical industries capable of producing war materials. Park, who served as an officer in the Japanese Army during World War II, reintroduced financial repression as a central instrument in this second import substitution episode. Distortions mounted and inflation worsened in the late 1970s, provoking another political crisis and prompting the stabilization measures of the 1980s (Haggard and Moon 1990).

98

Myung Soo Cha

In sum, unlike markets for goods or labour, not only did a capital market fail to take root under the Japanese rule, but its development was impeded by intervention in financial markets during the World War II period. The legacy of repression persisted after independence, although disrupted by periods of liberalization. The net effect of the policy cycle turned out to be positive in terms of financial development. Financial deepening achieved during the liberalization phase had permanent components which could not be undone during the subsequent reversion to repression. Summary and conclusions Grain markets in traditional Korea probably were less developed than in advanced parts of Qing China and Tokugawa Japan, let alone early modern England. Japan powerfully expanded the sphere of markets first by forcing Korea to open to external trade in 1876 and then by incorporating the country into its empire in 1910. The Japanese introduced modern transportation, communication technology, and fiscal and monetary systems. Although market activities proliferated in the four decades after port-opening, labour markets displayed a high degree of regional segmentation before 1910. Regional labour markets became rapidly integrated during the half century following the annexation by Japan, aided by new transportation and communication technology and autocratic colonial rule which prevented workers from combining. Industry-specific skills stood in the way of wage equalization across different sectors during the colonial period, but it was helped by high rates of labour input growth in post-1960 decades. Colonial government found capital markets more difficult to develop. The interwar financial system was characterized by dualism, with modern financial institutions remaining islands in a sea of traditional moneylenders dealing with small producers and households. The dualism deepened as a consequence of financial repression during World War II. Although government intervention persisted in post-colonial South Korea, liberalization intervened to encourage financial development. All in all, goods markets appeared to evolve first in Korea after the ports were opened in 1876, and this was followed by labour market integration after 1910 when the country was incorporated into the Japanese empire. Development of the capital market was delayed until after independence from colonial rule. While goods and labour markets were brought to Korea by colonialism, this cannot be said for the capital market. On the contrary, the colonial past—more specifically the Japanese war economy—appeared partly responsible for the persistence of government intervention in post-colonial financial markets, and this hampered financial development after independence.

Colonial origins of Korea’s market economy

99

Notes 1 2 3 4 5 6 7

8 9 10 11 12 13 14 15

16 17

18 19 20

21

The expression ‘great transformation’ is borrowed from Polanyi (1957). Tokugawa rice price series are available from Iwahashi (1981). Palais (1996). ‘Emergency conversion’ is a term first introduced by Bohannan and Galton, quoted in Jones (1993). For evidence on labour market integration in the Tokugawa period, see Saito (1978). I am grateful to William Collins for providing me with Indian regional wage series. Calculated from wage data provided in Margo and Villaflor (1987). Consider an economy, where prices are flexible, but nominal wages are less so. A negative demand shock will lower price level, raising both real wages and unemployment rate, which in turn will put downward pressure upon nominal wages. Hence the conclusion that demand shocks produce positive correlation between prices and nominal wages. On the other hand, a negative supply shock will raise price level, which alone will put upward pressure upon nominal wages. This is however outweighed by the downward impact due to the increase in unemployment as the supply shock shifts the labour demand schedule to the left. In Bengal district, average coefficient of variation of real wages for agricultural workers was 0.21 in 1873–1906, which is comparable with the above cited average coefficient of variation of nominal wages, 0.25. Taira (1970) made the same point about Meiji Japan. Rosenbloom (1990), Goldin and Margo (1992), Margo and Villaflor (1987), and Slaughter (1995). Boyer and Hatton (1994). Coefficients of variation from county level wage data are somewhat higher, fluctuating in the range of 0.15–0.18. See also Hunt (1986). A similar view is expressed in World Bank (1993, ch. 6). See also Kim and Topel (1995). All sectors being nonagricultural, the coefficients of variation of sectoral nominal and real wage are identical. Dependency ratio is the share of non-working young and elderly persons in the total population. Moneylender rates remained below bank loan rates in inter-war Japan, which led Teranishi (1994) to explain the presence of moneylenders in terms of the inability of modern financial institutions adequately to process information generated by agents of traditional types. Mizoguchi and Umemura (1988, p. 305). Financial intermediation ratio is a standard index of financial development and defined as the ratio of financial assets to national income. Another contrast with inter-war Japan, where coefficient of variation of regional moneylender rates were not only much lower than in Korea, but also declined as a matter of trend. Also, interest rate differential was smaller among moneylenders than among banks in Japan. See Teranishi (1994) and Lewis and Yamamura (1971). I am grateful to Professor Jongil Kim for kindly making his labour input series available. Dependency ratio and foreign savings ratio are from Economic Planning Board of Korea, various years and Park (1994), respectively. When the same specification was estimated for the colonial era, coefficients turned out either insignificant or of wrong signs. In contrast, nineteenth-century economic development in the USA and European countries was guided by a laissez-faire belief, which explains why more liberal financial regimes evolved in developed parts of the world. Even Japan had liberal financial markets with easy entry and market-determined interest rates before wartime control began in 1937 (Patrick 1984). Ueda (1993) finds the origin of post-1945 intervention of the Japanese government in financial markets in the war economy.

100

Myung Soo Cha

References Bank of Korea Kyongje Tong’gye Yonbo (Yearbook of Economic Statistics of the Republic of Korea), various years, Seoul: Bank of Korea. Bodenhorn, Howard and Rockoff, Hugh (1992) ‘Regional interest rates in antebellum America’, in Claudia Goldin and Hugh Rockoff (eds) Strategic Factors in Nineteenth Century American Economic History, Chicago: University of Chicago Press, 159–87. Boyer, G. and Hatton, Timothy (1994) ‘Regional labour market integration in England and Wales, 1850–1913’, in George Grantham and Mary MacKinnon (eds) Labour Market Evolution, London: Routledge, 697–734. Cardoso and Helwege (1995) Latin America’s Economy: Diversity, Trends, and Conflicts, Cambridge, Mass.: MIT Press. Cha, Myung Soo (1997) ‘Integration and segmentation in markets for rice and wheat, 1877–1994’, paper presented to the 3rd World Cliometric Congress, Munich. Cho, Yun Je (1988) ‘The effect of financial liberalization on the efficiency of credit allocation: Some evidence from Korea’, Journal of Development Economics 29, pp. 101–10. Cho, Yun Je (1990) ‘The financial policy and financial sector development in Korea and Taiwan’, in Jene K.Kwon (ed.) Korean Economic Development, Westport: Greenwood, 101–10. Chosen Sotokufu (various years) Chosen Sotokufu Tokei Nempo (Statistical Yearbook of the Government General of Korea) Seoul: Government General of Korea. Chuan, Han-sheng and Richard A.Kraus (1975) Mid-Ch’ing Rice Markets and Trade: An Essay in Price History, Cambridge, Mass.: Harvard University Press. Chun, Sungho (1996a) ‘18 Segi Miga Chui wa Migok Jongchaek e kwanhan Yongu’ (A study on rice price and policy in eighteenth century Korea), Sahak Yon’gu 51, pp. 115– 210. —(1996b) ‘1725–1761 nyon’gan Kyongsangdo Kosong Chibang eui Mulga Suchun e kwanhan Yon’gu’ (A study on price level in Kosong of Kyongsangdo, 1725–1761), Taedong Kochon Yon’gu 13, pp. 31–61. —(1997) ‘18–19 Segi Mulga Chuse (1744–1862)’ (Price trends in the eighteenth and nineteenth century (1744–1862), Choson Sidae Sahak 2, pp. 193–237. —(1998) ‘Choson Hugi Mika wa Jika Pyondong’ (Variations in rice and land price in the later Choson Dynasty), unpublished manuscript, Seoul: Sungkyunkwan University. Clark, G. (1988) ‘The cost of capital and medieval agricultural technique,’ Explorations in Economic History 25, pp. 265–94. —(1996) ‘The Political Foundations of Modern Economic Growth: England, 1540–1800, Journal of Interdisciplinary History 26, 4, pp. 563–88. Cole, David C. and Park, Yung Chul (1983) Financial Development in the Republic of Korea, 1945–1978, Cambridge, Mass.: Harvard University Press. Collins, William (1997) ‘Real wage convergence in nineteenth century India’, paper presented to the 3rd World Cliometrics Congress. Crawcour, E.S. (1961) ‘The development of a credit system in seventeenth century Japan’, Journal of ‘Economic History 21, 3, pp. 342–60. —(1989) ‘Economic change in the nineteenth century,’ in Marius B.Jansen (ed.) Cambridge History of Japan, Vol. 5, Cambridge: Cambridge University Press, 569–617. Davis, Lance E. and Gallman, Robert E. (1994) ‘Savings, investment, and economic growth: The United States in the nineteenth century’, in John A.James and Mark Thomas (eds) Capitalism in Context, Chicago: Chicago University Press, 202–29. De Vries, Jan (1976) The Economy of Europe in the Age of Crisis, 1600–1750, Cambridge: Cambridge University Press.

Colonial origins of Korea’s market economy

101

Economic Planning Board of Korea (various years) Daehan Minkuk Tong’gye Yon’gam (Statistical Yearbook of the Republic of Korea) Seoul: Economic Planning Board of Korea. Goldin, Claudia and Margo, Robert A. (1992) ‘Wages, prices and labor markets before the Civil War’, in Claudia Goldin and Hugh Rockoff (eds) Strategic Factors in Nineteenth Century American Economic History, Chicago: Chicago University Press, 67–104. Haggard, Stephan, Kim, Byung-kook and Moon, Chung-in (1991) ‘The transition to exportled growth in South Korea: 1954–1966’, Journal of Asian Studies 50, 4, pp. 850–73. Haggard, Stephan and Moon, Chung-in (1990) ‘Institutions and economic policy: theory and a Korean case study’, World Politics 42, 2, pp. 210–37. Hart, Marjolein T. (1997) ‘The merits of a financial revolution: Public finance, 1550– 1700’, in Jonker Joost and Jan Luiten van Zanden (eds) A Financial History of the Netherlands, Cambridge: Cambridge University Press, 11–36. Hicks, J.R. (1969) A Theory of Economic History, Oxford: Clarendon Press. Hunt, E.H. (1981) British Labour History, Weidenfeld and Nicholson: London. —(1986) ‘Industrialization and regional inequality: Wages in Britain, 1760–1914’, Journal of Economic History 46, 4, pp. 935–66. Hurd, J. (1975) ‘Railways and expansion of markets in India, 1861–1921’, Explorations in Economic History 12, pp. 263–88. Iwahashi, Masaru (1981) Kinsei Nihon Bukkashi no Kenkyu (A Study on the Price History of Modern Japan), Tokyo: Oharashinseishi. Jones, S.R.H. (1993) ‘Transaction costs, institutional change, and the emergence of a market economy in late Anglo-Saxon England’, Economic History Review 46, 4, pp. 658–78. Kajimura, Hideki (1968) ‘Richo kogi mengyo no ryutsu oyobi seisan kozo’ (Structure of cotton trade and production in the later Yi Dynasty), Toyo Bunka Kenkyusho Kiyo 46. Kim, Dae-il and Topel, Robert H. (1995) ‘Labor market and economic growth: Lessons from Korea’s industrialization, 1970–1990’, in Richard B.Freeman and Lawrence F.Katz (eds) Differences and Changes in Wage Structures, Chicago: Chicago University Press, 227–64. Kimura, Mitsuhiko and Uranagase, Ryu (1987) ‘Kaikogi Chosen no Kahei to Bukka’ (Money and prices in Korea during treaty port period), Shakai Keizai Shigaku 53, 5, pp. 607–35. Korea Institute of Labor Research (1996) Labor Statistics Database, Seoul: Korea Institute of Labor Research. Krueger, A. (1990) ‘Government failures in development’, Journal of Economic Perspectives 4, 3, pp. 9–23. Lewis, Kenneth H. and Yamamura, Kozo (1971) ‘Industrialization and interregional interest rate structure, the Japanese case: 1889–1925’, Explorations in Economic History 8, pp. 473–99. Margo, Robert A. and Villaflor, Georgia C. (1987) ‘The Growth of Wages in Antebellum America: New Evidence’, Journal of Economic History 47, 4, pp. 873–95. McCloskey, Donald N. and Nash, John (1984) ‘Corn at interest: The extent and cost of grain storage in medieval England’, American Economic Review 74, 1, pp. 174–87. Mitchell, Brian R. (1988) British Historical Statistics, Cambridge: Cambridge University Press. Miyajima, Hiroshi (1974) ‘Chosen kango kaikaku igo no shogyoteki nogyo’ (Commercial agriculture in Korea after the Kabo reforms), Shirin 57, pp. 38–77. Miyamoto, Matao (1981) ‘Edojidaini okeru sho komeshicho aida no rentosei ni tsuite,

102

Myung Soo Cha

1651–1850’ (Relationships among local rice markets in the Tokugawa period, 1651– 1850), Osaka Daigaku Keizaigaku 31, 2 and 3, pp. 274–307. Miyamoto, Matao and Toshimaru Harata (eds) (1985) Rekishi no naka no Bukka (Prices in History), Tokyo: Tobunkan. Mizoguchi, Toshiyuki and Umemura, Mataji (eds) (1988) Kyu Nihon Shokuminchi Keizai Tokei (Basic Economic Statistics of Former Japanese Colonies), Tokyo: Toyo Keizai Shimposha. Nihon Teikoku Tokei Nenkan, various years, Tokyo: the Japanese Government. North, Douglass (1990) Institutions, Institutional Change and Economic Performance, Cambridge: Cambridge University Press. North, Douglass and Weingast, Barry (1989) ‘The evolution of institutions governing public choice in seventeenth-century England’, Journal of Economic History 49, 4, pp. 803– 32. Ohkura, T. and Shimbo, H. (1978) ‘The Tokugawa monetary policy in the eighteenth and nineteenth centuries’, Explorations in Economic History 15, pp. 101–24. Palais, James (1996) Confucian Statecraft and Korean Institutions, Seattle: Washington University Press. Park, Yung Chul (1994) ‘Korea: Development and structural change of the financial system’, in Hugh Patrick and Yung Chul Park (eds) The Financial Development of Japan, Korea, and Taiwan, Oxford: Oxford University Press, 129–87. Patrick, Hugh T. (1984) ‘Japanese financial development in historical perspective, 1868– 1980’, in Gustav Ranis, Robert L.West, Mark W.Leiserson, and Cynthia Taft Morris (eds) Comparative Development Perspectives, Boulder: Westview. —(1994) ‘Comparisons, contrasts, and implications’, in Hugh T.Patrick and Yung Chul Park (eds) The Financial Development of Japan, Korea, and Taiwan, New York: Oxford University Press, 302–27. Perdue, Peter C. (1992) ‘The Qing state and the Gansu grain market, 1739–1864’, in Thomas G.Rawski and Lillian M.Li (eds) Chinese History in Economic Perspective, Berkeley: University of California Press, 100–125. Perkins, Dwight (1969) Agricultural Development in China, 1368–1968, Chicago: Aldine Publishing. Polanyi, Karl (1957) The Great Transformation, Boston: Beacon. Pressnell, L.S. (1960) ‘The rate of interest in the 18th century’, in L.S.Pressnell (ed.) Studies in the Industrial Revolution, London: Athlone Press, 178–213. Pyo, Hak K. (1998) ‘A synthetic estimate of the national wealth of Korea, 1953–1996’, paper presented to workshop held jointly by Institutes of Economic Research of Hitotsubashi and Seoul National University. Rosenbloom, Joshua L. (1990) ‘One market or many? Labor market integration in the late nineteenth century United States’, Journal of Economic History 50, 1, pp. 85–108. Saito, Osamu (1978) ‘The labor market in Tokugawa Japan: wage differentials and the real wage level, 1727–1830’, Explorations in Economic History 15, pp. 84–100. Shimbo, Hiroshi (1978) Kinsei Bukka to Keiza Hatten (Prices and Economic Development in Modern Period), Tokyo: Toyo Keizai Shimposha. Slaughter, Matthew J. (1995) ‘The antebellum transportation revolution and factor-price convergence’, NBER Working Paper no. 5303. Song, Hogeun (1991) Han’guk eui Nodong Jongchi wa Sijang (Labour Politics and Labour Markets in Korea), Seoul: Nanam. Taira, Koji (1970) Economic Development and the Labor Market in Japan, New York: Columbia University Press.

Colonial origins of Korea’s market economy

103

Teranishi, Juro (1994) ‘Modernization of financial markets: an analysis of informal credit markets in prewar Japan’, World Development 22, 3, pp. 315–22. Ueda, Kazuo (1993) ‘Kinyu Sistemu’ (Financial System), in Tetsuji Okazaki and Masahiro Okuno (eds) Kendai Nihon Keizai Sistemu no Genryu (Origins of the Contemporary Japanese Economic System), Tokyo: Nihon Keizai Shimposha. World Bank (1993) The East Asian Miracle, Oxford: Oxford University Press.

5

South Korea’s late industrialization in comparative historical perspective* Jaymin Lee

Introduction There have been many studies of the pattern of industrialization, most of them focusing on nineteenth-century Europe. But many countries have industrialized in the twentieth century, particularly in East Asia. So it is now possible to compare patterns of industrialization in Europe in the nineteenth century and East Asia in the twentieth. Few studies so far have tackled this problem. Economic historians are reluctant to deal with recent development, and development economists use historical facts only as a background for contemporary analysis. Gerschenkron’s analysis of industrialization in nineteenth-century Europe may be a starting point (Gerschenkron 1962a, 1968a). Can it cover twentieth-century East Asia as well, and provide an understanding of European and East Asian industrialization? Today, economic historians are sceptical about the validity of Gerschenkron’s concept of European industrialization in the nineteenth century. But development economists often use it to explain the pattern of industrialization in the twentieth century, implicitly accepting it as an accurate description of European industrialization.1 This paper argues that, although Gerschenkron’s scheme has been discredited for nineteenth-century Europe, it does explain South Korea’s late industrialization in the latter half of the twentieth century quite well. The second section examines Gerschenkron’s scheme and the modifications needed to make it applicable to particular examples of industrialization. The third section investigates how well Gerschenkron’s scheme explains the pattern of industrialization in European latecomer countries such as Germany and Russia, and East Asian countries like Japan and South Korea. The final section discusses the results and concludes. Pattern of late industrialization Gerschenkron proposed the following six propositions about European industrialization in the nineteenth century. The more backward a country’s economy was: *

The research for the paper was financially supported by the Center for International Studies, Graduate School of International Studies, Yonsei University.

104

South Korea’s late industrialization

105

1 the more likely its industrialization was to start as a sudden great spurt with a high rate of growth of manufacturing output; 2 the more its industrialization would involve large plant and enterprise; 3 the greater the stress on producer goods relative to consumer goods industries; 4 the heavier the pressure on the consumption level, particularly of the agrarian sector; 5 the greater and more coercive the part played by institutions to increase the supply of capital to nascent industries, and provide them with less decentralized and better informed entrepreneurial guidance; 6 the less likely it was that agriculture would play an active role in offering an expanding industrial market based on rising productivity. This paper will deal with four of these propositions. Propositions 4 and 6 will not be considered, although they are important issues in Gerschenkron’s scheme, especially with reference to the Russian Revolution (Gerschenkron 1968b). They are excluded here not because they are unimportant, but so as to focus on the pattern of industrialization itself rather than the interaction between the agrarian and industrial sectors in the process of industrialization.2 To test whether the four remaining propositions fit economic development in Europe and East Asia, several clarifications and modifications are needed. First, ‘producer goods’ industries are interpreted as ‘more dynamic’ sectors. By doing so, Gerschenkron’s proposition can be made more flexible so that it can be applied to twentieth-century industrialization. It does not destroy Gerschenkron’s own idea because by emphasizing producer goods, Gerschenkron apparently meant moving to more dynamic sectors (Hirschman 1971, pp. 94–5). Second, Gerschenkron’s scheme is added to by clarifying the problem of the international division of labour, one of its unexplained parts. Latecomer countries’ industrialization occurs as a result of interaction with advanced countries. Since Gerschenkron emphasized the importance of borrowed technology for late industrialization, he obviously did not ignore this interaction. But his position about the international division of labour is unclear, though his view about the emphasis by latecomers on producer goods industries may have some implications. Gerschenkron gives two reasons why latecomer countries tend to emphasize producer goods industries: a In latecomer countries, skilled and disciplined factory labour is a scarce resource, so it is rational for them to concentrate on more capital intensive producers’ goods industries. b Latecomer countries can move into dynamic and capital intensive industries through institutional arrangements like the introduction of long-term investment banks or state intervention.3 The problem is that (a) and (b) can contradict each other, (a) implies that latecomer countries can have comparative advantage in producer goods industries due to the current resource endowment from the early phase of industrialization. On the other

106

Jaymin Lee

hand, (b) implies that latecomer countries do not have comparative advantage in producer goods industries in the early phase of industrialization, but can make endeavours to move into producer goods industries through various institutional arrangements. In other words, they tend to promote producer goods industries as infant industries. To make Gerschenkron’s scheme more rigorous and applicable to non-European countries, it seems reasonable to assume that Gerschenkron means that latecomer countries try to move into producer goods industries in which they do not currently have comparative advantage. Gerschenkron said that Italy in the last decade of the nineteenth century could not experience a full spurt because the government failed to promote newly emerging dynamic industries (Gerschenkron 1962b), and argued similarly for Bulgaria (Gerschenkron 1962d). To create dynamic sectors, the role of institutional factors is essential, such as the role of the state and industrial-financial systems influenced by the state. At the heart of Gerschenkron’s proposition lies the industrial policy of the state. Crafts (1984) investigated structural change in nineteenth-century Europe applying a KuznetsChenery style pattern of growth. He found that, due to the pressure of the international division of labour, latecomer countries could not achieve comparative advantage in manufactured goods in the way that Britain had at the same level of per capita GDP. Latecomers had to decide whether they would adapt to the international division of labour or pursue an aggressive industrial policy. Small countries like Denmark or Finland belonged to the former category, while Germany, Russia and Italy, major countries in Gerschenkron’s scheme, belonged to the latter category. Once Gerschenkron’s scheme for late industrialization is characterized as industrial policy aimed at promoting infant industries, the success of industrial policy becomes an issue. Maturation and growth of infant industries promoted by industrial policy are important issues. A test of the success of an industrial policy is to judge whether the promoted infant industries grow and mature, and become competitive in international markets. Thus Gerschenkron’s scheme can be tested against the historical facts by asking the following questions: 1 Was there a sudden great spurt? 2 Was economic growth accompanied by a rapid structural transformation to dynamic sectors? 3 Was industrial policy for dynamic transformation systematically implemented? 4 Did industrial policy succeed, i.e. did the promoted infant industries grow and mature? The test of time: empirical evidence The test of Gerschenkron’s proposition can be carried out using the empirical findings done since his works were published. Germany and Russia will be examined first, and then Japan and South Korea. Germany and Russia are major countries in Gerschenkron’s scheme and Japan is the only non-Western country

South Korea’s late industrialization

107

that has fully industrialized. South Korea, notwithstanding the current crisis, is likely to become an advanced industrial country in the near future. Gerschenkron’s propositions were more or less based on the empirical findings for Europe accumulated up to the 1950s, but they have not stood the test of time since then. The current state of empirical evidence for Europe is not so favourable to Gerschenkron’s propositions. Germany Germany, a major country in Gerschenkron’s scheme, did not experience a discontinuous rise of the growth rate in 1840–70, when Gerschenkron says there was a great spurt. There is little evidence that the industrial growth rate jumped between 1830s and 1840s (Tilly 1991, pp. 176–81). As for the structural transformation to dynamic sectors, Germany experienced industrialization based on producer goods industries when compared with Britain. But during the 1840–70 period, economic growth was led by older technology rather than newer technology and industries. Germany began to experience growth based on the new technology and industry of the ‘Second Industrial Revolution’ only from the 1870s. This is also the case with the role of the banks and the state which supported them. It is true that large universal banks, copying the Credit Mobilier of France, played an important role in founding new firms, merger and acquisitions, and organizing cartels. The government also played an important role by imposing protective tariffs, supporting universal banks, and providing social overhead capital, etc. However, during the 1840–70 period, German trade policy was virtually free trade, and as Cameron (1972, pp. 13–14) pointed out, the role of the universal banks was not so pronounced. As in the case of structural transformation, the role of banks and government became salient from the 1870s. However, Gerschenkron (1989, ch. 1) severely criticized the introduction of protective tariffs in 1879 as unnecessary. He was wrong in his dating in this respect. Disregarding the dating errors, it is possible to apply Gerschenkron’s scheme to Germany from 1870 up to World War I. During this period, there was a rapid transition to dynamic sectors, based on the technology of the Second Industrial Revolution, and the role of banks and government became important. But there is no evidence that Germany experienced a sudden great spurt during this period. Russia Empirical evidence is also not favourable for Russia. According to Gerschenkron, Russia experienced a spurt in the 1890s and in 1906–13, but again there is no evidence that the growth rate increased discontinuously during this period (Crisp 1991, p. 264). Russian industrialization during Gerschenkron’s spurt period was led by the consumer goods industries, with a subsidiary role played by dynamic sectors such as new vintage steel mills. As late as 1913, textiles and shoes accounted for half of

108

Jaymin Lee

total industrial output, or twice as much industrial output as mining, metallurgy, and engineering added together (Gatrell 1986, p. 144). It is also questionable whether the state played an active role in Russian industrialization. Industrial subsidies were lacking in the Russian budget. Tariffs were high, but their main purpose was revenue rather than a coherent industrial policy. State enterprises were established, not for promoting hi-tech industries, which was the aim of industrial policy, but for things like monopolizing spirit production. Gerschenkron erred in mistaking intent for actual policy. He was fascinated by the plan that Witte proposed to the Tsar, but the ineffective state bureaucracy of Russia could not implement it (Gregory 1991, p. 73). In summary, Russian industrialization was market-led rather than state-led (Munting 1996). As with Germany, Gerschenkron’s scheme may explain Russia’s industrialization later than the period defined by Gerschenkron as the great spurt. Munting (1996) has pointed out that state-led industrialization described by Gerschenkron may fit Soviet industrialization better than imperial Russia.4 Soviet industrialization from 1928 may be defined as a sudden great spurt, accompanied by rapid structural transformation to heavy and chemical industries where there was dynamic growth. However, as Soviet industries never succeeded in gaining international competitiveness, the policy cannot be regarded as a success. Basically the same thing can be said for the other latecomer countries of nineteenth-century Europe. The broad picture of the pattern of European industrialization in the nineteenth century is that France was close to Britain and other continental countries experienced a gradual upswing of growth and structural transformation rather than a sudden great spurt. There were few systematic industrial policies implemented by the state during the spurt period outlined in Gerschenkron’s scheme (Sylla and Toniolo 1991, p. 22). Japan Gerschenkron’s scheme has rarely been applied to non-European cases. Gerschenkron himself (1962a: p. 27) was sceptical about the possibility that Third World countries of his day could industrialize. He did not pay much attention to Japan and failed to predict the ‘East Asian miracle’. Rosovsky (1961) is one attempt to apply Gerschenkron’s scheme to Japan, and he concludes that it is rather difficult to apply the idea to Japan. Empirical studies of Japan have shown that after modern economic growth began around 1885, the growth rate rose in long swings rather than a sudden great spurt (Ohkawa and Rosovsky 1973, ch. 8; Minami 1981, ch. 3). The rising growth rate through these long swings was accompanied by the growth of new industries. But there was no rapid structural transformation to dynamic sectors during the Meiji period when the breakthrough in industrialization was made. Japan initially depended on traditional industries like silk manufacturing, followed by light manufacturing industries like cotton textiles. Heavy and chemical industries developed slowly, only emerging in the 1930s in a full-fledged form.

South Korea’s late industrialization

109

Japan was the first country to implement industrial policy in a systematic way from the very beginning of industrialization. The ‘Shokusan-Ko¯gyo¯’ in the Meiji era deliberately fostered modern industries through government policy. As a result of this policy, there emerged a concentrated industrial organization dominated by zaibatsu with banks and large industrial firms within the same organization. The zaibatsu system was more or less the result of deliberate government policy, geared towards catching up with industrialized countries (Rosovsky 1972). Japan’s industrial policy aimed at moving resources to more dynamic industries, but the pace of movement was slow. Japan had difficulty with absorbing new technology even in simple modern industries like cotton textiles. However, Japan did succeed in developing international competitiveness in these industries over time, so the industrial policy could be regarded as successful. As in the case of Germany or Russia, Japan went through more than one phase of industrialization, and Gerschenkron’s scheme may be more relevant to a later phase of industrialization. After making a breakthrough in industrialization based on light manufacturing industries, Japan experienced full-fledged heavy and chemical industrialization in the 1930s. The growth rate was higher in the latter phase of industrialization, when industrial policy again played an important role. The heavy and chemical industries eventually gained international competitiveness, though in some cases only after World War II. But there is again little evidence that there was a sudden great spurt of growth in the 1930s. South Korea South Korea’s rapid industrialization from the 1960s is often described as ‘condensed growth’, which can be interpreted as Gerschenkron’s great spurt (Cho 1994, pp. 5–7). The growth rate of South Korea’s per capita GNP was on average 0.4 per cent from 1953 to 1962, but it suddenly jumped to 6.1 per cent in 1963 and stayed at about 7 per cent on average until 1996. Except for the severe crisis in the early 1980s, the growth rate was not much affected by the world business cycle.5 South Korea’s spurt was accompanied by a rapid structural transformation engineered by the industrial policy of the state. Whether South Korea’s industrialization strategy was a simple export-oriented one based on static comparative advantage or one characterized by active industrial policy to promote infant industries has been a subject of debate between ‘neoclassical’ and ‘revisionist’ views. It is now an established fact that the South Korean government heavily intervened in the industrialization process.6 While pursuing outward-looking development and export-promotion, the South Korean government actively promoted infant industries through protection, subsidy and other means. The government also deliberately promoted chaebol, or large business conglomerates, to take care of investment in large projects and more dynamic industries (see Jones and Sakong 1980). The relevant question now is whether industrial policy made any difference. Did it bring about the intended effect of fostering infant industries to maturation and growth? The empirical study by the World Bank (1993, ch. 6) suggests that

110

Jaymin Lee

South Korean infant industries did not mature or grow faster than mature ones. However, other studies (Enos and Park 1988; Jacobsson 1993; Lee 1997) have provided evidence that infant industries may have matured in South Korea. This unresolved issue in development economics cannot be discussed in depth here, but the Appendix at the end of this paper suggests that the heavy and chemical industries promoted as infant industries during the great spurt did mature and grow. Patterns of late industrialization The test results of Gerschenkron’s propositions for Germany, Russia, Japan, and South Korea are summarized in Table 5.1. The first row for Germany and Russia is for the period when the spurt occurred according to Gerschenkron. The second row is for Germany 1870–1913 and Russia’s spurt period under the Soviet planning system. In the case of Japan, the first row is for 1885–1920, when the breakthrough in industrialization was made and modern industries were first established. The second row is for 1920–39, when Japan experienced heavy and chemical industrialization after a period of stagnation. In the case of Korea it is 1963–96, when Korea experienced rapid economic growth. The periods of great spurt, rapid structural transformation, systematic industrial policy and its success are checked ‘Yes’ and ‘No’. The contents of Table 5.1 indicate that South Korea’s pattern of industrialization fits Gerschenkron’s scheme better than Germany’s, Russia’s or Japan’s. It will be interesting to compare Korea’s pattern in Table 5.1 with that of Germany and Japan during the periods in the first and second rows together.7 Though Korea’s spurt was shorter than these combined periods, Korea’s growth rate was much higher, so that in terms of the amount of transformation of the economy, Korea’s spurt period may be compared with them. The results for Germany and Japan for combined periods are presented in the third row. The third row is marked ‘Yes’ if either the first or second row for each category is ‘Yes’. The comparison of the third row for each country with the Korean case tells us that, even when the condition for ‘Yes’ is thus relaxed, Korea’s pattern fits Gerschenkron’s scheme better than Germany’s or Japan’s. This comes from the fact that Germany or Japan did not show a sudden great spurt but Korea did. Both Germany and Japan showed ‘trend acceleration’ after modern economic growth began.8 In contrast, Korea experienced a sudden great spurt from 1963, which persisted for more than 30 years, being accompanied by rapid structural transformation and a systematic industrial policy.9 Discussion and conclusion The reason why South Korea experienced a spurt while previous latecomer countries in Europe and Japan failed to do so is simple. South Korea in the 1960s had a much larger technological backlog to borrow from than the European countries or Japan in the nineteenth century or early twentieth century. In the nineteenth century, the gap of per capita GDP between early starters like Britain on the one hand and continental European countries or Japan on the other was at most one to

South Korea’s late industrialization

111

Table 5.1 Patterns of late industrialization: historical comparisons

three or one to four (see Maddison 1991, pp. 6–7). In the latter half of the twentieth century, the gap between the most advanced and the least developed is sometimes almost ten times as large. South Korea in 1960 had per capita GNP equivalent to one-fiftieth of that of the USA in official exchange rate (which was undervalued). To the extent that the per capita GNP represents the accumulation of knowledge and technology, the larger gap means a greater technological backlog to borrow from. Of course, a larger technological backlog means no more than a greater potential for growth. Realizing this potential is virtually the whole story of economic growth for latecomer countries. Recently, cross-country quantitative studies have supplied some determinants of growth rates, or the way the potential has been realized in the postwar era. High rates of investment and saving, stable macroeconomic policy, heavy investment in education, and political stability (which was in turn achieved through relatively equitable distribution of income) led to high economic growth. Hence the ‘East Asian miracle’, including the South Korean one (Barro 1991; Collins and Bosworth 1996). These factors, however, do not explain the sudden spurt that occurred from 1963. The sudden spurt was precipitated by the emergence of a politico-economic system summarized as the ‘developmental state’, which placed the goal of economic development above all other values.10 The rich endowment of human resources, which was the result of previous human capital accumulation through education, was an important background. The South Korean developmental state, rather than passively adapting to the international division of labour, aggressively pursued industrial policy to move resources into more dynamic sectors. Of course, it was important that the developmental state did not stifle the discipline of market forces. In South Korea, market discipline imposed by exports, imports and domestic competition contributed to the maturation and growth of infant industries (see Lee 1997). The ‘hard’ state, not captured by special interests, could utilize the market discipline effectively while intervening heavily in the economy.

112

Jaymin Lee

If a country coming after South Korea finds itself in a similar situation, it may be able to industrialize according to the pattern described by Gerschenkron. As in nineteenth-century Europe, there are smaller countries that are not likely to pursue industrial policy or adopt the developmental state, but for larger countries, this may be necessary. For example, if China is to be regarded as successful in industrialization, she will need to transform the heavy and chemical industries inherited from the pre-reform era to make them internationally competitive. The Chinese state has to lead the transformation, and it is trying to promote a business organization like chaebol for this purpose. If China manages to transform her heavy and chemical industries successfully, her industrialization may fit into the Gerschenkron scheme. It is also possible that other countries, for example North Korea, may tread a similar pattern of industrialization in the future. Empirical evidence presented in the third section and the discussion in this section suggest that Gerschenkron’s scheme explains the industrialization process of South Korea and possibly other East Asian countries better than that of European countries like Germany or Russia, though it was initially conceived for them. However, as Gerschenkron’s scheme aims to explain the pattern of late industrialization, and South Korea is a latecomer compared with European countries, the findings above endorse the scheme. Gerschenkron’s proposition seems to provide a good explanatory framework for analysing late industrialization in comparative historical perspective. Appendix Among various industrial policy measures that the Korean government employed, protection has had the largest redistribution effect. While financial policy has a strong effect in inducing production, its redistributive effect is limited to the difference between official and market rates of interest, and thus cannot match the effect of protection. The available data on cross-industry variation in the effective rate of subsidy for 1968, which include financial and fiscal subsidies as well as protection (Westphal and Kim 1976), are dominated by the variation in the effective rate of protection. The effective rate of protection (ERP) is a better measure of protection than the nominal rate of protection in a developing country like Korea, where production depends greatly on imported input. ERP, when calculated by comparing domestic and international prices, also represents the degree of international competitiveness. South Korea has an excellent data set for ERP. ERP has been calculated for 1963, 1970, 1975 and 1978 by Kim and Hong (1982) and for 1980, 1983, 1985, 1988, 1990, 1993 and 1995 by Hong (1997), both comparing domestic and international prices. Table 5.2 presents ERP figures for more dynamic heavy and chemical industries (HCI), less dynamic light manufacturing industries (LI), and the whole manufacturing sector (MS). ERP has been calculated for total sales (weighted average for domestic sales and export, with ERP for export taken to be zero) by the Corden method. The last row in Table 5.2 presents the difference in ERP between LI and HCI. This difference is important considering that we are interested in

South Korea’s late industrialization

113

Table 5.2 Trend of effective rate of protection (%)

Sources: Calculated from Kim and Hong (1982) and Hong (1997). Notes: 1 The effective rate of protection was calculated by the Corden method. 2 LI and HCI denote light industries and heavy and chemical industries respectively. LI includes food, beverages, tobacco, textile, apparel, leather, wood, paper, printing and publishing, rubber, plastic, non-metallic mineral and miscellaneous manufacturing. HCI includes chemicals, petroleum and coal, metal and machinery.

Table 5.3 Composition of value added in world market prices (%)

Sources: as Table 5.2. Note: Value added in world market price was calculated by the Corden method.

‘comparative’ advantage, and that the ERP figure itself may be affected by currency over- (or under-)valuation. In 1963, the year when the great spurt began, HCI had lower ERP than LI. This apparently means not that HCI had a comparative advantage over LI, but that LI was revealing the common characteristics of import-substituting industrialization in developing countries. Korea pursued import-substituting industrialization in LI in the 1950s while depending on imports for HCI products. HCI came to have higher ERP than LI between 1963 and 1970, apparently reflecting the fact that Korea’s developmental state began to actively promote more dynamic HCI. The ERP gap between HCI and LI widened until 1978 through the process of heavy and chemical industrialization effort by the government. The gap was reduced in the early 1980s when the excess of heavy and chemical industrialization was controlled, but stayed somewhat constant during the rest of that decade, until the gap narrowed quickly in the 1990s. This trend basically indicates that in the 1960s and 1970s, the government actively promoted HCIs as more dynamic infant industries, and in the 1980s and 1990s they matured, though with some short-run fluctuations. The trend of the amount of production for LI and HCI, represented as the proportion of the value added in world market prices, is presented in Table 5.3. Value added in world market prices is a more accurate measure of production than value added or output in domestic prices for a developing country like Korea. The figures in Table 5.3 show that the relative weight of HCI consistently rose after 1963. All in all, Korea’s HCI as infant industries have matured and grown in the developmental state.

114

Jaymin Lee

Notes 1 For a good summary of historians’ view, see Sylla and Toniolo (1991). For the references made by development economists, see, for example, Hirschman (1971), Cho (1994) and Shin (1996). 2 To the six propositions noted above, one could add Gerschenkron’s emphasis on the role of a particular ideology for latecomer countries. This paper will not deal with the issue. 3 According to Gerschenkron, another reason why latecomer countries emphasize the producer goods industry is that, owing to complementarity in demand, industrialization proceeds alongside a simultaneous growth of consumer goods industries, producer goods industries and social overhead capital, notably the railroad. This is based on the perception in the 1950s that, owing to the influence of Keynesian economics, shortage of demand is the source of backwardness (Gregory 1991, p. 76). However, industrialization is itself a process which increases supply capacity so this is seemingly a misperception by Gerschenkron. Since a country’s position in international division of labour is mainly determined by its supply capacity, it also cannot explain the international division of labour. 4 Gerschenkron (1962c) himself thought that the Stalinist development policy was the replication of Tsarist industrialization efforts. 5 South Korea’s industrialization did not begin in the 1960s. Some degree of industrialization occurred in the 1930s under the Japanese colonial rule, and in the 1950s relying on foreign aid. However, South Korea was one of the poorest countries in the world by the early 1960s, and it seems fair to say that South Korea’s industrial takeoff occurred in the 1960s. Industrialization before that could be regarded as a prototype. For latecomer countries, some earlier industrialization, even though based on modern technology, can be regarded as a prototype. See Trebilcock (1981, ch. 4). 6 For typical neoclassical views, see Krueger (1978) and Balassa et al. (1982); for revisionist views see Amsden (1989) and Wade (1990). Neoclassical economists have come to acknowledge heavy government intervention in East Asia so they have switched to a ‘market friendly’ view. See World Bank (1993). Some neoclassical economists still stick to the previous view (e.g. Krueger 1997). 7 In the case of Russia, it is difficult to consolidate the periods in the first and second rows due to the interruption by the revolution. 8 It is common to characterize Japanese modern economic growth as trend acceleration (Ohkawa and Rosovsky 1973; Minami 1981, ch. 3). In the case of Germany, according to Maddison (1991, p. 49), the annual growth rate of per capita GDP was 0.7 per cent during 1820–70 period and 1.6 per cent during 1870–1913 period. 9 If Korean industrialization under the Japanese colonial rule was heavily emphasized, Korea’s industrialization process might be characterized as trend acceleration. However, as mentioned in note 5, the task of linking colonial development and rapid economic growth from 1963 should be tackled first to make this proposition valid. Moreover, Korea’s level of development was too low in the early 1960s. 10 The ‘developmental state’ is basically a concept devised by political scientists or sociologists without giving rigorous theoretical grounds (see, for example, Johnson 1982; Weiss and Hobson 1995). But it could be given theoretical underpinnings. See Aoki, et al. (1997) and Lee (1999), for example.

References Amsden, Alice (1989) Asia’s Next Giant: South Korea and Late Industrialization, New York: Oxford University Press. Aoki, M., Murdoch, K. and Okuno-Fujiwara, M. (1997) ‘Beyond the East Asian miracle:

South Korea’s late industrialization

115

Introducing the market-enhancing view’, in M.Aoki, H.Kim and M.Okuno-Fujiwara (eds) The Role of Government in East Asian Economic Development: Comparative Institutional Analysis, Oxford: Clarendon Press, pp. 1–37. Balassa, B. and Associates (1982) Development Strategies for Semi-industrial Economies, Baltimore: Johns Hopkins University Press. Barro, R.J. (1991) ‘Economic growth in a Cross-Section of Countries,’ Quarterly Journal of Economics, 106(2), pp. 407–43. Cameron, R. (ed.) (1972) Banking and Economic Development, New York: Oxford University Press. Cho, Soon (1994) The Dynamics of Korean Economic Development, Washington, D.C.: Institute for International Economics. Collins, S. and Bosworth, B.P. (1996) ‘Economic growth in East Asia: accumulation versus assimilation’, Brookings Papers on Economic Activity, 2, pp. 135–203. Crafts, N.F.R. (1984) ‘Patterns of development in nineteenth century Europe’, Oxford Economic Papers, 36, pp. 438–58. Crisp, O. (1991) ‘Russia’, in Sylla, R. and Toniolo, G. (eds) Patterns of European Industrialization: the Nineteenth Century, London: Routledge, pp. 248–68. Enos, J.L. and Park, W. (1988) The Adoption and Diffusion of Imported Technology: The Case of Korea, London: Croom Helm. Gatrell, P. (1986) The Tsarist Economy 1850–1917, London: B.T.Bratsford. Gerschenkron, A. (1962a) Economic Backwardness in Historical Perspective, Cambridge: Harvard University Press. —(1962b) ‘Notes on the rate of industrial growth in Italy, 1881–1913’, in Gerschenkron, A. Economic Backwardness in Historical Perspective, Cambridge: Harvard University Press, pp. 72–89. —(1962c) ‘Russia: Patterns and problems of economic development, 1861–1958’, in Gerschenkron, A. Economic Backwardness in Historical Perspective, Cambridge: Harvard University Press, pp. 119–51. —(1962d) ‘Some aspects of industrialization in Bulgaria, 1878–1939’, in Gerschenkron, A. Economic Backwardness in Historical Perspective, Cambridge: Harvard University Press, pp. 198–234. —(1968a) Continuity in History and Other Essays, Cambridge: Harvard University Press. —(1968b) ‘Reflections on the economic aspects of revolution’, in Gerschenkron, A. Continuity in History and Other Essays, Cambridge: Harvard University Press, pp. 257–80. —(1989) Bread and Democracy in Germany, London: Cornell University Press. Gregory, Paul R. (1991) ‘The role of the state in promoting economic development’, in Sylla, R. and Toniolo, G. (eds) Patterns of European Industrialization: the Nineteenth Century, London: Routledge, pp. 64–79. Hirschman, A.O. (1971) ‘The political economy of import-substituting industrialization in Latin America’, in A.O.Hirschman A Bias for Hope, New Haven: Yale University Press, pp. 85–123. Hong, Sung-duk (1997) ‘The extended estimation of nominal and effective rates of protection’, mimeographed. Seoul: Korea Development Institute (in Korean). Jacobsson, S. (1993) ‘The length of infant industry period: Evidence from engineering industry in South Korea’, World Development, 21, pp. 407–9. Johnson, C. (1982) MITI and the Rise of Japanese Miracle: the Growth of Industrial Policy, 1925–1975, Stanford: Stanford University Press. Jones, L. and Sakong, I. (1980) Government, Business, and Entrepreneurship in Economic Development, Cambridge: Harvard University Press. Kim Kwang Suk and Sung-duk Hong (1982) ‘Long run change of nominal and effective rates of protection’, mimeographed. Seoul: Korea Development Institute (in Korean).

116

Jaymin Lee

Krueger, A.O. (1978) Liberalization Attempts and Consequences, Cambridge: Ballinger. —(1997) ‘Trade policy and economic development’, American Economic Review, 87(1), pp. 1–22. Lee, Jaymin (1997) ‘The maturation and growth of infant industries: the case of Korea’, World Development, 25(6), pp. 1271–81. —(1999) ‘East Asian NIEs model of development: Miracle, crisis, and beyond’, The Pacific Review, 12(2) pp. 141–62. Maddison, A. (1991) Dynamic Forces in Capitalist Development: A Long-run Comparative View, New York: Oxford University Press. Minami, R. (1981) Economic Development of Japan, Tokyo: Toyokeizaishinbosha (in Japanese). Munting, R. (1996) ‘Industrial revolution in Russia’, in M.Teich and R.Porter (eds) The Industrial Revolution in National Context, Cambridge: Cambridge University Press, pp. 329–49. Ohkawa, K. and Rosovsky, H. (1973) Japanese Economic Growth, Stanford: Stanford University Press. Rosovsky, H. (1961) Capital Formation in Japan 1868–1940, New York. —(1972) ‘What are the lessons of Japanese economic history?’, in A.J.Youngson (ed.) Economic Development in the Long Run, London: George Allen and Unwin, pp. 229– 54. Shin, Jangsup (1996) The Economics of the Latecomers: Catching-up, technology transfer and institutions in Germany, Japan and South Korea, London: Routledge. Sylla, R. and Toniolo, G. (1991) ‘Introduction: Patterns of European industrialization during the nineteenth century’, in Sylla, R. and Toniolo, G. (eds) Patterns of European Industrialization: the Nineteenth Century, London: Routledge, pp. 1–26. Tilly, R. (1991) ‘Germany’, in Sylla, R. and Toniolo, G. (eds) Patterns of European Industrialization: the Nineteenth Century, London: Routledge, pp. 175–96. Trebilcock, C. (1981) The Industrialization of the Continental Powers, 1780–1914, London: Longman. Wade, R. (1990) Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization, Princeton: Princeton University Press. Weiss, L. and Hobson, J. (1995) States and Economic Development: A Comparative Historical Analysis, Cambridge: Polity Press. Westphal, L.E. and Kim, Kwang Suk (1976) Korea’s Foreign Exchange and Trade Policy, Seoul: Korea Development Institute (in Korean). World Bank (1993) The East Asian Miracle: Economic Growth and Public Policy, New York: Oxford University Press.

6

Export dynamics in Taiwan and Mainland China 1950–2000 A Schumpeterian approach Hans H.Bass*

This paper deals with the export performance of two of Asia’s most dynamic economies, Taiwan and Mainland China. The paper uses a theoretical approach derived from J.A.Schumpeter, one of the founding fathers of dynamics as a concept in economic theory and economic history. To compare the economic history of Taiwan and Mainland China may be considered strange for various reasons—for example the differences in size, the present state of development, and the shape of their economic systems. However, the two economies can well be compared by focusing on their success in global markets. These days they increasingly interact and contribute to the ‘Greater China’ growth pole in the world economy. This paper will analyse the export performance of both Taiwan and Mainland China during the past 50 years. Then it will investigate the interaction between Taiwan and Mainland China in the 1990s with regard to exports. Finally, the paper will consider the export potential of these economies and discuss their respective needs for adjustment. Two success stories: some macro data 1950s to 1990s Dynamics and innovation in a Schumpeterian approach According to Schumpeter (1912), continuously new combinations of productive factors, or innovations, lead to overall growth in a dynamic economy. In this tradition this paper considers ‘export dynamics’ to mean a continuous upgrading in the structure of exports, i.e. the inclusion of more and more sophisticated commodities, and/or the developing of new markets, plus subsequent growth in export volume. ‘Innovation’ will not be confined to spectacular ‘new things’ (products) or processes (see also Schumpeter 1951, p. 218). Differing from many recent studies on innovation (e.g. Grupp 1998), this paper will consider non-technological forms of innovation as equally important, particularly ‘the opening of a new market, that is a *

The author would like to thank particularly Dr A.J.H.Latham (Swansea) for his very helpful comments on a previous version of this paper. Thanks are also due to Prof. Hitoshi Hirakawa (Tokyo), Mr Peter Kreutzberger (Beijing/Paris), Dr Bankole Oni (Bremen/Ibadan), and Prof. Karl Wohlmuth (Bremen) for their comments.

