China's Industrial Technology: Market Reform and Organizational Change (Unu Intech Studies in New Technology and Development, 8)

CHINA’S INDUSTRIAL TECHNOLOGY

In recent years, many studies have discussed the restructuring of national innovation systems in industrially developed countries. However, little attention has been paid to issues faced by developing countries involved in similar processes of change. In China’s Industrial Technology, Shulin Gu studies the transformation of the biggest industrial technology system which has ever emerged in a developing country. Offering a comprehensive review of reform policy, followed by an examination of major approaches to institutional restructuring, Shulin Gu explores the way in which China’s industrial technology has responded to economic reforms. At the heart of the work is the argument that market reform and organizational change are closely interdependent. Gu outlines the interaction of the two in China and reveals the damage which may result if market reform is not accompanied by new organizational design. Analysis of these issues is drawn from first-hand experience of Chinese technology systems, supported by insights from technological innovation economics and transaction cost economics. This is a ground-breaking work which offers insights into China’s possible future in the global and Asian economies and also provides policy lessons for other developing countries and former planned economies. Shulin Gu worked as a scientist and industrial engineer in a number of distinguished Chinese institutes in the late 1960s and 1970s. In 1983 Professor Gu began her studies of science and technology policy and in 1992 she joined UNU/INTECH as a Senior Research Fellow. Her work on technology policy and research institutions in China is internationally renowned and she has published widely in both English and Chinese.

1 WOMEN ENCOUNTER TECHNOLOGY Changing patterns of employment in the Third World Edited by Swasti Mitter and Sheila Rowbotham 2 IN PURSUIT OF SCIENCE AND TECHNOLOGY IN SUB-SAHARAN AFRICA The impact of structural adjustment programmes John Enos 3 THE POLITICS OF TECHNOLOGY IN LATIN AMERICA Edited by Maria Inês Bastos and Charles M.Cooper 4 EXPORTING AFRICA Technology, trade and industrialization in Sub-Saharan Africa Edited by Samuel M.Wangwe 5 TECHNOLOGY, MARKET STRUCTURE AND INTERNATIONALIZATION Issues and policies for developing countries Nagesh Kumar and N.S.Siddharthan 6 FLEXIBLE AUTOMATION IN DEVELOPING COUNTRIES The impact on scale and scope and the implications for location of production Edited by Ludovico Alcorta 7 GLOBALIZATION, FOREIGN DIRECT INVESTMENT AND TECHNOLOGY TRANSFERS Impact on and prospects for developing countries Nagesh Kumar and collaborators 8 CHINA’S INDUSTRIAL TECHNOLOGY Market reform and organizational change Shulin Gu

UNU/INTECH STUDIES IN NEW TECHNOLOGY AND DEVELOPMENT Series Editors: Charles Cooper and Swasti Mitter The books in this series reflect the research initiatives at the United Nations University Institute for New Technologies (UNU/INTECH) based in Maastricht, The Netherlands. This institute is primarily a research centre within the UN system and evaluates the social, political and economic environment in which new technologies are adopted and adapted in the developing world. The books in the series explore the role that technology policies can play in bridging the economic gap between nations, as well as between groups within nations. The authors and contributors are leading scholars in the field of technology and development; their work focuses on: • • • •

the social and economic implications of new technologies; processes of diffusion of such technologies to the developing world; the impact of such technologies on income, employment and environment; the political dynamics of technology transfer.

The series is a pioneering attempt at placing technology policies at the heart of national and international strategies for development. This is likely to prove crucial in the globalized market, for the competitiveness and sustainable growth of poorer nations.

CHINA’S INDUSTRIAL TECHNOLOGY Market reform and organizational change

Shulin Gu

London and New York

Published in association with the UNU Press

First published 1999 by Routledge 11 New Fetter Lane, London EC4P 4EE This edition published in the Taylor & Francis e-Library, 2005. “To purchase your own copy of this or any of Taylor & Francis or Routledge’s collection of thousands of eBooks please go to www.eBookstore.tandf.co.uk.” Simultaneously published in the USA and Canada by Routledge 29 West 35th Street, New York, NY 10001 © 1999 United Nations University 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 Gu, Shulin, 1942– China’s industrial technology: market reform and organizational change/Shulin Gu. p. cm. Includes bibliographical references and index. 1. Technological innovations—Economic aspects—China. 2. Technology and state—China. 3. Industrial policy—China. I. Title. HC430. T4G8 1999 338′.064′0951–dc21 98–20398 CIP ISBN 0-203-16520-9 Master e-book ISBN

ISBN 0-203-25957-2 (Adobe eReader Format) ISBN 0-415-19741-4 (Print Edition)

CONTENTS

List of tables

x

List of figures

xii

List of case texts

xiii

About the author

xv

Foreword

xvi

Series editor’s preface Author’s preface Acknowledgements

xviii xix xxiv

PART 1 A review of reform policy for the S&T system in China: from paid transactions for technology to organizational restructuring 1

2

3

Introduction to reform policy

2

The need to address reform policy

2

The approach of the study—empirical observation of the historical evolution of reform policy in China

4

Changes in the economic environment since the end of the 1970s

5

The institutional heritage of industrial R&D institutions

7

The pre-‘decision’ period: reformation of planning practice (1978–1985)

9

Rehabilitation and improvement of R&D institutions

9

Elaboration of planning practice

11

The decision on S&T management system reform (1985) and the creation of a ‘technological market’

13

Central policy measures—the ‘technology market’ approach

14

Other policy measures to support the technology market approach

15

Responses to the technology market solution

16

The growth of the technology market

19

vii

4

Merging R&D institutes into existing enterprises (1987)

22

The justification for merging—a policy response to the inefficiency of the technology market 22

5

6

7

Policy measures for merging

23

Explanation of the lack of success, for two industrial sectors

23

Transformation after merging into an enterprise—the case of the Automobile Technology Research Institute

25

Spin-off enterprises and the Torch Programme (1988)

26

The rationale for reform policy to support spin-off enterprises

26

Policy measures for spinning-off

27

The role of New Technology Enterprises

30

The transformation of established R&D institutes (since the 1990s)

31

General trends of transformations

31

An illustration of transformation into a market-profitable corporation

32

Policy response to the restructuring of industrial technology R&D institutes

34

Concluding remarks

36

The reform of industrial technology R&D institutions is indispensable to adapt to the new economic regime

36

The inadequacy of the pure technology market approach

38

The principle directions of restructuring, and some factors influencing the restructuring

38

The main purpose of reform policy—to address the lack of institutions

40

Appendix: Statistical data on China’s R&D system

41

PART 2 Spin-off enterprises: channelling the components of R&D institutions into innovative businesses 8

9

Introduction to spin-off enterprises

47

The essence of institutional restructuring in market reform

47

China’s S&T system

49

The scope and structure of Part 2

50

An overview: the launch of the Torch Programme and the development of spin-off enterprises

52

Origins of spin-offs

52

The launch of the Torch Programme

53

viii

10

11

12

The development of Zones and NTEs

55

The initiation of the NTEs

58

Forms of spinning-off

58

The initiators of NTEs

62

Decentralization and the emergence of NTEs

67

The technological activities of NTEs

70

A broad picture of technological activities

70

Characteristics of technological activities and user capability building

78

Concluding remarks

84

PART 3 The machinery technology R&D institutes: transforming the established industrial technology institutions 13

14

15

Introduction to machinery technology institutes

87

The importance of the transformation of existing industrial technology R&D institutions

87

Why the machinery industry?

88

Scope of the study

89

The development of the machinery industry in China prior to market reform: industry performance and R&D institutions

93

Introduction

93

Development of the machinery industry

94

Institutions for technological change in the machinery industry

107

Summary: institutional structure and technological learning

133

General trends in the transformation of the machinery industry: the extent and direction of the market reform of government-run R&D institutes

135

Income structure: government funds versus market earnings

135

Characteristics of various components of market earnings

138

Composition of market earnings: machinery technology R&D institutes affiliated to the central Ministry

141

The development of market earnings: machinery technology R&D institutes affiliated to the 143 central Ministry 16

Summary

145

The transformation of the ‘product technology’ R&D institutes

146

The transformation of ‘B’ institutes

147

ix

The transformation of an ‘A’ institute in a sector with rapid technological change

152

The transformation of an ‘A’ institute in a sector with a concentrated enterprise structure 160 The transformation of centrally affiliated institutes in sectors with numerous producers

168

The transformation of machinery technology R&D institutes affiliated to local governments 178 Summary: an overview of the transformations of product technology R&D institutes in the 184 machinery industry 17

18

19

The transformation of the ‘manufacturing technology’ R&D institutes

189

The manufacturing technology institutes

189

Contractual technology development

190

The transformation of an institute engaged in conventional manufacturing technology

191

The transformation of an institute engaged in microelectronics-based automation technology

197

Summary: comparing the transformations of manufacturing technology and product technology institutes

204

Technological trajectories

209

Introduction

209

Technological trajectory

210

Missing elements in the technological trajectory in the pre-reform period

213

Path-shifting and the roles of existing R&D institutes during the current reform

218

Summary

228

Institutional restructuring: internal and external contractual relationships

230

Basic assumptions of the transaction cost theory

230

Transformations of external transaction relationships

233

Transformations of internal organization

243

Concluding remarks: getting the institutions right

249

Notes

253

Bibliography

272

Appendix to the bibliography

280

Index

284

TABLES

3.1 3.2 6.1 6.2 6.3 7.1 A1.1 A1.2 A1.3 A1.4 A1.5 A1.6 A1.7 A1.8 A1.9 8.1 9.1 9.2 11.1 11.2 11.3

Transactions on the technology market (1987–1989) Transactions on the technology market (1993) Income structure of government-run R&D institutes (all industries, 1993) Composition of ‘horizontal’ earnings of government-run R&D institutes (all industries, 1993) Income structure of the remaining R&D institutes (1993) The most important events for economic reform and for the reform of the R&D system Government appropriations for science and technology from the state budget (1953–1991) National expenditure for R&D State-owned R&D institutes, by sector R&D institutes of the Chinese Academy of Sciences R&D institutes affiliated to ministries and commissions of the central government R&D institutes affiliated to local governments Government-owned R&D institutes, by fields (in 1988) R&D institutes of higher education Internal R&D of large and medium-sized enterprises Basic indicators of China’s R&D system The development of Zones and NTEs (1990–1992) Sample Zones of the study (1992) Distribution of technological activities of NTEs: a Zone level summary Typical products of the Beijing Zone Characteristics of NTE’s technological activities (in the area of computer and information technology) 14.1 Selected indicators of the development of the capital goods sector (ISIC 38) in China, India, Brazil and the Republic of Korea (1979–1980) 14.2 Size of selected manufacturing sectors, by share in total value added (1985–1987) 14.3 Size of selected manufacturing sectors, by share in total value added (1985–1987) 14.4 Output, input and productivity in China (1953–1990) 14.5 Comparative productivity by manufacturing branch, China and USA (1980–1990) 14.6 Exports and imports of machinery products (1978–1993) 14.7 Purposes of imports and origin of exports for the Chinese machinery industry (1993) 14.8 Exports, imports, production and consumption of machine tools in selected countries (1992–1993) 14.9 Chinese machine tool imports and exports (1993) 14.10 Industrial technology imports by sector (1950–1990) 14.11 Technology import agreements by category (1979–1990) 14.12 Government-run R&D institutes for the machinery industry in China in the mid-1980s 14.13 Entirely centrally commissioned and financed R&D institutes in the machinery industry, by technological function

20 20 31 32 32 36 41 42 42 43 43 43 44 44 44 49 55 56 71 73 79 95 97 98 100 102 103 103 104 105 105 107 116 125

xi

14.14 Technological development activities in large and medium-sized enterprises in the machinery 127 industry (1992) 15.1 Income structure of government-run R&D institutes in the machinery industry (1993) 136 15.2 Income structure of government-run R&D institutes for industrial technology in all industries 136 (1993) 15.3 Income structure of R&D institutes affiliated to the central Ministry for the Machinery Industry 137 (1993) 15.4 Definition of items of market earnings 139 15.5 Characteristics of various items of market earnings 141 15.6 Composition of market earnings of government-run R&D institutes, for various industrial groups 142 (1993) 15.7 Development of the structure of market earnings for centrally affiliated R&D institutes for the 144 machinery industry, 1984–1994 16.1 Basic statistics of selected industrial sectors (1993) 147 16.2 Change in income structure 1984–1994: Dalian Modular Machine Tool Research Institute 156 16.3 Change in income structure 1984–1994: Shanghai Power Equipment Research Institute 164 16.4 Change in income structure 1984–1994: Shanghai Electric Cable Research Institute 170 16.5 Change in income structure 1984–1994: Shanghai Internal Combustion Engine Research Institute174 16.6 Direction and characteristics of product technology R&D institutes transformation in the 185 machinery industry 16.7 Market size, technological change and the transformation of centrally affiliated product 186 technology R&D institutes in the machinery industry 16.8 Functions of various groups of institutes within the reconstituted technological infrastructure for 188 the machinery industry 17.1 Market earnings structure of the entirely centrally affiliated R&D institutes compared with the sub- 190 group of manufacturing technology institutes (RIMST) 17.2 Change in income structure 1984–1994: Beijing Research Institute for Mechanical and Electrical 193 Technology 17.3 Change in income structure 1984–1994: Beijing Research Institute for Automation in the Machinery 199 Industry A17.1 Change in income structure 1984–1994: Shanghai Material Research Institute 208 18.1 Types of imported technology 214 18.2 Characteristics of the technological trajectory in selected sectors of the machinery industry 215 18.3 Changes in the technological trajectory during the reform: actors and characteristics 221 19.1 Early adaptive strategies and consequent transaction governance conditions 236 19.2 M, I and β costs and choices of adaptive strategy, by institute group 238 19.3 Adaptive strategies for further transformation, group I institutes 240 19.4 Further transformations of internal organization, by institute 247

FIGURES

14.1 Organizational chart of Ministry departments and entirely centrally affiliated R&D institutes 19.1 Simple contractual schema 19.2 Relative stability of technological specialization 19.3 Organizational choice 19.4 Alternative forms of internal organization: influence of the number of niches and their interrelatedness

116 234 242 244 248

CASE TEXTS

3.1 3.2 3.3 4.1

The General Institute of Coal Mines Science and Technology The Automation Research Institute of the Ministry of the Metallurgical Industry (ARTMI) The Research and Design Institute for Chemical Engineering of Fushun city The restructuring of a research institute after merging— the case of the Automobile Technology Research Institute 6.1 The Automation Research Institute of the Ministry of the Metallurgical Industry (ARTMI) 10.1 Legend Computer Group Corporation (Legend), Beijing 10.2 Chutian Optical Electronics Corporation Ltd (Chutian), Wuhan 10.3 Physcience Optoelectronics Corporation of the Institute of Physics, Chinese Academy of Sciences, Beijing 10.4 Open Software System Corporation Ltd of Northeast University, Shenyang 10.5 The establishment and roles of the Wuhan Eastlake New-Tech Enterprise Incubator 10.6 The interface functions of Zone Administrations in securing investment in NTEs, illustrated by the Hangzhou Zone 11.1 Technological activities in Beijing Zone 11.2 Technological activities in Wuhan Incubator and Wuhan Zone 11.3 Technological activity in Shenyang Zone 14.1 The Research Institute for Machinery Science and Technology (RIMST) 14.2 Centrally commissioned and financed product R&D and design institutes for the machine tools and other tools sector 14.3 Product technology R&D and design institutes for the electrical equipment sector, by product categories 14.4 ‘Campaign for the development of precision machine tools’ (1960–1975) 14.5 The domestic development of fossil-fuel power plant equipment (1960–1985) 16.1 The Beijing No. 1 Machine Tool Plant and its Milling Machine Tool Research Institute 16.2 The Dalian Machine Tool Works 16.3 The Dalian Modular Machine Tool Research Institute, of the Ministry of the Machinery Industry 16.4 The Shanghai Power Equipment Research Institute, of the Ministry of the Machinery Industry 16.5 The Shanghai Electric Cable Research Institute, of the Ministry of the Machinery Industry 16.6 The Shanghai Internal Combustion Engine Research Institute, of the Ministry of the Machinery Industry 16.7 The Dalian Machinery and Electrical Research and Design Institute (DMERDI), of the Bureau of the Machinery Industry, Dalian Municipal Government 16.8 The Zhejiang Mechanical and Electric Design and Research Institute (ZMEDRI), of the Bureau of the Machinery Industry, Zhejiang Provincial Government 17.1 The Beijing Research Institute for Mechanical and Electrical Technology (BRIMET), of the Ministry of the Machinery Industry

17 17 18 25 32 59 60 61 62 65 66 72 75 77 117 120 122 129 131 148 152 154 160 168 172 178 180 192

xiv

17.2 The Beijing Research Institute for Automation in the Machinery Industry (BRIAMI), of the Ministry of Machinery Industry

198

ABOUT THE AUTHOR

Shulin Gu graduated from the Chemical Physics Department of the Science and Technology University of China with distinction. During the late 1960s and 1970s she worked as a scientist and engineer in a number of distinguished Chinese institutes, including the Institute of Mechanics of the Chinese Academy of Sciences, where she was Assistant to the Director and worked in the field of chemical dynamics and aerodynamics, and the Central Research Institute of the Beijing Petrochemical Corporation, where she took a leading role in a few areas of petro-chemical catalyst processes. Since the 1980s Shulin Gu has been engaged in science and technology policy studies. In the early 1980s she joined the Institute of Policy and Management of the Chinese Academy of Sciences, where she was Deputy Director, and then Director of the Department of Science and Technology Policy, and was appointed to a research professorship in science and technology policy studies. Through her contributions in areas such as the cooperation between academics and industry, the role of basic and fundamental research and science and technology system reform, she has become well known in the sphere of science and technology policy in China. Since 1992, Shulin Gu has been working at UNU/INTECH as Senior Research Fellow, where she has completed studies on two INTECH projects ‘The politics of technology policy institutions in China’ and ‘The evolution of high technology research institutions and new enterprise organizations in China’. Some outcomes of these projects have been circulated, and have received international notice. Her current interest is the impact of technology on economic development and national innovation systems in developing countries. She has published widely in both Chinese and English.

FOREWORD Bengt-Åke Lundvall, Aalborg University, Denmark

What will happen in mainland China in the next decade? Will it demonstrate the same technological dynamism as the smaller Asian tigers? What is going on in this giant economy in terms of innovation? Are there lessons to be learnt regarding the design of institutions and policies for other former planned economies and for developing countries? Professor Shulin Gu has written a path-breaking work on the Chinese innovation system that helps us answer these questions, so crucial for the global economy. Her book reflects her unique experience first as scientist and high level expert on science and technology policy in China and since 1992 as research fellow at UNU/INTECH in Maastricht. The general theme of the book is the close interdependence between market reform and organizational change. It is convincingly demonstrated that market reform cannot stand alone. If it is not combined with new modes of organization within technology institutes and with new organizational relationships between the institutes and the firms it does more harm than good. The book gives both the broader picture and very detailed information about how market change has interacted with organizational change in the case of China. Both successes and failures are registered and presented as steps in a process of policy learning at different levels of the innovation system. The book is structured into three parts. The first part gives a broad overview of reform policy in China with the emphasis on the period after 1985 up to the current period. The second part analyses the process of spin-off of new firms emanating from R&D institutions resulting in new technology enterprises mainly specializing in information technology. The third part focuses on how technology institutes, especially in engineering sectors, have been successfully transformed into knowledge-based firms producing machinery products for the market. The first part of the study shows the rapid transformation of Chinese policies in relation to its science and technology infrastructure. After a period of rehabilitation (the end of the 1970s) where science and technology capabilities that became disorganized during the cultural revolution were re-established, the 1980s, and especially the period after 1985, saw important shifts in policy strategy. First, there was a wave of merging institutes into existing firms, second came the period of spin-off where many new start-ups saw the light and, finally, technology institutes began to be transformed into corporations with a strong in-house engineering capability. Shulin Gu demonstrates how these different phases were accompanied by new regulatory regimes making the transformation possible. The second part of the study goes into some detail regarding the second phase. It is demonstrated that the spin-off phenomenon has been a major prerequisite for the creation of a great number of New Technology Enterprises (NTEs) and for the formation of a computer and information technology industry. Especially interesting is that some of the new capabilities created were in software rather than hardware (including the

xvii

transformation of US software programmes into Chinese). The author concludes this part of her analysis by emphasizing the general observation that for many developing countries facilitating spin-offs from public institutes may be one of the few operational ways to establish a dynamic interaction between the technology infrastructure and industry. The third part is focused on the part of the technological infrastructure directed toward the machinery sectors. It is a diverse set of institutes and the ways they have been transformed into enterprises are also diverse. Some become specialized in engineering services while others become niche producers. Shulin Gu uses transaction cost analysis to explain the different patterns of transformation but her analysis also highlights that the transformation is crises-ridden and costly and that basically it reflects a combined process of technological, institutional and policy learning. A general conclusion of great importance is that it is not enough to set up a new legal framework in order to make the transformation successful. To facilitate and stimulate the process of organizational change is a key action for avoiding extreme costs of transformation. The concept of national systems of innovation is a reasonably recent one developed in the last part of the 1980s. The author’s analysis of China is a major contribution to the further development of this concept. Her analysis makes the concept operational and demonstrates its analytical potential. One major strength of the analysis is that, in spite of its systemic, structuralist approach, it remains dynamic emphasizing the complexity of change and learning. Shulin Gu’s book on China deserves a wide audience among academics in innovation theory, development economics and economic history. Policy makers in charge of science and technology policy in all parts of the world will also have lessons to learn from how the technology infrastructure of the People’s Republic of China has been radically transformed during just one decade.

SERIES EDITOR’S PREFACE

Professor Shulin Gu, who has been a valued member of UNU/INTECH for the past five years, has produced this major study of science and technology policies in the People’s Republic of China. It is, in my view, a striking work of scholarship—carefully crafted, relevant and deeply interesting. I find it a valuable piece of work for a number of reasons. In the first place, Professor Gu has a range of knowledge and basic information about the Chinese situation which is special. She had direct professional experience of policies followed in the People’s Republic of China in the years before the reform. Her access to this early time has produced an authoritative account, based on Chinese sources, of a period of history which up till now has been covered—in the West, at least—by scholars who, perforce, worked from secondary and often informed sources. Subsequently, Professor Gu launches into a comparative discussion of the impact of reforms. She is led to emphasize that however radical the impact of economic liberalization may be in integrating science and technology to socially useful innovation, direct state intervention in the transformation of institutes has also been necessary. This is an informed and well argued empirical conclusion. In working out its historical basis, Professor Gu has given us a comprehensive analysis, covering the early pre-reform period, the reform process as it occurred in Chinese research institutions and the subsequent recent outcomes. The historical range of the analysis is another reason why the book is striking. And finally, one cannot help but be impressed by the wide range of approaches and analytic ideas on which Professor Gu has drawn from outside the usual boundaries of ‘China studies’. She brings several literatures to bear— old and new literature on the economics of innovation, for one example; the literature on institutional organization of science and technology, for another, and also the rapidly growing literature on ‘national systems of innovation’— to which this book should be considered a significant contribution. It is—to conclude—a book about policy, about vast experiments in policy for the productive use of science and technology in a country which is still materially poor (albeit scientifically rich). It is correct— from the standpoint of an institute like UNU/INTECH, devoted to policy studies—that the book should be so focused. We hope that it will have a real utility for Chinese policy makers as well as in other countries, where the problems it addresses are also very relevant. Furthermore, we hope the book will have the same attraction for policy researchers in the field in general, as it has within UNU/ INTECH. Professor Charles Cooper Director, UNU/INTECH

AUTHOR’S PREFACE

This book presents the outcomes of two UNU/INTECH projects: ‘The politics of technology policy institutions in China’ and ‘The evolution of high technology research institutions and new enterprise organizations in China’, conducted from 1992 to 1996. The intention is to address systematically some critical aspects of the transformation of a national innovation system with a special focus on issues encountered by a developing country, whereas previous studies of restructuring in national innovations systems have focused on industrially developed countries. Intensive field surveys in the People’s Republic of China for firsthand materials, as well as my intimate knowledge of the Chinese reform process, provide a sound and detailed empirical basis for the study, while insights developed in fields such as technological innovation economics and transaction cost economics are used in the analysis to discover points on which these perspectives may be applicable to the complicated Chinese system. The book is divided into three parts. An overall review of reform policy is provided in Part 1, and two major approaches to institutional restructuring are discussed in Parts 2 and 3. The three parts complement each other as variations on a single theme, of change on a spectacular scale: how the biggest industrial technology system which has ever emerged in a developing country has been modified greatly, in both its institutional patterns and in the direction and characteristics of its technological activities, in response to the market-oriented economic reform. Readers who are interested especially in one of the three topics may go directly to that part, since each part has been presented as a complete essay. This book should be of interest to researchers in the fields of science and technology policy, national innovation systems, the economics of technological innovation, institutional economics and industrial organization, as well as to policy makers, policy advisers and practitioners in developing countries and to international organizations with an interest in policies for science and technology, industry or trade. Part 1: Reform policy for the S&T system in China: from paid transactions for technology to organizational restructuring The first part of this book reviews reform policy for the science and technology system in China. The People’s Republic of China had committed significant resources since the 1950s to establishing extensive government-run R&D institutions to support industrial technology innovation in particular, and for the development of science and technology in the country in general. This resulted in a legacy characterized by the separation of the industrial technology system from its industrial users, a feature that is to varying extents pervasive not only in former centrally planned economies, but also in a number of developing economies in which the governments took a direct responsibility for industrial technology. In parallel with

xx

economic reform, a reform programme for the science and technology management system was initiated in China in the mid-1980s. This programme sought to link the R&D institutes to productive units by creating a market for paid technological services as a substitute for planning coordination. The results showed that the establishment of a technology market is only part of the transition process. Organizational restructuring has been involved in the transition, and the direction and success of this restructuring are closely related to policy measures as well as to adaptive responses by the generators and users of technology. The historical examination of reform policy in Part 1 asks these questions: 1) How did the demands on these technology R&D institutions change because of the changing economic environment and international relationships, and how were these changed demands perceived by policy makers? 2) What were the purposes of reform policies and the measures initiated by the policies, how did the R&D institutions and their users respond to the policies and for what reasons, and how were the reform policies amended in response? 3) What has come out of these interactions between the policy makers, the R&D institutes and their users? Part 1 shows that market reform requires organizational restructuring of the industrial technology system, and the restructuring usually involves integrating more activities within the territory of a previous R&D organization. This is in contrast to the broadly accepted view, on which many presently preferred reform policies are based, which emphasizes the soundness of pure market intermediation and gives little attention to organizational restructuring even when there is evidence of the need for restructuring. Three distinct approaches to organizational restructuring are identified: 1) merging R&D institutions into existing enterprises; 2) spinning-off new technology enterprises (NTEs) from R&D institutions; and 3) transforming entire individual R&D institutes into manufacturing or engineering corporations with intensive in-house R&D and design. The policies to facilitate these kinds of restructuring evolved along a time sequence, from merging R&D institutions into enterprises in 1987, to encouraging spinning-off in 1988, to the transformation of individual R&D institutes since the 1990s. The emphasis of policy measures also changed over time, concentrating first on one particular restructuring approach and then on another. This has been a process of policy learning, involving intensive experiments and interaction between policy makers and economic actors. As a result, the Chinese policy system has shifted from the direct policy intervention used before the current reform to indirect support and guidance, by setting up new regulations and incentives and creating regulatory and other supporting institutions. The creation of regulatory and supporting institutions, which can be observed to involve intensive institutional learning, is especially crucial in a transitional period, when these institutions are typically inadequate. Intensive policy learning and institutional learning have been crucial, underlying the refinement of the technology market mechanisms and the development of plural restructuring approaches. This Part has already been published, in a substantially finished form, as INTECH Working Paper 17. Part 2: Spin-off enterprises: channelling the components of R&D institutions into innovative businesses The second part of this book considers the spinning-off of new technology enterprises (NTEs), one of the three approaches to organizational restructuring mentioned above. Spinning-off gave birth to a unique and important part of the ‘non-state’ owned sector in China during the reforms. After emerging in the mid-1980s, the number of NTEs grew rapidly to about 10,000 by 1993, when the NTEs employed about 100,000 scientific and technical staff. This restructuring approach played a decisive role in China’s entry into the computer and information software industry, enabling the widespread applications of computer and information technology in the 1990s.

xxi

A systematic survey of the institutional dimension for the initiation of the NTEs shows how the existing organizations can play an important role in restructuring during a period of transition if there are proper policies to induce appropriate actions. In the years when NTEs were first evolving, the entrepreneurship of R&D institutions contributed to the birth of most of them. R&D institutions transferred personnel, technical and financial assets, while R&D institutions and local governments together provided infrastructural services for regulatory, monetary and incubatory purposes for the NTEs to grow. Thus the NTEs emerged from, and were assisted by, various elements of the old system, which had been freed to act by the decentralized delegation of decision-making authority. A survey of the technological activities of the NTEs reveals that the great majority of the NTEs are engaged in computer and information technology and concentrate on applications of the technology. Small systems development is the major feature of the technological activities, which promote ‘user capability building’. One of the most conspicuous achievements in this regard has been the adaptation of Englishlanguage-based computer and information technology into a Chinese-language context. This applicationoriented development by NTEs is in contrast to the hardware manufacturing orientation and military and scientific research purposes that the Chinese industry pursued in the pre-reform period. In the market place and based on autonomous business organizations, the direct encounter with users’ demands made the NTEs responsive to user capability building. The findings have important policy implications. Encouraging spin-offs could be a promising orientation for reform policies to restructure a largely mismatched R&D system, a demanding issue encountered by many transition economies. It could result in strengthening application capabilities for computer and information technology in former centrally planned economies and in some developing economies which are now in transition. It has been recognized that, for these economies, a weakness in application capability is the major supply-side impediment to their benefiting from the information revolution. Indeed R&D institutions are often the only suitable domestic resource in developing countries from which this strengthening of application capability can begin. This part of the study has already been published in a substantially complete form as INTECH Working Paper 16. A journal article in the Journal of Development Studies Vol. 32 No. 4, 1996, explores the findings of the study, especially those concerning China’s entry into the computer and information software industry. Part 3: The machinery technology R&D institutes: transforming the established industrial technology institutions Part 3 examines the third of the three approaches to restructuring, the transformation of established industrial technology R&D institutes on a whole institute basis. The machinery industry which has been chosen for the examination was one of the key industries in China prior to the current reform, and continues to be very important to China, as in other developing countries. There are almost seven hundred R&D establishments specializing in machinery technology, employing more than fifty thousand scientists and engineers. This represents roughly one fourth of all government-run industrial technology R&D institutions. After an introductory chapter, Chapter 14 of Part 3 provides background information on the development and performance of the Chinese machinery industry, and the R&D and design institutions for this industry in the pre-reform period. Both the nature of technological innovation in that period and the institutional framework in which it occurred are described.

xxii

A rich set of empirical evidence is analysed. Detailed case studies were conducted through field surveys at a number of institutes chosen to represent four subsectors (machine tools, fossil-fuelled electricity generation equipment, electrical cables and wire and small internal combustion engines) as well as each of the various types of affiliation between an institute and a local or national government body. Official statistical data on institutes’ income structure and other resources were used to verify the results from the field survey, so that the observations regarding changes in the institutes, and at an aggregate level of it, may be regarded as robust. Chapters 15, 16 and 17 of Part 3 present the empirical evidence. Chapter 15 offers an overview of the general trends in the transformation of the machinery technology R&D institutes. It also contains definitions and specifications of the indicators which are used in the official statistics on institute income structure. Chapters 16 and 17 present case materials, Chapter 16 for the institutes which previously specialized in ‘product technology’ and Chapter 17 for those which previously specialized in ‘manufacturing technology’. These two chapters conclude with summaries of analytical and policy implications in relation to the two functional groups. Each case text in these two chapters may also be read as a short historical review of the institute and of the subsector which the institute was assigned to serve. Chapters 18 and 19 go on to generalize observations from the empirical survey. As the empirical evidence has shown, the adaptations demanded of an established institute in the radical plan-to-market transition are no easy matter. Two aspects emerge as critical in this complicated transformation: the reorientation of technological change, and restructuring internal and external relationships. By the mid-1990s the machinery technology R&D institutes had fundamentally modified themselves through changes in both aspects. Some institutes became commercial developers of machinery engineering services, some had become in-house R&D departments for a host enterprise, and some became niche producers of machine products. Chapter 18 explains the re-orientation of technological change, described as a shift in technological ‘trajectories’. As a result of the trajectory shifting the industry became more responsive to changes in ‘specific’ technologies whereas the previous orientation was selectively sensitive to ‘general’ technologies. The shift can be attributed to both the incentive structure and changes in the institutional arrangements. Chapter 19 interprets the transformation of the internal and external relationships of the R&D institutes. Applying the analytical framework of transaction cost theory, the various restructuring strategies taken by different institutes at different stages of their transformation are explained as outcomes of rational strategic choices between alternatives which involve different transaction costs and other costs. The various strategies are responses to crises posed because of the radical changes in macro-economic parameters in which they operate. The most serious crisis, which endangers the productive value of specific assets accumulated in an R&D institute, appears in the early stages of the transition from a planned to a market economy when the specific assets of an institute have to be contracted out, while effective governance for the contracts has not yet developed. The R&D institutes in the study did encounter the greatest difficulties in making adaptive decisions during the early stages of transformation. This may well be an important reason why a transition process often involves very high costs in terms of the loss of technological strengths and international competitiveness. Part 3, like the studies reported in Parts 1 and 2, shows that extremely intensive learning is the essence of a positive process of transition. Technological learning drives the re-orientation of technological activities. Institutional learning is required by the changed economic environment, and provides necessary support to the technological re-orientation. Policy learning assists the development of the policy capacity to introduce radical reform programmes and to elaborate the policy framework in response to the immense and novel issues arising from the radical reforms. Technological learning, institutional learning and policy learning can only proceed through adaptive processes, and the adaptive decisions are largely made and implemented

xxiii

in micro-level organizations. By identifying the significance of micro-level learning and adjustment, the study supports the argument for ‘getting the institutions right’ which gives priority to factual (de facto) organization rather than merely to legal (de jure) organization. Focusing on the latter has led, in the conventional view, to an emphasis on the importance of ‘getting the prices right’ and ‘getting the property rights right’. But as the study demonstrates, if de jure changes are to be consolidated and endogenized, and so become really functional, these have to happen in micro-level de facto organizations. The more radical the de jure change that a reform programme introduces, the more complicated the adjustments required in de facto organizations. ‘Getting the institutions right’ hence provides the right framework for policy making and policy analysis, and one which is especially pertinent to economies in which programmes for profound systematic transition and industrial development are being introduced.

ACKNOWLEDGEMENTS

The author is greatly indebted to a number of persons and organizations. Charles Cooper has been the leader of academic efforts at UNU/INTECH. The design of the research projects, the development of the methods and concepts applied in the research, and the identification of the central questions to be addressed have all incorporated his sensible guidance. Nathan Rosenberg has had a great influence on the author’s understanding of technological innovation economics since the mid-1980s, through his published works and through continuing personal exchanges. He suggested the relevance of the transaction cost school to the studies reported in this book, and introduced literature on a number of themes, including the Soviet machinery industry, comparative economics of technological innovation in East and West, and even economic reforms in China. W.Edward Steinmueller has been a close critic of the author since 1994. His succinct comments have driven the author towards more accurate and refined understandings on a wide range of topics. Steinmueller’s comments on transaction cost explanations for the abrupt shift in the external contractual relations of R&D institutes when planning coordination are withdrawn, and his concerns about the development of customized machinery technology have been crucial to the study. Martin Bell and Zhu Sendi made their intellectual contributions in the implementation of the two INTECH projects during 1993–1994 and 1994–1996 respectively. Martin Bell joined part of the field survey in May 1993, working as a co-researcher with the author in the development of an interview questionnaire and in annotating part of the field survey materials. He has focused on the trends in integrating industrial R&D with manufacturing and other activities, which has been one of the pivotal ideas in the project design. Zhu Sendi, Chief Engineer at the Ministry of the Machinery Industry of China, with his profound knowledge of machinery technology and the Chinese machinery industry, provided valuable advice for the case studies, as well as advice regarding the collection of data and literature for the study presented in Part 3. He has also responded to many inquiries by the author regarding the technology, organization, management and history of the case institutes and the industry. This has provided vital support in fields where the relevant knowledge has not yet been adequately documented. In addition, the author wishes to express her gratitude to Richard P.Suttmeier, David Mowery, Paul David, Ye Dan, Shao Xinping, Ye Xiandong, Morris Teubal and Jorge Katz for their comments and reflections during and following the UNU/INTECH Workshop ‘The Restructuring of Industrial R&D Institutions in China’, held in Maastricht, 29–30 June 1994. Yang Wanhong provided assistance during the author’s field work in May to August 1993. Thanks are also due to the Institute of New Technologies of the United Nations University (UNU/ INTECH), the institute which provided the time, facilities and various kinds of administrative support for the two projects reported in the book, which are part of UNU/INTECH’s research plan. The State Science and Technology Commission of China through its Department of International Cooperation, the Research Institute of Policy and Management of the Chinese Academy of Sciences, and the Ministry of the

xxv

Machinery Industry of China supported and assisted the organization of field surveys carried out in May to August 1993, October 1994 and April to May 1995. The author acknowledges the enthusiastic responses and kind hospitality provided by engineers, directors and managers at a number of research institutes, NTEs, industrial enterprises and Development Zones for New Technology Industries in Beijing, Shanghai, Dalian, Shenyang, Wuhan and Shenzhen in China. Sen McGlinn has done excellent editing for the whole book. Finally, the conventional disclaimers hold. None of the persons and organizations acknowledged above have any responsibility for errors and inaccuracy in the book, and the opinions expressed should not necessarily be regarded as those of UNU/INTECH. Shulin Gu Maastricht 31 September 1997

Part 1 A REVIEW OF REFORM POLICY FOR THE S&T SYSTEM IN CHINA From paid transactions for technology to organizational restructuring

1 INTRODUCTION TO REFORM POLICY

The need to address reform policy Part 1 is intended to open discussions about science and technology policy, as part of market reform exercises, with a particular focus on China. Because of the great complexity and diversity of the R&D system in any modern country, it is limited to policy issues relating to industrial technology R&D. It is not the intention to consider academic (theoretical) R&D, or R&D in fields such as agriculture, health care, environmental protection, calamity prevention, etc., except to the extent that cross-over is unavoidable. When we have to deal with policies whose purpose is more general, the analysis will concentrate on those elements which are closely related to the transformation of R&D institutions for industrial technology. The terms ‘market reform’ and ‘market-oriented reform’ are employed here as synonyms, and in their most basic sense. They refer to the introduction of the basic rule of the market that transactions are determined in the market place through negotiations between the buyer and seller. This is genuinely crucial to the transition of China’s industrial technology R&D institutions, which were organized and developed in a substantially different regime. The central concern of the study is that market reforms require the restructuring of industrial technology R&D institutions. This is in contrast to the generally accepted view, which is usually restricted to the pure use of market power. This preoccupation has led to simplistic policy formulations that amount to no more than relying upon the expansion of a ‘free’ market. This has already considerably weakened the effectiveness of policies for the R&D system. The industrial technology R&D institutions inherited by countries currently undergoing market-oriented reforms are basically characterized by the separation of the industrial R&D and innovation system from its users in production. As several authors have indicated, these systems, in centrally planned economies, were characterized by 1) the centrally planned governance of investments related to industrial technology and 2) extensive externalization (outside firms) of elements relating to industrial technological changes such as research, development, design, standardization, etc. (Hanson and Pavitt 1987:25–30). This pattern of institutional arrangement developed mainly in the 1930s and 1940s, with the inception and consolidation of the first centrally planned economy in the former Soviet Union, and was later adopted by about a dozen centrally planned countries, including China, after the Second World War. A number of countries in Latin America, Asia, and later Africa, began to build up their own industrial R&D capabilities from the 1950s, following rather similar patterns. Their R&D capabilities were established in centralized institutions, and financed and operated by governments. The roles and effectiveness of these centralized industrial research and service institutes have been continuously debated.1

INTRODUCTION

3

Numerous attempts are reported to have been made by the Soviet Union and other Eastern European countries from the 1960s to the late 1980s, most of which were focused on intensifying links between the separate organizations and functions necessary for technological change. These efforts can be illustrated by the hard push for the establishment of ‘science-production associations’ in the former Soviet Union. Generally three types of measures were used by the Soviet Bloc countries until the late 1980s, all aiming at improving the efficiency of the centrally planned systems: 1) to merge the originally separate research institutes either with groups of enterprises or with large individual enterprises; 2) to institute full-cycle planning for research work, with each major project having planning targets to cover the development of technology and its application in production, specifying both users and expected effects; and 3) to increase incentives to both researchers and users of industrial technology, by means of favourable pricing, bonuses, etc. This packet of measures has been called by some authors ‘efforts under a planned regime’.2 The policy issues addressed here became sharper when the former centrally planned economies began to take drastic steps toward market reforms. Since around 1990, economic plans have been radically abandoned in these countries, and the funds granted from government budgets to both enterprises and R&D institutes have been dramatically reduced. R&D institutes are forced to sell themselves on the market. ‘Inability in selling themselves’ (on the part of specialized and former government-financed R&D institutions) was then widely seen as the most pressing difficulty, and this became the focus of attention in science and technology policy communities in these countries and worldwide.3 Assuming that there is a problem of ‘inability in selling R&D’, the next question which is commonly asked is, ‘Are the limited market functions sufficient to support market reform for the R&D system?’4 Authors who have posed this question begin with the reality that R&D institutes have proved to be bad sellers, which they argue is due to low demand from users and the inadequacy of legal protection for intellectual property. They conclude that government appropriations are urgently needed to prevent the existing R&D organizations being dispersed and lost. It must be admitted that the market institutions are definitely poor, and that the users of industrial R&D are weak, at a time when the reform efforts are just starting to transform them. In practice, the question raised by the debate is: Should the reform process go back to the old system? Can the transformation of the science and technology (S&T) system, which has proved to have limited abilities in selling itself and dealing with the market, proceed on the basis of a market orientation? An OECD report has put the question in slightly different terms, and suggests an answer: ‘The problem…was whether measures were designed to keep capacity or the present structures. The latter…were judged…to be unsuitable for S&T development. Government should fine-tune approaches designed to preserve capacity and to change present institutional arrangements’ (OECD 1992:168). Apart from the practical question of the ability of the S&T institutions to work through the market, there are theoretical questions about the suitability of the marketplace for mediating between the suppliers and users of technology and knowledge. Many presently preferred policies emphasize the soundness of market intermediation, assuming perfection in the technology market itself. There are reasons to question this assumption. In industrially-developed market economies, industrial firms are the institutional basis for industrial technology (Freeman 1982, especially section 5; Kline and Rosenberg 1986). A large part of industrial R&D and designing is internalized within firm organizations. The uncertainty of technological innovation, and the tacitness of technological knowledge, have favoured internal organizational mechanisms rather than pure market mechanisms. It is argued that the commercial success of industrial technology depends on continually seeking to match uncertain technological opportunities to changing market possibilities, and the match can be realized more easily within firms, with better information feedback between the various

4

REFORM POLICY FOR THE S&T SYSTEM

activities. This internalization has developed spontaneously in market economies, underlining the imperfections of the pure market mechanism in dealing with technology transactions. Williamson has discussed these imperfections extensively in the context of ‘market and hierarchy’ (Williamson 1985, especially Chapter 2). Market reform requires institutional restructuring, because the old institutions were developed to fit within the old economic regime. If the necessity of institutional restructuring was ignored (i.e., if R&D institutions were expected to produce the same products as in the past, within the same structures, but to sell them in a technology market), policy making would be unlikely to respond to the widespread evolutionary movement already underway towards the integration of production and research functions within organizations. For instance, in Hungary, during the last two decades ‘all types of research institutes adjusted themselves to the opportunities offered by business and restructured their activities accordingly… The R&D institutes themselves started manufacturing new equipment instruments.’ ‘In the 1970s production had become a common activity of R&D institutions…the market regulators [i.e., rules] made them interested in “in-house” production.’ ‘Thus, frequently, the mass production goods of the research institutes were also sold…’ (Balazs 1993).5 In East Germany, approximately 100 ‘research companies’ had been established from elements of former combines by November 1990 (Bentley 1992:156). Such realities have often been seen, in the light of neoclassic theory, as evidence of extraordinary chaos. This study, therefore, addresses industrial technology R&D reform policy, with the focus on the what in a conventional view would be regarded as ‘extraordinary’ institutional restructuring. The approach of the study—empirical observation of the historical evolution of reform policy in China No systematic framework has yet been developed for analysing reform policies. Recently the OECD has organized studies on science, technology and innovation policies in some formerly centrally planned economies, addressing transitional issues. These studies seem to be at an early stage, of seeking to describe and define problems through empirical examination.6 An analogous approach is taken in this study which, from an empirical survey of the formulation and implementation of policies, and responses to policies, is intended to provide ingredients for the conceptualization of the subject. China is a promising subject country because the reform has been going on there for fifteen years with little interruption, and may be supposed to illustrate the ‘natural’ evolutionary process in the medium term. Our study differs from the OECD projects in two respects: it is elaborated in a longer historical context, with greater attention to the causes and consequences of important policy making since the market-oriented reform began, and is more narrowly focused on R&D institutions for industrial technology, a segment of the national science and technology (S&T) system which was more urgently faced with the need for restructuring. The historical examination of reform policy asks these questions: 1) what demands to reform the R&D institutions arose out of the changing economic environment and international relationships, and how were these demands perceived by policy makers? 2) how were the related policies formulated and what measures were devised for the implementation of the policies? and 3) how did the R&D institutions respond to the policies, and how were the reform policies amended in response? These questions imply an examination of the most important events in the development of reform policies for the R&D system, in relation to the most important changes in economic reforms, at the expense of overlooking many other facets which may also have been influential, but to a lesser extent.