117

118

Hans H.Bass

market into which the particular branch of manufacture of the country in question has not previously entered’ (Schumpeter 1912, p. 66). This is important, as ‘learning to market in the newer technologies can be and often is more difficult than learning to produce’ (Freeman 1996, p. 169).1 ‘World novelty’ is not considered necessary for products from catching-up economies. Export growth The success of Taiwanese exports dates back to the 1950s. From 1952 to 1965 exports grew on average by 11 per cent per year, and from 1965 to 1980 by a staggering 29 per cent per year. Mainland China’s economic planners started to focus on export growth only after 1978. In the 1980s, few areas witnessed an export performance as dynamic as Mainland China2 and Taiwan. The average annual growth rate of export value 1980–95 was 15 per cent in China and 13 per cent in Taiwan.3 In 1979, Taiwan ranked twenty-second and the Mainland thirty-fourth as areas of origin in world merchandise trade flows. Today, both have bypassed many areas, which were previously leading export economies, and in 1977 were ranked fourteenth (Taiwan) and tenth (Mainland China). (Figures in this paragraph were computed with data in TSDB 1996; WDI 1997, pp. 154–6; GATT 1991 and WTO 1998; more details are given in Table A1).4 The following paragraphs will discuss the structural changes causing—rather than accompanying—the growth in volume. The history of Taiwan’s post-colonial foreign trade can be divided into five phases (Fei, Ohkawa and Ranis 1985; Chaponnière and Lautier 1998). Each one represented a substantial diversification of the export structure and an upgrading of capital-intensity in production and the technology content of commodities. Thus they ousted previous exports—a process which can be termed Schumpeter’s ‘creative destruction’. The five phases can be described as follows (for more details on data see Figure 6.1 and Tables A2-A4): 1 Export expansion based on agricultural commodities. Imports consisted of manufactured consumer goods and producer goods (up to 1950). 2 Import substitution growth, based on export of processed agricultural commodities. Imports consisted of producer goods and a decreasing share of manufactured consumer goods (1950–62). 3 Export substitution (export diversification) growth based on processed agricultural and industrial commodities (clothing, yarns, toys, footwear, sportrelated products). Imports consisted of food, manufactured consumer goods and producer goods (1962–70). 4 The second drive of import and export substitution growth, based on an increasing share of intra-industry trade in increasingly sophisticated commodities (especially electronics and NC-machine tools) (1970–86). 5 After 1986, due to a strong upward revaluation of the currency and increasing average unit labour cost, exports of all labour-intensive commodities decreased, and exports grew of hi-tech products, including newly developed ones (like computer notebooks). Between 1990 and 1996 exports in office machines and

Taiwan and Mainland China export dynamics

119

Figure 6.1 Taiwan’s export composition 1952–95. Source: See Table A2.

automatic data processing machines (SITC 75) and electrical machinery (Standard International Trade Classification (SITC 77)) increased threefold. An analysis on the 4-digit level of SITC groups reveals that exports of integrated circuits and automatic data processing machinery, including central processing units of computers, increased as much as fivefold (see Table A5). These subgroups represent the upper level in these commodity groups in terms of technology and knowledge content. Although time-lagged, post-war Mainland China’s foreign trade can also be interpreted as a series of upgrading steps (Lardy 1992; Bass and Wauschkuhn 1995; IBRD 1997; for more details see Tables A6 and A7 and Figure 6.2): 1 An inward looking economy, which only imported producer goods necessary to the framework of a central plan and could not be produced in the country itself. Textiles, raw materials, and agricultural by-products were exported to finance imports (1950–78). 2 Increased exports of petrol financed China’s turn to the outside world and global markets (1978–85). In this period Mainland China’s trade expansion was mainly due to petrol exports (having developed oil fields like Daqing in the 1970s), and other raw materials. Between 1981 and 1985 an increasing share of national petrol production was exported (1981:14 per cent, 1985:24 per cent (see Brender 1992, p. 35)). Yet world market prices of petrol declined sharply in 1986 (to 42 per cent of the 1983/85-average in 1986, and to 48 per cent in 1987). Chinese economic planners first tried to overcome deteriorating

120

Hans H.Bass

Figure 6.2 China’s export composition 1965–95. Source: See Table A6.

earnings by exporting increasing quantities, but soon redirected oil production to domestic industries, and began to concentrate on manufactured commodities for export. 3 ‘Export substitution’ growth based on labour-intensive manufactured consumer commodities: textiles, shoes, plastic fabrics, particularly those manufactured in Sino-foreign joint ventures and rural collective industries. Imports consisted of producer goods (1985-mid-1990s). However, a new drive to rural industrialization in the early 1990s (imports of textile machinery of US$10 bn) had triggered off overproduction of cottons. There was huge over-capacity in industry because state-owned enterprises could not be closed down. This led to flooding of world markets by Chinese textiles, which resulted in import restrictions by the USA, the EU, and Japan in 1996 (Becker 1997; see Table A8 for repercussions of previous import restrictions). China’s present situation is different from the late 1980s, when China had to face export problems due to rapidly declining oil prices: oil production is capital-intensive whilst the textile industry is labour-intensive. Due to the social problems involved, a similar swift reduction in exports does not seem possible. 4 The most recent phase in the history of China’s foreign trade can be described as an ‘industrial-policy’ attempt to export more sophisticated commodities (see Jing and Yang 1995). However, it is not yet clear whether these efforts towards a higher share of hi-tech exports will be successful.

Taiwan and Mainland China export dynamics

121

The following section will link the growth in volume and change in structure by analysing growth rates and discussing the role of ‘innovation’ on a macro-economic level. Decomposition of Growth Following a method developed by GATT in the 1960s and refined by the World Bank’s International Economics Department (see WDI 1997, p. 256–9), growth of nominal exports can be decomposed into three multiplicative factors: f1 measures the growth due to expansion of the world market for the country’s ‘traditional exports’,5 which will be called ‘passive expansion’; f2 measures the growth due to expansion of market share for its traditional exports or ‘active expansion’; f3 , measured as a residual, captures the growth in exports due to diversification into non-traditional exports.6 Taiwan’s exports between 1963 and 1969, the first export substitution phase, grew by 21 per cent on average per year. Growth due to world market demand and market share expansion in traditional commodities contributed 55 per cent of this growth rate. Diversification contributed 45 per cent (computed with figures from YITS, various years; due to lack of data, it is not possible to further split up these figures). A decomposition of growth of exports (data from WDI 1997, pp. 257–9) reveals that the bulk of Mainland China’s overall export growth between 1984 and 1993/ 1994 (about 22 per cent per year) was due to an increasing share in world markets for the country’s traditional export products. Nearly two-thirds of this export growth may be attributed to active expansion. Increase of world demand for traditional goods was responsible for less than one-third, whilst only about 10–15 per cent of growth in export volume was due to export commodity diversification. The difference between Taiwan and Mainland China becomes obvious also from a different comparison: the top ten export commodities on the 3-digit level of the SITC contributed 72 per cent to Taiwan’s overall exports in 1963, while the top ten SITC 3-digit groups contributed only 40 per cent to Mainland China’s overall exports in 1980. This means that Mainland China’s export commodities were considerably diversified at the start of its rapid export expansion phase. This resulted from China’s industrialization in the 1950s to 1970s, isolated from the world market. That was a different, though not unblemished Chinese success story, nowadays often forgotten. These findings imply that Mainland China’s export expansion was less innovation-driven than Taiwan’s export expansion in the same phase of export development. The pattern of Mainland China’s export performance was different also from other latecomers to industrialization in East Asia, like Indonesia, Malaysia, and Thailand, whose export growth was due to growth in world demand and export diversification (see Table A9). Furthermore the Chinese pattern was also different from the present pattern of more industrialized countries in East Asia, like Japan

122

Hans H.Bass

and the Republic of Korea as well as contemporary Taiwan. There, particularly in the period 1988–94, diversification predominantly meant expansion to the service sector. Twisting the cord in the 1990s Economic linkages between Taiwan and Mainland China Previously, it was argued in this paper that there are similarities in the development process of Taiwan’s and Mainland China’s exports. The foreign trade pattern of Mainland China can be seen to be one phase behind Taiwan. This explains why the export performance of both economies is now linked by a dynamic regional interaction.7 According to Akamatsu’s and Kojima’s flying-geese pattern (Hobday 1996; Hwang 1998), economic development in East Asia is partly perceived as a transmission process between countries. If industries lose cost competitiveness in one country because of rising unit labour costs, production is transferred to an economically less advanced country. At the same time, the more advanced country builds up new, more capital- and technology-intensive production lines. This framework links a micro-economic or sectoral product life cycle theory and a theory of foreign trade based on specialization. However, it can also be understood as a Schumpeterian process of innovation and imitation (Welfens 1989, p. 46). The basic linkage instrument between the economies of Taiwan and Mainland China is foreign direct investment. Mainland China policy planners after 1978 based their development strategy on the inflow of foreign investment and enforced policies to attract foreign investment in the early 1990s, as pointed out in the previous section. There were several reasons for Mainland China planners to be interested in investment from Taiwan. Taiwanese enterprises had experience in supplying world markets with light-industry products. They targeted markets in the industrialized economies rather than the domestic Mainland China market. Finally, there were intentions by policy-makers to strengthen non-political relations across the Taiwan Strait (Herrmann-Pillath 1994, p. 131, referring to Maruyama Nobuo (ed.) Kanan keizai-ken 1992). Meanwhile, Taiwanese enterprises were keen to invest overseas, and when Taiwanese restrictions on investing in Mainland China were revoked from 1990, they wanted to invest in China. This was because unit labour costs had been rising during industrialization due to changes in availability of labour, capital, and ecological resources. Since 1985 there had been continuing trade surpluses with the main export market, the United States, and this had led to pressure by the USA to adjust the trade balance. So, the Taiwanese currency was upvalued. Another reason was that there were lower transaction costs in an environment with cultural affinity. More recently, market-seeking investments have become important. Finally, investment incentives for Taiwanese tongbao (compatriots) may have played a role, such as the positioning of the four original Special Economic Zones in nearby Fujian and Guangdong provinces. Taiwanese enterprises were allowed to supply the domestic market earlier than foreign enterprises. However, the net effect of these

Taiwan and Mainland China export dynamics

123

Figure 6.3 Taiwanese direct investment in various countries 1985–96 (outflows per year in mill. S$). Sources: TSDB 1996; InvCom, Statistics…(Tongji Yuebao), Dec. 1996. 1993 data for Mainland China adjusted with an estimation by Long 1994.

incentives, set against a lack of bilateral investment protection agreements, can be questioned. Direct investment by Taiwanese enterprises—officially recorded since 1990 (see Figure 6.3)—has contributed heavily to the productive potential of mainland enterprises. According to Mainland China data, nearly 60 per cent of 1995 cumulative foreign investments came from Hong Kong—much from disguised Taiwanese sources, as Hong Kong dummy investors were used by Taiwanese companies to overcome legal obstacles for investment in Mainland China. This was particularly so prior to 1990. In 1995, about 18 per cent of gross investment in Mainland China came from abroad,8 but 30 per cent of Mainland China’s exports came from foreign-funded companies. According to official mainland sources, 10 per cent of foreign direct investment in the mainland currently comes from Taiwan. Taiwanese investments in Mainland China have significantly stimulated export volumes and have contributed to the upgrading of Mainland China’s export structure. On the other hand, there was product and process innovation in some sectors of the Taiwanese economy, which kept production lines in Taiwan in spite of rising wages. These observations now set the stage for investigating the performance of Chinese exporters in geographically defined commodity markets. Market share development Previously similar phases of export development were compared at different points in history.9 The present section compares the export performance of different devel

124

Hans H.Bass

Figure 6.4 Chinese SITC-76 import market shares in USA 1988–96. Source: See Table A8.

opment phases at the same point in time. With data from the three largest world markets, the USA, Japan and Germany (see Table A8) it is possible to distinguish three patterns of Chinese market share development in the 1990s. The first pattern depicts a situation where the market share of Taiwanese exporters (and those from Hong Kong) fell, while Mainland China’s share increased. This is a development which can be explained by the flying-geese model: the more advanced economy loses comparative advantages in labour-intensive industries, while the less advanced one is able to succeed in these markets, an effect triggered by the movement of Taiwanese industries to Mainland China, e.g. in electronics (see Table A10). A good example is telecommunications (SITC 76), including television and radio receivers, sound recorders, and television recorders. As can be seen in Figure 6.4, the Taiwanese share of the US import market for these products fell from 8 per cent to 4 per cent in the early 1990s, Hong Kong’s share fell from 3 per cent to 1 per cent, while Mainland China exporters increased their share from 5 per cent to 13 per cent. The overall increase of the ‘Greater China’ Economic area, however, is not really significant. The products, whose exports developed like this pattern, can be characterized as ‘Heckscher-Ohlin-type goods’ (Giersch 1984): their factor prices determined the export performance. In the second pattern, both Taiwan and Mainland China exporters increased their market share (see Figure 6.5). However, Mainland China may have only gained what Hong Kong lost because workshops were transferred from Hong Kong to Mainland China’s Shenzhen. Examples like this are office machinery and automatic data processing exports to Germany and Japan. The Standard International Trade Classification group 75 (SITC-75), office machinery, includes photo-copying machines, personal computers and other types of automatic data processing machines. Most of these can be classified as

Taiwan and Mainland China export dynamics

125

Figure 6.5 Chinese SITC-75 import market shares in Japan 1988–96. Source: See Table A8.

‘Schumpeter-type products’, incorporating a high share of R&D expenditure and rapid product and process innovation. Thus, these products are not only competitive by price, but also in technology (‘the Schumpeter-type competition’). Taiwanese notebooks—a particular case in point—constitute one-third of world exports, mice 80 per cent, and scanners 70 per cent (Gälli and Franzen 1995, pp. 163–6; Chaponnière and Lautier 1998, p. 244). For these products, Taiwan is a leading exporter, producer and developer. So, this case cannot be interpreted in the framework of the flying-geese model, the transfer of production and export capabilities from Japan to Taiwan (and other NICs). The third pattern of market share behaviour describes a situation where the gains of Mainland China exporters heavily outweighed Taiwanese losses in market share (see Figure 6.6). Striking examples are footwear (SITC 85) exports to Japan and the USA, and clothing in Japan. These products, although also of a HeckscherOhlin-type, are different from the first type, as they are easy to imitate, and product cycles are extremely short. We may conclude that only the second and third patterns are true examples of market innovation, because in the first pattern the shift from one area of origin to the other seems partly due to foreign direct investment in Mainland China from Taiwan. This is an innovation in production location rather than a market innovation. Yet, the two patterns do not fit neatly into the ‘innovators’ and ‘imitators’ of the flying-geese pattern, and suggest an innovation in its own right. In the following section factors will be discussed which may be relevant for the future development of such innovations in Mainland China and Taiwan. The discussion will refer to export promotion policies and to the two most striking differences between the Mainland China and the Taiwan export systems: the size of the enterprises in the export business, and the behavioural flexibility of the companies.

126

Hans H.Bass

Figure 6.6 Chinese SITC-85 import market shares in Japan 1988–96. Source: See Table A8.

A strong common future? Some conjectures Export promotion policies With regard to fostering innovation, three kinds of economic policy can be discerned. First, an optimistic view of the capability of markets to organize production efficiently, at least in the long run (as expressed in Schumpeter’s early views, see Schumpeter 1912). This allows the dynamic elements of an economy, the entrepreneurs, to realize their ideas without much interference by the state. It also tolerates the destruction of enterprises or branches which cannot adjust to the changing economic environment. With respect to catching-up economies, this sort of policy would basically provide information to economic actors, overcome institutional hindrances, and open up possibilities for the private sector, e.g. it allows high rates of profit for the entrepreneurs (see Laumas 1962). Second, an interpretation of the innovation process in line with the later Schumpeter (1942–47), favours government policy which allows institutional arrangements for ‘orderly’ structural change, and targets sunset industries and sunrise industries with detailed programmes. Third, innovation policy can derive from a more complex, ‘neo-Schumpeterian’ understanding of the innovation process (see Freeman 1987), which stresses the importance of innovation systems, ‘sets of institutions whose interactions determine the innovative performance of national firms’ (Nelson and Rosenberg 1993, p. 4). Such policy will promote the National Innovation System, i.e. the technological and institutional capability of a country to innovate, rather than target individual branches of industries. The export policies of the Taiwanese and Mainland China governments will now be evaluated in the light of these approaches. Particular items providing better export opportunities for enterprises in Taiwan

Taiwan and Mainland China export dynamics

127

included (cf. IBRD 1993, pp. 259–346; Chowdhury and Islam 1993; Herderschee 1995, pp. 69–102) tariff rebates on inputs for exports (in the early 1950s) and the establishment of export processing zones with fewer regulations on production. These began in 1965 in Gaoxiong. The share of export processing zone production in total exports was at most 9 per cent in the early 1970s. Furthermore, there were institutional or legal restrictions to entry into the domestic market. Meanwhile production for export was open to new entrants, thus upgrading the production and variety of exports (see Krugman 1983 for more general support for this argument). In Taiwan, industrial policy was geared towards developing new production lines, and was closely linked to trade policies targeting new export markets. This included intra-industry subsidies via taxes on domestic sales, export loans (see IBRD 1993, p. 282), the promotion of small and medium enterprises, and marketing and information assistance (such as fairs or linking up enterprises with counterparts overseas). Direct support for export enterprises seems to have been strong in the 1950s. However, government promoted exports by using a firm’s export performance as a basic criterion for judging access to imports. This was a self-enforcing process, a combination of private-sector dynamism with supportive rather than dirigist government (Chaponnière and Lautier 1998). Later support was indirect only, and by the beginning of the 1990s, most commodity-based export incentives had been phased out. More subtle interactions between government agencies and private enterprises emerged. The most important economic policy in Mainland China contributing to export success was the freeing of enterprises from the central plan. Mainland China’s foreign trade organization in the early 1980s concentrated on making enterprises more autonomous. In the second half of the 1980s the agency system allowed competing Foreign Trade Corporations to take over export commissions (HerrmannPillath 1995, p. 149). Until the mid-1980s, potential exporters were offered fixed prices by the Foreign Trade Corporations. For most of these commodities the prices were the same as the domestic market, thus giving individual enterprises no incentives to export and channelling any gains from exports into central funds. Price reforms, allowing exporters to gain from world demand, contributed to export expansion. Although these changes were an important step towards efficiency, Mainland China’s external sector still does not allow equal competition. Foreignfunded enterprises may import for their own purposes and export their own products, but Foreign Trade Corporations still market domestic-funded enterprises overseas, apart from a few large corporations. The number of Foreign Trade Corporations is very small in comparison to other economies: about 0.3 Foreign Trade Corporations per 100,000 inhabitants, compared to 8 in Japan, 31 in Germany, and 190 in Taiwan. The second crucial instrument in Mainland China’s export promotion policy was the attraction of foreign direct investment. There were regional investment incentives, like the reduction of corporate tax in the Special Economic Zones and other investment promotion zones (such as the Open Coastal Cities). Foreign direct investment significantly contributed to the upgrading of Mainland China’s production structure. From 1995 policy planners did not try to attract foreign

128

Hans H.Bass

investment with regional incentives but with technology-oriented selection criteria. Also new institutions were established, such as the Import/Export Bank, to facilitate hi-tech exports (see Bass 1996). This will be referred to later, when discussing responses to a changing international economic environment. Constraints include stagnating markets and income-inelastic demand for traditional Chinese manufactures, like textiles, and also the national need for labour intensive production. Other ways of promoting export in Mainland China in the 1980s included selective input subsidies and the gradual devaluation of the currency, thus reducing the general bias against exports (see Bass 1996). With regard to the three innovation-oriented approaches to public policy, we may place Taiwan’s present export promotion policy in the second and third category, while Mainland China’s policy is still in the first and second. In Taiwan, ‘guided capitalism’ or ‘governing markets’ (Wade 1990) was successful in the early, or easy periods of the catching-up, and under favourable international conditions. Whether this approach will continue to be successful in the more complex economic environment at the beginning of the twenty-first century, remains to be seen, as state administrations may not be able to keep pace with the need to process information adequately. Typical size of enterprises and innovative potential While Schumpeter originally (1912) saw the new small-scale entrepreneur as the source of innovation, he later thought that big business was the most innovative form of enterprise. Large enterprises have R&D departments and buy specialist knowledge. The ‘neo-Schumpeterian hypothesis’ thus argues that technological innovation is a function of size10 and has positive returns to scale (Wakelin 1997; for an overview of the vast literature about the related issue of market structure and innovation see Baldwin and Scott 1987;11 Rothwell and Dodgson 1996). Some neo-Schumpeterians (e.g. Wakelin 1997) tested the statistical relation between firm size and technological capability by econometric cross-country studies. They also examined the influence of technological capability on the economies’ export performance, including embodied and disembodied technology, such as workforce qualifications. Not surprisingly the impacts are strong when industrialized countries are considered. This section will just consider the effect of the different sizes of foreign trade enterprise in Taiwan and Mainland China on innovation. Taiwan generally is believed to be the paradise of small enterprises. According to a recent official definition, more than 95 per cent of Taiwan’s enterprises are of small or medium size. Economic planners in Taiwan in the past, impressed by the successes of the General Trading companies of Japanese conglomerates, the sogo shosha, attempted to support the development of similar enterprises. However, in Taiwan these large trading companies never really took off. Incentives were very modest, at least compared to what was offered in Korea to support a replication of the Japanese model; competition from foreigners, particularly Japanese trading companies, was fierce; and large trading companies got little support from local business groups (IBRD 1994, p. 112). The dominant form of trading company, of

Taiwan and Mainland China export dynamics

129

which there are presently about 40,000, is extremely small and specialized. These enterprises and small local manufacturers themselves handled two-thirds of Taiwanese exports in the 1980s. Now their share has declined to about half of Taiwanese exports (see Table A11).12 With increasing sophistication of products, small enterprises in Taiwan have clearly lost some of their export competitiveness. In China before 1978 all foreign trade was carried out by a dozen Foreign Trade Companies, each of which had a monopoly in a particular group of commodities. After 1978, in China’s move from a centrally planned economy to a market economy, new Foreign Trade Corporations were created. Provincial branches of national Foreign Trade Corporations became independent, and new Foreign Trade Corporations were created at the provincial or municipal level. These days, several thousand Foreign Trade Corporations exist, many trading in the same sort of commodities. There is a two-layer structure of exporters. There are actual traders, including the Foreign Trade Corporations and some large companies, which are allowed to trade their own products, and there are producers. At the production level, a high share of exports comes from enterprises with foreign investments, either joint ventures with a Chinese partner or wholly foreign-owned ones, and from rural ‘collective’ enterprises (see Taubmann 1996; Deng and Wang 1997). Apart from petty commodity production these are the most decentralized sectors of the Chinese economy. The domestically funded small- and mediumscale enterprises normally do not have an export license, so exports are still channelled through the bigger trading companies (see Tables A12 and A13). The larger ones still dominate: the 10 largest foreign trade companies in Mainland China traded about US$37 bn of imports and exports, i.e. approximately 15 per cent of Mainland China’s total foreign trade in 1995 (http://www.chinatoday.com/ trade/a00.htm).13 The institutional changes so far seem to have been a precondition for changes in China’s export structure. Primary products are homogenous goods sold on international markets at world prices, and little marketing knowledge is needed to sell them. Manufactured goods are more heterogenous and require a more sophisticated understanding of consumer tastes, quality standards etc. (Lardy 1992, p. 697). In future, China may have to carry out more institutional reforms in order to improve its export system, e.g. by further expanding the ‘agency system’. Economies of scale help to increase a company’s market share in traditional export commodities, but diversification seems to require a more decentralized marketing structure. Schumpeter’s later conjecture that ‘the perfectly bureaucratized giant industrial unit…ousts the small and medium-sized firm…’ (Schumpeter 1942; 1947, p. 134), does not seem plausible in China’s case. With the increasing sophistication of export products, the most favourable type of enterprise in the foreign trade sector seems to be neither very small nor very large. The bureaucratic management of large companies can counteract benefits from economies of scale and scope,14 and lack of access to local resources counteracts the benefits of the small-scale traders’ flexibility. Yet the issue of flexibility in economic behaviour is far more complicated

130

Hans H.Bass

than a straightforward correlation with the size of a company may reveal. This issue will be discussed below. Capabilities for adjustment This section discusses the adjustment processes exporters have to face due to changes in world demand. We will start with a review of two concepts helpful in this context, ‘creative/adaptive responsiveness’ and ‘flexibility’. In an article for the Journal of Economic History, Schumpeter distinguished ‘different kinds of reaction to changes in “condition”’ ‘Whenever the economy or an industry or some firms in an industry do something […] that is outside of the range of existing practice, we may speak of creative response’, in contrast to the ‘adaptive response’. There are three characteristics of creative response. First, ‘it cannot be predicted by applying the ordinary rules of inference from the pre-existing facts’; second, ‘creative response shapes the whole course of subsequent events and their long-run outcome […] or, to put it differently, it creates situations from which there is no bridge to those situations that might have emerged in its absence’; lastly, creative response has to do with ‘quality of the personnel’ available in a society, and in the particular field of activity, and with ‘individual decisions, actions, and patterns of behavior’ (Schumpeter 1951, p. 217). A related concept is flexibility, narrowly defined as adaptation to changing demand by reorganization of production (Piore and Sabel 1984; 1989). The precondition for economic flexibility is a low dependency on particular resources (hence: ‘flexible specialization’). A wider definition refers to the ability to react quickly to new circumstances, or to keep intellectual and other assets in a relatively fluid form, i.e. to have a great number of options available (Killick 1995; Klein 1988, p. 105). Schumpeter argued that innovation needs long-term planning and planning security, i.e. a stable economic environment, a temporary monopoly, or patent protection. However, this may create rigidities for short-term reaction. A trade-off between short-term and long-term flexibility exists: ‘using less specific assets may increase flexibility in the short run, but may reduce it in the long run’ (Chang HaJoon 1995, p. 204)—high short-term flexibility may create unnecessary failures. Chinese family businesses, as prevalent in Hong Kong and Taiwan, are typically very specialized in scope. Normally they focus on one market or one product only. Decision making is strictly hierarchical. According to studies in business sociology (Redding 1990; Fukuyama 1995), they are characterized by a high degree of centralization, i.e. personal direction. Ownership, control and family are closely interrelated. They are, incidentally, the prototype of Schumpeter’s bourgeois family background for capitalist enterprises. Normally speaking, growth in size of organization leads from centralization to structuralization. This includes specialization, i.e. division of labour between the persons and organizational departments concerned. It also includes standardization, like formal job descriptions, and formalization of work procedures. However, according to Redding this is not the case in the Chinese family businesses: because

Taiwan and Mainland China export dynamics

131

trust is limited to the family, power is person-embodied and authority subsists only with the owner’s family. Thus paternalism rules, i.e. big and small problems are decided by the boss. External relations of the enterprises are governed by personal relations rather than by formal arrangements, like subcontracting. This can ensure a high degree of flexibility or adaptability. Among the advantages of this type of organization are speed in decision making and a high degree of confidentiality in business information. As regards cross-border trade, the advantages of the Chinese family business can be seen in world market segments, characterized both by rapid change and by strong segmentation (Fukuyama 1995, p. 106), e.g. textiles, apparel, toys, plastics. The Chinese-type family business has a comparative advantage in gaining an overview of a particular market segment (see also Piore and Sabel 1984 for an assessment of market segments particularly suited for small-scale enterprises). There are, however, several disadvantages which occur as enterprises grow. If family relations are more important for promotion than expert knowledge, the middle management may get frustrated. According to Redding, organizations with these restrictions cannot divisionalize, but stay small or continue to grow with the same organizational structure. So Chinese family businesses are non-permanent structures and tend to break up in the third generation, as argued by Fukuyama. Although the Confucian family tradition is an important cultural heritage, the new global economic environment may be a stronger challenge. Getting into more sophisticated commodities needs adaptation, e.g. the incorporation of skills and knowledge from outside the family. According to Porter (1996) the typical ChineseAsian family enterprise, has some advantages, but they diminish in the present world economic environment (see Table below). Taiwan had a family-based business ethic and economic rationality, but family orientation did not have a place in business during most of the post-war history of Table 6.1 Advantages and disadvantages of Asian enterprises under different framework conditions

Source: Adopted from Porter 1996.

132

Hans H.Bass

Mainland China. With regard to behaviour, during the time of great political uncertainty risk avoidance was rewarded (see Redding 1993, p. 235). Some authors, however, point to the fact that the Sino-Stalinist development path did allow for some flexibility niches. The failure of formal institutions demanded flexible informal relations for survival (Herrmann-Pillath 1996, p. 13). Flexibility was a basic condition of life. With regard to enterprises, it should be mentioned that not all enterprises were subject to central planning. Rural industrial enterprises were allowed to act independently (Donnithorne 1972). Today’s managers, the onetime activists of the cultural revolution, may never have been used to stable institutions (Herrmann-Pillath 1995, p. 129). Economic flexibility depends also on access to information and informationprocessing capacity. The quantity of information increased with Mainland China’s opening to the outside world. The capacity to process this information increased in the 1980s with more and more market mechanisms allowed. Prices, as argued by von Hayek (e.g. 1967) contain more information on market demand and supply than a central plan can contain. Another aspect of flexibility becomes obvious when discussing the relation between individual Foreign Trade Corporations and the central authorities. In Mainland China, three categories of external trade can be distinguished according to the information, allocation and control mechanisms: centrally planned, government-guided, and market-guided. For the 1980s the share of the three categories is estimated to be 50–60 per cent centrally planned, 20 per cent guided, and 20–30 per cent free (Gerhald, cited in Sekiguchi 1990, p. 397). Sekiguchi estimates exports under direct plan in 1988 to amount to 30 per cent, those under the foreign exchange quota, i.e. ‘centrally guided trade’, as 20 per cent of total exports. In the 1980s, Foreign Trade Corporations decided their export supply volume according to international prices rather than simply filling their foreignexchange delivery quota. More independence for individual Foreign Trade Corporations, granted in the course of economic reforms from 1984, also allowed more flexible behaviour. In the 1990s, deregulation progressed further. In particular, exporters in the 1990s were permitted to retain a portion of their foreign exchange earnings. Finally, however, there is a strong constraint on the flexibility of the entire export structure due to the structure of domestic production. The national political rule is to hold back ‘destruction’ if an alternative does not create necessary job opportunities. State-owned enterprises in China are in a desperate condition. This is particularly true of the textiles industry. Over past decades central authorities reinvested too little profit from these enterprises, causing them to be short of capital and of too low a technological standard to be competitive. The implications of this situation have been discussed above. This brief overview suggests there is need for further institutional innovation in both economies in order to cope with new challenges of world trade. This may mean a more structured organization of enterprises in Taiwan, and more individual responsibility in enterprises in Mainland China. The idea of a convergence in the type of business organization does not seem out of place, suggesting there are

Taiwan and Mainland China export dynamics

133

potential opportunities for more interaction between the economies of Taiwan and the Mainland, which again can speed up dynamics in this sector. Appendix Table A2 Volume and composition of Taiwan’s exports 1952–95

Sources: GATT 1991; WTO 1998. Lardy 1992, p. 694, however, reports MOFTEC data which even indicate a percentage change of 316% for the first period for Mainland China. Note: Share of the Top-25 exporters in world trade: 83% (1989) and 84% (1997).

Source: TSDB 1988; computation from OECD, ITCS Rev. 3, 4/1998 [Food etc.=SITC 0–1; Minerals etc.=SITC 2–5; Textiles=SITC 65+84; Machinery and Transport Equipment=SITC 7 excluding SITC 75–77; Electrical Machinery=SITC 75–77].

Table A1 Change in merchandise exports, Taiwan and Mainland China 1979–97 (%)

Taiwan and Mainland China export dynamics

135

Table A3 Main export products of Taiwan (representing 75% of export value) 1951–69 (%)

Source: Own computations from YITS, United Nations, Yearbook of International Trade Statistics, various years, New York. Classification according to SITC, Rev. 1, is available for 1962–69 only. After 1969, Taiwan’s trade is no longer reported by this source. nes=not elsewhere specified.

136

Hans H.Bass

Table A4 Main 20 SITC 3-digit level export products of Taiwan 1990–95 (%)

Source: Computation from OECD, ITCS Rev. 3, 4/1998.

Taiwan and Mainland China export dynamics

137

Table A5 Main SITC 4-digit level export products of Taiwan in office machinery and electrical machinery (SITC 75 and 77) 1990–96 (in US$,000)

Source: Computation from OECD, ITCS Rev. 3, 4/1998.

Table A6 Volume and composition of Mainland China’s exports 1965–95

Sources: 1965–90: IBRD 1994:5; 1995: computed from OECD, ITCS Rev. 3, 2/1997. Foods: SITC 0+1+22+4; Agricultural raw materials: 2 less 22, 27, 28; mineral fuels: 3; ores etc.: 27+28+68; textiles and clothing: 65+84.

138

Hans H.Bass

Table A7 Main export products of Mainland China 1980–95 (%)

Sources: 1980, 1985; Yabuki 1995, p. 156 (from: China’s Commercial and Foreign Economic Statistical Materials 1952–1988, pp. 464–75). For COMTRADE statistics, using reported partner country data and being quite different from these data, see Yeats n.d., pp. 46–7. 1995: computed from OECD, ITCS Rev. 3, 2/1997.

Table A8 Greater China’s import shares in major markets of industrial economies

Table A8 Continued

*Starting 15.03.1994, EU restriction came in force for imports of many consumer goods from Mainland China, allowing only a 10 per cent growth of imports compared to the average of 1990–92 imports.

Source: Computation from OECD, Foreign Trade by Commodities 1990, Vol. 3–5, Paris 1991; OECD, ITCS Rev. 3, 1–5/1997.

Note: *Only commodity exports; diversification into service exports, which is particularly relevant for high-income economies like Japan, is not considered here. For the respective countries’ ‘traditional imports’ (i.e. 75% of basis year exports) as of 1983; China: basis 1984.

Source: WDI 1997, pp. 257–9.

Table A9 Decomposition of nominal export growth, various East Asian economies 1983/84–93/94 (%)

Taiwan and Mainland China export dynamics

143

Table A10 Volume and distribution of Taiwanese investment in Mainland China by branches, Taiwanese data 1991–96 (%)

Source: Computed from InvCom, Statistics…(Tongji Yuebao), Dec. 1996. Table A11 Economic relevance of small enterprises in Taiwan 1985–95 (%)

Sources: Chaponnière and Lautier 1998, p. 248 (Data from 1985); IBRD 1993, p. 162 (Data from 1990); CA 1995, pp. 1108–9 (Data from 1995). A different source (IBRD 1994, p. 112), however, indicates that the share of local trading companies is much smaller because of a large share of Japanese General Trading Companies in Taiwan’s foreign trade. The most recent official Taiwanese definition of SME includes enterprises –with a paid-in capital of less than NT$40 mill.; –or total assets less than NT$120 mill.; –or in the case of exporters/importers: annual sales revenues of less than NT$40 mill. (Chaponnière and Lautier 1998, p. 248). Table A12 Mainland China’s exports by type of exporting enterprise 1994

Source: DIW and ITC 1997, p. 59; von Kirchbach and Aguado 1996, p. 71. Although using the same source quite different data on number of firms: IBRD 1997, p. 13. The number for FTCs given in this source is 9,400. Table A13 Mainland China’s exports by type of producing enterprise 1994

Source: Computed with data from China Customs Statistics, as quoted in Taubmann 1 995.

144

Hans H.Bass

Notes 1 This argument is related to the debate on the causal relation between productivity growth and export performance. IBRD 1993, pp. 316–24 argues that exports helped East Asian economies to increase total factor productivity rather than vice versa: this was due not only to static factors such as economies of scale, but more by dynamic factors like improving knowledge and adopting international best-practice technologies. For an overview on this debate, and a contrasting opinion see Chowdhury and Islam 1993, pp. 79–87. 2 On the considerable data problems for Mainland China’s foreign trade, see e.g. Lardy 1995. 3 These figures of 1980–95 were outstripped only by three other ‘High Performing Asian Economies’: the Republic of Korea (14 per cent), Thailand (16 per cent), and Vietnam (18 per cent), and met by some European ‘newcomers’ (or: newcomers to the new European economic core): Portugal (12 per cent), Ireland (13 per cent), and Turkey (13 per cent). 4 The same is applicable for commercial services, where Mainland China is now ranking 16th, and Taiwan 19th, but this segment of world trade is not dealt with in the present paper as assessing the dynamics of international trade in service still poses considerable data problems (see Grupp 1998, p. 205) 5 ‘Traditional’ exports of a country are defined in this source as the 3-digit commodity groups that made up at least 75 per cent of the value of the country’s exports in 1983/84 and included at least the 10 largest commodity groups (WDI 1997, p. 259). 6 The algebraic relation can be expressed as X1/X0=f1. f2. f3

are the country’s traditional exports, W are world imports in these commodity groups, X are the country’s total exports, 0 and 1 designate the first and the last year of the period under consideration. 7 Whether interaction, interdependence or integration are the appropriate terms to describe what is happening between the two economies, may be disputed: e.g. see Crane 1993, Hong 1996, Bass 1998a. 8 As with figures on foreign trade, caution is also necessary with figures on foreign direct investment in China: first, there is the problem of re-routing indigenous capital via Hong Kong in order to get subsidies and other benefits for foreign investors; second, there are ‘fake’ joint ventures established to allow domestic enterprises to get hold of import (and export) licences. 9 For a comparable approach, using data from Mainland China and Japan, see Sekiguchi 1990. 10 More precisely, the size hypothesis in this formulation should be attributed to J.K.Galbraith (see Grupp 1998, p. 56). 11 The authors conclude: ‘There is no unambiguous evidence of an important, generally valid, relationship between competition and innovative activity. Case studies provide thought-provoking possibilities. But where statistically significant relationships have been found, the explanatory power is small.’ Baldwin and Scott 1987, p. 145. 12 According to a guess introduced into literature by Gälli (1980, cited in Wade 1990, p. 147), 30 or even 50 per cent of Taiwan’s foreign trade is said to be handled by Japanese so¯go¯ sho¯sha in the late 1970s. This figure is also referred to by the World Bank (IBRD 1994), but is contradicted by more detailed data from Taiwanese sources (see Table A11). 13 However, compared to countries like Brazil this may not be considered a huge concentration (see von Kirchbach and Aguado 1996).

Taiwan and Mainland China export dynamics

14

145

This is the general reason why in the 1980s and 1990s large enterprises divisionalized, used ‘outsourcing’, or even split-up into smaller semi-independent ones.

References ‘Top Ten Trading Companies in China in 1995’, http://www.chinatoday.com/trade/a00.htm Abramovitz, M. (1995) ‘The origins of the postwar catch-up and convergence boom’, in J.Fagerberg, B.Verspagen, N.van Tunzelmann (eds) The Dynamics of Technology, Trade and Growth, Aldershot: Edward Elgar, pp. 21–52. Baldwin, W.L. and Scott, J.T. (1987) Market Structure and Technological Change, Chur etc.: Harwood. Bass, H.H. and Wauschkuhn, M. (1995) ‘Chinas Außenhandel’, in H.H.Bass and M. Schüller (eds) Weltwirtschaftsmacht China, Hamburg: Institut für Asienkunde, pp. 76– 99. Bass, H.H. (1996) ‘Chinas Außenhandel mit den Industrieländern zwischen Exportförderung und Importprotektionismus’, in H.H.Bass and K.Wohlmuth (eds) China in der Weltwirtschaft, Hamburg: Institut für Asienkunde, pp. 32–53. —(1998a) Wirtschaftskooperation zwischen Festlandchina und Taiwan—Potentiale und Grenzen, Berichte des Arbeitsbereichs Chinaforschung im Institut für Weltwirtschaft und Internationales Management der Universität Bremen, Nr. 11: Universität Bremen. —(1998b) J.A.Schumpeter. Eine Einführung (Gastvorlesungen an der Aichi-Universität, Toyohashi/Japan), Materialien des Universitätsschwerpunktes ‘Internationale Wirtschafts-beziehungen und Internationales Management’, Bd. 12: Universität Bremen. Becker, J. (1997) South China Morning Post, 04.01.1997, cited from the German translation ‘Shanghais traditionsreiche Textilindustrie produziert nur noch Arbeitslose’, der überblick, 4/97, pp. 70–1. Brender, A. and International Monetary Fund (IMF) (1992) China’s Foreign Trade Behavior in the 1980s: An Empirical Analysis, IMF-Working Paper WP/92/5. Carlsson, B. and Jacobsson, S. (1996) ‘Technological systems and industrial dynamics: implications for firms and governments’, in E.Helmstädter and M.Perlman (eds) Behavioral Norms, Technological Progress, and Economic Dynamics. Studies in Schumpeterian Economics, Ann Arbor: The University of Michigan Press, pp. 261–83. Chang Ha-Joon (1995) ‘Explaining “Flexible Rigidities” in East Asia’, in T.Killick (ed.) The Flexible Economy. Causes and Consequences of the Adaptability of National Economies, London and New York: Routledge, pp. 197–221. Chaponnière, J.R. and Lautier, M. (1998) ‘Industrial policy for catching up. The case of Taiwan’, in M.Storper, S.B.Thomadakis and L.J.Tsipouri Latecomers in the Global Economy, London and New York: Routledge, pp. 224–53. Chowdhury, A. and Islam, I. (1993) The Newly Industrialising Economies of East Asia, London and New York: Routledge. Crane, G.T. (1993) ‘China and Taiwan: not yet “Greater China”’, International Affairs, 69, No. 4, pp. 705–23. CSYB=China Statistical Yearbook, various years, Beijing. Deng Li and Wang Yan Zhong (1997) ‘Gao hai zhong de xiao chuan—Zhongguo xiao qiye chukou ji fazhan fangxiang’, Guoji Maoyi, 5, pp. 23–7. DIW/ITC (1997)=Deutsches Institut für Wirtschaftsforschung and International Trade Centre UNCTAD/WTO (1997) China als Handelspartner und Produktionsstandort für deutsche mittelständische Unternehmen (Bearbeitung: F.von Kirchbach, H.Trabold et al.), Berlin: DIW. Donnithorne, A. (1972) ‘China’s cellular economy: Some economic trends since the Cultural Revolution’, China Quarterly, No. 52, pp. 605–19.

146

Hans H.Bass

Dosi, G., Pavitt, K. and Soete, L. (1990) The Economics of Technical Change and International Trade, New York: Harvester Wheatsheaf. Eliasson, G., Green, C. and McCann, C.R. Jr. (eds) (1998) Microfoundations of Economic Growth. A Schumpeterian Perspective, Ann Arbor: The University of Michigan Press. Fagerberg, J., Verspagen, B. and van Tunzelmann, N. (1995) ‘The economics of convergence and divergence: an overview’, in J.Fagerberg, B.Verspagen and N.van Tunzelmann (eds) The Dynamics of Technology, Trade and Growth, Aldershot: Edward Elgar, pp. 1–20. Fei, J.C., Ohkawa, K. and Ranis, G. (1985) ‘Economic development in historical perspective: Japan, Korea, and Taiwan’, in K.Ohkawa and G.Ranis Japan and the Developing Countries: A Comparative Analysis, pp. 35–64. Freeman, C. (1987) Technology Policy and Economic Performance. Lessons from Japan, London, New York: Pinter Publishers. —(1995) ‘Technological revolutions and catching-up: ICT and the NICs’, in J.Fagerberg, B.Verspagen and N.van Tunzelmann (eds) The Dynamics of Technology, Trade and Growth, Aldershot: Edward Elgar, pp. 198–221. —(1996) ‘Catching-up and falling behind: the case of Asia and Latin America’, in J.de la Mothe and G.Paquet (eds) Evolutionary Economics and the New International Political Economy, London: Pinter, pp. 160–79. Fukuyama, F. (1995) Trust. The Social Virtues and the Creation of Prosperity, New York: The Free Press; cited from the German translation: Konfuzius und Marktwirtschaft. Der Konflikt der Kulturen, München: Kindler. Gälli, A. and Franzen, J. (eds) (1995) Die Familie des großen Drachen. Band I: Die VR China, Hong Kong, Macao und Taiwan auf dem Weg zu ‘Großchina’?, München etc.: Weltforum-Verlag. GATT (1991)=General Agreement on Tariffs and Trade International Trade 89–90, Geneva. Giersch, H. (1984) ‘The Age of Schumpeter’, in American Economic Review, Papers and Proceedings, 74, 2, pp. 103–9. Grupp, H. (1998) Foundations of the Economics of Innovation. Theory, Measurement and Practice, Aldershot: Edward Elgar Publishing. Herderschee, J. (1995) Incentives for Exports, A case study of Taiwan and Thailand (1952– 1987), Aldershot: Avebury. Herrmann-Pillath, C. (1994) Wirtschaftsintegration durch Netzwerke: die Beziehungen zwischen Taiwan und der Volksrepublik China, Baden-Baden: Nomos. —(1996) Reflektionen über den Zusammenhang zwischen kultureller Modernisierung, Wirtschaftswachstum und Chinas Stellung in der Weltwirtschaft, hektogr. Beitrag zum Internationalen Symposium ‘Ordnungsreform und Entwicklung der chinesischen Wirtschaft in den 90er Jahren’, Giessen. Hobday, M. (1996) ‘Innovation in East Asia: diversity and development’, in M.Dodgson and R.Rothwell (eds) The Handbook of Industrial Innovation, Cheltenham and Brookfield: Edward Elgar, pp. 94–105. —(1995) Innovation in East Asia, The Challenge to Japan, Cheltenham: Elgar. Hong Zhong (1996) ‘Instrumente und Perspektiven der Integration des Großchinesichen Wirtschaftsraums’, in H.H.Bass and K.Wohlmuth (eds) China in der Weltwirtschaft, Hamburg: Institut für Asienkunde, pp. 127–48. Hwang Sun-Gil (1998) Das Wildgänsemodell für Ost-Asien, Arbeitspapiere zur sozialökonomischen Ost-Asien-Forschung, Bremen: Social Economic Action Research Institute, University of Bremen. IBRD-World Bank (1993) The East Asian Miracle: Economic Growth and Public Policy, New York and Oxford: Oxford University Press. —(1994) China. Foreign Trade Reform, Washington, D.C.: The World Bank.

Taiwan and Mainland China export dynamics

147

—(1997) China 2020: China Engaged. Integration with the Gobal Economy, Washington, D.C: The World Bank. InvCom (1996)=Investment Commission Ministry of Economic Affairs, R.O.C. (1996) Statistics…, Tongji Yuebao, December. Jing Chai and Yang Dan (1995) ‘A projection for China’s foreign trade in the upcoming 6 years’, in China Economic News, No. 5, pp. 6–8. Killick, T. (ed.) (1995) The Flexible Economy. Causes and Consequences of the Adaptability of National Economies, London and New York: Routledge. Klein, B.H. (1988) ‘Luck, necessity, and dynamic flexibility’, in H.Hanusch (ed.) Evolutionary economics. Applications of Schumpeter’s ideas, Cambridge: Cambridge University Press, pp. 95–136. Krugman, P. (1983) ‘New theories of trade among industrial countries’, American Economic Review. Papers and Proceedings, Vol. 73, pp. 343–47 Lall, S. (1995) ‘The creation of comparative advantage: Country experiences’, in I.ul Haque et al. (eds) Trade, Technology, and International Competitiveness, Washington: The World Bank, pp. 135–54. Lardy, N.R. (1992) Foreign Trade and Economic Reform in China (1978–1990), Cambridge: Cambridge University Press. —(1995) ‘The role of foreign trade and investment in China’s economic transformation’, The China Quarterly, pp. 1065–82. Laumas, P.S. (1962) ‘Schumpeter’s theory of economic development and underdeveloped countries’, in J.C.Wood (ed.) (1991) J.A.Schumpeter. Critical Assessments, London and New York: Routledge, pp. 407–13. Long, S. (1994) ‘Regionalism in Fujian’, in D.S.G.Goodman and G.Segal China Deconstructs, London, pp. 202–23. Nelson, R.R. and Rosenberg, N. (1993) ‘Technical innovation and national systems’, in R.R. Nelson (ed.) National Innovation Systems. A Comparative Analysis, New York and Oxford: Oxford University Press, pp. 3–21. North, D.C. (1991) ‘Institutions’, Journal of Economic Perspectives 5 (1), pp. 97–112. OECD, International Trade by Commodities ICTS, Rev. 3, CD-Rom, various years. Ohmae Keinichi (1995) The End of the Nation State. The Rise of Regional Economies, New York etc.: The Free Press. Okuda Satoru (1994) ‘Taiwan’s trade and FDI policies and their effect on productivity growth’, The Developing Economies, 32–4, pp. 423–43. Piore, M.J. and Sabel, C.F. (1984) The Second Industrial Divide. Possibilities for Prosperity, New York: Basic Books; cited from the German translation: Das Ende der Massenproduktion, Frankfurt: Fischer 1989. Porter, M. (1996) Far Eastern Economic Review, cited from AP-Newsletter 5/09.04.96. Ramamurthy, B. (1998) ‘The Asian Experience I: China and Taiwan’, in Ronnås, Ö.Sjöberg and M.Hemlin Institutional Adjustment for Economic Growth. Small scale industries and economic transition in Asia and Africa, Aldershot: Ashgate, pp. 37–80. Redding, S.G. (1990) The Spirit of Chinese Capitalism, Berlin and New York: de Gruyter (2nd printing 1993). Rothwell, R. and Dodgson, M. (1996) ‘Innovation and size of firm’, in M.Dodgson and R.Rothwell (eds) The Handbook of Industrial Innovation, Cheltenham and Brookfield: Edward Elgar, pp. 310–24. Schumpeter, J.A. (1912) The theory of economic development: an inquiry into profits, capital, credit, interest, and the business cycle, cited from the 8th printing (1968), Cambridge, Mass.: Harvard University Press. —(1947) Capitalism, socialism, and democracy, cited from the 2nd American edition, New York: Harper & Brothers.

148

Hans H.Bass

—(1951) Essays on Economic Topics (ed. by R.V.Clemence), Port Washington N.Y.: Kennikat Press. Sekiguchi Sueo (1990) ‘Foreign trade in Chinese economy: prices and price responsiveness’, The Developing Economies, Vol. 28–4, pp. 390–417. Taubmann, W. (1996) ‘Außenwirtschaftliche Aktivitäten ländlicher Industrieunternehmen in der VR China’, in H.H.Bass and K.Wohlmuth (eds) China in der Weltwirtschaft, Hamburg 1996: Institut für Asienkunde, pp. 54–65. TSDB=Taiwan Statistical Data Book (1996), compiled by Council for Economic Planning and Development, Taipei. von Hayek, F.A. (1967) ‘The economy, science, and politics’, in: Studies in Philosophy, Politics and Economics, London: Routledge and Kegan Paul, pp. 251–69. von Kirchbach, F. and Aguado, J. (1996) ‘China’s foreign trade expansion: The changing face of Chinese exporters and importers’, in H.-C.Bettignies (ed.) Business transformation in China, London etc.: International Thomson Business Press, pp. 61– 84. Wade, R. (1990) Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization, Princeton: Princeton University Press. Wakelin, K. (1997) Trade and Innovation. Theory and Evidence, Aldershot: Edward Elgar Publishing. Weber, M. (1972) Wirtschaft und Gesellschaft. Grundriss der verstehenden Soziologie, 5. rev. ed., Tübingen: J.C.B.Mohr (Paul Siebeck). WDI (1997)=World Development Indicators, Washington, D.C. Welfens, P.J.J. (1989) ‘Schumpetersche Prozesse in der Weltwirtschaft’, List Forum für Wirtschafts- und Finanzpolitik, Vol. 15, H. 1, pp. 40–60. WTO (1998)=World Trade Organization, World Trade Growth Accelerated in 1997, Despite Turmoil in Some Asian Financial Markets (19.03.1998) http://www.wto.org/wto/intltrad/ internat.htm Yabuki Susumu (1995) Zusetsu Chuugoku no keizai, cited from the English translation China’s New Political Economy. The Giant Awakes, Boulder etc.: Westview Press. Yamamoto Yasuko (comp.) (1997) Trade structure of Hong Kong, Taiwan and China, and Hong Kong re-export trade statistics, Tokyo: Institute of Developing Economies. Yeats, A, (n.d. [1991]) China’s Foreign Trade and Comparative Advantage: Prospects, Problems, and Policy Implications, World Bank Discussion Papers No. 141, Washington, D.C.: The World Bank. YITS=United Nations, Yearbook of International Trade Statistics, various years, New York.

7

Industrialization and institutional change in Hong Kong 1842–19601 David W.Clayton

This chapter will assess how institutions aided the operation of markets in Hong Kong and why they emerged, by outlining how chambers of commerce and the state provided market information and regulated transactions. It will also assess whether Hong Kong fits the model of institutional change developed by Douglass North. Institutions and economic development North and others argue that if an economy has good institutions it will develop and if it has bad institutions it will stagnate (Davis and North 1971; North and Thomas 1973; North 1990 and 1991).2 These institutions can be divided into informal ones, such as customary practices and codes of conduct, and formal ones, such as chambers of commerce, guilds and the state, which provide market information and regulate transactions by using bureaucracies and legal systems. North believes that as an economy develops, informal institutions will give way to formal ones, as part of a two-stage process. First merchant associations, by codifying and extending traditional rules governing trade, establish private property rights. Initially such institutions greatly improve the flow of information and reduce the need for expensive contractual agreements between members. But eventually, because they constrain market integration by encouraging intra-association trade, private property rights become an obstacle preventing further economic development. 3 Consequently, a second stage is required when the state codifies private (often regional) property rights into a set of public national property rights. North holds that this later development, which greatly expands the potential size of the market, is aided by the coexistence of a state large and powerful enough to protect merchants’ interests and mercantile élites sufficiently strong enough to protect their own interests vis-à-vis the state.4 Once established, these economy-wide institutions significantly reduce the ‘transaction’ and ‘information’ costs of doing business. Alternative agents for meeting and reducing these costs are the firm and the market (Casson 1997; Williamson 1975 and 1985). The firm internalizes these costs by creating bureaucracies and hierarchies to process information. Thereafter these structures reduce costs by reaping economies of scale or by applying new technologies, such as accounting practices or computers. By comparison, the market 149

150

David W.Clayton

sets a price for information and for assessing and managing risk—sometimes in the form of insurance policies—and reduces such costs by maintaining competition between providers. While it is virtually impossible to measure these ‘information’ and ‘transaction’ costs it is worth considering the factors that would have influenced their level in Hong Kong until 1960. The first point to note about Hong Kong is that its small size, and the high density and concentration of businesses within it, improved internal information flows, allowing manufacturers to exploit ‘external’ economies of scale. By the end of the 1950s if a customer wanted a product made or wanted to acquire one, there were clusters of manufacturers and merchants in Hong Kong, supplying information on price and quality in a cost-effective way. Equally Hong Kong, an international entrepôt since the nineteenth century, had excellent shipping, telegraph and telephone links with the outside world, complemented in the postwar period by the rapid development of passenger and freight air transport. Such excellent external communication links, delivered by a large number of highly competitive private firms, reduced transport costs and delivery times for Hong Kong manufacturers. Despite these technological and economic advantages, certain features of the Hong Kong economy pushed up ‘information’ and ‘transaction’ costs. First, as Hong Kong had a small domestic market and limited primary resource endowment, firms had to transact with a diverse range of overseas customers and suppliers—a market structure exaggerated from 1949 by the collapse of Hong Kong’s main pre-war market, China. To do so required extensive market knowledge and numerous regulatory strategies to deal with different legal and cultural systems abroad. Second, from the 1930s, in response to the growth of protectionist measures, access to most world markets required detailed documentation. In Hong Kong’s case these costs were potentially much higher because it transacted with a greater number of overseas markets, and because, after 1951, due to the economic embargo imposed on Communist China by the West, each Hong Kong firm required documentation to verify that its exports had not been sourced in that country. Such rules increased ‘transaction’ costs. If, for example, an exporter failed to supply the correct documents with a shipment of products, customs and excise would investigate them on arrival, damaging the reputation of Hong Kong as a supply centre by delaying or preventing delivery. Third, industrial firms in Hong Kong could not easily internalize ‘transaction’ and ‘information’ costs by vertically integrating into sourcing or marketing. Unlike their competitors in the USA, Europe and particularly Japan, they neither had the resources (capital, labour and especially land), the guaranteed large domestic market, nor, given the threat of a Chinese takeover, the long-term property rights to pursue this strategy. Instead, small-scale industries in Hong Kong, which emerged extremely rapidly from the late 1940s onwards, bought raw materials and marketed products through mercantile middle-men. These market transactions were underpinned by a complex web of (non-firm) institutions which had evolved over 100 years of Hong Kong’s modern history, which is a story we now need to tell.

Institutional Change in Hong Kong

151

Hong Kong’s institutional evolution There are four institutions we need to examine: guilds, sub-national community groups, community-wide chambers of commerce and the state. We will focus on the last two but begin with an overview of the others. Guilds Throughout this period most businesses were members of guilds, institutions which represent employers in the same trade. There have been no comprehensive studies of Hong Kong’s guilds so it is difficult to estimate how quickly they emerged in the nineteenth century and how their functions changed over time. Chung however does note that Chinese tradesmen such as boatmen and wine dealers set up ‘occupation organizations’ from the 1840s, and that thereafter their number and scope increased as Hong Kong developed (Chung 1998). Certainly by the end of our period, the vast majority of trades had their own organizations, with 70 trade associations registered under the colonial state’s 1965 Companies Ordinance, including manufacturers and merchants, making, or trading in, both ‘traditional’ Chinese products and ‘modern’, factory-produced ones.5 These associations provided members with information, represented them in discussions with the government and with other commercial organizations, and may have regulated price fixing and market sharing agreements between member firms. But, on the whole, it seems unlikely that guilds had a significant impact on Hong Kong’s institutional development. This is partly because most trades comprised a large number of small firms which were difficult to organize, but also because, in the absence of powerful trade unions and a state responsive to sectional interests, there was less need for lobbying bodies in Hong Kong. Consequently, guilds varied from very cohesive powerful bodies, providing good commercial information, to very loose organizations, acting more like social clubs. Sub-national community groups By comparison, native place or dialect groups, such as the Hokkienese from Fukien province and the Teochiu-speaking Chinese from north-eastern Guangzhou, provided good institutional support to migrants from different parts of China. Some of these bodies later developed into (or became incorporated into) larger regional groupings of Chinese migrants, an institutional response common to many overseas Chinese communities but slower to emerge in Hong Kong than elsewhere. There were also informal groupings of merchants from the Indian subcontinent, notably the tight-knit group of wealthy Parsi traders who traded in opium and other products (White 1994). All these sub-national community groups emerged primarily for social and political reasons rather than commercial ones; they acted as social clubs, burial agents, charity providers and a conduit for remittances and political funds back to home provinces. But they also reduced ‘information’ and ‘transaction’ costs by improving business ethics within the community. They often did this by

152

David W.Clayton

establishing ready-made authority figures to mediate during commercial disputes, a practice in keeping with the Confucian codes of conduct that ordered Chinese society. In Hong Kong, there were close membership overlaps between guilds and these sub-national groups given that trade in many primary commodities was monopolized by certain ethnic, or sub-regional groupings (Sinn 1990; Tsai 1993).6 In the twentieth century, new sub-national bodies emerged and existing ones drew up articles of association and registered with the state, which increasingly wanted to monitor any organizations perceived to be a threat to political stability in the colony. The most prominent of these new more formalized and politicized institutions in the early twentieth century was the Siyi Chamber of Commerce. Established in 1909, as a more formalized grouping of wealthy merchants from Xingshan and four other counties in China, it was banned by the colonial state in 1917, on the premise that it had close financial and political links with the government in Guangzhou (Chung 1998). But it proved to be an important catalyst for change, with fourteen further Chinese regional chambers established in the 4 years after it was founded. In the post-war period sub-national organizations continued to emerge and formalize, the most prominent of which were the Indian, American and Japanese Chambers of Commerce.7 Chambers of Commerce and the Manufacturing Association Higher up the institutional ladder in Hong Kong were community-wide organizations. We will discuss the three most important ones here: the Hong Kong General Chamber of Commerce, the Chinese General Chamber of Commerce and the Chinese Manufacturers’ Association of Hong Kong. The Hong Kong General Chamber of Commerce (henceforth the Chamber), founded in 1861, was the main organization supporting European firms. The Chamber’s remit was: to watch over and protect the general interests of Commerce, to collect information on all matters of interest to the Mercantile Community, and to use every means within its power for the removal of evils, the redress of grievances, and the promotion of the common good; to communicate with Authorities and others thereupon; to form a code of practice whereby the transactions of business may be simplified and facilitated; to receive references, and to arbitrate between disputants. (Hong Kong General Chamber of Commerce 1954)8 According to its official history, the Chamber formed to provide a more formal mechanism for communicating mercantile views to the government, and subsequently the Chamber did indeed influence colonial state policy over freight rates, postal provision, the currency, the system of commercial law and the level of taxation. The Chamber had 59 members in 1861, comprising all the main European merchant houses. Once established, the Chamber undertook economic functions which made it indispensable to expatriate firms.

Institutional Change in Hong Kong

153

Of some use to firms was the market information provided both formally and informally by the Chamber. By attending meetings and social gatherings, members learnt of business opportunities by word of mouth, supplementing information gained from other social arenas, such as horse racing meets, boating regattas and cricket matches. The existence of a cohesive and exclusive élite group of European businessmen made access to these informal networks vital for the success of Western expatriate businesses. The Chamber also formally received and disseminated business enquiries from abroad, a service labelled by the membership in 1953 as of ‘considerable value’ (The Chamber 1953). Annual reports by comparison updated members on legal changes, provided data on harbour and shipping services, and disseminated detailed market reports. The Chamber’s links with international organizations, such as the Federation of British and Commonwealth Chambers of Commerce and the International Chamber of Commerce, eased this flow of information, while the Chamber’s library, containing works of reference, chambers of commerce journals, market reports and periodicals and commercial directories, allowed members to verify and follow up contacts. The Chamber regulated intra-chamber trade by establishing rules on bankruptcy and business ethics, and also represented its members in contractual disputes with non-member firms, functions that were eventually formalized with the establishment of an arbitration subcommittee, to which members paid fees, costs and expenses.9 Once again, the Chamber’s links with organizations overseas, such as chambers of commerce in London, Manchester and Bradford, oiled this regulatory machine. And, if we are to believe the Chamber’s own audit of its regulatory functions, then it was successful; out of 179 commercial disputes it dealt with in 1956, over twothirds had been settled by the end of the year, while the vast majority were settled by early 1957.10 In addition, the Chamber also negotiated general contracts for its membership. These set levels for deviations from average quality and acceptable delivery dates for trade in products such as rice and cotton and woollen piecegoods.11 The Chamber was usually responsible for regulating these contracts, as it inspected goods before or after shipment and issued documentation to back up claims for compensation. These general contracts were a useful institutional response to the high but unstable levels of demand in the immediate post World War I period—economic conditions which encouraged speculative and corrupt business practices and caused a high number of broken contractual agreements (Pennell 1961). Unsurprisingly, given its racial bar on Chinese membership, Chinese businessmen did not use the services provided by the Hong Kong General Chamber of Commerce. Instead they turned to their own organizations. The Man Mo Temple, ‘the focal point of Chinese community life’ in the 1850s, was the first of such bodies to emerge. Its committee members, prominent merchants from many different dialect groups, settled community disputes and, more generally, looked after the civil and religious affairs of the Chinese community (Tsai 1993, pp. 50– 51). Its successor organization, the Tung Wah Hospital Committee, became the centre of the Chinese community from the 1870s, and, once again, used the ‘social influence and moral authority’ of the Chinese mercantile élite to regulate society

154

David W.Clayton

(Sinn 1989). When it declined from the late nineteenth century, it was because widening divisions within this élite, combined with greater state intervention in Chinese community affairs, undermined its authority. During the nineteenth century, the Nam Pak Hong was also an important institution. It was established by merchants of various dialect groups ‘to promote members’ welfare and market prosperity’ and originally comprised merchants trading in primary products between the north and south of China, who had become wealthy by selling to the large numbers of Chinese emigrants in the Pacific basin (Tsai 1993, pp. 62–64). A Chinese General Chamber of Commerce was originally established in 1896 but it was not until after 1913 that a more cohesive and financially sound organization emerged. It broke the previous institutional mould, by acting as an umbrella organization for many Chinese community and regional groups and by using bureaucratic techniques for transmitting market information. The Chamber’s official line is that it formed because prominent Chinese businessmen realized there was the need for a ‘good efficient organization’ to promote ‘trade and social intercourse’. Sinn by contrast suggests that the main motivating force behind its development after 1911 was political instability in China. Chung is more specific, noting that the Chamber was a deliberate attempt by the newly emergent Chinese regional chambers of commerce to counter the political influence of the Siyi Chamber of Commerce. Its aim, supported by the colonial state, was to attract members committed to Hong Kong’s development rather than political intrigue in China (Chung 1998). Once established on a sound political footing, the Chamber’s strength grew quickly: a membership of 400 in 1900 quadrupled by 1913, and then doubled again by 1953. It was attractive to the Chinese business community because there was now a greater need for a body which could lobby and negotiate both with the colonial administration in Hong Kong and with the increasingly autonomous Chinese provincial government in Guangzhou. But in addition membership was attracted by the Chamber’s commercial services, as it soon began to issue certificates of origin and provide market information in the form of its commercial year book and its more frequently issued trade bulletins (The Chinese Chamber of Commerce 1958; Sinn 1990). While these changes were ongoing, industrialization, which began slowly in the 1920s and 1930s and accelerated rapidly from the late 1940s, was beginning to transform the market for institutions in Hong Kong. The key change occurred in 1934, when the Chinese Manufacturers’ Union was established. This organization, later named the Chinese Manufacturers’ Association, was the first to represent industrial as opposed to service sector interests. It attracted support quickly— members numbered 70 in 1934, rising to 750 by 1953 and doubling again by 1965—because manufacturing, a very different economic activity as compared to commerce, required new forms of institutional support. Above all else Hong Kong’s industrializing economy required tighter regulation. This was because the manufacturing process needed a greater input of technical knowledge, while marketing strategies such as the development of brand names and the use of designs could be easily copied. Indeed, it is noticeable that during the 1920s and 1930s Hong Kong’s manufacturing sector was already acquiring its post-war reputation

Institutional Change in Hong Kong

155

for untrustworthiness and unreliability. The articles of association of the Union are an initial attempt to deal with this problem. They declare that the Union aimed to ‘promote and encourage unity and friendly relationships among manufacturers’ and ‘protect the reputation and good name of manufacturers’ (Chinese Manufacturers’ Union 1936 and 1947–48). Along with tighter regulation, Hong Kong’s industrializing economy also required greater state support. This was because industrialists, with high ratios of fixed capital to working capital in their investment profiles, required greater macro-economic stability and guaranteed access to overseas markets. The Union was an institutional response to these new demands and, once established, lobbied overseas governments, as well as the British colonial government in Hong Kong, to secure these goals. For example, when the Dutch government adopted protectionist policies in its colonial territories, the Union negotiated with the Dutch Colonial government in the East Indies and secured preferential access to what was, in the 1930s, Hong Kong’s largest market for manufactured goods. Existing organizations could not take on this lobbying role because they were dominated by merchants who desired a different political economy. In addition, the Union performed functions already undertaken by other chambers of commerce, guilds and the informal sector, by providing market information, issuing certificates of origin and settling contractual disputes. But once again, whilst there was some duplication here, the new requirements of its industrial membership encouraged the Union to innovate, by compiling a catalogue of industrial products and by organizing an annual exhibition of Hong Kong made products. The latter was highly successful. On the eve of the Pacific War, there had been four exhibitions, and the 1940 one, co-organized with the Society for Promoting the Use of Chinese Manufactured Products, was opened by the Governor and lasted twelve days. If the Union had not existed it is unlikely that this event would have occurred, as the Hong Kong General Chamber of Commerce had always rejected the idea of a trade exhibition in Hong Kong; it believed that, as most of its trade was with China, an exhibition of products for sale in China would be more effective if actually held there (Chinese Manufacturers’ Union 1936, 1947–48, 1950, 1960; The Chamber 1932, p. 74). The state The state was always a key player in Hong Kong’s institutional market place but the products it offered for sale changed over time. The most important institutional support provided by the state was its protection of property rights, achieved by importing the UK system of commercial law virtually wholesale into Hong Kong, and by paying for coercive agents to uphold it.12 This protection was particularly important because for large parts of the period 1842 to 1960 similar rights in most of China were insecure. Up until 1949 there was a great deal of social and political unrest in China, symbolized by events such as the Taiping rebellion of 1850–64, the Sino-Japanese War of 1937–45, and the Chinese Civil War of 1945–49; while even in periods of relative peace the reach of the central Chinese state was limited,

156

David W.Clayton

making property rights in south China less than fully secure. Thereafter, while a strong and stable Chinese state emerged, it was communist and thus abolished rather than protected private property rights. Consequently, the colony acted as an important refuge for Chinese capital (Department of Commerce and Industry 1953, 1959, 1965; Norton-Kyshe 1898; The Chamber 1926, 1927). Nevertheless, the impact of public property rights before the post-war period should not be overestimated. For one thing, criminal activity, especially piracy in the South China Sea, remained endemic, increasing the cost of insurance for Hong Kong businesses. More significantly, the failure of the colonial state to systematically codify local customs or alter British laws to meet local conditions made resort to legal processes less attractive to a Chinese business community, which, given its Confucian cultural roots, was inclined to rely on cheaper informal systems of regulation. Indeed, it was only after 1900 that Chinese firms began to register with the colonial state in any great number, while even then traditional practices did not die away, a point noted by the District Commissioner for the New Territories in 1956 when he said that ‘the Chinese do not need telling that litigation is a counsel of despair, profiting neither party’.13 In fact the state only adopted a more interventionist stance during the 1920s, when the depreciation of the Hong Kong dollar, inflation and rapid population growth caused rice riots and labour unrest which forced the government to cap the price and guarantee the supply of essential foodstuffs. It was not until the late 1940s that Hong Kong’s institutional map began to be redrawn by some new state bodies. The key organizational shift was the establishment of the Department of Commerce and Industry, a government organ with the remit ‘to give the economy an up-to-date and efficient organization in touch with all commercial activities in the colony and able to advise enquirers and the government on any aspects of economic activity in the colony and give information on commercial activities abroad’. This Department worked closely with other new state bodies, such as the overseas government offices established in London from 1946, in Tokyo from 1947 and in Sydney from 1960 (Department of Commerce and Industry 1952–65; The Chamber 1948).14 But the most important post-war change was the establishment of the Trade Advisory Committee in 1951. Renamed in 1952 the Trade and Industry Advisory Committee, and a successor to an earlier Board of Trade—established in the 1940s to co-ordinate post-war economic reconstruction—this body included representatives from the Hong Kong General Chamber of Commerce, the Chinese General Chamber of Commerce, the Chinese Manufacturers’ Union and the Department of Commerce and Industry, all of whom were appointed by the Governor (Department of Commerce and Industry 1952–60).15 It met on average ten times per year between 1951 and 1960 to discuss export promotion, the commercial image of Hong Kong, regulatory strategies and measures to counter protection overseas. It was an important body because it fed good quality information into government decision-making and legitimized public interventions in the market.16 It was able to achieve these ends because, in comparison to the pre-war period when the government mainly discussed commercial affairs informally with Western mercantile élites, there was now a

Institutional Change in Hong Kong

157

more universal and formal government/business relationship. The result was more effective government provision of information and regulation of transactions. The government provided market knowledge in a number of forms. First, the Department of Commerce and Industry published the biannual Commercial Guide to Hong Kong, which contained legal and financial information on subjects such as trademarks, freight rates and methods of payment, and the monthly Trade Bulletin, which provided information on market trends and listed inquiries from overseas. Both of these were delivered free of charge to overseas customers, often through Hong Kong government offices abroad.17 In addition the Department also periodically published a classified commercial directory and built up a commercial library. Second, the Department and the London and Tokyo offices handled individual trade inquiries by compiling databases of firms; by 1953 the Hong Kong government office held a list of 5,000 names.18 Third, the Department organized and partly funded trade fairs and missions, both at home and abroad. The aim of these initiatives was usually to diversify Hong Kong’s market base. For example, from the late 1940s to reduce reliance on South East Asian markets, the government encouraged links with Britain and the USA, and, from the mid-1950s, the government tried to develop trade with European, South American and African markets to break Hong Kong’s increasing dependence on Anglo-American demand.19 While it is difficult to assess whether these initiatives were successful, they did generate orders: for example, attendance at a trade fair in Vienna in 1959, which attracted 600,000 visitors, led to 300 commercial introductions and 45 inquiries submitted to the Trade Bulletin. If they were successful then this is partly due to the support of overseas government offices and to good business/state collaboration; the Trade and Industry Advisory Committee adopted a particularly hands-on approach, even taking time to evaluate the competing designs for stalls submitted by tendering firms. Such business/state links were even more important when it came to regulating transactions. One new initiative on this front was a list of state accredited factories. Firms that wanted to have an official stamp of approval for their products had to guarantee that their goods met specified standards and had to promise to pay compensation ‘to meet claims which are properly established’.20 This attempt to improve industrial standards was a useful regulatory stick for the state, which could threaten to expel firms to make sure they behaved, but overall its effectiveness was probably limited given the high turn-over of firms from the list. Of greater importance was the role of overseas offices in mediating between Hong Kong and external customers. Again the evidence is thin, but they do seem to have settled large numbers of commercial disputes; in 1953 the Hong Kong government office in London noted that ‘it has been possible for the office to prevent litigation following complaints made by UK importers against Hong Kong suppliers and claims made from Hong Kong against UK merchants’.21 Meanwhile, within Hong Kong itself the Trade and Industry Committee also emerged as a regulatory agent. A few examples will show how. In April 1952, the Committee agreed to a departmental recommendation that the Ching Wah Bulb Co., found guilty in court of forging trademarks, be warned that if it continued this practice then the ‘facilities

158

David W.Clayton

that the Commerce and Industry Department might be able to render would be withheld’.22 In 1952 the Committee used its good offices to settle a dispute between cotton merchants, weavers of cotton cloth and cotton spinners, concerning excess supply of raw cotton in the colony; it brokered a compromise deal, with 25 per cent of the surplus re-exported to keep prices high in the colony and 75 per cent remaining in Hong Kong to provide cheap cloth for the weavers. These examples show that the government had a competitive advantage when it came to certain forms of institutional provision. It had more resources to compile and distribute information, and could draw on the power and legitimacy of its government status to act as a more effective mediator in its dealings with states overseas and with businesses in Hong Kong. Indeed from the perspective of Hong Kong’s chambers of commerce it would have been rational once the state had intervened to withdraw from such provision, a development suggested by the North model. There is not the space in this chapter to examine this question in detail but it is evident that despite government intervention, private and government schemes continued to run parallel in the 1950s. Businesses requiring market information could use trade enquiry services and consult directories and commercial guides supplied by the state and private bodies; businesses wanting to complain about a Hong Kong firm and seek compensation could use the arbitration services of business organizations and the state, or could turn to the courts; businesses requiring import and export documentation could get the state or their own organizations to inspect their cargoes. Formal and informal institutions were thus acting in tandem. A clear example of this was the decision, taken by the Trade and Industry Advisory Committee at the very end of our period, to establish a new organization called The Hong Kong Exporters’ Association to deal with the indifferent reputation of some Hong Kong exporters. The economics and politics of institutional change in Hong Kong North would argue that as Hong Kong experienced high rates of economic growth for most of its history, Hong Kong’s institutions were efficient. (An alternative institutional perspective, say of Olson, would be that Hong Kong’s institutions were not so inefficient as to seriously constrain growth.) If we accept North’s position then our task is now to assess why this was the case. To do so we need to examine structural change within Hong Kong, the process of Hong Kong’s integration into world markets and the politics of institutional provision in a colonial setting. Before 1945 Hong Kong’s institutions were conducive to economic growth because it was an entrepôt centre which did not require extensive and powerful public institutions. Hong Kong was part of three, interrelated systems of trade, intra-China trade, intra-Asia trade and trade between Asia and the Western world (Latham 1994 and 1998), all of which were primarily supported by informal institutions and private property rights. Both intra-China trade and intra-Asia trade were dominated by Chinese merchants. Tsai notes that of the 1700 Chinese firms in the 1915 Anglo-Chinese Commercial Directory of Hong Kong, there were 84

Institutional Change in Hong Kong

159

Nam Pak Hong traders (General Exporters and Importers to Southern and Northern Ports and to Southeast Asian Ports), and 239 Kam Shan Chung traders (General Exporters and Importers to Melbourne, Sydney, San Francisco and Honolulu) (Tsai 1993, pp. 32–33). Such merchants would use family, guilds and sub-regional groups within Hong Kong to acquire and verify market information, and to regulate transactions. They would have only infrequently used formal institutions such as colonial commercial law. But they probably turned to chambers of commerce and trade associations from the early years of the twentieth century to settle some disputes and to verify information acquired from informal sources.23 This reliance on private institutions was a cost effective way of reducing ‘transaction’ and ‘information’ costs, as such mechanisms did not require extensive administrative and legal infrastructures, instead building on existing social networks. In theory reliance on them should have constrained the potential size of the market and restricted economic growth, but, in Hong Kong’s case, strong family, sub-regional and ethnic links between Hong Kong’s Chinese élites and the large number of sizeable overseas Chinese communities in Asia strengthened the process of market integration. Hong Kong’s trade with the Western world was dominated by expatriate Western merchants, who specialized in exporting manufactured goods to China and Chinese primary products to the West. These merchants were also engaged in intra-Asia and intra-China trade and, like their Chinese counterparts, also primarily used private institutions. But they also carried out more inter-ethnic transactions and formed more complex, non-kin based firms which employed Asians with knowledge of local markets and cultural systems. Perhaps in consequence they relied more on chambers of commerce and the state for institutional support. Hong Kong’s institutional world began to change after 1945 because the pattern of Hong Kong’s trade altered. The most significant shift was industrialization. To begin with this profound structural change could be handled by modifying the existing institutional system, as the most important markets for Hong Kong manufacturers were initially in the Chinese communities of South East Asia and also to a lesser extent in China. However after 1945 their key markets were the United Kingdom, the USA and Europe, a reorientation greatly accelerated by the closure of the China market in the early 1950s. The result was that the majority of Hong Kong’s trade now involved inter-ethnic transactions. There were ethnic linkages in Western markets that Chinese merchants and manufacturers could have exploited, but the scale of the trade with Europe and America in the 1950s (and the rapidity of its increase) suggests that inter-ethnic networks were more important.24 These took two main forms. Firstly Hong Kong Chinese manufacturers turned to Western merchant houses to supply inputs into production and to market their products. Secondly Western buyers, especially retailers, marketed Hong Kong made products abroad (Espy 1974). The existing institutional framework could not cope with this massive increase in inter-ethnic trade because business groups were too divided along ethnic and economic lines. Moreover trade in ‘modern’, factoryproduced lines required more stringent forms of quality control and the acquisition of design and technical knowledge, which established bodies had little experience

160

David W.Clayton

of providing. What was required were new, independent sources of market information and third party regulation. While structural changes and the market orientation of Hong Kong are the key determining factors, they do not fully explain institutional changes in Hong Kong, and nor do they fully account for why certain institutions were effective at certain times and not at others. To complete our analysis we need to bring politics into the story. In the nineteenth century the effectiveness of socially embedded private institutions, combined with the pattern of Hong Kong’s trade, reduced demands from business communities for greater state involvement in their affairs. Neither Western nor Chinese élites wanted a colonial state which intervened extensively in commercial affairs—or that taxed and spent a high proportion of the colony’s income. Western mercantile élites were less content with the existing set-up than their Chinese counterparts, desiring both more influence over colonial state decision-making and more state support for its sectional demands (Endacott 1973, pp. 4–25). But while Western merchants influenced law-making more than the Chinese community, they were relatively weak vis-à-vis the state: there were merchant representatives on the Legislative Council and the executive consulted merchants informally, but most power remained with the Governor. Furthermore, despite attempts to incorporate members of the Chinese community into the political system, there were very few Chinese merchants within the administration, a political deficit not compensated for by good channels of communication between the colonial authorities and the Chinese community. Consequently the government ruled Hong Kong by universalistic and liberal legal norms but at the same time was determined not to be seen as the representatives of expatriate capital. An example from a case in 1934 will illustrate how these two forces combined to block some legal changes. In that year, the Hong Kong General Chamber of Commerce made representations to the Colonial Secretariat for changes to the law on property transactions. The Chamber wanted the state to force buyers and sellers of property to advertise a transaction seven days before the transfer; the aim being to stop Chinese merchants, being sued for nonfulfilment of contract, from transferring the title deeds of land and buildings to members of their extended family before the courts seized their assets. The colonial state however refused to act, arguing that this was a departure from standard legal principle which would curtail trade in the property market.25 Before 1945, relations between the colonial authorities and two states curtailed its power, and thus its ability to dictate Hong Kong’s institutional evolution. The metropolitan state in London was the major constraint. It had to codify law-making in Hong Kong and, by opposing a higher tax burden in the colony and by preventing generous resource transfers from the UK, limited the size and scope of the colonial state. But in addition the Chinese state also undermined the effectiveness of institutional provision in Hong Kong because its own political grip on the area surrounding Hong Kong was less than secure. This meant, for example, that traders could escape prosecution by fleeing to Guangzhou and that piracy remained a

Institutional Change in Hong Kong

161

serious problem in the South China Sea. On the piracy issue the Governor noted in 1927 that Over and over again we have offered the Chinese authorities full naval and military co-operation for this purpose. They have, however, rejected these offers of help and they have done nothing whatever themselves, but have been scandalously forgetful of the first duty of any civilized and self-respecting government, namely the suppression of piracy and brigandage and the maintenance of law and order. (Pennell 1961, p. 41) The state’s strong and positive intervention in the post-war period was partly a response to the needs of post-war reconstruction.26 But unlike in the 1920s the colonial authorities also believed it had to meet a new demand for the long term provision of market information and regulatory frameworks. In 1950 the Governor informed the Secretary of State of Colonies that ‘What is really needed in the case of this colony is simply some medium for introducing prospective customers and buyers. The rest and particularly the financial side of any transactions is better left to the parties themselves to arrange as they wish.’27 As Hong Kong had traditionally adopted a laissez faire stance on such matters, it is surprising that the state chose to intervene itself and not leave it to the business community to fulfil this function— especially as the Colonial Secretariat had already acknowledged in 1949 that, even though it had taken on such a role since the war, it was receiving a disappointing number of trade inquiries.28 What may have persuaded the state to act were international controls on trade in the 1940s. The product of protectionism in the 1930s and the war economies of the 1940s, these potential obstacles to trade were now very extensive. As they were the creatures of state actions it was easier for states to provide information on them and also to lobby for their removal. More importantly however, the Chinese business community now wanted greater state intervention. Manufacturing interests, particularly immigrant entrepreneurs from Shanghai, wanted preferential access to government capital and supplies of raw material (under government control in the 1940s), and traditional Chinese commercial interests now had to deal with greater state regulation of world trade. As neither of these groups had strong political links with a China that had turned communist, they had more energy and inclination to lobby the colonial state.29 The state felt it could meet these new demands because there were now fewer constraints on its power. The metropolitan state for one was now more supportive of colonial development; in Hong Kong’s case it actively encouraged some initiatives believing they would increase Britain’s trade with China. Moreover, the budgetary implications of state intervention were not great: the colonial administration’s overall budgetary position had been eased by the introduction of income taxes in the 1940s and in any case much of the Department of Commerce and Industry’s revenue was derived from the fees.30 Interestingly, and in stark contrast to the pre-1945 period, it was the Hong Kong General Chamber of Commerce which now acted as a potential block on institutional

162

David W.Clayton

change. In 1948 a Chamber subcommittee, established to consider government plans to set up the Department of Commerce and Industry, concluded that as the Chamber represented all the leading business houses in the colony and fulfilled many of the functions to be undertaken by the state, the government scheme was too ambitious and ‘in large measure unnecessary’, as the Chamber could expand its role. A Chamber note to the government suggested that the pre-war practice of referring to the Chamber all matters affecting the commerce and industry of the colony might be resumed at an early date and that the adoption of this practice would prove of greater benefit to the Colony than would be achieved by setting up a complicated Government Department. (The Chamber 1948) The Chamber may also have been concerned that fees derived from documentation, an expanding source of finance to the Chamber, would be threatened by state provision. The twin forces of industrialization and the collapse of the entrepôt trade produced a change in the Chamber’s position. By 1951 European merchant houses were having severe difficulties continuing their entrepôt business because there was an embargo on trade with China, while their capital in China was being slowly expropriated by discriminatory practices introduced by the Chinese People’s Government (Clayton 1997). Consequently most switched resources to the marketing of Hong Kong-made products, away from their traditional core entrepôt business. The Hong Kong General Chamber of Commerce however could not regulate this trade because, while it had Chinese merchants and Chinese manufacturers as members, their numbers were limited.31 The vast majority of Chinese merchants and manufacturers were unlikely to turn to an organization perceived historically as representing European mercantile capital (The Chamber 1954).32 Consequently the Chamber quickly accepted the role now played by the Hong Kong state. By the early 1950s the Chamber noted that complaints against its members ‘could be easily settled after a discussion with the Secretary’ but the majority of complaints were against non-members. By the early 1950s it acknowledged that while it was ‘ready to help’ deal with manufacturers who made substandard products and conducted irresponsible transactions, it needed the Department of Commerce and Industry to ‘bring manufacturers into line’ (The Chamber 1952). Indeed, thereafter, in response to a growing number of complaints made against Hong Kong suppliers, by using its position on the Trade and Industry Advisory Council, it actively campaigned for more government intervention to regulate the behaviour of manufacturers in the colony. Conclusions In Hong Kong a complex pattern of regional and dialect groupings, communitywide organizations and sectional interests—representing traditional trades and

Institutional Change in Hong Kong

163

manufacturing industries—emerged to provide information and regulate trade. Most evolved slowly for a variety of reasons, with political and social factors important catalysts. By comparison, state institutional support mechanisms for trade and industry were imported from the West in the nineteenth century, and then developed slowly until the post-war period, when they quickly increased in scale, scope and effectiveness. In some respects this pattern fits the North model as there was a gradual shift towards more formal institutions over time, with the market orientation of the Hong Kong economy, the structure of state/business relations in Hong Kong and state/state relations in the area determining this institutional outcome. In other respects however Hong Kong differs. Its institutions did not develop in a neat linear pattern. There was, for example, a sharp break with the institutional past during the 1930s and 1940s. Until then Hong Kong’s private institutions were split along ethnic lines. From the North perspective this structure was sub-optimal, resulting in services being duplicated and making state intervention less effective because it divided the business community. But in Hong Kong the system worked well up until the post-1945 period because Hong Kong’s cosmopolitan society had good informal links with other markets abroad. Thereafter the system had to change because most of Hong Kong’s trade now involved inter-ethnic transactions. But while such a change fits into the North framework, it was political instability in Asia in the 1940s—resulting in the collapse of the entrepôt trade with China— which drove the process forward, a factor more in keeping with the institutional approach of David (1994) rather than North.33 Finally, it remains unproved from the evidence presented whether, as North would argue, institutional change predated market expansion and whether private institutions became redundant once the state stepped in. These are clearly subjects for further research but a tentative conclusion would be that the tale is perhaps more complex than that told by North. Notes 1 2

3

4

This article was aided greatly by discussions with Simon Smith, Bill Sheils, Norris Nash and Peter Lowe. North does not deny the role of other economic variables as contributing to economic growth—such as good factor endowment (physical, labour, capital) and the level of technological advance—but argues that institutions are the most important factor because their existence determines trade and not vice versa. Consequently instead of each individual merchant having to meet each potential buyer or seller personally and negotiate a price, and then use his own resources to enforce and oversee a deal, a guild/merchant association would provide market information, handle disputes between members, and, by threatening an ill-reputable merchant with expulsion, reduce the need for expensive contractual guarantees. In return for these services merchants contributed a subscription and attended meetings and social events. The merchant association thus reduced the transaction costs of doing business. For example, the guild could not protect the property rights of merchants visiting the state nor guild members who travelled abroad. Equally the state, acting as ‘a third force’, could ensure that all merchants would supply good information to the state and, with greater access to resources, the state could establish a regulatory system with more severe penalties.

164

David W.Clayton

5 Some of the more interesting include: Hong Kong and Kowloon Footwear Manufacturers’ Association (92 members), the Federation of Hong Kong Cotton Weavers (30 members), the Hong Kong Chinese Importers’ and Exporters’ Association (472 members), the Hong Kong Exporters’ Association (57 members), the Hong Kong Chinese Textile Mills Association (252 members), the Hong Kong Enamelware Manufacturers’ Association (15 members), the Hong Kong and Kowloon Plastic Products Merchants’ United Association, the Hong Kong and Kowloon Clock and Watch Trade Merchants’ Association (286 members), the Hong Kong and Kowloon Hide and Leather Traders’ Association (76 members), the Hong Kong Rattan Merchants’ Association (17 firms). For a full list see Hong Kong Trade Associations’ Circular No. 5, in HKPRO STAT/474. 6 In addition to China-centred organizations there were also Kaifong associations. These neighbourhood organizing committees tended to arbitrate on disputes between residents and local shopkeepers as well as funding and administering public works and charity. 7 Others included the Chinese Chamber of Commerce, Kowloon (500 members), the Federation of Hong Kong Industries (431 members), the Hong Kong Junior Chamber of Commerce (208 members), the New Territories General Chamber of Commerce (67 members), the Kowloon Chamber of Commerce (2582 members), and the Yuen Chamber of Commerce (375 members). Hong Kong Trade Association, Circular No. 5, in HKPRO STAT/474. 8 The organization was incorporated in 1928 and its rules were adjusted in 1954 because membership had increased by one-third and because it needed to amend racial bars in its constitution. See Certificate of Incorporation of the Hong Kong General Chamber of Commerce, C.D.Melbourne, Registrar of Companies, Hong Kong 1928. 9 The committee had powers to administer an oath to parties, to protect goods subject to arbitration, to make interim awards and to demand disclosure of evidence. The committee had a special account to retain money attached to cases, and could call evidence from witnesses from outside Hong Kong. In the 1950s the fee charged per case was HK$70 for cases where the amount involved did not exceed HK$500 and the time taken up did not exceed half an hour; in other cases the administration fees were HK$50 and the arbitrator’s fees HK$40 an hour; where the amount exceeded HK$ 10,000 fees were doubled. In the 1930s arbiters fees were HK$20 per hour (The Chamber 1929, 1954). 10 1956 is chosen because it was well after the Korean War induced boom and bust period of 1950–54 and before the British relaxation of the economic embargo on Communist China in 1957. 11 Note: the Hong Kong General Chamber of Commerce Rules Covering the Sale of Rice for Export from Hong Kong and the Hong Kong General Chamber of Commerce Standard Form of Fancy Piece Goods Contract 1924; The Hong Kong General Chamber of Commerce Terms of Contract with the Manchester and Bradford Chambers Relating to Cotton and Woollen Piece Goods 1922. 12 Even in the 1950s the Bankruptcy and Limited Liability Laws were based on British precedents; the Trade Marks Ordinance of 1954 on the British Trade Marks Act of 1938; the British Copyright Act of 1911 was adopted and an individual could apply for cover under the UK Registration of Designs Act (1949) and the Registration of UK Patents Ordinance. 13 Memorandum from the District Commissioner for the New Territories to the Colonial Secretariat Registrar General, 31 July 1956, in HKPRO A. SEC3/3511/56. 14 See monthly reports of the Department of Trade and Industry in HKPRO SEC7417. The Department also utilized the services of the commercial branch of the Board of Trade in London, which had officials posted at various embassies in the world. British commercial attaches in America passed on information on price, delivery dates, marketing organizations

Institutional Change in Hong Kong

15

16 17

18

19 20

21 22 23

24

25

165

in the colony and photographs of products. For examples see HKPRO SEC27/5401; United Kingdom Trade Promotion Organizations in the United States and Canada 1949. Minutes of the Trade Advisory Committee, 1 December 1952 to 7 December 1957, in HKPRO GR4/581/52. There was a lengthy discussion about the change to the Trade and Industry Advisory Committee in 1953; the minutes indicating that the change was the result of the increasing percentage of the colony’s trade made up of manufacturers, the belief by manufacturers that their interests were not being represented on the existing committee and a growing convergence of economic interest between merchants and manufacturers in the colony. See GR4/581/53; minutes of the Trade Advisory Committee, 20 August 1953. The Director of the Hong Kong government office in London also attended meetings when in the colony. These are taken from the minutes of the Trade Advisory Committee, various meetings 1952–57, in HKPRO GR4/581/53. In 1955, 8,500 copies of the Trade Bulletin were sent overseas and local sales were 900 per month; in 1959, 1,600 new Hong Kong businesses were listed, 1,500 overseas trade inquiries published, 10,000 copies/month circulated and 100 new requests per month were received requesting inclusion on the mailing list. A 1960 survey noted that 89 per cent of firms that had taken out adverts had had ‘success’, with 52.4 per cent having had orders placed with their firms (Department of Commerce and Industry 1960). In 1954, a key year for Hong Kong’s trade growth, the Hong Kong office received 2,000 trade inquiries and 200 status reports from Hong Kong on individual companies. By 1959, 300–400 calls were answered, 1,200 visitors received and a permanent exhibition of Hong Kong products set up. The Hong Kong government representative in Tokyo received 300 visitors per month in 1959. For a list of factories registered see HKPRO TD3141/3665/52. These incomplete records show that between 20 and 60 factories were listed each month, with a large turnover of factories registered (The Department of Commerce and Industry various years). See HKPRO FIN9/2141; Director of Commerce and Industry Department to Colonial Secretary, 7 October 1955. Other criteria included: ‘The factory is operated in accordance with labour and factory legislation in Hong Kong’; The factory is open to inspection at any time by persons with the authority of the Standards Institute or whatever body is formed to advise Government on such matters’; ‘Where Imperial Preference is sought, due attention will be paid to Imperial Preference costings’; ‘The factory has the recommendation and support of its own appropriate industrial organisation’ (The Chamber 1950). In 1954 the London Office received 20 claims and 30 in 1956 against Hong Kong suppliers, most of which it settled; in 1959 it successfully dealt with 25 trade disputes. Minutes of the Trade Advisory Committee, 24 April 1952, in HKPRO GR4/581/53. Without much evidence to draw on it remains difficult to ascertain whether transactions within the Chinese business community were covered by legal contracts. Tsai suggest that many were done by word of mouth and thus regulated mainly by informal channels— the family and ethnic linkages. Evidence from studies on early China suggests that most transactions were covered by written contracts, perhaps because the written Chinese script was central to Chinese culture and the spoken tongue so diverse that communication was difficult between merchants from different areas (Hansen 1996; Zelin 1991). Hong Kong exported 3.7 per cent (HK$163 million) of total exports to the US in 1951, 23 per cent (HK$679 million) in 1961 and 41.5 per cent (HK$5708) in 1971; the corresponding figures for the UK are 4.8 per cent (HK$215 million), 20 per cent (HK$589 million) in 1961 and 14.1 per cent (HK$1946 million) in 1971 (note the 1951 figures include reexports) (Census and Statistics Department: 1969; Hong Kong Government: 1971). For this exchange see HKPRO CSO1505; (58–1–178–16): Acting Secretary, Hong Kong

166

26

27

28

29

30

31

32

33

David W.Clayton

General Chamber of Commerce, to Colonial Secretary, 15 November 1934; D.W.Tratmen, Clerk of Council, to Secretary of the General Chamber of Commerce, 22 December 1934; minute by R.S.Linsell, Attorney General, to Colonial Secretary, 21 November 1934; and minute by T.S.Whytesmith, 21 November 1934. For the proposed formation of a trade and industry section of the Supplies, Transport and Industry Branch for the Rehabilitation of Hong Kong see FR8000/45 in HKPRO. For the government’s attitude to trade inquiries see incoming letter in HKPRO SEC8011; and minute by John Cowperthwaite 20 March 1946. Secretary of State of Colonies from the Governor of Hong Kong, 23 October 1950, in Department of Commerce and Industry minutes, 19 October 1950 in HKPRO SEC 27/ 5401. For communications on this point see HKPRO SEC10/5401; Statistical Bureau to the Colonial Secretary, 28 July 1949; minute by Oliphant (Supplies, Import and Export) to Honourable Secretariat, 5 August 1947; and minute by Cowperthwaite, 6 August 1947. For government-industry exchanges see HKPRO GR21/5361/48, general correspondence regarding the textile and cotton yarn industry in Hong Kong, 14 September 1948 to 14 January 1950 and HKPRO SEC8002/45, Industry, Application for Advance for Rehabilitation of Factories, Question of Providing Raw Materials. The Department’s trade promotion expenses varied from year to year. In 1960, when representatives were sent to a number of fairs and a directory was published, trade promotion’s expenditure was HK$940,971: HK$307,652 covered the expenses of overseas offices, with the London office taking up over two-thirds of this total; HK$633,319 comprised trade promotion expenses, with one-third covering attendance at three trade fairs, a trade mission to West Africa and miscellaneous travel by officials, and the other two-thirds covering the cost of publishing and distributing publications. There is no breakdown of revenue available but internal reports and communications indicate that advertising in the trade directories and bulletins contributed substantial amounts of revenue. For example all the expenses of British Industries Fairs were met by advertising in directories, produced and circulated before, during and after the fair. In 1960 a total of HK$320,216 was gained from advertising, one-third of total expenditure on promotions. See: Director of Commerce and Industry to Deputy Economic Secretary, 17 July 1961 and Memorandum for the Trade and Industry Advisory Board, ‘Hong Kong’s Public Relations Overseas with particular reference to trade and industry’, no author or date specified, both in HKPRO SEC1/4761/59; and Department of Commerce and Industry 1961. The 1957 Chamber directory listed 721 firms, of which only 50 listed their companies as manufacturers (or only exported a few closely related Hong Kong made products) (The Chamber 1957). The original rules of membership of the Hong Kong General Chamber of Commerce for example stipulated that all members had to be British; this regulation was changed later but until 1954 a subclause still stipulated that members of the General Committee had to be British. It was not until 1954 that the articles of membership were changed to read: The Chamber shall be International in its character and, subject to the provisions of the Memorandum and Articles of association, membership shall be open to persons of all races and nationalities’ (The Chamber 1954). Paul David argues North implies that institutions are brought about in a systematic, calculated way by humans to make the market work more efficiently. David disagrees, regarding institutions as accidents of history, with old institutions not changing incrementally but collapsing, dissolving, or being ‘dismembered and ingested by other competing organizations’. In his later work North accepts that there may not be a smooth progression to efficient institutions and that in many cases certain institutions can produce economic stagnation (David 1993).

Institutional Change in Hong Kong

167

References Casson, M. (1997) ‘Institutional economics and business history: A way forward’, Business History, 39, 4, pp. 151–76. Census and Statistics Department (1969) Hong Kong Statistics, 1947–1967, Hong Kong: The Government Printer. Chan, M.K. ed. (1994) Precarious Balance: Hong Kong Between China and Britain, 1842–1992, New York: M.E.Sharpe. The Chinese Chamber of Commerce (1958) Members of the Twentieth Committee, 1956– 1958, Hong Kong: The Chamber. Chinese Manufacturers’ Union (1936) Xianggang Zhonghau Changshang Chupin Zhina, Hong Kong: The Union. The Chinese Manufacturers’ Union (1947–48) Directory of Members, 1947–48, Hong Kong: The Union. —(1948) Industrial Bulletin, Hong Kong: The Union. The Chinese Manufacturers’ Union of Hong Kong (1950) New Articles of Association, Hong Kong: The Union. —(1960) New Articles of Association, Hong Kong: The Union. Clayton, D. (1997) Imperialism Revisited: Political and Economic Relations between Britain and China, 1950–54, London: Macmillan. Chung, S. (1998) Chinese Business Groups in Hong Kong and Political Change in South China, 1900–25, Basingstoke: Macmillan. David, P. (1994) ‘Why institutions are the carriers of history’, Structural Change and Economic Dynamics, 5, 2, pp. 207–8. Davis, L.E. and North, D. (1971) Institutions of Change and American Economic Growth, New York: Cambridge University Press. Department of Commerce and Industry (1952–65) Annual Reports, Hong Kong: The Department. Endacott, G.B. (1973) A History of Hong Kong, Hong Kong: Oxford University Press. Espy, J.L. (1974) ‘The strategy of Chinese industrial enterprise in Hong Kong’, unpublished DBA thesis, Harvard University. Hansen, V. (1996) Negotiating Daily Life in Traditional China: How Ordinary People Used Contracts, 600–1400, Yale: Yale University Press. Hong Kong General Chamber of Commerce (1926) Annual Reports, Hong Kong: The Chamber. —(1927) Annual Report, Hong Kong: The Chamber. —(1929) Memorandum and Articles of Association, Bye-laws, Hong Kong: The Chamber. —(1932) Annual Report, Hong Kong: The Chamber. —(1935) Annual Report, Hong Kong: The Chamber. —(1948) Annual Report, Hong Kong: The Chamber. —(1950) Annual Report, Hong Kong: The Chamber. —(1952) Annual Report, Hong Kong: The Chamber. —(1954) Memorandum and New Articles of Association, 1954, Hong Kong: The Chamber. —(1956) Annual Report, Hong Kong: The Chamber. —(1957) Classified Directory of Members, Hong Kong: The Chamber. Hong Kong Government, (1971) Annual Report, Hong Kong: The Government Printer. Latham, A.J.H. (1994) ‘The dynamics of intra-Asian trade, 1868–1913: the great entrepôts of Singapore and Hong Kong’, in A.J.H.Latham and Heita Kawakatsu, eds, Japanese Industrialisation and the Asian Economy, pp. 145–93, London: Routledge. —(1998) ‘The reconstruction of Hong Kong nineteenth-century Pacific trade statistics: the emergence of Asian dynamism’, in Sally M.Miller, A.J.H.Latham and Dennis O.

168

David W.Clayton

Flynn, eds, Studies in the Economic History of the Pacific Rim, 155–71, London: Routledge. North, D. and Thomas, R.P. (1973) The Rise of the Western World—An Economic History, New York: Cambridge University Press. North, D.C. (1990) Institutional Structure and Institutional Change, New York: Cambridge University Press. —(1991) ‘Institutions’, Journal of Economic Perspectives, 5, pp. 97–112. Norton-Kyshe, J.W. (1898) The History of the Laws and Courts of Hong Kong, London. Pennell, J.W. (1961) A History of the Hong Kong General Chamber of Commerce, Hong Kong: The Chamber. Olson, Mancur (1982) The Rise and Decline of Nations, New Haven, Conn.: Yale University Press. Sinn, E. (1989) Power and Charity: The Early History of the Tung Wah Hospital, Hong Kong, Hong Kong: Oxford University Press. —(ed.) (1990) ‘A history of regional associations in pre-war Hong Kong’, in Sinn, E. Between East and West: Aspects of Social and Political Development in Hong Kong, Hong Kong: Hong Kong University Press. Tsai, J.-F. (1993) Hong Kong in Chinese History: Community and Social Unrest in the British Colony, 1842–1913, New York: Columbia University Press. White, B. (1994) Turbans and Traders: Hong Kong’s Indian Communities, Hong Kong: Oxford University Press. Williamson, O. (1975) Markets and Hierarchies, New York: Free Press. —(1985) The Economic Institutions of Capitalism, New York: Free Press. Zelin, M. (1991) ‘The Structure of the Chinese Economy during the Qing Period’, in Kenneth Lieberthal et al. (eds) Perspectives on Modern China, Four Anniversaries, London.