INTRODUCTION

5

It will be evident that an historical perspective brings considerable benefits. It reveals an evolutionary sequence that highlights several steps, moving successively forward to adjust the S&T system in accordance with the economic reforms. In China, market-oriented reform was begun in the late 1970s, starting with the agricultural sector. This was followed by reforms for industrial sectors, launched in the first half of the 1980s. Following on from this process, market-oriented reform for the S&T system has been in place since 1985, when the Decision on S&T Management System Reform was promulgated. Many observers have stopped there, but a close examination will show that institutional restructuring has been developing ever since, in parallel with the expansion of market dynamics. The restructuring of R&D institutions thus far has produced a few recognizable forms: 1) merging R&D institutions into existing enterprises; 2) spinning-off new technology enterprises from R&D institutions; and 3) transforming whole individual R&D institutes into manufacturing or engineering corporations with intensive in-house R&D and design. Policies for facilitating these kinds of restructuring were created in a clear time sequence, from merging R&D into enterprises in 1987, to spinning-off in 1988, to the transformation of individual R&D institutes since the 1990s. The different mechanisms underpinning these transformations will be explored, along with the review of policy initiatives, in order to illustrate what kind of ‘fine-tuning’ reform policies were developed, in interaction with particular kinds of restructuring, at each stage of the process. This part of the book is organized in seven chapters. The remainder of this chapter will sketch changes in the economic environment since the end of the 1970s, and outline the institutional heritage of the system for industrial R&D in China, so as to provide a broad background for the following chapters. Chapter 2 introduces the main policy initiatives from the end of the 1970s to 1985, as a basis for comparison with those taken after 1985. This period saw an overall rehabilitation and enhancement of the planning apparatus and of state-run R&D institutions, to meet ambitious economic targets. During this period, market-oriented economic reform started in the agricultural sector and followed in the industrial sector. Chapter 3 analyses policy measures stipulated by the Decision on S&T System Reform in 1985. This launched the systematic introduction of technology market mechanisms into the operation of R&D institutes. Chapter 3 also outlines some responses by R&D institutes to the technology market. The following chapters deal with policies for various kinds of restructuring of industrial technology R&D institutes. Chapter 4 addresses restructuring by merging R&D institutes into existing enterprises; Chapter 5 is on the process of spinning-off enterprises; and Chapter 6 deals with the transformation of entire individual R&D institutes. Finally, the concluding chapter summarizes findings and suggestions for further study. Information for the study is drawn from policy documents, S&T statistics, and journalists’ reports, incorporating intensive interviews conducted in the summer of 1993 and over the previous ten years when the author worked on S&T reform policies with some important policy agencies in China. Changes in the economic environment since the end of the 1970s The centrally planned economy in China was introduced in the 1950s, along with massive imports of industrial technology from Soviet Bloc countries. The institutionalization of the R&D system in China was accomplished in the same period, along lines which were broadly coherent with the economic system. This R&D system existed for twenty years, with only marginal evolution during that time.7 Nevertheless, there were many efforts to reduce the degree of mis-matching in the system and to adapt it to Chinese conditions. These efforts were at times intensive, such as during the radical ‘Chinese style’ revolutions of 1958–1960 (‘Great Leap Forward’) and 1966–1976 (the Cultural Revolution).8 However, the reality was that after each

6

REFORM POLICY FOR THE S&T SYSTEM

radical revolution, the basic rules of the established institutions remained unaltered, while retaining Chinese characteristics which were different from other versions of the planned economy. The present reform in China was initiated at the end of the 1970s. In 1978, the top leaders of China decided to revise the objectives of economic development to give more attention to: 1) the efficiency of economic growth, at the same time as accelerating the speed of economic development; and 2) improvement in the people’s living standard. To realize these objectives, it was declared that political principle should no longer guide economic affairs (CCCPC 1981). Two accompanying changes in the strategy of economic development were crucial. First, the overwhelming priority of heavy industrial development was replaced with a more balanced pattern, with consumer goods and service sectors getting a greater share of resources. This stimulated the decentralization and diversity of the economy, and accelerated the development of local and non-state owned industry. Second, an open-door policy to the international community was announced, leading to a dramatic departure from the isolation from international participation which had prevailed during the 1960s and the first half of the 1970s. China has since taken corresponding steps to remove rigid planning controls and has increasingly introduced market elements.9 The market-oriented economic reforms have had vital effects on the existing S&T system, largely recasting the relationships of the system to the firms using its output. The first effect followed from the rural economic reforms. By the end of 1981, the previous ‘commune system’ had broken down. More than 90 per cent of farmers had begun to work under a ‘household responsibility system’ whereby land was entrusted to peasant families. They received more freedom to use the land and to sell their output in a free market, after having met a grain quota for the state. Moreover, rural residents were allowed to run various non-cropping businesses such as fish farms, animal husbandry, transportation, construction and industry. The rapid growth of rural industry, in particular, created huge demands for industrial technological expertise which were entirely beyond the capacity of the old institutions. By 1989 there were nearly one million township and village enterprises, employing more than fifty million workers (China Statistical Yearbook 1990, English version: 387, 390). Second, the extent of the state plan steadily shrank, and state-owned enterprises were granted more autonomy and thus took more responsibility. This occurred following the success of the agricultural reform. In 1984 the ‘Decision on Reform of Economic Management System’ was implemented. This granted stateowned enterprises autonomy in their product portfolios, marketing, purchasing, staffing and pricing (CCCPC 1984), but not in capital investment and enterprise equity. A new tax system allowed enterprises to retain their profits after taxation, for the first time since the 1950s. This redefined the relation between firms and the state significantly, although enterprises were still subject to many constraints (Yuan Baohua (ch.ed.) 1985: 342–343). Around 1987, a ‘management contract system’ for medium and large industrial enterprises, and a ‘leasing measure’, mainly for smaller (state-owned) enterprises, were widely implemented, with the aim of clarifying and consolidating firms’ autonomous responsibility on a contractual basis.10 The operation of state-owned enterprises was therefore increasingly moved out of administrative governance.11 As a result, the domestic users of industrial R&D were able, to a significant extent, to make their own decisions as to what should be bought, from whom, and what activities should be undertaken ‘in-house’. This was a fundamental alteration in the supplier—user relationship for industrial technology. Third, due to the open-door policy, China’s international economic links were substantially intensified, in terms of the inflow of investment, technology and capital goods, from the late 1970s on.12 Importation became the most important source of industrial technology, so that domestic suppliers faced fierce competition. To give an approximate idea, one survey indicates that at present about two thirds of the

INTRODUCTION

7

technology employed in production in the machinery industry is directly acquired from overseas suppliers (interviews at the Ministry of the Machine Industry, Sept. and Oct. 1994). The institutional heritage of industrial R&D institutions It is proper to summarize some of the institutional inheritance of the R&D system developed under the centrally planned economy, which has had a profound influence on the restructuring in the subsequent period of market reform. The system was at first basically transplanted from the Soviet Union, along with a vast inflow of industrial technology in the 1950s. This system was then consolidated by, and within, the centrally planned system in China, whose administrative machinery had complete authority in economic decision-making, including decision-making for the R&D system. The R&D system became an important component of the system as a whole, and continued to operate under planning control until 1985. As has been widely recognized, the separation (or externalization) of the R&D units from the production units is one of the most prominent organizational features of this kind of system. In comparison with countries with similar per capita GNPs, but under market economic regimes, this system was generally overextensive in terms of quantitative indicators such as numbers of institutions, manpower, expenditure. In China, there were more than 4,500 ‘independent’ R&D institutes affiliated to the governing machinery at levels higher than ‘county’13 in 1985, of which more than 2,000 were engaged in industrial technology (see Table 1.7 in the Appendix to Part 1). In this system, other functions necessary to technological changes were also segmentally organized within the administrative framework, the most notable being the plant design institutes.14 In addition to these generally recognized features, other institutional characteristics which appear not to have been adequately discussed thus far in the literature will be briefly mentioned below. Integral position in the administrative framework Having long been subordinate establishments, these institutions were ‘locked in’ to the administrative framework. Industrial R&D institutes were locked in to the specialized departments of industrial ministries or bureaux. Other functional institutions were locked in to other departments or ‘bureaux’. Design institutes, for instance, were locked in to ‘capital construction’ departments. This is characteristic of China, where the planned economy has a ‘stronger local authority’ (Tidrick and Chen Jiyuan 1987:180–186), and where the R&D system was widely extended, and separately locked in to different levels of administration (central and regional).15 This locking-in had a strong influence on the behaviour of industrial R&D institutes in the process of reform, both positively and negatively. Strong orientation to physical products Since they served industrial production units which were weak in in-house R&D, any output presented as generalized knowledge, theory or method was unlikely to be understood and employed by the firms. The only way was to make a physical prototype and perhaps even install it directly on site. This pulled R&D institutes very much ‘downstream’.16 It is safe to estimate that the majority of the work performed in industrial ‘R&D’ institutes was not R&D.17 The situation is also found in other formerly centrally planned economies.18 As the market reform proceeded, this characteristic favoured the transformation of some

8

REFORM POLICY FOR THE S&T SYSTEM

industrial R&D institutes into profitable businesses. Market-profitable manufacturing was easily started in the pilot plants or trial production workshops which already existed inside these institutes. The weakness of enterprises Because they were assumed to be pure ‘production’ units, the industrial enterprises were weak in R&D and design, and also in marketing and information collection. Besides, the enterprises had no means of financial accumulation. One should conceptualize the enterprises under a centrally planned economy as entirely different from the enterprises in market economies. The R&D institutes are also entirely different.

2 THE PRE-‘DECISION’1 PERIOD Reformation of planning practice (1978–1985)

Though the revision of economic development priorities and the reform of the agricultural and industrial sectors had begun in the late 1970s, as outlined in the previous chapter, science and technology policy in China was still run on rather traditional lines. Between the late 1970s and the mid-1980s, two initiatives were taken by the Chinese authorities in accordance with the revised economic strategy. One was the rehabilitation and improvement of R&D institutional establishments which had existed prior to the Cultural Revolution (1966–1976). This initiative was realized between 1978 and 1980. The second initiative was the elaboration of a planning system for S&T activities, which was begun in the first half of the 1980s, in parallel with the formulation and implementation of the Sixth National Five Year Economic Plan. This chapter discusses these two major initiatives to provide a glimpse of the characteristics and formal practices of traditional S&T policy institutions, under the planned regime. The traditional practice may be illustrated by two features which were common to both of the initiatives: first, the efforts were focused on the supply side, and on the array of externalized R&D establishments, and second, the government had twin roles: it formulated development strategy, and took action to implement it. Thus in ‘conventional’ planned economies, science and technology policy amounted to no more than establishing priorities, and conducting administrative implementation. In the case of China at that time, the first initiative (i.e., the rehabilitation and improvement of R&D institutional establishments) was adopted as a remedy for the damage wrought in the Cultural Revolution. This was done in the belief that the earlier established institutional base would have performed well without the disruption of the Revolution, and that the re-building of the system was critical to the success of the then newly defined economic development plan. The second initiative (the elaboration of the planning procedures for S&T resources) followed the first, so as to bring the re-built R&D system into line with the revised economic objectives, which placed a higher priority on the efficiency of development and on the people’s standard of living. In particular, the second initiative was triggered by the difficulties being encountered in the fulfilment of the economic objectives around 1980. Rehabilitation and improvement of R&D institutions The 1978–1985 National S&T Programme Efforts to rehabilitate R&D institutions were led by a national S&T programme—the ‘1978–1985 National Science and Technology Programme’, which was announced in 1978 (Fang Yi 1978). Eight S&T areas were chosen as national priorities. They were 1) agriculture; 2) energy resources; 3) materials; 4) computing; 5) lasers; 6) space science and technology; 7) high-energy physics; and 8) genetic engineering.

10

REFORMATION OF PLANNING PRACTICE

It is obvious that this programme focused on attaining a leading status in science and technology (‘White Paper’ No. 1:262–266). It is remarkable that the justification for the S&T policy made in this period was surprisingly analogous to the justification for the initial creation of the S&T system twenty years earlier.2 It was perceived that the emerging revolution in science and technology would have a far-reaching impact, would lead to an enormous improvement in productivity, and that science and technology should therefore be seen as an important factor in determining economic production. In particular, the scientists and engineers, who had been popularly denounced and dismissed from R&D in the Cultural Revolution, were declared to be part of the working (and therefore, leading) classes (Deng Xiaoping 1978). In institutional terms, efforts focused on the rehabilitation of pre-Cultural Revolution patterns at the macro level, and on the improvement of R&D institutes’ management at the micro level. After the experiences of the Cultural Revolution, which tended to disadvantage the supply side of S&T because of dissatisfaction with its separation from economic production, China now turned to the rejuvenation of the original R&D system, until the mid-1980s, when the broader economic institutions had been significantly altered. Several measures which were taken to rehabilitate and improve the R&D institutions will be described below. Rehabilitation and expansion of independent R&D institutions The drive to establish and complete a ‘national scientific research system’, which was part of the National S&T Programme (Fang Yi 1978), led to the almost complete rehabilitation of those R&D institutions, and other institutional establishments such as design institutions, which had been established before 1966 but which were later closed or down-graded to affiliation either with lower levels of the administration or with enterprises. In addition, a large number of new independent institutes were created, especially in the fields ‘where there was weakness previously, where regional development could not be adequately supported, and where the scientific and engineering professions are experiencing rapid progress which is of great importance’ (Fang Yi 1978). No official data is available for the expansion of R&D institutions for that period, but one indicator, the ‘expenditure for scientific research from, the government budget’ is, indirectly, significant: this figure reached a peak between 1978 and 1980, and has been falling since. The expenditures were 1.5, 1.6 and 1.5 per cent of GNP for 1978, 1979 and 1980, respectively. Since then this indicator has steadily decreased, to about 1.0 per cent in 1988 and 0.71 per cent in 1992 (SSB 1990a:202–203; China Statistical Yearbook 1993:23). Improvement of R&D institute management At the micro level, policy efforts in this period focused on the improvement of the management of individual R&D institutes. Much was learned from the lessons of the past, when political criteria had been substituted for professional standards. This was seen as an important reason for the lower performance of the system. The following measures were set in motion (Fang Yi 1978; ‘White Paper’ No. 1:14): • substitute experts for political cadres as directors of institutes; • set up academic committees at individual R&D institutes, as the authoritative bodies in academicallyrelated appraisal and review; • re-establish the excellence principle, which had been suspended for twenty years, as the basis for the assignment and promotion of S&T professionals.

REFORM POLICY FOR THE S&T SYSTEM

11

Creation of in-house R&D and design departments in enterprises Enterprises were also encouraged to create their own R&D departments. Though no statistics are available, in-house R&D and design departments were certainly widely established within big industrial enterprises in these years, and this continued throughout the 1980s and up to the present. The usual process was the enlargement and re-assignment of units which had previously functioned as testing, measuring, designing, or maintenance offices in the host enterprise (interviews, particularly at the Ministry of the Machinery Industry). Elaboration of planning practice The second policy initiative in this period was the elaboration of planning practices. This began in the first half of the 1980s, with the aim of aligning industrial technology R&D with the Five Year Economic Plan. For the first time in the history of planning in China, R&D projects were closely combined with the economic programme. One important reason was the frustration of the ambitious industrialization plan which had been set out around 1978, which projected large-scale procurement of foreign technology but proved to be unattainable because of a shortage of hard currency. The resulting critical review of S&T policy turned to address the effectiveness of the allocation of domestic S&T resources (SSTC 1981).3 The redirection of S&T resources was carried out by means of a planning apparatus which will be discussed in the paragraphs below. This was a significant initial step towards setting more realistic targets for technological development, although the efforts in this period were not very successful. They were motivated by a great concern to improve economic performance, which was widely accepted from the end of the 1970s. Before then, the lower standard of industrialization required no more than the duplication and minor diversification of technologies imported in the 1950s. The duplication process, behind a ‘closed door’, placed little pressure on industrial R&D to meet the requirements of technological upgrading. This period of qualitative stagnation was ending. Planning for the re-allocation of R&D resources to industrial technology Planning for the re-allocation of R&D resources was based, as before, on direct priority-setting and investment by the state. The intention was to achieve improvements ‘in the development of product and process techniques, and in the assimilation and dissemination of S&T achievements’ (Zhao Ziyang 1982). Eight areas were chosen for the ‘Key S&T Projects of the Sixth Five Year Plan’ (‘White Paper’ No. 1:114– 130) based on economic and technological criteria, as follows: 1 agriculture; 2 the consumer goods industry; 3 energy resource development and energy conservation; 4 the raw material industries and geological exploration; 5 mechanical and electrical equipment; 6 transportation; 7 new technologies; and 8 social development.

12

REFORMATION OF PLANNING PRACTICE

It can be seen that, of these eight areas, at least five, i.e. 2, 3, 4, 5 and 6, relate to industrial product and process technologies. The re-allocation of R&D resources was carried out by mandate, relying on the power of the state. The management of planned projects The management of the planned S&T projects utilized the existing administrative framework. Some major characteristics are listed below: • officers from industrial ministries served as the coordinators of the projects, supported by experts from the R&D community; • the technological work of the projects had to cover a wide range from basic research to small batch production, because the project targets were for the development of production technology, rather than for ‘generic’ technology;4 • participants of a particular project had to be drawn from different units, given the high degree of institutional segmentation which had developed. The project coordination had to correlate the tasks of the various participants; • the principal participants were industrial technology R&D institutes, not the academic institutes of the Chinese Academy of Sciences, nor enterprises (interviews); • industrial firms played secondary and complementary roles (interviews).5 Limitations of the elaborated planning approach As was said above, the re-allocation of R&D resources was mandatory, relying on the planning institutions. This seems to have had serious drawbacks as regards the economic performance targets. The following limitations have been derived from perceptions expressed in articles or by ministry managers. • the capacity of the administrative authorities was limited: administrative power was unable to deal with the innumerable different needs of various industrial users (interviews, Ministry of the Machinery Industry, and Ministry of the Electronics Industry); • there was a loss of efficiency due to the arm’s length collaboration among participants. This was caused by the ‘outside’ coordination of a large number of participants (Ou Wen 1991); • deficiencies in the dissemination of the resulting technologies were significant: the planning approach was found useful in acquiring some important industrial technologies, and in putting them into first use, but it proved far from efficient in achieving the widespread adoption of new technologies (Interviews at the Ministry of the Machinery Industry).6

3 THE DECISION ON S&T MANAGEMENT SYSTEM REFORM (1985) AND THE CREATION OF A ‘TECHNOLOGICAL MARKET’

From 1985 on, China’s science and technology policy came to concentrate on the reform of the system itself. The turning-point was the promulgation of the ‘Decision on Reform of the Science and Technology Management System’ (‘The Decision’), which was put into action in the same year. Note that this followed reform in the agricultural sector, which began at the end of the 1970s, and the reform of industrial sectors, which had been decided on just one year previously (in 1984), as mentioned in Chapter 1. The central concern of the Decision was with the problem of the lack of ‘horizontal and regular connection between science and technology and production’. Science and technology had become, in fact, increasingly incompatible with the operations of the agricultural and industrial sectors. In his speech at the 1985 National Working Conference of Science and Technology, former Premier Mr Zhao Ziyang spoke on the advantages and disadvantages of the existing system. This analysis may be seen as a reflection of disappointment with the old methods, based on long experience, including the intensive experiments with the S&T system since the late 1970s. He said: The current science and technology institution in our country has evolved over the years under special historical situations. The advantages embodied in this system manifested themselves in concerted efforts to tackle major scientific and technological projects, with great success. However, there is growing evidence to show that the system can no longer accommodate the situation in the four modernizations programme, which depends heavily on scientific and technological progress. One of the glaring drawbacks of this system is the disconnection of science and technology from production, a problem which is a source of great concern for all of us. By their very nature, there is an organic linkage between scientific research and production. For this linkage a horizontal, regular, many-levelled and many-sided channel should be provided. The management system as practised until now has actually clogged this direct linkage, so that research institutes were only responsible to the leading departments above, in a vertical relationship, with no channel for interaction with the society as a whole or for providing consultancy services to production units. This is the root cause of the inability of our scientific research to meet our production needs over the years…. This state of affairs can hardly be altered if we confine ourselves to the beaten track. The way out lies in a reform. (Zhao Ziyang 1985) This speech expressed a very strong inclination to appeal to the power of the market to solve these problems. This obviously revealed the agreed consensus of those guiding the market reform which was then going on in most economic sectors. He went on to say:

14

REFORM POLICY FOR THE S&T SYSTEM

Experience in the past thirty years shows that, as long as commodity transactions exist in an economy, we cannot achieve the desired results in any economic-related undertakings if we ignore the commodity—currency relationship, ignore the law of value and the role of economic levers. Often our efforts have got nowhere, have gone contrary to our wishes, or at the most were partially rewarded. You cannot force farmers to provide urban residents with food, cotton, and meat by administrative means. You have to do business with them in accordance with the law of value. The farmers would not hear of it if you expected them to produce for the sake of equalitarian idealism…. The same is true with scientific research. If you want scientific research to serve production needs, you must acknowledge the value created by mental labour and allow most technology achievements to become tradable. If you want research institutes to serve the whole society, you must break down hurdles of all descriptions to open up a technological market. If you want the scientific personnel of the research institutions to voluntarily and regularly go to enterprises to identify research items, you must alter the funding system in which research institutions depend entirely on appropriations from the state. To bind research institutions and production units in a common cause, you must adopt a variety of economic means linking them with ties of interests. (Zhao Ziyang 1985) The first section below analyses the main measures of the Decision, which focus on a ‘technology market’ approach, while the following section addresses other policy measures of the Decision intended to support the central ‘technology market’ approach. This is followed by two sections providing a preliminary summary of some consequences of the technology market approach. Central policy measures—the ‘technology market’ approach The chief policy thrust under the Decision was to introduce the market mechanism into the operation of the R&D system. Two types of practical measures were put in place concurrently: • diminishing government grants to put pressure on R&D institutes so that they would have to turn to real demands; • creating a technology market as the intermediate institution for transactions between R&D institutes and their industrial users. Diminishing government grants (Decision: sections I and II) A general survey of R&D institutes was carried out in 1986.1 On this basis, a categorization was made. The industrial technology R&D institutes were identified and their grants began to be progressively reduced.2 The process was to be completed within five years (from 1986 to 1990). The policy also offered incentives: both institutes and individuals were to be permitted to retain a proportion of their earnings. The incentives were open to all types of institute, not just the industrial technology institutes (‘White Paper’ No. 1:314– 315). As a result, by 1991 the 2,000 plus R&D institutes engaged in industrial technology had had their ‘operation fee’ entirely or partly cut, a total reduction of slightly less than one billion Chinese yuan per year, or about one tenth of the overall government S&T budget in 1985.3

CREATION OF A ‘TECHNOLOGICAL MARKET’

15

Creating a ‘technology market’ (Decision: section III) The term ‘technology market’ implies several things. First, it represents a new concept which legitimizes paid transactions for technology. This new concept was critical to the reform since the ideological tradition provided no ground for market transactions in technology. The ‘public good’ nature of scientific and technological knowledge had long been the basis on which the whole of the old system was constructed and run. Second, it represents a set of regulations and supporting agencies. The Law of Technological Contracts was promulgated in 1987 to govern this special type of economic contract (SSTC 1991a:146–152).4 The agencies to support technology transactions were established, derived mainly from administrative bodies related to science and technology. After a few years’ confusion, they were defined as ‘non-profit regulatory agencies, in charge of the registration of technological contracts’ (‘White Paper’ No. 4:45–47). Now a broad network of these agencies has been formed. Third, the term ‘technology market’ refers to a range of technology-related transactions. These are categorized in the Law of Technology Contracts (SSTC 1991a:372–378) as: • • • •

contractual development of technology; technology transfer; technological consultancy; and technological services (such as designing, engineering, testing and services for computer applications). Other policy measures to support the technology market approach

Other policy measures of the Decision to support the market approach provided for: 1) government and public financing of R&D, 2) the autonomy of R&D institutes and 3) the mobility of S&T personnel. Policy for government and public financing of R&D (Decision: section II) The distribution of the remaining government and public S&T funds was placed on a competitive basis. First, the National Natural Sciences Foundation was established in 1986, in charge of distributing central government funds for basic research and ‘fundamental’ applied research through peer review, based on the excellence of the applicants.5 Second, a competitive bidding procedure was adopted in 1986 for government investments in the ‘Key Science and Technology Projects’ of the five-year plan. Thus a quasi-market for government funds was created. The introduction of competition for funding for basic research and state S&T projects helped to improve the efficiency of the remaining public investment. This might be seen as an expansion of the market approach. The same procedure was later adopted for newly created governmental funds, such as the ‘leading funds’ of the Torch Programme, which aims to accelerate the commercialization of technology (see Chapter 4). The excellence-based competition also seems to have protected some more able institutes and individuals from the disturbance which was inevitable in the transitional period. Policy for the autonomy of R&D institutes (Decision: section VII) The policy for the autonomy of R&D institutes released R&D institutes from vertical controls, so that they could interact with the technology market which had been created. The directors of R&D institutes were given delegated authority in the following areas (‘White Paper’ No. 1:318–319):

16

• • • • •

REFORM POLICY FOR THE S&T SYSTEM

to decide on contractual R&D and contractual services with their users; to register various joint ventures with enterprises, design units and institutes of higher education; to decide on matters of institute personnel and internal organization; to dispose of their income from contracts; and to enter into international cooperation, and to retain foreign currency obtained, in accordance with state regulations. Policy for the mobility of S&T personnel (Decision: section IX) and the emergence of a quasi-market for S&T personnel

The policy to encourage the mobility of S&T personnel contained two principle elements: • the replacement of life-long recruitment with term appointments to defined posts; • permission for scientific and technological personnel to take second jobs, so long as they perform well in their primary assignments. The real mobility of S&T personnel was still very limited, at about 2 per cent per year in the second half of the 1980s (SSTC and NCSTD 1989:65). However a ‘quasi-market’ for well-educated workers has emerged as an alternative, with several opportunities for permanent or temporary mobility: • taking second jobs;6 • applying for temporary or permanent leave under terms such as ‘leave without pay, but with staff status reserved’ (tingxin liuzhi), ‘resignation’ and ‘early retirement’; • the re-entry of retired S&T personnel. An institutional framework to support the quasi-market is developing, consisting of a significant number of ‘personnel exchange and recruitment centres’. Most of the centres have, once again, emerged out of administrative agencies related to science and technology. The mobility of the talent needed for various transformations of the R&D institutions, including the development of New Technology Enterprises (see Chapter 5), is supported by these centres (SSTC and NCSTD 1989:62–68). This ensures the required mobility of S&T personnel, although a fully-developed labour market does not yet exist, by offering the following services: • protecting the professional title and welfare insurance which S&T persons enjoyed as public employees; and • providing S&T personnel recruitment services outside the planned personnel distribution system.7

Responses to the technology market solution Inefficiency of the technology market It is useful to review the responses of R&D institutes to the technology market, bearing in mind that it was the intention of the Decision that the technology market would intensify the links between scientific

CREATION OF A ‘TECHNOLOGICAL MARKET’

17

research and its productive uses, as a remedy for the chronic horizontal disconnection of the system. Having been granted nearly full autonomy, and being under pressure to survive financially, R&D institutes hastened to play the market. There is considerable evidence that, in the first years after the Decision, approximately from 1985 to 1987–1988, technology transactions took the form of either 1) once-and-for-all exchange, or 2) the formation of long-term contractual alliances, which were primarily intended to secure an economic return.8 Most industrial R&D institutes did not find that the technology market, as they had experienced it in these transactions, met their expectations. A few examples, presented in Case Texts 3.1, 3.2 and 3.3, may illustrate this. Some efforts toward structural transformations were in fact driven by these early experiences, with the outputs in the transactions shifting from ‘software’ know-how to ‘hardware’ products or integrated engineering services. These structural transformations will be discussed in Chapters 4, 5 and 6.

CASE TEXT 3.1 THE GENERAL INSTITUTE OF COAL MINES SCIENCE AND TECHNOLOGY This Institute belongs to the State General Corporation of Coal Mines, formerly the Ministry of the Coal Industry, with 17 branch institutes and centres scattered around the country, in charge of production technologies for coal mining. They detailed three phases of response to the reform. The first phase was from 1985 to 1986, when ‘technological consultancy’, ‘technology transfer’, ‘technological services’, and ‘technological training’ dominated in their efforts. They were basically ‘selling technologies’ (form 1 transactions, as above). At that time the payment from each contract was very low, and the yearly income was uncertain. As government funding diminished, they felt that it would not be possible to rely only on providing ‘software’; therefore they must also produce ‘hardware’. The second phase was from 1987 to 1988, when partnerships with firms were established based on technology transfer (presumably form 2 transactions, as above) or monetary investment. Some investments were made blindly in businesses such as beverages, food processing and kinds of manufacturing which were irrelevant to the institute’s strengths. Most of these failed either because they were acting outside their field of competency, or because the institute’s contractual rights could not be protected effectively. The Institute gained very little even when the enterprise side benefited considerably. The third phase began in about 1988. Since then various businesses have been established, with the main purpose of preserving the institute’s core technological abilities and physical installations developed and accumulated in the past. Source: SSTC 1991c:55–58.

CASE TEXT 3.2 THE AUTOMATION RESEARCH INSTITUTE METALLURGICAL INDUSTRY (ARTMI)

OF

THE

MINISTRY

OF

THE

This is a leading Institute belonging to the Ministry of the Metallurgical Industry. It is in charge of the development of metallurgical automation technologies, with more than 800 engineers and technicians. Beginning in 1985, the Institute attempted to respond to the ‘technological market’ by delegating responsibility and profitability to a number of smaller teams, so as to intensify the incentive for researchers to contract their services to outside users. This would be reasonable if the stronger incentives led these teams to produce technologies, which could be contracted more actively and carefully. However a strategic shift had to be made in 1987 to re-build the hierarchy of the Institute as a whole because:

18

REFORM POLICY FOR THE S&T SYSTEM

• these teams’ technologies were usually ‘un-packaged’, ‘not in the form of a complete set’, and ‘not good in reliability’; • these technologies were usually under-valued by buyers; and • there was fierce competition from foreign suppliers, who provided their technologies in well packaged sets, of high quality and accompanied by satisfactory engineering services. This further reduced the competitiveness of the Institute in the marketplace. Sources: Interview notes 6; SSTC 1991c:62–65; Information on Science and Technology of China No. 6, 1992:5–11.

CASE TEXT 3.3 THE RESEARCH AND DESIGN INSTITUTE FOR CHEMICAL ENGINEERING OF FUSHUN CITY This is an Institute at the municipal level. The fine chemicals they developed were mostly substitutes for imported ones, used as additives or intermediate inputs in various processes. Their scientific experiments used to be carried out at the scale of laboratory research. This led to some problems in technology transactions, which pointed to the necessity of expanding their activities to pilot-plant and trial production. The first problem with laboratory-scale research was that some technological problems such as the recovery of catalysts do not show up or cannot be solved at the laboratory stage. The technological uncertainties in largescale production use are liable to be higher, which damaged the Institute’s credibility with buyers. Second, market tastes cannot be tested, in even a preliminary way, without a pilot plant because the volume of laboratory samples is too small to do trial marketing. Market risks are therefore likely to be higher, and this is even more serious where competing imported chemicals are already available. There is no niche for a substitute without trial marketing to win the first buyers. Third, estimates of techno-economic terms cannot be based only on laboratory work. The lack of information about the commercial potential sometimes halted promising transactions. And finally, the limited capability of the user firms created difficulties. Many users, mostly smaller firms, ‘did not have even a limited understanding of technological problems. They do not understand that further testing and modification are indispensable for the establishment of a chemical process. If something could not be settled in one or two trials, they became disappointed.’ Source: Science and Research Management No. 1, 1992:45–47.

Factors influencing the failings of the technology market Three causative factors can be discerned from these cases, all of which are attributable to the difficulties encountered in selling laboratory software or segmented pieces of system technology. Note that while some of the factors below have been widely discussed by researchers both in China and abroad, some are still seriously neglected, particularly the weakness or deficiency of the market in dealing with uncertainties. Indeed, this is an issue deserving more study. Uncertainties of technological innovation Any new elements of technology, such as the new chemicals, new machine designs, and new automation controller units referred to in the cases above, entail a degree of uncertainty as regards their technical feasibility and market potential. The market mechanism itself is in essence not well adapted to dealing with

CREATION OF A ‘TECHNOLOGICAL MARKET’

19

the uncertainties which are intrinsic to technological innovation, because an accurate calculation by buyer and seller of their gain or loss is essential to efficient market transactions. The cases also show that the more complex the user’s system, the more difficult it will be to demonstrate the commercial value of some novel components of the automation systems (Case Text 3.2). Besides, the more robust the existing technology, the more demanding buyers will be in considering substitutes (Case Text 3.3). The existence of capable foreign suppliers reinforces this: it may have undercut the market potential of domestically developed technology. Inexperience of users Having long been passive recipients of new equipment, the industrial buyers of technology lacked experience in evaluating technological know-how. They also had limited abilities to assimilate technological knowledge or unpackaged technologies, as all the cases show. This is because the transfer of technology itself is a technology-intensive process (Mowery 1983), and the firms had had no chance to learn how to acquire technology at a time when they had just been thrust into the technology market. Underdevelopment of market institutions Breaches of contractual obligations by the parties, and the inability of the market mechanism to rectify the breaches, as Case Text 3.1 shows, indicate the underdevelopment of the market institutions. Weak market institutions are a common problem in transitional economies, and they have already been perceived and discussed by other authors, as was pointed out in Chapter 1. The growth of the technology market Despite these failings, the technology market has been growing since its establishment. Some characteristics of the market are analysed in this section, drawing on data which is obviously too incomplete to justify a more precise discussion. The rapid growth of the technological market is illustrated by its turnover. It is reported that the overall value of signed contracts increased from 2.3 billion yuan in 1985 to 8.1 billion by 1989 (‘White Paper’ No. 4:42). These figures should be treated with caution because analysis shows that 1) some transactions were not for technology but for non-technological commodities (SSTC 1989:86), and 2) that some of the transactions were for R&D projects proposed and funded under state and local plans (‘White Paper’ No. 3: 39). The rapid growth of the technological market is also illustrated by the development of market institutions. It is reported that the number of agencies in charge of the management of the technology market had increased to 21,132 units by 1991, up from 9,649 in 1987 (China Statistical Yearbook on Science and Technology 1992:342). The number of scientists and engineers formally employed by these agencies had risen to 124,000 in 1991 (ibid. 342). This means the market network has expanded to cover many mediumsized cities, which the Chinese call ‘counties’. As was mentioned above, these agencies have emerged mainly from administrative bodies related to science and technology (‘White Paper’ No. 2:43; SSTC and NCSTD 1990:71; SSTC and NCSTD 1989:83). The limited official data which is available on the contract structure in the early years, from 1987 to 1989, is shown in Table 3.1. The data indicates that technological consultancy and technological services, combined, were the most traded item in these years. They accounted for three quarters of the contracts, and

20

REFORM POLICY FOR THE S&T SYSTEM

more than half of the value of all contracts. This picture remained fairly constant over the three year period. Contracts for the ‘development of technology’ accounted for about 12 per cent of all contracts in these years, and in value terms for 27 per cent in 1987, increasing to 37 per cent in 1989. It is reported that some state-financed S&T projects relating to the development of technology entered the technology market, as has been mentioned (‘White Paper’ No. 3:39). Contracts for ‘technology transfer’ accounted for about 11 per cent of contracts signed, and 13 per cent in terms of the value of the contracts, in 1987, and this proportion is reported to have been quite stable between 1987 and 1989 (‘White Paper’ No. 4:43). Some trends during the recent several years are illustrated in Table 3.2, which is for 1993. The ‘contractual development of technology’ increased Table 3.1 Transactions on the technology market (1987–1989)

Contract quantity Contract value

Technological consultancy & technological services (approx %)

Contractual development of Technology transfer technology (approx %) (approx %)

75 50

12 30

11 13

Sources: ‘White Paper’ No. 3:39; No. 4:43. Note: Data on transactions in the technology market is incomplete for the early years. For prudence, the table presents aggregate figures, which have been summarized from, and checked with, those scattered in the sources below. Table 3.2 Transactions on the technology market (1993)

Contract quantity Contract value

Technological consultancy & technological services (approx %)

Contractual development of technology (approx %)

Technology transfer (approx %)

63 41

28 48

9 11

Source: Databank of Statistics on Science and Technology 1993:160. Note: To enable comparison, the data in Table 3.2 has been presented in the same format as Table 3.1.

significantly, accounting for 28 per cent of contracts, and 48 per cent of the value of contracts. The increase can hardly be explained merely by the partial inclusion of state S&T projects, given that overall expenditure under the state plan in the period was not growing. Technological consultancy and technological services, combined, still dominated in terms of numbers of contracts, but less than in the earlier period. Technology transfer also declined, but less markedly. These trends may be considered to be rather solid, since other scattered data for various years points toward a similar shift. Many features deserve further exploration. One concerns the quality of the technology market. Is it not homogeneous? Can it be said to be more ‘friendly’ to some kinds of transaction? The answer to the latter question seems to be affirmative, as indicated by the aggregate of ‘technological consultancy’ and ‘technological services’. This corresponds to observations with respect to contractual R&D undertaken by research enterprises in the US in the first half of the century (Mowery 1983; Mowery and Rosenberg 1989). The reason for the popularity of transactions in ‘technological consultancy’ and ‘technological services’ might be that they are more easily defined for paid services, and less involved in the core part of the inhouse activities of the buyers. As regards the quality of the technology market, the significant increase in the ‘contractual development of technology’, in particular, requires some interpretation. Such an increase seemingly contradicts the

CREATION OF A ‘TECHNOLOGICAL MARKET’

21

assumption in the paragraph above, supposing that such contracts involve more ‘core’ activities of the buyers. A subsequent study, which concentrated on the machine industry, has found that for that industry at least, the ‘contractual development of technology’ covered mainly complex manufacturing systems developed for a particular buyer’s purpose. The complex manufacturing systems usually incorporated testing or controlling units and user application software. This was, roughly speaking, because the existing machinery enterprises were not able to meet demands of this kind, while foreign suppliers were not able to fulfil the user’s specific requirements. This situation may also be illustrated a little in Case Texts 4.1 and 6.1. In these cases, the ‘contractual development of technology’ is not real R&D: it is more like the ‘contractual provision of custom-made products’ (or custom-made machinery, in the case of the machine industry). A study of the machine industry constitutes Part 3 of this book. The dynamics of the market also deserve further exploration. The technology market has been expanding rapidly so that all kinds of contracts were in fact increasing in absolute terms. There seems to have been some causal relation between the greater effectiveness of technology transactions and the widespread efforts of R&D institutes to integrate R&D with commercially profitable activities, discussed in both the following chapters of this Part and in Part 3.

4 MERGING R&D INSTITUTES INTO EXISTING ENTERPRISES (1987)

The justification for merging—a policy response to the inefficiency of the technology market Early in 1987, one and half years after the implementation of the ‘Decision’, reform policy took a bold step by urging industrial R&D institutes to enter into enterprises. Signed by the State Council, the document, ‘Stipulations of the State Council for Furthering the Reform of the S&T Management System’, presented the rationale for this move: [though the reform has achieved preliminary success over the past year and more…] one should have been conscious that the disconnection between S&T and production has not yet been fundamentally improved. The pattern of the organizational structure of the S&T system is basically untouched, the system remains closed (to the outside); the important R&D institutes are still affiliated to administrative organs rather than being bound up with the national economy; there are more qualified scientists and technicians than required in big research institutes belonging to central ministries and institutes of higher education, while there is a serious lack of S&T manpower in light industry, commercial enterprises and rural areas; the policy measures intended to intensify the links between research institutes and enterprises have been inefficient, so that a considerable number of research institutes are undertaking a kind of ‘self-accomplishment’ [of the commercialization of their technological strengths] without devoting much effort to making outside connections. (State Council 1987a) Here the disappointment with the responses of R&D institutes to the technology market solution is explicit. R&D institutes were mainly moving to capitalize on their technological know-how by themselves, and inside their own institutes, rather than transmitting it to productive units. The independent R&D institutes had in fact become more closed rather than more open to the outside. They were reserving their technological strengths for commercial exploitation. Reform policy thus turned to focus on organizational mergers, by breaking the organizational borders and combining the two types of organizations in a single unit, with enterprises as the basis of the combination. As the document stressed, ‘the majority of research institutes engaged in technology development, especially in the development of product technology, should enter into enterprises, or into groups of enterprises, or should closely cooperate with enterprises’ (State Council 1987a).