8

Education and development The experience of the Four Little Tigers* Derek H.Aldcroft

The East Asian miracle Today the vast majority of the world’s population has a smaller share of global income that at any previous stage in history. The high income countries (those with a per capita income of more than $6,000 in 1988) account for over 80 per cent of the world’s income, yet they contained only 16.6 per cent of its population. At the other end of the spectrum nearly 61 per cent of the world’s population (with incomes averaging $545 or less) has to make do with a mere 5.6 per cent of global income (World Bank 1990; Aldcroft 1991). While the gap between rich and poor nations has continued to widen during the twentieth century, the most striking feature of world development since World War II has been the growing disparity in economic performance within the Third World itself (Reynolds 1986, p. 45). The data in Table 8.1 illustrate this quite vividly. Africa, Latin America and Eastern Europe have performed very poorly, especially in per capita terms in the past two decades. During the same period the most dynamic region has been the Asian, and more especially East Asia which includes not only Japan, but also what have become popularly known as the four little tigers or dragons of Hong Kong, Singapore, South Korea and Taiwan. These four countries have had the highest growth rates in the world during the past two or three decades, with living standards more than quadrupling within the span of a generation, 1960–85 (Young 1994, pp. 965–6). This has brought income levels, especially in Hong Kong and Singapore, much closer to those of Western nations. The recent history of these countries is remarkable in many respects. They have had the fastest industrial revolution on record, with the region’s performance surpassing that of any other region in the world in the post-war years. Their high growth rates were maintained moreover through the 1970s and beyond, at a time when growth was slowing down or faltering in most other countries (Cowley 1991, 5; Rock 1993, p. 1788). From very small beginnings they have become the world’s most dynamic exporters of manufactured goods. Moreover, contrary to popular *

I am extremely grateful to Professor David Ashton of the Centre for Labour Market Studies in the University of Leicester for making available the text of the forthcoming study he and his colleagues have recently completed on education and training in the East Asian countries, details of which are listed in the References

169

170

Derek H.Aldcroft

Table 8.1 Growth performance by region 1950–92/4 (annual average compound growth rates)

Sources: Maddison 1995, pp. 80–83; World Bank 1992, 1995. Notes: (a) 1960–80; (b) 1980–93.

belief, their spectacular growth has been brought about largely through increased factor inputs (high investment, rising participation rates and inter-sectoral transfers of labour) rather than by improvements in total factor productivity, which have not been exceptional by international standards (Young 1994, p. 973 and 1995, pp. 673, 675). All this has been achieved with relatively modest rates of inflation apart from Korea, and with a remarkably equitable distribution of income. Their educational record has also been exemplary, the more so in that they have maintained their faith in the human capital revolution when Western nations and many developing countries were becoming increasingly disenchanted with it (Aldcroft 1998). The question is: did these countries find the key to their prosperity through human resource development or are we looking at a mirage? The achievements are all the more remarkable when one recalls that in the early post-war years the countries of East Asia were regarded as poor bets as far as modern development was concerned. They were poverty-stricken, they had limited resources and their future prospects looked distinctly bleak (Balassa 1988, S275; Mehmet 1988, pp. 28, 39). During the first half of the twentieth century they had achieved little or no real growth, at least in per capita terms, and even by the early 1960s economic conditions and levels of income, especially in Korea and Taiwan, were not very different from those in many African countries (Maddison 1989, p. 15; World Bank 1994, pp. 39, 219). Korea’s per capita GDP in 1960, for example, was the same as that of the Sudan, while Taiwan’s was very similar to that of Zaire (Morris 1995, p. 2). The World Bank’s description of the Republic of Korea’s economy around this time bears a striking resemblance to that of many poor Third World countries in the latter half of the twentieth century:

Education and development

171

The trade regime was characterised by import substitution, including the familiar set of complex multiple exchange rates, import licensing, and overvaluation. The economy was heavily reliant on foreign aid, with exports accounting for 3 percent of output, while imports amounted to over 10 percent. While inflation had been brought under control, output growth rates were stagnant. Furthermore, manufacturing accounted for just 11 percent of GNP, with over 45 percent of output concentrated in the primary sector. (World Bank 1994, p. 219) Educational strategies The East Asian countries did have one big advantage, however, which most Third World countries lacked. They had a high level of access to elementary education and a high literacy rate relative to their levels of GDP prior to their rapid expansion from the mid-1960s onwards (Barro 1991). Even at the beginning of the post-war era (circa 1950) more than one half the number of children of primary school age in East Asia were attending school; by the early 1960s primary school enrolment ratios almost matched those of Western nations. By then the majority of primary school children could read and write and adult literacy rates were around 71 per cent in Korea and Hong Kong and 50–54 per cent in Singapore and Taiwan. A decade later they were close to 80 per cent in all four countries (Oshima 1988, S109–10; Choudhury and Islam 1993, pp. 151–2). The widespread provision of basic education occurred mainly in the early postwar years. Though the colonial heritage had left a certain educational legacy, this fell far short of universal provision. In any case, the war had taken its toll on education as on everything else. Thus the educational system in Hong Kong was in a dreadful state at the time of liberation from the Japanese (August 1945), with only a few thousand children attending school (Crawford 1995, p. 105). In the case of South Korea the bulk of the population was still illiterate at that time and primary school enrolment rates were less than 50 per cent. Here, as elsewhere, the major advance came in the great educational drive following liberation from the Japanese which saw a strong rise in primary school enrolments and a dramatic fall in levels of illiteracy. Between 1945 and 1955 primary enrolment rates rose from 48 to 87 per cent, while the illiteracy rate plummeted from 78 to 12 per cent (over 13 years of age). Thus within a matter of little more than a decade South Korea had been transformed ‘into a developed country in terms of diffusion of primary education’ (Kimura 1990, p. 343). Much the same happened in Taiwan. Although, according to Woo (1991, p. 1030), the Japanese left Taiwan with one of the most literate populations in Asia, in actual fact the level of educational attainment was not very much better than in Korea. As in the latter case, it was the first post-war decade that saw the widespread provision of basic education, including rural education. Although public spending on education rose rapidly in the postwar years—faster in fact than the rise in GNP—by international standards it was not especially munificent. Measured as a proportion of GNP, educational budgets were no higher than those of many developing countries and by western standards (OECD countries)

172

Derek H.Aldcroft

Table 8.2 Total public spending as percentage of GNP

Source: UNESCO, Statistical Yearbooks, various years.

they were relatively low. The data in Table 8.2 for South Korea, Hong Kong and Singapore suggest that total public spending (current and capital) on education as a share of GNP rarely rose above 4 per cent, and for much of the time it was very similar to that in developing countries. Of course, in time, with their rapidly growing incomes, the volume of spending in absolute terms and in per capita terms was substantially higher than that in many poorer countries. Also one should bear in mind that the official figures take no account of the private funding of education which was very important in East Asia. Parental contributions in the form of fees and private tuition in South Korea for example, exceeded state spending in 1995 (Ashton, Green, James and Sung 1999, ch. 4.) Moreover, the delivery system was far more efficient than in the case of most Third World countries. In fact it is this latter aspect which is probably the most striking feature of the East Asian educational system. While there have been considerable variations in the structure and format of educational systems in terms of sources of funding, the role of the state, the split between general and vocational education, and the nature of the curriculum, some common features and practices stand out. The pattern of educational improvement in all four countries was essentially a sequential one, or bottom-up variety, which initially placed high priority on basic education, and then moved upwards and outwards at a later stage. Thus the main objective was first to secure a universal system of high quality primary education for all youngsters irrespective of gender. This was designed to achieve basic levels of literacy and numeracy for the majority of pupils at an early stage, which also helped to provide an element of social cohesion during early industrialization and national identity building. The latter sociopolitical philosophy featured prominently in the educational policies of most East Asian, and also Southeast Asian countries, in the post-war years. Thus in Singapore the government in 1946 declared that the aim of education should be that of ‘fostering and extending the capacity for self-government and promoting the ideal of civic loyalty and responsibility’. Similar sentiments were actually enshrined in Article 27 of the constitution of the Republic of Korea. It had also long been a basic tenet in Japan (Beauchamp 1987, p. 302). Only in Hong Kong, where schooling was seen as a

Education and development

173

Table 8.3 Primary, secondary and tertiary education ratios 1960–85 (enrolment ratios as percentage of relevant age group)

Source: Choudhury and Islam 1993, pp. 151–2.

Table 8.4 Educational attainment of the working population (%)

Source: Young 1995, p. 642.

means of advancing individual self-enhancement, was such a philosophy notably lacking (Crawford 1995, pp. 109–10). The move up the educational ladder came later, as the skill requirements of the economy became more demanding and therefore required the more intensive development of secondary and tertiary education, as well as vocational training. Thus by the mid-1980s all primary school graduates in Korea entered middle schools and the vast majority of these pupils later went on to high schools. Thereafter, opportunities to enter higher education were expanded rapidly, with a target enrolment rate for the tertiary sector set at the very high level of 50 per cent for the 1990s (Gannicott 1990, p. 44). The sequential pattern of this development can be seen from the data in Table 8.3. Universal primary education was more or less complete by the early 1960s, but secondary education was still fairly limited, while higher education was only available for the select minority of students. A generation later (circa 1985) secondary educational provision was nearing completion, and there had already been a big increase in tertiary enrolments, especially in Korea and Taiwan. The impact in terms of the educational attainment of the workforce has been little short of dramatic. As the data in Table 8.4 demonstrate, within the space of a generation (1966–90) the proportion of the working population having secondary education or better nearly tripled, and in the case of Singapore quadrupled, while about a fifth of the workforce had some exposure to tertiary education (junior college or ‘A’ level and above). Thus by the early 1990s the educational standards of the working population could easily stand comparison with those in Western countries (Bercuson 1995, p. 7; Young 1995, pp. 642, 645).

174

Derek H.Aldcroft

The funding of education as between public and private sources varied considerably between countries and levels of education. Basic primary education was funded largely by the state in all countries, but private sources made a significant contribution at the higher levels. In Hong Kong, for example, some 70 per cent of the secondary schools in existence in the early 1970s were financed by voluntary organizations or private bodies, while in Taiwan over 30 per cent of senior secondary or high school places were provided by the private sector. The tertiary sector was predominantly (two-thirds) funded by the private sector in South Korea and Taiwan, whereas in Hong Kong and Singapore it has played a less important role (Crawford 1995, p. 121; Morris 1996, p. 103; Morris and Sweeting 1995, p. 9). As for vocational and technical education, this did not feature prominently in the early educational programmes of the East Asian countries, though rural education and training were stressed in Korea and Taiwan as a means of raising productivity in agriculture through the use of new techniques and equipment as part of an agrarian reform package. The main priority at the onset was to ensure the widespread provision of good basic education so as to develop the cognitive skills of the labour force, after which on-the-job training could provide the requisite industrial skills. In any case, secondary education was much less élitist and less academically orientated than that in many Western countries. From the 1960s however there was a move to expand technical and vocational education in all countries, though more so in South Korea and Taiwan than in Singapore and Hong Kong, which continued to place greater emphasis on general academic education. The former countries developed a fairly extensive system of vocational and technical education in conjunction with their manpower planning strategies which were targeted at the growing demand for more sophisticated skills. Thus in the case of Taiwan, for example, some two-thirds of high school students were following a vocational track by 1980, compared with 40 per cent in the early 1960s (Woo 1991, pp. 1032–3). The effort seems to have paid dividends for, according to Cowley (1991, p. 7), both Taiwan and South Korea could draw upon some of the most skilled labour anywhere in the world. Perhaps what distinguishes the East Asian countries, including Japan, more than anything else is that they embraced the human capital revolution wholeheartedly and continued to retain their faith in it even when many other countries were becoming disenchanted with the results. Furthermore, unlike many Third World countries, they rejected, or rather adapted, the Eurocentric model of education in favour of one suited to their own needs. In other words, their educational systems were not simply carbon copies of the Western patterns, even though some of the better elements of the latter were incorporated in their educational programmes; and they were much more equitable than those of many poorer countries who pandered to the demand-driven pressures of élite groups (Altbach 1989, p. 26). As Mehmet commented. Education was harmonized with the labour market through curriculum design and other made-at-home reforms, to fit national manpower requirements for rapid economic development; entrepreneurship was

Education and development

175

recognized as a key development resource…[and] industrial and vocational (but not university) education was expanded and encouraged as part of ‘learning-by-doing’ industrial strategy to promote small enterprises as a basis for export-led industrialization. (Mehmet 1995, pp. 127–8) Education’s contribution to development The contribution of education to growth and development has been the subject of intensive analysis since the early days of the human capital revolution. While rates of return analysis and growth accounting exercises have provided some quantitative dimension to the debate, the results vary widely between countries and regions so that there is still no general consensus on the issue (Psacharopoulos 1973, 1985, 1988; Aldcroft 1998). In the case of the East Asian economies, many commentators have stressed the importance of education and training as a factor in their success, without providing much concrete quantitative evidence to support their statements. For example, Cowley (1991, p. 7) thought that ‘the tigers’ single biggest source of comparative advantage is their well-educated workers’. The World Bank, long an enthusiastic supporter of human resource development, believed that education and human capital formation were the largest single contributor to East Asian growth, and they reiterated this view in relation to Africa in 1994 (World Bank 1993, p. 349 and 1994, p. 25). Individual studies on Taiwan (Woo 1991, p. 1041) and Korea (Dornbusch and Park 1987, p. 397) also emphasized the importance of highly trained and productive workforces as a consequence of attention to human capital formation. On the other hand, some observers have been more circumspect in their judgement. Wheeler (1984, p. 29) was somewhat sceptical of the East Asian exemplar of educational prowess, given the wide range of conditioning factors which have favoured these economies. Similarly, the authors of the case studies in Morris and Sweeting (1995), while by no means dismissing education out-of-hand, were reluctant to assign a crucial role to education—supportive yes, but not the primary source or engine of growth. Several other studies dealing with the reasons for the East Asian success make only passing reference to the role of education and training (Brown and Liu 1992; Clark and Chan 1992; Hofheinz and Calder 1982; Kuznets 1988; Lin 1988). What is perhaps most significant in the context of education’s role in development, as far as the East Asian countries are concerned, is that they had a relatively high endowment of human capital prior to their modern economic growth. In a cross-section analysis of 98 countries for the period 1960–85 Barro (1991) found that while per capita growth rates were not related to the initial level of GDP (1960), they were substantially and positively related to the starting amount of human capital. Thus poor countries could hope to catch up with richer countries providing they had high levels of human capital endowment relative to their levels of GDP, but not otherwise. He cites the case of the Pacific Rim countries where school enrolment rates in 1960 were high relative to those

176

Derek H.Aldcroft

typically associated with their levels of per capita GDP. Thus in the case of Korea and Taiwan, the high primary school enrolment rates of 94 and 96 per cent respectively, as against predicted values of 61 and 66 per cent, helped to raise their respective growth rates by 14 and 12 per cent—although it should be noted that this factor alone was insufficient to account fully for the exceptionally strong growth of these countries. There was however a closer correspondence in the case of Japan. By contrast, in sub-Saharan African countries the reverse position held: school enrolments were low relative to those predicted on the basis of their 1960 per capita incomes and hence their subsequent growth performance was constrained. A similar point was made by Wheeler (1984, pp. 8, 10, 78–81; see also Mankin, Romer and Weil 1992, pp. 434–6) who showed that literacy change had a positive effect on output in underdeveloped countries and that only if educational provision was frozen at low literacy rates would a low level development trap emerge. This led him to conclude that the prospects for poorer countries was not all that bleak providing ‘the poor African and South Asian societies follow the human resource path which has already been blazed in Latin America and East/Southeast Asia, the predicted future is promising’. In other words, absolute poverty may not be a bar to progress as long as countries are geared up educationally to overcoming it. The East Asian experience is also in line with that of European countries. Sandberg’s pioneering work on the links between education and future income levels for European countries confirms the importance of the initial starting levels of human capital stocks (Sandberg 1982, pp. 687–97). Using adult literacy rates as a proxy for human resource stocks, he found that literacy levels circa 1850 provided ‘an amazingly good predictor of per capita income in the 1970s’. Countries that were highly literate at the start of the period, even though they may have been relatively poor in income terms, were generally the ones that subsequently attained high levels of per capita income. Conversely, the lower the initial stock of human resources, the slower the rate of modernization and income growth. It should also be noted that a high initial income combined with a high literacy level, as in the case of England and Wales, Scotland and Belgium, did not necessarily guarantee that their original relatively high income rankings would be replicated at some future date. Nevertheless, the results do broadly indicate that early educational attainment may well have been an important prerequisite for achieving advanced economic status. The reasoning behind the hypothesis is similar to that expounded by Easterlin in a seminal article published in 1981 (Easterlin 1981). That to modernize and exploit new opportunities, countries require a highly articulate and mobile labour force, an elastic supply of financial services, the ability to develop and utilize new technology, a ready supply of enterprising individuals, as well as a more rational and receptive approach on the part of the population to change and development (Cipolla 1969, p. 102). Without an adequate educational base, these attributes are unlikely to emerge. These findings also confirm the importance of basic or primary education which has frequently been stressed as the sector yielding the highest returns. Colclough (1982) emphasizes the relevance of investment in primary education for growth, productivity, fertility decline and improvements in health and nutrition, as well as helping to foster attitudes and behaviour that are conducive to change and

Education and development

177

development. Primary education is cheap relative to other forms of education, and the private and social returns to investment in this sector are higher than elsewhere (Psacharopoulos 1973, 1980, 1985, 1994; Todaro 1997, 385). Anderson and Bowman have also specified threshold levels of education or literacy which they believe are required to generate sustained modernization. According to their calculations, a literacy rate of 40 per cent or more is required to lift a country out of poverty, but self-sustaining growth in per capita income can only be achieved if primary school enrolment rates are 80 per cent or more. If not the economic system will be constrained in the long run and eventually it will retrogress (Anderson and Bowman 1976; Bowman and Anderson 1963; Bowman 1980). Reference to Table 8.3 shows that as early as 1960 the East Asian tigers had relatively high literacy rates and primary school enrolment rates that exceeded 80 per cent. In this respect they were far ahead of most Third World countries. Contrast their favourable position with that of most African and Southern Asian countries where, even as late as 1980, 60 per cent of the adult population on average was still illiterate (UNESCO 1995). The secret of the NICs success was that they concentrated initially on that sector of education with the highest pay-off and subsequently built on that foundation. This ensured equity and balance in contrast to the élitist paradigm which prevailed in many African countries. One of the remarkable features of the East Asian countries has been their success in international trade, especially the rapid shift to exporting manufactured products in an increasingly competitive world. No doubt low wage levels have played some part in this success, especially in the earlier years when labour-intensive products were the norm, but equally critical was the ability to adapt their production structures to the changing world environment, that is to shift to higher value-added, skillintensive products, which were the most dynamic in international trade terms. It is doubtful whether it would have been possible to make this transition without a skilled and educated workforce. As Castells argues: It is the ability to shift from one level of development to another and from one form of incorporation into the world economy to a more competitive one, generating higher value, that was the clue for a cumulative process of development that led to endogenous economic growth, in contrast with the short-lived phases of economic growth followed by stagnation and crisis that was the experience of most of Latin America in the 1970s and 1980s. (Castells 1992, p. 55) Choudhury and Kirkpatrick (1990, pp. 24–5) in their study of ASEAN comparative advantage, found that the rise in the quality of the labour force affected the pattern of factor use in manufacturing, as reflected in the shift to more skill-intensive forms of production. This was of crucial importance at a time when the comparative advantage in international trade was increasingly determined by the skill-intensity of production (Katrak 1982). And, as the authors point out, ‘comparative advantage is made, not given’. From this it follows that a policy strategy ‘which anticipates the pace and pattern of change in international competitiveness and seeks to

178

Derek H.Aldcroft

influence the economy’s accumulation of human capital and skills can make a significant contribution in enabling an economy to develop the production and trade structures needed to retain its competitive advantage’. On the more complex issue of the overall contribution of education or human capital improvement to economic growth, tentative estimates suggest that some 11–15 per cent of East Asian growth could be explained by this factor. This is probably higher than for most Western nations and for Latin America, but lower than for much of Africa (Barro 1991, pp. 417–18; Psacharopoulos 1983, p. 8; 1984, p. 337, 1988, p. 1). However, there is reason to suspect that most estimates of this type, which are based on some form of growth accounting framework, underestimate, for one reason or another, the true contribution of education to economic growth. Psacharopoulos (1983, pp. 4–5, 1984, pp. 341–6) has enumerated several reasons why this should be so. One of the most obvious drawbacks is that conventional methods of measuring and disaggregating the components of growth, which rely heavily on selected quantitative indicators such as school enrolments, literacy levels, number of years in schooling or investment in education, do not readily pick up qualitative improvements in the labour force arising from formal instruction. Even rates of return analysis, which rely heavily on earnings schedules, may suffer from the same defect if earnings inadequately reflect levels of productivity, though in this case there is a greater likelihood of overestimation in rates of return if earnings exceed marginal product. If, as Nancy Stokey (1991, p. 608) suggests, the qualitative component in education is of greater significance for long-term growth than the actual quantity of schooling, then the failure to identify its contribution would lead to a serious downward bias in the overall estimates of education’s contribution to growth. One of the weakest elements in the analysis of education’s return and contribution to growth is the use of labour earnings instead of productivity in the measurement of the costs and benefits of educational investment. Apart from the fact that earnings differentials due to education may not truly reflect productivity differentials, the more serious issue for less developed countries is the use of urban wage differentials which lead to a downward bias in the contribution of education in the rural sector. Studies have shown that rural education can be very important in improving productivity performance through the adoption of new methods and more efficient practices (Psacharopoulos 1983, p. 7). This point would also be relevant to those East Asian countries which encouraged the development of rural education in the early post-war years. An equally serious issue is the question of on-the-job training. If a comprehensive reckoning of all forms of education and instruction is to be obtained, this should be included in any growth accounting exercise. Yet since Mincer (1962) first identified the importance of on-the-job training for the USA economy—which in terms of costs was as important as formal education for the male labour force with rates of return comparable with those from formal education—there has been relatively little work done in this field, and certainly no attempt to incorporate it into growth accounting models, or even to relate it to more eclectic analyses of the sources of growth. For rates of return analysis this does not present a problem

Education and development

179

since, in theory at least, the relevant marginal wage should reflect the presence or otherwise of on-the-job training in any particular labour cohort. Conversely, in so far as growth accounting estimates rely heavily on measures such as years of schooling, school enrolments and the like, they clearly do not pick up the contribution of any instruction which occurs outside the formal education system. This does of course raise the issue as to what form education should take. Should it be all embracing and include the whole life-experience of an individual, including learning by doing and instruction in the home and on the job, or should it be confined more narrowly to education conventionally defined in terms of schooling and college instruction? Finally, there is the more nebulous issue, and one which eludes quantification, and that is whether education was important to the national identity building and social cohesion of the countries here under discussion. Certainly compared with Western models, education had more than an instructive role to play. Japan had, in this respect, been the forerunner since traditionally education had been designed with the purpose of serving the needs of the state: in particular as a means of inculcating law and order, encouraging integration, improving moral standards and fostering national identity building and social cohesion. Only later, in the postwar decades, was it geared more towards promoting economic development (Beauchamp 1987, pp. 302, 310). According to Hanley (1993, p. 80), it was ‘through the modern educational system that the Japanese were unified into one nation and inculcated with the goals that led to Japan’s industrialization and modernization’. As already noted, a similar purpose was enshrined in the early post-war education strategies of the East Asian tiger countries, and it was only subsequently that economic priorities took greater precedence in the formulation of educational strategies. Since all four countries were in general disarray, economically and socially, after the war, and two were of mixed nationality, it is conceivable that such policies did help to foster stronger national identities and social cohesion than would otherwise have been the case. Conclusion There are still many gaps in our knowledge as to the way in which education contributes to the development process. It is certainly important, though obviously it is only one of several factors which has been responsible for the rapid growth and modernization of the four tiger countries. On the other hand, while some writers would be reluctant to accord it the number one priority, it seems inconceivable that the four East Asian countries could have transformed themselves from ‘basket cases’ to super-growth states had they not made a concerted effort to raise significantly the quality of their human capital stock. What makes the East Asian experience distinctive is the continuity of the educational drive and the strategy of prioritizing goals in education and training which matched the stage of development. Such features are highlighted by David Ashton and his colleagues in their study of education and training in the four tiger countries:

180

Derek H.Aldcroft

…part of the distinctiveness of the Asian newly industrialised economies is not just the level and type of investment in human resource development, but their achievements in linking that investment to the requirements of the economy at different stages of economic growth. This has enabled them to sustain growth by moving from low valueadded to high value-added production without that change being hindered by a lack of skilled personnel. (Ashton, Green, James and Sung 1999, ch. 7) To achieve these objectives required the state to steer the economy and human skill formation in tandem so that rapid economic growth was matched by an equally rapid transformation in the stock of skills. For this purpose it was necessary to have a fairly strong centralist state with clear mechanisms for guiding both the direction of the economy and the pattern of human resource development. For the most part the tiger countries, with some reservations in the case of Hong Kong, conform to a statist development model whereby education and training were made an integral part of the process of economic transformation. Thus, as in the case of Japan, education may be seen as an integral part of the process of modernization. However, until formal growth accounting models can capture more of the unknown elements and until we learn more about the intricate ways in which education influences economy and society in general, the overall contribution of education to the growth process must remain rather speculative. As Hanley (1993, p. 80) has noted in the case of Japan, ‘…it is impossible to capture the ways in which it affected the country’s history using only economic variables that can be analysed through multiple regression analysis’. And, as Myllyntaus (1993, p. 166) has reminded us, however important human capital development may be for modern economic growth, it does not automatically produce it. ‘The widespread ability to read and write is not a magic charm that ignites the Industrial Revolution in any societal circumstances.’ References Aldcroft, D.H. (1991) ‘World income distribution’, The Economic Review, 9. —(1998) ‘Education and development: the experience of rich and poor nations’, History of Education, 27. Altbach, P.G. (1989) ‘Twisted routes: the western impact on Asian higher education’, Higher Education, 18. Anderson, C.A. and Bowman, M.J. (1976) ‘Education and economic modernization in historical perspective’, in L.Stone (ed.) Schooling and Society: Studies in the History of Education. Baltimore: The John Hopkins University Press. Ashton, D., Green, F., James, D. and Sung, J. (2000) Education and Training for Development: The Political Economy of Skill Formation in East Asian Newly Industrialised Economies, London: Routledge. Balassa, B. (1988) ‘The lessons of East Asian development; an overview’, Economic Development and Cultural Change, 36, Supplement.

Education and development

181

Barro, R. (1991) ‘Economic growth in a cross section of countries’, Quarterly Journal of Economics, 106. Beauchamp, E.R. (1987) ‘The development of Japanese educational policy 1945–85’, History of Education Quarterly, 27. Bercuson, K. (1995) ‘Singapore: a case study in rapid development’, International Monetary Fund Occasional Paper, 119. Bowman, M.J. (1980) ‘Education and economic growth: an overview’, World Bank Staff Working Paper, 402. Washington, DC: World Bank. Bowman, M.J. and Anderson, C.A. (1963) ‘Concerning the role of education in development’, in C.Geertz (ed.) Old and New States: The Quest for Modernity in Asia and Africa, New York: The Free Press of Glencoe. Brown, R.H. and Liu, W.T. (1992) Modernization in East Asia, Westport, Connecticut: Praeger. Castells, M. (1992) ‘Four Asian tigers with a dragon head’, in Appelbaum, R.P. and Henderson, J (eds) States and Development in the Asian Pacific Rim, London: Sage Publications. Choudhury, A. and Islam, I. (1993) The Newly Industrialising Economies of East Asia, London: Routledge. Choudhury, A. and Kirkpatrick, C.H. (1990) ‘Human resources, factor intensity and comparative advantage of ASEAN’, Journal of Economic Studies, 17. Cipolla, C. (1969) Literacy and Development in the West, Harmondsworth: Penguin Books. Clark, C. and Chan, S. (1992) The evolving Pacific Basin in the Global Political Economy, Boulder: Lynne Rienner Publishers. Colclough, C. (1982) ‘The impact of primary schooling on economic development: a review of the evidence’, World Development, 10. Cowley, A. (1991) ‘Asia’s emerging economies; burning bright’, The Economist, 16 November, 321. Crawford, L. (1995) ‘The development of secondary education in Hong Kong, 1945–71’, History of Education, 24. Dornbusch, R. and Park, Y.C. (1987) ‘Korean growth policy’, Brookings Papers on Economic Activity, 2. Easterlin, R.A. (1981) ‘Why isn’t the whole world developed?’, Journal of Economic History, 41. Gannicott, K. (1990) ‘The economics of education in Asian–Pacific developing countries’, Asian–Pacific Economic Literature, 4. Hanley, S. (1993) ‘The relationship of education and economic growth: the case of Japan’, in Tortella, G (ed.) Education and Economic Development since the Industrial Revolution, Valencia: Generalitat Valenciana. Hofheinz, R. and Calder, K.E. (1982) The Eastasia edge, New York: Basic Books. Katrak, H. (1982) ‘Labour skills, R&D and capital requirements in the international trade and investment of the United Kingdom, 1968–78’, National Institute Economic Review, 101. Kimura, M. (1990) ‘Diffusion of primary education in Korea, 1911–1955’, in Tortella, G (ed.) Education and Economic Development since the Industrial Revolution, Valencia: Generalitat Valenciana. Kuznets, P.W. (1988) ‘An East Asian model of economic development: Japan, Taiwan, and South Korea’, Economic Development and Cultural Change, 36. Lin, C. (1988) ‘East Asia and Latin America as contrasting models’, Economic Development and Cultural Change, 36. Maddison, A. (1989) The World Economy in the 20th Century, Paris: OECD. —(1995) Monitoring the World Economy 1820–1992, Paris: OECD.

182

Derek H.Aldcroft

Mankin, N.G., Romer, D. and Weil, D.N. (1992) ‘A contribution to the empirics of economic growth’, Quarterly Journal of Economics, 107. Mehmet, O. (1988) Human Resource Development in the Third World, Kingston, Ontario: Ronald P.Frye. —(1995) Westernizing the Third World, London: Routledge. Mincer, J. (1962) ‘On-the-job training: costs, returns, and some implications’, Journal of Political Economy, Supplement, 70 (1962). Morris, P. (1995) ‘Education and development: an introduction’, in Morris, P. and Sweeting, A. (eds) Education and Development in East Asia, New York: Garland Publishing. —(1996) ‘Asia’s four little tigers: a comparison of the role of education in their development’, Comparative Education, 32. Morris, P. and Sweeting, A. (eds) (1995) Education and Development in East Asia, New York: Garland Publishing. Myllyntaus, T. (1993) ‘Education in the making of modern Finland’, in Tortella, G (ed.) Education and Economic Development since the Industrial Revolution, Valencia: Generalitat Valenciana. Oshima, H.T. (1988) ‘Human resources in East Asia’s secular growth’, Economic Development and Cultural Change, 36. Psacharopoulos, G. (1973) Returns to Education, Amsterdam: Elsevier. —(1980) ‘Returns to education: an updated international comparison’, World Bank Staff Working Paper, 402. Washington, DC: World Bank. —(1983) ‘Educational research at the World Bank’, The World Bank Research News, 9. —(1984) ‘The contribution of education to economic growth: international comparisons’, in Kendrick, J.W. (ed.) International Comparisons of Productivity and Causes of the Slowdown, Cambridge, Mass.: Ballinger Publishing Company. —(1985) ‘Returns to education: a further international update and implications’, Journal of Human Resources, 20. —(1988) ‘Critical issues in education: a world agenda’, International Journal of Educational Development, 8. —(1994) ‘Returns to investment in education: a global update’, World Development, 22. Reynolds, L.G. (1986) Economic growth in the Third World: an Introduction, New Haven: Yale University Press. Rock, M.T. (1993) ‘Twenty-five years of economic development revisited’, World Development, 21. Sandberg, L.G. (1982) ‘Ignorance, poverty and economic backwardness in the early stages of European industrialization. Variations on Alexander Gerschenkron’s grand theme’, Journal of European Economic History, 11. Stokey, N. (1991) ‘Human capital, product quality and growth’, Quarterly Journal of Economics, 106. Todaro, M.P. (1997) Economic Development, 6th edition. London: Longman. UNESCO (various years) Statistical Yearbooks, Paris: UNESCO Publishing. Wheeler, D. (1984) Human Resource Policies, Economic Growth and Demographic Change in Developing Countries, Oxford: Oxford University Press. Woo, J.H. (1991) ‘Education and economic growth in Taiwan: a case of successful planning’, World Development, 19. World Bank (1990) World Development Report 1990, Washington, DC: World Bank. —(1992) World Development Report, Washington, DC: World Bank. —(1993) Adjustment in Africa: Reforms, Results and the Road Ahead, Oxford: Oxford University Press. —(1994) The East Asian Miracle: Economic Growth and Public Policy, Oxford: Oxford University Press.

Education and development

183

—(1995) World Development Report, Washington: World Bank. Young, A. (1994) ‘Lessons from the East Asian NICs: a contrarian view’, European Economic Review, 38. —(1995) ‘The tyranny of numbers: confronting the statistical realities of the East Asian growth experience’, Quarterly Journal of Economics, 110.

Part III

9

Another monetary economy The case of traditional China Akinobu Kuroda

Introduction John Hicks believed that money was created by the merchant. Merchants found it convenient if the value of a commodity could be ‘stored’ in some way to carry out exchange, so Hicks saw money originating in the ‘store of value’ function (Hicks 1969, p. 63). He noted in his Theory of Economic History, however, that ‘Chinese money did not pass through a stage in which the “store of value” function was preeminent, [it] passed straight into being a means of payment’ (Hicks 1969, p. 68). This sentence suggests that Hicks admitted the possibility of several paths of monetary evolution, but he did not explicitly outline what caused these diverse paths of development. Max Weber thought about monetary evolution in a similar way. In his view, money in traditional China was a typical case in which currencies issued by the authority circulated as a means of payment, and their intrinsic value was so small that the government could make a good profit from seigniorage (Weber 1964, pp. 285–86). The same idea is found in Hicks, referring to money in ancient Greece: ‘There are bronze coins, unquestionably, token money, pure means of payment, which must have circulated at values in excess of their intrinsic values’ (Hicks 1969, p. 67). Both Hicks and Weber share the view that instead of the units being of intrinsic value, the authorities, by virtue of their power, create a currency that is accepted by society. This was especially the case with nonprecious metal currencies. Copper cash had been the main currency in China up to the sixteenth century when silver began to circulate in large quantities. But are Hicks and Weber right in assuming that the government obtained a large profit from seigniorage? Actually, the reverse was the case. It is true that sometimes the nominal value of copper cash issued by the Chinese dynasties was less than its production cost, even when issued in large quantities. The eleventh century under the Northern Song dynasty (960– 1127) and the eighteenth century under the Qing dynasty (1644–1911) are good examples of this. It is interesting that less private counterfeit cash with low intrinsic value circulated during these periods than at other times (Peng 1958, pp. 300–301; Kuroda 1994, pp. 54–55). That is, the official copper cash acted as a currency with good intrinsic value. Hence, in this instance, good money drove bad money out of circulation.1 The counter-Gresham’s-law phenomenon was not an instantaneous one, but it lasted for a half century. What does this paradoxical phenomenon mean? 187

188

Akinobu Kuroda

The purpose of this chapter is to examine the distinct characteristics of the monetary system under the Chinese empire, and to clarify its meaning in comparison with the rest of the world. In order to achieve these objectives, this chapter is divided into five sections, including the introduction. The second section examines how copper cash was issued and used in traditional China. It will show that copper cash was actually in circulation coming from local sources, although copper cash was officially supplied by the government alone. Local currency supplies were created by loose agreements, here called ‘currency circuits’, which were honoured by local merchants. This is followed in the third section by a more precise look at these currency circuits. Besides the circulation of copper cash, other local currencies existed, including monetary substitutes like rice or cloth, throughout China’s long history. The absence of the monetary use of precious metals, i.e. gold and silver, before the sixteenth century in China shows us how a market can exist even when there is little money suitable for inter-regional settlement. What happened to China when a great quantity of silver flowed into the country? This is the next question. The fourth section shows how Chinese society between the sixteenth and eighteenth century linked inter-regional convertibility of silver to local liquidity sustained by copper cash. A flexible relationship between silver and copper cash was established during the period, which preserved the independent characteristics of the local economy. The final section offers a brief summary of the arguments presented in this chapter, and provides general insights for monetary history from the point of view of nonprecious metal currencies. The copper cash economy in traditional China What is common sense in one society is not always so in another society. It is quite understandable that a society with only copper cash had its own view of money, utterly different from that of a society using precious metal money. According to common sense, it may be assumed that a government with the power to issue currency tends to supply more face value currency than the production cost in order to profit from seigniorage. Indeed, profit from seigniorage was an important source of income from the medieval era for kings or feudal lords in Europe in order to supplement the fiscal deficit. Sultans in west Asia or India did the same thing, and so did the shoguns in Tokugawa Japan.2 This, however, was not the case with the Chinese dynasties. They had another reason for the issuing currency and conducted monetary administration in a different way. Look at a memorial written by an official in the sixteenth century, when China suffered from a serious shortage of copper cash. If the ministers of Works [responsible for the mint] argue that coining 5 wen [copper cash] by spending one fen (a hundredth liang) of silver means suffering a 2-wen loss on the grounds that one fen of silver equals 7 wen, they do not understand that changing silver into copper

Traditional China: A monetary economy

189

cash is not like pouring water into fire. Besides spending one fen of silver between heaven and earth [that is, human society], adding copper 5 wen equalling 8 li (a thousandth liang) means that human society gets 80 per cent interest in addition to 100 per cent principal. (Yang C. n.d.) The official apparently thought that, in spite of suffering a loss, the government should issue copper cash. In his view, something ‘beneficial to people’ (bianmin) ought to be superior to profits for the government. Although the high cost of issuing the currency did not allow the dynasties to follow this principle easily, the same argument appears repeatedly throughout Chinese history. In reality, the production cost of copper cash sometimes surpassed the face value, as mentioned above. An essay written in 1125 shows us that, although according to the price of copper set by the government the intrinsic copper value of 1000 wen was 600 wen, adding the cost of transport, fuel, salaries and so on, the total cost of 1000 copper cash amounted to 1400–1500 wen (Yang S. 1125, p. 132).3 Even this cost was probably underestimated, because the copper price in the market was almost certainly higher than the government’s bargaining price. During the second half of the eleventh century and just before the essay was written, the Northern Song dynasty issued the largest amount of copper cash in Chinese history. Throughout the period, the good coins issued by government were melted down, in spite of a legal ban on melting.4 That means that people calculated the value of the official coin was higher than its face value. The same story of melting down copper cash was repeated during the eighteenth century under the Qing dynasty (Miyazawa 1998, pp. 342, 368–69; Kuroda 1994, pp. 42, 53). Thus there was a world in which the government issued currency, while gaining little profit from seigniorage, and sometimes even suffering a deficit. Naturally, Chinese dynasties sometimes broke this principle by issuing less costly currencies when forced to raise extra money for fighting intruders or suppressing rebels. Issuing bad money, however, heralded the end of a dynasty. In peaceful times, the fiscal pressure of high issuing costs usually made the dynasties reluctant to supply copper cash. That is why, although copper cash was the only legal tender, its issuance was frequently suspended. Throughout the two-thousand-year long history of the Chinese empire, there were three centuries when the governments supplied copper cash in sufficient quantity. These were the second century BC under the Western Han dynasty (BC 206-AD 8), the eleventh century under the Northern Song dynasty, and the eighteenth century under the Qing dynasty. The first two periods followed the political unification of a divided China, while the last period followed a huge inflow of silver. But why did they not try to issue less costly cash with a lower copper content? If the production cost of 5 wen was 7 wen at that time, why did they not lower the copper content to a 5-wen cost level? The main reason given in memorials written by officials is that lowering the standard of official coins would encourage counterfeiters to issue more illegal coins. They seemed to believe that a high copper content in cash made illegal minting less profitable. However, if the copper cash supply was not sufficient for the demands of daily transactions, the official standard

190

Akinobu Kuroda

became irrelevant, because counterfeiters would issue illegal cash to meet the demands of the local markets. So the higher the copper content of official copper cash was, the larger the profits from counterfeit cash became. So it is doubtful that the high quality of copper cash issued by the authorities effectively discouraged counterfeiters. The relationship between the circulation of private copper cash and the circulation of official cash must be examined. Take the case of Quanzhou, Fujian province, in 1606. For two or three decades previously, the Ming dynasty issued a large quantity of copper cash called Wanli Tongbao, and this official copper cash prevailed along southern coastal China, including Quanzhou. However, as soon as famine struck the Quanzhou region in 1606, counterfeit cash with less intrinsic value appeared in the region as the price of rice rose. Despite lacking any institutional guarantee, such as given by the merchant guilds, the forged cash instantly prevailed in the local market. Clearly the circulation of the counterfeit cash was related to the panic caused by the shortage of rice during the famine. The market of Quanzhou needed a rapid increase in the supply of currency to support a price level which was high enough to attract rapid rice imports quickly. That is why, after sufficient rice had been brought in by sea, the price of rice fell and the counterfeit cash disappeared (Chen n.d., v.2, 12b–13a). This process shows that the local market had the capacity to adjust the currency supply to local demand, in this case to respond to the increased saleability of rice.5 In other words, the local market could regulate local liquidity without institutional support. Had it not been copper, the local market would have created a monetary substitute, like cloth, as other regions in China did at that time. On the other hand, what made the official cash prevail in Quanzhou before the famine year was that it was in sufficient supply for normal demand.6 As long as official copper cash issued in sufficient quantities was better money than private cash, local people preferred the former to the latter. So if the central government wished to eliminate private currencies from local markets throughout the empire, they had no other choice than to issue good money in large quantities, as the Northern Song and the Qing dynasty actually did. The point about monetary policy under the Chinese empire is not that local currencies were eliminated, but that they were restricted. Restriction was necessary because the government did not want local liquidity to become too independent. The reason why the dynasties issued a large supply of good cash of small denomination was to induce local markets to use the unified currencies. However, authorities could not help but allow the markets to create local units of account. Even if local merchants used official copper cash, they always made their transactions with them according to various units of account they agreed on among themselves. The mutual use of the unit of account by local merchants is called a ‘currency circuit’, as will be examined more precisely in the next section. Local liquidity and currency circuit Compared with silver, copper cash was too heavy to carry long distances. If the exchange ratio of silver to copper cash was one silver tael (liang or 37.3 gram) to

Traditional China: A monetary economy

191

1,000 wen, as the Qing dynasty tried to establish in the seventeenth century, merchants had to carry approximately 4 kg of copper cash to buy one tael worth of goods. It is not surprising that transporting 1,000 copper cash sometimes cost more than 1,000 wen. Copper cash was also too bulky to store as an asset. So merchants tried to create a more convenient system for high-value transactions. The ‘short string’ custom was one of them. Although one string (guan or diao) of copper cash originally meant tying 1,000 copper cash coins by string, merchants used a string bearing less than 1,000 coins as one guan or less than 100 cash as bo (100 wen). For example, ‘77 bo’ meant that a string of 77 copper coins was used as if 100 wen. An essay written in the middle of the twelfth century depicted the monetary custom in Kaifeng, the capital city of the Northern Song, as follows. Officials were using the seventy-seven bo, in the streets the seventyfive bo was generally used, in transactions of fish, meat, and vegetable the seventy-two bo was used, in gold and silver they used the seventyfour bo, in pearl transaction, hiring women slaves, and buying insects they used the sixty-eight bo. (Meng 1147) Although the 75 bo was a popular short string in the city, people used different units of account depending on the transaction, and other values of short string were in use in other regions. The custom of the short string in the Song period originated at the turn of the eighth century in the Tang period. It began as a private custom. To begin with, the dynasty banned it and tried to enforce full-string use, but failed. Under the Song, the government also set 77 bo as the official short string and overlooked the use of various units of account (Miyazawa 1998, pp. 302–7). The same practice in transactions based on copper cash existed in the early twentieth century. Although the Qing government established 980 coins as one diao, local people used various short strings depending on the region and on the goods, just as people in the Song period had done earlier (King 1965, pp. 58–68). The Chinese dynasties tried to keep uniform copper cash as the only legal tender, as mentioned in the previous section, but they adopted a laissez-faire policy as far as local units of account were concerned. Thus the combination of a unified currency with adjustment to local needs was one of the main characteristics of the monetary system of the Chinese empire. The adjustment to the needs of local liquidity is found in other periods. From the turn of the fifteenth century to the end of the seventeenth century, the custom of preferring old coins became a remarkable phenomenon. In the Ming period, the dynasty did not issue copper cash in large quantities until the last seventy years. Lacking legal tender, many regions used commodity currencies, such as rice, cloth, and so on. Even in regions using copper cash, old coins were used. Song coins were preferred, especially in southeast China. The coins in circulation did, however, include a large number of counterfeit units. A local gazette edited in

192

Akinobu Kuroda

the early seventeenth century explicitly depicts counterfeiters producing copper cash impressed with a name of the Song era, even changing the era name every so often (von Glahn 1996, p. 103; Kuroda 1998, p. 9). The quality of forged coins ranged from those close in quality to the original to coins made mostly of lead. Local merchants would differentiate cashes between standard cash and current cash, but rejected some kind of cashes. In the case of Putian county, Fujian province, in the second half of the seventeenth century, local people used Song cash at three times the value of the current dynasty’s cash. Since the intrinsic value of the former cash was twice the latter, Song cash was apparently overvalued against cash issued by the Qing dynasty. Wealthy merchants in Putian saved Song cash for their assets, meanwhile using Qing cash in daily transactions. The authorities often tried to ban the use of Song cash, and merchants resisted by closing their shops (Kuroda 1998, p. 7). Thus local merchants differentiated between copper cashes by dividing them into currency for saving assets (standard cash) and currency for daily transactions (current cash). This process is found over the periods. For example, in Shanxi province under the Northern Song, people saved copper cash while using iron cash imposed by the government for ordinary transactions (Miyazawa 1998, p. 473). Paper money can be interpreted in the same way between the twelfth and the early fifteenth centuries under the Southern Song (1127–1279), the Jin (1115– 1234), the Yuan (1234–1368), and the early Ming (1368–1644). These dynasties issued paper money in large quantities instead of copper cash. In the case of the Southern Song, the government actually depended on paper money (huizi). But an essay, written in the latter half of the thirteenth century, shows that peasants in a rural market near the state capital were using rice as a means of exchange (Miyazawa 1998, p. 58). Commodity currencies like rice as well as copper cash were used in daily transactions, while paper money issued by the governments were probably used as a unit of account or to store assets, so long as they were not overissued.7 Two points emerge. Firstly, quite apart from legal tender issued by the government, local markets tended to create local currencies, including commodity currencies, and to adjust independently to the need of local liquidity. Secondly, local markets also tended to separate a currency for daily transactions from a means of storing value. The second point relates to the reason China absorbed silver in large amounts from the sixteenth century, and what subsequently happened to China. It should be noted that no authorities or local oligarchies issued orders to the local markets. All rules of local markets were loose agreements among merchants. They used local arrangements on ‘short string’ or standard cash. However, if these became unprofitable, the merchants could create other agreements to replace them. A currency circuit was an independent arrangement, but was not sealed off from other circuits, or inflexible. Silver and copper cash A bureaucrat in China, in a memorial of 1567, reported that silver was extremely inconvenient for the small transactions of the people. He also added, with reference

Traditional China: A monetary economy

193

to Shanxi and Shaanxi provinces (northwest of China) where silver was used, that travellers had to cook their own meals because copper cash was not available there, so there was no way to pay for a meal (Ge 1582, 16a). But the opposite situation is given in a report of the Korean envoy to Japan in 1429. The envoy reported with surprise to the king that in Japan copper cash prevailed, so that travellers did not carry rice but copper cash (Kuksa P’honch’an Wiwonhoe 1982, v.3, p. 207). These reports show that, in East Asia where silver was not popular as money until the sixteenth century, copper cash was generally used for daily transactions. What was the case of societies using silver? An example is found in medieval England, where silver was the only currency in circulation. In England during the thirteenth and fourteenth centuries, loaves of bread varied in size according to the price of wheat. According to P.Spufford’s explanation, ‘it would have been impossible to charge a variable price for a fixed size of loaf, since there were no coins in circulation sufficiently small to accommodate the gradations in price involved’ (Spufford 1988, p. 238). This reveals that, under circumstances where small-denomination currency was lacking, the fluctuation of prices was expressed by increasing or decreasing scales of weights or measures per fixed monetary unit. In 1279 and 1280, King Edward I ordered the issue of one half penny and one quarter penny coins to supply the demand for petty payments of less than one penny. However, even a half penny equalled about 2lb weight of white bread or twice the weight of coarse bread (Spufford 1988, p. 235). Under these circumstances, the daily transactions had to be based on barter, and money units used just as means of account.8 Wages of construction workers in southern England in the 1280s under King Edward I were only ninepence per week. At the end of the thirteenth century the mint in England issued the highest number of silver coins prior to the eighteenth century (Spufford 1988, pp. 204–5). In contrast, in 1480 when the Ming dynasty had issued very little currency for half a century, vegetable sellers or transport workers in Beijing were said to spend twenty to thirty copper cash coins to feed their families for one day (Dai n.d., v. 2, p. 252).9 Although this was probably the time least legal tender was issued between the tenth and twentieth centuries, there is no doubt that ordinary people in China were using currency in their daily transactions much more frequently than their counterparts in England. When the Ming dynasty ceased supplying legal tender, from the turn of the fifteenth century, the ‘currency circuit’ in local markets became more independent and diverse. Some regions returned to rice or cloth use, while others created a system dividing saving cash from current cash, as mentioned previously. Some regions, however, began to introduce silver as unit of account and as a means of exchange. As more silver was imported from Japan illegally (wokou) from the 1540s, more regions began to use silver (Hamaguchi 1969). Silver was, however, ill-suited to daily transactions and caused the system of currency use for small transactions to change. Adoption of silver as a unit of account for taxation by the Ming dynasty and the continuous silver inflow from the second half of the sixteenth century resulted in excessive currency convertibility between regions. This made the grain trade within

194

Akinobu Kuroda

a region unstable, as grain could be easily sold to other regions in exchange for silver, irrespective of local demand for grain. This in turn resulted in a seventeenthcentury crisis. Societies responded by providing a variety of local currencies in quantities never seen before. In China, there was no general increase in prices during the seventeenth century despite a continuing silver inflow due to the trade surplus (Kishimoto 1997). This suggests that, though the massive influx of silver could stimulate inter-regional trade, it did not always result in an increase in the currencies in daily use within a region. The crisis in China saw regional riots against grain being sold to other regions. As more regions turned to silver, local people gathered to stop grains being sold from their locality. These riots peaked in the first half of the eighteenth century. By the middle of the century, however, the riots had died out, even though grain prices continued to rise through the second half of the century (Kuroda 1994, pp. 64, 121). Why? Two strong policies carried out by the Qing government changed the situation. First, the government issued more copper cash than had been issued since the eleventh century. The mints issued more than 3 million guan annually of copper cash of good quality, the Qianlung Tongbao. This discouraged the circulation of counterfeit cash. Second, the dynasty encouraged grain to be stockpiled locally. The amount of stock in the local official granaries reached an unprecedented level that century. The Qing government tended to prevent the import of grain from other regions to make up for local shortages. Even grain-deficit districts were compelled to form their own grain stocks. The renewal of grain stocks was done on a local basis by paying with copper cash, the local supply of which was also increased in this period. Thus the granary system created an easy exchange between grain and copper cash (Kuroda 1994, pp. 74–91). While the dynasty did not issue any silver coins, it continued issuing copper cash, especially during the eighteenth century. No exchange ratio was fixed between the two currencies. The Chinese empire thus stabilized local economies by disconnecting local currencies within a region from silver by issuing huge amounts of copper cash and by stockpiling local grain. As a result, in the Qing dynasty, silver bullion came to be used as the inter-regional currency, and copper cash as local currency. The monetary system naturally affected the pattern of price trends. In terms of the exchange rates for grain, copper cash, and silver, the quotations of both grain and copper cash fluctuated against silver, rather than the price of grain (goods) fluctuating against copper cash and silver (moneys). In the middle of the eighteenth century, a structure was created whereby local assets, grain and copper cash, were overvalued against silver, the inter-regional currency (Kuroda 1994, pp. 84–91). The situation stimulated the flourishing rural markets, like the periodic local fairs (shiji). Thus the introduction of silver did not change the independent nature of local currencies in China. On the contrary, it reinforced this characteristic of the Chinese monetary system. Silver also came to be used in different units of account by the tael system. Just as with the short string custom of copper cash, people used various silver taels that differed by region or by goods. Although the Qing government

Traditional China: A monetary economy

195

enforced the use of the official tael (kuping tael) in the payment of taxes, it was just one of many coexisting taels (King 1965, pp. 70–81; Kuroda 1994, p. 35). The situation was like the official short string (shengbo) in the Song period. From the turn of the eighteenth century, silver dollar coins from Mexico began circulating in southeast coastal China, and lesser dollars replaced copper cash in some regions. Even in these regions, however, the silver tael did not disappear but survived as a unit of account until the 1930s. The ratio of the silver tael to the silver dollar also fluctuated, as did the ratio to copper cash (King 1965, pp. 88–90). Regardless of metallic content or form, various ‘currency circuits’ coexisted in nineteenth-century China. Conclusion It has been possible to confirm the independent features of local currencies in traditional China. If the interpretations above are correct, how could local markets maintain their monetary separation? Agricultural societies have always seen large fluctuation in demand for money. During the harvest season, large amounts of small-value currencies are needed to pay scattered peasants, and these currencies are hoarded within the local area during the farmers’ slack season. So currencies need to perform two functions which are inconsistent with each other. One is that of a local currency that circulates and is stored in accordance with local demand. The other is the function of an interregional currency with which local currencies can be convertible. This links with the functions of store of value and unit of account. Local markets tended to separate the two functions into current cash and standard cash, as mentioned in the third section. In the case of traditional China, these became copper cash and silver.10 Hicks distinguished local currency from ‘great’ currency. The ‘great’ currencies are ‘widely used in international or distant commerce’, and the circulation of local currency is ‘confined to the area which the government controlled’ (Hicks 1969, pp. 89–90). The case of the monetary system of the Chinese empire, however, suggests that more attention must be given to the relation between local liquidity and inter-regional convertibility within one country.11 Local economies can be classified into two types according to how they link local currencies with currencies with inter-regional acceptance. In the first type, local currencies are not linked to inter-regional currencies, while in the second, local currencies are linked to inter-regional currencies and their quantity fluctuates according to the balance of external trade. This classification is similar to our concept of a monetary economy, and some mercantilist nations succeeded in attaining it. By contrast, the Chinese empire allowed the local economies to preserve their own seasonal currency fluctuations and price adjustments, and at the same time tried to sustain political unification. The huge inflow of silver did not change the system. China maintained unification through establishing a dual system of silver and copper cash, as seen in the fourth section. Thus, returning to Hicks’ suggestion at the beginning of this paper, there are two points. The first is related to the political and economic structure of the Chinese

196

Akinobu Kuroda

empire. The aim of the monetary policies of the empire was not to profit from seigniorage, but to restrict the autonomy of local currencies so as to keep the vast territory under control. The second point has a broader implication that can be applied to other societies. The local market, based on transactions among peasants and merchants, has different features from markets ruled by oligarchies. The seasonal fluctuations of demand for money and the inelasticity of its supply, resulted in an autonomous local currency supply independent of movements of currency with inter-regional convertibility. The case of traditional China is a very typical one. The significance of this aspect of monetary history can be overlooked when the concept of a monetary economy based on precious metal currencies alone is examined.12 Notes 1 We must accept the operation of Gresham’s-law in traditional China. Bad money often drove good money out of circulation, or made local people put a premium on good money. However, in the latter cases the premium on copper cash was not always related to the metallic content. 2 For example, in Ottoman Turkey the government usually took a seigniorage profit from issuing copper as well as silver or gold coins (Pamuk 1997). It was not until the sixteenth century that states in Western Europe began to issue copper coins. They issued them in large quantities in order to cover their fiscal deficit (Motomura 1994). 3 In his book Miyazawa insists that the production cost of copper cash could not surpass the face value. This material and Yang Cheng’s letter, however, are clearly inconsistent with Miyazawa’s interpretation. Even if his understanding of the materials themselves is correct, the fact that copper cash coins were melted down through the Northern Song period proves that markets valued the copper contents of cash higher than the face value, irrespective of what official value was. 4 An official reported in 1075 that copper ware made of copper cash was worth five times the face value of the original coins (Li 1183, v. 19, p. 6594). 5 The concept of ‘saleability’ was originally used by Menger to indicate what makes a good become a money. In his view, the most saleable good can become money for itself. In the case of this famine, however, the saleability of rice as the most saleable good had to be realized by other measures (Menger 1892). 6 Even the Wanli Tongbao, however, was only in circulation in part of the empire. In Jiangxi province, Wanli Tongbao was no longer in circulation a few years after they were issued (von Glahn 1996, p. 102). 7 Maeda persuasively argues that paper money issued by the Yuan had actually been in circulation. The materials he cites, however, also prove that the paper money was not usually issued in small denomination for daily use (Maeda 1973, pp. 83–86) 8 In some cases, however, the scarcity of currency encourages the use of credit transactions. Britnell’s study of a local town in medieval England shows that the burghers maintained the local market through the frequent use of credit transactions in the period of silver ‘famine’ (Britnell 1986). 9 Another memorial reported that the poor earned only 20 wen per working day, but that it was easy to buy daily supplies at ten or five wen (Dai n.d., v. 1, p. 340). 10 The dual monetary system inevitably intensified the seasonal fluctuation of demand for copper cash. Traders of local goods had to obtain copper cash in large quantities to buy peasant products after harvest. For example, in the second half of 1890s, 90

Traditional China: A monetary economy

197

million coins were transported from the port of Jiaozhou to rural areas to purchase groundnuts every winter (Zheng 1991, p. 126). 11 K.Polanyi also referred to ‘the distinction between local and external money’ (Polanyi 1977, p. 258). Bronze coins in ancient Greece as well as cowries in West Africa are cited as examples of ‘local money’. According to his view, however, the redistributive mechanism or the reciprocal relationship within the region lies behind the distinction. Even if his images of ancient Greece and West Africa are accurate, the case of ‘currency circuit’ in traditional China above is utterly different from the redistributive or reciprocal system. Within the currency circuit, not only merchants but also peasants competed with each other to maximize their profits. 12 We can find that the cowry shares similar attributes with copper cash (Hogendorn and Johnson 1986). In Bengal, during the seventeenth and eighteenth centuries, the exchange rate of cowry to silver continued to rise as copper cash to silver did in China during the same period (Mahapatra 1969–70). The tendency for appreciating local currencies to appreciate in China and India during this time was in contrast to Western Europe of the seventeenth century where people tended to put a premium on silver against debased copper coins (Glassman and Redish 1988). Cowry has been circulated as a currency in Yünnan province, Southwestern China, even in this century, and possibly its circulation had already begun as early as the ninth century (Vogel 1993, p. 220). Although cowries are also found in remains of ancient Chinese dynasties, such as the Yin dynasty (circa BC 1700-circa BC 1100), it is not clear whether they were used as currencies or as ornaments (Peng 1958, pp. 7– 20).