MERGING R&D INSTITUTES

23

Policy measures for merging Policy measures encouraging R&D institutes to enter into existing enterprises fell into two categories. The first category encouraged them by means of protecting the preferential position they enjoyed as independent institutes. Under these measures (State Council 1987b): • R&D institutes, after being combined into enterprises, would still retain relative independence from the host enterprise in financial matters and professional activities, provided that they accomplished the tasks required by the host enterprise; • R&D institutes, after entering into enterprises, would continue to enjoy tax exemption for income from sales of both technology and pilot plant products; • R&D institutes, once absorbed into enterprises, would continue to get ‘operational funds’ and ‘capital construction investments’ from the Government, based on their appropriations in the last year before they entered the enterprise. In addition, policy guidelines were set for host enterprises to promote merging (State Council 1987b): • the wages and bonus standards of the host enterprise could be adopted, if these were higher than those of the institute; • host enterprises were requested to increase investments in the R&D units they obtained; • host enterprises were recommended to rely mainly on the R&D unit they had acquired to deal with affairs relating to technology imports etc. But the results were not up to expectations. Few industrial R&D institutes, out of a total of 2,000, responded. In the Ministry of the Machinery Industry, for instance, only two of the sixty-four ministry-level R&D institutes were merged with enterprises during the whole of the 1980s. Frustration with the lack of success in encouraging mergers was admitted officially. The central focus of the S&T reform policy was further modified one and a half years later, in mid-1988, with a shift to facilitating the establishment of ‘new technology enterprises’, which were largely spin-offs from R&D institutions. This will be the subject of Chapter 5. Explanation of the lack of success, for two industrial sectors Three factors which were thought to be attributed for the lack of success in mergers became evident from the explanations made by some responsible managers from the Ministries of the Machinery and Electronics Industry. The inability of enterprises All the legacies inherited from the past, as summarized in Chapter 1, particularly the enterprises’ lack of financial reserves, contributed to the frustration of enterprise-based mergers. A responsible officer from the Ministry of the Machinery Industry (Interview note 7:6) estimated that sustaining a centrally affiliated R&D institute would entail an annual expenditure of several million yuan. To shoulder such a financial burden, the host enterprise would need an annual turnover to the order of a billion yuan. For instance, in 1987 not a single enterprise in the instrumentation sector of the machinery industry had reached the point at which they could afford to accept one of the thirteen ministry-level R&D institutes engaged in instrumentation

24

REFORM POLICY FOR THE S&T SYSTEM

technology. This resembles problems encountered in other attempts to re-structure enterprises in China, where the average size of enterprises is relatively small.1 Compared with the machinery industry, enterprises in the electronics industry were even weaker. They had to renew their key installations, relying heavily on large-scale technology imports throughout the 1980s, because of the rapidity of technological change in this area.2 As a result, the production technology of the industry became more heavily dependent on foreign suppliers, leaving domestic R&D no place as a supplier to the sector. Managers in the electronics industry seemed not to see merging as having any significant potential in the near future, while the Ministry of the Machinery Industry is now, in the 1990s, actively promoting mergers (Interview notes 7, 8). Lack of congruity of R&D institutes with enterprises The knowledge and physical equipment of R&D institutes were seriously outdated. They lacked experience in working with rapid change, and organizational rigidity impeded steps to keep up with change. During the 1980s, the domestic R&D institutes were basically barred from importing technology, and their value as suppliers of industrial technology was considerably reduced. In fact, not only the electronics industry, but also, to a less extent, the machinery industry saw the physical installations of R&D institutions lagging behind those of leading enterprises during the 1980s. In a sense, the massive importation of industrial technology has undone the compatibility which had developed prior to the reform (Interview notes 7:3, 4). Enterprises in general did not welcome merging. In addition, R&D institutes have an institute culture which is different from production enterprises. They enjoyed more independence in searching knowledge, and received more prestige from the society than production units. Painful failures experienced when trying ‘to combine with production’ in the past have in fact inhibited their response (interviews in a number of R&D institutes, 1994). Besides, the work programmes of many R&D institutes were much more extensive than could be maintained if they merged with any single enterprise. All these were institutional barriers arising from their previous institutional separation from enterprises. The attitude of industrial ministries The industrial administration has ample power to accelerate or delay the merging process. The market seemed not to be strong enough to guide organizational merging. In 1987, the Ministry of the Machinery Industry discreetly set up five criteria to guide possible mergers.3 This resulted immediately in one case of merging. During the 1980s, two institutes, of the total of 64 (at the ministry level), entered into two big enterprises. One of these mergers had been achieved prior to the 1987 national policy push, on the initiative of the Ministry: the Automobile Technology Research Institute of the Ministry merged into the No. 1 Automobile Factory in Shenyang city. Some apparently feasible mergers were suspended at that time, to be completed a few years later.4 Efforts to promote merging revived in the early 1990s. Another six or so mergers have been realized recently, and it was estimated that a dozen of the 64 institutes would be transformed in this way (Interview notes 7:5). It was explained that some of the reasons for the lack of success in merging during the mid-1980s are now changing. There was no indication of a possible revival of merging in the electronics industry, however.

MERGING R&D INSTITUTES

25

Transformation after merging into an enterprise—the case of the Automobile Technology Research Institute The first case which was completed, when the Automobile Technology Research Institute of the Ministry of the Machine Industry became part of the No. 1 Automobile Factory in 1980, illustrates some profound transformations which resulted from such a merger. Being an integral part of an enterprise ensured that the previously independent R&D institute would work to support the core technological efforts of the host enterprise. The embodiment of this focus in the institute structure was accompanied by significant structural remodelling.

CASE TEXT 4.1 THE RESTRUCTURING OF A RESEARCH INSTITUTE AFTER MERGING—THE CASE OF THE AUTOMOBILE TECHNOLOGY RESEARCH INSTITUTE The Automobile Technology Research Institute was combined into the No. 1 Automobile Factory in 1980. This was decided and effected under the direct supervision of the Ministry of the Machine Industry. Ten years later, the Institute has been thoroughly transformed as an integral part of the host enterprise. Change in staff occupation structure In 1980, the Institute staff totalled 1,280, with 476 engineers and technicians, 220 administrators, and 584 workers. In 1991, there were 2,245 staff, of whom 1,318 were engineers and technicians, 42 administrators and 885 workers. Thus the staff almost doubled, largely due to the huge increase in technological professionals, while the administrative staff was cut drastically. Change in financial structure The main source of finance after the merger came from the factory. In 1991, the factory contributed about six sevenths of the total Institute income, and the Institute earnings from outside contracts accounted for less than one tenth. In the period 1980–1991, the annual expenditure of the Institute multiplied ten times, with the increased inflow almost entirely coming from the factory, while funds from the state were reduced to a few per cent. The change in the financing structure indicates that the Institute has shifted to serving mainly the needs of the host factory. Change in the priority of investment and technological activity The investment priority was explicitly shifted to testing. More than 40 per cent of the increased income was used for constructing and equipping testing facilities such as a road test site. This is a reflection of the fact that there had been virtually no such installations and activities within the factory. In addition, product development and designing has become more important. Eight new models of automobile, a large number of modifications based on these models, and a few new series of automobile engines were developed, some of which have been put into production. Before 1980, only very marginal modifications to imported designs had been made since the factory was erected in the 1950s. Becoming part of the factory has ensured that the previously independent R&D Institute works to support the core technological efforts of the host factory. Indeed testing, product development and designing are really at the heart for an automobile manufacturing factory. Sources: SSTC 1991c:323–327; SSTC 1989:59–67.

5 SPIN-OFF ENTERPRISES AND THE TORCH PROGRAMME (1988)

In mid-1988, a new initiative under the reform policy, known as the ‘Torch Programme’, was launched. This was one and half years after the 1987 national policy to promote merging. The launch of the Touch Programme was partly a policy response to frustration with the merging initiative. Instead of institutes being incorporated into existing industrial enterprises, the Torch Programme supported the integration of R&D institute assets, including experts, technological know-how, and some physical equity, with commercial production and service activities within newly-created business organizations. These enterprises, which were spun-off from the R&D institutes, were known in the Programme as New Technology Enterprises (NTEs). Part 2 of this book provides a systematic survey of NTEs, describing the development of NTEs themselves and supporting institutions to the NTEs, and the resulting impact on the widespread dissemination of computer and information technology. This chapter concentrates on the policy formulated for the spin-off approach to institutional restructuring during the market reform. The rationale for reform policy to support spin-off enterprises In fact, spin-off enterprises had emerged long before the Programme began. The first was reported to have been established in 1980. More appeared in several large cities from around 1984–1985. By 1985, there were about 100 such enterprises clustered in north-western Beijing, where many of the best R&D institutes and universities in the country were located. Most of the new businesses were engaged in computer technology, starting with sales and user services of imported personal computers (PCs). The first Science and Industry Park was co-sponsored by the Chinese Academy of Sciences and the Shenzhen Municipal Government in 1985. Thus the archetypes of NTEs and the Development Zones for New Technology Industries, both of which were endorsed by the Programme as institutions to promote restructuring, had already emerged. It is safe to say that the emergence of spin-off enterprises was authorized by the 1985 Decision, although it had not exactly envisioned the advent of NTEs. The 1985 Decision had largely released R&D institutes and S&T personnel from rigid control, thus empowering them to begin ventures in response to various opportunities. This was happening at a time when the ‘computer revolution’ had reached the point of having a huge potential for incorporating computers in various operations, while the rapid expansion of the economy produced a high demand for computer applications. Selling computers and developing user-specific applications of computer and information technology required an intellectual service sector. The old planning framework had not been designed to provide services of this kind, which need on-going institutional innovation.

SPIN-OFF AND THE TORCH PROGRAMME

27

The Torch Programme was founded to foster spinning-off as a new reform thrust for the better use of technological strengths in economic development, as Mr Song Jian, Chairman of the State Science and Technology Commission, stressed in his speech at the 1988 National Working Meeting of the Torch Programme (‘White Paper’ No. 3:415): We acknowledge that we can expect to find better solutions to the problems which have been faced by R&D institutes and universities, i.e. the limited ability of large and medium-sized enterprises to absorb [external] technologies and the difficulties arising from excessively small [technology] markets. The business entities which have been initiated by scientific and technical experts, based on their scientific and technological strengths and on the integration of [technological] development, production and marketing, are engaged in transforming accumulated S&T achievements into productive power and commodities. Policy measures for spinning-off Two strands of reform policy We have examined reform policies for R&D institutions in a chronological sequence, with the 1985 Decision representing a decisive turn towards market solutions. It has been shown that market reform entails a whole set of policies which differ from those of the past, particularly in using indirect rather than direct interventions. It will be useful, before proceeding with the analysis of the Torch Programme, to introduce a rudimentary conceptualization, based on observation, which divides the policies put in place to foster the transition of the R&D system into two strands: • the establishment of regulations and incentives; and • the creation of a regulatory agency framework. Various initiatives under the market reform required regulations and incentives. We have seen that regulations and incentives were set up by the Decision to promote the ‘marketization’ of industrial R&D. These included the delegation of autonomy to R&D institutes, laws for technological contracts, and stipulations for rewarding institutes and individuals with earnings from contractual technological activities, and so forth. We have also seen the second strand of policy that the Decision announced, the creation of ‘technology market agencies’, which amounted to a regulatory infrastructure for marketization. Moreover, the Decision provided an impulse for the emergence of a quasi-market for S&T personnel, in the context of the lack of a normally developed labour market. The creation of a regulatory agency framework is the most interesting of these two strands of policy. The organizational foundations which are required if regulations are to be put into practice had to be established. Regulations can be drafted in months, but the regulatory framework took years to develop (naturally the development of new economic organizations such as New Technology Enterprises was also needed, and was also a long-term process). Organizational evolution is a very complex process, entailing various modifications of every component in a society. We have seen that the technology market agencies and the personnel exchange and recruitment centres, which supported the quasi-market that encouraged the mobility of S&T talent, were derived from existing administrative agencies related to science and technology. We

28

REFORM POLICY FOR THE S&T SYSTEM

will discuss the policy measures devised under the Torch Programme, distinguishing between these two strands. Policies for the establishment of regulations and incentives1 First, the Commission named several criteria which were to guide the licensing of NTEs (SSTC 1991a:563– 566): • the technology underpinning the activities of the enterprise should be in specified areas of ‘new and high’ technology defined by the State Science and Technology Commission;2 • the enterprises should have appropriate capital and physical resources, market potential, and acceptable organizational and managerial abilities; • the chief manager should be a scientific or technical professional. Second, incentives were established for licensed NTEs, mainly in the form of preferential taxation (SSTC 1991b:233–238). Third, preferential stipulations were articulated for NTEs in export and import licensing, finance and investment, pricing, and employment (ibid.). The taxation incentives and other preferential stipulations resembled those for foreign investments, since NTEs are technology-intensive and there was no regulatory practice dealing with domestic non-state initiatives.3 In addition, intellectual property began to be included in incentives (‘White Paper’ No. 3:247). Not only patented and other proprietary technology, but also the special technological skills of individual initiators could in some cases be counted as equity (interviews in 1988–1989 and in 1993). However this policy has not yet been systematically realized, because of the lack of the institutional framework for property transactions. Policies for the regulatory and investment institutions First, Development Zones for New Technology Industries (Zones) were encouraged (‘White Paper’ No. 3: 249), in order to create a favourable regulatory environment for NTEs. The Zones regulate NTEs in matters such as licensing, taxation, international trading, financing and investment, employment, intellectual property, etc. Some Zones have invested in physical infrastructure such as roads, buildings, communications infrastructure, etc. to attract more domestic and foreign investors, as is the case in many ‘science parks’ and ‘technopolises’ around the world (Castells and Hall 1994). Second, Service Centres for Scientific and Technical Entrepreneurs (Centres) were established as ‘incubators’ for spin-offs, especially those initiated by individual S&T persons (‘White Paper’ No. 3:248). In practice, both the Centres and Zones were established at the initiative of city governments. They are also administered, from their inception, by local authorities. This had become possible in China because of the decentralized delegation of economic power under the market-oriented economic reform programme. The policy of the central government functioned 1) to put the government’s authority behind the campaign for the commercialization of technology; and 2) to guide the experimentation and disseminate the lessons on how to achieve better performance which were learned during the campaign. Third, a variety of sources of finance were opened up or encouraged, such as governmental ‘leading funds’, bank loans, foreign capital, etc. (‘White Paper’ No. 3:247), in order to broaden the resources invested in the commercialization of technology.

SPIN-OFF AND THE TORCH PROGRAMME

29

A new kind of bank loan, known as ‘loans for the development of science and technology’, intended for the commercialization of S&T achievements, was announced. The National Construction Bank and its city and provincial branches, for instance, decided to issue loans of this kind. The Bank and the State Science and Technology Commission (SSTC 1991b:458–459) stipulated that: • the loans are available to R&D institutes, collectively-owned enterprises, and state-owned enterprises. This implies that commercial loans were officially available to R&D institutes and most NTEs; • the loan could been used for the development of new product and process technologies; for pilot plants and the trial production of new technology; for the assimilation of imported technology; and for smallscale capital investment for commercial production based upon new technology; • the State Science and Technology Commission and its local branches were in charge of loan appraisal, although the bank retained the right to make the final decision. Thus there was a system of ‘outside’ professional appraisal to support the new banking service. This was necessary, given that banks had little experience in appraising loans for technology because they had been used to doing no more than accounting, under the planned regime.4 A network for financing NTEs developed under the umbrella of decentralization, with three investors involved: R&D institutions, banks, and Zones. A division of labour was forged among these investors: the R&D institutions provided mainly ‘venture’ capital for the initiation of NTEs, the banks provided funds for the expansion of NTEs when they had passed their first stage of development, and the Zones provided mainly investment in infrastructure (Zhao Wenyan et al. 1989).5 To a lesser extent, the Centres also worked on providing venture capital to NTEs initiated by individual scientific and technical experts (Interview notes 9:2, 3). The differentiated functions of these investors were based on their special positions. The R&D institutes, for instance, depending on their immediate knowledge of the underlying technology, and close involvement in the initiation of the NTEs, were better suited to manage the risks and rewards of their investments. They acted as the main risk-capital investors in the 1980s, although there was also a government financing agency known as the Venture Investment Corporation. While national ‘leading funds’ for the commercialization of technology were established by the Torch Programme, the projects set up under the Programme were predominantly funded by the local branches of banks.6 Bank operations had also been altered by the decentralization of the banking system, under which ‘each regional branch of the specialized banks was required to link their credit to deposits collected within the region’, (Qian and Xu 1993: section 4.3). The involvement of the banks was strongly influenced by the policies of local governments. In fact, each of the booming Zones had very active branch banks which sustained its prosperity.7 As a result of these policies, spin-off enterprises expanded rapidly and continuously. In 1992, there were fifty-two Zones scattered throughout the country and approved as national level Zones. In that year, 5,569 NTEs were registered in these Zones, producing products and services worth 231 billion yuan, and spending 15.2 billion yuan on their company R&D (China Statistical Yearbook on Science and Technology 1993: 307). In comparison, the S&T funds from the state budget in that year were 17.6 billion yuan (ibid.: 24). In the same year, the total expenditure for ‘research and development’ of the country amounted to 16.9 billion (ibid.: 23).

30

REFORM POLICY FOR THE S&T SYSTEM

The role of New Technology Enterprises According to our survey (see Part 2), the great majority of the NTEs were engaged in computer and information-related technology. A small proportion were engaged in other fields such as new materials, fine chemicals, medicines and biotechnology products. The technological activities of the NTEs in computer and information-related technology concentrated on what we have called user capability building for applications of these technologies. Three types of capability building emerged from the empirical evidence: adaptation of computer and information technologies based on the English language to the Chinese environment; development of user-specific automatic operation systems; and design and assembly of special purpose single devices and machines. The latter two usually embraced the incorporation of computer or information technologies in various devices or user systems. All three demanded a moderate degree of complexity in the systems, and in a few cases the NTEs have a quite competitive mastery of sophisticated technology. The capabilities which NTEs have acquired are impressive, given their short history. The success of NTEs has been widely accepted in China. The fact that most NTEs are engaged in computer and information technology strongly suggests that a surge of computer applications may be expected in China and in some other developing and formerly centrally planned economies as well if their accumulated technological capabilities can be released to support the applications. In these countries, computer applications have apparently been hindered thus far, and one significant cause has been poor interfacing, which divorces local users from the new technological opportunities. A combination of accumulated R&D experience with local user-specific engineering has proved to open the way to accelerate the applications in countries where the revolutionary expansion of the computer and information industry has not been indigenously cultivated. Spin-off restructuring bridges the gap by creating innovative and autonomous NTEs, and should also be applicable to the exploitation of other new technologies which are emerging.

6 THE TRANSFORMATION OF ESTABLISHED R&D INSTITUTES (SINCE THE 1990s)

General trends of transformations As a continuation of the market reform programme, there have been widespread transformations of existing industrial technology R&D institutes since the mid-1980s. These transformations involve internal restructuring to adapt to external changes, to the extent that the nature of the existing organizations is altered. In comparison, restructuring by means of spinning-off NTEs transfers only some ingredients of institutes into new establishments, while the parent institutes continue with little interruption. The general results of the transformations are shown by aggregate statistics which clearly indicate that institutes’ market earnings have increased strongly. Table 6.1 shows that the main source of income for industrial technology R&D institutes as a whole has shifted, and now comes from ‘horizontal’, i.e., market earnings, which accounted for almost 80 per cent of total income by 1993. Government funds contributed a remarkably small share of their overall income, 17 per cent in 1993, a large part of which was for S&T projects sponsored by state and sectoral plans, and state investment in laboratory installations. As for the structure of the ‘horizontal’ earnings, Table 6.2 shows that ‘other production and activities’ accounted for the largest part, 45 per cent. This income is obviously derived from conventional products and services. ‘Technology development’ and ‘trial production’ followed as the second and third sources. Together they contributed 37 per cent of the horizontal earnings. These two activities probably relate to new products and services, and involve searching for commercially promising technologies. Table 6.3 presents the income structure for the ‘remainder’ group of independent R&D institutes, for comparison. It shows that government funds are much more important for this group, accounting for 44 per cent of their total income, compared with 17 per cent for the industrial technology group. The market component, that is their ‘horizontal earnings’, of 49 per cent, is evidently less important than for the industrial technology group, yet the ‘remainder’ group has also been significantly marketized. Table 6.1 Income structure of government-run R&D institutes (all industries, 1993) Overall income

Structure of income

Government funds

‘Horizontal’ earnings

Other sources

13.57 (billion yuan) 100%

2.33 17%

10.59 78%

Source: Databook of Statistics on Science and Technology 1993:13, 52.

0.66 5%

32

REFORM POLICY FOR THE S&T SYSTEM

Overall income

Structure of income

Government funds ‘Horizontal’ earnings Other sources Note: The table (and Table 6.2) covers not only institutes belonging to central industrial ministries, but also those belonging to local industrial bureaux at the municipal and provincial levels. This group totalled 1,804 institutes in 1993. Table 6.2 Composition of ‘horizontal’ earnings of government-run R&D institutes (all industries, 1993) Overall ‘horizontal’ earnings

Structure of ‘horizontal’ earnings

Technology development

Technology transfer Technological Trial production Other production and consultancy & sales technological services

10.59 (billion yuan) 100%

2.00 19%

0.57 5%

1.32 12%

1.97 18%

4.75 43%

Source: Databook of Statistics on Science and Technology 1993:13, 70. Table 6.3 Income structure of the remaining R&D institutes (1993) Overall income

Structure of the income

Government funds

‘Horizontal’ earnings

Other sources

10.67 (billion yuan) 100%

4.75 44%

5.20 49%

0.71 7%

Source: Databank of Statistics on Science and Technology 1993:13, 52. Note: The ‘remaining R&D institutes’ refers to all the independent R&D institutes, excluding the government-run industrial technology R&D institutes defined in Table 6.1. This group totalled 3,701 institutes in 1993.

In accordance with the general trend shown in these figures, some consensus seems to have been reached recently in the community of policy researchers and institute directors in China, that the major direction of reform for R&D institutes engaged in industrial technology is to develop technology-intensive production.1 However we cannot simply conclude that the transformation of existing industrial technology R&D institutes involved only a general shift towards the market. Before turning to alternative kinds of transformations, we will provide some observations of the general trend at institute level. An illustration of transformation into a market-profitable corporation The general trends of the transformations point to the development of a number of manufacturing or engineering corporations with relatively intensive in-house R&D, as illustrated in Case Text 6.1. This institute is a well-known example, which we visited personally. Many similar cases have been emerging recently (SSTC 1991b; and Interview notes 5 and 8).

CASE TEXT 6.1 THE AUTOMATION RESEARCH INSTITUTE OF THE MINISTRY METALLURGICAL INDUSTRY (ARTMI) (CONTINUING CASE TEXT 3.2)

OF

THE

TRANSFORMATION OF R&D INSTITUTES

In 1987, the Institute made a strategic shift toward re-building the hierarchy of the Institute as a whole, in response to the failure of their delegation of decision-making power to small research teams contracting separately on the technology market. The new strategy, which was intended to enable the Institute to provide automation systems competitively, required a thorough reorganization of the Institute. Six larger departments were set up for 1) automation systems, 2) motor drives, 3) metallurgical instrumentation, 4) power generation components, 5) a.c. servo devices and 6) engineering design. Most of these departments embrace R&D, design, manufacturing, and marketing, i.e., almost every department is organized as a profit-making unit with integrated functions. Design, which is thought of as the ‘gateway through which R&D enters into the economy’ has been integrated at both the department level and the Institute level. At the Institute level, the engineering design department was established to provide design services for automation systems. Quality control has also been enhanced. It has been incorporated into the routine operations of departmental workshops, using the ISO 9000 as basic quality standard. Decision-making powers were re-centralized, away from the small teams to a two-level structure. The central power is at the Institute level, while the departments have flexibility to make business decisions within their specialized field in accordance with the overall targets of the Institute. The central power at the Institute level acts in several ways:

• the Institute is in charge of decisions regarding contracts for big projects involving more than one department; • the accounts office of each department is a branch of the Institute accounting office, working under the control of the Institute; • investment decisions are exclusively the domain of the Institute. Foreign involvement is sought to complement the Institute’s capabilities in the design and manufacture of some devices which the new Institute strategy requires. Four joint ventures have been set up: one with a German company for a.c. and d.c. drives; one with an American company for PLC and CNC devices etc.; one with a company from Hong Kong for instruments; and one with a French company for a.c. servo devices and CNC. The greater mobility of manpower in general, resulting from the reform of the S&T system, made it easier for the Institute to strengthen their engineering capabilities through the aggressive recruitment of key design personnel. Capabilities in complex metallurgical automation systems were built up by contacting international suppliers of the domestic market. The Institute worked as a sub-contractor for a large system initially contracted by foreign suppliers, or as a co-contractor with foreign contractors. They were attractive to their foreign partners because of 1) the Institute’s technological strengths in particular technology areas, such as drives, 2) their cheaper technologically-skilled labour, and 3) their closeness to, and awareness of, the user’s environment, leading to advantages in adapting imported systems for the operational site. These advantages make the Institute an indispensable local partner for the transnationals. As a result of their learning, the Institute has become quite competitive in the domestic market for smaller and less sophisticated metallurgical automation engineering. The transformation can be illustrated in financial terms. In 1993 ‘market sources’ (contracted engineering services and product sales) were estimated to constitute more than 90 per cent of gross income, and more than 80 per cent of net income. This is a reversal of the situation in 1987, when ‘state sources’ constituted two thirds of net income. The change was mainly due to the expansion of the ‘market sources’: contractual engineering projects increased twenty-fold over that period. ’Spin-off’ enterprises have been created from the Institute, but with entirely different roles from the NTEs. Twenty-six spin-offs have channelled about 300 staff members, mainly from the procurement and logistical departments of the Institute, into retail shops and restaurants. There is no intention of spinning-off R&D

33

34

REFORM POLICY FOR THE S&T SYSTEM

capability; rather, there is a partial externalization of some general and social services which were previously over-internalized within various ‘units’ of the R&D institute Sources: Interview notes 6; Information on Science and Technology of China No. 6, 1992:5, 8; SSTC 1991c: 62–65; Introduction to the major projects accomplished by the Institute, provided by the Automation Research Institute of the Ministry of the Metallurgical Industry.

The case demonstrates that the formulation of institute strategy is critical. Information on market opportunities does not provide adequate guidance for individual institutes on how to interact properly through effectively exploiting their institutional strengths. Internal restructuring has to be initiated as an institute strategy. The directors of the Automation Research Institute in Case Text 6.1 said emphatically that reformulating their own strategy to shift away from the internal delegation of powers to research teams was the most difficult step during the reform (SSTC 1991c:62–65; Interview notes 6). Profound internal restructuring proved to be indispensable to put institute strategies into effect. Usually pilot plants and trial production workshops, which already existed within many institutes, were re-organized as the basis for commercially exploiting the institute’s technological strengths. Departmental structures were adjusted. Decision-making power was re-divided. Abilities in design, manufacturing, quality control, marketing, etc., were usually weak, and were complemented in various ways which were more accessible because of the market reform. The profundity of the restructuring, illustrated to some extent in Case Text 6.1, explains why the transformation of R&D institutes took some time, before the institutes themselves were convinced this was the right path, and it was widely accepted by the S&T community. Policy response to the restructuring of industrial technology R&D institutes Several policy measures have followed, mainly in response to the transformation of individual industrial technology R&D institutes. Clarification of alternative restructuring approaches Up to the beginning of the 1990s, the reform policy attempted to answer the questions of what the future of the restructuring of industrial technology R&D institutes would be, and how it could be achieved. A document on the acceleration of S&T system reform, drawn up jointly by the State Science and Technology Commission and the State Economic Reform Commission, clarified some alternative paths for transformation which were encouraged. This document was circulated in late 1992. The many experiments and experiences which had been built up by then are evidently reflected in the five possible transformations which were suggested for industrial technology R&D institutes:2 • merging with existing enterprises, mainly for institutes engaged in the development of product technology; • continuing with the establishment of NTEs, and expanding the Zones; • transforming entire R&D institutes into new and high technology enterprises; • transforming R&D institutes into 1) productivity centres, providing technological services to small and medium enterprises in management, training, and information, or 2) into consultancy centres; • retaining a small number of state-supported institutes engaged in providing technological infrastructure for the whole society. Their total costs must be low enough to be sustained in the long term by the state.

TRANSFORMATION OF R&D INSTITUTES

35

National Engineering Centre Programme The National Engineering Centre Programme was started around 1990. Its aim is to reorganize industrial technology R&D capabilities by establishing ‘National Engineering Centres’. While most of these centres are located within existing industrial R&D institutes, the Programme has been expanded since 1994 to include large industrial enterprises. The Automation Research Institute of the Metallurgical Industry, for instance, has been authorized to establish the ‘National Engineering Technology Centre for Metallurgical Automation’ (in 1991).3 It is reported that by the second half of the 1990s there will be nearly one hundred national engineering centres in operation, with investment funds coming partly from the State Planning Commission and the State Science and Technology Commission, and partly from the host institute.4 Some ambiguity can be seen in the targets of the policy initiative, and its effects on the transformation process will require further study. The delegation of autonomy in doing business with foreign companies It was announced in 1993 that one hundred R&D institutes were to have autonomy in doing business directly with foreign companies.5 The majority of the one hundred are industrial technology R&D institutes. The remainder are from the Chinese Academy of Sciences (seven institutes) and institutes of higher education (seven units). This is a privileged position equivalent to the best-performing state-owned firms, which have also just started to enjoy autonomy. This seems to be intended to encourage the transformation of R&D institutes which are becoming engineering or manufacturing corporations. It is too early to evaluate the effectiveness of these policies. Moreover, the present policies are still far from sufficiently coherent to guide the complicated transformations, although progress has been made. Some institutes will unavoidably be dispersed, in the sense that they will cease to function as a unit undertaking any R&D or technologyrelated work (Interview notes 7:6).

7 CONCLUDING REMARKS

Some findings emerged from the earlier survey. The reform of industrial technology R&D institutions is indispensable to adapt to the new economic regime This historical review of the relationship between the producers and users of industrial technology before and after the current market reform in China has clearly shown that, in economic systems undergoing a transition, parallel transformations of the industrial technology R&D institutions are indispensable. Under the centrally planned economy, policy measures for industrial technology were restricted to the supplier side. While these policies were inefficient, they were consistent with that institutional framework. They were sustainable on the macro level by means of the allocation and re-allocation of investment, and on the micro level through the adjustment and improvement of institutional establishments. All the policy measures were characterized by direct intervention empowered by state planning. The rigid and uniform ‘externalization’ of the organization established for industrial technology was in essence an integral part of the institutional arrangement of an economy operating under pervasive administrative coordination. Marketoriented economic reform therefore required a fundamental reform for the institutions of industrial technology. This indispensability is clearly shown by linking the most important events in the economic reform with the stages of the reform of the R&D system which we have examined in the previous chapters. The parallels are summarized in Table 7.1. The table shows that the new strategy for economic development in China was set up in 1978, divorcing political doctrine from economic affairs and opening the Chinese economy to international participation. The aim was to improve the efficiency of economic growth. This led to economic reform of the agricultural sectors, which was under way around 1980, and that in turn inspired the industrial economic reform which was in full stride by about 1984. The reform of the ‘scientific and technological Table 7.1 The most important events for economic reform and for the reform of the R&D system Year

Important economic events

Important policy for the R&D system

Year

Before 1978

•

•

Before 1978

•

the establishment of a planned economy (the 1950s) ‘Great Leap Forward’ (1958–1960)

the institutionalization of the R&D system (the 1950s)

REFORM POLICY FOR THE S&T SYSTEM

Year

Important economic events •

1978

•

1979 1980

•

1981

1982 1983 1984

Cultural Revolution (1966–1976) ‘Open Door’ Policy (1978)

rural economic reform (around 1980) •

Important policy for the R&D system

Year

•

1978

rehabilitation and expansion of R&D system (1978–1980) 1979

37

1980 elaboration of R&D planning practice (1981– 1985)

1981

1982 •

1985

urban economic reform (1984) •

1983 1984 market reform decision for R&D system (1985)

1986 1987

•

1988 1989 1990

•

1991

•

1992

•

1993

1993

reform for taxation and banking system; and for state-owned firms (1993) transformation of R&D institutes (in the 1990s)

1985 1986 merging R&D institutes into enterprises (1987) 1988 Torch Programme (spinoffs) (1988) 1991

1987

1989 1990

1992

management system’ followed, beginning in 1985, although there had been previous wide-ranging attempts to adjust it using the planning approach. Both before and after 1985, the reform of the industrial technology system was driven by the gulf between a system which had been shaped by the central planning framework and the needs of the users of industrial technology, who had already been substantially freed to act within a market framework. Once the system had moved out of the centrally planned framework, plurality or diversity, a key characteristic under market regimes, started to be much more evident, as can be seen from the great variation in restructuring approaches and related policy measures. One striking feature of the policy process during the market reforms is that interaction between policy makers and affected parties, as well as between the users and producers of industrial technology, has become much stronger. Since R&D institutes and industrial enterprises have been given the right to make decisions independently, in response to government policy and the economic environment, they have a greater influence on the evolution of reform policy and reform practice. This greater plurality of constituents in the policy institution that evolved during the market

38

CONCLUDING REMARKS

reform provides a basis for more open experimentation, and therefore for much more intensive learning about both policies and institutions. The government, having been released from duty as the sole player in both policy formation and policy implementation, is becoming stronger in the role of ‘picking the winner’ and circulating the best practice. The inadequacy of the pure technology market approach The pure technology market approach alone proved to be inadequate in adapting the old R&D system to market-oriented economic regimes. Restructuring is the essence of the transition of industrial technology R&D institutions. The technology market was originally devised, in the reform policy of 1985, as the main bridge linking suppliers and users of industrial technology. It has been shown to be inadequate on its own in meeting this target. The uncertainty of technology transactions limits the efficiency of the market, and this problem is intrinsic to the process of technology innovation. In particular, the fierce competition resulting from the opening to foreign suppliers of technology, and the increasing complexity of the systems employed in various industrial operations, seem to heighten the uncertainty. It can also be seen that the inexperience of buyers, and the under-development of market institutions, were additional obstacles to concluding technology transactions. Releasing R&D institutes from rigid controls, though not sufficient in itself, did enable R&D institutes to act with the technology market, and it opened the way for various kinds of organizational restructuring involving functional redefinitions in the environment created by the market reform in general. The technology market turned out to be complementary to various transaction agents linking firms and the restructured R&D units. A tentative analysis indicates some characteristics of the technology market. First, ‘technological consultancy’ and ‘technological services’ were the main subjects of market transactions in the first ten years of recorded activity in China. These transactions were probably less deeply involved in the core activities of the buyers. Second, ‘technology transfer’ transactions were very rare in the market. Third, the ‘contractual development of technology’ displayed a significant increase, in terms of both contract numbers and contract value. This seems to point to some peculiar aspects of the market in linking R&D units and firms which are in the course of restructuring. Focusing on this hopeful sign might produce good results. The study of the machine industry reported in Part 3 reveals that some R&D institutes in the industry are restructuring themselves as suppliers of customized producer goods, while the technology market helps to link them with their clients. These transactions are recorded as the ‘contractual development of technology’. However it is measured, the technology market has been expanding, in absolute terms, since it was created, in step with the deepening of the market reform. The principle directions of restructuring, and some factors influencing the restructuring The common direction of restructuring in the overwhelming majority of cases thus far has been to integrate more activities within the same organizational territory. The integrations were intended to capitalize on institutes’ technological assets under the market rules. Transaction costs, which were even higher because of the inexperience of the actors, underdeveloped market institutions, and increasing complexity of technology, have pushed the restructuring toward more hierarchical forms of integration. The plurality of the restructuring process is illustrated by the fact that a few restructuring schemes have been widely experimented with, with varying degrees of success:

REFORM POLICY FOR THE S&T SYSTEM

39

• Merging R&D institutes into existing enterprises. This was initially prompted by the reform policy in 1987, as a response to dissatisfaction with the technology market solution. Successful cases of merging were rare, but all those achieved under the supervision of the responsible ministries in the two sectors survived. Many aspects of institutional incompatibility, mostly inherited from the past, were the main factor obstructing the organizational fusion. Technology imports tended to widen the gap between the institutes and the enterprises, in terms of technological acquisition and physical installation, but at the same time the increasing wealth of enterprises in a rapidly growing economy tended to make merging more affordable.1 Once the organizational barriers were surmounted, R&D assets tended to be well combined into the core technological activities of the host enterprise. Careful guidance of potentially feasible mergers was seen to be indispensable to overcome the institutional barriers, which seems to be beyond the capacity of the market. In this connection, the responsible ministry can have a positive function where there is a lack of well-developed regulatory agencies for merging. • ’Spinning-off’ new technology enterprises. This kind of restructuring began to take place as a spontaneous response to the market reform, on the initiative of autonomous R&D institutes and S&T personnel. It is characterized by channelling ingredients of the technological assets accumulated in R&D institutes into newly established enterprises. The integration of the assets with other factors necessary for commercial success is thus realized within new organizational setups. The policy measures in response to this spontaneous development, in the initiatives of the Torch Programme of 1988, sanctioned and accelerated the process. This kind of restructuring has expanded very significantly in China. It would be reasonable to devote more attention to it in formulating policy, since it has proved to be the most popular method of restructuring thus far, in both China and other countries.2 • The factors influencing the widespread boom of the ‘spin-off’ approach will be discussed in detail in Part 2, but two can be mentioned here. First, both the academic community and the local governments have been active in spinning-off. R&D institutes and S&T personnel were the main initiators of New Technology Enterprises in various forms, while the local governments supported the initiation of both NTEs and Zones. This was achieved in the context of the decentralization of decision-making authority in China’s reform. The second factor concerns the impact of the personal computer revolution, which has reached the stage at which widespread applications have become feasible. Skill-intensive and knowledgeintensive work is urgently needed to localize the global technology trajectory in the various sectors of national economies. In China, spin-off enterprises concentrated on this area, and the importance of this factor seems far from unique to China. • Transforming whole individual R&D institutes into manufacturing or engineering corporations with relatively intensive ‘in-house’ R&D. This kind of restructuring also began as a spontaneous response to market reform. The integration of institute assets with other factors necessary for market profitability was achieved within the territory of particular individual institutes, usually accompanied by profound internal restructuring. The profundity and severity of the internal restructuring required to remodel the institute, perhaps including shifting the focus of its specialization, explain why successful transformations of this kind came to be known and accepted only very recently, much later than ‘spinoff’ restructuring. Thus far, institutes restructured in this way have almost exclusively been moving towards becoming market profitable manufacturing or engineering corporations. Two factors modified the general direction which may be identified from this review of China’s reforms. First, very tough financial pressure was imposed on institutes throughout this period. This was deliberately designed to goad the institutes into the market. In other words, the policy environment in China’s case was quite unique, although a trend to reduce government funds for R&D institutes is currently very common in

40

CONCLUDING REMARKS

many other countries. Second, the strong orientation to physical output, which was characteristic of institutes in China, underpinned the direction of the transformation. This characteristic is shared by many former centrally planned economies to some extent, but there are some in which the analogy can be applied only with caution, such as India, where government-financed industrial technology R&D institutes have been relatively more academically oriented. Moreover, as far as the success of transformation in a certain institute goes, proper institute strategy seems to be the most important factor. The transformation of institutes is basically a matter of institute management, given that a high degree of autonomy has been granted to them. The outline of directions and influencing factors above should be seen as an effort to illustrate, rather than exhaustively describe, the restructuring possibilities. Nevertheless this typology of restructuring is quite solid as regards the types of transformation which have been tried on a large scale under China’s reforms. It should be noted that the Chinese practice has not yet been able to show how to manage successfully the transformation of some existing R&D capabilities to serve functions which are, in market terms, unprofitable. Such functions include the provision of technological services for small and medium-sized enterprises in fields such as training, management, and information3 and R&D for more advanced and fundamental industrial technology. These issues are being addressed by reform policy in China at present. The analysis of restructuring outlined above has been carried out only at an aggregate national level. Sectoral research is desirable and is likely to produce insights as to the dynamics of restructuring in the context of the relationship between R&D institutes, enterprises, foreign suppliers, and government policies. The characteristics of the underlying technologies of particular sectors warrant close examination in such research, since these usually have a large impact on the pattern of restructuring. The main purpose of reform policy—to address the lack of institutions Plurality or diversity has become the main characteristic of S&T policy in China, since the reform began to recast the political—economic regime, as has been pointed out above. Two strands of policy measure were provided for China’s S&T system reform: one setting up new regulations and incentives, and another for the creation of regulatory and other supporting frameworks. Both point to the establishment of the new institutions necessary if players are to operate under the new rules. Policy for the creation of regulatory and supporting frameworks, i.e., creating the organizational basis for putting regulations into effect, is particularly necessary in a transitional period, when these frameworks are especially inadequate. Different reform targets were embodied in different sets of supporting institutions. To introduce the technology market, policies under the 1985 Decision concentrated on laws, regulations, and agencies for technologyrelated transactions. An extensive and impressive framework of ‘technology market agencies’, and ‘personnel exchange and recruitment centres’ to support a quasi-market for S&T skilled labour resulted. To endorse and accelerate ‘spin-off’ restructuring, the policies under the 1988 Torch Programme concentrated on regulations, incentives, and agencies for technology-intensive, non-state small and mediumsized enterprises. The development of Zones, Centres, and financial institutions as vigorous organizational networks supporting NTEs was thereby encouraged. One significant highlight of this review is the cumulative nature of institution-building. Each successive effort resulted in some components of institutional construction which were useful in the next stage of the reform. The framework of centres for personnel exchange and recruitment, for instance, which was created earlier, largely supported the mobility of S&T persons required for the establishment of New Technology Enterprises. Both the technology market and the Development Zones for New Technology Industries helped the transformation of individual R&D institutes. Throughout this cumulative process, market institutions

REFORM POLICY FOR THE S&T SYSTEM

41

were expanded, elaborated, reinforced, and adjusted, so paving the way to further economic transition which was the goal of the reform. APPENDIX: STATISTICAL DATA ON CHINA’S R&D SYSTEM Appendix Table 1.1 Government appropriations for science and technology from the state budget (1953–1991) Year

Appropriations, in million yuan

Percentage of

Government budget

National income

GNP

1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984

56 122 213 523 523 1,124 1,915 3,381 1,949 1,373 1,861 2,427 2,717 2,506 1,535 1,480 2,415 2,996 3,768 3,610 3,459 3,465 4,031 3,925 4,148 5289 6,229 6,459 6,158 6,529 7,903 9,472

0.3 0.5 0.7 1.7 1.7 2.7 3.5 5.2 5.5 4.5 5.5 6.1 5.8 4.6 3.5 4.1 4.6 4.6 5.1 4.7 4.3 4.4 4.9 4.9 4.9 4.8 4.9 5.3 5.5 5.7 6.1 6.1

0.1 0.2 0.3 0.6 0.6 1.0 1.6 2.8 2.0 1.5 1.9 2.1 2.0 1.6 1.0 1.0 1.5 1.6 1.8 1.7 1.5 1.5 1.6 1.6 1.6 1.8 1.9 1.8 1.6 1.5 1.7 1.7

1.5 1.5 1.5 1.3 1.3 1.4 1.4

42

CONCLUDING REMARKS

Year

Appropriations, in million yuan

Percentage of

Government budget

National income

GNP

1985 1986 1987 1988 1989 1990 1991 1992

10,259 11,810 11,574 11,800 12,787 13,912 16,069 17,563

5.6 4.8 4.6 4.4 4.2 4.0 4.2 4.0

1.5 1.5 1.2 1.0 1.0 1.0 1.0 0.9

1.2 1.3 1.0 0.8 0.8 0.8 0.8 0.7

Sources: China Statistical Yearbook on Science and Technology 1991:24, 31; 1993:24, 31. Notes: 1 The appropriations in this table do not include funds used for R&D purposes by enterprises and institutions of higher education. 2 Before 1977, there were no available statistics on GDP in China. 3 The science and technology appropriations from the government budget were used for wages, R&D expenses and prototype testing. Appendix Table 1.2 National expenditure for R&D Year

Expenditure for R&D (million yuan)

Ratio of R&D expenditure to GNP

1990 1991 1992

12,543 14,230 16,900

0.71 0.72 0.70

Source: China Statistical Yearbook on Science and Technology 1993:23. Appendix Table 1.3 State-owned R&D institutes, by sector Year

Chinese Academy of Sciences

Institutes affiliated to central government

Institutes affiliated to Institutes affiliated to local government at county government higher than county level

Total

1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991

– – – – – 122 122 123 123 123 123 123

722a 757a 811a 857a 909a 622 922 910 877 887 904 912

3,495 3,597 3,647 3,593 3,796 3,946 4,227 4,189 3,933 4,001 4,057 4,092

4,217b 4,354b 4,458b 4,450b 4,705b 7,957 8,631 7,583 7,431 5,011b 5,084b 5,127b

n.a. n.a. n.a. n.a. n.a. 3,267 3,360 2,361 2,498 n.a. n.a. n.a.

REFORM POLICY FOR THE S&T SYSTEM

Year

Chinese Academy of Sciences

Institutes affiliated to central government

Institutes affiliated to Institutes affiliated to local government at county government higher than county level

Total

1992

123

915

4,078

5,116b

n.a.

43

Sources: SSB 1990a:207; China Statistical Yearbook on Science and Technology 1991:67, 68, 69; 1993:65, 67–69. Notes: a The sum of the institutes of the Chinese Academy of Sciences and institutes affiliated to ministries and commissions of the central government, b Do not include institutes which are affiliated to county governments. Appendix Table 1.4 R&D institutes of the Chinese Academy of Sciences Year

Numbers of institutes affiliated

Scientists and engineers (1, Annual expenditure (million 000) yuan)

1985 1986 1987 1988 1989 1990 1991 1992

122 122 123 123 123 123 123 123

32.2 34.5 37.0 38.9 n.a. 41.0 41.5 41.5

833 850 886 1,034 n.a. 1,319 1,398 1,828

Sources: ‘White Paper’ No. 1:232, Table 1–1; No. 2:269, Table 2–6; No. 3:344, Table 2–1; No. 4:212, Table 1.1.1, 215, Table 1.1.4.; China Statistical Yearbook on Science and Technology 1991:75–77, 125–126; 1992:75–77, 125–126; 1993:75–77, 124, 126, 128; SSB 1990a: 215, Tables 2–10. Appendix Table 1.5 R&D institutes affiliated to ministries and commissions of the central government Year

Numbers of institutes affiliated

Scientists and engineers (1,000)

Annual expenditure (million yuan)

1985 1986 1987 1988 1989 1990 1991 1992

622 922 910 877 887 904 912 915

93.0 175.1 197.1 206.2 n.a. 229.3 261.0 264.7

2,525 6,033 6,176 7,299 n.a. 9,682 12,049 14,194

Sources: See Table A1.4. Appendix Table 1.6 R&D institutes affiliated to local governments (excluding the institutes affiliated at the county level) Year

Numbers of institutes affiliated

Scientists and engineers (1,000)

Annual expenditure (million yuan)

1985 1986 1987

3,946 4,227 4,189

105.8 114.7 127.6

3,375 3,875 3,620

44

CONCLUDING REMARKS

Year

Numbers of institutes affiliated

Scientists and engineers (1,000)

Annual expenditure (million yuan)

1988 1989 1990 1991 1992

3,933 4,001 4,057 4,092 4,078

138.4 n.a. 154.0 153.7 154.2

4,490 n.a 5,697 6,989 9,068

Sources: See Table A1.4. Appendix Table 1.7 Government-owned R&D institutes, by fields (in 1988) (excluding the institutes affiliated at the county level) Field

Number of institutes Scientists and engineers (1,000)

Government appropriations (million yuan)

Expenditure (million yuan)

Natural Science Engineering Science and Technology Medical Science Agricultural Science Total

344 2,704

50.2 265.9

1,087.8 4,493.1

1,385.9 9,352.1

389 1,496 4,933

23.7 43.6 383.5

453.9 949.3 6,984.1

757.5 1,327.2 12,822.8

Sources: See Table A1.4. Appendix Table 1.8 R&D institutions of higher education Year

Number of R&D laboratories affiliated Scientists/engineers

engaging in R&D (1,000) (in full time equivalents)

Annual expenditure for R&D (million yuan)

1986 1987 1988 1989 1990 1991 1992

1,490 1,514 1,715 1,739 1,666 1,676 1,819

83.9 94.2 106.9 112.1 116.4 117.1 122.3

597 736 874 991 1,194 1,353 2,080

Source: China Statistical Yearbook on Science and Technology 1991:209; 1992:209; 1993:209. Appendix Table 1.9 Internal R&D of large and medium-sized enterprises Year

Number of enterprises having internal R&D departments/total number of enterprises

Scientists/engineers engaged in R&D (1,000)

Annual expenditure for internal R&D and other technological development (million yuan)

1987 1988 1989 1990 1991

4,633/9,681 5,119/10,738 6,424/12,222 7,289/13,475 7,899/14,935

198 269 307 314 334

8,798 11,604 12,377 13,306 16,599

REFORM POLICY FOR THE S&T SYSTEM

Year

Number of enterprises having internal R&D departments/total number of enterprises

Scientists/engineers engaged in R&D (1,000)

Annual expenditure for internal R&D and other technological development (million yuan)

1992

8,576/16,991

373

20,881

Sources: China Statistical Yearbook on Science and Technology 1992:149; 1993:149.

45

Part 2 SPIN-OFF ENTERPRISES Channelling the components of R&D institutions into innovative businesses

8 INTRODUCTION TO SPIN-OFF ENTERPRISES

Part 2 examines the massive and influential restructuring which began to take place in the second half of the 1980s through the creation of new business enterprises spun off from R&D institutions. ‘Spin-off’ enterprises represent one of the mechanisms for integrating the capabilities (or outputs) of R&D institutes with industrial production. Instead of institutes being incorporated into existing industrial enterprises, or their R&D outputs being ‘transferred’ to such enterprises through market or other mechanisms, in restructuring through spinning-off some of the technological assets of institutes, mainly knowledge, skills, product or process specifications, are channelled out of the old structure, to form the nucleus of new ventures. The new organizational base of the spin-off enterprise enables other assets (i.e. finance, production facilities and competence, market knowledge and marketing expertise) to be recombined with the technological assets needed for new economic activities more effectively. The emergence of the spin-off enterprises has proved critically important for the economic development and economic reforms in China. The experience of Chinese policy makers during the spin-off restructuring has been an important source of the new ideas applied in reforming industrial firms. The creation of spin-off enterprises has led to the birth of the Chinese computer software industry which, in turn, has been functioning as a major supportive capacity for the explosively increasing applications of computer and information technology in China in the 1990s. Spinning-off has proved to be a fruitful approach to the institutional restructuring of the R&D system, which is indispensable for various economic reforms, not least because it offers a way of making good use of accumulated capabilities to contribute to the widespread dissemination of computer and information technology. The remainder of this introduction will recap on the essence of the institutional restructuring which is indispensable to market reform, and some features of the Chinese R&D system that are relevant to spin-off restructuring. The final paragraphs of the introduction outline the scope and structure of Part 2. The essence of institutional restructuring in market reform In China Market-oriented reform for the S&T system was formally introduced in China in 1985, by the declaration and implementation of the ‘Decision on the Reform of the Science and Technology Management System’ (CCCPC 1985; see Part 1). The government funding for industrial technology R&D institutes was to be gradually diminished, and a ‘technology market’ to be created.1 The tremendous development of spin-off enterprises from R&D institutions took place in this context.

48

SPIN-OFF ENTERPRISES

The immediate experience of the technological market was not very pleasant, for either buyers or sellers. The attempt to merge R&D institutes into existing industrial enterprises which was strongly promoted by the reform policy launched in 1987, was not successful at that time. But in the same period, the mid-1980s, spin-off enterprises were emerging as a response by R&D institutes and scientific and technical professionals to the reform in general, and to their frustrating experiences with the pure technological market solution in particular. This initiative was soon incorporated in the reform policies with the launch of the Torch Programme. Spinning-off, like most of the options for restructuring R&D institutions which have been developed since the market reform (discussed in Part 1) were directed towards the internalization of the various functions necessary for the commercially competitive development of technology. This has been happening in parallel with the slow but steady strengthening of in-house R&D and other technologically innovative activities within productive enterprises. The ‘technology market’ introduced to mandate paid transactions for technologies has served as a complementary means of transferring technology between organizational entities. In some former centrally planned economies Outside China, a growing literature is reporting poor sales records for the independent R&D institutions of the former planned economies when they function in the marketplace.2 This is related to a number of problems encountered there: a brain drain, crises in R&D organizations and, most critically, acute disputes regarding reform policy. Reform policy for the R&D sector has in fact been weak and muddled. The serious question which has arisen is: how to re-deploy the scientific resources which have built up within, and were grounded in, the previous institutional framework? Chris Freeman put the problem in these terms: Most of the former Communist countries did actually invest fairly heavily in R&D and related technology activities. But this investment suffered from great distortions such as the huge military projects in old USSR, and the tendency to neglect the enterprise-level scientific and technical activities. Therefore the re-orientation and re-deployment of the science-technology resources of the former Communist countries is one of the most urgent problems which they confront. (Freeman, in Bentley 1992:xiii–xiv) These observations highlight the importance of focusing on the restructuring of the old S&T system. It seems that a blanket approval of marketization, which is pervasive under reformist policies as a talisman to cure problems produced by dysfunctional administrative control, is far from adequate to offer sensible guidance to a smooth transition. This blocks the thinking and hampers the formulation of proper S&T policies under the present reforms. Issues such as 1) a deeper understanding of the various phases of restructuring since the reforms began, and 2) a more exact theoretical interpretation of the abilities and limitations of pure market mechanisms with respect to the transformation of the S&T system cannot be overlooked in any analysis of intellectual labour. Other historical observations While the restructuring of the S&T system is still developing in many places where market reforms are now taking place, other restructuring responses to market forces have been observed previously. Among many others, one useful analysis has been made by D.Mowery and N.Rosenberg, with respect to institutional

INTRODUCTION

49

restructuring in the first half of this century in the US. At that time science, or more precisely, the utilization of scientific knowledge, methods and principles for industrial purposes, was beginning to be widely applied by industries in a market environment. The outcome was that science was integrated into industrial enterprises, in the form of in-house R&D departments. At the same time, independent research enterprises were decreasing in importance, in relative terms, and came to serve as complementary sources of industrial technologies. This analysis is a useful reference-point for our study, in the sense that market rules are just now being introduced in China into a domain in which all industrial enterprises and scientific institutes were previously institutionalized under a quite different system of control. As for the restructuring which the introduction of market rules entails, the following lessons can be learnt from the American experience: • ‘the highly uncertain nature of the research enterprise, the imperfect character of knowledge about a given project, and the thin market for specialized research service’ undermines ‘the effectiveness of contracts in the provision of research’ (Mowery and Rosenberg 1989:82); • ‘commercially successful innovation requires the combination of skills and information from a wide range of functions within the firm and often exploits firm-specific knowledge emerging from complex production process’ (ibid.: 91); • ‘to be effective, industrial research requires complementary changes in the structure and organization of firms and markets. Where these changes have not occurred, industrial research has been more modest in scope and less effective’ (ibid.: 92). China’s S&T system It is useful for our current focus to recapitulate some features of the Research and Development (R&D) system developed in China which have already been mentioned in Part 1, but from the perspective of reform policy development. Table 8.1 presents a number of basic indicators of this system, showing the Table 8.1 Basic indicators of China’s R&D system 1 2 3

4

5 6

Expenditure for R&D: usually slightly less than 1% of GNP. In 1990 it was 0.7%. R&D Scientists and Engineers: 653,000 (in full time equivalents, in the fields of natural science and engineering, with tertiary or higher educational background/or equivalent qualification). Distribution of expenditure in 1990: Independent R&D institutes: 70% Higher educational institutions: 5% Enterprises: 25% Distribution of R&D scientists and engineers in 1990: Independent R&D institutes: 59% Higher educational institutes: 18% (in full time equivalents) Enterprises: 23% Total number of independent R&D institutes in 1990:5,084. Sectoral distribution of independent R&D institutes in 1990: i) Agriculture, Forestry, Animal Husbandry, Fishing, Water Conservancy: 1,552 institutes, 9% of the expenditure by all ‘independent’ R&D institutes; ii) Industry, Transportation, Post and Telecommunications: 2,109 institutes, 54% of expenditure;

50

SPIN-OFF ENTERPRISES

iii) iv) v)

Comprehensive Scientific and Engineering R&D: 242 institutes, 14% of expenditure; Medicine and Public Health: 332 institutes, 5% of expenditure; Others (including Geological Survey and Exploration; Construction; Environmental Protection; Measurements; Meteorology, etc.): slightly more than 10% of expenditure.

Source: Calculated from data in the China Statistical Yearbook on Science and Technology 1991:1, 31, 70, 118. Richard Conroy provides a similar description of the situation for the early 1980s (Conroy 1992: Chapter 2).

size of the Chinese R&D system and its sectoral distribution in terms of manpower and expenditure. The first significant feature is the size of the system, indicated by items 1 and 2 in Table 8.1. This is a huge system in terms of both total manpower and overall R&D expenditure, though it is moderate in per capita terms even as compared to some developing countries.3 Second, R&D institutes for industrial technology were separated from the industrial users of their outputs. Those carrying out industrial R&D were organized in, and confined to, their designated departmental administrations according to the institutional rules of the time—as were the end-users of their research. Organizational segmentation was so serious that, until 1990, only about one fourth of R&D activities, in terms of expenditure as well as manpower involved, were conducted in enterprises, as shown by indicators 3 and 4. This gave rise to serious problems in ‘linking’ R&D institutes and production enterprises, or ‘transferring’ R&D output to its industrial users. Third, the numerous R&D institutes are widely dispersed among the levels of the administrative hierarchy, as implied indirectly by indicator 5. This dispersal occurred in parallel with the decentralization of the economy in China from the 1950s to the 1970s. The resulting widespread geographical distribution of R&D institutions facilitated the participation of local governments in the initiation of the Development Zones for New Technology Industries (‘Zones’), encouraged by the Torch Programme to foster the development of the spin-off enterprises, as we will see in the following chapters. Finally, the independent R&D institutes are also widely dispersed among sectors, including agricultural, industrial, medical and public health care, ‘comprehensive scientific and engineering’ (such as the institutes of the Chinese Academy of Sciences) and higher educational sectors. Our survey reveals that most of the spin-off enterprises were initiated from the comprehensive scientific and engineering and the higher educational sectors (together with a small segment of industrial sectors), whereas the restructuring of established industrial R&D institutes which will be the focus of Part 3 involves mainly R&D institutes in the industrial sector. The scope and structure of Part 2 This study examines two fundamental dimensions of the way in which the components of R&D institutions have been channelled into innovative businesses: 1) the initiation of spin-off enterprises, and 2) the technological activities which the spin-off enterprises are engaged in. Both are examined in an attempt to understand the basic aspects of the restructuring, i.e. how the existing organizational frameworks can be restructured, how far the spin-off enterprises have come in terms of gaining capabilities to provide users with products and services, and where they are going. The exploration of these two dimensions was not easy. Neither official statistics nor published data could be directly applied as indicators in some key respects. For instance, the ownership records of spin-off enterprises which were maintained by the regulatory agents (the ‘Zones’, see Chapter 9) gave a picture of the

INTRODUCTION

51

policies and political choices made by the initiators of the spin-offs, rather than an accurate definition of the nature of each enterprise’s initial capital. Field work, conducted from May to August 1993, was thus heavily drawn on for firsthand material. The field survey was designed to apply our examinations to three principle kinds of establishments: R&D institutes, Zones and spin-off enterprises. They are recorded in the Appendix to the bibliography, which gives details of interviewees cited in the text, institutes and enterprises visited, and the interview notes cited in the text. Chapter 9 provides an overview of the origin of spin-offs, the launch of the Torch Programme and the further development of both spin-off enterprises and the Development Zones for New Technology Industries, the policy instrument for the implementation of the Torch Programme. Chapter 10 presents the outcomes of our examination on the first dimension mentioned above, the initiation of spin-off enterprises. Various forms of spinning-off are identified, based on case analyses which specify the spinning-off process in each case, what kind of R&D institute assets was transferred, and the nature of the resulting relationship between the R&D institute and spin-off enterprise. Then a quantitative estimate is made of the relative importance of organizational initiators versus individual initiators of spinoff enterprises. This shows that R&D institutes were the predominant initiators during the early years of spin-off restructuring, and were responsible for the birth of nearly half of the spin-off enterprises. Individual initiators, although they were less prominent and not properly recorded, accounted for the creation of about two tenths of the total. The remainder were mainly established by existing industrial enterprises, which is outside the focus of this study. The roles of R&D institutions and of local governments are also revealed in the case studies. Together, and with some division of labour, they provided not only technological assets, but also capital investment and social and political credit, and served as regulatory and monitoring agents for the spin-off enterprises, a new type of business organization in the Chinese context. These roles are discussed further in the final section of Chapter 10 in the context of the decentralization of decision-making in the Chinese system, which has been one of the major approaches to implementing reform programmes since the end of the 1970s. Chapter 11 analyses the technological activities of the spin-off enterprises. On the basis of field work, these activities are grouped into two basic classes: 1) computer and information technology and their integration with mechanical technology; and 2) new biological, medical or chemical materials. The technological activities of the majority of the spin-off enterprises were found to fall in the field of computer and information technology. Much of Chapter 11 is then devoted to detailed analyses of the activities relating to computer and information technology in individual Zones. By showing that the software-intensive development of small systems is the key characteristic of these activities, the analysis uncovers a picture of the way in which a computer software industry originated in China along with the emerging spin-off enterprises in response to the computer revolution during the second half of the 1980s. A short chapter of concluding remarks summarizes the findings of Part 2 regarding spin-off enterprises, as a promising approach to institutional restructuring.

9 AN OVERVIEW The launch of the Torch Programme and the development of spin-off enterprises

By the early 1990s, a very large number of spin-off enterprises had emerged from R&D institutions. They grew very rapidly following the launch of the Torch Programme in 1988, which provided policy support for the development of such enterprises. However, the origins of this kind of institutional development can be traced back to the first half of the 1980s, coming from autonomous initiatives by individual scientific and technical experts as well as R&D institutions. The number of spin-offs was already significant in 1988 when reform policy makers acted in response to them. Hence the process outlined in this chapter is in accordance with one of the major arguments developed in Part 1, from the review of reform policies over the longer period since the end of the 1970s. That is, that a positive policy process is possible where there is a certain degree of consensus between policy makers and economic actors as to the target of changes, and where intensive interactions between policy makers and the practical experiences of the economic actors are feasible. Spin-off restructuring, which by the late 1980s had developed as one of the major approaches to reforming the scientific and technological system in China, is typical of the way in which policy formalities can respond to practice. Policy sanctioned existing autonomous initiatives and provided further guidance and support for the emerging institutional innovations. The uncertainties that are often associated with radical institutional changes could be considerably reduced because of the openness of the interactive process that enabled partially tested experience to be better absorbed. Origins of spin-offs In 1980, Cheng Chunxian, a Research Professor at the Institute of Physics of the Chinese Academy of Sciences (CAS), with the support of the Science and Technology Association of Beijing, created the first technological development entity that was not initiated, financed or owned by the state. It later became a Chinese-American joint venture: the Beijing Huaxia Guigu (China-Silicon Valley) Information System Corporation Ltd (Chen Zhaoying et al. 1992:2). Other enterprises of this type followed in Beijing, initially in small numbers and with considerable caution in the face of official suspicion and sometimes hostility. At first most were very small ventures, although the boundaries between the ventures themselves and the originating institutions were sometimes blurred. Nevertheless, several rapidly became quite independent businesses. By the mid-1980s the best known of them were the so-called Two Tongs and Two Hais—Sitong (i.e. Stone), Xin-tong, Ke-hai and Jing-hai. All four began as businesses focusing on computer technology. By 1985 there were about 100 such ventures, and the main street on which they were concentrated had begun to be called Electronics Street. The sixteen largest of these ventures were reported to have a combined turnover of 120 million yuan in 1985 (Wang Xiaolong (ed.) 1993:41).

SPIN-OFF ENTERPRISES

53

In the mid-1980s these spatial concentrations of spin-off enterprises were more formally recognized and encouraged. In July 1985, the Shenzhen Science and Industry Park was founded as a joint initiative by the CAS and the Shenzhen Municipal Government. The park was set up to create a base area where advanced technology from CAS and other institutes could be combined with foreign investment and technology to open up a new way of developing commercial high-technology products (Zhang Yiyi 1989). In May 1988, the State Science and Technology Commission launched the Torch Programme as a mechanism to consolidate and further encourage two linked types of institutional change: the emergence of ‘spin-off’ enterprises and the development of geographical areas in which they could be concentrated. The launch of the Torch Programme The Torch Programme was developed in response to concerns for the future and frustrations from the past. New national objectives were being formulated and it was recognized that they would have to be achieved in new international and technological contexts. On the other hand, previous efforts to link R&D capabilities with industrial production were seen to have been unsatisfactory. The future-centred concerns involved broad and ambitious targets to mobilize domestic scientific and technological strengths to support the development of new and high technology industries and an exportoriented strategy. The export strategy had been announced in 1988, and applied mainly to the coastal provinces of China. Much of the debate about how this export orientation might be achieved emphasized the necessity of ‘releasing’ science and technology manpower from the constraints of existing institutions in order to organize new businesses, i.e., to create ‘spin-offs’. Some main points were frequently stressed:1 • China’s international market-oriented strategy for industrialization could not repeat the pattern of fast growth in South Korea and Taiwan in the 1960s and 1970s, which had been characterized by the labourintensive assembly of components and final products. To drive China’s economic take-off, technologyintensive and skill-intensive sectors should be given high priority at the same time as developing labourintensive industry. • China had developed strong manpower capabilities in science and technology which should be effectively employed without delay, otherwise other Asian counties would increase their efforts to borrow these capabilities to upgrade their own industries. • The most important issue hampering the commercialization of science and technology was the immobility of personnel who were locked into their existing institutions. However, in the changing political and economic climate, it had become increasingly feasible to release scientific and technological talent to create new forms of business enterprise. • To commercialize technological know-how and expertise, various complements would be indispensable, which were then separate or lacking. However, better access to international and domestic markets under the Open Door policy would make it possible to acquire these complements from the marketplace. • Creating new, internationally competitive, business entities would introduce international management and institutional norms, so injecting an ongoing momentum to reform and development in China. The above concern about the broad development objectives was aggravated by frustration with previous efforts to link R&D institutes and industrial enterprises during the 1980s. Both efforts to incorporate R&D institutes into existing enterprises and the use of market-mediated and other mechanisms to strengthen technology transfer were seen as unsatisfactory. Referring to the weakness of the latter, Mr Song Jian,

54

THE TORCH PROGRAMME

Chairman of the State Science and Technology Commission, stressed in his speech at the inaugural National Working Meeting of the Torch Programme in 1988 (‘White Paper’ No. 3:415): We acknowledge that we can expect to find better solutions to the problems which have been faced by R&D institutes and universities, i.e. the limited ability of large and medium-sized enterprises to absorb [external] technologies and the difficulties arising from excessively small [technology] markets. The launch of the Torch Programme also reflected a growing recognition of the role being played by the spin-off enterprises that had emerged over recent years. Again this was highlighted by Mr Song Jian: The business entities which have been initiated by scientific and technical experts, based on their scientific and technological strengths strengths and on the integration of [technological] development, production and marketing, are engaged in transforming accumulated S&T achievements into productive power and commodities. The Torch Programme was thus designed to concentrate on creating new institutions to support the development of spin-off enterprises, or New Technology Enterprises (NTEs) as they were formally known. The concentration on supportive institutions in the Torch Programme marked a significant departure from earlier government programmes which had been characterized by direct control over the actions of economic agents. An array of policy measures were announced under the Torch Programme (for details see ‘White Paper’ No. 3:245–250): • to encourage the science and technology institutions to start up spin-off enterprises under a variety of forms of ownership, fiscal and other incentives would be available to approved NTEs, and patented and other proprietary technology would be allowed to be accounted as equity; • to create a favourable environment for NTEs, Development Zones for New Technology Industries would be established; • Service Centres for Scientific and Technical Entrepreneurs would be established as ‘incubators’ for spinoff enterprises, especially those initiated by individuals; • to finance the commercialization of technology, the government would set up ‘leading funds’ and various other financial sources, mainly banks, were encouraged to open and expand credit facilities for commercializing technology. New Technology Enterprises were to be formally licensed by Zone authorities, but had to meet several criteria (SSTC 1991a:563–566): • the technology underpinning the activities of the enterprise should be in specified areas of ‘new and high’ technology defined by the State Science and Technology Commission;2 • the enterprises should have appropriate capital and physical resources, market potential, and acceptable organizational and managerial abilities; • the chief manager should be a scientific or technical professional. Through the implementation of the Torch Programme, the Development Zones for New Technology Industries (hereafter referred to simply as Zones) grew rapidly. The Zones served as the main policy vehicle

SPIN-OFF ENTERPRISES

55

for promoting and regulating the development of spin-off enterprises. And most of the spin-offs which had been established before the Programme soon became formally licensed NTEs. The development of Zones and NTEs Responses to the Torch Programme from academics and local governments were enthusiastic. Many municipal governments initiated Zones, applying to the central government for approval of their status as ‘National Level’ Zones. The pace of development generated concerns about the demand for investment resources and the quality of the Zones being created, and the central government warned local governments about excessive haste and placed restrictions on the establishment of new Zones. Nevertheless the number of Zones expanded rapidly, as shown in Table 9.1 By 1992 there were fifty-two National Zones incorporating 5,669 NTEs. In addition, there were a number of Zones initiated by local governments, but not approved by the State Council as national. The Zones were geographically widely spread. By 1992, fifty-two Zones had been spread in twenty-seven of the twenty-nine mainland Provinces (including Hainan). The educational Table 9.1 The development of Zones and NTEs (1990–1992) Indicator

1990

1991

1992

Zones NTEs of which NTEs with foreign capital involvement Employment Turnover of which turnover from products1 turnover from technological services2 turnover from trading3 Annual exports Expenditure for technological development

27 1,652

27 2,587

52 5,569

persons m. yuan

75 122,889 7,567.1

167 138,231 8,729.5

564 340,346 230,924.9

m. yuan m. yuan

56% 16% 28% 688.7 418.3

51% 20% 29% 714.6 781.3

65% 11% 23% 16,359.1 15,238.1

Source: China Statistical Yearbook on Science and Technology 1993:307. Notes (for Tables 9.1 and 9.2): 1 Turnover from products refers to income earned from sales of the enterprise’s own production. 2 Turnover from ‘technological services’ refers to turnover from: a technological transfer; b contractual design, engineering, etc.; c technological consultancy services and other technological services; d royalties for intellectual property in various uses outside the company; e contractual R&D projects for outside parties; f sales of products produced in the internal pilot plant. 3 Turnover from trading refers to income earned by selling commodities not produced by the company itself.

backgrounds of employees in NTEs were typically much higher than average. In Beijing Zone, for instance, about 50 per cent of employees had educational qualifications at the level of a university degree or higher (Beijing Zone 1993: 21). Although no recent information was available, the role of foreign investment in

56

THE TORCH PROGRAMME

the development of NTEs was limited until 1991. Only 176 of the 2,473 NTEs in the Beijing Zone involved foreign capital as reported by the Zone in 1992. Not surprisingly, there are considerable differences between the Zones. One attempt to reflect some of this diversity has been made in a study based on a databank of information on the Zones (Chen Zhaoying et al. 1992: Chapter 3). This suggests that most of the national Zones can be grouped into three fairly clear-cut categories. • The first involves Zones located in urban areas where R&D institutes and universities are concentrated. Often spin-offs had already been established to some extent in these areas and the Zones were intended to accelerate the process. Typical cases of this category include the Zones in Beijing, Tianjing, Wuhan, Shenyang and Changchun. • The second includes Zones located in existing or planned industrial areas. A typical example is the Caohejing Zone in Shanghai where there were already firms in technologically advanced industries and several universities and R&D institutes. By establishing this Zone, the Shanghai municipal government attempted to take advantages of the Torch Programme to further its strategic efforts for regional development. However, it is not yet clear how establishing a Zone will, on its own, stimulate the development of new ways of linking the R&D capabilities of existing institutes with industrial enterprises. • A third category is similar to the well-known research or science park, in which a relatively small area is newly opened to attract foreign and domestic investors in technology-intensive projects. High standard infrastructure is constructed for the purpose. A typical example is the Shenzhen Science and Industrial Park. There seems to be a convergence between Zones of this category and the Special Economic Zones, whose purpose is to attract foreign investment in general. The Zones chosen to be surveyed in the study are all in the first category, the category which is most pertinent to the focus of the study. Table 9.2 presents the basic indicators of the four Zones chosen. Tables 9.1 and 9.2 show some features of NTEs. First, a significant part of the business portfolio of NTEs relates to trading and technological services, as can be seen from their income structure. This is particularly true for the Beijing Zone, and for Zones in their earlier years, indicating that most NTEs started by providing retail and user services for computers. NTEs are widely regarded in China as being characterized by a higher degree of integration Table 9.2 Sample Zones of the study (1992) Indicator

Beijing Zone Shenyang Zone Wuhan Zone Hangzhou Zone

NTEs of which NTEs with foreign capital involvement Employment Turnover of which turnover from products1 turnover from technological services2 turnover from trading3 Exports

1,512

631

246

81

persons m. yuan

155 43,567 58,414

89 35,014 15,166

46 14,905 9,659

3 2,405 2,207

m. yuan

34% 18% 48% 5,052

58% 13% 28% 892

62% 14% 24% 79

66% 7% 27% 81

SPIN-OFF ENTERPRISES

Indicator Annual expenditure for technological development

57

Beijing Zone Shenyang Zone Wuhan Zone Hangzhou Zone m. yuan

3,990

898

320

124

Source: China Statistical Yearbook on Science and Technology 1993:308–309. See notes for Table 9.1.

between technology, production and retailing, and by being run autonomously. The second striking feature is that these integrated ventures have relatively intensive technological inputs, as can be seen from the high ratio between their expenditure for ‘technological development’ and their turnover. The national average ratio for all the Zones ranged between 5 and 9 per cent during 1990 to 1992, and the figure for the Beijing Zone in 1992 was 7 per cent. In comparison, the national average for all industrial enterprises was less than 1 per cent. A third feature is that the production and services of NTEs are oriented to the domestic market, as can be seen from the small number of NTEs which involve foreign capital, and by the limited value of exports as compared to gross production turnover, at least until 1992. In fact, spinning-off represents a move to respond to, and localize, new technology opportunities, by releasing domestically accumulated strengths. This is in contrast to firms in the Special Economic Zones in the coastal area of China. The following chapters will discuss the characteristics of the R&D institutes, Zones and NTEs in more detail.

10 THE INITIATION OF THE NTEs

This chapter explores the institutional framework in which spin-off enterprises were brought into being. Two dimensions of the institutional framework are examined: the forms of spinning-off, and the initiators of spin-off enterprises. In the first perspective, we ask what kinds of technological assets were transferred into a new NTE and how were they transferred? This perspective focuses on the sources of the technological assets which are essential for the start of an NTE. The second perspective looks at the ‘initiators’ of spin-off enterprises, based on observations of existing NTEs. We ask who owns an NTE, either actually or officially. We seek to understand the relationship between the institute which provided technological assets for an NTE, the NTE which received the assets and other initial capital for its establishment, and the Zone and Centre institutions who were established as policy instruments for spin-off enterprises. These two perspectives are certainly interwoven and overlapping, but they highlight different facets of the complicated system which supported the emergence of spin-off enterprises. A discussion then offers tentative interpretations for some questions which are frequently asked in the broader context of the decentralized socio-political structure in China, such as how the decisions about channelling technological assets and other investments were made, what were the incentives for these decisions, and those related to property rights and the efficiency of NTEs. Forms of spinning-off Information about the forms of spinning-off came from our field work at both R&D institutions and selected NTEs. There are no official statistics which could be used to differentiate the forms, but from the survey three forms of spinning-off could be distinguished. They are: spinning off an organized part of an institute’s assets, spinning-off individual talent and transforming an internal institute department to engage in commercial activities while remaining as an integrated part of the parent institute. Each form shall be illustrated by case material. Form 1: Spinning-off an organized part This is a form of spinning-off in which part of the organized structure of an institute, embracing manpower, technological and, frequently, physical assets, is diverted into the establishment of a new, independent NTE. Obviously, this form of spinning-off generates an NTE which is at its outset stronger than those created by spinning-off individual talent, not only in terms of technological know-how, initial capital, physical installations and real estate, but also in terms of internal cooperative relationships, the staff’s trust in the leadership, and social links. All these are associated with the organized structure and inherited from

INITIATION OF NTES

59

the parent institute. The following case (Case Text 10.1) of the Legend Computer Group Corporation is a good example. The Corporation has benefited from its perceptive strategy, which was largely realized through the efforts of a rather cohesive and self-assured group of leaders, who were engineers in computer technology and some of whom had built up extensive management experience in the parent institute.

CASE TEXT 10.1 LEGEND COMPUTER GROUP CORPORATION (LEGEND), BEIJING Legend is an example of a form 1 spin-off enterprise. It began in 1984, with about forty professionals in computer science and technology spun off from the Institute of Computer Technology of the Chinese Academy of Sciences. Legend now ranks among the top NTEs, and is more productive and competitive than most stateowned computer enterprises in terms of turnover, exports, etc. Turnover has reached several hundred million yuan. Legend’s ownership is complex: ‘Legend Beijing’, the original spin-off, is ‘publicly’ owned and licensed as an NTE in the Beijing Zone. Legend Beijing has in turn engaged in joint ventures with Hong Kong and other foreign private capital. Technologically, Legend began with selling and after-sales service for imported personal computer sets, and soon added their own development work, especially in Chinese character processing systems. They produced a series of ‘Legend CCSs’ (Chinese Character System), which became the ‘core technology’ of Legend. The products of Legend have penetrated into international markets at the ‘boards’ and ‘cards’ level, i.e. subsystems of PCs, where they have a small share of this segment of the international market. Domestically, Legend has been strong at the level of complete computers. Their range extends from 286 to 486 PCs.1 Legend has developed intensive international links in various forms. Autonomous decision-making although under official public ownership made this possible. Legend Technology Ltd., Hong Kong, founded in 1989, is a joint venture of Legend Beijing with a private Hong Kong-based computer company. This meant that the technological strength of mainland R&D could be complemented by the competence of the Hong Kong company in information access and marketing capacity. The joint venture Legend Hong Kong then furthered its links with companies in North America and Europe, to gain access to advanced design and manufacturing, as well as the supply of key components, which have enabled Legend to translate their core technology into VLSI circuitry. Sources: Field interviews; Liu Chuanzhi 1991:219–224; Lu Tanpin: ‘Legend Going to the World’, mimeo 1990; Science and Technology Daily, 9, 13, 28 May and 2, 6, 8, 10, 11 June, 1992; Electronics International, 8 May 1992; Electronics Commercial, 26 May 1992; Wenhui Daily (of Hong Kong), 5 April 1992; Far Eastern Economic Review, 23 September 1993.

Form 2: Spinning-off individual talent This is a form of starting up an NTE through the mobility of scientists and engineers, moving individually from their previous R&D institutes. Sometimes a few of them united to combine their personal embodied knowledge, experiences, innovative ideas and management competence for a new company. When an NTE starts in this way other assets, especially capital, are required to complement the individual embodied technological assets, because no individual persons could afford to invest in an NTE in the 1980s, though the investment would not necessarily be very high. Besides, socio-political support was indispensable for spinning-off based on individual talent, especially in the 1980s, so that the NTE could be recognized in an officially acceptable category of ownership (i.e., non-private ownership). Being ‘private’ meant more difficulty in being licensed as an NTE, and in obtaining bank loans. These difficulties could be overcome by joining a ‘Centre for Scientific and Technical Entrepreneurs’ (see The Launch of the Torch

60

SPIN-OFF ENTERPRISES

Programme, Chapter 9). Most Centres offer institutional support for talented individuals to start their businesses, by providing ‘social-political’ assets and some capital. Chutian Optical Electronics Corporation Ltd. (Case Text 10.2), for example, was created by an individual engineer with the help of the Wuhan Centre. Local administrative authorities were also widely involved in providing capital (usually very limited) and political support for individual initiators. Jinghai and Stone, two of the four famous NTEs of the Beijing Zone (the ‘Two Tongs and Two Hais’ mentioned in Chapter 9) were started in this way.2 The average starting scale of individually initiated NTEs was smaller than the previous group, and their performance results have been mixed: some of them are strikingly excellent, but the number of enterprises of this form among the ‘malpracticing’ NTEs was reported by some Zones as ‘extraordinarily high’.

CASE TEXT 10.2 CHUTIAN OPTICAL ELECTRONICS CORPORATION LTD (CHUTIAN), WUHAN Chutian was founded in 1985 by Mr Sun, now its Chief Manager, who was previously an engineer at the Institute of Optical Technology of Wuhan city. This company was initially incubated in, and a few years later left (or graduated from) the Wuhan Eastlake New-Tech Enterprise Incubator, a ‘Centre’ encouraged by the Torch Programme. The ownership of Chutian was originally ‘collective with the Incubator authority as initiator and supervising unit’ (see Case Text 10.5). It has now been transformed into a Limited Liability Corporation, with the equity shared between the founder, some of the employees and the Incubator. Chutian produces lasers, mainly special laser welding machines, but also other laser products such as surgical lasers. Their technological activities are centred on designing. Their designers combine elements of technologies which are not particularly new—laser devices, electric circuitry, mechanical parts and computeraided controllers—into new machines with the performance desired by their customers. Design is also the forte of the initiator. A close relationship with his previous research institute and with a few local universities is reported to be very helpful in gaining access to related knowledge. The designs are assembled from components which are usually bought or manufactured to order externally. The company also offers after-sales services such as training and maintenance. These products are quite competitive in the domestic market, and have begun to penetrate the international market. Sources: Field Interviews at Wuhan Eastlake Enterprise Incubator and with Chutian Corporation Ltd.; Interview notes 9:

Form 3: Transforming an internal institute department In this form, a part of the organized structure of the initiating R&D institution is licensed as a business unit by a Development Zone for New and High Technology Industries, while at the same time remaining an integrated part of the R&D institution. This form of NTE initiation is taken by the initiating R&D institute as a way of adjusting to the market reform. The spinning-off usually starts with the direct aim of creating profitable businesses based on institutional strengths to supplement institute income. This start in turn paves the way for the wider and more routine commercial exploitation of academic output. Only academically strong R&D institutes are able to make this step while maintaining their academic quality, and this form of spinning-off is not very common. Of a total of 3,000 NTEs in the Beijing Zone, perhaps ten or slightly more seem to have been established in this form.3 Case Text 10.3 provides one example.

INITIATION OF NTES

61

CASE TEXT 10.3 PHYSCIENCE OPTOELECTRONICS CORPORATION OF THE INSTITUTE OF PHYSICS, CHINESE ACADEMY OF SCIENCES, BEIJING The Institute of Physics originated in 1928, as the National Institute of Physics, Academy Sinica, and has long remained among the top institutes in terms of its academic quality in some areas of physics. It now has more than 500 researchers working in condensed material physics, optical physics and atomic and molecular physics. Two National Laboratories, and one Open Laboratory of the CAS, operate within the Institute. The Institute is identified in the reform policy as a ‘basic research’ institute, enjoying a ‘full budget’ from the state. The meaning of this funding was clarified in 1986, when a fixed amount was set as the budget. This budget has since suffered significantly from inflation, and has been insufficient to cover routine operational needs (for general research requirements and staff salaries) since 1992. Funding for project research is now granted by the National Science Foundation and other such foundations, on a competitive basis depending on excellence. Within the Institute there is a ‘Department of Development and Applications’, which is also registered in the Beijing Zone as an NTE, the ‘Physcience Optoelectronics Corporation’, in order to enjoy the benefits of that status. This Department is responsible for commercializing research output which appears suitable to be produced in small batches and has a high value added, such as a molecular beam epitaxy system, crystals and related devices. The Department therefore serves as a window to transform some of the Institute’s output into products. It also returns profits to the Institute, which have been used mainly to increase the average bonuses for the Institute staff faster than inflation. This has proved to be very important for the stability of the Institute. To keep the Department united with the Institute, the accounting office of the Department is controlled by the Institute. And to maintain certain incentives for the Department, there is a reward system for individuals whose bonuses may be much higher than average, according to their direct contributions to the Department’s profitability. From this, a sort of ‘symbiosis’ has emerged between the two parts of the Institute: on the one hand, the small batch fabrication in the Department needs the expertise and installations of the Institute; on the other hand, the researchers of the Institute like to have a base for the commercial development of their achievements. Some researchers have worked in the Department for a time and then returned to their laboratories. In other cases, a researcher would work on the two activities simultaneously. The Institute has also served as a base for form 1 and 2 spin-offs. One research laboratory of the Institute, with its researchers, moved out and became a main power behind San Huan New Material R&D Incorporation, an influential NTE in the Beijing Zone engaged in the commercial development of permanent magnets. This is an example of form 1 spinning. A group of researchers left and initiated the Beijing Kehai High-Technology Corporation, another powerful NTE in the Beijing Zone. This is a case of form 2 spinning. Professor Chen Chunxian, the founder of the first NTE as mentioned in Chapter 9, was previously a researcher at this Institute. Sources: Interviews; Institute of Physics (edited and published): Institute of Physics, Chinese Academy of Sciences, 1990–1991; Wang Xiaolong (ed.) 1993.

Mixed forms The boundaries between these three forms are actually ambiguous. Ambiguity is more likely to be found between forms 1 and 3, reflecting the fact that many R&D institutes are in the midst of experiments to restructure themselves. This can be seen in the Open Software System Corporation, described in Case Text 10.4. A similar situation may also be observed in the Development Centre for Seawater Desalination and Water Treatment Technology, and in Zhejiang University, which we interviewed in the field survey (see the list of R&D institutions visited, in the Appendix to the bibliography).

62

SPIN-OFF ENTERPRISES

CASE TEXT 10.4 OPEN SOFTWARE SYSTEM CORPORATION LTD OF NORTHEAST UNIVERSITY, SHENYANG This is an NTE licensed by the Shenyang Zone. It is also the Research and Development Centre of Computer Software (CSC) of Northeast University. The same group of professionals has initiated a joint venture with a Japanese company, called the Shenyang Northeast University Alpine Software Institute Ltd (NAS). The three titles correspond to three orientations of one entity: the CSC focuses on research and teaching (since 1989), Open Software (since April 1991) undertakes commercial activities as an NTE in the Zone, and the NAS (since July 1991) exports software to Japan. This experiment is inspired and directed by Prof. Liu Jiren, General Manager of Open Software System, Director of CSC, and Co-Director of NAS, together with his young colleagues with computer expertise (average age, 27) ‘to manage, in parallel and mutually beneficially, the commercial development of computer software, research and training’. Open Software produces common (application) software which is competitive in the domestic market. The purpose of establishing a joint software venture with the Japanese is to gain closer access to Japanese experience in management, quality assurance and working discipline. The Japanese side also provided capital, and serves as the first overseas user of the products of the venture. The importance of the introduction of Japanese quality assurance may be indicated if one considers that, before the reform, computer training and software production in China was never oriented to commercial applications. Professor Liu and his qualified team (some 20 to 30 of the 80 in the team have been trained abroad) are actually re-casting some features of the sector in a changed environment. At the moment it is hard to know whether the disparate targets can be achieved harmoniously within the present organizational form. From the point of view of the University, the Centre is one unseparated part of the University but, in contrast to the Department of Development and Application at the Institute of Physics, this part of the University has been delegated full autonomy in its financial and other business affairs. This may thus be seen as an example standing somewhere between spin-off forms 1 and 3. Many Zone managers recognize this as form 1. Sources: Interviews at Open Software; NEU-ALPINE: An Introduction of Software Products, 1993; Nanhu Development Zone Newspaper, 1 July and 1 August 1993; China Electronics Newspaper, 4 June 1993; Far Eastern Economic Review, 23 September 1993.

The initiators of NTEs This section examines the initiators of spin-off enterprises in relation to the officially recorded ownership of NTEs. The relationship between the forms of spinning-off outlined above and the recorded ownership of the NTEs that received the assets is interesting, because it is indicative of the initiators’ sophistication in managing the establishment of NTEs under a rather traditional socio-political framework. The managements of the Zones recognize three kinds of ownership in licensing an NTE: public, collective, or private. Public ownership is assigned to an NTE whose initial capital is publicly owned. The public capital may come directly from the state budget allocated to the initiating institute, or it may be earned by the institute itself during the period of market reform. In the latter case the investment can be defined as public because it is possessed by a publicly owned institute. ‘Collective’ ownership is assigned to an NTE whose initial capital was thought to be collective in nature. An investment made by a group of individuals was often designated as collective rather than private. A ‘public’ R&D institute might also generate a ‘collective’ NTE if its initial investment was (or was regarded as) coming from the institute’s own, and thus ‘collective’, market efforts.