References Britnell, R.H. (1986) Growth and Decline in Colchester, 1300–1525, Cambridge: Cambridge University Press. Chen, M.R. (n.d.) Quannan Zazhi, in Cao Rong (ed.) Xuehai Leibian, Liuan (1831) 18. Dai, J. (ed.) (n.d.) Huang Ming Tiaofa Shirei Zuan, reprint, Tokyo: Koten Kenkyukai (1966). Ge, S.L. (1582) ‘Guangzhu Zhiqian Zuyong Shu’, in Ge Duansu Gong Ji, Jinan (1791) 3. Glassman, D. and Redish, A. (1988) ‘Currency depreciation in early modern England and France’, Explorations in Economic History 25, pp. 75–97. Hamaguchi, H. (1969) ‘Mindai no Beika Hyozi to Gin no Ryutsu’, Niigata Chuo Koto Gakko Nenpo 15, pp. 17–31. Hicks, J. (1969) A Theory of Economic History, Oxford: Clarendon Press. Hogendorn, J. and Johnson, M. (1986) The Shell Money of the Slave Trade, Cambridge: Cambridge University Press. King, F.H.H. (1965) Money and Monetary Policy in China 1845–1895, Cambridge, Mass.: Harvard University Press. Kishimoto, M. (1997) Shindai Chugoku no Bukka to Keizai Hendo, Tokyo: Kenbun. Kuksa P’honch’an Wiwonhoe (ed.) (1982) Choson Wangjo Sillok, Seoul. Kuroda, A. (1994) Chuka Teikoku no Kozo to Sekai Keizai, Nagoya: Nagoya Daigaku Shuppankai. —(1998) ‘16–17 Seiki Kan-Shinakai Keizai to Senka Ryutsu ‘, Rekisigaku Kenkyu 711, pp. 2–15. Li T. (1183) Xu Zizhi Tongjian Changbian, reprint, Beijing: Zhonghua Shuju (1986). Maeda, N. (1973) Gencho Shi no Kenkyu, Tokyo: Tokyo Daigaku Shuppankai. Mahapatra, P.R. (1969–70) ‘Currency System in Medieval Orissa’, Quarterly Review of Historical Studies, 9, 2, pp. 72–80.

198

Akinobu Kuroda

Meng, Y.L. (1147) ‘Dushi Qianbo’ in Tongjing Menghua Lu, reprint, Beijing: Zhongguo Shangye Chubanshe, (1982) p. 23. Menger, K. (1892) ‘On the Origin of Money’, Economic Journal 2, pp. 239–55. Miyazawa, T. (1998) Sodai Chugoku no Kokka to Keizai, Tokyo: Kyuko Shoin. Motomura, A. (1994) ‘The best and worst of currencies: seigniorage and currency policy in Spain, 1597–1650’, Journal of Economic History, 54, 1, pp. 104–27. Pamuk, S. (1997) ‘In the absence of domestic currency: debased European coinage in the seventeenth-century Ottoman Empire’, Journal of Economic History, 57, 2, pp. 345– 66. Peng, X.W. (1958) Zhongguo Huobi Shi, second edn., Shanghai: Shanghai Renmin Chubanshe. Polanyi, K. (1977) The Livelihood of Man, New York, Academic Press. Spufford, P. (1988) Money and its Use in Medieval Europe, Cambridge: Cambridge University Press. Vogel, H.U. (1993) ‘Cowry trade and its role in the economy of Yünnan: From the ninth to the mid-seventeenth century’, Part I, II, Journal of the Economic and Social History of the Orient, 36–3, pp. 211–52, 36–4, pp. 309–53. von Glahn, R. (1996) Fountain of Fortune: Money and Monetary Policy in China 1000– 1700, Berkeley: University of California Press. Weber, M. (1964) The Theory of Social and Economic Organization, trans. A.M.Henderson and T.Parsons, New York: Free Press. Yang, C. (n.d.) ‘Yu Tan Erhua Dasima Shu’ in Chen Zilong (ed.) Huang Ming Jingshi Wenbian, reprint, Beijing: Zhonghua Shuju, (1987) 4, pp. 3898–99. Yang, S. (1125) ‘Lun Shishi’, in Guishan Ji 4, in Wenyange Siku Quanshu, reprint, Taibei: Shangwu Yinshuguan (1986) 1125, pp. 128–134. Zheng, Y.K. (1991) Zhonguo Jindai Duiwai Jingji Guanxi Yanjiu, Shanghai: Shanghai Shehui Kexueyuan Chubanshe.

10 Money and growth without development1 The case of Ming China Dennis O.Flynn and Arturo Giráldez

Introduction Silver played a central role in the emergence of trade at the global level.2 Beginning in the sixteenth century, hundreds of tons of silver were extracted each year from mines in Japan, Mexico, and Peru. The bulk of this silver ultimately found its way to China. Silver gravitated ineluctably to the world’s largest economy because the price of silver was higher in China than elsewhere in the world. Why was China’s silver price dramatically higher than in any other place in the world? Because Ming monetary and fiscal systems had been converting to a silver standard in the aftermath of the collapse of China’s paper-money regime in the fifteenth century. China’s advanced economy, comprising a quarter of the world’s population, converted to a silver foundation over an extended time period; as a result, demandside forces raised the price of silver within China to double its value in the rest of the world. Tens of thousands of tons of silver eventually gravitated from the world’s major silver mines in Spanish America and Japan to China—via Nagasaki, Taiwan, India, Indochina, the Atlantic, Cape Horn, Acapulco-Manila and the Pacific, the Mediterranean, the Red Sea, the Baltic, Russia, the Ottoman Empire, the Silk Road, and other routes—because private and public entrepreneurs throughout the world profited by participating in the shipment of silver to its most lucrative end-market. …in the Far East, silver was valued much more highly than gold in comparison with western Europe, so that the western merchant had everything to gain from paying for his purchases in the east in silver. (van der Wee 1977, p. 297) Indeed, merchants the world over shipped silver toward China because it was profitable to do so; this was equally true of (domestic and foreign) merchants involved in the transshipment of silver in Africa, America, Asia, and Europe. As just summarized, the silver side of the birth-of-world-trade equation is now relatively clear. Organization of Chinese exports for exchange against foreign silver, on the other hand, is a more complex matter. Indeed, the relationship between goods exported from China and Chinese silver imports will merit careful consideration later in this essay. 199

200

Dennis O.Flynn and Arturo Giráldez

Perhaps the greatest single beneficiary of the immense Chinese appetite for silver was imperial Spain because her American colonies contained the world’s richest silver mines.3 Profitable production for any product typically depends upon a robust end-market, of course, and China was by far the world’s most dynamic end-market for silver during the period under consideration. Chinese silver demand determined the fate of the Spanish Empire because strong end-market demand was required to maintain profitable mining economies in Spanish America. Since the Spanish Empire was financed by direct and indirect profits based upon New World mines, the ‘silverization’ of China had a profound impact on European and world history. A second notable beneficiary of Chinese demand for silver was the Tokugawa Shogunate; silver-mine profits played a key role in consolidating and unifying Japan during the late sixteenth and early seventeenth centuries (Flynn 1991). In both hemispheres, substantial economic rewards also awaited entities with access to or— better yet—control of trade routes linking silver mines to end-market China. Portugal’s Estado da India, for example, reaped enormous profits by shipping vast quantities of silver to Asia, as did the Dutch and English East India Companies. For the purposes of this essay, suffice to say that innumerable large and small enterprises on all five continents profited directly and indirectly via participation in a Sinocentred silver trade that enveloped markets throughout the world. The purpose of this essay is to investigate whether the multi-century influx of silver into Ming China was beneficial or detrimental to the long-run economic development of end-market China itself. Some influential silver-as-benefit arguments are discussed briefly in the next section, along with select counter-arguments that contend that silver imports were detrimental to China’s long-run development. Our argument belongs to the silver-as-detriment camp: we conclude that the multi-century importation of tens of thousands of tons of silver into China involved an immense system-wide drain on China’s wealth. However, we deny neither that significant Chinese economic growth occurred in response to the massive multi-century influx of silver from abroad, nor that silver imports prompted significant market-oriented restructuring of China’s domestic economy. Still, we maintain that overall Chinese development was impeded during the process. In other words, economic growth is a necessary condition for economic development, but growth alone is insufficient to guarantee development. Silver imports contributed to economic growth in China, but economic development was simultaneously undermined.4 Before delving into the underlying logic of our contention that China’s silver imports simultaneously contained pro-growth and anti-development facets, we wish to specify boundaries of our discussion. First, we are aware that uncoined silver only partially supplanted bronze coin as the monetary standard in China. Von Glahn (1996a, p. 8) depicts the Ming-Qing monetary system as ‘parallel bimetallism’ because uncoined silver and bronze coin operated side-by-side. Thus, when we argue that silver imports ultimately impeded Chinese development, we are at the same time cognisant that bronze monies played a prominent role throughout the late Ming period. Second, while China’s failure to industrialize during Europe’s Industrial Revolution is both a classic and recently rejuvenated debate of great importance, we make no claim to connect our silver-focused arguments to debates

Money and growth without developmen

201

about China’s failure to industrialize during subsequent periods.5 This essay focuses solely on linkages between the deleterious effect of silver imports on China’s longrun economic development, nothing more and nothing less.6 Pro-development versus anti-development interpretations of silver imports Silk was the dream commodity of the project-makers; at the beginning of the seventeenth century silk was for Asian trade what gold and silver had been for the conquistadors. (Steensgaard 1974, p. 367)

Sixteenth- and seventeenth-century Chinese policy prescriptions that favoured unfettered importation of foreign silver into China, as well as those opposed, were remarkably similar to familiar free-trade versus mercantilist arguments found in the contemporaneous European literature. In 1630, Fujian-born He Qiaoyuan emphasized that silver acquired through foreign trade necessarily implied huge Chinese exports, thereby ensuring ‘employment for weavers, potters, and merchants, whose waxing affluence augured higher standards of living for all’ (quoted in von Glahn 1996, p. 127). And there is certainly an element of truth in He’s logic. Incontrovertible evidence of millions of pesos worth of annual foreign silver imports into China does indeed imply millions of pesos worth of Chinese exports. Prodigious Chinese silk exports were essentially bartered directly for Japanese silver, for instance, and the same pattern obtained in the case of the vast Acapulco-Manila silk-for-silver trade across the Pacific. Moreover, world silver traversed countless Afro-Euro-Asian trade routes linking Atlantic silver with the Chinese marketplace.7 It is true that large quantities of porcelain, gold and other items were exported from late Ming China in exchange for silver,8 but silk was by far China’s dominant export.9 Ironically, focus on Chinese silk exports requires de-emphasis of direct silk-for-silver linkages when speaking of the Atlantic-European leg of the global silver trade; Europe’s main source of silk was Persia10 (Chaudhuri 1965, pp. 203– 6; Glamann 1981, ch. 6; Steensgaard 1974, ch. 9). European economies were intimately tied to trade with China just the same, but linkages were much more complex and convoluted than was China’s direct trade with either Japan or America. Estimates of the volume of silk production during the Ming period (1368–1644) are crude at best, but it is known that countrywide production was enormous. Souza (1986, p. 46) cites Bocarro, a seventeenth-century Portuguese historian who estimated that China produced about 2,500 tons of silk annually, 800 tons of which were exported. Most of the exported silk went either to Japan, through Manila to the New World (some of which continued on to European markets),11 or to India. The Manila trade alone carried a sufficient volume of raw Chinese silk such that ‘silk manufacturers in Mexico City, Puebla, and Oaxaca gave work to more than 14,000 people’ (Borah 1943, p. 90) by the late sixteenth century; the Pacific trade boiled down to a straightforward swap of silk for silver (Flynn and Giráldez 1996).

202

Dennis O.Flynn and Arturo Giráldez

Portuguese exports from China—in exchange for Japanese silver—increased eightfold between 1600 and 1637, when silk comprised 66.8 per cent and 89.9 per cent of the total value of the cargo respectively (Souza 1986, pp. 50–1). Technical considerations confined traditional Chinese silk production to smallscale, peasant households. Silk production—from planting mulberry trees, to raising silkworms, and through the formation of cocoons—was exceedingly labour intensive. Notwithstanding the existence of large-scale (and often unprofitable) production facilities at the three imperial silkworks at Hangchow, Nanking, and Soochow—which averaged over 600 looms and 2,000 workers per location in 1685—inexpensive hand-reeling machines guaranteed that Chinese silk weaving remained decentralized (Shih Min-hsiung 1976, pp. 40–1). The fact that both silk production and weaving were based upon a decentralized, peasant foundation throughout the Ming and early Qing periods implies that silk exports were responsible for a huge Chinese employment effect. It was precisely this exportbased employment which prompted He Qiaoyuan’s 1630 free-trade argument cited above: vast silver imports implied Chinese exports on an equally massive scale, which in turn generated jobs and contributed to a fundamental restructuring of China’s domestic economy.12 Scholarly opinion today generally supports the silver-as-benefit arguments of He and others. In a thorough literature search of this topic, von Glahn (1996, p. 142) concludes that: The influx of foreign silver coincided with rapid advances in the commercialization of China’s domestic economy…. By 1550, the population was growing in virtually every region of the empire. Current scholarly opinion tends to view silver imports as the key stimulus to commercialization in the late Ming.13 Atwell (1977, 1982, 1986, 1988a, 1988b) has been perhaps the most persistent modern-day scholar to emphasize the stimulative nature of foreign silver imports into the late Ming economy.14 Andre Gunder Frank (1998, especially ch. 3) provides a distinctive global perspective, including a conclusion that Japanese and American silver stimulated the Chinese economy (while simultaneously failing to stimulate economies in Europe). Predictably, there were influential late Ming Chinese thinkers who challenged the prevailing view that the domestic economy benefited from massive silver imports. The leading statesman of Ming China’s version of the ‘physiocratic school’ was Xu Guangqi (1562–1633), whose anti-silver Omnibus of Husbandry has been summarized in translation by von Glahn (1996a, p. 199): But neither coin nor silver are wealth; rather, they are merely measures of wealth. When the ancient sage-kings spoke of wealth, they meant grain for feeding the people and cloth for clothing them…. The greater the quantity of silver and coin, the more expensive grain and cloth become, and the more severe is the dearth that ensues.

Money and growth without developmen

203

Xu’s reservations about the wisdom of massive silver imports were echoed centuries later when Adam Smith (1776, p. 324) expressed his famous anti-bullionist views: ‘we may imagine the real wealth and revenue of the country to consist in…the value of the annual produce of its land and labour, as plain reason seems to dictate; or in the quantity of the precious metals which circulate within it, as vulgar prejudices suppose’. Either way, according to Smith, one must admit that the importation of precious metals from abroad is costly in terms of real domestic resources surrendered to obtain the metals. If domestic mines were sufficiently rich to satisfy the requirements of a domestic monetary system, according to Smith, then society’s cost of maintaining a silver-based monetary system would be obvious to everyone; society’s cost would simply consist of the value of the land, labour and capital required to produce the domestic silver comprising domestic monies. Real resources would have to be expended on production of monetary substances, resources that would therefore be unavailable for production of non-monetary commodities. The same reasoning holds true when silver mines lie outside of the country, said Smith, except that the monetary system’s cost in terms of land, labour and capital inputs would be reflected in the cost of producing the goods exported in exchange for imported silver. From society’s point of view, the resource cost of the silver money is the same, irrespective of whether the silver is produced domestically or in foreign lands. In the first case, domestic inputs are applied directly to the production of domestic silver, while in the second case domestic inputs are applied to the production of domestic export items, which are then swapped for foreign-produced silver. Domestic resources are expended in the acquisition of the silver in either case. This conceptual foundation underlies Smith’s (1776, p. 324) statement that the ‘food, clothing, and lodging, the revenue and maintenance of all those whose labour or stock is employed in bringing them from the mine to the market, is the price paid for them in Peru as well as in England’. Whether the silver is produced domestically or is imported, society’s cost of acquiring additional silver equals either (1) the direct resource cost of domestic silver production, or (2) the resource cost of the goods exported in exchange for imported silver. Smith therefore rebuked bullionists for failing to recognize this fundamental fact of life: what sense can there be in a national policy which advocates the exportation of scarce consumer items in exchange for non-consumable precious metals from abroad? Due to the diversion of resource expenditures toward silver imports, he reasoned, citizens are able to buy fewer non-monetary goods and services. Thus, the accumulation of silver imports within a country is bound to lower the economic quality of life domestically. Money-and-growth models: paper monies and economic development Explicit integration of money into growth models occurred a generation ago. Money-and-growth models were primarily concerned with the effects of monetary expansion on capital intensity15—and thus economic growth through capital intensification—but the logic of these models is consistent with the anti-bullionist

204

Dennis O.Flynn and Arturo Giráldez

sentiments of Adam Smith. Compatibility between money-and-growth and Smithian views is perhaps most obvious when viewing nineteenth-century transformations from gold standards to paper-money systems.16 According to money-and-growth reasoning, the nineteenth-century replacement of commodity monies by paper monies should, other things being equal, enhance economic development in paperissuing countries. Conversion of a hypothetical, nineteenth-century country from a gold-coin standard to a paper-money standard illustrates the point. Under a gold-coin standard, the stereotypical nineteenth-century country would need to devote substantial resources to the production of gold for use as money. If this country were devoid of domestic gold mines, however, significant resources would nonetheless have to be devoted to the production of items produced domestically, but then exported in exchange for foreign-produced gold. In either case—whether gold is produced domestically, or other domestic products are exchanged for foreign gold instead—a fraction of the country’s resource base would have to be allocated for production of the gold which comprised the intrinsic content of the country’s gold coin. Substitution of a paper-money regime in place of such a gold-money regime would avoid expenditure of resources previously devoted to gold production. Since paper money is relatively costless to produce—compared with resource-intensive production of gold coins—the paper-money regime frees up domestic resources for usage in the non-monetary sector of the economy. By the nineteenth century, commodity money was almost exclusively limited to metals like silver and gold…. Because money had intrinsic value, there was no need for the government to guarantee its value, and the quantity of money was regulated by the market through the supply and demand for gold and silver. But metallic money has shortcomings because scarce resources are required to dig it out of the ground. (Samuelson and Nordhaus 1995, p. 480) It is through this resource-freeing mechanism that standards of living would rise in countries that issue paper monies in substitution for commodity monies; resources previously tied up in gold production could be redeployed to non-monetary sectors of the economy. It is the freeing up of resources previously bound to the goldmoney regime that facilitates augmented domestic development. A benefit of papermoney systems, then, is that they mimic the creation of new resources. Existing resources are simply freed up, of course, rather than being newly created, but the positive effect in terms of domestic development is much the same. Money-and-growth logic for late Ming China It is instructive to view the ‘silverization’ of late Ming China in terms of the moneyand-growth logic just outlined. The nineteenth-century case (where there was a transition toward paper money and away from gold money) is the reverse of the sequence of events that concern us here—because in late Ming China a silver-

Money and growth without developmen

205

money regime replaced a paper-money predecessor—but the same logic applies. China’s paper-money regime had been around since at least the eleventh century, half a millennium prior to similar developments within Europe (von Glahn 1996a, 1996b). Finding itself entangled in expensive military confrontations, a fiscally strapped Ming government succumbed to the temptation to overproduce paper monies. Predictably, the excessive issue of paper money caused its market value to plummet, which in turn forced the Chinese government to abandon its long-standing commitment to a fixed exchange rate between its paper money and silver. This abandonment of redeemability, in terms of a fixed quantity of silver guaranteeing the value of the paper money, reinforced decline in the purchasing power of Ming paper monies. Hyperinflation ensued. In need of a stable monetary regime, Chinese merchants chose to base transactions on a silver standard (instead of worthless paper money); this trend was especially pronounced initially along the southeast coast of China. Other regions followed. The white metal eventually ‘conquered’ China’s economy and became a de facto monetary standard of the realm. Although resistant to silver’s intrusion for generations, the Ming Dynasty ultimately abandoned opposition to silver as a monetary standard. By the 1570s its empirewide tax system had also switched to silver with implementation of the so-called ‘Single-Whip Tax Reform’. Thus, one-quarter of the world’s population (not to mention residents of neighbouring tributary states) had largely converted both monetary and fiscal systems to a silver regime during the sixteenth century. The resultant demand-side surge had an enormous impact on the world silver market, as previously mentioned, because the price of silver within China consequently rose to double the metal’s price in the rest of the world (Geiss 1979, p. 165). It was for this reason that silver flowed from markets around the world; profitmaximization involved shipment of the white metal toward markets offering the highest price (ultimately, those in China). The eventual purchase of tens of thousands of tons of silver within Chinese market places required prodigious Chinese exports. Silk was by far China’s dominant export commodity, comprising up to 90 per cent of the total value of China’s exports to Japan—in exchange for Japanese silver—in the mid-1630s (Souza 1986, pp. 50–1). Silk comprised a similar (and probably higher) proportion of Chinese exports heading for the New World on board the Acapulco-Manila galleons—in exchange for Spanish-American silver—between the 1570s and 1700 (Flynn and Giráldez 1996). This contention is fortified by the research of Borah (1954, p. 123), who claimed that 90 per cent or more of the manufactured goods exported from Mexico to Peru in the mid-1630s were Chinese silks imported via Manila. Application of money-and-growth logic reveals that importation of silver implied a huge resource drain for the late Ming economy as a whole. Resources devoted to the export of silk represent resources needed to facilitate the transition from paperbased to silver-based monetary and fiscal regimes. Resources previously available to the non-monetary sector under China’s earlier paper-money system, in other words, became tied up by monetary and fiscal sectors now resting on a silver foundation. Silverization of the Chinese economy required that resources previously

206

Dennis O.Flynn and Arturo Giráldez

available for production of myriad non-monetary industries be subsequently diverted to a monetary/fiscal sector now requiring a vast stock of (resource-hungry) silver. The profit motive that prompted merchants from around the world to ship silver toward China has already been emphasized. Exports from China represent the flip side of this market process, however, and there is no reason to doubt that Chinese silk exporters must have enjoyed profits comparable to the legendary gains attributed to those involved in the importation of silver. The statement that silver fetched many times as much silk inside China as outside China—which is the way things looked from American, European, and Japanese perspectives—also implies that Chinese silk fetched multiple times as much silver abroad as within the domestic Chinese marketplace. Otherwise, there would not have been such explosive growth of China’s silk industry. Thus, tremendous growth in China’s silk industry (itself intimately intertwined with silver imports) did stimulate overall growth, in the sense of enhanced gross domestic product, of the late Ming economy. This pro-growth role for silver imports is compatible with our claim that silver imports concurrently impeded Chinese economic development. Economic growth and economic development are not equivalent concepts, notwithstanding an abundance of present-day analysts who treat the terms as synonyms. The GDP of late Ming China may indeed have grown impressively overall, in the sense that expanding resources may have been utilized efficiently to raise output, but a significant portion of employed resources were tied up in the exportation of silk products, the function of which was to attract foreign silver. Since silver was itself ‘sterile’—in the sense that resource embodied within the metal had been diverted away from use in non-silver sectors of the Chinese economy—fewer resources were available for capital accumulation or consumption by either the private or public sectors. In other words, although evidence points toward a secular rise in late Ming gross domestic product, late Ming gross domestic wealth (especially excluding the portion tied up in silver) may have simultaneously declined. Even if some future economic historian were to argue that late Ming gross domestic wealth rose—a proposition we doubt could be substantiated—such evidence would still fail to contradict our contention. We simply argue that economic development was weaker under the silver regime than would have been the case had China maintained a paper-money regime. Could a private-market response be detrimental to development? This essay’s main contention—that the importation of silver impeded the overall economic development of late Ming China—introduces dissonance in terms of mainstream economic logic.17 The silverization of China emanated from private (rather than government) initiative. Since the economics profession generally views private-sector responses to exogenous shocks as solutions, as opposed to problems in their own right, it is suspicious for an economist to claim that a private-sector response could have exacerbated a major social problem.18 How then could it be that the silverization of China, a spontaneous and unplanned free-market response to government mismanagement of a paper-money regime, was associated with

Money and growth without developmen

207

anti-development forces that persisted for centuries? Economist colleagues claim that, contrary to our unflattering money-and-growth view, the silverization of China should be viewed as a private-market solution to the government mismanagement involved in the initial collapse of China’s fiduciary-money system. The multicentury free-market response to this fifteenth-century collapse, in the form of the silverization of China, is viewed by them as offering a market solution to a preexisting problem, rather than creating a costly problem in its own right. According to this line of reasoning, the multi-century importation of silver into China may indeed have been coterminous with a massive loss of social wealth (as our moneyand-growth argument suggests). Yet, this loss should be attributed to the governmentcaused collapse of China’s paper-money regime during the middle of the fifteenth century. Destruction of a functioning paper-money system was an economic tragedy of epic proportions, to be sure, but the tragedy occurred in the fifteenth century; it was at that time that Chinese society suffered its loss. The private-sector response, in the form of spreading ‘silverization’ during the late Ming period, should be viewed as a beneficial solution to a pre-existing problem. In this way, massive imports of silver involved social gain, rather than the social loss for which we have argued. This counter-argument can be summarized as follows: (1) a fifteenth-century loss of wealth occurred with collapse of the paper-money regime, while (2) the subsequent centuries-long importation of silver was a private-sector response that enabled late Ming China to partially offset the fifteenth-century social loss. In this way, silver imports benefited China in terms both of the country’s gross domestic product as well as in terms of (what we refer to as) its gross national wealth. Spreading costs over the centuries We believe that the pro-market criticism just outlined is valid as far as it goes. It is important to separate two issues: (1) the existence of social losses stemming from China’s monetary collapse, versus (2) the time period to which social losses should be assigned. We agree that the fifteenth-century collapse of China’s paper-money regime was the root social calamity leading to the social loss under discussion; the subsequent importation of silver from around the world was indeed merely the market’s response to this earlier mid-fifteenth-century disaster. From this vantage, the importation of silver appears beneficial to late Ming society because silver imports contributed to stabilization of China’s monetary system over the centuries (a requisite ingredient for a flourishing market economy). Having acknowledged this point, however, we are nevertheless hesitant to assign all social costs associated with China’s subsequent importation of silver back to the fifteenth-century period during which the Ming paper-money regime collapsed. True, money-and-growth logic does lead to the conclusion that the original implementation of China’s papermoney regime must have generated a positive social wealth effect. Thus, logical symmetry requires us to conclude that the fifteenth-century collapse of this same paper-money regime must have reversed the initial positive wealth effect from creation of the paper-money system. In this way, the fifteenth-century social cost of China’s monetary collapse should be viewed as our economist colleagues say.

208

Dennis O.Flynn and Arturo Giráldez

Yet we want to be careful to distinguish between (1) the (substantial) mid-fifteenthcentury social costs associated with the collapse of the paper-money regime, and (2) additional social costs associated with subsequent, post-fifteenth century growth of the late Ming economy. We are sceptical about attributing all (post-fifteenthcentury) social costs back to events which occurred in the mid-fifteenth century. Otherwise, one would also be forced to attribute the social costs of importing silver into China during the eighteenth, nineteenth, and twentieth centuries back to the mid-fifteenth-century collapse of the paper-money regime! We believe that such a procedure is questionable. Perhaps it makes more sense to distribute the social costs of silver importation over the centuries during which they occurred; each period could be assigned social costs of silverization to the extent that each period required additional silver imports. According to this procedure, huge social costs would still be attributed to the initial collapse of China’s paper-money regime in the fifteenth century. But additional social costs, not attributable to the period of the mid-fifteenth-century paper collapse, were subsequently generated because rapid demographic and economic expansion between the mid-fifteenth and midseventeenth centuries necessitated the importation of massive quantities of new foreign silver. The social cost of the additional silver imported during this twocentury interval could be assigned to the centuries during which the silver regime was expanding. The central tenet of our argument holds: late Ming importation of silver simultaneously implied robust economic growth and stunted economic development. Chinese society no doubt suffered a severe developmental setback during the mid-fifteenth-century collapse of its paper-money regime, but the antidevelopment force thereby set in motion was exacerbated by expansion of the silver regime during succeeding centuries. The initial shock due to government mismanagement of its paper-money regime implied social costs which reverberated across centuries as market-driven silverization became increasingly institutionalized throughout an expanding economy. Staggering Chinese resource costs can be assigned both to the original mid-fifteenth-century monetary collapse, as well as to the subsequent periods of silverization. Whether government failure is assigned all, as opposed to most, of the blame is not our primary concern. What we do find interesting is the process through which rational market mechanisms, operating over several centuries, dovetailed with a notable failure in monetary policy in a way that must have stunted Chinese development over time. Summary and conclusion Scholars have long speculated that increased money supplies, caused by an unprecedented eruption of precious metals from the New World, promoted early European capitalism during the sixteenth and seventeenth centuries. Mainstream economists like Earl J.Hamilton and John Maynard Keynes argued that importation of American metals led to both price inflation and ‘profit inflation’, which in turn stimulated early capitalism:

Money and growth without developmen

209

Never in the annals of the modern world has there existed so prolonged and so rich an opportunity for the business man, the speculator and the profiteer. In these golden years modern capitalism was born. (Keynes 1930, Vol. II, p. 159) Subsequent historical research discredited the profit-inflation thesis, and it had to be abandoned.19 Half a century later, neo-Marxist Immanuel Wallerstein revived the notion that New World bullion was a critical ingredient in the emergence of European capitalism. He argued that economic surplus was somehow attached to American silver as it was transshipped through Spain—a country characterized by Wallerstein as ‘semi-peripheral’ and a ‘passive conveyer belt’ of surplus value—and onto Northwest Europe. And hence without it [bullion], Europe would have lacked the collective confidence to develop a capitalist system, wherein profit is based on various deferrals of realized value…bullion must be seen as an essential crop for a prospering [European] world-economy. (Wallerstein 1974, p. 46) Wallerstein evidently thinks that American silver accumulated within the capitalistic core of Europe, a crucial point for him because he clearly believes that surplus value was somehow attached to the metal itself.20 Much of the silver, specifically in monetized form, had to remain within Europe; otherwise, Wallerstein’s conceptual treatment of silver collapses: The production of gold and silver as a commodity made the Americas a peripheral area of the European world-economy insofar as this commodity was essential to the operation of this world economy, and it was essential to the extent that it was used as money. Had the bullion of the Americas all flowed out to Asia, the Americas would have been just another external arena and Europe would have been merely an axis of three arenas—America, Europe, and Asia—obtaining its Asian luxuries at the price of goods sent to the Americas. (Wallerstein 1980, p. 109) Wallerstein’s words from 1980 were prescient because we now know that most of the American silver did in fact flow to China and (to a much lesser extent) to India (Flynn and Giráldez 1996). (Wallerstein never discussed Japanese silver.) As stated earlier, the massive flow of silver to China occurred because of tremendous demandside growth within China, but also because of reduced European monetary demand for silver. The seventeenth century is referred to as the ‘copper century’ in European monetary history precisely because copper monies (and other devices) replaced silver monies. On the whole, Europe no doubt contained less monetized silver at the end of the Price Revolution than at its outset.

210

Dennis O.Flynn and Arturo Giráldez

Following the Wallersteinian paradigm, confident capitalism should have emerged in China—the final resting place of Japanese and American silver—rather than in Europe, which was merely an area of intermediary transshipment. In a sense, such a claim of stimulation of China’s economy via silver importation is what scholars such as Atwell and Frank argue (although through mechanisms different from those imagined by Wallerstein). So on the one hand, theories of Chinese market place stimulation via the importation of silver seem themselves as if they could have been imported from European historiography. On the other hand, beneficial impacts from silver imports hark back to the views of seventeenthcentury Chinese thinkers who directly observed the phenomenon. The prevailing view then was that silver imports stimulated massive Chinese exports, which in turn bolstered both national output and employment. This view contained truth when enunciated in the seventeenth century, and it contains truth today. The argument of this essay is consistent with aspects of both silver-as-benefit and silver-as-detriment interpretations of the late Ming period. It supports the silver-as-benefit school in the sense that growth in Chinese output was dramatic, not only in the export sector, but also in domestic sectors of the economy that underwent fundamental restructuring in response to external market pressures (Marks 1997). If late Ming estimates of real GDP were to exist (which they do not), they would probably reveal a growth trend. But our argument also supports the silver-as-detriment school because expanding real GDP does not necessarily imply a wealthier society. Replacement of a paper-fiscal-and-monetary regime by a silver-fiscal-and-monetary regime implied enormous resource costs for Chinese society. As money-and-growth proponent Stein (1970, p. 85) correctly observes, the ‘opportunity cost of producing gold (or export to purchase gold) is output which is no longer available for consumption or investment’. Although domestic resources may have been fully employed in the context of an expanding real GDP, a significant portion of these employed resources was bound up with sustenance of the commodity-money system itself; the switch to this commoditymoney regime reduced available wealth for non-monetary sectors of the society. Economic growth occurred, in other words, but at the cost of reduced economic development. Silver’s overall deleterious impact on China (which had moved away from a paper-money regime) leads to the question of whether Europe (which had partially moved toward paper monies during the same period) may have benefited as a result of the globalization of the silver market. Partial replacement of commoditymoney regimes by paper-money regimes probably did contribute something to expansion of usable wealth within Europe. This is the root of money-and-growth logic. Thus, parts of Europe may have benefited by the replacement of commodity monies with paper substitutes. Yet it does not follow that such benefits are attributable to the global silver market. The freeing of domestic resources within certain parts of Europe occurred because silver was partially supplanted by paper devices. In other words, a lack of silver was a symptom of money-and-growth gains within certain areas of Europe. Paper-money substitution for European silver monies did partially cause silver to vacate European money markets (in favour of

Money and growth without developmen

211

the Chinese market place). The absence of silver was an effect, however, while the issuance of paper-money substitutes was a collateral cause of silver’s flight from Europe. It was the issuance of paper substitutes, not the absence of silver per se, which engendered augmented European wealth. Still, we should be careful not to exaggerate European benefits, given that the seventeenth century is termed Europe’s ‘copper century’ in monetary history, and copper monies were commodity monies (just like silver). While the fifteenth-century collapse of its paper-money regime engendered huge social costs at the time of that event, we find interesting the notion that social costs to China continued to mount during subsequent centuries of private-sector response to the initial mishap. Once silverized, China’s vast and expanding economy required growing imports of silver from abroad. Silk products were the main export category for centuries, while tea finally became China’s most important export item during the eighteenth century. To the extent that silver imports mandated huge exports of tea during the eighteenth century, it seems that a centuries-old monetary catastrophe was a seminal event indeed. It propagated mischief right into the twentieth century, a long-term anti-development Chinese legacy that money-and-growth reasoning allows us to glimpse. We should emphasize in closing that silver per se did not engender an antidevelopment bias. Had China adopted a gold, copper, cowry, or other commodity money standard in place of its collapsed paper-money regime, deleterious welfare impacts on China would have been much the same. Chinese exports would have simply purchased the monetary commodity that substituted for silver. The volume of foregone resources—those used to produce requisite exports—would probably have been much the same. World trade history would have deviated dramatically from the actual history we seek to understand, to be sure, but we intentionally refrain from counterfactual speculations about potential trade-history patterns that might have resulted. China in fact ‘silverized’. Unplanned consequences were profound, both domestically and for the rest of the world. Notes 1 2

3 4

Criticisms of an earlier version of this paper by Ken Pomeranz are gratefully acknowledged, although the authors alone are responsible for views expressed herein. The global context of silver’s role in the birth of global trade is discussed in Flynn and Giráldez (1995a and 1997) and Frank (1998). See von Glahn (1996a) for a classic account of the demand-side forces within China which attracted silver. For discussion of the importance of China’s silver demand for fiscal support of the Spanish Empire, see Flynn and Giráldez (1996). The phrase ‘economic development’ means different things to different people. Our terminology is no doubt different from that normally found in Chinese historiography. For example, for Huang (1990) ‘growth’ means a rise in total output, while ‘development’ refers to rising output per labour hour. Our concept of ‘growth’ is the same as his, but we use ‘development’ to mean augmented system-wide wealth (an issue that is separable from issues of income). Chinese income may in fact have risen during certain periods of silver importation, but a portion of the increase production was ‘taxed away’ in the form of Chinese exports exchanged for imported silver. We

212

5

6

7 8

9 10

11

Dennis O.Flynn and Arturo Giráldez

place ‘taxed away’ in quotation marks to indicate that these Chinese exports were no longer available to the Chinese populace (thus, it is as if there were an external tax levied on them). Our wealth-versus-income distinction is similar to other situations that might result in the simultaneous rise of pre-tax income while after-tax welfare declines. It is this negative impact on social wealth—from the ‘tax’ arising from a switch in monetary regimes—that we identify with inhibited Chinese development. The economic system may have simultaneously experienced growth in national income (even on a per-capita basis), but per-capita wealth was lower than it would have been in the absence of a collapse of Ming China’s paper-money system. For recent insights into China’s failure to experience a European-style Industrial Revolution, see Wong (1997). In a study focusing on the relationship between trade and domestic restructuring within China’s domestic economy, Marks (1997, pp. 16, 20–1) states that in ‘effect, until the early nineteenth century, China was the industrialized country in the world…. To be sure, Europeans introduced opium into the equation, but basically the linking of China to the European world economy was less a new phenomenon than a continuation of existing patterns set deep within the functioning of the Chinese economy.’ Also, see von Glahn’s cogent comparison of the ‘sprouts of capitalism’ (Chinese perspective) and ‘early modern’ (Western perspective) interpretations of Chinese history. Our argument that Chinese imports of silver reduced the availability of Chinese resources is consistent, however, with the work of Pomeranz (forthcoming), who argues that European access to New World resources rendered possible an Industrial Revolution in Europe. If Pomeranz is correct about Chinese bottlenecks of an ecological nature, then silver imports exacerbated Chinese disadvantages in terms of resource limitations. The Introductory essay in Flynn and Giráldez (1997) makes a case for linking certain economies of Africa into the global silver marketplace. Atwell (1998, p. 396): ‘by 1614, Ming blue and white wares are said to have been “in daily use” among the ordinary citizens of Amsterdam.’ And (pp. 393–4) ‘as early as the 1530s, perhaps 40,000 to 60,000 pieces of Chinese porcelain were arriving in Lisbon from Asia each year. By the 1540s, the Lisbon élite is said to have been wearing Chinese silks, drinking Chinese tea, and placing special orders for Ming porcelain with Portuguese motifs.’ Moreover (p. 399), ‘between the late 16th century and the early 1630s, Japanese imports of raw silk, most of which were from China, rose from an estimated 60,000–90,000 kilograms per year to perhaps as much as 280,000 kilograms.’ All (p. 401) ‘indications are that Ming porcelain was in common use in Peru as well. Shards of Chinese porcelain have been found on the shore of Lake Titicaca some 15,000 feet above sea level.’ While one (p. 402) ‘Spanish observer during the 1630s [document dated 1637] even claimed that because the domestic supplies of raw silk in New Spain were insufficient to meet the demand, trade with China helped to keep 14,000 weavers in Mexico City, Puebla, and other cities employed.’ For a long-term overview of global silk connections, including eras of Chinese dominance in this industry, see Ma (1998). According to Atwell (1998, p. 402), some Chinese silks have been mis-identified as ‘Persian’: ‘Future research may reveal that at least some of the famous “Persian silk” sold on Middle Eastern and European markets during the 16th and 17th centuries was also of Chinese origin. It is known, for example, that the Ottoman Court received “Chinese fabrics” as war booty from Iran during the mid-16th century.’ See de la Fuente, Garcia, and Iglesias (1996, p. 101) for mention of the re-exportation of Chinese goods to Spain by way of Havana, a common pattern.

Money and growth without developmen

12 13

14

15 16

17

18 19

20

213

Detailed linkages between the trade sector and China’s domestic economic restructuring during the Qing dynasty are explored in Marks (1997 and 1998). Von Glahn’s (1996a, p. 142) next sentence, however, states that ‘it is equally plausible that causation ran in the opposite direction: commercial expansion in the domestic economy, by raising the demand for media of exchange, attracted silver from abroad and promoted foreign trade’. Moloughney and Xia (1989, pp. 67–8), Goldstone (1991, pp. 438–40), and von Glahn (1996b) have been highly critical of Atwell’s work, while Flynn and Giráldez (1995, pp. 438–40) maintain that the arguments of these authors are not necessarily incompatible. See Stein (1970) for a review of both neoclassical and Keynes-Wicksell approaches to analysis of the effects of monetary policy for growing economies. Note that paper monies replaced commodity monies in the nineteenth-century case, whereas the opposite occurred in late Ming China, where the commodity money (silver) replaced paper money. The logic is the same in each case, but the former involves social gain while the latter involves social loss. This discussion in this section stems from criticism by University of the Pacific colleagues, Benjamin Dennis and Timothy Opiela, who argue that the social costs discussed in the previous section should be confined to the mid-fifteenth century disintegration of China’s paper-money regime. This event involved a public sector failure. The private-sector ‘solution’ should not be blamed. Except in the case of negative externalities, an issue intentionally ignored here. The Hamilton and Keynes profit-inflation thesis was inconsistent with empirical evidence; industrial prices did not rise faster than wages, so there were no inflated profits with which to launch incipient capitalism. See Felix (1956), Nef (1936) and Outhwaite (1969, p. 40). For a more complete criticism of Wallerstein’s views on silver, see Flynn (1996).

References Atwell, William S. (1977), ‘Notes on silver, foreign trade, and the late Ming economy’, Ch’ing-shih wen-t’i, n. 3, pp. 1–33. —(1982), ‘International bullion flows and the Chinese economy circa 1530–1650’, Past and Present, n. 95, pp. 68–90. —(1986), ‘Some observations on the “seventeenth century crisis” in China and Japan’, Journal of Asian Studies, Vol. XLV, n. 2, pp. 223–44. —(1988a), ‘Ming observations of Ming decline: some Chinese views on the “seventeenthcentury crisis” in comparative perspective’, Journal of the Royal Asiatic Society, n. 2, pp. 316–48. —(1998b), ‘Ming China and the emerging world economy, c. 1470–1650’, in Dennis Twitchett and Frederick W.Mote (eds) The Cambridge History of China, Vol. 7, Cambridge: Cambridge University Press, pp. 585–640. Borah, William (1954) Early Colonial Trade and Navigation between Mexico and Peru, Berkeley and Los Angeles: University of California Press. Borah, Woodrow (1943) Silk raising in Colonial Mexico, Berkeley: University of California Press. Chaudhuri, K.N. (1965) The English East India Company, London: F. Cass. De la Fuente, Alejandro, Garcia del Pino, Cesar and Delgado, Bernardo Iglesias (1996) ‘Havana and the fleet system: Trade and growth in the periphery of the Spanish Empire, 1550–1610’, Colonial Latin American Review, Vol. 5, n. 1, pp. 95–115.

214

Dennis O.Flynn and Arturo Giráldez

Felix, David (1956) ‘Profit inflation and industrial growth: The historic record and contemporary analogies’, Quarterly Journal of Economics, 70 (August), pp. 441–63. Flynn, Dennis O. (1991) ‘Comparing the Tokugawa Shogunate with Hapsburg Spain: Two silver-based empires in a global setting’, in James D.Tracy (ed.) The Political Economy of Merchant Empires, Cambridge: Cambridge University Press, pp. 332–59. —(1996) World Silver and Monetary History in the 16th and 17th Centuries, Aldershot: Variorum. Flynn, Dennis O. and Giráldez, Arturo (1995a), ‘Born with a “silver spoon”: The origin of world trade in 1571’, Journal of World History, Vol. 6, n. 2 (Fall), pp. 201–21. —(1995b) ‘Arbitrage, China and world trade in the early modern period’, Journal of the Economic and Social History of the Orient, Vol. 6, n. 2, pp. 429–48. —(1996) ‘China and the Spanish Empire’, Revista de Historia Economica, XIV, n. 2, pp. 309–38. —(1997) ‘Introduction: monetary substances in global perspective’, in Dennis O.Flynn and Arturo Giráldez (eds) Metals and Monies in an Emerging Global Economy, Aldershot: Variorum pp. xv–xxix. Frank, Andre Gunder (1998) ReOrient: Global Economy in the Asian Age, Berkeley: University of California Press. Geiss, James (1979) ‘Peking under the Ming (1368–1644)’, PhD Dissertation, Princeton University. Glamann, Kristof (1981) Dutch-Asiatic Trade, 1620–1740, The Hague: Martinns Nijhoff. Goldstone, Jack (1991) Revolution and Rebellion in the Early Modern World, Berkeley and Los Angeles: University of California Press. Huang, Philip (1990) The Peasant Family and Economic Development in the Yangzi Delta, 1350–1988, Stanford: Stanford University Press. Keynes, John Maynard (1930) A Treatise on Money, New York: Harcourt, Brace. Ma, Debin (1998) ‘The great silk exchange: how the world was connected and developed’, in Dennis O.Flynn, Lionel Frost and A.J.H.Latham (eds) Pacific Centuries: Pacific and Pacific Rim history since the sixteenth century, London: Routledge, pp. 38–69. Marks, Robert B. (1996) ‘Commercialization without Capitalism: Processes of environmental change in South China, 1550–1850’, Environmental History, Vol. I, n. 1, pp. 56–82. —(1997) Tigers, Rice, Silk and Silt: Environment and Economy in Late Imperial South China, New York and London: Cambridge University Press. Moloughney, Brian and Xia, Weizhong (1989) ‘Silver and the fall of the Ming: a reassessment’, Papers on Far Eastern History, n. 40, pp. 65–6. Nef, John U. (1936) ‘Prices and industrial capitalism in France and England, 1540–1640’, Economic History Review, 7 (November), pp. 155–85. Outhwaite, R.B. (1969) Inflation in Tudor and Early Stuart England, London: Macmillan. Pomeranz, Kenneth (2000) The Great Divergence: China, Europe, and the Making of the Modern World Economy, Princeton: Princeton University Press. Samuelson, P.A. and Nordhaus W.D. (1995) Economics, New York: McGraw-Hill. Shih, Min-hsiung (1976) The Silk Industry in Ch’ing China, Ann Arbor: Center of Chinese Studies, The University of Michigan. Smith, Adam (1776, 1937) The Wealth of Nations, New York: Modern Library. Souza, George B. (1986) The Survival of the Empire: Portuguese Trade and Society in China and the South China, 1630–1754, Cambridge: Cambridge University Press. Steensgaard, Niels (1974) The Asian Trade Revolution of the Seventeenth Century, Chicago: Chicago University Press. Stein, Jerome L. (1970) ‘Monetary growth theory in perspective’, American Economic Review, 60 (March), pp. 85–106. van der Wee, Herman (1977), ‘Chapter V: monetary credit and banking systems’, in E.E.

Money and growth without developmen

215

Rich and C.H.Wilson (eds) The Cambridge Economic History of Europe, Vol. V, The Economic Organization of Early Modern Europe, Cambridge: Cambridge University Press, 1977. von Glahn, Richard (1996a) Fountain of Fortune: Money and Monetary Policy in China 100–1700, Berkeley: University of California Press. —(1996b) ‘Myth and reality of China’s seventeenth-century monetary crisis’, Journal of Economic History, Vol. 56, n. 2, pp. 429–54. Wallerstein, Immanuel (1974) The Modern World-System, Vol. I, Capitalist Agriculture and the Origins of the European World-Economy in the Sixteenth Century, New York: Academic Books. —(1980), The Modern World-System, Vol. 2, Mercantilism and the Consolidation of the European World-Economy, New York: Academic Press. Wong, R.Bin (1997) China Transformed: Historical Change and the Limits of European Experience, Ithaca: Cornell University Press.

11 California and Nevada minerals in the Pacific Rim 1850–1900 David J.St. Clair

Many aspects of the California Gold Rush and Nevada’s Comstock Silver Boom are familiar stories. The world rushed to San Francisco and the gold fields in 1849 and gold poured into the world economy. In 1859, the rush shifted to Virginia City and silver joined the flow of gold through the Golden Gate. The magnitude of California and Nevada precious metals production is generally appreciated and there is no point in retracing this familiar ground. Instead, the goal of this paper is to suggest connections and consequences that have hitherto received too little attention. Three points will be developed. First, while both gold and silver poured out of the American West, they sought different destinations. One metal joined longstanding trade flows across the Pacific. The other took a decidedly non-Pacific track. Both altered world trade and monetary policies. Second, a discussion of important California metals should include quicksilver. The dramatic increases in world gold and silver production during the second half of the nineteenth century were very much influenced by California quicksilver production. Third, gold and silver production in Nevada and California were the catalyst for developing California manufacturing. California’s role as a producer and exporter of manufactured products is often taken as a twentieth-century phenomenon. This view, however, ignores the industry’s nineteenth-century roots in precious metals production. Each of these developments had an impact on California and on the development of the Pacific economy. A full or even adequate treatment of these issues is beyond the scope of a single paper. However, some preliminary assessments will be suggested. Gold and silver: in or out of the Pacific? California produced more than $318 million of gold during the second half of the nineteenth century, and at least another $11 million of silver (Boalich 1914, p. 20, 25). Josiah Whitney estimated that United States gold production in 1854 (virtually all from California) amounted to 41.5 per cent of world output (Whitney 1854, p. 506). Together, California and Australia, the other major mid-century gold discovery, accounted for 72.6 per cent of world output in 1854, dwarfing traditional producers. 216

California and Nevada Minerals

217

Most California gold quickly left the state.1 More than 90 per cent of gold produced in California from 1854 through 1859 was shipped out of San Francisco. Close to three-quarters of the gold and silver exported from California from 1854 through 1863 was destined for eastern US ports. England received 19 per cent, while China took only 5.4 per cent. China and Pacific markets did not receive much of the treasure flow from California so long as gold was the principal treasure export. This changed with the Comstock Silver Boom. Nevada produced $335.7 million of silver and gold from 1859–1900 (Smith 1943, pp. 291–7). Most of this silver was exported from San Francisco; however, its course differed markedly from that of gold. Silver exports were overwhelmingly destined for Asian markets, especially China. From 1882 through 1898, exports to four countries—China, Japan, India, and Hawaii—accounted for 97.4 per cent of California sea treasure exports (St. Clair 2000). During this period, silver made up 99.6 per cent of sea treasure exports. Consequently, it is clear that these four countries received almost all California treasure exports, and that these exports were virtually all silver. It is interesting to note that more than half of the silver exported from San Francisco during this period was made up of Mexican silver dollars. This is a reflection of the important role that Mexican silver dollars played in Chinese commerce. To increase the flow of Comstock silver in the China trade, the United States began coining the American Trade Dollar in 1873 (St. Clair 1998b). From 1873 through 1878, Trade Dollars were coined in quantity and Nevada silver largely displaced Mexican silver dollars in American exports. However, in 1878, American policy changed, abandoning the Trade Dollar in favour of the re-monetization of silver and silver coinage for domestic circulation. Mexican silver dollars once again took a significant share of American treasure exports. The figure cited above is for the period after the termination of Trade Dollar coinage. Despite the advantages enjoyed by Mexican silver dollars, Comstock silver continued to flow into Pacific markets and US silver exports increased through the century. Silver flowing across the Pacific had been a long-established feature of world trade for almost three centuries before 1859.2 Nevada silver joined this flow, whereas California gold did not. Both metals, however, were part of the surge in world precious metals production that took place in the second half of the nineteenth century (see below). World liquidity and trade, within the context of the very different destinations of the two precious metals, increased. Contemporary writers were aware of the divergent paths of these metals. They were also aware of how increased gold production augmented silver flows to Asia and led to silver’s commoditization (and eventual de-monetization). In 1864, the US Treasury Department summarized this as follows: The absorption of silver in Asia has never been so great as since the gold discoveries of California and Australia. With the increase of bullion Europe ceased to regard with apprehension the Oriental demand for silver in exchange for silks, teas, indigo, and other staples of eastern production. When it was known that the Pacific gold stream was yearly increasing in volume, and could readily fill any vacuum which the

218

David J.St.Clair

shipment of silver to India and China might produce, a great expansion of trade to Asia followed. The precious metals came to be regarded as merchandise, and it seemed wholly unessential whether payment was made for eastern products in the coin or the manufactures of Europe. (US Treasury 1864, p. 198) Viewed in this way, both gold and silver flows from California and Nevada were integral parts of both the increasing flow of silver to Asia and increasing trade with that region. Increasing gold and silver production alleviated Western fears of money being drained off to Asia. Silver was essentially converted to a commodity, while gold becomes the sole monetary metal in Europe and in the United States. This transition was neither immediate nor smooth, but it is a fair characterization of the process. Viewed in this way, it is not primarily the differential quantities of silver versus gold produced during this period of time that was important for silver demonetization. Rather, it was the increasing quantities of both metals that permitted the flow of silver to Asia to proceed unimpeded. Quicksilver A third California metal should be added to this scenario. California mercury, or quicksilver as it was more commonly called in the nineteenth century, played an integral part in increasing the production of both gold and silver in the second half of the nineteenth century. While quicksilver was used in the manufacture of many products such as vermilion, inks, china, pottery, textiles, explosives, mirrors, and thermometers, its most important application was in the extraction of gold and silver from ores. Mercury has an affinity for gold and silver, and will form an amalgam with these metals on contact. The great outflow of silver from Spanish America that began in the sixteenth century was made possible by the invention of the Patio Process of mercury amalgamation in Mexico in 1554. The Washoe Process, invented in the 1860s to treat Comstock silver ores, improved upon the Patio Process (Smith 1943, pp. 41–5). Until the invention of cyanide processing in 1890, mercury amalgamation was the primary, often the sole means of extracting silver and gold from low-grade ores. Workable mercury deposits have been rare (Young 1970, p. 92). Throughout history, only a handful of mines have produced the bulk of the world’s supply of quicksilver. The Almaden Mine in Spain has dominated world quicksilver production for over 2,000 years. Colonial silver mines in Mexico and Peru relied on quicksilver from Almaden, supplemented with mercury from a rich quicksilver mine discovered in 1563 at Huancavelica, Peru. From the mid-sixteenth century to 1850, Almaden produced 50.7 per cent of world output, while Huancavelica produced 26.9 per cent. Mines at Idria, Slovenia, supplied 22.4 per cent (St. Clair 1998a, p. 211). Almaden and Huancavelica were usually able to meet the demands of the Mexican and Peruvian silver mines, but the threat of quicksilver shortages and high quicksilver prices were a constant concern (Bakewell 1971, pp. 150–80). A mercury shortage halted silver production at Potosi in 1715 (Tandeter 1993, p. 13).

California and Nevada Minerals

219

Two significant developments affected the supply of mercury prior to 1850. First, significant production at Huancavelica had ceased by 1800. Second, by 1835, the two remaining quicksilver mines at Almaden and Idria had come under the control of a cartel organized by the Rothschilds. Until the advent of California production, the Rothschild cartel enjoyed a virtual monopoly in quicksilver. Quicksilver was discovered at New Almaden, near San Jose, California, in 1845.3 Sporadic production at New Almaden occurred between 1846 and 1848, but full production did not begin until 1850. New Almaden output quickly surpassed both Almaden and Idria, and its success lead to the discovery of more than 40 quicksilver mines along the California coast. However, no more than seven of these ever produced significant quantities of quicksilver. Quicksilver quickly became one of the state’s largest mining industries. Until the end of the nineteenth century, quicksilver was second only to gold in value of mineral output (California Miner’s Association 1899, p. 430). The value of California quicksilver produced in the nineteenth century was about a quarter of the value of all gold and silver mined from Nevada’s Comstock Lode. The New Almaden quicksilver mine became one of the richest mines (of any type) in California history, producing profits that rivalled the profits from all the Comstock mines combined (St. Clair 1998a p. 210). California quicksilver production is shown in Table 11.1, while Almaden and Idria quicksilver output are shown in Table 11.2. From 1850 to 1900, California mines produced half of the world’s quicksilver. California mines were thus able to meet the rapidly increasing demand for quicksilver stemming from gold and silver discoveries in California, Nevada, other western states, and abroad. In the late 1870s, California mines were producing two-thirds of the world’s output of quicksilver and putting pressure on the Rothschilds cartel. The cartel was unable to maintain price discipline, leading to vigorous competition for world markets in early 1880s (St. Clair 1998b, p. 221). California quicksilver exports are also shown in Table 11.1. During the period 1852–90, 57.1 per cent of California quicksilver was exported, with exports usually exceeding domestic consumption. Exports were greatest during the first two decades, the period that includes both the California Gold Rush and the Comstock Silver Boom. Exports took 67.5 per cent of output in the 1850s and 72.0 per cent in the 1860s. Exports in the 1870s took 51.0 per cent of output, with exports lower in the early part of the decade and higher in the second half. During 1870–74, exports only took 34.9 per cent of output. Overseas shipments recovered in 1875 and remained substantial through 1883. Exports in the 1880s amounted to 45.3 per cent of output, but lower exports during this decade were due to the dramatic decline in exports after 1883. Exports in the period 1880–83 were 61.9 per cent, but fell to half this rate thereafter. The dramatic reduction in overseas shipments that occurred after 1883 followed record California production in the 1870s that precipitated a price war with the Rothschild cartel, especially in the China market. Table 11.3 shows the steep drop in the share of California exports to China in 1884. California producers lost most of the China market that year and never recovered it, beginning a long period of distress for California quicksilver producers.

220

David J.St.Clair

Table 11.1 California quicksilver production and exports 1850–90 (in 76.5-pound Flasks)

Source: St. Clair 1998a, p. 220. *New Almaden Quicksilver Mine closed by court injunction.

California and Nevada Minerals

221

Table 11.2 Annual quicksilver production from Almaden and Idria (in 76.07-pound flasks*)

Source: St. Clair 1998a, p. 213. *Flasks reported here are smaller than those used in the USA.

Table 11.3 also shows the distribution of California quicksilver exports. China was the largest market, taking 43.4 per cent of California exports. Mexico took 37.1 per cent and South America 11.0 per cent. China and Mexico combined accounted for more than 80 per cent of California quicksilver exports. California quicksilver exports were almost entirely a Pacific phenomenon. The impact of California quicksilver How would gold and silver miners around the world have fared without California quicksilver? Would output have been affected? Would the massive gold and silver flows into the world economy after 1850 have been as large in the absence of California quicksilver? The answers to these questions are clearly counterfactual and difficult to answer with any certainty, but insight is possible. The growing demand for quicksilver after 1850 On the supply side, the significance of California quicksilver must be viewed against the backdrop of the curtailment of production at Huancavelica and the Rothschild control over Almaden and Idria. On the demand side, gold and silver discoveries in the latter half of the nineteenth century suggest that world demand for quicksilver increased substantially. While earlier precious metals flows from Spanish America were dramatic, world silver and gold production accelerated after 1850. Silver output from 1493 to 1880 amounted to 193,233,450kg. 4 Of this

222

David J.St.Clair

Table 11.3 Distribution of California quicksilver exports to Pacific Rim countries (as a % of total annual California quicksilver exports)

Source: St. Clair 1998a, p. 223. * =Less than 1%. – =No exports. *** =Not calculated

amount, 149,826,750 kg were produced between 1493 and 1850. From 1851 to 1880, another 43,406,700 kg were produced. This amounted to 22.5 per cent of the total for the entire period. The 30 years of production from 1851 through 1880 was equivalent to 29 per cent of the total produced during the earlier 358-year period (1493–1850). Silver output in the three decades after 1850 was 2.36 times the production during the three decades prior to 1850. In fact, before 1850, world silver output had grown most rapidly between 1781 and 1810. Yet silver production

California and Nevada Minerals

223

in the three decades after 1850 exceeded the 1781–1810 silver production by 63.7 per cent. Gold production shows an even more pronounced acceleration after 1850. From 1493 to 1900, world gold production amounted to about 487,227,000 ounces (Ridgeway 1929, p. v). Of this, 151,058,913 ounces were produced prior to 1850, while output from 1851 to 1900 amounted to 336,168,087. Production in the last 50 years of the nineteenth century was 2.23 times the amount of gold produced in the preceding 358 years. These are certainly rough and ready calculations and intended only to support the claim of an increase in the demand for quicksilver after 1850. The real question is whether Almaden and Idria could have, or would have, met this increased demand? It is hard to see how Almaden and Idria could have increased output to levels comparable to California output, even if the cartel had wanted to. Over the period 1850–86, Almaden and Idria would have had to double output in order to replace California quicksilver. Table 11.2 shows Almaden production increasing over the period, but a doubling of output, particularly in view of Almaden’s earlier supply problems, does not seem possible. Increasing output at Idria seems even more problematic. Turning to the cartel question, it is hard to see why the Rothschilds would have wanted to replace California output. Indeed, unsuccessful attempts were made to negotiate a deal with California producers to restrict output. It should also be noted that production at both Almaden and Idria were lower during the 1800–50 period than during the 1700–1800 period. While we do not know for sure if this was solely due to Rothschild control, it is certainly consistent with the logic of cartels. Alternatives to amalgamation The significance of California quicksilver was a function of the availability of substitutes to mercury amalgamation. The more and better the substitutes, the more likely that gold and silver output could have been maintained in the absence of California quicksilver. One possible alternative to amalgamation in gold recovery was mechanical separation. Simple panning (without quicksilver added to the pan) illustrates this approach. The action of swirling water, often combined with ridges placed in pans or rockers, trapped the heavier gold, parting it from mud and rock. Mechanical separation might be practical for placer gold, but it did not work well with hard rock gold ores or silver ores. And while placer gold could be mechanically separated, it appears this was very inefficient, often leaving half of the gold behind (Young 1970, p. 96). Only larger flecks of ‘dimensional’ gold could be recovered, leaving the fine flood or flour gold to wash away. Placering in one form or another (panning, dredging, or hydraulic) dominated the first three decades of the California Gold Rush. From 1861–70, about 90 per cent of the gold extracted in California was placer gold (Paul 1947, p. 286–7). From 1871–80, placer gold accounted for 70 per cent. If Young’s observation is correct, placering without mercury would have certainly reduced gold output significantly.

224

David J.St.Clair

Until 1890, the primary alternative to amalgamating hard rock gold ore silver ores would likely have been smelting. This was probably the earliest method of gold and silver extraction. Prior to the development of the Patio Process, silver was extracted at Potosi by smelting in small ovens. Smelting also persisted after amalgamation was introduced. Brading and Cross report that 13 to 30 per cent of Mexican silver was extracted through smelting (Brading and Cross 1972, p. 570). Zacatecas miners resorted to smelting almost half of their silver during the 1685– 1705 period when mercury was in short supply, and some regions, such as Sombrerete, relied exclusively on smelting (Brading and Cross 1972, p. 556). Some silver ores were smelted because they could not be amalgamated. Others were smelted prior to amalgamation. Galena, a common lead-silver ore, was particularly troublesome. Kustel states that galena required a ‘perfect roasting’ before successful amalgamation was possible (Kustel 1863, p. 129). Fortunately, galena was particularly disposed to smelting because its native lead provided the flux needed for smelting. Even where smelting was possible, it was seldom an economical alternative to amalgamation. Smelting was limited by high fuel costs, high labour costs, high flux costs if the ore did not naturally contain a fluxing metal, and by the need to construct expensive smelters. Large capital outlays could only be justified where sufficient quantities of high-grade ore permitted continuous operations, a condition seldom met in most regions. These problems might be overcome by shipping ores to distant smelters, but shipping costs made this impossible for all but the highest-grade ores. For example, the Calico Mine, one of California’s largest silver mines, sent its highest-grade ores to San Francisco for smelting (Steeples 1995, p. 147). Calico’s more abundant lower-grade ores were amalgamated locally. Finally, successful smelting was not as easy as it might appear, especially for silver. Gold is found in commercial quantities in only a few minerals, but silver is found in at least fifty-one silver minerals (Paul 1963, pp. 98–9). Cost-effective smelting required a good knowledge of chemistry, metallurgy, and mineralogy (Paul 1963, p. 99). Rossiter Raymond noted that very few Nevada smelters possessed enough knowledge to make their operations profitable, even when conditions were otherwise favourable for smelting (Paul 1963, p. 99). Guidio Kustel dismissed smelting as a very expensive method that was never preferred ‘unless the amount of lead or some particular circumstances should decide for it’ (Kutsel 1863, p. 145). Brading and Cross concluded that smelting was generally ‘uneconomic’ and suitable only for high-grade ores (Brading and Cross 1972, pp. 552, 554–5). Phillips claimed that smelting was actually inefficient, often leaving 15 to 25 per cent of the silver behind (Phillips 1867, p. 502). Other methods such as Cuzo Reduction of silver ores, or the use of chlorine gas, never proved very effective for most ores (Paul 1963, p. 93). We must therefore conclude that alternatives to amalgamation were usually very poor substitutes. For most ores, mercury amalgamation remained the only practical technique until the development of the cyanide process in the 1890s.

California and Nevada Minerals

225

Price effects While we do not know how much Rothschild quicksilver would have been produced in the absence of California quicksilver, we can be assured that it would have only been available at much higher prices. The Rothschilds had gone to great lengths to acquire what they thought was a virtual quicksilver monopoly. Whitney pointed out that the price of quicksilver in Mexico in 1844 was nearly triple its level in 1777, due largely to the quicksilver monopoly (Whitney 1854, p. 194). The Gold Rush and Silver Boom would certainly have driven prices higher, but how much higher? In the 1840s, the Mexican government had offered a bounty for a Mexican quicksilver mine. At the time, quicksilver was selling in Mexico for $150 per flask. This was before the California and Nevada discoveries increased demand. When serious production began at New Almaden in 1850, a flask of quicksilver sold for around $100. It seems reasonable to expect that, in the absence of California production, quicksilver prices would have certainly gone back up to $150 per flask, and probably much higher. Instead, California quicksilver production dramatically lowered quicksilver prices. Average annual quicksilver prices in San Francisco are shown in Table 11.4. In 1850, quicksilver sold at an average price of $99.45 per flask, with prices reaching as high as $114.75 that year. With full production at New Almaden, prices dropped quickly. In 1856, the average price was less than half the 1850 level. In the 1870s, Nevada’s Washoe mills were using California quicksilver that sold for as little as $25.25 per flask. Lower quicksilver prices lowered the cost of charging processing mills. Rossiter Raymond reported in 1871 that one pound of quicksilver was needed to process 10 tons of gold ore (Raymond 1871, p. 29). At the Comstock, about 300–500 pounds of quicksilver were needed to process 3,000 pounds of ore (Smith 1943, p. 43). Lower quicksilver prices also reduced the cost associated with lost quicksilver. Only part of the quicksilver was recovered after amalgamation. Raymond reported that 60 to 65 per cent of the quicksilver employed in gold refining was lost (Raymond 1871, p. 29). Silver refined with the Patio Process lost about 1.5 pounds of quicksilver per pound of silver produced (Raymond 1869, p. 10). Raymond estimated that quicksilver losses with the Washoe Process were only one-third the losses incurred with the Patio Process (Raymond 1869, p. 10). Grant Smith put Washoe losses at between 1.0 and 1.5 pounds of quicksilver lost per ton of ore processed, with richer ores incurring higher losses (Smith 1943, p. 43). Smith considered Comstock quicksilver losses over the years to have been quite substantial, estimating that 14 million pounds of quicksilver had been lost by 1943 (Smith 1943, p. 257). The cost of this lost quicksilver had to be incorporated into milling prices, thus affecting the cost and profitability of silver mining. But how great an impact would these higher costs have had on gold and silver output? Precise measurements are not possible at this time, but some crude assessments can be made. The effect of higher quicksilver prices on gold, silver, or vermilion output essentially depended on the price elasticity of demand for quicksilver users. Elasticity in turn depended on demand characteristics such as the size of the input cost in the

226

David J.St.Clair

Table 11.4 Average San Francisco quicksilver prices 1850–90 (dollars per 76.5-pound flask)

Source: California State Mining Bureau 1917, p. 47.

production process and the availability of substitutes. Price elasticity of quicksilver also depended on whether it was to be used in the recovery of gold, in processing silver ores with the Patio Process, in refining silver with the Washoe Process, or in the manufacture of vermilion. The size and significance of quicksilver costs in each of these endeavours have been estimated and are summarized here (St. Clair 1998a, pp. 224–9). In the recovery of gold, quicksilver costs were about 2 per cent of milling costs, a tiny fraction of overall production costs, and an even smaller portion of the value of gold recovered.5 Even at higher quicksilver prices such as $150 per flask, it is unlikely that the demand for quicksilver for gold refining would have been significantly reduced. Gold output would have therefore been only marginally impacted. Silver refining was another matter. With the Patio Process, quicksilver costs amounted to about 4.4 per cent of the value of silver recovered. Higher quicksilver prices would have had a significant impact on the profitability of silver mining. For example, with quicksilver selling at $150 per flask, quicksilver costs would have accounted to 18 per cent of the value of silver recovered, enough to eradicate profits on all but the richest ores. By the same token, lower quicksilver prices could stimulate silver production. Whitney reported that Mexican silver producers responded to lower quicksilver prices in the 1770s by increasing silver output. The US Treasury Department cited lower quicksilver prices as one of the reasons for the increase in silver output in the 1860s (Whitney 1854, p. 194; US Treasury 1864, p. 197). Silver produced with the Washoe Process was less sensitive to higher quicksilver

California and Nevada Minerals

227

prices, but still significant. For example, during the period 1859 through 1881, quicksilver prices averaged $45.90 per flask. Quicksilver costs during the period amounted to about 2 per cent of the value of the silver recovered. With Comstock silver mines showing average profits of 6.25 per cent, these quicksilver costs were significant. If quicksilver had been priced at $150 per flask, Comstock silver profits would have been reduced by 70 per cent. It is not hard to see why Grant Smith called quicksilver costs ‘enormous and startling’ (Smith 1943, p. 254). In the manufacture of vermilion, the quicksilver cost accounted for more than 70 per cent of the price of vermilion. Even small increases in quicksilver prices had to be passed through to higher vermilion prices in order to maintain profits. However, with intense competition from substitute dyes, vermilion prices were quite elastic, making it difficult if not impossible to raise prices. Consequently, vermilion producers were certainly the most sensitive to higher quicksilver prices. This may explain why California and Rothschild quicksilver competed most intensely in New York and China, two markets dominated by vermilion producers. This crude survey of elasticity factors suggests that gold production would have been the least affected by higher quicksilver prices. Silver would have been significantly impacted, with Patio Processed silver affected more than Washoe Processed silver. The differential impact of higher quicksilver prices on gold versus silver would have altered the relative quantities of the two metals. This in turn might have impacted the political and monetary disputes over bimetallism. Vermilion producers were probably the most sensitive to quicksilver prices. Gold, silver, and the development of California industry California on the eve of the Gold Rush was a remote Mexican outpost with a nonIndian population that did not exceed 8,000. Settlement was confined to a coastal strip from San Diego to Sonoma. Economic activity was centred on the ranches, the large cattle ranches that produced hides and tallow, the two leading commodities that connected California with the outside world. Along with soap-making, processing hides and tallow were the only activities that might be described as industrial. The hides, minimally dressed and processed, were sold to foreign merchants or exchanged for imported products. Cattle brought from four to six dollars per head, their price largely restricted to the value of their hides and fat. Ample supply and very limited demand made the meat almost worthless. The pre-gold California economy was certainly rudimentary. Some historians, such as Robert Cleland and Osgood, have gone even further, arguing that it was stagnant (Cleland and Hardy 1929, p. 1). However, the Gold Rush unleashed a torrent of change on this pastoral economy. Its first effects were disruptive. Workers, ranch hands, and shopkeepers rushed off to seek their fortunes in the gold fields. Although we have no statistics, production must have suffered. Fortunately, the disruption was only temporary, as many who rushed off in search of fortune returned after finding only hard work and little gold. On returning, they found a very different economy. Overnight, gold transformed California’s lethargic business world into a surging boom. People rushed in as

228

David J.St.Clair

California became the centre of world precious metals production. By the beginning of 1849, California’s population had reached 26,000 (Rolle 1987, p. 166). By the summer, it jumped to 50,000. San Francisco became the world’s fastest growing city, its population exploding from 812 in 1848 to 25,000 in 1850. The official census of 1850 recorded 92,597 people living in the state, while unofficial estimates put the correct figure at 115,000. California’s population rose to 380,000 in 1860, 560,000 in 1870, and 865,000 in 1880 (US Department of Commerce 1975, p. 25). This surge of population brought an unprecedented demand that turned the traditional economy upside down. The scrawny Spanish-stock cattle that had earlier sold for $5 per head now brought $300 to $500 per head to feed hungry miners with gold in their pockets (Cleland and Hardy 1929, p. 36). Mining surged, and California agriculture was soon booming as well. Herds of cattle and sheep driven to California augmented supplies. Wheat output increased dramatically. By 1860 California was producing five times as much wheat as all other Western states and territories combined. California wheat exports poured into world markets (Paul 1959; and St. Clair 2000). Vineyards were planted and a wine industry took root within a couple of years of Marshall’s discovery. The impact of gold on California agriculture has generally been appreciated, but what about California industry? While some historians have acknowledged that important first steps were taken during these years, most have argued that industrial development lagged until the last decade of the nineteenth century, or even later. Consequently, the Gold Rush is accorded little role in the development of industry.6 Often cited reasons for the delay in the development of industry during the Gold Rush include high transportation costs, distant markets, high wages, high interest rates, lack of coal, lack of iron, lack of cotton supplies, lack of water power near cities, expensive water in large towns, expensive land prices near deepwater ports, insecure land titles, and the prospects of better profits in agriculture and mining. According to John Hittell, these obstacles prevented California from exporting any manufactures, kept industry only producing crude industrial products, and limited the state’s exports to unfinished or semi-finished products (Hittell 1879, pp. 183–4). Hittell concluded, ‘California agriculture and mining industries had reached advanced development in some branches, while our manufactures are backward’ (Hittell 1879, pp. 183–4). Symptomatic of the state’s retardation was, according to Hittell, its failure to embrace steam power, relying instead on its human muscle to its ‘great disadvantage’ (Hittell 1879, p. 184). Only a few historians have expressed positive views regarding the development of California industry during the Gold Rush. John Caughey wrote that California manufacturing developed ‘hand in hand’ with mining, commerce, and agriculture in northern California (Caughey 1970, p. 202). Carey McWilliams took a different tack, arguing that California actually enjoyed the advantages of a head start in the competition for industry (McWilliams 1949, p. 216). He argued that California became a manufacturing centre almost at the same time that it became a state. According to McWilliams, California’s early start in industrialization was a distinct departure from the norm, a great exception brought about by the novel conditions

California and Nevada Minerals

229

Table 11.5 California manufacturers

Source: United States Census of Manufacturers 1850, 1860, 1880; United States Population Census 1870.

created by the Gold Rush and California’s unique environment (McWilliams 1949, p. 214). Census data problems One reason for the generally negative view of California industry during the Gold Rush may stem from a problem with United States Census data.7 Table 11.5 shows California manufacturing from census data. The 1850 census ranked California manufacturing sixteenth (by value of output) among the 36 states territories, a remarkable achievement for a new state. By 1860, California manufacturing output had risen to seventh place, growing by 430.6 per cent during the 1850s. This growth was far faster than that of any other state. Table 11.5 also shows the number of manufacturing establishments. Between 1860 and 1870, the number of establishments appears to drop precipitously along with a modest drop in the value of manufacturing output. Historians have generally attributed the decline to the revival of competition following the Civil War, and the opening of the transcontinental railroad in 1869. The state’s early start at industry, they argue, proved too fragile to survive. However, this view is incorrect. The decline of manufacturing appears to be due solely to incorrect categorization in the census. Census data for both 1850 and 1860 include ‘mining’ in the ‘manufacturing’ category. Consequently, the Gold Rush of the early 1850s and the consolidation of mining in larger companies after those years distort manufacturing statistics and give the impression that early industrial development during the Gold Rush was artificial and transient. Table 11.6 shows California manufacturing with gold mining removed from industry figures. These figures reduce the size of the ‘manufacturing’ sector reported in the census, but still show California with an industrial sector larger than nine other states and territories. More importantly, between 1850 and 1860, California’s industrial sector (excluding gold mining) grew by 510.6 per cent, faster than the 396.4 per cent growth in gold mining. By 1860, California industry (again excluding gold mining) was ranked eighteenth, and was larger than the manufacturing sectors (with mining still included) of 21 other states and territories. In addition, when the distortion of gold mining is removed, there is no decline in the number of establishments or output after 1860.

230

David J.St.Clair

Table 11.6 California manufacturers, excluding gold mining

Source: St. Clair 1999, p. 192. Table 11.7 Manufacturing per capita in 1870

Sources: The author’s calculations from US Commerce Department 1975; United States Census of Manufacturers 1880.

By 1870, California ranked twenty-fourth in population and sixteenth in manufacturing output (gold mining excluded). Table 11.7 shows 1870 population, manufacturing output, and output per capita for California and six other states with larger populations and larger manufacturing sectors. California manufacturing output per capita exceeded that of Ohio and Illinois, but was still well behind the others. While output per capita is not a flawless measure of the manufacturing sector, it is indicative. By 1880, California still ranked twenty-fourth in population and fifteenth in agricultural output, but had moved up to twelfth in manufacturing output (US Census of Manufacturers 1880, p. xii). California’s per-capita manufacturing output in 1880 was about the same as Illinois’s, and still greater than Ohio’s. California manufacturing per capita also increased relative to all of the states shown in Table 11.7, except Illinois. By any measure then, California manufacturing grew rapidly during the Gold Rush. There was no significant delay or lag in developing California industry, and while the often-cited obstacles to growth were formidable, the history of California industry is a story of overcoming these obstacles not succumbing to them. The impact of gold and silver on California industries Gold and silver mining influenced California industries in three ways. First, the Gold Rush precipitated the population boom that created a soaring demand for a wide range of consumer and producer goods. These products often had little or no

California and Nevada Minerals

231

direct connection to gold mining. In these cases, there was nothing unique about the gold industry; it was merely the sector that fuelled an economic expansion from which other industries benefited. Second, direct gains accrued to industries linked to gold and silver mining. Booming gold and silver mining increased the demand for inputs and technologies from supplying industries. Third, and perhaps most important, technologies and industrial infrastructure initially developed for gold and silver mining were transferred to other sectors and products. Precious metals mining was the catalyst for the creation of an industrial infrastructure centred on a foundry-machine shop core. Links to consumer goods industries The increased demand for food, clothing, shelter, transportation, and construction materials was initially met mostly with imports. For example, glass bottles were in such demand that old bottles from Honolulu, Tahiti, and Mexico were collected and shipped to San Francisco (Hittell 1882, p. 524). Imports from the East Coast followed, but breakage and transportation costs doubled their price. Local production of glass began in San Francisco as early as 1862. Similarly the first stone house built in San Francisco was constructed of imported Chinese marble in 1854 (California Division of Mines 1951, pp. 235, 238). But within two years, stone from California quarries was replacing imports. Likewise, California initially imported all of its flour, and the first flour mills in the state got their start by remilling imported flour that had spoiled in transit. Flour production expanded rapidly as California’s wheat crop grew and flour imports ceased in 1860. California flour production continued to expand, exporting to world markets. San Francisco grew first as a bustling trade centre before becoming the centre of California industry. At least 29 different consumer goods industries were established in and around San Francisco by 1860 (St. Clair 1999). This does not include the numerous producer goods industries established in the city (discussed below). By 1860, California’s largest industries, in order of size of output, were: flour milling, lumber, sugar refining, machinery (including steam engines), and malt liquors. The largest industries in 1870 were flour milling, lumber, machinery and shoe findings, sugar refining, quartz milling, and cigar making. These were mostly consumer goods industries (except machinery, quartz milling and, to a certain extent, lumber) that thrived in the general prosperity initiated by the Gold Rush. Flour and lumber mills proliferated. Ninety-one flour mills and 279 lumber mills were in operation by 1860, compared to only two and ten, respectively, in 1850. The growth of California flour milling is shown in Table 11.8. Flour mills produced half of the food (by value) produced in the state. In addition, flour and lumber mills were capital intensive, further augmenting the demand for California machinery. California flour mills employed more than 12 per cent of the state’s steam engines in 1870. In addition, flour exports, along with silver and quicksilver, made up the bulk of California’s exports to China in the nineteenth century (St. Clair 2000).

232

David J.St.Clair

Table 11.8 California flour mills

Source: US Census of Manufacturers 1850, 1860, 1880; US Population Census 1870.

Links to producer goods industries Many industries were connected to gold and silver mining through suppliercustomer links. These could be either forward or backward linkages. Forward linkages are connections with ‘downstream’ industries, that is, industries that utilize the product. In contrast, backward linkages are upstream connections to industries that provide raw materials or machinery used in the production of the product in question. Forward linkages to gold mining included jewellery making, gold beating (to produce gold leaf), and coin minting. Jewellery and gold beating were undertaken in San Francisco, but neither was particularly significant. However, in 1850, private mints in San Francisco began minting gold coins to alleviate California’s currency shortage. These private mints operated until a United States Mint opened in San Francisco in 1855.8 While forward linkages were not extensive, backward linkages were very significant. To appreciate these connections, it is important to see mining, especially gold mining, as an industry, rather than as a discovery or find. Perhaps the most common image of the California gold miner is that of a bearded, grizzled prospector bent over a stream, panning for gold. While this may have been typical of many of the early Forty-niners, it does not accurately reflect gold mining after it quickly became more of an industry and less of an adventure. Different types of gold and silver mining had different links to industry. Placer gold mining (including panning, rockers, toms, and sluices) used water, motion, quicksilver, and trapping mechanisms such as ridges and cleats to separate gold from mud and gravel. The backward linkages from placer mining included links to quicksilver, lumber, and the acid industry (for parting gold and silver). To meet the demand from mints and mines, San Francisco acid production began in 1854. Lumber was needed for rockers, toms, and sluices. By the mid-1850s, however, simple placer mining sites had been played out. Hydraulic mining and dredging, more advanced forms of placering, were developed to work less accessible ores. Both hydraulic mining and dredging are very capital intensive, with more extensive and significant links to other enterprises. California industry expanded to meet the demand for leather hoses, pumps, and nozzles. The dams and flumes required for hydraulic mining also dramatically increased the demand for lumber. Lumber mills responded with special planks, narrower at one

California and Nevada Minerals

233

end so they could be readily attached end-to-end, to construct the long wooden channels for hydraulic sluices. Leather hoses were made in San Francisco starting in 1857. California oak-tanned leather was stronger than leather used by eastern and European hose producers (Hittell 1882, p. 520–2). As a consequence, California hoses were superior products, stronger and less expensive. California leather hoses were exported around the world and were used extensively by fire departments until rubber hoses replaced them after 1874. Nozzles, first made of wood, were soon crafted out of metal in California foundries. Dredging was initially tried in 1850 on a river boat converted to the task of capturing gold from river bottoms in Marysville.9 However, dredging did not become important until after 1880, when court rulings limited hydraulic mining. The Risdon Iron and Locomotive Works of San Francisco produced a larger dredge in the 1890s that re-ignited interest in the technology. Dredging remained an important mining technique into the 1940s. Quartz or hard rock mining of both silver and gold had the greatest impact on California industry. Gold embedded in quartz was discovered as early as 1849, and was followed by a wave of speculative excitement. But the excitement ended in a bust, and the quartz mining that survived was carried out on a small and unprofitable basis for many years. Rodman Paul observed that, as late as 1859–60, the cash returns to quartz mining could be written off as unjustified were it not for the unique technologies invented in this activity (Paul 1963, p. 33). The California mining equipment industry owed much to the persistence of these early ventures and to Comstock silver mining. For example, in 1863, there were more than 100 quartz mills in northern Nevada, most within fifteen miles of Virginia City (US Treasury Department 1864, p. 202) These were capital intensive, with each mill having an average of 4 to 5 stamps and costing from $10,000 to $100,000 per mill. Three-quarters were steam driven, the remainder water powered. By and large, the equipment in these mills was developed and manufactured in California. In addition, techniques developed on the Comstock for milling, tunnelling, draining, ventilating, shoring, etc. were later employed in California gold mines. Thus, Nevada’s Silver Boom played an invaluable role in refining the techniques and technologies that were to nurture and sustain both California mining and California mining equipment manufacturers. Hard rock mining entailed tunnelling to reach the ore, digging ore out and bringing it to the surface, and finally crushing and processing the ore. All stages of this activity were capital intensive and required specialized machinery. To get at the ore, drills and explosives were used to dig through rock. San Francisco foundries and machine shops developed drills that reduced friction, breakage, and fuel consumption (Hittell 1882, p. 657). Hand drills were quickly replaced by steamdriven patent drills. Steam engines were originally taken down into the mines to power the drills, but this drastically reduced their efficiency. The air-compressor permitted the steam engines to remain above ground with hoses supplying the compressed air to the drills. More leather hoses were needed. Explosives were also used to get at ore. Imported black powder was originally used, but transporting it was dangerous and shipments were disrupted by the Civil War. Within the state, the California Powder-Works opened near the city of Santa

234

David J.St.Clair

Cruz in 1861 to produce black powder (Hittell 1882, p. 709). The company subsequently opened a second facility near Point Pinole to produce its high explosive ‘Hercules’ powder. Acids were used in the manufacture of these explosives, leading to the development of yet more satellite industries. California explosives were shipped throughout the West for use in mines and railroad construction and were exported to Canada, Hawaii and Latin America, especially Mexico (Hittell 1882, p. 707). California’s explosives industry, however, did not lead to the development of an armaments industry, at least not in the nineteenth century. The manufacture of guns remained an eastern specialty. Blasting techniques developed in the gold mines, on the other hand, were applicable to the mining of other minerals. One of the more interesting applications was found in California’s marble quarries where precision blasting of marble blocks was perfected. All but the shallowest of hard rock mines required drainage, venting, and hoisting. Timbers and lumber were needed for hoists, supports, and shoring. Hoisting machines and steam engines were produced by California foundries and machine shops and San Francisco wire and cable makers made cable for hoists and ore trams. In addition, most mining machines used leather belts in conveyers and drives. By 1861, four San Francisco firms manufactured leather belts superior to competing eastern and European products (Hittell 1882, p. 521). San Francisco foundries also produced most of the pumps used to pump water out of California and Comstock mines (Hittell 1882, p. 657). The Risdon Iron and Locomotive Works manufactured water pipe for use in Virginia City, as well as irrigation pipe for Hawaiian plantations, and made the much-acclaimed pumps for the Chollar-Norcross Mine (Hittell 1882, p. 660). Pumps provide an interesting example of how California firms overcame the obstacles working against west coast manufacturing. California foundries produced mostly mining pumps, which were large and designed and manufactured to order. California foundries relied on their design expertise, their proximity to mining company customers, and superior service to compete against cheaper eastern imports. Although they succeeded in securing the bulk of the mining business, they could not compete with eastern firms in the market for smaller pumps for cisterns, household use, or small business applications (Hittell 1882, p. 658). Small eastern pumps were mass-produced, employing cheap child labour, and sold for up to 60 per cent less than local products. California producers enjoyed neither the labour force, the low wages, nor the market size that enabled them to compete in this market. California steam engines were also developed for the mines. The 1870 census recorded 42 steam engines in use in California mines, many made by California companies (US Population Census 1870, p. 760). California ranked ninth in the number of steam engines used in mines. This is all the more impressive in light of California’s water-power resources. California ranked first in the use of water wheels as a power source in mines, employing 70 of the 134 water wheels in use in the United States in 1870 (US Population Census 1870, p. 496–8). While steam engines were developed for the mines, they spread to other California industries. In 1870, there were 604 steam engines in use in the state’s manufacturing establishments. The lumber industry, flour mills, distilleries, and the iron trades often utilized steam

California and Nevada Minerals

235

power. The assertion that California industry lagged because it failed to embrace the steam engine is simply not correct. Processing quartz ores proved to be a formidable challenge. Ore-bearing rock was first broken into smaller, more manageable pieces. This was followed by grinding. California initially imported grinding machines from Europe for this task, but these, according to John Hittell, proved to be ‘fancy and usually worthless’ (Paul 1963, p. 31). They were quickly abandoned in favour of arrastras, simple Mexican devices that dragged heavy stones over the ore. A Chilean version substituted a millstone. Slow and ineffective, arrastras were soon replaced by California stamp mills which used heavy iron feet, mechanically lifted and dropped, to grind the ore. They were produced by California foundries and machine shops. Rodman Paul called hydraulic mining and the California stamp mill the crowning technological achievements of the California Gold Rush. The stamp was especially important in encouraging the development of local foundries and mining technology. Because about two-thirds of the gold in quartz ores was not recovered by early processing methods, California miners, working with local foundries and machine shops, rapidly developed other techniques to improve yields. Paul claims that more progress was made in the first twelve years of the California Gold Rush than had occurred over the previous several centuries (Paul 1963, p. 31–2). Californians invented concentrators, machines that generally used conveyer belts and shaking motion to further concentrate ores before amalgamation. They also developed a second grinding process, often with mercury added. In the late 1850s, metal pans with mechanical stirring devices and steam heat emerged to facilitate amalgamation. These techniques were later incorporated into Washoe Process vats used on the Comstock. All of these mining developments stimulated California industry. The effect on the metal working industries appeared immediately in census data. In 1850, half the state’s non-gold manufacturing establishments were blacksmith shops. Since there were no separate census categories for ‘foundries’ or ‘machine shops’, these were included in blacksmithing. In terms of value of output, California blacksmithing ranked third in the nation behind only New York and Pennsylvania. This is remarkable for a state that was less than two years old! By 1880, the outputvalue of California foundry and machine shops, now separately enumerated, ranked eleventh among the states, but seventh on a per-capita basis (US Population Census 1870, p. xxi). In blacksmithing, California ranked seventh in output, and second in output per capita. The growth of the iron-working trades is obscured by the increasing complexity of the census. This category includes blacksmithing, foundry and machine shops, wire and cable making, iron pipe, pumps, steam engines, saws, shipbuilding, wheelwrighting, and other types of enterprises. These activities formed the core of nineteenth-century industry. After 1850, successive censuses expanded the reporting categories. While this was more accurate and useful for some purposes, the growth of iron working in the aggregate is lost. Table 11.9 recombines these separate categories in the census into an aggregate ‘iron working trades’ industry. The

236

David J.St.Clair

Table 11.9 California iron working trades

Source: St. Clair 1999, p. 203. Note: ‘Iron Working Trades’ includes all iron working census categories.

dramatic increase in iron working trades, despite the state’s poor natural endowment of iron ores and coal, is striking. The spread of gold and silver technologies Technologies developed for gold and silver mining were not confined to the mining sector. As pointed out above, such devices as hoses, steam engines, and pumps all found their way to other sectors of the economy. Nathan Rosenberg has called this process ‘technological convergence’, and maintains that it was vital to creating the machine tool industry on the East Coast in the early nineteenth century (Rosenberg 1963, pp. 414–63). The textile industry, he argues, was the initial catalyst for technological convergence on the East Coast. Technological convergence can also be seen in California, but with mining equipment serving as the catalyst. The blacksmith shops, foundries, and machine shops that produced mining equipment also created technologies and an industrial base that could later be employed in shipbuilding, in the defence industry, and in other types of manufacturing. By the 1880s, for example, California firms were supplying most of the machinery used on Hawaiian plantations and in sugar cane processing, replacing European imports (Hittell 1882, p. 653). California’s hydroelectric power also had early connections to the Gold Rush (Hutchinson 1969, pp. 218–19). The first hydroelectric operation in the state was undertaken in northern California in 1879. Soon after, Lester A.Pelton, a millwright and carpenter in the Mother Lode town of Camptonville, created the turbine wheel generator, building on technology developed for gold mining. The cable industry provides another example of technology dissemination. A.S. Hallidie, president of The California Wire-Works Company, made screens for quartz mills and flour mills, riddles, bird cages, fenders, fire guards, and many other wire products for use in kitchens and industry. In 1868, Hallidie invented a wire ropeway for transporting ores. Soon after, he used the same technology to invent the cable railway, which powers San Francisco’s famous cable cars (Bailey 1996, pp. 115– 26; Hittell 1882, pp. 425–6, 668). California’s oldest foundry, the Union Iron Works, also illustrates how gold mining technologies were transferred to other industries. Founded in 1849, the Union Iron Works initially overcame a lack of iron by buying scrap iron made

California and Nevada Minerals

237

plentiful by the fires that destroyed San Francisco in the early 1850s. The Union Iron Works produced a large share (90 per cent by one estimate) of the mining equipment used by California and Comstock mines (Teiser 1946, pp. 39–52). From mining equipment, Union Iron Works branched out to supply other iron-working industries, building the first locomotive on the West Coast in 1865, and the first steel ship made on the West Coast in 1885. One striking difference between California producers and their eastern counterparts that encouraged technology dissemination was the degree to which California producers did not specialize. Eastern foundries and machine shops tended to specialize in the production of a few products. However, with many smaller local markets, California foundries often made more than twenty products, ‘everything that is in demand, from mining-machinery, locomotives, steamship engines, sugar-mills, and architectural iron-work, down to the various small articles required for every-day use’ (Hittell 1882, p. 659). Diversity was also typical of California’s agricultural equipment producers (McWilliams 1949, p. 224). These differences between eastern and western foundries were probably due to wider markets in the East which facilitated specialization. It was also due to the mining origins of western foundries. Mining equipment was very diverse and often custom-made. San Francisco foundries survived by staying flexible, by experimenting, by innovating, and by producing a wide array of products. McWilliams argued that a willingness to experiment was a long-standing hallmark of California that had taken root in the state’s mining past (McWilliams 1949, p. 221). Equally important, the diversity of California foundries and machine shops probably speeded up the process of technological diffusion because it became more of an in-house process on the West Coast, facilitating easier transfer of technology from product to product. California and the development of the Asian-Pacific economy The developments discussed above shaped and stimulated Pacific-Rim and world trade. Sometimes the effects were immediate; at other times, the consequences took decades to unfold. In each case, precious metals production was the catalyst for bringing the Pacific Coast of North America into a developing Pacific economy. Gold was California’s first major export, and while not primarily destined for Pacific markets, it nonetheless freed up silver for Asian trade. On the other hand, Nevada silver joined the long-established flow of silver across the Pacific. California and Nevada gold and silver mining were important parts of the worldwide increase in precious metals that increased world liquidity and trade in the second half of the nineteenth century. In addition, the abundance of both silver and gold allayed Western concerns over a drain of silver due to Asian trade. As a consequence, silver was eventually de-monetized. Quicksilver was not only a major California export, and primarily a Pacific export, but was also indispensable to gold and silver production. World gold and silver production after 1850 would not have been as large in the absence of California quicksilver. This was especially true in the case of silver; less in the case of gold.

238

David J.St.Clair

Counterfactual speculation suggests that late nineteenth-century disputes over bimetallism and the de-monetization of silver may have taken very different courses in the absence of California quicksilver. The longer term effects of gold and silver production can be seen in the role of these metals in developing California’s industrial base. The backward linkages from gold and silver mining created an industrial nexus centred around foundries, machine tool companies, and the iron working trades The Asian-Pacific consequences of these developments are noteworthy. They are best appreciated by considering the state of Pacific trade prior to 1848. Before the Gold Rush, Pacific trade might be viewed as consisting of two separate, largely distinct activities. On the one hand, Asian trade and Asian-European trade was carried on in the West Pacific. In contrast, limited trade was taking place across the Pacific. While limited, this trans-Pacific trade, essentially consisting of silver-forsilk trade, had been carried on since the sixteenth century. In the early nineteenth century, American ships joined the trans-Pacific trade, adding tea to the list of traded products. From an American perspective, the Pacific was initially a mere water extension (a very long water extension) of the Atlantic. It was a trade route to Asian markets. The Pacific Coast of North America was essentially a way stop on this trade route. It was a place to acquire provisions, or a small market for yankee products, or perhaps a place to acquire hides or furs. After 1848, California set off on a different path. It became an economy in its own right and, equally important, it became an important member of a developing Asian-Pacific economy. Gold, silver, and quicksilver initiated this transformation. California agriculture and industry were ignited by precious metals production. California industrial development in particular added a dynamic dimension to transPacific trade. Asian-Pacific markets were important to California mining equipment manufacturers and became the destination for California’s early industrial exports. This trade was not merely an adjunct to Atlantic trade; rather it was a central feature of California trade. It was a west-coast phenomenon with a keen eye on AsianPacific export markets from its very inception. The development of a robust economy on the eastern shores of the Pacific impacted Asian-Pacific economies as well. California became a growing market for western Pacific producers, especially from China and Australia. In 1851, The Economist took note of the increase in trade between Hong Kong and California and suggested that this trade offered the best hope of making Hong Kong a useful settlement (The Economist, March 8, 1851). Pacific trade, as opposed to trade through the Pacific, required trading partners on both shores. The development of the California economy, lead by precious metals production, contributed to building such a Pacific economy. It would be more than a century before America’s Pacific trade surpassed its Atlantic trade, but the foundations for Pacific trade were very much a product of precious metals production on the eastern shores of the Pacific.