INITIATION OF NTES

63

These definitions of ownership are vague and arbitrary. In addition to ownership, another important indicator of the nature of an NTE is the body designated as its ‘initiator and supervising unit’ (zhuban danwei). An ‘initiator and supervising unity’ is officially identified in licensing an NTE. The term is specific to a traditional planning system, in which people were organized in a vertical system and higher levels of organization were assigned greater powers as well as heavier responsibilities for activities at lower levels. The fact that an ‘initiator and supervising unit’ was named in the official data on NTEs is an aid in discerning the true owner of an NTE. In the case of the ‘collectively owned’ NTEs, which is the largest ownership category, it is possible to distinguish between institutional initiators (especially R&D institutes, but including other bodies) and individual initiators. This gives five modified categories of ownership which are applied in our survey: 1) public ownership; 2) collective ownership with an R&D institute as initiator and supervising unit; 3) collective ownership with an organization other than an R&D institute as initiator and supervising unit; 4) collective ownership without any organizational initiator and supervising unit; and 5) private ownership. R&D institutions as initiators R&D institutions have been the most important initiators of NTEs. The ownership of their NTEs falls in categories 1 and 2 above, which correspond roughly to forms 1 and 3 of spinning-off. Our 1993 survey indicates that 50 per cent, 40 per cent, 30 per cent and 30 per cent of the NTEs present at that time in Beijing, Wuhan, Shenyang and Hangzhou Zones, respectively, should be considered as directly initiated by R&D institutions. A study based on data from 1989 and 1990 indicates that NTEs initiated by R&D institutions accounted for 48 per cent and 52 per cent, respectively, of the total number of NTEs in all Zones in these two years (Chen Zhaoying et al. (eds) 1992:154– 155). Thus, at a rough estimate, more than a thousand NTEs had been spun off in this way by 1992–1993. An attempt has been made to further differentiate the contributions to the creation of NTEs by various groups of R&D institutions. Using the classification of R&D institutions as 1) institutes of the CAS, 2) R&D institutes belonging to central ministries, 3) R&D institutes belonging to local governments, and 4) R&D institutes of higher education. Data for 1990 indicates that these four categories were responsible for the initiation of approximately 12 per cent, 12 per cent, 17 per cent and 11 per cent of all the NTEs at that time. If we compare these figures with the numbers of institutions in each category (see the Appendix to Part 1), it would appear that the institutes of the CAS and R&D units in higher education are more vigorous than other categories of R&D institutes. In addition, information provided by the Beijing Zone shows that, in the central ministry category, R&D institutes in the electronic, chemicals, aeronautical, and posts and telecommunications industries were more active in initiating NTEs (Interview notes 11:2). R&D institutions contributed to transmitting a critical part of the technology-related assets which constituted the core of the technological capability of these NTEs. However, many observers have overlooked other functions of the R&D institutes. One was the provision of venture capital to the NTEs they initiated,4 and the other was to serve to some extent as a regulator or monitor of the management of the initiated NTE. These two functions are linked with the implications of being an ‘initiator and supervising unit’, and are similar to, but much less extensive than, the functions of the Zones and Centres (see the section on the role of local governments below).

64

SPIN-OFF ENTERPRISES

Existing enterprises as initiators Existing enterprises were also involved in the initiation of NTEs. These are either enterprises which existed in the area prior to the establishment of a Zone and were suitable to be licensed as an NTE, or newlyestablished subsidiaries or reorganized versions of existing enterprises. According to the published data for 1990, about 22 per cent of existing NTEs were initiated by existing enterprises. Our field survey shows that in 1993 5 per cent, 15 per cent, 25 per cent and approximately 33 per cent of the NTEs in Beijing, Wuhan, Shenyang and Hangzhou Zones, respectively, were initiated in this way. In some regions such as Shenyang, the local government encouraged existing enterprises to create their ‘corners’ in the Zone area to reinvigorate state-owned industrial enterprises (See Case Text 11.3). However initiation by enterprises is not particularly relevant to the focus of this study. Individual initiators and foreign initiators Individuals who have a certain degree of scientific and technological expertise are encouraged to initiate NTEs by the policy which stipulates that the chief manager of an NTE should have such qualifications (see Chapter 9: Torch Programme). However there are no statistics relevant to this type of initiation. They are identified by looking for two categories of ‘collective’ NTEs: those with an organization other than an R&D institute as initiator and supervising unit, and those without any organizational initiator and supervising unit (ownership categories 3 and 4 above). Given that in 1993 almost no NTEs, except those involving foreign capital, were recorded as having a private ownership, these two categories should provide a good proxy for the number of individually initiated NTEs at that time. The reliability of the method was agreed by the managers of the Zones which we visited, who added more details to confirm it. The individual initiations of NTEs examined here correspond to form 2 spinning-off identified above. Having ensured that the Zone managers had a clear definition of true individual initiation in mind, they were asked for estimates and comments.5 In brief, it can be said that 30 per cent of the NTEs in the Wuhan and Shenyang Zones, and 20 per cent of those in the Beijing Zone, were directly initiated by individual scientific and technical professionals in 1993, when the field survey was conducted. The proportion initiated by individuals in Hangzhou Zone was difficult to estimate, but should lie somewhere between 5 per cent and 30 per cent. If these four Zones were typical, we can estimate that individual initiations accounted for about or slightly above 20 per cent of all NTEs at the time, and there is evidence that the proportion is increasing. Of this 20 per cent, almost half were collectively owned with an organization other than an R&D institute as initiator and supervising unit (category 3 ownership). This figure can be derived from the fact that category 3 NTEs accounted for 6 per cent, 7.5 per cent and 10 per cent of the total NTEs in all Zones for 1989, 1990 and 1991, respectively (Science and Technology Industries of China, No. 10, 1992:13). The organizations involved in these individual initiations were various local agencies, including the Service Centres for Scientific and Technical Entrepreneurs which were set up in a number of Zones and the Development Zones for New Technology Industries, as well as regional (district and town) administrations. Their involvement demonstrates the role played by local administrations, which will be considered below. Foreign initiators were not very active in initiating NTEs (mostly joint ventures) in the Zones until about 1993. Foreign capital was involved in between 5 per cent and 15 per cent of NTEs, with the Beijing Zone having the highest proportion (15 per cent). It was often the Chinese NTEs which had developed strong special assets and adopted an internationally-oriented strategy which were most vigorous in seeking ventures with foreign counterparts, to gain access to components and other complementary technology as well as management skills (Case Texts 10.1 and 10.4; Interview notes 11: 10). Evidence in late 1993 revealed that some of the biggest transnationals in computer and information technology were actively

INITIATION OF NTES

65

seeking a place in China’s Zones. For example, DEC had decided to set up joint ventures with NewTech of Beijing University and others (People’s Daily, Overseas Edition, 25 November 1993), Motorola was creating a Science and Technology Centre in Beijing Zone (People’s Daily, Overseas Edition, 20 November 1993), and AST of America developed a computer model with Chinese writing input jointly with an NTE— Xin Tiandi in Beijing Zone (People’s Daily, Overseas Edition, 9 December 1993). It would be very interesting to extend the survey of the role of foreign involvement in NTEs to spin-off restructuring after 1993, but this is beyond the scope of this study, which was conducted in 1993–1994. The spin-off approach as it existed in 1993 originated from an internal need for structural transformation, and would have been necessary even without the participation of foreign capital. The role of local governments As has been shown in the above, local agencies such as the Service Centres for Scientific and Technical Entrepreneurs and the Development Zones for New Technology Industries, directly helped a not insignificant proportion of individually initiated NTEs. However, it is necessary to view the role of local governments in a broader context. The most important contribution by local governments has been the widespread establishment of the Zones and the Centres. These have served as special regulatory institutions which license and monitor NTEs, and as supporting instruments which implement the preferential policies for NTEs stipulated by the Torch Programme. Moreover, the Centres and Zones have assisted spin-off restructuring by providing every other missing mechanism necessary for the initiation and development of NTEs. They can be regarded as an interface between the new business undertakings and the existing socioeconomic framework, in which some of the facilities which NTEs require are still absent or unsatisfactory, since China is undergoing a painful transition from the old socio-economic framework. Case Text 10.5 illustrates the entrepreneurship of a city government in the establishment of a Centre, and the Centre’s role in supporting the individual initiation of NTEs.

CASE TEXT 10.5 THE ESTABLISHMENT AND ROLES OF THE WUHAN EASTLAKE NEW-TECH ENTERPRISE INCUBATOR The Wuhan Incubator, the first of its kind in China, was founded on 7 June 1987, before the launch of the Torch Programme. Oral reports indicate that the local (Wuhan city) Science and Technology Commission (STC) first raised the idea as early as 1985. Two events helped in its implementation. One was support from the State Science and Technology Commission (SSTC). This support was unequivocal by the end of 1986 and gave the Wuhan government confidence. Another was a study of the feasibility of ‘incubators’, which was jointly sponsored by the Science and Technology Foundation, the United Nations, and the SSTC, starting in 1987. The study produced a series of feasibility reports and probably had a vital influence on the dissemination of the incubator concept in China. What the Wuhan initiators perceived was a dissatisfaction with the performance of the ‘technology market’. According to one author, ‘at that time (1985–1986), though the technology market had developed to a certain scale, the disconnection of scientific research from production was still very serious’ (Yan Zulin 1991). During the same period, individual scientists and engineers started to leave their R&D institutes and universities, initiating their own businesses. The incubator idea was thus intended to help to nurture enterprises initiated by individual scholar-businessmen, transforming their technological knowledge and skills into products. There were about 120 enterprises in the Incubator by the summer of 1993, of which more than half had been founded since 1992. They were selected by the Incubator from about 600 applicants. Since the establishment of

66

SPIN-OFF ENTERPRISES

the Incubator, three enterprises have been closed because of bad marketing prospects, or because of managerial malpractice and five enterprises have matured and moved out of the Incubator. About half of the employees working in the Incubator came from R&D institutes, universities and colleges, and the rest came from enterprises or had been unemployed. The products and services developed in the Incubator were based mainly on personal professional experience and knowledge gained in previous employment, in the form of technological know-how and skills. Avoiding intellectual property conflicts with related universities and institutes has been one of the main concerns of the Incubator managers. The roles of the Incubator were introduced as follows:

1

Help with initiation: Apart from administrative support during the initiation procedure, the Incubator authority plays an important role as ‘initiator and supervising unit’ for most of the incubated enterprises. This is recorded in the official licence. The Incubator authority thereby accepts, explicitly or implicitly, certain liabilities for the incubated enterprises: i ii iii

2 3

as guarantor for bank loans, contractual obligations, etc. made by the enterprise; as the ‘supplier’ of their non-private ownership nature; and as a branch of the public body, they could protect the seniority and privileges (such as housing and insurance) accumulated by the scientific and technical staff during their careers in state-owned R&D institutions.

Providing physical space and utilities. Financing—help in gaining access to several sources of finance: i ii iii

loans from local banks, with the Incubator as guarantor; loan quota under the Torch Programme; venture capital. A venture capital company has just been created by the Bureau of Finance of the Wuhan Municipal Government, implying that more funds will be available, but no details were provided by the Incubator authority. However it was reported that a significant part of the funds needed by the incubated enterprises was still collected from the personal friends and relatives of individual initiators.

Sources: Interview notes 9:1–4; Dong Guilan and Peng Ying 1992: 153–164; Yan Zulin 1991:128–129, 134.

Case Text 10.6 offers a further illustration of how a Zone administration can assist the development of spin-off enterprises by intermediating between them and financial sources. It is interesting to note that the Zone administrations serve something of a dual function in this respect: they function as expanded components of the existing planning system where the remaining part of that system is working, and they also have a function in providing technical and market advice to banks which lack the necessary ability because they have for so long operated as instruments of central planning. Banks are intended to be run more as investing agencies now, sometimes under the guidance of priorities set by the government.

CASE TEXT 10.6 THE INTERFACE FUNCTIONS OF ZONE ADMINISTRATIONS IN SECURING INVESTMENT IN NTES, ILLUSTRATED BY THE HANGZHOU ZONE

INITIATION OF NTES

67

The Zone administration links three actors together: enterprises, the economic planning system and sources of finance (the banks). One aspect of the interface role is to help firms find finance. In some cases this may be via the central planning process, as illustrated by the case of the Semiconductor Research Institute of Zhejiang University. The Institute planned to scale up its high-purity silicon crystal production, and the Zone administration helped to prepare the project proposal to be presented to the State Planning Commission. After the project was approved, the Zone administration helped to secure a bank loan. The Zone manager explained that: ‘The Government has the funds, but it does not have the knowledge about projects like this. We can provide them with that information.’ In other cases, the link to finance operates more directly, via the Zone which can itself provide authorization to secure bank loans for projects, up to a certain ceiling. This ‘approval’ would probably be redundant in the absence of a central planning bureaucracy: the process is necessary because the banks are not permitted to make loans without such authorization. In this connection, the Zone administration may be seen, partly, as a new expansion of the planning bureaucracy. The Zone also provides technical and market advice to the banks about the projects submitted by enterprises. This is very similar to the technical appraisal work carried out by investment banks in the context of free capital markets. The main difference here is that, in the Chinese system, this takes place almost entirely outside the banks, which have only a very narrow financial and accounting function in the planned system. Source: Interview notes 1:2, 3.

Decentralization and the emergence of NTEs The analyses of the forms of spinning-off by which technological assets were channelled into NTEs and of the initiators of NTEs and their roles have revealed that institutions of the old centrally planned system, including both R&D institutions and local governments, have been participating widely in the spin-off restructuring. A new kind of economic organization, the NTE, has been evolving as a result. But many questions remain unaddressed. My Western colleagues ask, how could it happen so massively and pervasively? Who decided on the transfer of assets? Was the state still involved in the process of assets transfer? How was responsibility and liability for the transferred assets shared between the institutes and spin-offs? What incentives motivated the many actors? And, how was the ownership of the transferred assets identified and reinforced? In the first place we attempt to give tentative and partial explanations to these questions, by referring to the decentralization of decision-making in the Chinese system, which has been one of the major approaches to implementing reform programmes. Qian and Xu (1993) have explained the emergence of rural industry as the non-state sector, in the context of decentralization, and considered that it has been the engine behind the rapid economic growth in China since the 1980s. They contend that a ‘Chinese version’ of central planning had been developed before the 1970s, resembling an ‘M-form’ corporate structure, whereas the versions developed in Eastern European countries and the former Soviet Union were more comparable to ‘U-form’ structures.6 Since the 1980s, the market reform has deepened the ‘M-form’ structure in China and transformed it to such an extent that the bottom-level bureaucrats in many cases act like entrepreneurs. Just like rural industry, the spin-off enterprises are non-state owned, and have emerged from this decentralized structure, but they have often been neglected by analysts. The spin-off enterprises are unique in having intimate access to the best-educated manpower, and they have contributed considerably to upgrading the technological capabilities of the Chinese economy. Both rural industry and the NTEs have been driving forces behind China’s economic growth and economic reform since the 1980s.

68

SPIN-OFF ENTERPRISES

Decentralization and asset channelling The re-allocation of R&D institute assets was realized in a ‘two-step’ process. Firstly, the decentralization resulted in a partial commitment of state-owned institute assets to individual R&D institutes. This was legitimated by the 1985 Decision. The legitimization then made it legally appropriate to contract out institute assets, subject to the R&D institutes being monitored by their ‘principal’. Having been created from the autonomy delegated to state R&D institutions, the NTEs acquired even larger autonomy in decisionmaking. They had much looser, or very little, direct links with the state. Likewise, when local governments received greater autonomy to make economic decisions for regional development, and to mobilize capital for the purpose, they were able to set up Zones and Centres. Rural industries were initiated in a similar way, with local governmental authorities playing central starting-up roles and the firms being partly owned and run by their managers and staff. This was achieved in the context of administrative decentralization.7 Incentives Greater incentives resulted from the ‘M-form’ structure because of the deeper delegation of both responsibility and rewards. The incentives to local governments were directed towards achieving better performance in local development. This led them to participate actively in the Torch Programme. The incentives to academics were induced by the changed rules which confirm the market value of technological assets. They were then led to exploit their special assets through various means including the establishment of NTEs. Rivalry between cities and between academics to achieve better performance in the implementation of the Torch Programme reinforced the incentives, helped to expose and resolve the uncertainties which are inevitable in the course of radical institutional change, and accelerated learning from each other. Indeed with the opening to international contacts and the high powered incentives, many managers of NTEs learned impressively quickly to go to the international market, as shown in Case Texts 10.1 and 10.4. The Zones likewise competed to develop ‘internationally compatible norms’. This encouraged the managers of Zones, many of whom came from government branches, to refrain from their accustomed direct intervention in their current managerial responsibility. Financing the commercialization of technology in NTEs The deeply decentralized structure provided a framework for financing the initiation and expansion of NTEs. Three kinds of actors; R&D institutions, banks and Zones were motivated and enabled to do so. The R&D institutions provided venture capital for the initiation of NTEs, the banks provided funds for expansion, and the Zones (from local governments) contributed mainly to infrastructure. To a lesser extent, Centres also provided venture capital to NTEs initiated by individuals (Case Text 10.5). The differentiated functions of these actors stemmed from their special positions. The immediate knowledge which the R&D institutions had of the technology and the managers of the initiated NTE made them better suited than others to manage the risks and rewards that venture capital involves, although there is a government financing agency known as the Venture Investment Corporation. Local banks were very active in each of the booming Zones,8 since under the decentralized structure, the banks operation was strongly influenced by local government policies. The nation-wide Torch Programme merely provided ‘leading funds’.9

INITIATION OF NTES

69

Property rights and hardened budget constraints Technological assets and capital investment were shifted on a massive scale to NTEs, despite an ambiguous definition of property rights and weak property right enforcement through legislation and the courts. The efficiency of the investment proved higher than it had been in the past. The commitment to decentralization can be attributed to the hardening of budget constraints. Investors who had been delegated more responsibility monitored the efficiency of their investments more closely. NTEs which incurred losses were often re-structured by their organizational initiators. On the other hand, some analyses show that the individually initiated NTEs did not demonstrate a clearly superior performance in the early years (Beijing Zone 1992:23–24; Lin Chenhui 1992). One important reason was that organizationally initiated NTEs were monitored by external initiators, whereas at that time no bodies existed to monitor and discipline individually initiated enterprises. The ambiguity of property rights did not cause any serious difficulty at first, but did prove to be problematic some time after start-up. One problem was how to return profits to the initiating institutes when the spin-offs developed successfully. Another problem related to liability where a spin-off failed. Moreover, when it was necessary to re-structure an NTE’s assets this had to be realized through administrative means, which proved very costly. There is strong evidence that these problems have been widely recognized to be harmful for the further development of NTEs. The principle of ‘corporation organization’ to normalize the internal management of NTEs and their relations to investors has been forcefully suggested.10 Recognizing the problems may provide a good starting-point for efforts to solve them, as has often been the case during the restructuring of the R&D system in the past. However it remains to be seen whether, and when, real progress will be made on the property rights issue.

11 THE TECHNOLOGICAL ACTIVITIES OF NTEs

A substantial part of the study is devoted to identifying the technological fields in which the NTEs had developed their advantages and their characteristics in the advantageous areas. The simple findings from the study are that the NTEs were overwhelmingly concentrated on applications of computer and information technology, and that within the broader area of computer and information technology, the strengths of the NTEs lie in a few classes of products and services. These will be discussed in the first sections of this chapter. The next section of this chapter outlines the characteristics of the NTEs in their contributions to applications of computer and information technology. These characteristics include ‘small systems development’, which was the major outcome of the technological efforts at NTEs, and ‘user capability building’ which was the key orientation of the systems development. The chapter also touches on the learning process which was involved in the development of application systems, as well as the impact of the personal computer revolution on the spin-off restructuring approach. A broad picture of technological activities To begin with, it is necessary to devise a categorization of these activities. The SSTC stipulates the ten technological areas to be encouraged (see Chapter 9 note 2), but Zone managers said they used simpler, and in some respects different, categorizations which varied slightly from Zone to Zone. After making an inventory of these categorizations, it was decided to adopt a simple one as the first step of our analysis. This categorization was originally introduced by the managers of Wuhan Centre, with only two groups of technological activities: • group 1: computers and information technology and integrated electronic and mechanical technology; • group 2: biological, medical, fine chemical products and other new materials. This categorization gives two groups with clearly distinctive characteristics. The products of group 1 are machines, machine systems and subsystems such as parts and components, and the products of group 2 are materials of homogeneous composition. Group 1 reflects the strong trend for micro-electronic and mechanical technologies to blend, a tendency which is even more evident in the activities of NTEs since they were devoted to assimilating and adapting, rather than pushing the frontier of the field. Hence it is understandable that the managers of Zones tended not to follow the SSTC’s more elaborate classification. Since group 1 activities will be discussed extensively below, they will also be referred to simply as activities in computer and information technology, or more precisely, in computer, information and related technology.

SPIN-OFF ENTERPRISES

71

The technological activities of NTEs have been allocated to one of these two groups, relying on Zonelevel aggregate data provided by the managers of the four Zones visited. Table 11.1 summarizes the estimates of the distribution, and comments made by the Zone managers. It can be seen that, in all the four Zones, about 60 per cent–70 per cent of the technological activities fall in group 1. Although the data on which the estimates were based is not entirely comparable, there is no indication that there could be an error serious enough to undermine the conclusion that the majority of NTEs were engaged in computer and information technology. Technological activities in detail: computer and information technology Following the broad outline of NTE activities above, this section reports on a Zone-by-Zone survey focusing on computer and information technology. The diversity between Zones was expected to be an informative means of furthering our understanding of NTEs’ contribution to the technology. Once again, we had to develop a common indicator by which comparable search and analysis might be carried out in different Zones. This was found in the concept of ‘competitive products’, used by Zone managements to refer to those products and services that were thought to be market competitive. Many Zones were compiling information on their competitive products, although they were not using a standard classification. By making an inventory of the competitive products from different Zones, and consulting with Zone managers and managers from the relevant industrial ministries, a profile of typical products and services was drawn up as follows: 1 computers; 2 computer parts and peripherals; 3 character and graphic processing technology and apparatus; 4 automatic operation or production systems; 5 industrial control machines; 6 telecommunications equipment; 7 integrated electronic and mechanical devices and machines. Table 11.1 Distribution of technological activities of NTEs: a Zone level summary Zone

Distribution

Group 1

Group 2

Beijing

70%

Wuhan

60%

16%

Comments and references 1

About 50% of NTEs are engaged in electronics and information technology, and these account for about 50% of annual sales; (51.98%)a 2 Integrated optical, electronic and mechanical technology account for more than 20%; (22.84%)a 3 New materials, about 8%; (7.18)a 4 Biological and medical products, about 8%; (9.69%)a Sources: Figures marked ‘a’ are from Beijing Zone 1993. The remaining figures are from Interview notes 11:3–5. 20–30% 1 Micro-electronics & computers and integrated electro-mechanical technology account for about 50% of the number of NTEs and 30% of the profits; 2 Biological & medical products/fine materials, about 20–30%;

72

TECHNOLOGICAL ACTIVITIES OF NTES

Zone

Distribution

Group 1

Group 2

Comments and references 3

Shenyang

60–65%

35%

Hangzhou

80%

20%

Telecommunications, about 10% in terms of the number of NTEs, but 20% in terms of profits. The difference is because a few large enterprises (WRIPT, and Chang-Fei Company, a joint venture with Philips producing optical fibre) are included. Source: Interview notes 9:8. 1 Integrated electro-mechanical technology accounts for 35–40% of gross sales; 2 Micro-electronics and information technology, including computer software, about 25%; 3 Biological products, including refined Chinese medical products, about 25%; 4 New materials, about 10%. Source: Interview notes 12:4–6. 1 Electronics and information technology, 58% of NTEs; 2 Integrated optical-electronic-mechanical technology, 23%; 3 Materials, 10%; 4 Biological and medical, 4%; 5 New Energy and conservation technology, 4%; Source: Interview notes 1:4.

In this profile, the group 1 technological activities have been dis-aggregated into seven smaller categories. The term ‘micro-electronics’ which appears in the categorization by the Torch Programme (see Chapter 9, note 2) has been dropped because at that time there were no technological activities for the development of ‘real’ micro-electronics such as integrated circuits. All seven typical products and services involve applications of micro-electronic technology. Technological activities in Beijing Zone The Beijing Zone was found to be strong in categories 1, 2, 3, 5 and 6. Underlying these is the mastery by the NTEs in Beijing of sophisticated Chinese Character processing technologies. This mastery was the key to the development and success of many competitive products there, as shown by a number of giant NTEs, such as Legend (Case Text 10.1) which produces computer parts and personal computers, New-Tech of Beijing University which produces computer compiling and printing systems, and Stone, which produces Chinese typesetting systems and similar products. Being at the centre of the emerging Chinese market for computer and information technology seems to have had an important impact on the structure of their technological activities. Case Text 11.1 provides details of the typical products and technological specificities in the Beijing Zone.

CASE TEXT 11.1 TECHNOLOGICAL ACTIVITIES IN BEIJING ZONE In 1993 about half of the NTEs (3,400 in total) and half of the annual sales were in the field of computer and information technology. Five typical areas of products and related services were identified: computers; special

SPIN-OFF ENTERPRISES

functional boards and cards; character and graphic processing apparatus; industrial control machines; and telecommunications equipment. Based on the identified typical categories, further discussions with the Director of the Enterprise Development and Management Department of the Zone were required, to ask questions such as ‘Which NTEs are typically engaged in each type of activity?’, ‘What are the characteristics of technological activities in a certain category by representative NTEs?’, ‘How were the representative NTEs initiated?’ and ‘How did they gain the critical technological capability?’ Behind these questions, the fundamental question was: ‘What makes the Beijing Zone so distinctive?’ These discussions generated the findings summarized in Table 11.2. Table 11.2 Typical products of the Beijing Zone Category of product

Typical products

Typical NTE(s)

Computers

Personal computers (PCs), mini computers, work stations;

Special functional boards and cards Character and graphic processing apparatus

—

Legend for personal computers; Taiji for minicomputers and work stations; —

Industrial control machines Telecommunication equipment

STD 5000 series

New-Tech of Beijing Univ. for compiling and publishing systems; Stone for typesetting equipment; Kangto

—

—

Office automation, TV and teaching image, scientific graphic, and publishing systems;

Computers The techniques for the Chinese character processing which were more available in the Beijing Zone, and the ability to integrate them into various brands of computers, was the main strength for NTEs in the Beijing Zone to have computers as one area with certain competitiveness. Initially, the Chinese character processing technologies were embodied in cards or boards which were then to be added to imported computers. Gradually a small number of NTEs acquired the capability to develop specific IC chips which embody these technologies and used them in their computers. Design capability was the other strength responsible for the development of computers in the Zone. This advantage in design, however, has been limited by the lack of advanced equipment used for the purpose. Design work, in around 1993, was said to focus largely on simplifying structures to reduce costs to suit the needs of domestic users. Of the producers of PCs, the NTE selected as a typical example was Legend (Case Text 10.1). Legend was spun off from the Institute of Computer Technology of the Chinese Academy of Sciences. In 1992 its sales of PCs on the domestic market had surpassed the biggest state company, ‘Great Wall’, making it the number one domestic producer of PCs. In the international market, Legend sold computer components (as distinct from the complete PCs sold by Legend on the domestic market). The international competitiveness of Legend in computer parts stemmed from the sophisticated Legend Chinese character processing systems, which had been combined into some functional PC subsystems such as graphic displays, printers and RAM expansion. These subsystems had been adopted by a number of computer producers in the world.

73

74

TECHNOLOGICAL ACTIVITIES OF NTES

The domestic market for personal computers and work stations was dominated by Taiji Computer Corporation and the China Computer System Engineering Corporation. The former is also the No. 15 Institute of the Ministry of the Electronics Industry, and the latter is the No. 6 Institute of the same Ministry. Taiji used DEC technology to produce VAX systems, while Computer System Engineering had introduced technology and equipment from Sun Micro. The state provided support for these imports. Special functional boards and cards The boards and cards in the Chinese market served two main functions: Chinese character processing, to be added to computers which had been developed based on the English language; and networking, to be used to connect differing types and series of computers. As has been mentioned, a few of the producers of the boards and cards, including Legend and NewTech of Beijing University, were able to embody their technology into IC chips which brought them big competitive advantages. But most of the boards and cards at that time were still based on printed circuits. A great variety of networking cards were developed because almost every type of computer made in the world came into China from the early 1980s. Because of a tight financial situation, Chinese users tended to keep their old machines in operation when they obtained new ones, and networking was increasingly in demand. The Hong Kong brokers who sold most of these computers did not provide adequate services, and services such as network building became the job of NTEs, many of whom started their business by selling and maintaining imported computers. By 1985 special functional boards and cards comprised the biggest segment of the computer market in the Beijing Zone, a phenomenon perhaps never seen elsewhere in the world. Character and graphic processing apparatus It was commented that the basic hardware technologies for the character and graphic processing apparatus were imported. The NTEs contributed to this category through 1) adding-on Chinese character processing systems, which had to be continually up-dated to keep pace with the international market and 2) the development of specific applications software for local users. The number and variety of users was increasing, including government offices, shopping centres, banks, video and broadcasting agencies, industrial firms and scientific information networks. Stone was one of the competitive producers in this field. It was founded in 1984 by a group of engineers in computer engineering from the Chinese Academy of Sciences. The key products which the Company developed were their 4S Typesetting Systems, which captured 80 per cent of the domestic market and had replaced traditional manual typesetting of Chinese characters. New-Tech of Beijing University, founded in 1986 by Beijing University, was another example of the competitive producers. New-Tech developed their brand, ‘Computer Compiling and Printing Systems’, which had resulted in revolutionary advances of the Chinese publication industry. The market for these systems was expanding to neighbouring Asian countries where Chinese is an officially or popularly used language. Industrial control machines The technological efforts in relation to industrial control machines were mainly for import substitution. Some standard control machines, such as the STD series, had entered large-scale production to replace imported products. Kangto, an NTE spun-off from the No. 502 Institute of the Ministry of Aerospace and the Aeronautical Industry, was the leading producer in this area. Telecommunications equipment According to the Zone manager there was a heavy reliance on imported hardware technology in telecommunications equipment. However the advantages of NTEs in the use of Chinese character processing techniques and other adaptations were manifest in some products, such as an ‘auto-switchover’ device used for replacing the operators who have to work in China’s many small exchanges from morning till night. Another feature was the small size of the equipment developed. The programmed telephone exchange systems, one of the products in this category, were applied in exchanges for less than 1,000 lines. Summary of comparative advantages The survey led to some ideas regarding the comparative advantages of NTEs in the Beijing Zone in general. Compared with state enterprises, the NTE managements were recognized to be more sensitive to information

SPIN-OFF ENTERPRISES

75

regarding both market demands and technological change. The NTEs were more responsive in adopting and adapting foreign technologies, and they were more active in marketing their products and services as well. Being close to specific local demands, and the complexity of the Chinese language in particular, were seen to be the major sources for the NTEs in developing their advantages compared with foreign companies. The uniquely complicated structure of Chinese characters had challenged the Chinese developers to improve their capabilities in graphic information processing for instance, which constituted largely the strengths taken by both Legend and the New-Tech of Beijing University. Sources: Enterprise Development and Management Dept. (Beijing) 1992, 1993; Beijing Zone 1993; Wang Xiaolong (ed.) 1993; Interviews and Interview notes 11:6–15.

Wuhan Incubator and Zone In Wuhan Incubator and Wuhan Zone, the technological activities were considerably different from those in Beijing. NTEs in the Wuhan Zone and Incubator were more concentrated on types 4 and 7 above, i.e., the development of automatic operation or production systems and of integrated electronic devices and machines. The latter did appear in Beijing Zone, but was much less important in related terms. The development of automatic operation or production systems did not appear at all in the list of competitive products compiled by the Beijing Zone. Although both type 4 and 7 activities relate to applications of computer technology, the emphases differ in some respects. The former concentrates on systems development, involving modelling operation processes, compiling the systems, and installing the systems in the work place. The latter concentrates on the necessary technological work of designing and assembling the devices or machines developed. Chutian Optical Electronics Corporation Ltd., introduced in Case Text 10.2, is one example of an NTE with type 7 activities. It was reported that the NTEs in Wuhan who were regarded as engaging in these two fields, and especially those with type 4 activities, were often also dealers in personal computers. The same observation was made in the Beijing Zone: many of the NTEs who were identified as producers of computers, functional boards and cards or character and graphic processing technology, such as Legend and Beijing Xinghe Electronic Co., were also dealers in personal computers and developers of small automatic systems for particular users. Once these companies had specialized in a certain area, other activities such as selling computers or small systems development were no longer recorded among their major activities.

CASE TEXT 11.2 TECHNOLOGICAL ACTIVITIES IN WUHAN INCUBATOR AND WUHAN ZONE In the Wuhan Incubator A picture of the technological activities in the Wuhan Incubator was derived from descriptions by the Incubator managers in response to our concrete questions such as what businesses were undertaken most commonly in the NTEs, their principal clients, the usual requirements of these clients and how the NTEs meet them, what outputs were generated from the businesses and how the NTEs deliver them, and so forth. The managers were also asked for specific cases when their initial answers were limited to more general statements. Because there was no documented material available, lengthy interviews were required to get answers corresponding to the classifications and concepts developed for the purpose of the study. Eventually a common understanding was reached, that the typical areas for the activities of the NTEs in the Wuhan Incubator were the development of automatic operation or production systems, and of integrated electronic devices and machines (type 4 and 7 activities).

76

TECHNOLOGICAL ACTIVITIES OF NTES

Many of the NTEs were reported to be engaged in the development of small automatic operation or production systems. Their users came from a range of sectors, including medicines, ceramics, textiles, tobacco, food processing, oil production, energy generation and experimental operation in scientific laboratories. The NTEs themselves concentrated on systems development, mostly in software technology, i.e., modelling the processes for the operations, then computerizing them to operate automatically. The hardware needed for the systems, such as sensors and industrial control devices, was procured in the marketplace. Contracts for these systems would only be considered complete when the systems had been installed in the work places, so installation services and debugging after installation were certainly included. The technology employed seemed to be unsophisticated, both in terms of the hardware involved and the automatic systems which were centrally developed. Another type of technological activity in the Incubator was the development of integrated electronic devices and machines. Products included sensors, numerical indicators used in numerically controlled machinery, laser welders, and instruments for the telecommunications and electricity nets. Design, and to a less extent manufacturing, were the key tasks. This type of activity did not involve a high degree of technological sophistication either. It was reported that these two types of activity were often intertwined with each other. An enterprise whose main business was the development of small automatic systems might also produce some related device, and vice versa. In the Wuhan Zone The NTEs in the Wuhan Zone, as in the Wuhan Incubator, had the same two major activities, according to the head of the Enterprise Management Department of the Wuhan Zone. It was said that the NTEs in the Wuhan Zone which engaged in the development of automatic operation systems were serving mainly industrial users. These developments involved the design of automation systems, which would often be constructed with a simple computer and selected sensors. An interface between the automation system they developed and the production process that was to be controlled by the system also had to be designed. What differentiated the Wuhan Zone from the Wuhan Incubator was some 10 per cent of the NTEs in the Zone, producing about 20 per cent of the total profits, worked in the field of telecommunications equipment. The difference between the two figures arises because a few big establishments in the Zone, such as the Wuhan Research Institute for Posts and Telecommunications and the Chang-Fei Company (a joint venture with Philips, producing optical fibre) worked in the field of telecommunications equipment. Source: Interview notes 9:5, 8–9.

Shenyang Zone In the Shenyang Zone, the most important category of technological activity was the development of integrated electronic and mechanical devices and machines (type 7), which accounted for 40 per cent of the gross sales of the Zone’s NTEs. This strong commitment to integrated electro-mechanical apparatus is distinctive for Shenyang, although at that time no NTEs in this Zone were as outstanding in their field as the big Beijing-based NTEs in the area of Chinese character processing technology. The history of the city largely explains the concentration: Shenyang had been, and still is, one of the most important bases of the machinery industry in China. R&D institutes and universities in Shenyang were created to serve the industry. In addition to type 7 activities, computer software, oriented mainly to general applications and led by the Open Software System Corporation, was being pushed as one of the key businesses in the Zone. Automated operation systems were also being developed by many computer dealers. These two activities fall in types 1 and 4 as defined above. Together with other scattered items, they accounted for another 25 per cent of overall sales from the Zone. An interesting point which emerged from the survey of Shenyang Zone was that the

SPIN-OFF ENTERPRISES

77

NTEs were mainly serving the service sectors in their need for automated operation systems, whereas the state was supporting automation in the big and heavy manufacturing sectors. The knowledge acquired from the Shenyang Zone may also shed some light on the development of the computer and information market in China. A unified domestic market for computer and information technology seems to be taking shape, in which the Zones are network hubs, and the newly emerging giant NTEs are becoming trans-regional corporations.

CASE TEXT 11.3 TECHNOLOGICAL ACTIVITY IN SHENYANG ZONE The technological activities in Shenyang Zone were identified from interviews with Zone managers and the managers of selected NTEs. An inventory of the competitive products and services and their producers was derived from these sources. This showed that the first area of technological strength of the Zone was in integrated electronic and mechanical products (type 7). NTEs devoted to this field accounted for 35–40 per cent of the Zone’s gross sales. Other competitive products of the Zone were scattered over several types of activity, of which computer software was noteworthy. Together they accounted for another 25 per cent of gross sales. Integrated electro-mechanical products The integrated electronic-mechanical (type 7) products produced embraced various kinds of power supply devices and transformers, machine tools and parts (saws, numerical controllers for machine tools), some production lines (such as those used in meat packaging), and robots. The producers of almost all of these competitive products had recognizable strong links with an R&D institute or a university, except for one whose origin was unclear. There was one producer of bar-code scanning equipment, a new product for Chinese users with strong potential demand. This was an NTE called the Xianda Bar Code Scanner Technology Corporation Ltd., initiated by the Shenyang Engineering University. The technological sophistication incorporated in these products seemed to vary greatly. Some were obviously advanced, such as the robots produced by an NTE spun off from a famous research institute in the field. Some were rather conventional. These products seemed to be specifically designed and produced in small volume, although there was little information from which their technological characteristics could be discerned. They were sold mainly in domestic markets. It was not surprising that type 7 activities were a typical area for NTEs in Shenyang, given that this city has long been an important base for the machinery industry in China. What was surprising was that the NTEs in the Zone initiated by big enterprises were not equally active. Many such NTEs had been established since the Shenyang Municipal Government sponsored a programme known as ‘One Enterprise, One Corner (in the Zone)’. In response, big existing enterprises established branches in the Zone, as a step to renew themselves. In 1993 about 200 to 300 of the 1,100 NTEs in the Zone were initiated by big enterprises. These branch NTEs usually received their new product designs through technology transfer either from other NTEs, or directly from an R&D institute or a university, under the coordination of local government, according to the Zone managers. Considering the unimpressive performance of these ‘corner branches’, it must be asked whether such one-off transfers of technology can be effective without continuous communications, and how the new institution of Zones can contribute to solving the daunting difficulties encountered by the state enterprises. Computer software and automatic operation systems About 25 per cent of the technological activities were scattered over types 1, 4, 5 and 6, i.e., computers (in this case computer software), automatic operation systems, industrial control machines, and telecommunications equipment. The Open Software System Corporation Ltd., an NTE initiated by the Northeast University, was the dominant producer of computer software (see Case Text 10.4). This NTE, together with its home University and the Zone administration, had ambitions to promote software development as one of the Zone’s key businesses, to make it competitive both domestically and internationally. A Park for computer software, based on the strength achieved by the Open Software System Corp., was under construction at the time. Industrial control machines were produced by the Automation Engineering

78

TECHNOLOGICAL ACTIVITIES OF NTES

Development Company, also initiated by the Northeast University. Another NTE whose origin was unclear was appraised by the Zone as a typical producer of programmed exchange equipment. The three NTEs just mentioned were introduced as typical producers of the Zone’s competitive products of types 1, 5 and 6. Type 4 activities, for the development of automatic operation systems, were not very visible in the published list of ‘competitive products’, but in conversation it was said that this business was undertaken by NTEs which were often licensed as computer retailers. The automatic operation systems were developed to serve users in the service sector, such as government agencies, banks, hotels and restaurants. It is interesting that the banks have changed their position in the supply-demand relationship in this field several times. Initially they relied on outside technicians for their automation programmes. Later they set up internal computer divisions for this purpose. By the time the survey was carried out, some banks were actively initiating an NTE to externalize their capability in computer applications. More than ten NTEs of this kind were created in 1993 in the Shenyang Zone. In Shenyang the users of type 4 products did not include big manufacturers. The Zone managers and managers from the central Ministry of the Machinery Industry both commented that any automatic systems for the needs of large state enterprises were developed mainly through the planning approach, and rely on internal software expertise. The technology was largely imported and the systems were mostly conventional. External support could be drawn on if the job was too sophisticated, mainly through planning coordination. The computer market in Shenyang The computer ‘giants’ with a home base in Beijing Zone, such as Legend and Stone, had established branches in Shenyang Zone. Branches of both Legend and Stone were among the largest 50 NTEs of the Shenyang Zone. Domestic products, including brand computers and boards and cards, were being sold alongside imported computers and peripherals at many computer shops, some of which were sales points of the emerging giants. These observations symbolized the growth of a unified domestic market for computer and information products and services. This contrasts with conventional commodity markets in China, which many observers have noted are becoming more segmented as economic decision-making is delegated to regional authorities (see: Wu Jinglian and Liu Jirei 1991, especially Chapter 7; Shirk 1993). It would be useful to explore why the market for computer and information technology has developed along more healthily lines, if our observation is correct. Sources: Interviews in Shenyang and Beijing; Interview notes 12:4–8; Nanhu Development Zone Newspaper, 6 May, 5 June, 1 July, 1 August 1993; Wang Chengxiang: Speech at Working Meeting of Shenyang Manhu Science and Technology Development Zone, mimeo, 1993.

Summary of typical products and services To sum up, Table 11.3 brings together the main characteristics of NTEs’ technological activities highlighted from the Zone level analysis. The summary is presented according to the classes of ‘typical products and services’, and compares the NTEs with their domestic and foreign counterparts. This leads to a concise picture of NTEs which highlights the areas in which the NTEs had developed specific strengths. We will discuss the characteristics of NTEs’ technological activities in the following section based on this summary. Characteristics of technological activities and user capability building As one can recognize from Table 11.3, NTEs have not developed substantial competitive advantages in all seven classes of products and services. Types 2, 3, 4 and 7 are advantageous areas for NTEs. Some of the most conspicuous characteristics of the technological activities involved in these four types of products and services will be outlined below, under the headings of small systems development, user capability building, the learning process, and the impact of the personal computer revolution.