California and Nevada Minerals

239

Notes 1 2 3 4 5 6

7 8 9

This discussion of San Francisco exports is drawn from St. Clair (2000). On silver flows to Asia from the sixteenth century, see Flynn and Giráldez (1995, 1996). This discussion of the New Almaden Quicksilver Mine and California quicksilver production is from St. Clair (1994–95). These figures are the author’s calculations based on figures presented in Laughlin (1888, p. 42). The author’s calculation (St. Clair 1998a, p. 227) based on data from Raymond (1871, p. 29) The views of historians on the pace of California’s industrial development, and the reasons for its delay and retardation, are surveyed in St. Clair (1999). Also see Bullough, Orsi and Rice 1996, p. 279; Cleland and Hardy 1929, pp. 133–34; Hittel 1862, p. 304; Hittel 1879, pp. 183–84; Hutchinson 1969, p. 207; Nash 1973, pp. 39– 53; Pomeroy 1965, pp. 111–13; Rolle 1987, p. 229. For views about the effects of the Civil War on California industry, see Cleland and Hardy 1929, p. 134; Bullough, Orsi, and Rice, 1996, p. 279; Rolle, 1987, p. 229. This problems with census data was originally, and more fully, discussed in St. Clair (1999, pp. 190–93). For a discussion of private coinage in California, see Adams (1913). This discussion is drawn from Bullough, Orsi and Rice (1996, p. 196) and Aubury (1910).

References Adams, E. (1913) Private Gold Coinage of California, 1849–55: Its History and its Issues, Brooklyn: Edgar H.Adams. Aubury, L. (1910) Gold Dredging in California, Sacramento, California: State Printing Office. Bailey, L. (1996) Supplying the Mining World: The Mining Equipment Manufacturers of San Francisco, 1850–1900, Tucson, Arizona: Westernlore Press. Bakewell, P. (1971) Silver Mining and Society in Colonial Mexico: Zacatecas, 1546– 1700, Cambridge: Cambridge University Press. Boalich, E. (1914) Mineral Production for 1912, Sacramento, California: State Printing Office. Brading, D. and Cross, H. (1972) ‘Colonial Silver Mining: Mexico and Peru’, Hispanic American Historical Review, 52, pp. 545–79. Bullough, W., Orsi, R. and Rice, R. (1996) The Elusive Eden: A New History of California, New York: McGraw-Hill. California Division of Mines (1951) Geologic Handbook of the San Francisco Bay Counties, San Francisco: Division of Mines. California Miner’s Association (1899) California Mines and Minerals, San Francisco: Louis Roesch Company. California State Mining Bureau (1917) California Mineral Production for 1916, Sacramento: California State Printing Office. Cleland, R. and Hardy, O. (1929) The March of Industry, San Francisco: Powell. Caughey, J. (1970) California: A Remarkable State’s Life History, Englewood Cliffs, N.J.: Prentice-Hall. The Economist, 8 March 1851.

240

David J.St.Clair

Flynn, D. and Giráldez, A. (1995) ‘Born with a “silver spoon”: The origins of world trade in 1571’, Journal of World History, 62, pp. 201–21. Flynn, D. and Giráldez, A. (1996) ‘Monetary substances in global perspective: An introductory essay’, in D.Flynn and A.Giráldez (eds) Metals and Monies in an Emerging Global Economy, Aldershot, UK: Variorum. Hittell, J. (1862) The Resources of California, San Francisco: A.Roman & Co. —(1879) The Resources of California, San Francisco: Bancroft. —(1882) The Commerce and Industries of the Pacific Coast of North America, San Francisco: Bancroft. Hutchinson, W. (1969) California: Two Centuries of Man, Land, and Growth in the Golden State, Palo Alto, California: American West Publishing. Kustel, G. (1863) Nevada and California Processes of Silver and Gold Extraction, San Francisco: Frank D.Carlton. Laughlin, J.L. (1888) The History of Bimetalism in the United States, New York: D.Appleton and Co. McWilliams, C. (1949) California: The Great Exception, New York: A.A.Wyn. Nash, G. (1973) ‘Stages of California’s economic growth, 1870–1970: An, interpretation’, in G.Knoles (ed.) (1973) Essays and Assays: California History Revisited, Los Angeles: Ward Ritchie Press. Paul, R. (1947) California Gold, Lincoln, NB: University of Nebraska Press. —(1959) ‘The wheat trade between California and the United Kingdom’, The Mississippi Valley Historical Review, XLV, pp. 391–412. —(1963) Mining Frontiers of the Far West, 1848–1880, New York: Holt, Rinehart and Winston. Phillips, J. (1867) The Mining and Metallurgy of Gold and Silver, London: E. and F.N.Spon. Pomeroy, E. (1965) The Pacific Slope, Lincoln, Nebraska: University of Nebraska Press. Raymond, R. (1869) Mineral Resources of the States and Territories, Washington, D.C.: Government Printing Office. —(1871) Mines, Mills, and Furnaces of the Pacific States and Territories, New York: J.B. Ford and Co. Rolle, A. (1987) California: A History, Arlington Heights, Ill.: Harlan Davidson. Rosenberg, N. (1963) ‘Technological change in the machine tool industry, 1840–1910’, The Journal of Economic History, XXII, pp. 414–43. Ridgeway, R. (1929) Summarized Data of Gold Production, Washington, D.C.: Government Printing Office. Smith, G. (1943) The History of the Comstock Lode, 1859–1920, University of Nevada Bulletin No. 3, Geology and Mining Series No. 37. St. Clair, D. (1994–95) ‘New Almaden and California quicksilver in the Pacific Rim economy’, California History, LXXIII, 4, pp. 278–95. —(1998a) ‘California quicksilver in the Pacific Rim economy, 1850–90’, in S.M.Miller, A.J.H.Latham and D.O.Flynn (eds) Studies in the Economic History of the Pacific Rim, London: Routledge. —(1998b) ‘American Trade Dollars in nineteenth century China’, in D.O.Flynn, L. Frost and A.J.H.Latham (eds) Pacific Centuries, London: Routledge. —(1999) ‘The Gold Rush and the beginnings of California industry’, California History, LXXVII, 4, pp. 185–208. —(2000) ‘San Francisco’s Pacific exports, 1850–1898’, in D.O.Flynn, A.Giraldez and J. Sobredo (eds) Studies in Pacific History: Economic, Social, and Political History, Aldershot, UK: Variorem. Steeples, D. (1995) ‘Calico Silver and the Fabric of Western Development’, in W.Childs (ed.) Essays in Economic and Business History, Vol. XIII, Chelsea, MI: Book Crafters.

California and Nevada Minerals

241

Tandeter, E. (1993) Coercion and Market: Silver Mining in Colonial Potosi, 1692–1826, Albuquerque: University of New Mexico. Teiser, R. (1946) ‘The Charleston: An industrial milestone,’ California Historical Society Quarterly, XXV, pp. 39–52. US Bureau of Mines (1886) Mineral Resources of the United States, 1885, Washington, D.C.: Government Printing Office. US Census Bureau (1850) Census of Manufacturers. —(1860) Census of Manufacturers. —(1870) United States Population Census. —(1880) Census of Manufacturers. US Department of Commerce (1975) Historical Statistics of the United States, Colonial Times to 1970, Washington, D.C.: Government Printing Office. US Treasury Department (1864) Statistics of Foreign and Domestic Commerce of the United States, Washington, D.C.: Government Printing Office. (Citations are to the Johnson Reprint (1973) New York: Johnson Reprint Corporation.) Williams, A. Jr. (1885) US Bureau of Mines, Mineral Resources of the United States, 1883 and 1884, Washington, D.C.: Government Printing Office. Whitney, J. (1854) The Metallic Wealth of the United States, Philadelphia: Lippencott, Grambo & Co. Young, O. Jr. (1970) Western Mining, Norman, OK: University of Oklahoma Press.

Part IV

12 The transmission of corporate cultures International Officers in the HSBC Group1 Frank H.H.King When the 1998 first half-year’s results were announced to the HSBC Group’s staff, the headline in their house organ read ‘Strength in our diversity’.2 Writing of the economic downturn now popularly referred to as ‘Asian flu’, the Chairman of HSBC Holdings plc, John Bond, said, ‘While economic difficulties in Asia have affected our business, increased profits from our operations in the Americas and Europe vindicated our strategy of geographical diversification.’ Some few months later, with a world economic domino effect threatening, this comment could be taken without query, but in August an employee or shareholder whose original connection was with or in, say, Britain’s Midland Bank, might well have asked whether the comment was directed at the ‘Group’ as a whole or rather at that portion of it originating in Asia, that is, the Hongkong and Shanghai Banking Corporation Ltd (HongkongBank or HKBL), reflecting indeed the origin of the Chairman.3 The intention of playing with this particular quotation is to note in a perhaps sub-conscious sense that the Hongkong Bank was somehow the real core of the business, that HSBC Holdings is in fact the old Hongkong and Shanghai Banking Corporation writ large, and that the acquisition and/or establishment of banks in, for example, the United States, the United Kingdom, Brazil, and Argentina represented decisions intended to diversify HongkongBank’s position in parallel fashion to their establishment some forty years earlier of that bank’s British Borneo branches!4 Or to express it in the context of this collection, the dynamism of the East enabled a Hong Kong bank, structured in the regional, colonial mould, to break loose and to become a multi-based, multinational financial group. Perhaps a little too dramatically expressed, but that is the sentiment underlying the more qualified comments which follow. The subject of this paper is a case study of the role of corporate culture in the inter-member functioning of a financial conglomerate. The focus is on the origin and role of the International Officer staff (IOs) as executives who, by transcending particular company identification, are able to function on a Group-wide basis and thereby facilitate the working relationship of the various components of the HSBC Group. We must accept the proposition that there have been corporate mergers which have failed, surprisingly, not because of the objective data but because of other incompatibilities, for example, clashes in the corporate cultures. The International Officers are one instrument of several in the moulding of a joint 245

246

Frank H.H.King

trans-corporate culture which has permitted the Group to operate with the success which the corporate balance sheets and the ‘figures’ experts predicted. To write off a ‘corporate culture’ in the singular, especially when referring to HSBC Holdings, is to beg the question. Presumably the holding company might, initially at least, have no particular corporate culture. And indeed this is often the problem: all that exists are the several cultures of the jointly held subsidiary corporations. But, as suggested in the opening paragraphs, the holding company in our case retained the culture of the Hongkong Bank. The initial task of the holding company relative to culture was, therefore, to have knowledge of the ‘other’ cultures and to modify and co-ordinate, but not necessarily to eliminate, the differences, with the assumption that, all other things being equal, the original Hongkong Bank’s culture, having proved successful, should prevail.5 If participating in cultural adjustment has been the purpose of the IO in the initial stage of the holding company’s management, the second phase is a natural progression: the IO will, as he or she is transferred from one assignment to another, bring to each posting something of the positive aspects learned in previous postings in several subsidiaries, in several corporate cultures. Even with the acquisition of the Mercantile Bank in 1959 the changes were not entirely one way. And this particular method of cross-pollination will be supplemented by others. The culture of HSBC Holdings and its subsidiaries is not moving to a replica of China-coast policies, even if HongkongBank itself retains the tradition of curry tiffin on Thursdays; something new must develop.6 But for the moment the IO is part of a group of executives whose origins are directly traceable to HongkongBank and whose role and priorities can be understood in this historical context. Indeed, that is the justification for an economic historian reflecting on the matter at a time of undoubted transition. This brings us closer to the focus of this essay. However, some general historical background is essential, not only of the origins of the IO staff but also of the holding company itself and of the previous cultural peculiarities.7 The path to HSBC Holdings The Hongkong and Shanghai Banking Corporation was founded in Hong Kong in 1865 as a ‘local bank’, with provisions consistent with the Colonial Banking Regulations and with the intention of operating in both Hong Kong and Shanghai. Almost immediately, however, the bank established agencies on the China coast and in the Philippines; it was also involved in exchange operations between London, India, and the Far East. But by the end of the century the bank’s main area was recognized as Japan, China, the Philippines, Vietnam, and the Straits Settlements, with additional operations in Ceylon and Indonesia and exchange with India, San Francisco, and London; there were also agencies in Lyon, France (primarily for the silk trade) and in New York. Although its role in major China loans had given it prominence, by the 1930s the bank was seen once again simply as a British overseas bank, or, to detractors, as a ‘colonial bank’, one of many London and ‘X’ type banks, distinguished only for the fact that its head office remained in Hong Kong.8

Transmission of corporate cultures

247

When China went off silver in 1936 and Hong Kong was on a sterling exchange standard, the bank’s reserves were consequently and fortuitously vested in London; on the eve of the surrender of Hong Kong to the Japanese in 1941 an emergency Order in Council transferred the Hongkong Bank’s head office to London. By 1946 the head office was back in Hong Kong, and the bank was participating first in the recovery of Hong Kong’s economy and its trade relations in the East, and then with the financing of the territory’s first efforts at industrialization. Although by the mid-1950s the bank’s role in China had been restricted to an office in Shanghai, its activities elsewhere in the region prospered. Thus the Hongkong Bank remained a regionally based colonial-type bank at a time when the status and profitability of British colonial banking was threatened, on the one hand by the changed political status of the territories in which such banks operated and on the other by competition from local and USA banks. The Hongkong Bank with its Eastern base had advantages, including the stability and high profitability of Hong Kong itself, the prosperity of Southeast Asia, now to include the developing areas of British Borneo, and the increasing significance of Japan. In a sense it was the bank’s traditional role as a ‘local bank’ that led it first to extra-regional expansion; the need of its local customers to diversify overseas, especially in the United States, suggested that the Bank follow, and, accordingly, the Board of Directors, recognizing that the existing San Francisco and New York agencies could not legally fulfil the necessary tasks, authorized the establishment of the Hongkong and Shanghai Banking Corporation of California as a State bank. At the same time the Hongkong Bank, as a regional bank, faced the threat of American competition through the possible acquisition of a London and ‘X’ type bank, specifically the Mercantile Bank of India by Chase Manhattan Bank. Too small to compete, the Mercantile had been unable to expand significantly from its high tax bases in India, Pakistan, and politically sensitive Sri Lanka. In 1959 the Hongkong Bank acquired Mercantile; the following year, Hongkong Bank, by then becoming a major player in the post-war rationalization of British colonial banking, acquired the British Bank of the Middle East (BBME), now known as ‘BritishBank’, another extra-regional move. There then followed some twenty years of consolidation and reorganization, placing the increasingly successful Hongkong Bank in a position to envisage acquisition of banks in the United States and in the United Kingdom. Giving priority to the former, the Bank, after lengthy negotiations, commercial, regulatory, and political, acquired a majority share in the holding company which held 100 per cent of the shares of New York’s Marine Midland Bank. This acquisition required the Hongkong Bank to sell its California subsidiary. In its first attempt to acquire a United Kingdom bank, the Royal Bank of Scotland, the Hongkong Bank failed. Subsequently it purchased a significant holding of Midland Bank shares with ‘partnership’ in mind; the changing fortunes of Midland led first to a withdrawal of interest and then to a full acquisition of this major British clearing bank. As this is not a complete history, suffice to note that, with changes in the several banking regulations, the Hongkong Bank was able to establish subsidiaries in Australia and Canada, to establish or acquire companies in other financial specializations, and to own service companies subsidiary to these.

248

Frank H.H.King

The special Hong Kong ordinance by which the Hongkong Bank was incorporated required that its Head Office be in the territory. To amend the ordinance, especially after the return of Hong Kong to China had been agreed in 1984, might, even if possible, undermine faith in the bank’s commitment to the Colony, and yet, if the bank were to acquire a major United Kingdom bank, its head office had, for practical political reasons, somehow to be in Britain. There was a further problem. The Hongkong and Shanghai Banking Corporation had become both a bank operating under its own name and a bank holding company with subsidiaries operating under their own names, with greater or lesser public acknowledgement of their association. The solution to both was for the Hongkong and Shanghai Banking Corporation to enable a minor British incorporated overseas company, a subsidiary, HSBC Holdings, to take-over its parent company and then to redistribute the subsidiaries rationally. Thus the Hongkong Bank, with headquarters in Hong Kong, retained, inter alia, the Hang Seng Bank as a subsidiary while Midland Bank became parallel to the Hongkong Bank as a direct subsidiary. At first direction of both the Hongkong Bank and its holding company, HSBC Holdings, remained in Hong Kong, but, with the final acquisition of Midland Bank, the direction of the holding company and thus the ultimate authority for HongkongBank became situate in London. Culture and control Although this essay is too brief to contain a full analysis of the internal corporate culture of the Hongkong and Shanghai Banking Corporation, certain of the main characteristics can be listed. One banking observer commented that the HSBC Group’s corporate culture in 1998 simply reflected the strong personality of the holding company’s chief executive, Sir William Purves. But that is superficial. Purves brought it with him from Hong Kong; he learned it on the China coast. Inherited from the old China-coast bank, the historical characteristics have been modified and adapted but still bear the imprint of the successful regional bank. The first characteristic is particularly relevant to a study of the International Officer staff: the belief that a generalist banker can play a major role in an industry of specialists. Second, and at another level, management has traditionally given priority to a high level of liquidity, to the rapidity of on-line decision making, and the absence of loan committees. The willingness to question has been coupled with the expectation of total compliance, a certain ‘buccaneering’ tradition has been tempered by the acceptance of regulations, and there has been stress on a personal knowledge of colleagues, which again suggests the need for an ‘élite’ corps sufficiently small that such personal knowledge can be maintained. Indeed, before the communications revolution, personal knowledge of all executive members of the staff was the real basis of the chief manager’s (CEO’s) control.9 This last assumed a socialization period for future executive staff in the London office, in the Bachelor’s Mess in Hong Kong’s Peak district, the P&O’s long passage to the East, and the continued stress on sports during years of routine tasks while juniors, apprentice-like, learned their profession.

Transmission of corporate cultures

249

The structural developments described in the previous section have impacted on this corporate culture as the Group evolved over the post-war period in several diverse ways. One might argue first that the special purpose, service-type subsidiaries, with specific tasks of their own, did not need full cultural integration. Their chief executives would only need to know how to deal with the appropriate bank executive; the latter would not expect to be dealing with a banker. As for the specialist financial institutions, their expertise lay outside the experience of the general bankers who staffed HongkongBank, and their often long City traditions carried definite cultures of their own. Their cultural integration would wait while the environment of the world of finance itself changed, and general bankers became competent in, to them, new fields of operation. This decline in corporate specialization was matched by the frequent reorganizations which brought the specialists closer to the parent bank or to the holding company, a trend often illustrated by changes in name designed to link the specialists more obviously to the ‘group’, specifically to the ‘HSBC Group’.10 At any rate, although over-generalized, this must suffice for the present study. These companies set aside, we are left with the main banking subsidiaries existing either by new creation or through acquisition. An example of the former would be the California subsidiary. Here the problem was specific: the culture should have been transferred through the new bank’s top management, seconded from HongkongBank. However the first manager was, through no fault of his own, unqualified; the mixture of HongkongBank executives, local American bankers, and the California scene was inadequate for the task, and the California bank had a difficult history.11 Similarly HongkongBank of Australia had initial problems. One officer, formerly a local employee of HongkongBank’s Singapore office, recalled that the atmosphere, that is, the culture, bore no resemblance to his recollections of his former experience.12 But both the HongkongBank of Australia and the Hongkong Bank of Canada were tempted to take on new business which established banks may well have avoided in the initial efforts to get established in the market. The Canadian subsidiary, however, acquired the troubled Bank of British Columbia, operated in an increasingly Chinese/Hong Kong immigrant environment, tended to understand and retain HongkongBank’s culture, and with appropriate management became the largest foreign bank in the country. The Mercantile Bank and the British Bank of the Middle East had cultures similar to that of HongkongBank itself. The problem in both cases was that the culture was naturally company specific, and HongkongBank had to bring them in. The solution was absorption—joint recruiting, down-grading of the several head offices, and, eventually, removal of head offices from London, where tradition provided a base for independence, to Hong Kong. This was a process of some twenty years. In contrast, with Marine Midland Bank the policy was at first described as ‘partnership’; HongkongBank executives could not run an American bank, it was proclaimed; co-ordination would be insured at the board of directors level. Perhaps the model had been the relationship of HongkongBank with Hang Seng—

250

Frank H.H.King

the British executives could not, did not intend to, run a ‘Chinese bank’. As for Hang Seng, the relationship remained unchanged; there was consequently no transfer of ‘culture’. However, when Marine Midland Bank had renewed problems in the mid-1980s, HongkongBank had to protect its investment, acquired 100 per cent of the shares, and moved in its ‘International Officers’ (see below). ‘Partnership’ in the sense of a holding company permitting a major financial subsidiary virtually equal status, failed as a long-run policy. With the establishment of a separate holding company, that is, of HSBC Holdings, plc, the source of authority and of responsibility became clear. A series of policies designed to bring closer association between the holding company and the principal subsidiaries included rationalization of those subsidiaries, a process involving divesting them of their historical confines and even the names that linked them too closely to a particular niche and/or way of doing business. There are examples where the name actually was retained but with ‘HSBC’ preceding. This process was facilitated by external changes, which revolutionized the financial world in the twenty years since acquisition of Marine Midland, the one single act which linked the fate of the old Hongkong and Shanghai Bank to a multinational future. The evolution of HSBC staff policies and categories In 1941 the Hongkong Bank consisted of 45 branches or agencies and an executive staff of 249 with total assets of £72 million. There were in addition a clerical staff of Portuguese and other local employees and a compradore’s staff of Chinese. This compares with 1998 figures of over 1,000 full branches and 21,000 employees in HSBC Bamerindus (Brazil) alone; the Group staff of 130,000 overall operated 5,500 offices in 79 countries with total assets of £286,391 million. The assets have grown by a factor of over 4,000, a figure which in itself is meaningless. It merely suggests that the problem of retaining the corporate culture is real. Furthermore, a head office in London is contrary to the traditions of the Hongkong Bank. The argument that a head office in the East is essential since that is where decisions must be made, however, at one level remains intact; the head office of HongkongBank (HKBL) is in Hong Kong. At another level the very principal of decision-making suggests the holding company should be in a major centre, London. There was, however, a second objection to a London head office. Sir Vandeleur Grayburn, the Bank’s aggressive chief executive in the 1930s, argued that a strange debilitating disease, Londonitis, turned a dynamic, sound, clear-thinking, hardworking Eastern banker into something spineless and useless. And some would argue today that London is still not Hong Kong with its notorious work ethic. Of all Hongkong Bank executives Willie Purves was aware of the problem—on missions from head office and a brief tour as temporary London manager, he had studied and reported the ‘disease’ first-hand. Then too, London has itself changed and, with it, the role of London office. Yet here is another problem in the retention of the corporate culture. HSBC Holdings has several facilities and policies which encourage crossfertilization, which permit an integration of corporate cultures into an apparently

Transmission of corporate cultures

251

workable whole. Some parallel earlier policies, but most are new. Gone are the years of socialization in London office and New Beckenham playing fields—highcost young executives cannot waste time to be replaced by a short London stay for International Officers, by in-career training, and Outward Bound sessions. Gone is the personal knowledge senior management ideally (but not always in practice) had of junior staff, replaced by more complete personnel policies, career planning, meetings during training courses, and the Group’s training school at Bricket Wood.13 In addition the Group has policies which encourage temporary assignments in some other member company; the Group has its Archives to provide historical backgrounds and summaries, its Public Affairs Department to provide internal Group information, focusing on the contributions in business, in sports, drama, and in public affairs based on the reports of correspondents in the several subsidiaries and in all regions. All this was supplemented, especially after 1980, by exercises in ‘corporate identity’. Each of these deserves separate consideration, but in this essay we must turn to the International Officers, their identity and their role. In mid-1998 there were 330 International Officers with some 50 more so designated on a temporary basis.14 These are the direct successors to the 249 executives who managed the Hongkong Bank in 1941, but their role has developed and changed, and the evolution is important. When that is stated, and the focus on Hongkong Bank executives understood, their present role can be better placed in perspective. For the International Officer is a heritage of the Hongkong and Shanghai Banking Corporation (with the Mercantile Bank and the BBME), of the overseas, expatriate tradition. There was nothing comparable in the early history of other Group members. The staff structure of the pre-World War II bank Having been founded in Hong Kong on the basis of a local decision made by a local provisional committee, the new Hongkong Bank had to employ those available in Hong Kong and the Treaty Ports, using merchants as agents in the outports. But experience had shown that such random hiring could lead to difficulties; it would also involve the employment of non-British Europeans. Although the bank was particularly successful in its initial selection, a more cohesive, dedicated staff was immediately seen as desirable. Thus in the second stage, London office was assigned the task of recruiting ‘juniors’, preferably with some experience, who would train in London until required in the East, replacing the randomly employed as these were discharged or retired. The juniors performed simple clerical tasks in London and during their early assignments in the East, but they were, from the first, recruited as career executives.15 Indeed, if lucky a junior might be given a special assignment or be otherwise noticed. They and their earlier and more senior colleagues, including the Chief Manager (or CEO) constituted the variously named ‘British Staff’ or ‘European Staff or, singularly, ‘Foreign Staff’, and notices declared for the information of customers that only they could sign for the Hongkong Bank.16 One should further characterize this staff by noting that (i) with rare exceptions they came through London, (ii) again with rare exceptions—three come to mind—they were British,

252

Frank H.H.King

(iii) they were generalists, but de facto might develop a speciality, even an expertise, for example foreign exchange dealing or knowledge of Chinese, Malay, or Indian dialect; they might, beginning in the 1930s, be seen as ‘figures’ men and assigned to an embryo head office team, (iv) there were no lateral transfers from or to other banks, and (v) barring some exceptional circumstance they remained throughout their working life, retiring from the East in their early 50s. Not all became great bankers, but eventually even the least adapted could be given the managership of a small agency, one for which major decisions were made by the nearest branch— Singapore, for example, made the decisions for Malacca. Post-war these arrangements would have to change, if only because the bank could not afford the luxury of juniors performing simple clerical tasks nor could it afford to retain a highly paid officer in anything but an important post. The role of the generalist would also come under scrutiny. As already noted, the Foreign Staff were supported by local employees. The system was three-tiered: Foreign Staff were officers and could sign for the bank, the Portuguese (or Burghers in Ceylon, or other local staff) were the clerks, and the Chinese staff, employed and vetted by the compradore, performed tasks ranging from the handling of cash (the shroffs) to menial work—and in Shanghai there were Chinese clerks, but on teams separate from the Portuguese. Post-war adjustments Post-war the sharp differentiation between Foreign Staff who were officers authorized to sign for the bank and all others could not survive. The casual on-the-job training of executives became impractical for both financial and professional reasons. Many Portuguese emigrated and the compradoric system became as unacceptable at the practical level as it was at the political level. The corps of officers had to be expanded; it had then to be differentiated and rationalized, at the end of which process the International Officer staff survived, one might say, as a distillate. The arrival of the specialist There had been pre-war specialists. Sir Charles Addis rejected a proposal to make him a Chinese-language specialist; instead he trained himself, as it were on the job, as a China policy adviser and investment banker. But he had no successors from within the bank. A political adviser was seconded from the Foreign Office, but he was not considered ‘one of the team’, was not integrated, and reported his special assignments directly to the Chief Manager. The changes post-war at all levels of operation and the need for personnel training, the introduction of computers, personal finance, public relations, and many others, required special training. A member of the Foreign Staff, a generalist, hopefully one who showed some interest in the subject, was sent off. On his return he introduced his new specialty, convinced the Chief Manager, and developed a department. But at some crucial point he would reach his limit. There had been too many new developments; he could initiate but he could not keep up as the tasks

Transmission of corporate cultures

253

and the bank expanded. The generalist as specialist was seen as an amateur; he was returned to general banking and, let us say, a computer expert was employed. And he had to have officer status—the erosion of the ‘élite’ International (vice ’Foreign‘) Staff was beginning. Certain specialties, however, required a separate subsidiary company. This might be staffed, at least initially, by a member of the bank’s foreign staff, facilitating a cultural transfer, especially if the subsidiary were financial. And yet if an International Staff member remained as chief executive, there would be a further strain on the bank’s officer corps. Steps to an expanded officer corps To relieve the strain on the bank’s small officer base, the first step could be seen as the designation of long-serving clerks as officers in the limited sense that they could now sign, committing the bank, to work they previously initialed and passed to a member of the Foreign Staff to sign. In a process beginning in the mid-1950s and not completed for some fifteen years, these local officers eventually were assigned tasks previously the sole responsibility of the Foreign Staff. Rather than merely saving the time of an FO, the local officer was performing his task. Once this point had been reached, natural evolution resulted in the eventual appointment of a local officer to a managerial position, making possible the great expansion of local branches in the late 1960s. Nor was the appointment as local officer any longer confined to senior members of the Portuguese staff; the ending of the compradoric system and the direct hiring of local staff gave the bank the control needed, and Chinese officers, male and female, made possible the changing role of the bank within the communities it served. As for numbers, the officer corps was further enlarged not only by the specialists but also by Foreign Staff of the Mercantile Bank and the BBME as they became integrated into a single service. Rationalization and demarcation With an expanded officer corps of highly varied origins and status (corporate) cultural factors could not be neglected. For the local recruit or employee appointed to officer status there was the need to acculturize him (or her). Initially, the newly appointed officer was actually sent to London for ‘training’, but London was no longer the London of P.G.Wodehouse or even of the 1930s.17 (Indeed nostalgia led post-war chief executives to retain for at least the first passage East the opportunity to go by sea. The dubious socializing role of the Bachelors Mess in Hong Kong’s Peak District survived into the early 1990s!) The Mercantile Bank had sent its local officers to the formal training school established in Britain and still functioning in the early 1960s; those on graduation were, predictably, often more British than the British, but the school was closed when direct recruitment into Mercantile ceased, and the Hongkong Bank’s training had been in the actual London office. But all this missed the point. If the Hongkong Bank’s traditional corporate culture were to be maintained, it could not be done by attempting to replicate pre-

254

Frank H.H.King

war practices which had, in any case, been replaced.18 It would prove sufficient to assign the newly appointed local officer to an office either in another country or at some distance from his home base. There was always a distinction between British expatriate officers and locally appointed officers, the former being available for assignment throughout the area in which the Hongkong Bank functioned, the latter being subject to a contract which limited their assignment to an office, a country, or perhaps a region—they were ‘regional officers’. This basic difference was reinforced by the realities on the ground. Although the Bank could assign the International Staff to any office without restriction in the colonial and immediate post-colonial periods, there would by the mid-1960s be national policies which severely limited the number of expatriates granted work permits. On the other hand, perhaps a great majority of local officers did not (or their spouses did not) desire a peripatetic existence. Furthermore, while a newly independent country would accept a limited number of ‘European’ expatriates, the likely reception of overseas Chinese, Arabs, or Indians, for example, was then unknown but in any case seemed at odds with the concept of the Hongkong Bank as a ‘British overseas’ bank. All this would change or be modified, but for the roughly twenty years between 1965 and 1985 there had to be experimentation. Once again the corporate culture was involved. The bank, pressured by the logic of its own success, had to consider itself not so much British as multi-national. This in turn brought into question the exclusively British, male composition of the expatriate International Staff. Another cultural challenge resulted from the working permit restrictions. The bank could not afford to select any but mid-career or senior expatriates for the permits, resulting in juniors being barred from retail, behindthe-counter experience, a fact particularly regretted by one IO interviewed. Only in Hong Kong, where no such restrictions existed, was the young expatriate officer able to gain a variety of experience. As for the country-bound regional officer, he suffered from lack of overseas experience at a time when banking was becoming more international. The regional officer could, however, be sent overseas for training or for temporary assignment in the International Staff. There was cross-fertilization. With the acquisition of Marine Midland in 1980 the division of staff would approximate the following: (i)

International Officers—generalists, deriving from the Foreign Staffs of the Hongkong Bank, the Mercantile Bank, and the BBME, plus those recruited directly into the Group—that is, de facto, into the Hongkong Bank for service interchangeably in HongkongBank, Mercantile or BBME; (ii) Specialists, many on overseas terms of service but whose assignments in the three banks would depend on the requirements of their speciality—thus a China specialist might be transferred from his head office in Hong Kong, but only within the terms of a contract; (iii) Regional officers and other local officers, who were recruited for service in a particular location or area;

Transmission of corporate cultures

255

(iv) executives and officers of Hang Seng Bank; (v) executives and officers of the Marine Midland, all of whom would remain with Marine Midland; (vi) officers of various subsidiaries, who would normally remain on a career basis with their initial organization or with a similar subsidiary. For the purposes of this essay we may ignore further consideration of the executives of Hang Seng and of the officers of specialist subsidiaries. The International Officers The international staff and the former ‘foreign staff compared During the immediate post-war period, several bank branches suffered labour disputes and, in several cases, there were strikes. These disputes were, be it noted, with local employees, and had a different cause and purpose in each location. Those branches were nevertheless able to operate through the strikes, however, because the Foreign Staff could, in a sense, do everything themselves. In one Japanese strike, the bank flew in additional foreign staff, who, remaining in their hotels, undertook the clerical work brought to them ‘secretly’. In the mid-1980s an international officer might claim to be a generalist, correctly, but he would most probably fail to function in any strike-bound, hi-tech office operating in markets requiring some specialist input. In today’s banking industry even the endangered generalist is not that general. When the Foreign Staff was renamed ‘International Staff’, an officer moved directly from one to the other, the latter was the unquestioned successor of the former. But more than the name had changed; more than the name would change. Recruitment and selection Personnel policy had not been a strong point in the pre-war Hongkong Bank, and in this the bank was not alone.19 Once the process of London recruiting had been developed in the years between say 1868 and 1880, managers overseas accepted the new juniors (with only the most unsuitable weeded out—Wodehouse, for example) on the basis of seniority just as they came and, over periods as long as ten years, waited for, as one manager put it, the cream to rise to the top. Those who lacked ability could, as noted before, be shunted to harmless managerial positions in sleepy outports. They were not, on the whole, highly paid, the salary of the Chief Executive being, as a 1954 recruiting brochure put it, ‘in a class by itself’.20 Pre-war the London manager would interview a young applicant, a youth under 21 with public school, preferably some experience, or Scottish background, and ask questions appropriate to a very basic situation: what do you play? why do you want to join the bank? And, despite efforts by the London manager, Sir Charles Addis, in the early part of the century, there was no continuing effort to recruit university graduates. Surprisingly, this approach continued virtually to the eve of

256

Frank H.H.King

the Hongkong Bank’s negotiations with Marine Midland and the review of the bank’s organization by the Federal Reserve. Perhaps this can be excused. The bank had proved itself able to compete with bankers boasting degrees in business administration; the bank’s manager for Japan, for example, was at one time the only foreign banker in Tokyo with no university experience—but his father had been a planter in Malaya. Obviously the early post-war recruits with their military experience, with the challenges of the developing world and the continued ‘romance’ of banking ‘east of Suez’ brought something else to their careers—a spirit consistent with the traditional culture of the Hongkong Bank. And yet, not at all surprisingly, when the Bank’s personnel policy was professionalized, its approach was formalized and its standards raised. For with training in London virtually eliminated, with the high initial salaries, with the immediate professional demands, and with the now universal expectations of a university degree, the bank could and did ask interview questions at a graduate level and sought entrants, who, while still expected to be general bankers, could bring some expertise to even their first assignment. Or, in brief, the International Officer staff was from the first more professionally oriented. A concomitant of this up-grading was the review of the then employed foreign staff in mid-career and the termination of several appointments. There were no easy berths, and the simpler tasks could be undertaken more cost effectively by junior ranking local officers. In consequence the Foreign/International Officer staff was no longer a cross-section from future chief executive to Manager, Malacca, from junior behind the counter to senior manager advising Chinese mandarins or colonial governors. The former foreign staff had been ‘élite’ only in the sense that they alone were ‘officers’ of the corporation and able to sign in the name of the bank. The new international officer staff retained the heritage and, to the extent relevant, absorbed the culture, but to all this they brought a new dimension. A more recent scaling down of the International Officer staff was consequent to (i) the high cost of an IO relative to a regional officer performing the same task— even to a factor of three, and (ii) the realization that, with large, highly qualified regional or country-bound staffs, many tasks undertaken by an IO could be equally efficiently performed either by a regional officer or by an IO re-designated, even if this involved overseas terms of service.21 Thus in the past decade the number of IOs has fallen from 550 to the 380; even this figure overstates the number of ‘pure’ IOs, those permanently in that status, probably 330 in mid-1998.22 Now like the old foreign staff, they were a small group, all known to the most senior management, all intending a long-term commitment. Their careers could be overseen by the personnel officers in HSBC Holdings, plc; they belonged to the Group, not to any single banking subsidiary. Further developments and characteristics By the 1980s the Hongkong Bank, both bank and holding company, was by any standards a multinational. The International Officer staff had to reflect this; it had also to reflect current politically correct standards, especially where PC, once

Transmission of corporate cultures

257

traditional prejudice was overcome, was also cost effective. The Hongkong Bank pre-World War I had been characterized as a British overseas bank, staffed by British officers, controlled by a multinational board of directors, and catering to a multi-racial, multi-communal constituency.23 Although there were French, German, Japanese banks active in the Far East, none could boast the same multinational credentials, despite the Hongkong Bank’s close association with the British and Hong Kong governments. The Hongkong Bank had become a holding company with foreign registered subsidiary banks; subsequently, HSBC Holdings gathered an even broader sweep of companies, and the ‘British managed’ claim was no longer valid. These developments had to be recognized in the composition of the International Staff. In mid-1980, although 75 per cent of those newly recruited into the International Staff are British citizens, there are 25 per cent who include Commonwealth citizens, Americans, citizens of South Asian countries, Malaysians, and Hong Kong originating Chinese. The number is supplemented by an equal number of regional officers who apply for and are accepted on permanent transfer or on secondment for an agreed period of years.24 There are, however, regional officers who hesitate to sacrifice the real opportunities in their own country for a life apparently more acceptable to ‘Europeans’ experienced in the traditions of overseas service and the expatriate life. One cultural feature of the HSBC Group has been the tendency to employ, at all levels, with the expectation of long service. For a promising young banker in a single office country the long-term prospects are not bright. The opportunity for other assignments in the Group may be attractive. At the other extreme, in a country such as India where the Group is attracting candidates from, for example, the Alhamabad Indian School of Banking, service outside the country serves both the personal needs of the Indian bank officer and the priorities of the Group. The last hurdle was the inclusion of women as International Officers. The bank objected, or pretended to object, that women would not be accepted in Muslim territories and therefore…. It was never clear where this argument, even if based on fact, was leading. Side-stepped while the bank was negotiating for the acquisition of Midland Bank, the issue was eventually faced. The role of the International Officer The International Officer is not designed as purveyor of old-time Hongkong Bank culture. However, from the very nature of his career he plays a role in furthering inter-cultural understanding within the Group, and from his background or the background of his seniors in the IO staff he brings a touch of HongkongBank however much transformed in detail. The essential element in management of a multinational, multi-corporate financial group is to have knowledge of these historically developed differences, which the balance sheets and more formal reports do not always reflect. In this the IO is one of many agents. There are secondments, Group-wide training courses, joint operations, circulated newsletters and reports. And yet the role of the IO is unique and warrants the consideration given both by HSBC Holdings management and by outside observers and students.

258

Frank H.H.King

Uniqueness and structural problems There are other large international banking groups with officers qualified to work overseas, but I have no evidence to indicate they form a distinct and long-term staff group within the organization. To the extent this is confirmed, the study of HSBC Holdings use of the IOs is of special importance. Although the IOs are intended to retain the status held by the earlier ‘Foreign Staff’ as an élite, this arises not from a unique right to sign for the firm but on the more significant factors, several of which have been noted earlier: (i)

they remain generalists but receive continued training including specialist training and experience; (ii) they usually are assigned to established posts of varying character and responsibility, but they can be moved, transferred by management without recourse; (iii) they are available as trouble-shooters; (iv) consequently they are justified in supposing that their names are on the hypothetical list from which senior managerial officers will usually be selected. The IO corps is sufficiently small for individual careers to be subject to holding company management; IOs travel more frequently and are familiar with each other. Their knowledge of other IOs and their whereabouts leads to direct contacts on business matters which can bypass the hierarchies in the initial stages, thus speeding the decision-making process. One indirect consequence of this is that personnel officers in the several subsidiaries can focus on their own officers, who are thus better managed, and can recommend them for transfer or secondment. The essential factor is international experience for an international financial group. The IO’s higher cost arises in part from his expatriate terms of service, which assume the officer’s continued overseas service and his immediate availability for transfer.25 Not surprisingly, these costs include education and travel expenses of dependants, payment of taxes, provision of housing, etc. The concept of an officer corps the members of which can be assigned overseas at the will of HSBC Holdings seems to run counter to the impact of work permit/visa problems. Hong Kong was an exception, and the easier regulations plus the large presence of the Hongkong Bank and the Group in that territory meant that an IO was often, de facto, based in the territory. But this would not meet the requirements of a Group-wide IO corps. Indeed, the problem of obtaining work permits is one involving constant negotiations between the bank and the authorities. The restrictions were usually initiated by governments to force the pace of localization; this has been achieved, but, in any case, the high cost of expatriates, once qualified locals became available, meant the foreign firms had less motivation for resisting. An international financial group, however, needs bankers with international experience; a multinational is not, in this, pressing for permits for officers of any one nationality and a quid pro quo can be argued. It is possible to point out to the Malayan authorities for example, that Malaysian nationals have been appointed as IOs and that the bank has been able

Transmission of corporate cultures

259

to obtain a working permit for one such in, say, Thailand. Alternatively, it may be argued that the assignment of an IO in, say, Brazil, will bring an expert into the country with the latest skills, which should improve local banking and which could be made available to the Central Bank. Generalists and Hongkong Bank culture In the nineteenth century the Hongkong Bank, with its major branches in Hong Kong, Shanghai, Singapore, and Yokohama, was described as a collection of banks trading with a common capital. This depended upon a particular (and very narrow) definition of ‘bank’ and did not touch on the legal corporate realities. But it did highlight the high level of autonomy which, per force, the manager of each of these branches possessed. His resources depended on the common capital assigned him and on the deposits he garnered in his own branch and in the agencies subject to his policy decisions. His decisions could not be made subject to review before the fact; they could be reviewed and assessed on the basis of the figures reported to the Chief Manager in Hong Kong, and there was one travelling inspector. Offices on the route of the P&O steamer from London to Hong Kong could be visited by the Chief Manager en route. A branch manager was, then, actually in charge of a varied range of activities coming under the heading ‘banking’, and his juniors had to be prepared in each of these fields—to a point. Post-World War II a young banker might, as noted, be assigned to follow a new area of banking to the point that he became a specialist. Again postwar, the decisions of the branch officers made within specified limits would be reviewed after the fact; decisions involving funds beyond those limits would be referred to a head office for on-line approval. But the traditional concept remained: the banker in the branch made the decision, subject to fast (sometimes overnight) approval when limits were exceeded. The customer dealt with the branch; its manager was part of the process. The customer was not dealing with a specialist agent sent to the branch by head office to represent a particular head office department. There was no micro-management and the branch manager was not relegated to the role of administrative custodian. This tradition has remained in the region in which the present Hongkong Bank, Hongkong Bank Malaysia Berhad, and BritishBank (BBME) operate. Regional officers can, to the extent not country-bound, be reassigned to meet requirements, but the International Officer corps has officers who, as generalists, could be moved in to meet a routine need or to face up to a crisis. Alternatively a regional officer might be seconded to the IO corps before being assigned to the overseas branch. In either case he would act, once in place, as an officer of that branch. He would, however, have been briefed, he might bring a message, and he would have a network of IOs with whom he could consult informally. This latter ability is not without criticism. When IOs were assigned to top Midland Bank positions, some Midland Bank veterans were relieved, seeing their arrival as the end of a long period of policy uncertainty, an end to a ‘policy of the month’ approach. Others saw the IOs as ‘white WOGs’ and resented the inside channels available to these favourite sons.26

260

Frank H.H.King

Is a generalise career in banking still viable in today’s fast changing financial world? The specialist is often dubious. He may have submitted a report only to have it questioned by a generalise To others this is the strength of the system. But an IO interviewed for this essay confessed that competition forced a degree of specialization, and that this, once admitted, the retention of general skills was, although justified, extremely difficult.27 Another commented that being a generalist was difficult in an industry in which every task was a specialty or which involved competing with specialists. On the other hand, it is difficult to argue with success, and the focus of discussion today has moved on. At risk is the very role of banks in a financial environment which appears to encourage disintermediation. HSBC Holdings intends to be prepared. International officers, case studies The most dramatic interventions by IOs have been in senior management positions with Marine Midland Bank and, later, with Midland Bank in the United Kingdom. These might be observed in the decade from the mid-1980s and, particularly in the case of Marine Midland, the IOs involved, Keith Whitson for example, had been recruited to the foreign staff of the Hongkong and Shanghai Banking Corporation and were on the ‘through train’ to the new IO staff. Keith Whitson and Richard Orgill were both assigned to Midland Bank. They were illustrative of the dynamism of East Asia, the products of the Hong Kong experience. One could argue that these are not sound examples. Although the Hongkong Bank had promised ‘partnership’ with Marine Midland, it was obvious that if anything went wrong Hongkong Bank would have to interfere with the management of its subsidiary, with or without something called an International Officer staff. Furthermore the senior executive so parachuted in would expect to bring some of his own ‘people’ along. This is a fair criticism, but the fact is there were IOs and they brought with them a particular culture, that of the Hongkong Bank. Both with Marine Midland and with the Midland the intervention of IOs brought, additionally to any other benefit, something of HongkongBank’s traditional corporate culture. With Midland Bank the exchange was not entirely one way. At least one former executive had resigned to become an IO with HongkongBank and had been sent back to London.28 Another major IO operation came with the acquisition by HSBC Holdings of Bamerindus (now HSBC Bamerindus) of Brazil. In the agreement for the takeover HSBC Holdings returned 70 per cent of the assets to the central bank; there were problems. At this point 20 IOs plus 20 seconded officers (of whom 2 spoke Portuguese and some 12 spoke Spanish) were brought in to effect a turnaround. Other less dramatic operations involved (i) the request for a bank with offices in Brunei which was capable of effecting the transfer of all savings accounts from the Brunei State Bank to themselves virtually over a weekend. Only HSBC Holdings had the large number of officers needed immediately, if for a short period, for immediate assignment in the Borneo sultanate. Then (ii) when HSBC Holdings decided to open in Kazakhstan, an IO was assigned the task—he was available

Transmission of corporate cultures

261

without recourse on an emergency basis. There are bankers who would not accept an assignment in Kazakhstan; the IO had no choice. But (iii) there are Midland bankers who did not want to go even to Aberdeen, a branch without a manager. To prove a point, HSBC Holdings moved in an IO. In Malaysia (iv) the Hongkong Bank had been moving in traditional lines, arguing for continuity within a range of central bank restrictions. An IO was transferred into a planning position, and brought with him a message from Head Office. A British-educated Malaysian Chinese himself, this officer was facing, with apparent success, a very real test of acceptance. Earlier in Sri Lanka (v) the Hongkong Bank had proudly announced that a local regional officer had become the officer in charge of Group operations in the country. Unfortunately he was a Tamil on the eve of serious disturbances and, when he considered it sensible to leave the country, an IO took over.

*

*

*

After this random account of specific interventions I should remind readers that the normal role of an IO is one of assignment to a key position consistent with his experience and to which he brings the international background, as interpreted in the context of the corporate culture, from one subsidiary to another or from one region/country to another. For both emergency and routine the careful posting of a small corps of International Officers is key to the control by the holding company of the HSBC Group. The corporate culture which these officers pass on is that of the old Hongkong Bank, modified by the parallel cultures of Mercantile and BBME, modified further by external changes and by influences from the non-integrated banks, including Marine Midland and Midland. In time the impact of London, the distant memories of the days of Treaty Ports and dynamic Hong Kong will fade as specific influences, subsumed in so much that must be new. Nevertheless, the role of a unique International Officer corps is not dependent on the specific content of their message, provided that the banking culture the members disseminate is current and sound. For the present, however, this story belongs in a collection describing the dynamism of the East. Or, put another way, this story is vindication of the proposition that the history of the corporation not only matters but must be studied and taken seriously; a corporation’s history may well set the context in which the officers of that corporation operate, affecting the efficiency of their decisions. Perhaps more immediately the Hongkong Bank’s history and the development of its corporate culture answer the question of how this particular regional, even ‘colonial’ bank became the nucleus of a multinational financial conglomerate. That is to say, why the Hongkong Bank and not the British Bank of the Middle East, or even, why the Hongkong Bank and not Marine Midland? Traditions which had been established to assure survival in another financial era, when redesigned and restructured, made possible the extraordinary development of a corporation fired by the dynamism of the East; there was a great leap forward.29 But this essay does not necessarily provide a blueprint for the future, nor is it a

262

Frank H.H.King

‘how to succeed in multinational banking’ text. Although the Bank’s history has shaped its culture and in turn its culture has encouraged its success, there is a new question: can history at some point nevertheless become a burden and, if so, can or should it be forgotten? Very recent developments have suggested an impatience with the past; marketing a maze of products from an apparent maze of what seems to outsiders as unrelated companies, albeit ones which carry some sign of the Group very small on their stationery, is said to be difficult for both the member banks of the Group and for the potential customers.30 Students of management may find, in the history of the Hongkong Bank, useful ideas for consideration, particularly, for example, in the basic discussion relative to the use of generalists in banking today. But any specific feature the management consultant recommended borrowing would have to be tested against and adapted for a different corporate culture. These are not interchangeable parts. All this is, indeed, matter for a further essay, one which an observer of the postwar bank may be disqualified from writing—disqualified perhaps by a tendency to sentimentality, even nostalgia, traits which, in the research for this paper proved essential. Notes 1 This is a fully revised version of a paper presented in the seminar entitled ‘Asian Pacific Dynamism, 1550–2000’ at the Twelfth International Economic History Congress, Madrid, Spain, August 1998. I am indebted to Edwin Green, Archivist, and Michael Broadbent, Director Group Corporate Affairs, for providing me both information and the facilities for conducting interviews. I am also grateful to the managers in Hong Kong, Kuala Lumpur, and Singapore both for their time and for recommending International Officers for interview. 2 HSBC Group News (London), August 1998, interim results special issue. 3 The name of the founding bank of the Group is The Hongkong and Shanghai Banking Corporation’, abbreviated by the initials ‘HSBC’. This series of letters has now become an acronym, a word in its own right, and is so used in this essay. ‘HSBC’, the acronym, is also used as part of the legal name of a subsidiary to identify it as a member of the Group, see note 4 below. The use of ‘Group’ is for ease of reference to the collection of companies including HSBC Holdings and its subsidiaries, whether or not that particular subsidiary uses the initials in its legal name. Thus to avoid confusion the original bank, HSBC itself, has been redesignated ‘HKBL’. 4 In the USA, Marine Midland Bank; in the UK, Midland Bank; in Brazil, HSBC Bamerindus; in Argentina, HSBC Banco Roberts. The two ‘Midland’s are a coincidence! 5 In my article, ‘Does the corporation’s history matter?…’ in Andrew Godley and Oliver M.Westall (eds) Business History and Business Culture, p. 117, I quoted International Business Week with reference to HSBC Holdings Chairman (officially known as ‘Group Chairman’), Sir William Purves, supporting the proposition that he was remaking the group in the image of the old Hongkong Bank and that the basis of the historical culture was intact. 6 This is a reference to an article in the Financial Times espousing the cause of Lloyds’ Bank in their bid to acquire Midland Bank. See my ‘Does the corporation’s history matter?’, p. 137, n. 5. Those following this argument on colonialism will wish to know that curry is now served at the Group’s Bricket Wood training centre. See note 13.

Transmission of corporate cultures

7

8

9 10 11 12

13

14

15

16

17

18

19

20 21

263

Background for the present article may be found in my The Hongkong Bank in the period of development and nationalism, 1941–1984 (Cambridge 1991), cited hereafter as being Vol. IV of The history of the Hongkong and Shanghai Banking Corporation. For definitions of corporate culture as used in this article, see my ‘Does the corporation’s history matter? HongkongBank/HSBC Holdings: a case study’, in Godley and Westall, Business History and Business Culture (Manchester 1996), pp. 116–37. The ‘X’ refers to the region being served, as, for example, the Chartered Bank of India, Australia and China. For a discussion of this type of bank, see my ‘Structural alternatives and constraints in the evolution of exchange banking’, in Geoffrey Jones (ed.) Banks as Multinationals (London 1990), pp. 85–98. See my article, ‘Does the corporation’s history matter?…’ for a fuller discussion. The ‘HSBC Group’ is a description, not a legal term. It encompasses the holding company and its subsidiary and associate companies. King, The history, IV, p. 492. References to comments by bank officers are consequent to a series of interviews undertaken over the past three years with the kind co-operation of key executives in Group Public Affairs, Group Archives, branch managers and the officers themselves. By pure coincidence, the main house at Bricket Wood has associations with the Mercantile Bank and with India. The paintings recall the Group’s Eastern heritage. On the role of curry, see note 6. These figures and other current material were provided me by several officials of HSBC Holdings, plc, in August 1998. I wish to express my appreciation for their courtesy and especially to Edwin Green and his colleagues in the Archives for arranging the interviews. I am equally indebted to Michael Broadbent, Director, Group Corporate Affairs, but newly arrived from Hong Kong. The juniors’ mess in the Colombo Office was transformed, significantly, into the first computer room for the computers that completed in seconds, tasks over which they laboured for hours. In the 1980s the term ‘International’ was substituted for ‘Foreign’, see below. The term ‘European’ in this context is usually a euphemism for any white man with ancestors originating west of Suez, including North America. Wodehouse was briefly a junior in the London Office of the Hongkong Bank. See his ‘My banking career’, in The Hongkong Bank Group Magazine, No. 6 (summer 1975), pp. 13–16; or see the fictionalized (but accurate when dealing with the London Office) version in Psmith in the City (London 1910, Penguin edition 1970). Extracts are quoted in King, History, II, pp. 178–81. One cannot help but make a parallel with the authentically reconstructed VeniceOrient Express. The hardware is without blemish and the ride great fun, but the prewar atmosphere cannot be replicated; the statesmen and secret agents have been replaced by fellow tourists. The Hongkong Bank’s personnel policies reflected those of major hongs on the China coast. Equally questionable, for example, were the actions of Sir Robert Hart, Inspector-General of the Imperial Maritime Customs; see Paul King [no relation], In the Chinese Customs Service: a personal record of forty-seven years (London 1924). Quoted in King, History, IV, p. 321. An example of redesignating an IO is R.Tennant, the present (mid-1998) General Manager, Group Human Resources. Originally an executive in the Midland Bank, he was seconded to work in Hongkong Bank, Hong Kong. Given top management problems in Midland, Tennant resigned and joined Hongkong Bank as an IO. He was later transferred to London to head personnel, a position which did not require IO status; he was accordingly reclassified.

264

22 23

24 25

26

27

28 29

30

Frank H.H.King

These figures are based on material provided during interviews held in HSBC Holdings, August 1998. Between its founding in 1865 and 1914, there had been Parsees, Jews of Middle Eastern origins, Americans, and Germans and a Norwegian on the Provisional Committee and/or the Board of Directors of the Hongkong and Shanghai Banking Corporation. There had even been a Swiss and a French chief executive, but these were anomalies; management and staff were British—additional exceptions: one German with close English connections, one American originally recruited as a local employee in the New York agency. Those on secondment may constitute some 50 per cent of the total number of IO’s. Source: August 1998, interviews. If an IO from say Malaysia is, for example, assigned to a position in Malaysia, he remains on expatriate terms of service—assuming that the assignment was made as an IO, that there is no thought of changing his status, and that it is expected he will subsequently have another overseas assignment. In the sincere hope that some readers will not know the meaning of ‘WOG’, I provide the explanation: ‘worthy Oriental gentleman’. And it used to be said that, for some Britons, WOGs began at Calais, a former English possession on the ‘other’, i.e. wrong, side of the British Channel. Charles Addis confronted, at a key point in his career, with the choice between becoming the bank’s China specialist (with linguistic skills) or a general banker, decided on the latter. A bank China specialist on hearing this story expressed surprise. ‘I made that decision when I applied for a position with the bank,’ he said. This is correct; today the decision is made up-front. Being a generalist in today’s banking world is a challenge which may or may not appeal. See note 21. The Hongkong Bank, for example, had had a New York agency since 1875, as the Bank’s representative pointed out at Congressional hearings. But this was countered by the comment ‘most of us had not even heard of you’. The implication being that, quibbles about agencies aside, here was a bank few had heard of applying to buy America’s thirteenth largest bank. Something had ‘clicked’; the culture developed and shaped by the Bank’s history suddenly became not an anachronism but the key to its successful growth. It does not necessarily follow that the same holds true today. See King, The History… IV, p. 841. This refers to plans for all constituent members of the group to be referred to as HSBC Bank, thus HSBC Holdings would join the ranks of long-lived corporations who find the association of their historical name a hindrance, even an embarrassment, giving the wrong signal to the potential customer. ‘CP’ for ‘Canadian Pacific’ comes to mind. The Hongkong and Shanghai Banking Corporation in establishing its California subsidiary took the opposite position, stubbornly insisting on the name ‘The Hongkong and Shanghai Banking Corporation of California, Incorporated’ (‘ink’ for short), despite the obvious consequence that in the 1950s it would be seen as Chinese and Communist and, therefore, bad.

13 Chaos and instability The Asia Pacific rice trade in the 1990s A.J.H.Latham

China, Indonesia, the Philippines, India and Bangladesh consume some 70 per cent of world rice production, most of it long grain (FAO 1991, p. 2). To them must be added Korea, Taiwan and Japan, who consume short grain rice. There are also the Pacific countries of South America. The basic diet in Peru is rice and potatoes, and Chile too is an important rice consumer, as is Mexico. Apart from their own harvest, these countries have to import rice and most of their supplies come from other parts of America: Argentina, Uruguay and the United States. Brazil, the largest rice consumer in America, also obtains supplies from the same sources (Grain 1996 October, pp. 10–16; December, pp. 13–15; 1997 March, pp. 8–10; Latham 1998, pp. 88–92). That South America is rice-dependent is surprising because the rices she consumes are not indigenous. In exchange for giving maize and potatoes to the world she obtained rice. The fact that both South America and East Asia are rice-dependent has important economic implications for the Pacific region. Up to the Pacific War of 1941 the three great world rice exporters were Burma (Myanmar), Indo-China (Vietnam) and Siam (Thailand). After the Pacific War both Burma and Vietnam declined as exporters for political reasons, but the United States emerged as a major exporter to fill the gap. She does grow rice in Pacific California, the rice being largely short grain of the kind preferred in Korea and Japan, but this rice is mainly consumed in the United States. Her main export supplies come from the Southern States, mainly Arkansas, and also Louisiana and Texas. These are not Pacific areas but can easily supply the Pacific via the Panama canal (Grain 1996 November, pp. 9–15). Nearly all the rice now milled in Mexico comes from the United States, and 25–30 per cent of rice milled in Central America comes from the USA. The USA also exports heavily to the rest of South America, including Peru and Brazil. The main reason why Latin America has become the main export market for US rice is that the USA exports rice in the husk, which is what Latin American millers seek. However she does export milled rice, and has exported to Indonesia, Japan and China (Grain 1997 March, pp. 8–10). In the 1990s the international rice trade has seen some chaos. Table 13.1 shows world rice exports. The first point to note is that the rice trade more than doubled in this period from 11,661 metric tons in 1990 to 27,428 metric tons in 1998. Thailand is the predominant exporter, followed normally by the USA and Vietnam, which has now returned to the international market in a big way after previous 265

266

A.J.H.Latham

Table 13.1 World rice exports 1990–98 (thousand metric tons: milled basis)

Source: USDA, Grain: World Markets and Trade.

difficulties. Other major Pacific exporters are China, which sometimes exports and sometimes imports, and Australia. Australia has a true international comparative advantage in the production of short grain japonica rice types, being a temperatezone rice producer in similar latitudes to Japan, Korea and California (Rice Facts 1996, p. 1). Note too the importance of Uruguay, with substantial exports to Peru, Chile and Brazil. Argentina has also recently become an important exporter to these markets (Grain 1999 March, p. 8). Table 13.2 shows the major importers, and Pacific importers include China, Indonesia, Japan, Korea, Malaysia, the Philippines, Mexico and Peru. Few countries export and many countries import. But the crucial and disturbing feature for the 1990s, with serious implications for the economy of the Pacific region, is the instability of the trade. Japan in 1994 is a good example. Japan pursues a rice self-sufficiency policy, but in 1993 the worst cold wet weather for 50 years meant a severe crop shortfall, forcing her to import heavily in 1994. Although the preferred rice in Japan is short grain, much of the imported rice was long grain not suitable for her domestic market (LRBAC 1993 October, p. 1). Indonesia too had to make substantial imports in 1994, followed by even bigger imports in 1995 and yet more in 1996, even though she too had pursued rice selfsufficiency policies. A combination of floods and drought at the wrong times of the year were responsible (LRBAC 30 September, p. 1, 1 November, p. 1 1994; 1 February, pp. 2, 28 February, pp. 2, 31 March, p. 2 1995). China is of particular interest because she is sometimes a major exporter and sometimes a major importer. Table 13.3 shows how in the early nineties she exported rice each year with a peak in 1993 when her net exports were over 8 per cent of the world total, followed by a switch-about leading to the situation where in 1995 her net imports were over 9 per cent of the world total, a shift of nearly 20 per cent. The crucial year was 1994. By July 1994 she had been hit by flooding in the southern and central provinces. But she was already committed to export agree-

1990s Asia Pacific rice trade

267

Table 13.2 World rice imports 1990–98 (thousand metric tons: milled basis)

Source: USDA, Grain: World Markets and Trade.

ments (LRBAC 31 July 1994, p. 1; Grain 1995 February, pp. 10, 12). She went ahead with these, leaving herself desperately short in the domestic market and being forced to place import orders. This led to the net import situation of 1995 and there was yet more flooding (LRBAC 1 September 1995, p. 1). 1996 saw further net imports, with a return to net exports in 1997. 1998 saw net exports equivalent to over 12 per cent of world trade, but there was also extensive flooding in the autumn which endangered the situation for 1999. 1995 was a year of crisis both for China and the rice trade in general. Bangladesh made huge imports of 1,567 thousand metric tons, China 1,964 thousand metric tons and Indonesia a massive 3,011 thousand metric tons. How could the rice trade cope with one of its major suppliers becoming an importer, at the same time that Indonesia and Bangladesh required enormous quantities? There was no great increase in supplies from the big three, Thailand, the USA and Vietnam. The situation was saved by India, which suddenly came into the market. The Food Corporation of India had a stockpile which she had built up over the previous years. She released this into a seller’s market, 4,179 thousand metric tons in 1995 and another 3,549 thousand metric tons in 1996 (LRBAC 1 September 1995, p. 1). It led to an almost

268

A.J.H.Latham

Table 13.3 China rice exports and imports 1990–98 (thousand metric tons: milled basis)

Source: Tables 13.1 and 13.2.

comical situation whereby in February 1995 there was a 20-day wait for ships to get a berth at an Indian port! (LRBAC 31 March 1995, p. 2). There were not enough railcars or motor trucks to get the rice to the ports, and insufficient loading equipment to load the rice when it did get there. In November there were 37 ships waiting at Kakinada, 20 plus at Kandla, and sixteen at Bombay. By December there was a loading delay of 40 to 50 days at Kandla and at Kakinada ships could not even be given an expected loading date! (LRBAC 1 August, p. 2, 1 November, p. 2, 1 December, p. 2 1995). 1996 was another bad year for China, with imports of 832 thousand metric tons, and Indonesia also imported 1,029 thousand metric tons. The situation stabilized in 1997. But this respite was not for long. In March 1998 it was reported that Indonesia was likely to be the biggest buyer of rice that year. The situation was disastrous, as El Niño weather conditions had coincided with economic collapse. There were reports of more fires in Borneo so the situation was likely to be get worse. Indonesia needed more rice than it was logistically possible for her to take. The problems were finance, ability to unload the ships and distribute, and a shortage of available rice (LRBAC 4 March 1998, p. 2). As can be seen in Table 13.2, Indonesia’s imports in 1998 were eventually a colossal 6,081 thousand metric tons. It is notable that the Philippines, in the same general climatic area of the western equatorial Pacific, also saw rising imports in these years, with 768 thousand metric tons in 1996, 814 thousand metric tons in 1997 and 2,187 thousand metric tons in 1998. The Financial Times reported on 18 August 1998 that the El Niño weather phenomenon had devastated crop production in the Philippines, with rice down 27 per cent on the previous year and corn down 44 per cent. According to William Dar, the Agricultural Secretary, this was the worst performance record for 20 years. There was drought in Mindanao in the Southern Philippines (Financial Times 18 August 1998, p. 4). As for China, the Yangzi in 1998 had the worst floods since 1954. In July there was heavy flooding in Jiangxe, Hunan and Hubei, all important rice producing provinces. There was substantial destruction of bridges and railway lines, and strong currents made it difficult to work barges. Many mills were flooded, and rice could not be transported from the mills to the ports. The only consolation was that damage to rice crops in some areas would be partly offset by extra production in drier areas due to more water, and in the badly flooded areas, the deposition of fertile sediment would stimulate the later crop (LRBAC 31 July 1998, p. 1).

1990s Asia Pacific rice trade

269

The situation appeared dire (Financial Times 1/2 August 1998, p. 3; 10 August 1998, p. 4; 12 August 1998, p. 5). As a Mr Wun, a farmer in one of the Yangzi provinces, said on the main BBC London evening news at 9.00 pm on 14 August: ‘All the grain has been flooded and there won’t be any autumn harvest at all!’. By August several flood peaks had passed down the Yangzi each breaching more of the temporary defences put up to protect cities and industrial areas. But the floods went on for much longer than expected. Extra rain in some areas would be of benefit and partly offset flood damage, but as usual reliable damage estimates were difficult to get. Contracted shipments however were still being made, despite substantial delays because of the difficulty of getting rice from the Government store to the mills, and from the mills to the ports (LRBAC 28 August 1998, p. 1). By the beginning of October it was possible to evaluate the damage. Where harvest had not been completed rice had been lost, but the biggest problem was that the new crop could not be planted before the cold weather came. Some estimates suggested losses of 20 per cent, not including increased production in areas which benefited from extra rain. The Government stated that all contracted sales would be shipped, and that further sales might be made in the new year (LRBAC 2 October 1998, pp. 1–2). As is clear from Table 13.1 China did indeed continue to export rice despite the floods, her total exports in 1998 amounting to 3,734 thousand metric tons, the highest figure for the decade, and more than twice the level of the previous high in 1994 when exports were 1,519 thousand metric tons. Clearly the exports were determined by the fact that China had already contracted for these sales. 1999 may have seen compensating imports as happened in 1995. It was reported that the US Government had done a deal with China to help Indonesia, by agreeing to ship 500,000 metric tons of wheat to China whilst China would ship 250,000 metric tons of rice to Indonesia (LRBAC 6 November 1998, p. 4). By the end of 1998, the features of the year had become clear. Indonesia had made record imports of 6,081 thousand metric tons, double the level in the crisis year of 1995, and the Philippines also had made huge imports of 2,187 thousand metric tons. This came at a time when Bangladesh also had to make huge imports due to flooding, and so had Brazil. The fact that China continued to export despite her own floods was of considerable assistance to the world market, but the real saviour was again India as in 1995. Her exports rose to 4,491 thousand metric tons, even higher than her total for 1995 of 4,201 thousand metric tons. But the reports from India were disturbingly similar to those of 1995. There was congestion at the main port of Kandla, with 4 vessels on berth loading, 3 waiting, 15 with cargoes waiting for berths, 9 arrived whose cargo was not yet ready, and a further 4 vessels were expected shortly, a total of 35 vessels. Matters were made worse by the fact that the Indian government had changed truck legislation limiting trucks to 9 metric tons instead of close to 19 metric tons as previously. Freight costs had doubled overnight, but it was not clear that this was due to trucks carrying smaller loads or drivers having to make larger payments at the checkpoints between mill and port! (LRBAC 6 November 1998, p. 2). Clearly much of the situation in 1998 affecting Indonesia and the Philippines was linked to the El Niño phenomenon, or more properly the El Niño-Southern

270

A.J.H.Latham

Oscillation (ENSO). This irregularly recurring climatic variation sees warm water in the Western Pacific moving eastward taking rain clouds with it causing droughts in Indonesia and Australia but rains and floods in the Eastern Pacific, particularly Peru and other parts of South America. The eastward movement of warm water and recurrent droughts in Indonesia was first recognized by Sir Gilbert Walker in 1924 (Walker 1924) but its direct connection to the high rainfall and floods in Peru was not recognized until 1966 by Bjerknes (Bjerknes 1966). As early as 2 February 1998, it was reported from Argentina that the new crop ought to be in full swing by mid-February, but due to colder and wetter weather the harvest would be delayed until the end of the month. The area planted had been 5 per cent greater, but a smaller crop was expected because of the poor weather blamed on El Niño (LRBAC 2 February 1998, p. 1). These adverse conditions affected all the main rice growing and exporting areas of South America, Argentina, Uruguay and Brazil, and continued into March (LRBAC 4 March 1998, p. 1). This was the month that the catastrophic affect of the drought in Indonesia made itself felt (LRBAC 4 March 1998, p. 2). The adverse situation continued in South America through April (LRBAC 30 April 1998, p. 1). The ENSO effect was making its mark on rice growers across the Pacific. By May there was strong import demand from Indonesia, the Philippines and much of Latin America. This was due to the fall in production because of the bad weather attributed to El Niño (Grain May 1998, p. 6). The effects of this climatic trauma is shown fully in Tables 13.1 and 13.2. What is clear is that the ENSO phenomenon is not an isolated occurrence and will certainly return again in the future. There was a major occurrence in 1982–83, and also in 1972–73 (Glantz 1996, pp. 59–69). Quinn and Neal have listed these occurrences over the last 450 years (Quinn and Neal 1987). The impact on the rice trade, and economic activity in general, in the Pacific region is likely to be as serious as the one in 1997–98. Associated with this phenomenon has been the economic crisis of 1997–98. The chaos in the rice trade did not cause the economic crisis, but it featured strongly in the background. The crisis appears to have begun when partly as a response to recession and falling prices in Japan, the Chinese devalued the Yuan in the autumn of 1994 (Financial Times 20 May 1998, p. 4; Daily Telegraph 18 January 1999, p. 26 (Patrick Minford)). Prices were rising at a rate of 25 per cent per year making China’s exports prohibitively expensive, and drawing in cheaper imports from abroad. The swap rate for the Yuan fell heavily and recognizing this Zhu Rongji, China’s economic tsar and vice-president, abolished the managed rate of Yuan 5.7 to the dollar and adopted the swap rate of Yuan 8.7, an effective devaluation of 33 per cent (Financial Times 16 January 1998, p. 8. See also Sunday Telegraph 18 January 1998, p. 25). As a result of the fall in the swap rate and the related devaluation China was less able to import from her Asian neighbours, reducing their exports. More crucially their exports were hit in other world markets because China was now a lower price competitor. Thailand was badly affected and when she was forced to devalue on 2 July 1997, the sky fell in, bringing down currencies around the Pacific (Financial Times 1 July 1998, p. 23).

1990s Asia Pacific rice trade

271

But why was the swap rate falling and China forced to devalue in 1994? The economic crisis in Japan had put downward pressure on world prices and currencies. The massive flow of inward investment to China as she modernized her economy was also part of the story. The inflow of foreign capital stoked up inflationary forces. But China had other problems as well in 1994. As has been shown, flooding on the Yangzi in July caused a fall in rice production at a time when she had committed herself to rice exports. The rice crisis impacted on prices in a region where south of the Yangzi probably a third of per capita income is spent on rice alone (see FAO 1991, pp. 26, 70). But if one effect of the fall in the currency was to save foreign reserves by cutting imports in general and boosting exports, as far as rice was concerned it had the opposite effect. China was already committed to rice exports and had to import rice to replace the rice she was committed to sell. She imported heavily in 1994, massively in 1995, and was still importing in 1996. These were also years of heavy imports in Indonesia. But in 1998 a new rice crisis threatened. In Indonesia there was ENSO related drought, leading to the fall of Suharto. The destruction of Chinese-owned supermarkets where the local people bought their rice was a reaction to shortages and rising prices (Financial Times 12 August 1998, p. 4). Imports were urgently required by Indonesia, but the letters of credit were not available. The situation was desperate, as she needed the rice but did not have the money to pay for it (LRBAC 2 October 1998, p. 3). It was the fall in the Chinese Yuan in 1994 which had triggered the general crisis. Presumably a further depreciation would plunge the whole of the AsianPacific region into a further spiral of currency depreciations. But the situation on the Yangzi in the fall of 1998 was much worse than in 1994 when the collapse of the Yuan occurred. Overseas funds began to be repatriated from China. The failure of the Guangdong International Trust and Investment Corporation (GITIC), which helped raise billions of dollars for infrastructure including a nuclear power station and a superhighway, created further fears (Financial Times 7 October 1998, p. 4; 8 October 1998, p. 7; 20 October 1998, p. 8; 26 October 1998, p. 6; 28 October 1998, p. 4; 29 October 1998, p. 7; 3 November 1998, p. 29; 5 November 1998. p. 6. 7/8 November 1998, p. 4). What is more, the Asian crisis had widened to become an Asian-Pacific crisis, with Brazil and other Latin American countries drawn in. Pacific currencies, including those of Australia, New Zealand and Canada had already fallen severely, and on 13 January 1999 (Daily Telegraph 14 January 1999. p. 29) Brazil floated the Real resulting in a depreciation of 40 per cent plus. This threw more strain on the China exchange rate (Financial Times 18 January 1999, p. 19; 5 March 1999, p. 5; 8 March 1999, p. 19; LRBAC 2 February 1999, p. 1). Serious doubts about the stability of the Chinese exchange rate continued, as news came of more failures and defaults in China. The flow of money into China began to dry up (Financial Times 11 January 1999, p. 1; 12 January 1999, p. 4; 20 January 1999, p. 6; 22 January 1999, pp. 3, 38; 25 January 1999, p. 22; 27 January 1999, p. 4; 1 February 1999, pp. 19, 24; 2 February 1999, p. 33; 3 February 1999, p. 4; 8 February 1999, p. 5; 2 March 1999, p. 6; 25 March 1999, p. 19; 22 April 1999, p. 6; 5 May 1999, p. 34). Rumours and fears

272

A.J.H.Latham

of devaluation abounded (Financial Times 23/24 January 1999, p. 13; 26 January 1999, p. 33; 28 January 1999, p. 29; 5 February 1999, p. 4; 10 March 1999, p. 51; 11 March 1999, pp. 4, 19; 24 March 1999, p. 6). The domestic market was severely depressed with stocks unsold, and overseas purchases of aircraft and trucks from abroad were deferred (Financial Times 5 February 1999, p. 4; 9 February 1999, p. 1; 10 February 1999, p. 33). In February China halted crude oil sales to Japan, because the cost of producing crude oil in China was higher than oil could be bought on the international market due to the fall in international prices (Financial Times 9 February 1999, P. 4). So what are the conclusions to be drawn from this? It is clear that the AsiaPacific rice trade in the 1990s was a picture of chaos and instability and will continue to be so. China was the most unstable player swinging from exporter to importer unpredictably, for climatic reasons beyond her control. Unfortunately this instability in the Asia-Pacific rice trade was played out in the background of the currency and general economic crisis in the region of 1997–98. China was the weak link in the chain of exchange rates, and her weakness was compounded by the consequences for her rice trade of the Yangzi floods of 1994 and 1998. Thus the rice trade is an indicator of more general problems in these countries. Sudden imports of unusual size point to a country with domestic difficulties. In countries where consumers spend a third of their income on rice, rice shortages result in upward price movements which diminish the amount of disposable income remaining to be spent on manufactures and services, with obvious consequences. Increased imports of rice also adversely affect the balance of trade and payments of these countries, many of which are already heavily indebted. Downward pressure on exchange rates will result. The rice trade is a crucial characteristic of the Asia-Pacific economy, and will continue to be an unstable element in the overall dynamism of the region. References Bjerknes, J. (1966) ‘A possible response of the atmospheric Hadley circulation to anomalies of ocean temperature’, Tellus 18, pp. 820–9. Daily Telegraph, London. Financial Times, London. FAO (1991) Demand Prospects for Rice and Other Foodgrains in Selected Asian countries, Rome: FAO. Glantz, M.H. (1996) Currents of Change: El Niño’s Impact on Climate and Society, Cambridge: Cambridge University Press. Grain: United States Department of Agriculture (USDA). Grain: World Markets and Trade Latham, A.J.H. (1998) Rice: The Primary Commodity, London: Routledge. Latham, A.J.H. (1999) ‘Rice is a luxury, not a necessity: The sources of Asian growth’, in D.O.Flynn, L.Frost and A.J.H.Latham (eds) Pacific Centuries: Pacific and Pacific Rim History since the Sixteenth Century, London: Routledge. LRBAC: London Rice Brokers’Association Circular, Farnham, Surrey, various dates. Quinn, W.H. and Neal, V.T. (1987) ‘El Niño Occurrences Over the Past Four and a Half Centuries’, Journal of Geophysical Research, Vol. 92, no. C13, pp. 14,449–14,461. Dec 15.

1990s Asia Pacific rice trade

273

Rice Facts (1996) Leeton, New South Wales: Ricegrowers’ Co-operative Limited. Sunday Telegraph, London. Walker, G.T. (1924) ‘Correlation in season variations of weather IX: A further study of world weather’, Mem. Indian Meteor. Dept. 24 (9), pp. 275–332.

Index

acid production 232, 234 Addis, Sir Charles 252, 255 African countries 170, 176 agriculture 30, 228, 238 Aldcroft, Derek 5–6 Allen, Robert 31–2 Almaden Mine, Spain 218, 221, 223 Anderson, C.A. 177 Argentina 9, 266, 270 ASEAN 177–8 Ashton, David 179 Asia: Californian trade 217–18, 222, 237– 8; enterprises 131; and Europe 16–17, 32; garment industry 50, 64; GDP 170; globalization 1, 18, 33, 38; real wages 19–20, 21–3, 24, 29; regional differences 20–1, 37–40, 41; textile trade 49, 65–6; see also individual countries Asian Tigers 64–5; economic development 64–5, 169–71; education 111, 170, 171–5, 179–80; human capital 6, 170; Japanese rule 56–7; labour costs 56–7; literacy 177; see also Hong Kong; Korea, South; Singapore; Taiwan Atwell, W.S. 202 Australia 9, 16, 216, 266, 270 Bangladesh 10, 65, 265, 267, 269 banking: China 77–8; gender 257; generalists 248–50, 252–3, 258, 259– 60; specialists 252–3, 254, 258; see also Hongkong Bank; HSBC Holdings Bass, Hans 4 Bjerklos, J. 270 blacksmithing 235–6 Blainey, Geoffrey 16

Bombay 268 Bowman, M.J. 177 Brading, D. 224 Brazil 9, 269, 270, 271 Britain: cotton 49, 50, 51, 52, 62; industrialization 14; interest rates 94; living standards 28, 30–1; real wages 19–20, 21, 23, 25, 27; textile trade 49, 62, 64; unskilled wages 91–2; wheat prices 89 British Bank of the Middle East 249 BritishBank 247, 259 bronze coinage 88, 89 Brunat, Paul 75–6 Brunei State Bank 260 Bulbeck, David 16 Bulgaria 106 bullion 203, 209 Burma 19, 22–3, 40, 265 cable cars 236 California: acids 232, 234; agriculture 228, 238; Asian trade 217–18, 222, 237–8; blacksmithing 235–6; cable cars 236; dredging 232, 233; economic effects of gold rush 227–9; explosives 233–4; flour mills 231, 232; glass manufacture 231; gold 8, 216, 217, 223; gold coin 232; hydraulic mining 232–3, 235; industrialization 229–31; iron working 235–7; lumber 231, 232, 234–5; population 228; pumps 232–3, 234; quicksilver 216, 219, 220, 221–7, 237– 8; silver 8, 216, 237–8; steam engines 234; stone quarries 231, 233; Union Iron Works 236–7 capital market 94–8 274

Index

Castells, M. 177 Caughey, John 228 Ceylon 40; see also Sri Lanka Cha, Myung Soo 3 chaebol system 109 Chambers of Commerce 152–4, 155 Chambon silk spinning 75, 79, 82n8 Chen Qi-Yuan 78, 79 Chile 9, 265, 266 China: banking 77–8; communism 5; copper coins 187–95; cotton 51, 61, 66; currency devaluation 10, 128, 270, 271– 2; enterprises 76, 80–1, 129, 130–2, 206–8; export policies 127–8; exports 4–5, 119–20, 120, 126, 137, 138, 140; FDI 77; flooding 267, 268–9, 271, 272; Foreign Trade Corporations 121, 129; Guangdong 78–9, 80, 82n7, 271; industrialization 112, 212n5; Japanese industry 52–3; living standards 15, 28, 30; market share 141–3; money systems 6–7, 187–8, 200–1; oil exports 272; paper currency 7, 192, 196n7, 199, 205, 208, 213n6; Pearl River Delta 58; porcelain 212n8; Qing dynasty 191–2; Quanzhou region 190; real wages 20, 25, 40; rice imports/exports 9–10, 88, 265–7, 268, 269, 271, 272; Selfstrengthening movement 70, 81, 82n16; silk industry 2–3, 70–1, 72–3, 81n3, 201, 202, 205, 206, 211; silver 7, 8, 200, 201–2, 206–8, 210–11; silver coins 190–5, 199; social/political unrest 155–6; Special Economic Zones 127–8; Taiwanese investment 122–3, 124; tea 211; textile trade 2, 52–3, 61–2; trade liberalization 18; transport 77; woollen trade 61; yarn imports 51; see also Ming China Chinese dialect groups 5, 151–2, 162 Chinese family businesses 130–1 Chinese General Chamber of Commerce 154 Chinese Manufacturers’ Association 154–5 Ching Wah Bulb Co. 157–8 Cho, Yun Je 97 Choudhury, A. 177 Chung, S. 151, 154 Clayton, David W. 5 climate: drought 270, 271; El Niño 10, 268, 269–70, 271; flooding 267, 268–9, 271, 272; Korea, South 88; rice production 9–10, 268, 270 clothing: see garment industry

275

coal 17 coffee trade 16–17 Colclough, C. 176–7 Collins, William 38 commodities 14, 16–17, 33 communications 150 comparative advantages 105–6, 177–8 Confucianism 152 copper coins 7, 187, 188–95 core-periphery approach:convergence 13– 14; European-Asian trade 16– 17;inequality 15, 28, 31–2; real wages 19–20 corporate cultures 245–6, 253–4, 261–2 cotton 50, 66; Britain 49, 50, 51, 52, 62;China 51, 61, 66; Hong Kong 58; India 2, 50–1, 53–4, 55–6, 65; Japan 51–2, 53–4; price fall 17 cowries 197n11, n12 Crafts, N.F.R. 106 creative destruction 118 Cross, H. 224 currency: devaluations 10, 128, 270, 271– 2;local 195, 196, 197n12; revaluation 118–19 currency, types: bronze 88, 89; copper 7, 187, 188–95; cowries 197n11, n12; gold 204, 232; paper 7, 192, 196n7, 199, 204, 205, 208, 213n16; silver 190– 5, 196n8, 199 currency circuits 188, 190–2, 193 Dar, William 268 David, P. 163, 166n33 Department of Commerce and Industry 156–7, 158 developmental states 111, 114n10 dialect groups, Chinese 5, 151–2, 162 dredging 232, 233 drought 270, 271 Easterlin, R.A. 176 economic crisis 10, 27, 245, 270–1 economic development 65, 170; Asian Tigers 64–5, 169–71; China 211–12n4; education 5–6, 175–9; institutions 149– 50, 158–62; money 203–6;polarization 169; Thailand 64 The Economist 238 education: Asian Tigers 5–6, 111, 170, 171–5, 179–80; economic development 5–6, 175–9; funding 174; GNP 171–2; human capital 111, 170, 174, 175–

276

Index

6;Japan 173, 174, 179; modernization 177;social cohesion 6, 172–3, 179; training 174, 178–9; wages 176, 178 El Niño 10, 268, 269–70 El Niño-Southern Oscillation 269–70, 271 England, silver coins 193, 196n8 enterprises: Asia 131; China 80–1, 129, 130–2; family-based 130–1; flexibility 130, 131, 132; private 76, 206–8; size 128–9 entrepreneurship 80–1, 153, 161 Europe 16–17, 24, 32, 170 explosives 233–4 export policies 126–7, 128 export processing zones 135 export substitution 118, 120, 121 exports 117–18; China 4–5, 119–20, 120, 126, 137, 138, 140; flexibility 130; GDP 17, 60; growth, 121–2, 139; Hong Kong 165n24; Indonesia 17, 64–5; Korea, South 60; productivity 144n1; Taiwan 4–5, 60, 119, 124, 133, 134, 135, 136, 137, 141; Thailand 17, 265 factor prices 19, 33, 40–1 Farnie, D.A. 2 FDI 4, 77, 123 financial liberalization 97 flooding: China 267, 268–9, 271, 272; rice production 267, 272 flour mills 231, 232 flying-geese pattern 122, 124 Flynn, Dennis O. 7 Food Corporation of India 267–8 Foreign Trade Corporations 127, 129 France 90, 91 Frank, Andre Gunder 202 free trade, Japan 17–18, 71 Fukuyama, F. 131 galena 224 garment industry: Asia 50, 64; homogenization 66; Hong Kong 2, 56– 8; India 65; Korea, South 60–1, 62; mass-produced 2, 49–50; Taiwan 2 GDP: Asia 170; exports 17, 60; India 17, 35, 36; Indonesia 17, 35, 36; inequality 15; Japan 35, 170; Korea, South 35, 60–1, 170; Latin America 170; Philippines 35; real wages 34, 35–7, 41; Taiwan 35, 60, 170; Thailand 17, 35, 37; USA 170 gender, banking 257

Germany 107 Gerschenkron, Alexander: Germany 107; Japan 15, 108–9; Korea, South 109–10, 112; late industrialization 3–4, 15, 104– 6, 110, 111, 114n3; Russia 107–8 Giráldez, Arturo 7 glass manufacture 231 globalization 13–16, 41; Asia 1, 18, 33, 38; commodity markets 33; Japan 52–6; silk industry 212n9; silver market 210– 11 GNP: education 171–2; Korea, South 4, 109, 111 gold: California 8, 216, 217, 223, 232; California/Asia trade 217–18, 222, 237–8; mechanical separation 223; placer gold 223, 232; quicksilver extraction 218; smelting 224 gold coin 204, 232 grain market 17, 88–9 Grayburn, Sir Vandeleur 250 Green, F. 179–80 Guangdong 78–9, 80, 82n7 Guangdong International Trust and Investment Corporation 271 guilds 151, 152, 163n3 Hallidie, A.S. 236 Hang Seng Bank 248, 249–50, 255 Hanley, S. 179, 180 von Hayek, F.A. 132 He Qiaoyuan 201, 202 Heckscher, Eli 33 Hicks, John 6, 187, 195–6 Hindu equilibrium 32 Hittell, John 228, 235 HKBL: see Hongkong Bank Hong Kong: Chambers of Commerce 152– 4, 155; commercial wealth 58–9; communications 150; cotton exports 58; Department of Commerce and Industry 156–7, 158; dialect groups 151–2; education 5–6, 171, 173, 174, 180; as entrepôt 58, 162; exports 165n24; garment industry 2, 56–8; immigrant entrepreneurs 153, 161; industrialization 5, 154–5, 162; information costs 150, 151–2, 159; institutions 151–62; Japanese rule 247; Kaifong associations 164n6; Manufacturing Associations 154–5; non-intervention 57; Pearl River Delta investments 58; piracy 160–1; racism

Index

153; state 155–8; textile trade 57–9, 65; trade growth 165n18; transaction costs 150 Hong Kong Chamber of Commerce 5 Hong Kong Exporters’ Association 157–8 Hong Kong General Chamber of Commerce 155, 160, 161–2, 166n32 Hongkong and Shanghai Banking Corporation 246, 262n3, 264n30; see also HSBC Holdings Hongkong Bank 8–9, 245, 246, 247, 256– 7; branch manager 259; branches and staff 250, 251–2, 253, 254–5; corporate culture 253–4, 261–2; cotton 58; labour disputes 255; Marine Midland 254–5, 260; New York agency 264n29; regional officers 254, 259; Sri Lanka 261; subsidiaries 249, 255; see also International Officers Hongkong Bank Malaysia Berhad 259 Hongkong Bank of Australia 249 Hongkong Bank of Canada 249 HSBC Bamerindus (Brazil) 250, 260 HSBC Holdings 8–9, 248; Bamerindus (Brazil) 260; Brunei State Bank 260–1; corporate culture 248–50, 250–1; economic crisis 245; International Officer staff 257; Kazakhstan 261; staff policies 250–5 human capital: Asian tigers 6, 170, 175–6; economic development 180; education 111, 170, 174, 175–6; World Bank 175 hydraulic mining 232–3, 235 hydroelectrics 236 immigration 14, 15–16 imperialism 37, 41 import substitution 97, 113, 118, 171 income distribution 18–19, 170; see also real wages; wages India: cotton 2, 50–1, 53–4, 55–6, 65; garment exports 65; GDP 17, 35, 36; industrialization 55–6; labour costs 55; living standards 15, 28, 30; real wages 20, 21, 22, 25, 27; rice consumption 265; rice exports 10, 88–9, 267–8, 269; trade liberalization 18; transport 50–1; wheat prices 88–9 Indonesia: drought 270; exports 17, 64–5; GDP 17, 35, 36; real wages 20, 21, 22– 3, 27, 35, 39, 40; rice consumption 265; rice imports 10, 266, 267, 268, 269, 271; trade liberalization 18

277

industrialization 38, 40; Britain 14; California 229–31; China 112, 212n5; Germany 107; Hong Kong 5, 154–5, 162; India 55–6; institutions 105, 106; Italy 106; Japan 15, 108–9; Korea, South 3–4, 60–1, 95, 109–10, 112, 114n5; late 3–4, 15, 104–6, 110, 111, 114n3; Russia 107–8; see also Gerschenkron inequality: core-periphery approach 15, 28, 31–2; GDP 15; growing 169 infant industries 106, 109–10 inflation 170, 205, 208–9, 213n19 information costs 150, 151–2, 159 information processing 132 Ingram, James 30–1, 32 innovation 117–18, 120, 125, 128, 130 institutions: economic development 149– 50, 158–62; formal/informal 162–3; Hong Kong 151–62; industrialization 105, 106; transaction costs 5 interest rates 94–5, 96 International Officers 245–6, 254; case studies 260–1; corporate culture 245–6, 248, 251, 254; and foreign staff 255; generalist 248–50, 252–3, 258, 259–60; multinational culture 256–7; recruitment 255–7; role 257–60, 261–2; university graduates 256; women 257 International Scientific Committee 31 investment: flying-geese pattern 122, 124; see also FDI iron working 235–7 Italian Piedmontese silk reeling 71, 74–5 Italy 106 James, D. 179–80 Japan: cotton 51–2, 53–4; education 173, 174, 179; free trade 17–18, 71; GDP 35, 170; globalization 52–6; imperialism 37, 41; industrial plant in China 52–3; industrialization 15, 108– 9; Meiji Restoration 70, 76, 81; modernization 3; piece-goods 54; private enterprises 76; real wages 1, 20, 21, 22, 23, 25, 27, 35, 40; recession 270, 271; rice 9, 88, 265, 266; silk industry 2–3, 70–1, 72, 73–4, 75; silver exports to China 193, 200, 201–2; suspicions of West 82n11; terms of trade 17–18; textile trade 2, 52–5; transport 17; zaibatsu system 109 Japanese rule 56–7; China 54–5;

278

Index

educational legacy 171; Hong Kong 247; Korea, South 3, 37, 60, 89, 98; Taiwan 37, 59 Jardine, Matheson & Co. 75, 152–3 jewellery 232 Journal of Economic History 140 jute 17 Kaifong associations 164n6 Ka Kinada 268 Kandla 268, 269 Kawakatsu, Heita 51 Kazakhstan 261 Keynes, J.M. 208–9, 213n19 King, Frank H.H. 8–9 Kirkpatrick, C.H. 177 Korea, South 96; bronze coinage 88, 89; capital market 94–8; chaebol system 109; climatic shocks 88; education 5–6, 111, Korea—contd. 171, 172–3, 174; exports 60; financial liberalization 97; garment industry 2, 60–1; GDP 35, 60– 1, 170; GNP 4, 109, 111; government/ markets 86, 97; human capital 175–6; import substitution 113; industrialization 3–4, 60–1, 95, 109–10, 112, 114n5; interest rates 94–5, 96; Japanese rule 3, 37, 60, 89, 98; labour market 89–93; protectionism 112–13; real wages 20, 23, 25, 35, 89–91; repression 92, 97; rice 3, 86–7, 265; savings 95; technological catching-up 110, 111; textile trade 57, 61, 65; trade liberalization 18; USA 97 Krugman, Paul 38 Kuroda, Akinobu 6–7 Kustel, G. 224 labour costs 55, 56–7, 118–19, 122 labour market 89–93, 174–5 labour surplus 33 Lal, Deepak 32 Lancashire yarn 49, 50–1, 52, 62 Latham, A.J.H. 9 Latin America: currency crisis 271; GDP 170; real wages 19, 20; rice 265; see also individual countries leather hoses 233 Lee, Jaymin 3–4 Lewis, W.Arthur 15, 22, 32, 33 literacy 171, 176, 177; see also education living standards 13–14, 15, 19; Asia 1, 15– 16, 41; Britain 28, 30–1; China 15, 28,

30; Europe 41; India 15, 28, 30; Thailand 30–1 lumber 231, 232, 234–5 Ma, Debin 2–3 Mackintosh, W.A. 65 McWilliams, Carey 228–9 Maddison, Angus 18, 34 Maebashi 75 man-made fibres 2, 49–50 Man Mo Temple 153 Manufacturing Associations 154–5 marble 231, 234 Marine Midland Bank 249, 250, 254–5, 260 market information 157, 159 market share 123–4, 141–3 mass production 2, 49–50 Mauritius 65 Mehmet, O. 174–5 Meiji Restoration 70, 76, 81 Meng, Y.L. 191 Menger, K. 196n5 Mercantile Bank 249, 253 merchant associations 149, 163n3 mercury: see quicksilver Mexican silver dollars 8, 217 Mexico, quicksilver 225 Mexico, rice imports 265–6, 267 Midland Bank 245, 247, 259, 260 millet prices 88 Mincer, J. 178 Ming China: money/economic development 204–6; paper currency 213n16; physiocrats 202–3; silk exports 201–2; silver 200, 205–8; silver coins 7, 193–4, 199 mining techniques 232–3 Miyamoto, Matao 88 Miyazawa, T. 196n3 modernization 3, 177 money 6, 187–8, 190–2, 200–1, 203–6; see also currencies Multi-Fibre Arrangement 58, 62 Munting, R. 108 Myllyntaus, T. 180 Nam Pak Hong 154 Neal, V.T. 270 neoclassical economists 109, 114n6 Netherlands 94 Nevada: mining techniques 233; silver 8, 216, 217–18, 227

Index

Nordhaus, W.D. 204 North, Douglass 5, 149, 158, 163 office machinery 124 Ohlin, Bertil 33 oil exports 272 Opium War 71, 88 Orgill, Richard 260 Osaka Spinning Company 54 Pacific War 2 Pakistan 65 Palais, James 89 Panama Canal 41, 265 paper currency 204; China 7, 192, 196n7, 199, 205, 208, 213n16 Park Chung Hee 97 Parthasarathi, Prasannan 28, 30, 32 paternalism 131 Patio Process, silver 218, 225, 226 Paul, Rodman 233, 235 Pearl River Delta investments 58 Peel Holdings PLC 63 Pelton, Lester A. 236 Pennell, J.W. 161 Peru 9, 218, 265, 266, 270 petrol exports 119–20 Philippines: GDP 35; real wages 20, 21, 23, 35, 40; rice trade 10, 268, 269, 270 physiocrats 202–3 piece-goods 54 piracy 160–1 Polanyi, Karl 197n11 Pomeranz, Kenneth 30, 32 porcelain 212n8 Porter, M. 131 Portugal 200 prices-wages 86, 99n7 private enterprises 76, 206–8 productivity 19, 144n1 profit, inflation 208–9, 213n19 property rights 149, 155, 156, 158, 163n4 protectionism 62, 107, 112–13 Psacharopoulos, G. 178 pumps 232–3, 234 Purves, Willie 250 Quanzhou region, China 190 quartz 233, 235 quicksilver: California 216, 219, 220, 221– 7, 237–8; demand 218, 221–3; Peru 218; prices 225–6; Slovenia 218, 221, 223

279

Quinn, W.H. 270 racism, Hong Kong 153 Raymond, Rossiter 224, 225–6 real wages 1, 18, 40–1; Asia 19–20, 21–3, 24, 29; Britain 19–20, 21, 23, 25, 27; China 20, 25, 40; core-periphery approach 19–20; cross-country comparisons 14, 15, 21–3; GDP 34, 35–7, 41; income distribution 18–19; India 20, 21, 22, 25, 27; Indonesia 20, 21, 22–3, 27, 35, 39, 40; Japan 1, 20, 21, 22, 23, 25, 27, 35, 40; Korea, South 20, 23, 25, 35, 89–91; Latin America 19–20;nominal wages 91; Philippines 20, 21, 23, 35, 40; Taiwan 20, 23, 25, 35; Thailand 20, 21, 22–3, 27, 35, 37 recession 270, 271 Redding, S.G. 132 Reid, Anthony 16 repression 92, 97 revisionists 109, 114n6 rice: Bangladesh 10; China 9–10, 88, 265– 7, 268, 269, 271, 272; as currency 86– 7, 192; India 10, 88–9, 265, 267–8, 269; Indonesia 10, 265, 266, 267, 268, 269, 271; Japan 9, 88, 265, 266; Korea, South 3, 86–7, 265; Latin America 265; long/short grain 265, 266; Philippines 10, 268, 269; prices 1, 3, 17, 87–9; Thailand 9, 265; USA 9, 265; Vietnam 9, 265; world exports 266; world imports 267 rice production: climate 9–10, 268, 270; drought 271; flooding 267, 272 riots, weavers 3, 79, 80 Rosenberg, Nathan 236 Rosovsky, H. 108–9 Rothschild cartel 219, 223 Russia 107–8 saleability 196n5 Samuelson, P.A. 204 Sandberg, L.G. 176 savings 95 Schumpeter, J.A. 117–18, 128, 129, 130 seigniorage 187, 188–9, 196, 196n2 Sekiguchi, Sueo 132 sericulture: see silk industry Shanghai 54–5, 73, 75, 80, 82–3n18 Shanghai Ewo Filature 75, 79–80 Siam 18, 22; see also Thailand

280

Index

silk exports 2–3, 201–2, 205, 211; raw silk 2–3, 70–1, 72–3 silk industry 201; China 81n3, 201, 202, 206; globalization 212n9; Japan 2–3, 70–1, 72, 73–4, 75; labour intensive 76–7, 78, 79; production 81n1; raw cocoons 2, 77–8; silk reeling sector 70– 5, 78–9, 80, 82–3n18; spinning 75, 79, 82n8; technology adapting 2–3, 76–9; technology transfer 74–8; transport 71; twisting mechanism 75; weavers’ riots 3, 79, 80 silver: California/Asia trade 8, 216, 237–8; Chinese imports 7, 8, 193, 200, 201–2, 206–8, 210–11; globalization of market 210–11; Japanese exports 193, 200, 201–2; Mexican dollars 8, 217; Ming China 200, 205–8; Nevada 8, 216, 217– 18, 227; production 218, 223, 224, 225–7; quicksilver 218; surplus value 209–10 silver coins: China 190–5, 192–5, 199; England 196n8 Singapore: education 5–6, 172, 174; as entrepôt 65; Hong Kong mills 58 Sino-Japanese naval war 70 Siyi Chamber of Commerce 152, 154 Slovenia 218, 221 Smith, Adam 203, 204 social cohesion 6, 172–3, 179 Song, Hogeun 92 Soviet Union 108 Spain 200, 218, 221 spice trade 16–17 spinning mills 52–3 Spufford, P. 193 Sri Lanka 261 St. Clair, David 8 steam engines 234 Steensgaard, Niels 201 Stokey, Nancy 178 stone quarries 231, 233 Suez Canal 17, 41, 71 sugar 17 Sung, J. 179–80 Sweden 90, 91 Taiping Rebellion 88 Taiwan 59–60; Chinese family businesses 130–1; currency revaluation 118; education 5–6, 171, 174; export policies 126–7, 128; export substitution 121; exports 4–5, 60, 119, 123, 133, 134,

135, 136, 137, 141;FDI 128; garment industry 2; GDP 35, 60, 170; human capital 175–6; import substitution 118; investment in China 122–3, 124; Japanese rule 37, 59; labour costs 118, 122–3; real wages 20, 23, 25, 35; small enterprises 128–9; textile trade 57, 59– 60, 65; US aid 59 Tan, Lay Cheng 16 tariffs 14 Tavelle silk spinning 75, 79, 82n8 tea, China 211 technological convergence 110, 111, 236 technology: adaptations 2–3, 76–9; gold and silver mining 236–7; silk industry 74–8; textile trade 52–3; see also innovation telecommunications 124 textile trade 64; Asia 49, 65–6; Britain 49, 62, 64; China 2, 52–3, 61–2; Hong Kong 57–9, 65; Japan 2, 52–5; Korea, South 57, 61, 65; man-made fibres 2, 49–50; mass production 2, 49–50; Taiwan 57, 59–60; technology 52–3; transport 17; wages 28, 30; yarn exports 2; see also cotton; silk; woollens Thailand: devaluations 10, 270; economic development 64; exports 17, 265; GDP 17, 35, 37; living standards 30–1; real wages 20, 21, 22–3, 27, 35, 37; rice trade 9, 265 Tomioka Mill 76, 80, 82n11 Trade Advisory Committee 156–7, 158 trade associations, Hong Kong 151 trade liberalization 14, 17–18, 71 trade unions 92 training, on-the-job 178–9 transaction costs 5, 150, 151–2 transport: China 77; commodities 16–17; India 50–1; Japan 17; Panama Canal 41; silk industry 71; Suez Canal 17, 41, 71; trade liberalization 14 Tsai, J.-F. 159 Tung Wah Hospital Committee 153–4 turbine wheel generator 236 Union Iron Works 236–7 Uruguay 9, 266, 270 US Treasury Department 217–18 USA: aid to Taiwan 59; Census data 229– 30; GDP 170; Korea, South 97; protectionism 62; rice trade 9, 265;

Index

Taiwanese imports 129, 130; see also California; Nevada van der Wee, Herman 199 vermilion 225, 226, 227 Vietnam 9, 40, 265, 266, 267 vocational training 174 von Glahn, Richard 200, 202 wage-rental ratios 1, 33–4, 38, 41 wages: convergence 30, 93; education 176, 177, 178; prices 86, 99n7; skilled/ unskilled 34, 89–90, 91–2; textile trade 28, 30; see also real wages Walker, Sir Gilbert 270 Wallerstein, Immanuel 209–10 Washoe Process, silver 218, 225, 226–7 weavers’ riots 3, 79, 80 Weber, Max 6, 187

wheat prices 88–9 Whitney, Josiah 216, 225, 226 Whitson, Keith 260 Williamson, Jeffrey G. 1 women in banking 257 woollen industry 61 World Bank 109–10, 170–1, 175 Wu, Yiqi 16 Xu Guangqi 202–3 Yamanobe, Takeo 53 Yang, C. 189 yarn trade 2, 51; see also textile trade yuan devalued 10, 128, 270, 271 zaibatsu system 109 Zanier, Claudio 71 Zhn, Rongi 270

281