SPIN-OFF ENTERPRISES

79

Small systems development Small systems development characterizes the contents as well as the complexity of the technological activities in all these areas. Although there was wide Table 11.3 Characteristics of NTE’s technological activities (in the area of computer and information technology) Typical products and services 1

• Domestic production is based on imported technologies (See Case Text 11.1); • Simplified design and advantages in character processing have given NTEs and other domestic producers a small niche in the domestic market (See Case Text 11.1). 2

• NTEs have a dominant position in the development of this kind of product, especially NTEs in Beijing (interview, Ministry of the Electronics Industry); • International competitiveness is developing in character and graphic processing and display boards and cards (Case Text 11.1). 3

• Very significant achievements in compiling, publishing and typesetting Chinese texts (Case Text 11.1); • The NTEs of Beijing Zone have taken the lead, and are penetrating throughout the Chinese market (Case Text 11.3); 4

• User-specific design and services could not be covered by the planning approach

Characteristics of technological activity Computers

• Foreign machines are still dominant (interview, Ministry of Electronics Industry);

Computer parts and peripherals

• Centred on functional boards and cards; special adaptation for Chinese market; stimulated by the peculiarity of Chinese language and users (Case Text 11.1);

Character and graphic processing apparatus

• Some competitive products have been developed, based on advantages in character processing technologies (Case Text 11.1);

Automatic operation or production systems

• Smaller and non-manufacturing users are supplied by NTEs; these users are mainly in the service and light industrial sectors (interview, Ministry of the Machinery Industry, and Case Texts 4.2 and 4.3);

80

TECHNOLOGICAL ACTIVITIES OF NTES

Typical products and services (interview, Ministry of the Electronics Industry, and Case Text 11.2); • Local specificities of the operation and management environment cannot easily be dealt with by foreign suppliers (interview in Beijing Zone). 5

Characteristics of technological activity

Industrial control machines

• NTEs seem to contribute to the assimilation and dissemination of imported technologies for smaller and simpler product series (Case Text 11.1);

• The development of larger industrial control systems is still organized under coordination by the government, from technology imports to the assimilation of the imported technology (interview, Ministry of the Machinery Industry); Typical products and services

Characteristics of technological activity

6

Telecommunications equipment

7

Integrated electronic and mechanical devices and machines

• The NTEs seem to contribute to the development of smaller systems, based heavily on imported technology (Case Text 11.1); • Development in this area covers a variety of products such as specific electric devices, machine tools, robots, analytical and testing instruments, by combining electronics-based information processing and controlling with mechanical technologies (Case Texts 10.2, 10.3, and 10.3);

• NTEs are envisaged as important potential contributors to these smaller but specific devices, machines and instruments (interview with the Manager of the Instrumentation Department, Ministry of the Machinery Industry).

variation between NTEs, they had acquired technological capabilities of sufficient sophistication to sustain their efforts to win competitive positions in these fields in the face of fierce competition from both foreign suppliers and big state enterprises. It is useful to reiterate the features of small systems development, looking at what the small systems are for and what key elements have to be mastered to develop them. The features of the several kinds of small systems to which NTEs had contributed can be summarized as follows: • Localization or ‘Sinologization’ of computer language. This is the development work involved in adapting English-language based computer and information technology for a Chinese language environment. Software development is the key to language localization, which has to be compatible with

SPIN-OFF ENTERPRISES

81

the design of standard computers. But the hardware, such as functional boards and cards to add a Chinese character processing function to ordinary computers is also important. Language localization has led on to the development of character and graphic processing apparatus such as Chinese compiling and publishing systems and Chinese typesetting machines. In addition, language localization was one of the key elements in the development of computer applications software, as shown in the case of Hope Corporation (interview at the Hope Corporation).1 Domestically-produced computers and industrial control machines all incorporated Chinese processing functions. • The development of small automatic operation systems. This development work focuses on incorporating computer and information technology into existing operations in various sectors. Software development is again the key, with hardware often being procured from the market. The technological activities involved centre on modelling a production or operation process, designing of automatic systems for the process, and the engineering services for on-the-spot installation of systems, training and maintenance. The system must be developed to match user-specific conditions, including both language localization and management procedures. Some NTEs were specialized developers, and others were also computer dealers. • The development of small and specific devices and machines. This development work aims at generating hardware outputs based on good design, by incorporating computer and information technology into conventional products such as machine tools, medical equipment, instrumentation, electric and telecommunications equipment. Design is central, but the assembly of final products was also often included in the core area of an NTE engaged in this business. The products produced by the NTEs thus far were smaller, and more functional specific, than those produced by large state enterprises in this sector. The latter were the major producers of larger, heavier, and more conventional varieties. User capability building From the objectives and clients of the three kinds of small systems described above, it is clear that small systems development is oriented principally to building up user capabilities, so that they can make better use of computer and information technology. The localization function focuses on the linguistic modification of computer and information technology in order to bridge the language gap which restricts the technology from being employed in the Chinese-speaking economy. The other functions the systems embody are directed at upgrading various operations and devices. All the systems aim at the widespread applications of the technology, and have their clients which are final users in various sectors. This contrasts with electronic fabrication, which aims to produce the basic technology products which are used as intermediate inputs in various applications. These application-oriented systems have became available in the Chinese market only since the NTEs began to contribute to them. The planning approach seemed not to be able to deal with widespread applications of computer and information technology, because the planned economy essentially favoured quantitative outputs of mass-produced hardware. Under the planning approach, the investment nomenclature was dominated by categories such as ‘plant for mainframe machine manufacturing’ and ‘plant for peripheral equipment manufacturing’ which left little room for applications or services, as a senior manager from the Ministry of the Electronics Industry commented (interview at the Ministry, 1993). NTEs contributions to user capability building were also highly valued, as highlighted by the statement that, ‘It is owing to the NTEs, which have accelerated the development of computer applications, that computer technology has started to become a useful instrument for various sectors, rather than only for scientific calculations as it was until the end of 1970s’ (ibid.).

82

TECHNOLOGICAL ACTIVITIES OF NTES

The learning process The learning process of NTEs is closely connected to the strong orientation of their technological activities towards building user capabilities. One illustration of this close connection is that NTEs were initially prompted to learn to master systems development by selling and providing after-sales services for imported computers. This was around 1984 and 1985, when about 100,000 personal computers were imported into China.2 Technical professionals were called on to offer their skills in the procurement, distribution and user services for the new high-tech products. Computer retailing and after-sales services were the weak links in the long chain which was dealing with computer imports at that time, starting from American manufacturers, through Hong Kong middlemen, to coastal cities or firms which had been granted autonomy in the use of hard currency. The old system for computer distribution was incapable of meeting the new challenge from the personal computer, which was being imported in great quantities. Clustering on ‘Beijing Electronic Street’, the predecessors of the NTEs emerged to take over the distribution function. Spin-off NTEs were thus born together with the emergence of the Chinese personal computer market, as twins. Having started by selling computers, NTEs’ learning continued to be heavily induced by selling as well as by market demands. Selling provided NTEs with information on overseas producers and domestic users, along with opportunities to accumulate capital to invest in learning. The information learned was soon applied in their in-house development. Over a period of several years, the NTEs upgraded themselves from computer dealers to developers of small application systems, as illustrated by the changing composition of their income. In 1984, ‘trading’, ‘products’, and ‘services’ contributed 75 per cent, 4 per cent and 21 per cent, respectively, of the overall turnover of the NTEs in the Beijing Zone (Yu Weidong (ed.) 1988:129). By 1992, these proportions had shifted to 48 per cent, 34 per cent and 18 per cent respectively (Table 9.2, also see notes to Table 9.1 for definitions of the income items). Technological efforts which were oriented to building user capabilities, and which were strongly led by market demand, were sensitive to the specificities of users’ requirements and conditions, as has been fully demonstrated in the previous sections. This must have involved a radical change in the way in which R&D activities were organized in NTEs compared with academic R&D institutes. And this in turn must have meant a dramatic alteration for the working style of the scholar-businessmen who created NTEs. It would be very interesting for further studies to explore how they were able to make this transition so rapidly. Some evidence suggests that they were consciously aware of, and stressed, the need to re-orientate their technological activities (compared with their previous work in R&D institutes), and that this was achieved by intensive study of the technological skills which the market demanded, such as product and engineering design. It would also be interesting to study how their previous experience in academic R&D was (or was not) transferred to learning the skills needed for market profitability. The orientation to building user capabilities is a common theme along a learning sequence, which starts with selling and after-sales services, followed by the emergence of simpler add-in technology developed to meet local needs (such as boards and cards for Chinese character processing). The next stage is building up some core competence which serves as a basis for the development of application systems. This process of learning centres around systems development and strongly emphasises software, which is rather different from the learning process in many other developing countries such as South Korea and Malaysia, where the development of the electronics industry centred on manufacturing. An examination and comparison of such different approaches to the development of computer and information technology would be another interesting theme for further study.

SPIN-OFF ENTERPRISES

83

Impact of the personal computer revolution The personal computer revolution had a significant and continuing impact. In parallel with the worldwide rise in demand for computers to be applied in banks, airports, hotels, publishing, broadcasting, advertising and government agencies, the service sectors were observed to be the most important users for NTEs’ computer and information technology products and services. The ‘open door’ policy, a component of the economic reform programmes in China, made it essential that these basic services of the social infrastructure should conform to international standards. We also observed that the cheapening and availability of microprocessors and other core components had been crucial in enabling the NTEs to concentrate on application systems for local needs without having to develop most of the basic devices themselves. This is what Ernst and O’Connor call the ‘comodification’ of computer hardware, i.e. the standardization of the basic designs of the computer system, which have been transformed into mass-produced and general-purpose black boxes (Ernst and O’Connor 1992:72–75), applicable in various applications. This lowered entry barriers to the point at which organizations and individuals with no exceptional wealth could afford to initiate businesses for applications of the technology.3 The widely-adopted strategy for NTEs was to develop a core technology and obtain its complements externally (Yu Weidong (ed.) 1988:127). This is in accordance with other evidence that there is a trend towards increasing availability and tradability of the basic components of the technology in the marketplace.

12 CONCLUDING REMARKS

This study has established that spin-off restructuring of R&D institutions in China is pervasive, important and practicable. It has shown that the entrepreneurship of R&D institutions is central for the large-scale establishment of New Technology Enterprises in China. The technological assets accumulated in R&D institutions and then channelled into NTEs through the restructuring become an important source of competitive advantages that enables the NTEs to mature quickly in their newly created business areas. The study has identified several forms of spinning-off, that is, ways of channelling the technological accumulation to new enterprises. Spinning-off part of the organized structure of an R&D institution has been shown to be more effectual, in the early stages of the restructuring in China, than the initiation of NTEs by individual technical professionals. The fact that organizational initiators have played a predominant role in initiating NTEs points to one essential aspect of a positive restructuring process. Existing organizations have to and, as demonstrated in the study, can play an important part in the restructuring if the process proceeds with a certain degree of social consensus and participation. R&D institutions and local governments have been very active in the spin-off restructuring in China, not only in transferring accumulated technological assets but also in providing financial, regulatory and physical infrastructure, as well as socio-political support. Since the institutions to supply all kinds of necessary support are missing, the lack has to be supplied by the constituents of the existing structure, temporarily and in more or less ad hoc ways, for radical restructuring inevitably means doing things in unfamiliar ways. Incentives and the delegation of responsibility, as well as regulations to provide new rules for new games, are preconditions to achieve broad participation in restructuring. In the case of NTE development in China, these preconditions were provided by the launch of the Torch Programme and the implementation of that Programme was rooted in a decentralized decisionmaking structure. The results of the wide, and indeed vast, participation in institutional experiments for spinoff restructuring are also visible: the birth of new business establishments (NTEs), the development of various kinds of regulatory institutions which are growing beyond the temporary and ad hoc stage, and the transformation of many of the existing state-owned R&D institutes themselves (which will be examined in Part 3). The study has revealed that the spin-off enterprises devote themselves overwhelmingly to computer and information technology. User capability building has been the principal orientation of technological efforts in the field. This is highlighted by the several kinds of application system to which the NTEs provide a most important contribution: 1 systems for localization or sinologization of English-language based computer and information technology;

CONCLUDING REMARKS

85

2 small automatic operation systems for information processing and control, used in various processes in the service and production sectors; and 3 small and specific devices and machines incorporating computer and information technology. These achievements of the spin-off enterprises in China may point to a different approach to the development of computer and information industry, an approach which is open to latecomers. This approach is based on software-intensive engineering, aiming at building up user capabilities so that local users can make better use of the technology for local needs. This contrasts with achievements elsewhere in the less developed world thus far, which have been based on hardware manufacturing, largely led by multinational companies seeking low-cost fabrication facilities. A combination of elements of the two approaches would open up much broader opportunities for poorer countries to benefit from this revolutionary technology. The study provides strong reasons for supposing that spin-off restructuring is possible and indeed necessary. User capability building in computer and information technology is one of the most promising directions for the spun-off enterprises. In the light of the dramatic fall in prices and increased tradability of computer and information technology, it is widely agreed that widespread applications of the technology are now much more affordable and applicable even for poorer countries. Such applications should be a critical factor for economic development, but have been delayed in part by weak capabilities in application engineering to translate the technologies into various uses. As Ernst and O’Connor say: What is increasingly clear is that to be an efficient user of new information technologies, a country must possess a degree of familiarity with the design and development of the hardware components and software which determine the basic functionality of various information-based systems or subsystems. It must also be knowledgeable about the interface between the new information technologies and other (e.g. mechanical) technologies to be able to combine them effectively. (Ernst and O’Connor 1989:39) While Ernst and O’Connor think of multinational as well as domestic producers of computers as potential contributors to building up the required interface, our study indicates that the interface for applications has started to function in China, but stems from domestic R&D institutions. Once the knowledge and familiarity with computers which is stored in R&D institutions is released so that it can be re-oriented for application purposes, the economic returns on the knowledge can be very high. Centralized coordination through government planning has proved to be ineffective for the widespread applications of computer and information technology.

Part 3 THE MACHINERY TECHNOLOGY R&D INSTITUTES Transforming the established industrial technology institutions

13 INTRODUCTION TO MACHINERY TECHNOLOGY INSTITUTES

Part 3 focuses on the transformation of the existing industrial technology R&D institutions, to complete our survey of the institutional restructuring of the Chinese industrial technology R&D system during the current market-oriented economic reforms. The transformation of existing R&D institutions is the most complex part of the restructuring, more complicated than either the spinning-off approach examined in Part 2, which leads to the establishment of a new type of organization operating according to new rules, or the approach of merging an institute with an enterprise, described in Part 1 in the context of the policy process for the S&T system and illustrated by cases in which the merging enterprise offers the organizational basis for the restructuring of the institute to be merged. The transformation of an existing R&D institute involves a process in which the old components of an institute are preserved as it passes through the transformation process, but are reorganized to match changes in both the functions and the operational norms of the institute. Two of the major questions to be studied in this part of the book are what these changes in technological functions and operational norms were, for the existing industrial technology R&D institutes, and how were these institutes able to achieve them? The importance of the transformation of existing industrial technology R&D institutions The importance of the transformation of the industrial R&D institutes is obvious. It is critical for the success of the reforms, both within the scientific and technological infrastructure and for the Chinese economy as a whole. A large part of China’s scientific and engineering manpower has been employed in the ‘independent’ industrial R&D institutes. In 1990, for instance, when the S&T reforms had been under way for five years, there were 2,109 such institutes employing 220,000 scientists and engineers. These ‘industrial technology R&D institutes’ include R&D institutes specializing in technologies for various manufacturing sectors, and for the transportation and post and telecommunications sectors. There were 1, 974 institutes for the ‘manufacturing’ sectors, employing 207,000 scientists and engineers. The ‘transportation’ sector accounted for 100 institutes and 3,000 scientists and engineers, while ‘post and telecommunications’ had 35 institutes with 2,000 such workers (China Statistical Yearbook on Science and Technology 1991:70–72, 78–80). This can be compared with the manpower involved in the spinning-off approach to restructuring. In the same year, 1990, there were 1,652 NTEs employing a total of 122,000 workers (Part 2: Table 9.1). No data is available to show how many of the NTE employees were scientists and engineers who had come directly from the industrial R&D institutes, but it may well be that the NTEs have had only a trivial impact on the transformation of machinery technology R&D institutions which are examined in this part of the book. Moreover, the NTEs are engaged in only a limited number of fields, mainly

88

MACHINERY TECHNOLOGY INSTITUTES

computer and information technology. Spin-off restructuring has made an insignificant contribution to the traditional areas of industrial technology, which were the major professional areas of the existing industrial R&D institutes. A similar comparison can be made between existing industrial R&D institutes and R&D-related establishments in industrial enterprises. In 1990, there were 7,289 industrial enterprises which had established in-house R&D departments, and these employed 314,000 scientists and engineers who were engaged in ‘technological development’ (a term used in the Chinese industrial statistics which is intended to cover a wider field than ‘research and development’). Just three years earlier (in 1987, the earliest year for which data is available), industrial enterprises had 4,633 in-house R&D departments employing 198,000 scientists and engineers (see Part 1: Appendix Table 1.9). This suggests that during the reform period industrial enterprises have surpassed the independent R&D in quantitative terms. But while there may be more in-house R&D departments employing more scientists and engineers, it does not follow that the R&D departments of industrial enterprises have surpassed the existing independent institutes in qualitative terms. The level of sophistication in industrial technology which the existing R&D institutes have achieved can hardly be surpassed in a short time because, as we will see in the following chapters, technological learning is a cumulative process in which capabilities, once acquired, lead to further learning-by-doing and so to greater capabilities. To attach importance to mastering technological sophistication implies that we consider design and testing capabilities to be very critical to the innovativeness of a manufacturing sector. These engineering capabilities are accumulated mainly in the existing industrial R&D institutes because product and process engineering, which mainly involve design and testing activities, were among the major assignments of these institutions under the old economic system. Chudnovsky and Nagao (1983:9–13) define a machinery industry in a developing economy as a producer of complex capital goods if it is able to do product design, whereas those without product design capability are categorized as early entrants. In the ‘chain-linked model’ of innovation, suggested by Kline and Rosenberg (1986) and aimed at describing the innovation process in manufacturing sectors, design is considered as the start of an innovation. While the Chinese economy had for a long time been investing heavily in design and testing activities for industrial technology, it had achieved only limited improvements in innovativeness and competitiveness until the recent reforms. The problem lay not in the level of investment but in the institutional arrangements of that period, including weak incentives for innovation. However the vast investment made in the past had brought with it an accumulation of engineering capability which could become more valuable. Success or failure in transforming the existing industrial R&D institutes, as part of the present reforms, will determine the extent to which the productive value of the ‘specific assets’ (to use a term from transaction cost theory, see Williamson 1975 and 1985) or ‘higher ordered factor creation’ (to use Porter’s term, see Porter 1990) which the industry possesses can be preserved and further developed. Why the machinery industry? There are several reasons for concentrating the study on machinery technology R&D institutes. In the first place, the machinery industry was one of a few industrial sectors which were given top priority and which received huge investments from the 1950s to 1970s. The share of the machinery industry in total industrial output increased from 11 per cent in 1952 to 17 per cent in 1957, 22 per cent in 1965 and 27 per cent in 1975. It then fluctuated between 22 per cent and 27 per cent in the following years. As a result, the R&D institutes for machinery technology grew significantly in relation to the total number of industrial technology R&D institutions. In 1990, for instance, there were 551 R&D institutes assigned to the broadly

INTRODUCTION

89

defined ‘machinery industry’ (ISIC 381 to 385), of which 26 were for metal products (ISIC 381), 347 for non-electrical machinery (ISIC 382), 63 for electrical machinery (ISIC 383, but excluding electronics and communication equipment), 15 for transportation equipment (ISIC 384, but excluding aircraft and ship building) and 100 for instrumentation (ISIC 385). These institutes together employed 54,600 scientists and engineers (China Statistical Yearbook on Science and Technology 1991:70–72, 78–80). Thus, in terms of the number of institutions and R&D manpower, the machinery industry possesses about one fourth of China’s total industrial technology R&D resources. In addition, a study of the machinery technology R&D institutions can be expected to be more widely illustrative of the transformation of industrial technology R&D institutes, even though three quarters of the 2,109 industrial technology R&D institutes are not officially affiliated to the machinery industry. This is because a number of those institutes were in fact assigned to the development of machinery technology for related industries. An example can be seen in Case Text 3.1 of Part 1, which is an institute in the mining industry. Other cases show a similar engagement observable in the food processing sector (ISIC 31) and the textile sector (ISIC 32). This means that the Chinese industry before the current reforms was heavily based on traditional mechanical technology. Indeed electronics, the emerging generic technology, played only a trivial role from the 1950s to the 1970s, and was very much restricted to military purposes in that period. It only began to be applied in diverse economic sectors in China from the mid-1980s (Gu and Steinmueller 1997). But it is necessary to remark that while the technological activities of the machinery industry R&D institutions may be representative, this does not mean that the approaches to transformation adopted by the various industries are uniform. As we will see, there is a considerable variety of approaches to institutional restructuring even within the machinery technology R&D institutes. Scope of the study As in the previous study of spin-off restructuring, the main approach adopted here is case studies of individual R&D institutes. These studies rely on intensive field work conducted in late 1994 and early 1995, supplemented with statistical data analysis and a literature survey and analysis. Some degree of generalization is then achieved on the basis of the empirical cases and supplementary data. The sample R&D institutes chosen for this survey are basically confined to those under the jurisdiction of the central Ministry of the Machinery Industry, with some extension to manufacturing enterprises and to machinery technology R&D institutes affiliated to local governments. A few manufacturing enterprises are included for comparison. The sample covers roughly the ISIC 38 industries (ISIC 381, 382, 383, 384 and 385, but excluding electronics and communication equipment and aircraft and ship building), the areas that have been described above as comprising the ‘machinery industry’ R&D institution. It is necessary to note that the categorization used in the management of the Chinese machinery industry is slightly different from the ISIC code. We will in most cases follow the Chinese sectoral categorization, from which it is not difficult to derive the ISIC code. Chapter 14 offers an historical perspective of the development of the Chinese machinery industry, and especially of the development of the industry’s design and R&D institutions. The chapter focuses on institutional aspects of the R&D and innovation functions in relation to the main features of the central administrative coordination of the planned economy in China. The outline of the institutional arrangements for R&D and of the reasons behind them which is presented in this chapter is intended to provide a background for the analysis of the present transformation in the following chapters. This chapter may also be read separately as a history of the Chinese machinery industry from the 1950s to the 1970s, explained from the point of view of the acquisition and adaptation of technology and its institutional basis.

90

MACHINERY TECHNOLOGY INSTITUTES

Chapter 15 analyses the official Chinese statistics on the income structure of the machinery technology R&D institutes at the industrial aggregate level. This analysis has a twofold purpose. First, it aims to identify how far these institutes have moved into the marketplace during the reform period. It shows that the degree of ‘marketization’ of the machinery technology R&D institutes has been very high, but no significant difference is found between this group and industrial R&D institutions as a whole. Second, a review of statistical indicators is used to explore the characteristics of the various kinds of market earnings, since there will be analytic implications for the use of official statistics at the individual institute level in the following chapters. Chapters 16 and 17 present the outcomes of institute case studies. Chapter 16 focuses on ‘product technology’ R&D institutes and Chapter 17 on ‘manufacturing technology’ R&D institutes, reflecting a functional division of the machinery technology R&D institutes that was used in the pre-reform system. To ensure that the case studies reflect the diversity of transformation paths, the subject institutes were selected with regard to the characteristics of the technology in which they engage (i.e., the complexity of the technology and/or the rapidity of changes in the technology), the firm structure of the sector which they were assigned to serve (i.e., the size of the technology market since the reforms began), as well as the importance of foreign technology imports in the field in which the institute specialized. From the ‘product technology’ institutes, a group that constitutes the overwhelming majority of the machinery technology R&D institutes, these criteria led to the selection of institutes in three subsectors: the machine tool sector, with a smaller technology market and in rapid technological change; the electric power plant equipment sector, which has a very small number of producers and a moderate pace of technological change; and the electric cable and small-capacity internal combustion engine sectors. The final two sectors have a large number of small and medium-sized firms which use the institutes’ output, so these institutes have a large ‘technology market’, a situation typical for a significant number of subsectors in the machinery industry. The emphasis in the case studies is on the centrally affiliated institutes, which have accumulated the most sophisticated technological capability. As in the case studies of spin-off restructuring, the aim is to understand the transformation of existing R&D institutes from both technological and institutional perspectives. However the context in which existing R&D institutes are transformed is different, since their past situation and their existing links with their industrial users are important. We must therefore examine questions such as what roles they are now playing in the technological change of a particular sector, what channels they rely on for the delivery of their technology, how they acquire and organize resources for the roles they play, and so on. By asking these questions we can trace the process by which the institutes have moved from the past to the present. Thus each case could be read as an epitome history of an institute and of the related sector. The official institute-level statistics for income structures are used to provide a point of reference for the interviews and other sources, and a basis for cross-institute comparisons and generalization. Chapters 16 and 17 both conclude with a generalization about the direction of the transformations and the factors which have influenced them, for the product technology institutes and the manufacturing technology institutes respectively. Chapter 16 shows that the ‘product technology’ institutes have developed in different ways according to their original institutional affiliation. Those which were under central coordination but based in a host enterprise have become key elements in the technological capabilities of their respective host enterprises, whereas previously they had served many firms in the sector to which they were assigned. Those which were affiliated to local governments have become active niche suppliers of selected machine products, while most of those which were ‘entirely’ centrally affiliated are moving to provide product engineering services, in some cases combined with plant engineering services, on the technology market. It is also found that the centrally affiliated product technology institutes have more opportunities to remain as

INTRODUCTION

91

independent market suppliers of product and plant engineering if the sector which an institute was previously assigned to serve has proved to offer a larger technological market, that is, it is a sector with a more decentralized firm structure. Chapter 17 illustrates the transformation of manufacturing technology R&D institutes. There is a tendency for these institutes to become commercial developers of complex manufacturing systems, where the conditions are right. The commercial development of manufacturing systems is largely a matter of systematically matching a set of techniques to the user’s conditions to achieve optimal operating performance. This is a new approach to technological innovation for manufacturing processes which the planning system could not support. What emerges from the empirical analyses of Chapters 15, 16 and 17 is that the innovation system for machinery technology, which was constructed separately from the industrial users of the technology, and which operated in the context of vertical planning coordination, had been decisively transformed by the mid-1990s. The transformation has produced an innovation system in which productive enterprises play an increasing role in innovative activities. Parts of the separate R&D structure have been incorporated into enterprise organizations, either through being merged in a host enterprise or by transforming an R&D unit in effect into a productive enterprise. Other institutes, most of which were previously affiliated to the central Ministry, continue to be independent from production but have turned to the commercial delivery of their product engineering and complex manufacturing systems. They have switched from vertical planning coordination to horizontal contracting transactions for the various engineering services they provide. This group is critical in a transitional period when industrial firms are still relatively weak, a point which readers may also recognize from the several enterprise cases which are covered in Chapter 16. Chapters 18 and 19 attempt to explore further two themes. Chapter 18 searches for changes in the characteristics of technological change due to the transformation of the R&D and innovation system, that is, it looks for changes in the ‘technological trajectory’. Technological change is seen as a selectively cumulative process in which change takes place in some particular directions and not in others, and in which learning from previous change sets the options which are open for future changes. Technological change in a particular context therefore develops a certain trajectory which has particular characteristics. Yet, with sufficient effort and incentive, it is possible to shift to a new technology trajectory, as is happening in the course of China’s economic transition. In the case of the Chinese machinery industry, there has been a shift during the reforms from a trajectory characterized by the development of ‘general technology’ to one which is increasingly sensitive to ‘specific technology’. The market-oriented reforms contribute to this pathshifting in two respects, by providing an incentive structure which has led technology producers and users to give more attention to factors such as product quality and production efficiency, and by supporting the development of information channels, in the context of contractual relations, which assist horizontal communications. Horizontal communications are very critical to the learning required for specific problemsolving, and were weak under vertical planning coordination. Learning to solve specific problems leads to the development of technologies specific to individual firms and to particular manufacturing contexts. These specific technologies are the basis for market competitiveness. Thus the rapid improvement in the international competitiveness of the Chinese machinery industry in terms of exports (introduced in Chapter 14) can be explained from a change in the learning process at the micro level. This also offers an example of the considerable impact of social dimensions on shaping technology trajectories and the shift in trajectories which is involved in the transition from one economic system to another. Chapter 19 interprets the institutional restructuring of the centrally affiliated institutes from the transaction cost perspective. The analysis of restructuring focuses on those institutes which have remained independent from productive enterprises while shifting to a trajectory which is sensitive to the development of specific technology. Drawing upon Williamson’s contractual schema, the chapter shows how the external

92

MACHINERY TECHNOLOGY INSTITUTES

contractual relations and internal organization of these institutes have moved from one state to another and the factors which have affected their movements. The analysis reveals that there are unavoidable intermediate states of transformation in both external contractual relations and internal organization once an institute is released from planning coordination. These intermediate states are very unstable and there is a risk that specific institute assets which have been accumulated in the past may be lost. The Chinese reform programmes offered a period of several years in which these institutes could adapt to the necessary and fundamental restructuring. These programmes also delegated decision-making autonomy to the institute level where specific institute assets could be more closely managed. These two aspects of the reform programmes have probably contributed to lowering the transition costs faced by the Chinese innovation system for machinery technology, in comparison with the transition costs suffered in the former centrally planned economies of Eastern and Central Europe. Throughout the period of market reform, shifting the trajectory of technological change demands intensive technological learning. This technological learning requires an appropriate institutional framework, which also has to be continuously reformed. Hence institutional learning has been as intensive as, and no less crucial than, technological learning. ‘Getting the institutions right’ is highlighted in our analysis as part of a strategic schedule of economic development and economic transition. We believe that the schedule of getting the institutions right offers a better balance, and a better integration of the various factors involved, than those schedules which have solely emphasized ‘getting the prices right’ or ‘getting the property rights right’. This is because it is institutional learning as well as technological learning that determines the actual progress of market reform, and institutional learning involves not only macro-level factors such as price and property right structures, but more critically adjustments in the organizations and operational rules of economic players, to fit market mechanisms. These adjustments are determined and mastered by people working in institutes at the grass-roots. As this study demonstrates, institutional analysis and technological analysis are useful and informative for studying an economy in transition, and a combination of the two makes it possible to focus on the endogenous process of learning and transformation. Other recurring themes are dealt with throughout the following six chapters, as the matter requires. For instance, the limitations in reform policy and practice thus far are discussed in the course of each chapter and especially in the chapter summaries.

14 THE DEVELOPMENT OF THE MACHINERY INDUSTRY IN CHINA PRIOR TO MARKET REFORM Industry performance and R&D institutions

Introduction This chapter describes the context in which the transformation of the machinery industry R&D institutes in China began. Two facets of this background will be introduced: the development of the industry, and the design and R&D institutions of the industry. The focus will be on the institutional aspects, but these will be related to main features of the industry’s development. The ‘machinery industry’ is defined here, as in the following chapters, as the industry producing machinery and equipment which becomes part of capital formation, so that it is roughly synonymous with the term ‘capital goods’. Bulk intermediate goods such as metal and non-metal materials are not included. The discussion will be confined, so far as possible, to ISIC (International Standard Industrial Classification) division 38, excluding ‘electronic and communication equipment’.1 The development of the machinery industry in China has been characterized by two seemingly contradictory aspects. There was a high growth rate from the 1950s to 1970s resulting in a great capacity in volume terms at the end of this period, but the efficiency of the development has generally been low: measured by some indicators, even lower than many other developing countries. Both of these characteristics were in fact embodied to a great extent in the institutional framework in which the industry developed. The current reforms may well improve the development efficiency, but the transformation of the industry still has a long way to go. The development of the machinery industry in China can be divided into three stages. The first stage covers most of the 1950s, when technological efforts focused on introducing a modern machinery industry. During this period the institutions for large-scale production and for research and development in machinery technology were created. The second stage covers the 1960s and 1970s, when industry development centred on extensive capacity expansion. During this period the institutions which had been established were expanded and elaborated, despite some interruptions, especially during the ‘Cultural Revolution’ (1966– 1976). The third stage covers the current reforms, beginning in the late 1970s. The industry is moving from a centrally planned form. Direct government control is being withdrawn and enterprises are being granted more autonomy, while international exchanges are expanding dramatically. The second part of this chapter will describe the evolutionary path of the machinery industry’s R&D and design institutions during the first and second of these stages, but with a selective focus on issues with a significant impact on the current transformation. It covers two kinds of institutional establishments: plant design institutes charged with capital investment tasks, and R&D institutes for the development of product

94

MACHINERY TECHNOLOGY INSTITUTES

and manufacturing technology. However the focus will be on the R&D institutes, which are the subject of the current INTECH study. It is necessary to mention that a third kind of institution, productive enterprises charged with the production of machinery products, is not covered in this study. The three kinds of institutions, all essential to the machinery industry, started to take shape in China in the 1950s. Under the centrally planned regime2 which prevailed from the 1950s to the 1970s, plant design institutes, R&D institutes, and productive enterprises were separately organized, with each kind of institution having a clearly specified function. Decisions on investment, technology imports, and product development were made from above, and what coordination there was between these separate functional establishments was also achieved through the planning administration. Before considering the development of the machinery industry during the first phase, a few words are needed regarding the literature utilized in this chapter. Two works stand out as the most useful, and will be the most frequently cited here, apart from reference materials such as the various statistical yearbooks. One is a massive book in two volumes entitled The Machinery Industry in Contemporary China. This was edited by an authoritative editorial committee of forty-two members, headed by Jing Xiaocun. The committee members are the most responsible and experienced technological and managerial experts in this industry in China. This book is published as part of the ‘Contemporary China Series’, which comprises two hundred volumes covering various sectors of the Chinese economy. The series is intended to summarize the experiences and lessons achieved during the first thirty or so years of post-war development. The second preeminent work is a three-volume book entitled A History of the Electrical Equipment Industry in China, also edited by a large and authoritative editorial committee of fifty-three members, with Zhou Jian’nan as Chief Editor. These two sources will be cited here as ‘Jing Xiaocun (ch.ed.) 1990’, and ‘Zhou Jian’nan (ch.ed.) 1990’, respectively. They are especially useful for information on historical events which have important effects on institutional evolution. Such sources are valuable in making an institutional analysis, because they often indicate details and subtleties relevant to understanding the institutional evolution. Ordinary statistics do not convey this sort of information. The other important source of information is my own field work from 1994–1995, which is extensively utilized here. Development of the machinery industry While industrialization efforts can be traced back to the second half of last century, by the end of the Second World War China had, for many reasons, developed no more than a limited industrial foundation. In 1952 the machinery industry, in the modern sense, was trivial. Its activities were predominantly in the fields of repair, maintenance, and assembly (see Cheng 1972: 25–29; Jing Xiaocun (ch.ed.) 1990: Vol. A, 3–5). In 1952, agricultural output constituted about 60 per cent of national income, and per capita GNP was 104 yuan, equivalent, at the official exchange rate, to 52 US dollars (China Statistical Yearbook 1989:28, 29). Entry and growth, 1950–1970 China’s entry into the modern machinery industry was realized through large-scale technology imports made possible by the powerful central mobilization of resources. During the first five-year plan (1953–1957), technology import centred on four sectors: power-generating equipment, transportation equipment, metallurgical and mining equipment and machine tools (Cheng 1972: Table 3.5 on page 31, Jing Xiaocun (ch.ed.) 1990: Vol. A, 19–21). In addition to the four priority areas, investment in this period also covered various branches of the machinery industry, embracing agricultural machinery, oil-refinery and chemical

MACHINERY INDUSTRY DEVELOPMENT

95

engineering equipment, textile machinery, ball bearings, measuring instruments, etc. (Cheng 1972:40–41; Jing Xiaocun (ch.ed.) 1990: Vol. B, 53–54). These technology imports formed the hard core of the machinery industry and laid a foundation for a Chinese industry capable of producing heavy-duty and precision machinery, which had not existed before. The central mobilization of resources helped to overcome barriers to the development of a modern machinery industry. Industrially underdeveloped countries commonly face barriers in developing a capital goods industry because of a shortage of capital and lack of skilled manpower. Three routes to overcoming these barriers can be differentiated: 1) evolutionary entry, through the gradual accumulation of capabilities from repair and maintenance experience, 2) policy-induced entry, and 3) entry based on a vehicle assembly industry (Chudnovsky and Nagao 1983:38, 44–45). China’s entry followed the second of these paths: it was strongly and directly policy-induced. Table 14.1 Selected indicators of the development of the capital goods sector (ISIC 38) in China, India, Brazil and the Republic of Korea (1979–1980) Indicator 1

China

India

Brazil

Korea, Republic of

107.2

18.3

15.8

6.0

13,600

1,283

1,031

417

50.6

8.0

16.9

7.2

2.9

1.4

4.0

6.1

4

Number of establishments1 (x 1,000) Number of workers employed1 (x 1, 000) Gross output ($ billion)2 Imports ($ billion)

5

Exports ($ billion)

0.4

0.5

2.4

2.4

6

Export ratio of domestic production (5):(3) (%) (5) as percentage of all manufactured exports Apparent consumption (3)+ (4)–(5) ($ billion) Domestic supply ratio ((3)–(5)):(8) (%) Value added ($ billion)4

0.8

6.2

14.2

33.3

n.a.

13.3

29.0

19.7

53.1

8.9

18.5

10.9

94.5

84.3

78.4

44.0

15.2

2.1

7.2

2.7

2

3

7

8

9

10

3

3

96

MACHINERY TECHNOLOGY INSTITUTES

Indicator 11

12 13

14

15

Production of machine tools5 ($ million) of which, exports ($ million) Imports of machine tools5 ($ million) Apparent consumption of machine tools (11)+(13)–(12) ($ million) Domestic supply ratio for machine tools ((11)–(12)): (14)

China

India

Brazil

Korea, Republic of

420

165

315

135

28

25

71

26

140

76

175

344

532

216

419

453

74

65

58

30

Source: Chudnovsky and Nagao 1983:96–97. Notes: 1 The data for Brazil (1977) and Korea (1979) covers establishments employing five or more workers; for India (1978), those with ten or more workers and using electric power, and with twenty or more workers not using electric power; for China (1980), it is all state-owned or collectively owned enterprises. Establishments making radio and TV sets (ISIC 3832) are not included in the data for India and the Republic of Korea. 2 The information relates to 1979 for China, Korea, and Brazil, and to 1978 for India. Calculated on the basis of official conversion rates to the US dollar. 3 The data is calculated based on SITC 7, except that passenger vehicles and TV sets are excluded. The information relates to 1979 for China, Brazil and Korea, and 1978 for India. Note that the value of Chinese exports given here is different to that given in Table 14.6 for the same year, which is 0.81 billion US dollars. The disparity may be caused by differences in definitions. For instance, SITC 69 (metal manufactures) is included in the data in Table 14.6 but excluded from this table. 4 The information relates to 1979 for China, Brazil and Korea, and 1978 for India. The value added for China is estimated as 30 per cent of gross output. 5 The data is for 1980, estimated on the basis of figures in the American Machinist, February 1982.

Once established, the machinery industry grew rapidly. The average annual growth in gross output between 1953 and 1985 was 15 per cent, which is higher than the 11 per cent average growth rate of industry as a whole over the same period3 (State Council and SSB 1987:134, 135). The share of the machinery industry in total industrial output increased from 11 per cent in 1952 to 17 per cent in 1957, 22 per cent in 1965, and 27 per cent in 1975 (ibid.: 132–133). It then fluctuated, falling as low as 20 to 22 per cent in 1981–2 and 1991, and reaching 24 to 27 per cent in the remaining years (China Machinery Industry Yearbook 1994:VI–2).4 A change in the industrial planning system which began to take effect in the late 1970s was largely responsible for the fluctuation. Pushed by central planning, which gave high priority to the development of ‘heavy’ industries, the machinery industry was faced with enormous demands. Manufacturing enterprises were continuously

MACHINERY INDUSTRY DEVELOPMENT

97

constructed to cope with the demand, with plant design institutions planning the capacity expansion, as we will see in the description of institutional evolution below. The large size of the Chinese machinery industry as it developed in the three decades from the 1950s can be clearly seen from a comparison with the Indian, Brazilian and Korean machinery industries (Table 14.1). These countries are regarded as typical of the few developing countries which have successfully entered into complex capital goods production. The table shows that Chinese achievements are impressive in all quantitative measurements, such as number of workers, number of enterprises and gross output. The comparison is quite striking, although some of the data used in Table 14.1 is not strictly comparable. Branch structure The Chinese machinery industry has a very big ‘non-electrical machinery’ branch (382), and a smaller ‘transportation equipment’ branch (384), as compared with counterparts in other developing countries (see Table 14.2). This seems to be largely due to demand side factors. The pursuit of ‘self-reliant’ industrialization created enormous demands for domestically manufactured production equipment from the machinery industry, while passenger vehicles were not, until recently, in great demand because of the low level of family incomes and limited mobility of Chinese residents. The combined machinery industry had a major role in manufacturing industry as a whole. By the mid-1980s it accounted for about 30 per cent of the value added in the entire Chinese manufacturing industry,5 which was Table 14.2 Size of selected manufacturing sectors, by share in total value added (1985–1987) Branch (ISIC code)

China

India

Brazil

Korea, Republic of

Food products (311/2) Beverages (313) Tobacco (314) Textiles (321) Wearing apparel (322) Leather and fur products (323) Footwear (324) Wood and cork products (331) Furniture, fixtures excl. metal. (332) Paper (341) Printing and publishing (342) Industrial chemicals (351) Other chemicals (352) Petroleum refineries (353) Rubber products (355) Plastic products (356) Pottery, china, earthenware (361) Glass (362) Other non-metal. min. prods. (369) Iron and steel (371)

11.4

11.6

14.3

12.6

14.1

15.2

10.4

16.6

1.4

0.5

4.3

1.4

2.0 1.2 7.4 3.1 4.6 1.9 1.7 0.6 1.0 6.1 7.8

1.6 1.9 7.3 7.5 3.5 2.4 0.9 0.2 0.5 4.6 10.8

3.6 2.4 6.6 5.5 4.3 1.3 2.8 0.2 0.7 5.1 5.7

2.5 2.3 3.9 4.6 3.0 3.1 2.4 0.3 1.0 3.3 6.2

98

MACHINERY TECHNOLOGY INSTITUTES

Branch (ISIC code)

China

India

Brazil

Korea, Republic of

Non-ferrous metals (372) Metal products (381) Non-electrical machinery (382) Electrical machinery (383) Transport equipment (384) Prof., scient. equipment (385) Total ISIC 38

2.2 3.3 14.0 7.7 4.7 1.2 30.9

0.9 2.7 9.1 8.1 8.1 0.8 28.8

2.5 4.7 8.4 6.3 5.9 0.6 25.9

1.1 4.3 5.4 13.2 8.7 1.1 32.7

Source: Handbook of Industrial Statistics 1990, UNIDO: 94, 96, 105, 115.

approximately the same as in India, Brazil and the Republic of Korea, as shown in Table 14.2. For comparison, Table 14.3 presents data for Japan, the US and West Germany, illustrating the situation in these typical industrially-developed countries at about the same time. The ISIC 38 industry in Table 14.3 Size of selected manufacturing sectors, by share in total value added (1985–1987) Branch (ISIC code)

Japan

US

Germany, Federal Rep. of

Food products (311/2) Beverages (313) Tobacco (314) Textiles (321) Wearing apparel (322) Leather and fur products (323) Footwear (324) Wood and cork products (331) Furnit., fixtures excl. metal. (332) Paper (341) Printing and publishing (342) Industrial chemicals (351) Other chemicals (352) Petroleum refineries (353) Rubber products (355) Plastic products (356) Pottery, china, earthenware (361) Glass (362) Other non-metal. min. prods. (369) Iron and steel (371) Non-ferrous metals (372) Metal products (381) Non-electrical machinery (382) Electrical machinery (383) Transport equipment (384)

9.6

11.8

9.6

5.4

5.3

4.1

2.7

3.0

2.5

2.5 5.3 4.3 5.2 1.1 1.3 3.4 0.4 1.0 3.1 5.5 1.2 6.5 12.6 15.1 10.4

4.2 7.4 4.4 5.5 1.4 1.1 2.5 0.1 0.8 2.0 2.4 1.2 6.1 11.4 11.1 12.8

2.4 1.9 7.4 5.2 4.3 1.3 2.6 0.3 0.9 2.2 4.1 1.5 6.4 15.1 12.8 12.8

MACHINERY INDUSTRY DEVELOPMENT

Branch (ISIC code)

Japan

US

Germany, Federal Rep. of

Prof., scient. equipment (385) Total ISIC 38

1.6 46.2

4.0 45.4

1.5 48.6

99

Source: Handbook of Industrial Statistics 1990, UNIDO: 102, 107, 120.

these three industrialized countries accounts for about 45 per cent of the value added in the entire manufacturing industry, reflecting an apparently better-developed capital goods industry (although the internal structures of the ISIC 38 industries in these countries differ). In the mid-1980s the ISIC 38 industries of most other developing countries, which have not been included in Table 14.2, accounted for 20 per cent or less of the entire manufacturing industry. China, India, Brazil and South Korea can thus be regarded as leading the developing countries in this field. These very brief and preliminary comparisons highlight some features which may characterize the process of industrial development, and suggest that the development of the capital goods industry is highly significant for industrialization.6 Product diversification The products of the Chinese machinery industry have been much diversified over the past forty years, as shown by a continuing increase in the numbers of product varieties. Several thousand new products were introduced in each decade from the 1950s on (Jing Xiaocun (ch.ed.) 1990: Vol. B, 249–255; Cheng 1972: 132–133). By 1985, the Chinese machinery industry was able to produce more than fifty thousand varieties of products (Jing Xiaocun (ch.ed.) 1990: Vol. B, 248), while in the early 1950s it was able to produce only one thousand, and in the mid-1960s about ten thousand (Cheng 1972:133). Product diversification was accompanied by a high ratio of domestic supply, which increased from about 50 per cent in the mid-1950s to 90 per cent in 1979 (Jing Xiaocun (ch.ed.) 1990: Vol. B, 250; Vol. A, 146). The domestic supply ratio has declined under the current reforms (see below). Product diversification has been aimed predominantly at adapting machinery technology to local needs. General-purpose machine products were the most common products of the diversification. Typically, basic designs that had originally been developed abroad were modified to produce what Chinese managers called a ‘partly modified (original) design’ (jubu gaijin sheji) or ‘analogously-made (from original) design’ (leibi sheji) (Jing Xiaocun (ch.ed.) 1990: Vol. B, 256; and interviews). Simplifications and down-scaling modifications were also quite common, under a ‘two legs’ policy which encouraged the simultaneous development of small enterprises, for instance in agricultural machinery, metallurgical equipment, chemical fertilizer equipment and small-scale power plant equipment, and large-scale enterprises. The development of more sophisticated machinery was less common, but a few instances in the fields of precision machine tools and large-scale power plant equipment will be given in the section discussing institutional evolution. Another type of product diversification which has occurred is the improvement of products already in production, but this has been rather less common than the localization of foreign designs. One notable example in recent years has been the development of more energy-efficient machinery. Since the first half of the 1980s, improvements have been made in the energy-consumption levels of thousands of varieties of products in twenty categories of machinery such as industrial boilers, water pumps, diesel engines, electrical motors, air compressors and electric welders (Jing Xiaocun (ch.ed.) 1990: Vol. B, 247, 253–254). This effort has underpinned the great improvement in the energy efficiency of the Chinese economy since the early 1980s.7

100

MACHINERY TECHNOLOGY INSTITUTES

Since the mid-1970s the Chinese machinery industry has taken on product diversification in more challenging producer goods such as large oil-based nitrogen fertilizer plants (Jing Xiaocun (ch.ed.) 1990: Vol. A, 263–265). More remarkably, some machinery for producing consumer goods has been significantly modified and improved. For example, food processing and packing machinery, which was absent from Chinese machinery industry statistics before the reform, has been developed since the late 1970s. It is reported that by 1993 there were 3,000 enterprises producing food processing and packing machinery, with an annual output of 11 billion yuan, some of it for export (China Machinery Industry Yearbook 1994:IX–7, 8, 9; China Machinery Industry Yearbook 1991:111–74, 75, 76). Productivity The productivity of the Chinese machinery industry has been low, even lagging behind that of other major developing countries. This can be clearly seen from the number of workers employed and their gross output as shown in Table 14.1. According to these figures, the labour productivity of the Chinese machinery industry in the 1979–1980 year was lower than that of Korea, Brazil or India, although some statistical incompatibilities should be noted.8 A study carried out jointly by Chinese, American and Japanese scholars9 reveals that the average annual growth rate of the Chinese economy from 1953 to 1978 was 5.92 per cent. This growth resulted almost entirely from increases in capital stock, which grew by 5.51 per cent per annum, and in labour input, which grew by 1.21 per cent per annum. Productivity actually declined by 0.8 per cent, as shown in Table 14.4. This is comparable to the outcomes of a World Bank study (World Bank 1985: Ch.7), which indicated that stateowned industry in China had not shown any improvement in total factor productivity until the current reforms. Since the reforms began, however, the situation has been changing. Increased productivity has become a source of economic growth. The same joint study revealed that, of the 8.35 per cent average annual growth between 1979 and 1990, 2.53 per cent was contributed by productivity improvements, 4.25 per cent by increased capital investment, and 1.57 per cent by increased labour input (Li Jingwen et al. 1993:52). Another study, by Szirmai and Ruoen (1995) suggests that in the 1980s Chinese manufacturing as a whole had attained a productivity level of Table 14.4 Output, input and productivity in China (1953–1990) Year

Growth rate (%)

Output

Labour input

Capital input

Labour

Capital

Increase in productivity

1953 1954 1955 1956 1957 1958 1959 1960 1961 1962

12.34 5.59 6.21 13.19 4.38 19.97 7.82 −1.44 −33.46 −6.75

2.89 2.57 2.00 2.37 3.00 8.39 5.55 −0.86 −1.07 −0.73

20.83 15.51 13.42 15.31 16.94 22.06 26.83 20.91 3.39 −0.09

1.36 1.26 0.97 1.12 1.49 4.16 2.76 −0.44 −0.55 −0.39

11.04 7.90 6.88 8.09 8.52 11.14 13.46 10.33 1.65 −0.04

Contribution to growth rate (%)

−0.06 −3.57 −1.64 3.98 −5.63 4.69 −8.40 −11.23 −34.56 −6.32

MACHINERY INDUSTRY DEVELOPMENT

Year

Growth rate (%)

Output

Labour input

Capital input

Labour

Capital

Increase in productivity

1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990

10.19 15.30 15.64 15.72 −7.52 −6.77 17.69 20.58 8.74 2.80 7.97 1.14 7.99 −2.69 7.50 11.57 7.33 7.51 4.39 8.37 9.82 13.61 11.97 7.98 10.45 10.34 3.49 4.91

1.57 3.40 3.71 3.64 3.55 3.37 3.74 3.87 3.69 2.22 1.54 2.12 2.20 2.15 1.76 1.84 2.21 2.85 3.32 1.26 5.18 3.17 3.66 3.19 2.90 2.96 2.34 2.14

2.53 4.22 8.69 7.86 3.78 3.03 4.12 9.40 9.23 7.98 9.45 8.02 8.14 7.00 7.53 9.96 1.36 8.27 6.62 6.76 7.73 8.77 11.32 12.42 9.89 10.00 9.07 8.59

0.84 1.73 1.74 1.59 1.57 1.58 1.70 1.56 1.43 0.86 0.61 0.86 0.91 0.92 0.78 0.80 0.99 1.36 1.66 0.64 2.66 1.68 2.02 1.78 1.63 1.68 1.34 1.20

1.17 2.07 4.61 4.42 2.10 1.60 2.25 5.60 5.66 4.82 5.59 4.74 4.76 3.99 4.30 5.62 5.72 4.31 3.30 3.29 3.77 4.09 5.07 5.49 4.32 4.32 3.87 3.75

101

Contribution to growth rate (%)

8.18 11.50 9.29 9.71 −11.19 −9.95 13.74 13.72 1.65 −2.88 1.68 −4.46 2.32 −7.60 2.50 5.15 10.62 1.84 −0.57 4.44 3.39 7.84 4.88 0.71 4.50 4.35 −1.72 −0.04

Source: Li Jingwen et al. 1993:56. Note: Data is calculated based on 1978 constant prices. Statistics have been transformed in accordance with the United Nations’ SNA (i.e. ‘System of National Accounts’).

roughly one twentieth of that attained by US manufacturing (see Table 14.5). This study compared productivity by means of purchasing power parity (PPP).10 While labour productivity in the Chinese ‘machinery and transport equipment’ sector was just 4 per cent of the US level, labour productivity in the Chinese ‘electrical machinery and equipment’ sector was 8–10 per cent of the US level. Another surprising result of this study was that, in spite of the high growth which was achieved in the period 1980–1992, the comparative productivity of Chinese manufacturing and of the Chinese machinery industry has not improved significantly. Some sectors, such as the ‘machinery and transport equipment’ sector, appear to

102

MACHINERY TECHNOLOGY INSTITUTES

have improved slightly, but by 1992 the gap between Chinese productivity and the world leaders in productivity remains as wide as it was a decade earlier.11 Table 14.5 Comparative productivity by manufacturing branch, China and USA (1980–1990) (USA=100) Branch

1975

1980

1982

1984

1986

1988

1990

1992

Food and beverages Tobacco products Textile mill products Wearing apparel Leather products and footwear Wood products, furniture and fixtures Paper products, printing and publishing Chemicals, petroleum and coal products Rubber and plastic products Non-metallic mineral products Basic and fabricated metal products Machinery and transport equipment Electrical machinery and equipment Other manufacturing Total manufacturing Source: Szirmai and Ruoen 1995: Table 14.

— — — — — — — — — — — — — — 4.4

3.5 2.3 8.2 5.4 11.1 2.8 3.0 7.7 5.8 6.2 8.9 2.5 7.5 3.8 4.8

3.3 2.7 7.4 5.9 10.6 3.3 3.0 7.4 5.3 6.6 10.1 2.9 9.0 3.6 5.0

3.2 2.9 6.3 6.0 9.8 3.0 3.1 6.1 4.8 6.2 9.7 2.5 10.0 3.2 4.6

3.3 3.2 5.9 6.3 12.3 3.3 3.1 5.7 4.4 5.6 9.8 2.8 10.8 3.3 4.7

3.6 5.0 5.9 6.2 10.6 3.9 3.1 5.3 4.4 6.0 9.5 3.0 10.1 2.8 4.7

3.7 6.4 4.8 5.8 10.3 3.7 2.8 5.1 4.0 5.2 8.2 3.0 8.2 2.6 4.4

4.6 7.5 4.6 6.7 8.6 5.4 3.2 6.1 4.3 7.4 9.1 4.2 8.2 3.0 5.2

Exports and imports From the 1950s to the 1970s, exports and imports were insignificant, largely due to the policy of selfreliance. Table 14.1 gives a comparative picture of four major developing countries in the late 1970s, a picture which is broadly representative of the situation throughout these three decades. In 1979 just 0.8 per cent of China’s domestic ISIC 38 products were exported, compared to 6.2 per cent for India, 14.2 per cent for Brazil and 33.3 per cent for the Republic of Korea. Chinese imports of ISIC 38 products were also low, with a ‘domestic supply ratio’ of 94.5 per cent in the late 1970s, compared to 84.3 per cent for India, 78.4 per cent for Brazil and 44 per cent for the Republic of Korea. China’s machine tool exports were worth 28 million US dollars in 1979, slightly higher than India and the Republic of Korea in absolute terms, but lower than Brazil. The domestic supply ratio for machine tools alone was 74 per cent in 1979, compared to 65 per cent for India, 58 per cent for Brazil, and 30 per cent for the Republic of Korea. Under the current reforms there have been dramatic changes in international trade. Both exports and imports of machinery products have expanded steadily, as can be seen from Table 14.6. Imports, in absolute terms, have increasingly been surpassing exports. In 1993, China imported about 50 billion US dollars worth of machinery products and exported more than 22 billion dollars worth. The share of Chinese machinery exports in the world market increased from 0.15 per cent in 1978 to 1.61 per cent in 1993, while

MACHINERY INDUSTRY DEVELOPMENT

103

the domestic supply ratio for all machinery products has fallen from more than 90 per cent to about 50 to 60 per cent during the past fifteen years (China Machinery Industry Yearbook 1994: I–24, 25).12 The purpose of machinery imports and the origin of exports (Table 14.7) show the effects of the reforms on trade patterns. In 1993, 20 per cent of machinery imports were related to export-oriented assembly or manufacturing, 31 per cent were associated with direct international investment, and imports in ‘ordinary terms’, i.e. those not intended for such special uses, accounted for 34 per cent. In the same year, 71 per cent of machinery exports were assembled or manufactured from imports, and 34 per cent were ‘ordinary’ exports. These figures show that Chinese machinery imports have to a significant extent been driven by international investment and export-oriented assembly or manufacturing, and that Chinese machinery exports have been predominantly based on imports. Thus it is clear that the Chinese machinery industry has been substantially ‘globalized’. Turning now to machine tools in particular, it can be seen from Table 14.8 that exports and imports have also grown substantially compared to the figures for 1979 given in Table 14.1. In 1993, China exported machine tools Table 14.6 Exports and imports of machinery products (1978–1993) Year

Imports of machinery products (billion US dollars)

Exports of machinery products (billion US dollars)

Machinery product China’s machinery exports as a percentage of exports as a percentage of total exports world machinery product exports

1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993

— — 5.65 6.35 3.61 4.67 8.24 18.29 19.35 17.39 20.70 22.50 21.43 25.05 34.89 49.46

0.65 0.81 1.39 1.81 1.94 1.94 2.19 1.68 2.48 3.85 6.16 8.32 11.09 14.12 19.55 22.70

6.75 5.95 7.67 8.18 8.71 8.72 8.40 6.13 8.02 9.79 12.96 15.83 17.86 19.64 23.00 24.75

0.15 0.16 0.23 0.30 0.33 0.33 0.34 0.24 0.30 0.39 0.61 0.77 0.90 1.11 1.46 1.61

Source: China Machinery Industry Yearbook 1994:1–15, 16. Note: ‘Computer and telecommunication equipment’ is included in ‘machinery products’ in this table. Table 14.7 Purposes of imports and origin of exports for the Chinese machinery industry (1993) Imports in 1993

Exports in 1993

Total: 49.49 billion US dollars 1 ordinary imports:

Total: 22.70 billion US dollars 1 ordinary exports:

34%

26%

104

MACHINERY TECHNOLOGY INSTITUTES

Imports in 1993 2

3

Exports in 1993 imports for export assembly or manufacturing: imports associated with direct foreign investment:

20%

2

exports from assembly or manufacturing of imports:

71%

31%

Source: China Machinery Industry Yearbook 1994:1–56, 57. Note: The data here includes the ‘electronics industry’, which accounted for 22 per cent of the imports (10.7 billion US dollars) and 36 per cent of the exports (8.12 billion US dollars) in 1993 (China Machinery Industry Yearbook 1994:1–22, 1–23). The percentage breakdowns of imports and exports do not add up to 100 per cent, because other categories such as imports based on barter trade agreements or compensatory trade agreements are excluded.

worth 200 million US dollars, compared to just 28 million dollars in 1980. Chinese machine tool exports in 1993 exceeded those of India or the Republic of Korea, and were similar to Brazilian exports. On the other hand, Chinese imports of machine tools increased even faster, reaching 1,522 million US dollars in 1993, which made China the second biggest importer of machine tools in the world. As a result, by 1993 the domestic supply ratio for China, calculated on the basis of the data in Table 14.8, had fallen to 50 per cent, while the Korean domestic supply ratio had risen to 59 per cent and the ratio for Brazil to 66 per cent. India’s domestic supply ratio, on the other hand, had fallen to 43 per cent. The structure of machine tool imports and exports is shown in Table 14.9. Imports are biased towards the technologically sophisticated end of the product Table 14.8 Exports, imports, production and consumption of machine tools in selected countries, in billions of US dollars (1992–1993) Country

Exports

1992

1993

1992

1993

1992

1993

1992

1993

Japan Germany United States China Republic of Korea Brazil India

3.532 4.685 1.214 0.197 0.110 0.19 0.017

3.644 3.329 1.010 0.200 0.110 0.195 0.017

0.547 1.367 1.874 0.838 0.967 0.086 0.191

0.378 1.150 2.50 1.522* 0.700 0.091 0.185

8.355 7.686 3.074 1.685 0.576 0.306 0.203

7.154 5.145 3.275 1.753 0.822 0.326 0.156

5.37 4.848 3.734 2.326 1.433 0.213 0.367

Imports

Production

Consumption 3.887 2.968 4.765 3.075 1.213 0.200 0.324

Source: WMEM, 1994, Vol. 3:43, citing the American Mechanist for March 1994. Note: *According to the China Customs statistics, machine tool imports were 1.9407 billion US dollars in 1993.

MACHINERY INDUSTRY DEVELOPMENT

105

Table 14.9 Chinese machine tool imports and exports, in thousands of US dollars (1993) Imports

Metal cutting machine tools Metal forming machine tools Total

Exports

Value

Unit value CNC machines, percentage by value

Value

Unit value CNC machines, percentage by value

1,149,700

14.4

56%

175,500

0.1*

8.4%

790,770

15.8

28%

41,080

1.8

4.2%

1,940,470

216,280

Source: WMEM 1994: Vol. 2, 61. Note: *The metal cutting machine tools include some grinders and bench drills, which lower the average unit value of this category. Table 14.10 Industrial technology imports by sector (1950–1990) Branch (equivalent ISIC code)

Number of contracts/percentage of expenditure

1950–1959

1963–1968

1971–1978

1979–1990

1963–1990

(21–29) Coal mining, crude petroleum and natural gas, ironbearing ore, nonferrous metalbearing ores, construction material ores and other ores (31) Food products, beverages tobacco and prepared animal feeds (32–33) Textiles, wearing apparel, leather and fur products, wood and cork products and furniture (34) Paper, printing and publishing (35) Petroleum refineries, industrial chemicals, pharmaceuticals,

38/6.6

2/2.1

51/12.3

139/2.5

192/4.8

8/1.6

–

–

112/1.7

112/1.3

11/2.6

1/9.6

9/1.8

69/1.0

79/1.1

5/1.2

2/2.4

1/0.02

66/1.8

69/1.3

22/11.0

13/28.8

86/38.5

685/25.2

784/28.4

106

MACHINERY TECHNOLOGY INSTITUTES

Branch (equivalent ISIC code)

Number of contracts/percentage of expenditure

1950–1959 synthetic fibre, rubber and plastic products (36) Construction materials and other non-metal mineral products (37) Iron, steel and non-ferrous metals (381) Metal products (382) Nonelectrical machinery (383) Electrical machinery, of which: Electronic and communication equipment Computers Household electronic equipment (384) Transport equipment (385) Professional and scientific equipment (41) Electricity, gas and heat production Total expenditure for industrial 1. 694 technology imports

1963–1968

1971–1978

1979–1990

1963–1990

21/4.2

2/0.8

5/0.9

175/2.5

182/2.1

17/29.2

13/33.7

22/26.3

272/18.1

307/20.2

–

–

–

24/0.2

24/0.1

19/7.6

7/5.7

32/12.6

981/7.0

1020/8.4

10/3.0

9/5.4

9/2.5

860/7.7

878/6.5

2/1.4

9/5.4

3/0.4

365/3.3

377/2.6

1/0.03

84/0.6

85/0.5

4/2.0

131/2.1

135/2.1

3/3.6

1/3.4

10/1.0

256/7.8

267/6.1

3/0.6

1/0.3

–

207/0.6

208/0.5

66/28.7

3/8.1

13/4.1

114/23.7

130/18.9

1.694 billion USSR roubles

0.266 billion US dollars

7.728 billion US dollars

24.319 billion US dollars

32.313 billion US dollars

Source: SPC et al. (eds) 1992:6, 7, 16, 40, 53, 83 and 84. Notes: 1 All expenditures are in current prices. 2 Technology imports for agriculture, transportation, post and telecommunications, and some other fields are not included. 3 The source does not explain the breaks in the time series of data (1960–1962 and 1969–1970).

MACHINERY INDUSTRY DEVELOPMENT

107

Table 14.11 Technology import agreements, by category (1979–1990) Branch (ISIC code)

Number of agreements by category Total

I

II

III

IV

V

(382) Non-electrical machinery (383) Electrical machinery, excluding electronic communication equipment, computers and household electronic equipment (384) Transport equipment (385) Professional and scientific equipment Total for machinery industry Total for all industrial technology imports

655 167

65 2

127 15

73 44

60 51

980 279

144 162 1,128 1,780

12 9 88 402

49 10 201 229

32 17 166 1,187

19 8 138 704

256 206 1,721 4,302

Source: SPC et al. 1992:80, 185–278 and 316–328. Note: The technology import agreements are differentiated in five categories: I technology licensing; II consultant and technical services; III cooperative manufacturing, usually associated with cooperative design; IV key equipment imports; and V imports of entire production lines, i.e. turn-key engineering.

spectrum, while exports are predominantly of more conventional vintages, as can be seen from the wide gap between imports and exports in terms of average unit value, and the shares of CNC machines in imports and exports. China is clearly faced with critical challenges. There is an increasing demand for CNC machine tools from various manufacturing sectors, while domestic production remains oriented to rather traditional technologies. Technology imports increased dramatically from the 1970s onwards, with the ending of closure to international technological exchange. Expenditure on technology imports between 1979 and 1990 was double that over the preceding thirty years, as illustrated in Table 14.10. In the period 1979–1990, machinery technology imports under 1,748 agreements accounted for 17.3 per cent of the total expenditure for technology imports, or 4.2 billion US dollars. The categories of electronics separated from ISIC 383 accounted for another 1.46 billion US dollars (6 per cent of technology import expenditure) and 580 agreements in 1979–1990. The technology imports take various forms, with technology licensing, cooperative manufacturing, and the import of key equipment being especially significant, as can be seen from Table 14.11. Turn-key engineering is now regarded as a second-best option for technology import agreements, especially for the machinery industry. The technologies employed in the machinery industry have been substantially renewed through these imports, which are regarded as the most important factor in the modernization of the industry (Sheng Shuren (ch.ed.) 1991:94–98). Institutions for technological change in the machinery industry This section will describe the evolutionary path of the machinery industry’s plant design and R&D institutions from the 1950s to the 1970s. It covers two kinds of institutional establishments: plant design institutes charged with capital investment tasks, and R&D institutes for the development of product and manufacturing technology. However the focus will be on the R&D institutes, and on issues which have had a significant impact on the current transformation of the institutional structures.

108

MACHINERY TECHNOLOGY INSTITUTES

Plant design institutes and capital investment Founding and early experience The plant design institutes of the Chinese machinery industry, like those in other Chinese industrial sectors, were initially founded in the early 1950s ‘to deal with the tasks of capital construction set out in the First Five Year Plan’ (Jing Xiaocun (ch.ed.) 1990: Vol. B, 87). They were established by central decision,14 and with the central mobilization of resources. Scarce manpower, in some cases trained in Europe and America, was quickly mobilized from enterprises and universities and other sources into the newly-established institutions. During 1953, seven plant design institutes were established, and 1,037 experienced persons were transferred from enterprises and another 780 from universities and schools. By the end of the year, the design institutes had a combined staff of 4,361 persons, and the engineering survey institutes had 1,106 (Jing Xiaocun (ch.ed.) 1990: Vol. B, 88–89; Vol. A, 18). Plant design institutes, together with ‘civilian engineering survey institutes’, were in effect the technical agents of the capital investment institutions established in the early days of the centrally planned regime in China. In addition to these technological agents, there was an administrative network in charge of capital investment, operating within the Planning Department and the Capital Construction Department of the Ministry of the Machinery Industry. Through these two ministry departments, the State Planning Commission and the State Capital Construction Commission15 exerted their authority on behalf of the central planning system (Yuan Baohua (ch.ed.) 1985: Chapter 10; Jing Xiaocun (ch.ed.) 1990: Vol. B, Chapter 17). Plant design was learned in the early 1950s through technology imports from the Soviet Union and Eastern Europe. Turn-key plant construction was the major form of technology transfer, with project implementation being supported by Chinese designers from these newly-established plant design institutes, who contributed surveys of the engineering environment, collected background information, translated the masses of technical documents and blueprints, and made some modifications to original designs to adapt them to local conditions. Much was learnt in the process of doing, but there were also intensive training courses, taught by Soviet experts, in both the Chinese design institutes and Soviet factories and design institutes (Jing Xiaocun (ch.ed.) 1990: Vol. B, 89). Soviet consultants were also ubiquitous in the Chinese design institutes (Jing Xiaocun (ch.ed.) 1990: Vol. B, 89– 91), leading to the adoption of the Soviet management system for plant design which was followed for the next twenty years (Jing Xiaocun (ch.ed.) 1990: Vol. B, 94–95; SSTC 1985b:117, 213). Expansion of industrial capacity: the 1960s and 1970s From their first establishment until the start of the current reforms, plant design institutes contributed to the enormous expansion of capacity in the Chinese machinery industry, which was realized through the construction of new plants designed by these institutes. In the first five year plan (1953–1957) new investments were concentrated on power-generating equipment, transportation equipment, metallurgical and mining equipment and machine tools. In the late 1950s attention shifted to the metallurgical and mining equipment sectors. Eighteen key factories for metallurgical and mining machinery were established, and an independent sector in this field had taken shape by 1960 (Cheng 1972:137–139; Jing Xiaocun (ch.ed.) 1990: Vol. A, 192–193). From 1961 to 1966, the emphasis in plant construction was on chemical and petroleum equipment. Medium-size fertilizer plants with capacities of 25,000–50,000 tonnes per year, and small plants for 3,000–

MACHINERY INDUSTRY DEVELOPMENT

109

10,000 tonnes per year, were developed during this period (Jing Xiaocun (ch.ed.) 1990: Vol. A, 258–262; Cheng 1972:139). Petroleum refining plants were developed in response to the rapid growth in oil production in the 1960s, with a standardized capacity of 1–1.5 million tonnes of oil per year. Later in the 1970s this was increased to 2.5–3 million tonnes per year (Jing Xiaocun (ch.ed.) 1990: Vol. A, 272–274; Cheng 1972:141). Agricultural machinery16 was also strongly emphasized, in accordance with the goal of mechanizing agricultural production which was announced in the second half of the 1950s. Between 1958 and 1965, investment in agricultural machinery reached its highest level as a proportion of total investments in the machinery industry, accounting for 15 to 18 per cent. Investment in agricultural machinery remained a priority in the 1970s (Jing Xiaocun (ch.ed.) 1988:24–28; China Statistical Yearbook 1987: 230). The machinery production capacity also expanded geographically into the hinterland area of south-west China between 1965 and 1979, when more than half of the entire investment in the machinery industry went to the hinterland provinces (Jing Xiaocun (ch.ed.) 1990: Vol. A, 64). The management of plant design Plant design was managed in the context of ‘taut’ investment planning. ‘Taut’ investment planning, according to Kornai (Kornai 1980: Chapters 9 and 10), is one of the central features of capital investment under a centrally planned regime, where it reflects the ‘expansion drive’, resulting from the dominant behaviour of investors. Because investment is not limited by fear of financial loss or failure in such a regime, the investors push the economy to the limit of available resources. Under a permanent taut investment regime, there is little sensitivity or capacity to adapt to either radical technological changes or complementary improvements. In the Chinese case, taut investment planning was a central characteristic of the planned regime, and drove it to an emphasis on quantitative expansion to the neglect of qualitative aspects of investment. Duplication of imported technology, in a large part due to the quantitative expansion, was a major source of technology and the primary work of the plant design institutes in the 1960s and 1970s. This was especially true during the early years of the regime, and especially in 1958–1960, when ‘plant design institutes had to disseminate their designs by printing and circulating (simplified) “stylized” blueprints of various machinery plants to meet the massive demands from numerous investors’ (Jing Xiaocun (ch.ed.) 1990: Vol. B, 93). The sacrifice of investment quality owing to this quantitative drive can be seen from the very large fall in productivity in that period, shown in Table 14.4. Throughout the decision-making and implementation of capital investment the key players were the administrative machinery of planning and the plant design institutes, which acted as technical agents. User enterprises had little involvement. As a result, the experience accumulated in plant construction (learning by doing) was not widely communicated. The long-distance coordination among the organizational establishments delayed and hindered its transmission. In addition, because the government manages industrial investment directly in such an institutional structure, it was not possible to use selective intervention to promote enterprise learning by ‘punishing poor performers and rewarding only good ones’ (Kim 1993:357–383). From 1964, plant design institutes were required to give more attention to adopting new technology, through visiting production enterprises and absorbing their useful experiences into new plant designs (Jing Xiaocun (ch. ed.) 1990: Vol. B, 94–95). The Chinese media called this a ‘design revolution’. The revolution’s achievements were limited: since innovating enterprises were denied the possibility of investing on their own initiative, their experiences could only be implemented in a new plant, if at all, after a long waiting period during which the ‘capital construction project’ was being formulated.

110

MACHINERY TECHNOLOGY INSTITUTES

As the technical agents in capital investment, the operation of the plant design institutes was closely related to the planning procedure. ‘Project specification (jihua renwushu)’ was one of the jobs assigned to the institutes. Their specifications served as technological certificates for investment planning decisions. Once a project was finally approved, the design institutes would go on to undertake plant design work and produce ‘construction blueprints (shigong tu)’ that were used as technical instructions for plant construction in implementing the investment (Jing Xiaocun (ch.ed.) 1990: Vol. B, 77). The procedure was thus technology-oriented, with no serious attention given to the economic return on a particular investment. It was not until the 1980s that feasibility studies and project assessment were introduced as part of plant design work, following visits by plant design experts to France and other European countries (Jing Xiaocun (ch.ed.) 1990: Vol. B, 129). It was possible to obtain technological inputs into new plant designs from several sources (Jing Xiaocun (ch.ed.) 1990: Vol. B, 86–131):1) information about the best practice of domestic plants, from visits to existing enterprises; 2) information about advances in other countries gained from technological imports and/or visiting the countries concerned; 3) information embodied in new equipment available domestically or abroad; 4) the domestic development of new equipment in R&D institutes, sometimes at the suggestion of a plant design institute (such links are considered in more detail below); and 5) some experiments conducted in the plant design institutes themselves. In short, the plant design institutes were the gateway through which new elements of technology, from whatever sources, were channelled into productive lines. It appeared that the first of these sources, i.e. learning from experience in plants of similar design, was most important in the 1960s and 1970s, which were characterized by duplicative expansion. The expansion and reform of plant design institutes The plant design institutes developed as the administrative investment planning system was elaborated. It is reported that by 1985 the Ministry of the Machinery Industry had one General (plant) Design Institute at the central level and eleven plant design institutes specialized in plants for particular sectors,17 employing a total of 9,000 staff including 3,300 engineers (Jing Xiaocun (ch.ed.) 1990: Vol. B, 51, 100–103). There were another 135 units, with 6,000 employees, engaged in plant design and engineering survey (Jing Xiaocun (ch.ed.) 1990: Vol. B, 101–102). Many of these were affiliated to local administrative bodies of the machinery industry (Jing Xiaocun (ch. ed.) 1990: Vol. A, 30, 58). Some were integrated in a large enterprise or a locally affiliated R&D institute, as illustrated by the cases of the Dalian Machinery and Electrical Research and Design Institute and the Zhejiang Mechanical and Electrical Design and Research Institute, both of which are examined in Chapter 16 (Case Texts 16.7 and 16.8). Reforms of the plant design institutes and the capital investment management system started in the late 1970s and have produced some important changes (Jing Xiaocun (ch.ed.) 1990: Vol. B. 57). First, decisionmaking has been decentralized, and is now shared by central and local governments and enterprises to varying degrees, according to the project cost. Second, financing for capital investment is now provided to user enterprises mainly through bank loans, rather than being granted in the government budget. Third, a contract system with rewards and penalties has been introduced for the implementation of capital investment projects, to bind both user enterprises and plant design institutes. Fourth, plant design has been provided on a paid basis since 1983, and an obligatory competitive tendering system established for the provision of engineering design services. The main response of the plant design institutes to the market-oriented reform has been to move towards vertical integration (Jing Xiaocun (ch.ed.) 1990: Vol. B, 104). On the one hand, the institutes have been extending their function ‘backwards’, i.e. to the pre-design stage, with engineering consultancy services

MACHINERY INDUSTRY DEVELOPMENT

111

such as plant feasibility studies, project assessment, and investment planning. On the other hand, they have been extending ‘forwards’, to the post-design stage, with the provision of engineering service such as the supervision of engineering work, equipment purchases, installation and adjustment, and so on. The result is something like a turn-key plant investment service. The transformation of plant design institutions has not yet become a focus of reform policy in China. The transformation which has occurred has not be well reported and analysed, and it is rather difficult to learn from experience in other countries, partly because the plant design organization described above was largely confined to the former centrally planned regimes.18 This does not mean that reform of this system has not proceeded. In fact integrated plant engineering services have constituted a large part of China’s technology exports recently, although they are still limited in terms of technological sophistication and value. In 1993 there were 396 packaged technology export projects, worth 1.79 billion US dollars, which accounted for 90. 9 per cent of China’s total earnings from technology exports in that year. Major exports of this kind include complete sets of equipment for power generation and cement and glass production, in which plant design institutes obviously played an important role (China Science and Technology Indicators 1994:98–99).

R&D institutes: early developments Founding: mid-1950s The establishment of R&D institutes in the machinery industry began in 1956, several years after the plant design institutes, as part of the nationwide campaign of ‘marching towards modern science and technology’. This campaign was initiated under the first five year plan (1953–1957), when the goal of technology imports and industrial construction was a conspicuous success. The intention was to invite modern science and technology to play greater roles in the further development of the country. The second half of the 1950s is therefore reasonably seen as the period when China instituted its modern science and technology system. In addition to industrial R&D institutes in many industrial sectors, most influential national S&T institutions, such as the Chinese Academy of Sciences (CAS), the Chinese Academy of Agricultural Sciences (CAAS) and the Chinese Academy of Medicine Sciences (CAMS) were founded or rehabilitated during this period. In the 1950s many R&D institutes were being established throughout the world: there was a belief in the ‘far-reaching importance of the new technology revolution’ (Zhou Enlai 1956:181– 182).19 In China at that time, the highest emphasis was given to advanced and theoretical S&T. This may be illustrated by a key policy speech delivered by Premier Zhou Enlai at an important conference in January 1956. He said that ‘the most excellent scientific manpower and the best university graduates should be organized to work in scientific research in the Chinese Academy of Sciences’, and that ‘each ministry of the government…should establish and strengthen its research institutions, sharing responsibilities with the Chinese Academy of Sciences, to provide the most advanced technology’ (Zhou Enlai 1956:185). In fact the technological efforts for the machinery industry were very much restricted to the industry itself rather than being shared by the Academy as originally expected. The central task of the R&D institutes established in the machinery industry was ‘to promote product design and processing technology’ (Jing Xiaocun (ch.ed.) 1990: Vol. B, 242). In addition, ‘some more generic work’ should also be covered, including: 1) design techniques; 2) standardization, systematization, and testing; 3) new manufacturing technology, new materials, and new processes. All these tasks were seen as necessary to achieve the central purpose ‘to make the industry capable of independent design’.20

112

MACHINERY TECHNOLOGY INSTITUTES

Three types of R&D institute were quickly set up: 1) a few central institutes engaged in generic manufacturing technology (known as ‘generic technology’ (gongxin jishu), or ‘comprehensive technology’ (zonghexin jishu); 2) about twenty central institutes, in charge of product design; 3) design offices, standardization offices, or testing laboratories in 105 ‘key’ enterprises which were directly subordinated to the central ministry (Jing Xiaocun (ch.ed.) 1990: Vol. A, 25; Vol. B, 243). As in the capital investment system, the functions of the R&D institutes were to be administratively coordinated, specifically by two lines of ministry departments: the Technology Department, later called the Department of Science and Technology, and bureaux specializing in the affairs of particular sectors or sub-sectors. The ‘sector bureaux’, specialized by product categories, actually managed both production and technology in a certain subsector. The system at this stage closely resembled the Soviet system in every respect, from the relationship between government bodies and R&D establishments to the structure of R&D units. For instance, the central institutes engaged in generic manufacturing technology, such as the Research Institute for Machinery Manufacturing Science (jixie zhizao kexue yanjiu yuan), the Research Institute for Processing and Production Scheduling (gongyi yu shengchan zuzhi yanjiu yuan), and the Research Institute for Tools and Tooling Science (gongju kexue yanjiu yuan), were all established according to Soviet advice (Jing Xiaocun (ch.ed.) 1990: Vol. A, 25; Vol. B, 243), and all had their equivalents in the Soviet system (see Shi Jianzhong 1992; Amann and Cooper 1982:41–42, 95). Long-term and annual planning Long-term National Science and Technology Development Programmes and annual plans were the major planning instruments for national scientific and technological work. Long-term National S&T Development Programmes were used to outline medium-term development objectives, and the machinery industry was included as part of the programme.21 The first National S&T Programme covered the period 1956–1967; the second was for 1963–1972, and the third for 1978–1985. The National S&T Programmes analysed trends in relation to science and technology on both supply and demand sides, and set objectives and guidelines for various sectors, which would in turn be formulated in ‘scientific and technological projects’. One characteristic of the Chinese planning system was the strong reliance on internal consensus. Drawing up the programme involved wide participation from politicians, industrial ministry managers, and scientists and technologists (but little participation from enterprise managers). The process was intended to achieve not only information exchange but also the cross-sectoral coordination which was often required in national programmes. Repeated communication, both top-down and bottom-up, went into shaping the final conclusions, although the decisions were finally taken at the top (Luo Wei 1983). The first National S&T Programme contained fifty-seven projects, of which seven were directed at machinery technology and were entrusted to the Ministry of the Machinery Industry for execution. These projects centred on new product development in heavy machinery, precision machinery, automated machine tools and some complete sets of industrial apparatus (Jing Xiaocun (ch.ed.) 1990: Vol. B, 243). Actual product development was then organized within the industry, although the planned programme suffered some disruption during 1958–1960. Under the second National Programme, the Ministry worked out a ‘1963–1972 S&T Development Programme for the Machinery Industry’, which set the industry the target of developing 10,000 new products, and thirty seven sets of equipment for the metallurgy, chemicals, petroleum, electrical power and defence industries (Jing Xiaocun (ch.ed.) 1990: Vol. B, 244). Implementation was again disrupted to some extent, this time by the ‘Cultural Revolution’ (1966–1976). After the Cultural Revolution, the third national programme (1978–1985) included computer-aided

MACHINERY INDUSTRY DEVELOPMENT

113

manufacturing technology and energy conservation technology (Jing Xiaocun (ch.ed.) 1990: Vol. B, 246– 247). China’s good performance in energy conservation since the 1980s has already been noted. Although the long-term programmes laid out guidelines, it was the annual plans which transformed the guidelines into implementation plans. These were structured around ‘working projects’ which were granted the required funds. Two types of working project may be distinguished: numerous projects for incremental adaptations, and some projects for relatively radical development. The first type was routinely financed, from sources known as the ‘technical measures fee’ and ‘new product trial manufacturing fee’. The funds from these fees were to be used for maintaining ‘simple re-production’.22 The second type of project was financed by ‘special funds’, approved on a case-by-case basis, and financed from the national budget. Projects of the first type were subsumed under the routine planning management, while those of the second type were often coordinated by a specially assigned higher-level leadership. Since the first half of the 1980s, some projects of the second type have been organized and financed as ‘Key S&T Projects in the Five Year Plan’ (Part 1, Chapter 2). A management procedure for product innovation Since the second half of 1950s, the Chinese machinery industry has acquired the capacity to manage technological innovation. This implies an upgrading from the simple copying stage, when production was simply and entirely based on imported blueprints. Two key features in the management of the product innovation process were the introduction of a conceptual scheme of the innovation process and the promotion of standardization. If the industry was to cope with continuous changes, a concept of the processes by which new product designs would be developed and adopted and new products brought into production was required. It is reported that this was introduced at the urging of Soviet experts (Jing Xiaocun (ch.ed.) 1990: Vol. B, 260). In the case of ‘new product design’, the management procedure identified three steps in the process: preliminary design (chubu sheji), technical design (jishu sheji) and blueprint design (shigongtu sheji) (Jing Xiaocun (ch.ed.) 1990: Vol. A, 148). In the case of bringing new products into normal production, the management standards also identified three steps: making the prototype (yangji shizhi), small batch trial manufacturing (xiaopi shizhi) and beginning normal large-scale production (chengpi shengchang) (ibid. Vol. A, 148). Amann and Cooper report that in the Soviet system the regulations for the introduction of a new product in the early post-war period covered six stages: ‘technical specification’, during which the main requirements and justification of a new design were outlined, a ‘technical project’ which covered the main design work, a ‘working project’ culminating in trial manufacture, and ‘prototype manufacturing’ to test and approve the design for ‘batch production’, followed by ‘normal production’ (Amann and Cooper 1982:44). This confirms the close resemblance between the Chinese and Soviet innovation management patterns, both of which roughly replicate the chronological development of an innovation. Some such regular scheme was required, under a centrally planned regime, to facilitate the upgrading of the infant industry. The functions of designing a new product and bringing the new design in production, which were organized in separate units, were linked by a coherent conceptualization of the innovation process, and its management implications, and then by planning coordination. The management schemes thus reduced coordination costs sufficiently to enable the young Chinese machinery industry to realize some innovations. It was not a very adequate solution to the problem of communication at the interface between design and production, however. This problem arises mainly from the organizational separation of these functions. In the traditional Soviet system (from the late 1940s to the 1960s), solutions relied largely on the elaboration of approval procedures using complicated technical documentation (Amann and Cooper 1982:

114

MACHINERY TECHNOLOGY INSTITUTES

42, 45). In contrast, solutions in the traditional Chinese system (from the late 1950s to the late 1970s) tended rather to promote inter-organizational activities on the part of designers working in enterprises. For instance, the prototype of a new design might be made in a production enterprise under the direction of designers from an R&D institute, rather than in the experimental workshops of the designing institute, as in the Soviet system. So even before the current reforms, Chinese designers were used to taking their design blueprints to productive enterprises and working there, often for several months per year (Interviews, 1994– 1995 field survey). Compared to the Soviet system (Amann and Cooper 1982:53, 94–95), the Chinese industrial R&D system seems to have been less isolated from firms’ production, although this was realized at the expense of placing strong pressure on designers to work in factories. But neither the Soviet nor Chinese solutions could entirely overcome the shortcomings inherent in their planned regimes. The second key feature of innovation management was reliance on standardization, which assumed greater importance from the second half of the 1950s. Standardization was intended ‘to augment the capacity to organize production, accelerate new product development and shorten the lead-time for bringing newly developed products into production’ (Jing Xiaocun (ch.ed.) 1990: Vol. B, 277). The major task in standardization was the formulation and compilation of series of standards specifying the type and size of products (chanpin xilie xingpu). These standards contained a detailed categorization of machinery products. Information on a particular type of machine could then be clearly codified. Given these standards, incremental innovations to machinery products could be incorporated in the routine of planning management. The coordination of ordinary production was also guided by the ‘type and size of product’ standards: existing models and the new models to be brought into production were all compiled in, and guided by the standards series (interview at the Beijing No. 1 Machine Tool Plant, October 1994). Once again, this system paralleled the Soviet system (Amann and Cooper 1982:44) which had emerged in the 1930s but was not effectively implemented until the late 1940s. By greatly reducing the cost of communicating producer information, standardization did have the effect of accelerating product diversification and the widespread diffusion of product and processing information. In the late 1950s and early 1960s, standardization for electric motors, transformers, switches and electric ceramics is reported to have accelerated new product design and production in these sectors; similar benefits were achieved for machine tools, pumps, industrial fans, bearings, workbench fixtures, and so on (Jing Xiaocun (ch.ed.) 1990: Vol. B, 277–278). Standardization also facilitated the widespread dissemination of a more modern type of producer information throughout the industry while it was still in the process of moving from repair and maintenance roles to production. But these benefits came at the cost of locking the industry into very rigid technological norms. The newly established R&D system was assigned a central role in standardization. At first only some R&D institutes were nominated as ‘standard centres’. In 1963, having built up some experience from the work of these institutes, the Ministry of the Machinery Industry decided to disseminate this experience to the other R&D institutes in the machinery industry and make all of them officially responsible for standardization. They were assigned the function of ‘organizing technological work for the industry’ (hangye jishu zuzhi gongzuo) (Jing Xiaocun (ch.ed.) 1990: Vol. B, 292), which meant in effect carrying out technical work for innovation planning and production planning, for which standardization offered a necessary basis. With the 1963 decision, the R&D system was functionally integrated into the planning administration, especially because of its responsibility for standardization. Standardization was organized by product, and since the 1963 decision R&D institutes, which provided the technical basis for standardization, have generally grown to be more precisely specialized by product. Productive enterprises, under the centrally planned regime, were already specialized and organized by product lines. Centrally planned regimes generally place greater weight on product indicators, and less on

MACHINERY INDUSTRY DEVELOPMENT

115

other information, such as prices. But it would appear that the Soviet system relied less heavily on standardization as an instrument in innovation planning, and did not implement standardization rigorously, even after the Second World War (Granick 1967:113). Granick suggested several reasons for this, including substantial Soviet imports of machinery products, especially in the 1930s, the simultaneous use of imperial and metric measurement systems, the lack of emphasis on standardization in planning in the prewar period, and producers’ monopolistic position in a certain category of product, so that standardization had only minor value (Granick 1967:43–47). The last of these highlights an important difference in enterprise structures between the two systems. The Chinese machinery industry is composed of less monopolistic enterprises, because of the decentralization of planning authority in 1958–1960 and in 1966– 1972. For instance in 1980, a typical ISIC 38 factory in the Soviet Union had a staff of 1,800, while its Chinese counterpart employed just 120 people. Although the ISIC 38 industries in the two countries had virtually the same number of workers (13.3 m. for China versus 14.7 m. for the USSR), China had 107,000 enterprise units, thirteen times as many as the USSR’s 8,000 (Guo Rui (ch.ed.) 1994:12). The much more broadly decentralized enterprise structure in China provided a stronger rationale for more determined standardization to reduce planning management costs. R&D institutes: expansion and further development As a result of the great efforts in the development of the machinery industry under the Chinese centrally planned regime during the 1960s and 1970s, the institutional framework for machinery technology R&D expanded considerably, and with some unique characteristics. This section will describe these developments in some detail, since they are very relevant to the focus of the current study. The R&D institutes will be considered under three headings: centrally affiliated institutes, locally affiliated institutes, and institutions with other relationships. Of these, the centrally affiliated will be considered in most detail because these played an important role in the technological changes in the industry. Rather than describe the process of development during these two decades in detail, this section will in most cases give a picture of the situation around 1984–1985, when the current reform programme was about to start.23 Centrally affiliated R&D institutes ‘Centrally affiliated’ R&D institutes were commissioned and financed by the central planning administration, i.e. the Ministry of the Machinery Industry. These institutes can be subdivided into a few which specialized in manufacturing technology and a greater number which specialized in product technology. The first group was supervised by the Department of Science and Technology and the second by bureaux within the Ministry of the Machinery Industry specializing in ‘sectoral affairs’. The organizational chart in Figure 14.1 shows the two groups of centrally affiliated institutes and the Ministry departments which supervised them. The chart shows only two of the eight sectoral bureaux which operated for most of this period (from the 1950s to the 1980s). The eight bureaux covered 1) parts and components for general use; 2) machine tools and other tools; 3) general industrial equipment; 4) heavy and mineral machinery; 5) electrical equipment; 6) instrumentation; 7) agricultural machinery; and 8) automobiles (Zhu Sendi 1994). The bureaux were assigned responsibility for all issues relating to the productive operation of enterprises in their sector. They can therefore be regarded as production coordination departments of the Ministry, in charge of both production planning and planning for product innovation. This contrasts to the Department of Science and Technology, which provided a particular service to the machinery industry as a whole, and can be regarded rather as a resource coordination department. It is useful to note that it was the

116

MACHINERY TECHNOLOGY INSTITUTES

Figure 14.1 Organizational chart of Ministry departments and entirely centrally affiliated R&D institutes RIMST=Research Institute for Machinery Science and Technology.

production coordination departments, and not the resource coordination departments, which had administrative power extending down to production enterprises in the period of the traditional planned economy. The centrally affiliated R&D institutes can also be subdivided into those which were entirely centrally funded and directed (’A’ institutes) and those which were part of a host enterprise and only partly funded and directed by the Ministry and its bureaux (’B’ institutes). The B institutes, created by Table 14.12 Government-run R&D institutes for the machinery industry in China in the mid-1980s Affiliation

Central government

entirely centrally commissioned and financed (group I institutes) 53

product technology (‘A’ institutes) partly centrally product technology commissioned, (‘B’ institutes) enterprise-based (group II institutes) Local governments (group III institutes)

Functional assignment Number of institutes

Scientific and engineering personnel

manufacturing technology

4,000

8

18,000 138

10,000

technology diffusion, technological support to local enterprises, some product

493

24,000

MACHINERY INDUSTRY DEVELOPMENT

Affiliation

Functional assignment Number of institutes

117

Scientific and engineering personnel

development for local specificities Source: Jing Xiaocun (ch.ed.) 1990: Vol. B. 284–285. Note: The table covers R&D establishments under the jurisdiction of the central Ministry of the Machinery Industry or the bureaux of the machinery industry of provincial or municipal governments. Institutes for electronic and telecommunication equipment and the aircraft and shipbuilding industries are not included.

commissioning an enterprise’s in-house laboratory to perform centrally directed research, was most important in enabling the centrally affiliated system to specialize by products. Table 14.12 summarizes the government-run R&D institutes for the machinery industry, in terms of affiliation, functional assignment, number of units and personnel. CENTRALLY AFFILIATED INSTITUTES FOR MANUFACTURING TECHNOLOGY

The centrally affiliated institutes originally assigned for manufacturing technology are known as institutes for ‘generic’ or ‘comprehensive’ technology, but are in fact engaged in R&D relating to manufacturing technology, as can be seen from the description of their objectives in Case Text 14.1. The eight institutes in this group are supervised by the Department of Science and Technology of the Ministry of the Machinery Industry. Six of these were among the earliest R&D institutes to be established, and of the two which were ‘established’ in the 1970s, one institute for manufacturing automation was evolved in 1976 from a previously existing institute, and the other was a previous standardization department which was redesignated as an independent institute in 1975. The eight institutes of this group collectively comprise the Research Institute for Machinery Science and Technology (RIMST). By 1992, the RIMST had a total of 6,671 staff, two thirds of them being technological professionals. The RIMST has the highest level of S&T manpower of the entire machinery technology R&D system. It is also the sole pool for the best research and development experience in manufacturing technology (in contrast to product technology). The eight institutes are listed in Case Text 14.1. Of these, we have chosen three institutes, the Shanghai Materials Research Institute (SMRI), the Beijing Research Institute for Mechanical and Electrical Technology (BRIMET) and the Beijing Research Institute for Automation in the Machinery Industry (BRIAMI), to visit and use as illustrations in our study.

CASE TEXT 14.1 THE RESEARCH INSTITUTE FOR MACHINERY SCIENCE AND TECHNOLOGY (RIMST) The RIMST originated from two of the ‘generic technology’ institutes which were established in the 1950s: the Research Institute for Machinery Manufacturing Science (jixie zhizao kexue yanjiu yuan), and the Research Institute for Processing and Production Scheduling (gongyi yu shengchan zuzhi yanjiu yuan) which has already been mentioned. These two predecessor institutes were combined under the present name in 1958. The combined RIMST had four subordinate research units at that time. In 1969, during the ‘Cultural Revolution’, the RIMST was moved to Zhengzhou in Henan Province, and renamed as the Zhengzhou Research Institute for Machinery Engineering (ZRIME), while the portion of the old RIMST which was not relocated was reorganized as the Beijing Research Institute for Mechanical and Electrical Technology (BRIMET). In 1978 the previous name of RIMST was restored, covering the few institutes born of the relocation and still surviving.

118

MACHINERY TECHNOLOGY INSTITUTES

In 1978 and again in 1985, the objective of the RIMST was specified by the Ministry of the Machinery Industry as research and development of ‘new, fundamental and generic’ machinery technology. The meaning of ‘new, fundamental and generic’ can be seen from the task descriptions of the RIMST’s component institutes below. By 1992, the RIMST had a total staff of 6,671, of whom 65 per cent, or 4,347, were technology professionals. The component institutes in 1985 were as follows:

1

Harbin Research Institute for Welding (HRIW) (1956), engaged in comprehensive research and development in welding technology, is the sole institute in this field in China. 2 Shenyang Research Institute for Foundry Technology (SRIFT) (1956), engaged in research and development in foundry technology and materials, including casting processes, melting processes, alloy castings, moulding materials, and related testing and instrumentation. 3 Shanghai Materials Research Institute (SMRI) (1956), engaged in research, development and application of materials for the machinery industry, including various mechanical engineering materials and their applications and testing. 4 Wuhan Research Institute for Material Protection (WRIMP) (1956), engaged in research and development in surface engineering, including surface protection, metallic corrosion and tribology. 5 Zhengzhou Research Institute for Machinery Engineering (ZRIME) (originally established 1958, redesignated in 1969), engaged in research and development in the structural strength and vibration of machinery, and gear transmissions. 6 Beijing Research Institute for Mechanical and Electrical Technology (BRIMET) (originally established 1958, redesignated in 1971), engaged in research and development in forging and stamping, heat treatment, and forging dies for the machinery industry. 7 Beijing Research Institute for Automation in the Machinery Industry (BRIAMI) (derived from the BRIMET in 1976), engaged in the development of computer-integrated manufacturing technology, including computer applications software, automatic control and application engineering, industrial robots and various special-purpose components and devices. 8 Research Institute for Standardization in the Machinery Industry (RISMI) (1975), engaged in standardization and measurement technology for the machinery industry, and responsible for technical planning and coordination in this field for the machinery industry. Sources: Introductory material: ‘Research Institute for Machinery Science and Technology (RIMST)’, provided by the RIMST, 1994; Shi Jianzhong (research team head): An Analysis of RIMST Institutes: Their Scientific and Technological Strengths and Policy Recommendations for Further Development, Mimeo, Beijing, July 1992; Interview with Mr Fang Guiru, October 1994.

CENTRALLY AFFILIATED INSTITUTES FOR PRODUCT TECHNOLOGY

The second, and much larger, group of centrally affiliated institutes were originally assigned to the development of product technologies. Although institutes of this type were created in the early days of the development of the machinery technology system, this group expanded and developed significantly in the 1960s and 1970s. The particularly Chinese characteristics of the machinery technology system are largely associated with institutes of this group.

MACHINERY INDUSTRY DEVELOPMENT

119

Of this group, fifty three institutes which we will call the A institutes were entirely commissioned and financed by the central ministry, and another 138 institutes were partly commissioned and financed by the central ministry but were also partially financed by, and serving, a host enterprise. These will be called the B institutes. By 1985, 18,000 technological professionals were working in the fifty three A institutes, and another 10,000 in the 138 B institutes. The major difference between A and B institutes is the relationship of the latter to a host enterprise. Most B institutes relied on their host enterprise for physical accommodation and administrative matters, contributed a large part of their services to the host enterprise, and received much of their finance from it. B institutes also had duties assigned to them by the Ministry of the Machinery Industry, and received funds from the Ministry in proportion to these duties. These duties fell under the heading of ‘organizing technological work for the industry’, and centred on providing technical support to both innovation planning and production planning.24 As mentioned, this required them to develop and rely on standardization. Most B institutes were specialized by product, in accordance with their standardization work, which was also organized along product lines. Furthermore, most B institutes originated as an in-house laboratory in their host enterprise, and the host enterprises were also specialized by products, as was common in the former centrally planned regimes. The B institutes, in other words, were created by the central Ministry commissioning their establishment in a number of leading enterprises. This extended the R&D and innovation system and enabled the centre to provide effective coordination, at less cost, to innovations for the huge but still underdeveloped Chinese machinery industry. The origin of the B institutes as in-house laboratories within enterprises has meant that they have been directly owned by the host enterprise, implicitly in the pre-reform period and more explicitly under the current reform, and ultimately they are owned by the local governments. In contrast, the A institutes were owned by the central government, exerting its control through the Ministry. To further clarify the characteristics of A and B institutes, two sectors of the machinery industry have been chosen for an in-depth institutional analysis with respect to the functions of the institutes and the relationships among them, and between the institutes and the responsible departments of the Ministry. The two sectors chosen are 1) the machine tools and tools sector and 2) the electrical equipment sector. The size of these two sectors can be indicated by noting that they comprise seven and sixteen A institutes, respectively, of the fifty three A institutes in the industry, and forty and nineteen B institutes, respectively, out of the 138 B institutes in the industry. PRODUCT TECHNOLOGY R&D IN THE MACHINE TOOLS AND OTHER TOOLS SECTOR

Under the supervision of the Bureau of Machine Tools and Other Tools, seven A institutes and forty B institutes were established in this sector by the mid-1980s. A hierarchy-like order developed among the A and B institutes. The A institutes tended to specialize in more general objectives of product development, covering a greater range of product categories, while most of the B institutes received technological guidance from an associated A institute. That guidance was provided by the A institute on behalf of the Bureau of Machine Tools and Other Tools. The sectoral bureau subdivided its sector into eight product sub-categories: 1) machine tools, 2) foundry machinery, 3) forging machinery, 4) woodworking machinery, 5) cutting and measuring tools, and measuring instruments, 6) abrasive and grinding apparatus, 7) accessories and built-up jigs, and 8) machinetool electric devices. Both the planning management of the sector Bureau and the establishment of R&D institutions for the sector were organized largely along the lines of this product categorization. Some of these categories were further subdivided, so that the product category of ‘machine tools’, for example, includes institutes specializing in ‘milling machines’ and ‘grinding machines’. This points to the very low

120

MACHINERY TECHNOLOGY INSTITUTES

level of ‘product variety’ in the ‘product technology’ R&D system. The B institutes were rather strongly specialized by product variety. It was the creation of the B institutes that provided the central innovation planning system with the technological agents at the most elementary level which it required.

CASE TEXT 14.2 CENTRALLY COMMISSIONED AND FINANCED PRODUCT R&D INSTITUTES FOR THE MACHINE TOOLS AND OTHER TOOLS SECTOR

AND

DESIGN

Four institutes were assigned to machine tools (product category 1):

T-A 1:

T-A 2:

T-A 3: T-A 4:

The Beijing Research Institute for Machine Tools (established in 1956), originally called the Beijing Research Institute for Metal-cutting Machine Tools, with the objective of providing R&D and design for advanced metal-cutting machine tools; The Dalian Research Institute for Modular Machine Tools (1959) focused on modular machine tools and transfer lines for machine tools. This was the only A institute in this sector which specialized at the lowest level of product categories. Modular machine tools are a subcategory within ‘machine tools’. This highly-specialized A institute was established because of the special importance of this variety of products; The Guangzhou Research Institute for Machine Tools (1959) focused on machine tools for tropical environments; The Suzhou Research Institute for Electric Spark Machines, originally established by Jiangsu Province and taken over by the Ministry in 1978, focused on electric spark machines.

One R&D institute was assigned to foundry and forging machines (pro duct categories 2 and 3): T-A 5:

The Jinan Research Institute for Foundry and Forging Machinery (1961) focused on the development and design of foundry and forging machinery.

One institute was assigned to tooling and tools (product category 5):

T-A 6:

The Chengdu Research Institute for Tooling and Tools (1965), originally part of the Research Institute for Tooling and Tool Science (1956), focused on the development and design of cutting tools and tool materials.

One institute was assigned to abrasive and grinding apparatus (product category 6):

T-A 7:

The Zhengzhou Research Institute for Abrasive, Grinding and Grinding Apparatus (1962) focused on abrasive and grinding apparatus.

MACHINERY INDUSTRY DEVELOPMENT

121

Sources: Jing Xiaocun (ch.ed.) 1990: Vol. A, 117–187 (Chapter 5); Zhu Sendi 1994; Interviews, Beijing, Shanghai, and Dalian, October 1994 and April 1995.

Below the group A institutes were forty B institutes, of which twenty were for machine tools (product category 1). Most of these twenty B institutes for machine tools were under the technological guidance of the Beijing Research Institute for Machine Tools (T-A 1), which was in charge of ‘advanced metal-cutting machine tools’. This is a typical example of an A institute being assigned more ‘general’ objectives in the development of product technology, where the related product category is important and was subdivided, for management purposes under that regime, into finer Varieties’. In such cases, it was the B institutes that contributed the elementary and primary work required for standardization, especially the formulation and revision of the ‘size and type of product’ series of standards. Two examples of B institutes for machine tools were selected for study during the field work: The Milling Machine Institute, based in the Beijing No. One Machine Tool Plant, and the Grinding Machine Institute, based in the Shanghai Machine Tool Works. There where three B institutes working in product categories for which no A institute existed: woodworking machinery (product category 4), accessories and built-up jigs (category 7), and machine-tool electrical devices (category 8). The managers from the Ministry recognized that these three institutes were slightly different from the other B institutes (Interview with Mr Yu Chengting). They were ranked somewhat ‘higher’ than the ordinary B institutes and were more heavily commissioned and financed by the Ministry. It is widely accepted that the B institutes were created by the Chinese system. According to the literature (e.g. Jing Xiaocun (ch.ed.) 1990: Vol. A, 136), B institutes were created between 1959 and 1962 in the light of the experience of the Shanghai Machine Tool Works in the mid-1950s. They were assigned ‘to play roles in planning, coordinating and organizing product development’, to cope with ‘the fact that enterprise establishments in the sector became increasingly numerous as the second five year plan (1958–1962) approached’. These institutes were important because of their product specialization. Each institute concentrated on one concrete type of machine, and this was the basis for the effective compilation of the size and type standards for machine products and the organization of product diversification in the conditions which applied at that time. A case study of a B institute in this sector, the Milling Machine Tool Research Institute of the Beijing No. 1 Machine Tool Plant, is provided in Chapter 16 (Case Text 16.1). PRODUCT TECHNOLOGY R&D IN THE ELECTRICAL EQUIPMENT SECTOR

The sector was divided into four product categories: electricity generation equipment; transmission and distribution equipment; electrical appliances; and electrical materials, components and special facilities for the production of electrical equipment. Along the lines of this product categorization, fifteen A institutes and nineteen B institutes were developed in the sector. Another A institute, the Institute of Technical Economics for the Electrical Industry (established in 1985), focused on economic and technological policy for the sector and is not included in this analysis. As can be seen from Case Text 14.3, the nineteen B institutes in the electrical equipment sector were highly specialized by product category, as was the case in the machine tools and tools sector. But in the electrical equipment sector many more A institutes specialized in a single product sub-category (eight of the sixteen A institutes). These eight, along with the nineteen B institutes, were the technological agents at the most elementary level which the central innovation planning system required. The guidance and supervision given by A institutes to B institutes was less significant in the electrical equipment sector than in the machine tools and tools sector, and especially in the product category of ‘machine tools’.

122

MACHINERY TECHNOLOGY INSTITUTES

An A institute might be established for a single important product, such as the Lanzhou Design and Research Institute for Diesel Powered Vehicle Engines (E–A 2 in Case Text 14.3), whose product was originally for military needs, or the Xi’an Research Institute for Micro-Electric Control Machines (E–A 9), whose product was critical to the sector and technologically complex. Other A institutes were established for the development of a product whose production required smaller investments, but for which there were numerous manufacturing users (E–A 5, 10, 11 and 13) (interview with Mr Zhou Zhang). B institutes, in contrast, were usually established for products demanding greater production investments, and with fewer manufacturing users. For the development of such products, commissioning a leading enterprise reduced investment costs significantly while planning effectiveness was scarcely affected (interview with Mr Zhou Zhang). Apparently, the less hierarchical structure of institutes in the electrical equipment sector was mainly due to the less hierarchical product categorization. This would indicate that the aim in developing the system was to promote product specialization at a level as low as necessary, while retaining effective planning coordination.

CASE TEXT 14.3 PRODUCT TECHNOLOGY R&D AND DESIGN INSTITUTES FOR THE ELECTRICAL EQUIPMENT SECTOR, BY PRODUCT CATEGORIES Four A institutes, and three B institutes were assigned to electricity generation equipment (product category 1):

E–A 1:

E–A 2: E–A 3: E–A 4: E–B 1: E–B 2: E–B 3:

The Shanghai Power Equipment Research Institute, established in 1979 with the objective of providing R&D and design work for integrated power plant equipment. The institute was originally established in 1958, as the Shanghai Research Institute for Turbines and Boilers, was dismantled in 1969 and re-created ten years later; The Lanzhou Design and Research Institute for Diesel Powered Vehicle Engines (1966), with the objective of designing and developing vehicle power plants and other mobile plant; The Harbin Power Equipment Research Institute, whose tasks were similar to those of the Shanghai Power Equipment Research Institute (E– A 1); The Shanghai Research Institute for Industrial Boilers (1984), with the objective of providing design, testing and research work with respect to industrial boilers; The Harbin Research Institute for Large Capacity Generators (1958) focused on the development of hydro-electric generation equipment, and was based in the Harbin Electric Generator Factory; The Shenyang Research Institute for Accumulators (1972) focused on the development and design of accumulators and was based in the Shenyang Accumulator Works; Hangzhou Research Institute for Industrial Steam Turbines (1977) focused on the development and design of industrial steam turbines, and was based in the Hangzhou Steam Turbine Works.

MACHINERY INDUSTRY DEVELOPMENT

Five B institutes and no A institutes were assigned to electricity transmission and distribution equipment (product category 2): E–B 4: E–B 5: E–B 6: E–B 7: E–B 8:

The Shenyang Research Institute for Transformers (1958) focused on the development and design of transformers, and was based in the Shenyang Transformer Works; The Xi’an Research Institute for High-Voltage Electrical Apparatus (1958); The Xi’an Research Institute for Rectifier Devices (1966). Both this institute and the Xi’an Research Institute for High-Voltage Electrical Apparatus (E–B 5) adjoined the Xi’an Switch and Rectifier Works; The Xi’an Research Institute for Electrical Capacitors (1972), based in the Xi’an Electrical Capacitor Works; The Xuchang Research Institute for Relay Devices (1973), based in the Xuchang Relay Device Works.

There were eight A institutes, and seven B institutes engaged in product technology R&D for electrical appliances (product category 3):

E–A 5: E–A 6: E–A 7: E–A 8: E–A 9: E–A 10: E–A 11: E–A 12: E–B 9:

E–B 10:

The Shanghai Scientific Research Institute for Electrical Apparatus (1955), for the development and design of small and medium-size electric motors; The Guangzhou Scientific Research Institute for Electric Apparatus (1956), for environmental technology in relation to electrical apparatus; The Tianjing Design and Research Institute for Electric Drives (1957), focusing on research and development for electric drives and automation systems; The Xi’an Research Institute for Electric Furnaces (1964) focused on the development and design of electric furnaces; The Xi’an Research Institute for Micro-Electric Control Machines (1965); The Chengdu Research Institute for Electric Welding Equipment (1965); The Shanghai Research Institute for Electric Powered Tools (1965); The Kunming Scientific Research Institute for Electric Apparatus (1970s), focusing on environmental technology in relation to small and medium-size electric motors operating at high altitude. The Jiamusi Research Institute for Explosion-Proof Electrical Apparatus (1959) focused on the development and design of explosionproof electric apparatus, and was based in the Jiamusi Explosion-Proof Electrical Apparatus Works; The Xiantan Research Institute for Electric Traction Equipment (1964) focused on the development and design of electric traction equipment, and was based in the Xiantan Electric Machine Works;

123

124

MACHINERY TECHNOLOGY INSTITUTES

E–B 11:

E–B 13: E–B 14: E–B 15:

The Nanyang Research Institute for Explosion-Proof Electric Apparatus (1967) was based in the Nanyang Explosion-Proof Electrical Apparatus Works; E–B 12: The Shenyang Research Institute for Electric Drives (1972) focused on the development and design of lowvoltage switch devices, and was based in the Shenyang Low-Voltage Switch Device Works; The Changsha Research Institute for Electrical Automobile Apparatus was based in the Changsha Electrical Auto-apparatus Works; The Tianshui Research Institute for Electric Drives adjoined the Tianshui Electrical Apparatus Corporation; The Shanghai Research Institute for Electrical Automation, belonging to the First Bureau of the Machinery and Electrical Industry of the Shanghai Municipal Government.

Three A institutes and four B institutes were assigned to electrical materials, components and specific facilities for electrical equipment (product category 4): E–A 13: E–A 14:

E–A 15: E–B 16: E–B 17: E–B 18: E–B 19:

The Shanghai Research Institute for Electric Cable (1957) focused on the development of technology and materials for wire and cables and the development and design of cable plants; The Guilin Scientific Research Institute for Electrical Apparatus (1969) was originally established in Beijing in 1954 as the Beijing Research Institute for the Electrical Apparatus Industry. It focused on the development of insulating materials, electrical alloys and magnetic materials; The Xi’an Research Institute for Electro-electronic Engineering focused on R&D for electro-transistor devices and engineering. The Xi’an Research Institute for Electrical Ceramics (1958), was based at the Xi’an High-Voltage Ceramics Works; The Harbin Research Institute for Electrical Instruments (1960), based in the Harbin Electrical Instrument Works; The Shenyang Research Institute for Electro-Engineering Equipment (1973), based in the Shenyang Electro-Engineering Machine Works; The Harbin Research Institute for Electro-carbon Devices (1974), based in the Harbin Electric Carbon Works.

Sources: Zhou Jian’nan (ch.ed.) 1990 Vol. A, B, C, various chapters; Jing Xiaocun (ch.ed.) 1990 Vol. A, 313–379; Interview with Mr Zhou Zhang, October 1994; Interview with Prof. Chen Binmuo, October 1994; Zhu Sendi 1994.

Summary: the A institutes in the sectors of electrical equipment and of machine tools and other tools The R&D institutes which are entirely centrally commissioned and financed are of particular interest in examining the transformation of R&D institutes under the market reform. The institutes which were

MACHINERY INDUSTRY DEVELOPMENT

125

attached to enterprises and only partly financed by the government could be transformed by simply making them an integral part of the host enterprise and terminating the central funding, but the transformation of A institutes was not so simple. The descriptions of the two sectors given above suggests a classification of A institutes for product technology R&D into four classes according to their technological function, as shown in Table 14.13. Institutes devoted to the development of single machines, and those with general objectives or assigned to developing complex mechanical products (functions I and II in Table 14.13) account for sixteen of the twenty-two A institutes in the two sectors examined above. Product technology institutes in these functional groups, along with institutes for manufacturing technologies, will be the principal focus of our study of the transformation of R&D institutes. The institutes devoted to the development of single machines can be further subdivided into those serving many manufacturing users and those in sectors with only a few users for their technological output. The distinction is important because the number of users is a factor in determining the size of the market for their technologies once market-oriented reforms had begun. Table 14.13 Entirely centrally commissioned and financed R&D institutes in the machinery industry, by technological function Technological function: assigned to the development of

Example

Manufacturing technology

Most institutes of RIMST, except (3) and (8) and perhaps (4) (see Case Text 14.1);

Product technology I

II with fewer manufacturing users with many manufacturing users III

IV

General objectives or complex mechanical products Single machines Institutes T–A 3, 4 and 5 (see Case Text 14.2); and institutes E–A 5, 10, 11 and 13 (see Case Text 14.3); Institutes E–A 2, 4, 8 and 9, (see Case Text 14.3); and institute T–A 2 (see Case Text 14.2); Materials and components

Environmental and application technology

Institutes E–A1, E–A3, E–A7 (see Case Text 14.3) and T–A 1 (see Case Text 14.2);

Institute (3) (see Case Text 14.1); institutes T–A 6 and 7 (see Case Text 14.2); and institutes E–A 14 and 15 (see Case Text 14.3); Institutes E–A 6 and 12 (see Case Text 14.3);

Locally affiliated R&D institutes Locally affiliated R&D institutes were commissioned and financed by local governments at either provincial or municipal levels.25 By the mid-1980s, there were 493 locally affiliated R&D units engaged in machinery technology, employing about 24,000 technical personnel (Jing Xiaocun (ch.ed.) 1990: Vol. B, 285).

126

MACHINERY TECHNOLOGY INSTITUTES

The locally affiliated R&D institutes had broader technological functions than the centrally affiliated institutes, which were highly specialized by product. Two kinds of technological objectives were assigned to the locally affiliated institutes: somewhat less than half were engaged in the development of agricultural machinery technology and the remainder in the development of mechanical and electrical technology in general (interview with Mr Zhu Sendi, September and October 1994). Their major roles were in the dissemination of machinery technology to local machinery producers through information, standard-setting, and testing services, and providing technical support to the local planning administrations. Nevertheless some locally affiliated institutes developed specific institute strengths, mostly in connection with local specificities in resource endowments and/or industrial structure. The centrally affiliated group and the locally affiliated institutes together comprised the government-run system for machinery technology R&D and innovation. This system was rather centralized in terms of the development of innovative machinery technology, although it was widely dispersed in terms of organizational establishments. The locally affiliated institutes played an insignificant role in the creation of technological change in the industry, partly because they received more limited resources but more particularly because the functions assigned to them generally differed from the design and productdevelopment functions of the centrally affiliated institutes. This is not to deny the role that the locally affiliated institutes played in the diffusion of technological changes and in supporting the system of rigid technical specification required in a regime under which innovations were made only at the initiative of the planning authority. Because of the role they played in the system, locally affiliated institutes are not the main focus of our study, but a few cases have been chosen to examine the broad directions of the transformation of this group under the current reforms. Other R&D institutions In addition to the government-run system of locally and centrally affiliated institutions, enterprises had their own in-house R&D and design departments. Until the second half of the 1970s, technological activities within enterprises were also planned and coordinated by the planning administration. They had only minor importance, as a secondary and complementary part to the R&D system. Since the end of the 1970s, enterprises have been encouraged and stimulated to establish their own R&D and design units, which are perceived as strategically important for enterprise development. By the first half of the 1990s many large enterprises had enhanced their in-house R&D and design efforts, although these were still inadequate in most cases. Table 14.14 outlines the situation of enterprises’ in-house technological activities in the machinery industry in 1992. It can be seen that 118,000 scientists and engineers were engaged in in-house technology development activities, 15 per cent of the total ‘engineering and technical personnel’ employed in the industry and almost double the number of scientists and engineers working in the government-run R&D institutes for machinery technology. The machinery industry spent 2.4 per cent of its gross sales on technological development, compared to an average of just over 1 per cent in all manufacturing. Government funds accounted for only 8.4 per cent of expenditure on technology development in the machinery industry, which is similar to the average level of government technology development funding for manufacturing. R&D in relation to machinery technology was also undertaken in academic settings, in twenty-four higher educational institutions which fell under the jurisdiction of the Ministry of the Machinery Industry, and in other institutions including the R&D institutes of the Chinese Academy of Sciences. But it was the centrally affiliated R&D institutes described above, assisted by the locally affiliated R&D institutes, which

MACHINERY INDUSTRY DEVELOPMENT

127

made the largest contribution to technological change in the industry. This is still the case, although large enterprises are becoming increasingly important. The limited role of academic R&D in machinery technology innovation was largely due to the concentration of innovative efforts in this industry on product diversification (Jing Xiaocun (ch.ed.) 1990: Vol. B, 286, 287). This study touches on the role of academic R&D only in a few cases which demonstrate the role of outside R&D as perceived by industrial enterprises. Our field-work interviews with the Ministry and R&D institutes revealed that higher engineering education faces some challenges, including engineering courses for undergraduate and graduate students which are excessively specialized, largely by product, at the expense of lower standards in basic scientific and engineering knowledge. The graduates found it difficult to adjust to slightly different subjects from those in which they were trained. Another difficulty which has emerged during the reforms is duplication of efforts in product development. The potential of university R&D in developing fundamental engineering knowledge remains largely untapped. However these problems will not be addressed in this study. Table 14.14 Technological development activities in large and medium-sized enterprisesa in the machinery industry (1992) ISIC code

381

382

383

384

385

Machinery industry All manufacturing

(A) number of enterprises (B) number of in-house technological development units (TDUs)b (B/A) proportion of enterprises having a TDU (C) total employees (×1, 000) (D) engineering and technical personnelc (×1, 000) (D/C) eng. and techn. personnel as a proportion of all employees (E) scientists and engineersd engaged in technological developmente (TD) (×1, 000) (E/C) scientists and eng. in TD as a proportion of all employees (F) scientists and eng. employed in TDUs (×1, 000) (F/D) scientists and eng. in TDUs as a proportion of all eng. and techn. personnel (G) annual gross sales (million yuan)

384 152

2,656 1,354

774 407

677 331

219 158

4710 2,402

16,137 6,469

40%

51%

53%

49%

72%

51%

40%

422

4,867

1,109

1,935

337

8,670

27,800

25

404

96

196

46

767

1,917

6%

8%

9%

10%

14%

9%

7%

4.4

100.4

25.8

36.9

16.2

183.7

341.5

1.0%

2.1%

2.3%

1.9%

4.5%

2.1%

1.2%

2.8

55.9

17.6

35.3

6.7

118.3

229.2

11%

14%

18%

18%

15%

15%

12%

18,209

148,013

61,132

114,781

7,897

350,032

1,403,746

128

MACHINERY TECHNOLOGY INSTITUTES

ISIC code

381

382

383

384

385

Machinery industry All manufacturing

(H) actual expenditure for TD (H/G) expenditure for TD as a percentage of gross sales (I) portion of expenditure on TD granted from government funds (million yuan) (I/H) gov. funds as proportion of expenditure for TD (J) portion of expenditure for TC used for product development (million yuan) (J/H) expenditure for product development as a proportion of expenditure for TD

281

4,750

1,397

1,686

332

8,446

19,636

1.5%

3.2%

2.3%

1.5%

4.2%

2.4%

1.4%

6

333

50

279

42

710

1,578

2.1%

7.0%

3.6%

16.5%

12.7%

8.4%

8.0%

130

2,154

766

833

165

4,048

8,557

46%

45%

55%

49%

50%

48%

44%

Source: China Statistical Yearbook on Science and Technology 1993:158, 159, 164, 169, 175, 184, 185, and for indicator definitions, 355–359. Notes: a The classification of enterprise size in Chinese statistics is based on production capacity and on the original value of an enterprise’s fixed assets. For instance, for the sector of power plant equipment, an enterprise that produces equipment of 100 or more megawatts is classified as ‘large’, and one which produces equipment of between 30 and 100 megawatts is classified as medium-sized; while for the sector of non-electric machinery, an enterprise whose fixed assets worth 30 million yuan or more at purchase price is classified as large, and an enterprise with assets worth between 8 and 30 million yuan is classified as medium-size. b Defined as units which are not only officially established as in-house technological development units but also undertake regular activities with defined physical equipment and which are supported with a regular budget. c Defined as those who were educated in the engineering departments of secondary or tertiary institutions, or who have otherwise been granted an engineering or other technical title. d Scientists and engineers are those who were educated in science or engineering departments in higher education institutions, or who have otherwise been granted a scientific or engineering title of middle rank or higher. e Industrial enterprises’ technological development (TD) covers the activities, such as R&D, design, testing and prototype development, which are needed for the introduction of a new product, design, material, process or apparatus, or to modernize those which are already in use or production. Thus technological development is the internal generation of new technologies. Two other terms are in use in Chinese statistics: ‘technology imports’ refers to obtaining technology from international or Chinese sources outside the enterprise and ‘technology reformation’ refers to capital investment in the means of production already in use.

Planning coordination of complex product development: institutional limitations Whilst minor and incremental product adaptations were systematically and thoroughly incorporated into the planning routine, which was supported by the product-specialized R&D system and technically reliant on standardization, this was not the case for more substantial innovative steps such as products which did not fit into the relevant series of standards. The latter could not be routinely handled within the rules which

MACHINERY INDUSTRY DEVELOPMENT

129

governed the institutions as they had developed in the 1960s and 1970s, without special initiatives and appropriate organizational arrangements made by the higher planning administration. Two cases have been selected to illustrate some key features of the innovation of complex products. These are the development of precision machine tools between 1960 and 1975, and the development of large capacity power plant equipment between 1960 and 1985. The two cases are typical, well known in China, and of great significance to the industrial development in China during the 1960s and 1970s. These two cases show that the approach of the planning coordination system to complex product innovation differed in each case in terms of the hierarchical level involved, the roles of R&D institutes and enterprises, and the matters which were centrally coordinated. It is notable that hierarchical coordinating structure was relatively thinner in the case of the power equipment sector than for the machine tool sector. This is in accordance with the description of the institutional structures in these sectors which was provided above, and partly explains the success in exporting electricity generating equipment which has already been mentioned and the relatively high productivity shown in Table 14.5. Nevertheless, the chief feature of complex product innovation as it was organized under the centrally planned regime, a feature evident in both the cases selected here, was that the planning centre was the only source of the initiatives and coordination required for innovation. The mass of the economic agents, especially the productive enterprises, were basically passive. The development of precision machine tools (1960–1975) The development of precision machine tools was one of the most important innovations for the Chinese machinery industry. It began in 1960 as a central initiative involving the coordination of a number of participating ministries and commissions to achieve effective resource mobilization. A leading R&D institute in the sector, the Beijing Research Institute for Machine Tools, functioned as the technical centre in the acquisition of imported basic designs and the planning and implementation of domestic innovative activities for the development. Thus the coordination by the planning system relied heavily, in technical aspects, on the R&D system through which separate pieces of innovative work were defined, scheduled, and synthesized. The innovation plan succeeded in substantially improving the standard of domestic supplies of precision machine tools. However it took rather a long time, some fifteen years, and the development was based on rather traditional mechanical technology.

CASE TEXT 14.4 ‘CAMPAIGN FOR THE DEVELOPMENT OF PRECISION MACHINE TOOLS’ (1960–1975) Foreign supplies of precision machine tools, which had come from the Soviet Union, stopped in 1960, at a time when demands for precision machine tools were increasing as economic development continued. A decision was made by the State Council, in 1960, to develop the more advanced machine tools domestically. A ‘programming group’ was accordingly set up, with several members from responsible ministries and commissions, including the State Science and Technology Commission (the commission in charge of S&T affairs), the State Planning Commission (responsible for budget coordination), the Ministry of Foreign Trade (responsible for the coordination of imports), the State Construction Commission (responsible for the coordination of capital investment), and the Ministry of the Machinery Industry, which was then responsible for machine tools. A ‘steering office’ was set up in the Bureau of Machine Tools and Other Tools of the Ministry, to serve as the executive agent for the implementation of the decision. In addition, the Beijing Research Institute for Machine Tools, a leading centrally funded institute specializing in the development of

130

MACHINERY TECHNOLOGY INSTITUTES

advanced machine tools (an ‘A’ institute for the product category of ‘machine tools’, see Case Text 14.2), was assigned to work with the Bureau in providing technical planning and coordination for the campaign. A development plan based on industrial demands was then formulated, specifying fifty-six varieties of precision machine tools to be developed in the period 1961–1970. The development plan also divided up the work needed for the development, scheduled each part of the work, specified the outputs, and mapped out the resources and measures which would be necessary. This is plainly an innovation plan, i.e. one aimed to bring a set of new machine tool products into normal production, rather than a plan limited to ‘R&D’. The resources needed were guaranteed by the central planning administration. The basic design of the precision machine tools required for the development came from abroad. In 1963, a number of sample machine tools were imported. Intensive reverse engineering was organized by the Beijing Research Institute for Machine Tools. Knowledge about their design and manufacture was gained mainly through testing and experiments. The extensive participation of personnel from a number of enterprises and institutes fostered the broad dissemination of the knowledge acquired. Some design adaptions were made and the measurement methods necessary for the manufacturing of the designs were devised. In 1964 a ‘research team’ was organized at the Beijing Research Institute for Machine Tools. Members of the team were trained in testing and measuring methods using imported machines so that skilled persons would be available to work on the newly introduced machines at various production sites. The equipment and testing instruments necessary for the production of the precision machine tools were either imported or developed within China. Key equipment which was domestically developed included equipment for the manufacturing of lead screws, worm gears, index plates, screw nuts, and involute cam formworks. All of these items demanded a high degree of precision. Measuring and testing equipment such as photo-electric comparators and graduators were also developed. Plans were coordinated to have various special complements such as bearings, electric motors, accessories, alloys, lubricant oils and cooling agents developed and provided by producers around the country. Capital investments were made as needed, notably in the construction of a few constant-temperature plants. The target of developing fifty-six varieties of precision machine tools had been fulfilled by the mid-1970s. Total production capacity reached 500 units per year in 1965, from three or four key producers. Manufacturing precision had improved: lead screws and worm gears, for instance, were produced at a precision level of ‘1’ or ‘0’. This success was significant for the machine tool sector and the industries using its products in China. On the other hand, the development took fifteen years, which was even longer than the planned ten years, and the success was limited to rather conventional mechanical technology although the machine tool industry worldwide had begun to experience the ‘micro-electronics’ revolution at about that time. In fact, the challenge of the revolutionary innovation of CNC machine tools showed that the planning institution had serious deficiencies in responding to rapid technological change. Source: Jing Xiaocun (ch.ed.) 1990: Vol. A, 150–153.

The development of power plant equipment (1960–1985) The development of fossil-fuelled power plant equipment was another important achievement in innovation for the Chinese machinery industry. It was centrally initiated around 1960 and continued until the beginning of the current reforms. In the case of power plant equipment, central coordination extended more directly to the innovative activities in productive enterprises, so that the leading R&D institute and the ‘government-run’ independent R&D system as a whole had a less central and exclusive role than in the campaign for the development of precision machine tools. This greater role for central coordination from the ministry, and the less elaborate hierarchical structure which resulted, was partly due to the more concentrated structure of enterprises in this sector. Cross-industry coordination for the development of power plant equipment was also prompted largely by the need for user-producer information feedback.

MACHINERY INDUSTRY DEVELOPMENT

131

The great efforts made in the development of power plant equipment have given the Chinese sector experience and some strengths in this field, but its capabilities are still not adequate to provide the scientific foundation for the creation of a new basic design, and shortcomings in manufacturing, design, and testing technology have recently seen the Chinese sector turning to closer alliances with foreign sources of technology.

CASE TEXT 14.5 THE DOMESTIC DEVELOPMENT OF FOSSIL-FUEL POWER PLANT EQUIPMENT (1960– 1985) Fossil fuel is the major power source for electricity generation in China. By 1989, 72.7 per cent (89,972 MW) of the installed power generation capacity was fossil-fuelled, and 81 per cent of this equipment was domestically produced (Zhou Jian’nan (ch.ed.) 1990: Vol. A, 39). This compares with 3,616 MW of fossilfuelled capacity in 1957, at which time just 14 per cent of the equipment was domestically produced (Zhou Jian’nan (ch.ed.) 1990: Vol. A, 15). In the early 1950s power generation technology was imported, with generators of 6–12 MW coming from Czechoslovakia, and the Soviet Union supplying generators of 6–12 and 25–50 MW Based on the imports of the 1950s, the industry in China was already able to produce 25–50 MW power plant equipment by the end of 1950s, but there continued to be many problems with the quality of its products (Zhou Jian’nan (ch.ed.) 1990: Vol. A, 11–14; Jing Xiaocun (ch.ed.) 1990: Vol. A, 334–336). Sustained domestic development began during the 1960s and 1970s, with the aim of self-sufficiency and, in the light of economies of scale in the electricity generating sector, of successively raising the upper size limit. The development began with efforts to improve the 25–50 MW power plant equipment (1961–1965), then to develop equipment of 100 MW to 300 MW equipment (1963–1978), and finally to improve the domestically developed 300 MW equipment while assimilating imported 300 MW equipment (1980–1985). The basic designs were derived from the smaller-scale equipment imported in the 1950s. These were combined with intensive experiments in air dynamics, combustion, structural strength, etc. The lack of an adequate foundation for the creation of a new basic design was one reason that led the industry to turn to import basic designs in the 1980s. As has been noted, the coordination provided by the Ministry extended directly to the development work in manufacturing enterprises, and no institute was appointed as the technical planning centre for the programme. Several factors explain the choice of this structure:

•

the enterprise structure of this sector is much more concentrated. Until the 1970s there were only two enterprises involved (Harbin and Shanghai): a third was established in the 1970s. The central planning was capable of coordinating this simple structure directly; • enterprises were technologically stronger because they were bigger and had inherited talent which had gained experience with 10 MW imported equipment in the 1940s (Zhou Jian’nan (ch.ed.) 1990: Vol. A, 10); • investments in the R&D work in these big enterprises had been made from the 1960s onwards. For instance, an investment from the State Science and Technology Commission in 1963 had helped to establish an in-house research institute for water-cooled electric generator technology at the Shanghai Electrical Machinery Manufacturing Works. Water-cooled generator technology was an important Chinese invention in the sector; and • it appears that the development of power plant equipment was expected to remain a high priority, and the resources required could more easily be obtained by direct coordination by the central planning system.

132

MACHINERY TECHNOLOGY INSTITUTES

The major role of the Shanghai Power Equipment Research Institute (originally the Shanghai Research Institute for Turbines and Boilers), a leading research institute in the electrical equipment sector, appears to have been in: 1) testing and experimentation in relation to some sophisticated questions in the areas of design and manufacturing, and in the development of materials; 2) accident analysis, which is indispensable to learning from failures and which involves sophisticated theoretical knowledge; and 3) organizing standardization and information exchanges for the sector (Zhou Jian’nan (ch.ed.) 1990:20, 21, 42–55, and interview at the Shanghai Power Equipment Research Institute, October 1994). The research institute also assisted on occasion in equipment design. There was substantial cross-industry cooperation between users and producers, which were administratively organized in two powerful ministries. Higher coordinating organs were established to coordinate feedback from the users. For example, in the programme to improve the domestically developed 300 MW plant equipment, a ‘leading group’ was organized by the Machinery Industry Ministry and the Ministry of Water Conservancy and the Electrical Power Industry, at the urging of the State Council. Meetings were held in 1978, 1979, 1980, 1981, 1982 and 1985 to organize inventories of problems encountered and assign responsibility for projects to solve the problems. The coordinated improvement of domestically developed equipment was increasingly mixed with the assimilation of imported 300 and 600 MW technology: Westinghouse turbine and electricity generators from ABB and Siemens. In 1994 the Shanghai producer decided that their 300 MW product should be based on the imported design. Production of the domestically developed 300 MW has since stopped in the Shanghai base, after a total of about thirty units. The experience built up over a long period of independent development is now being directed to the assimilation and modification of Westinghouse technology, in particular in turbine blades, while the weaknesses of domestic technology with regard to design methods and the manufacturing process have largely been rectified by importing technologies. Sources: Zhou Jian’nan (ch.ed.) 1990: Chapter one; Jing Xiaocun (ch.ed.) 1990: Vol. A, 312–339; Interviews at the Shanghai Research Institute for Electric Machinery, Shanghai Turbine Works, Shanghai Electrical Machinery Manufacturing Works, October 1994.

The elaboration of planning practice (the 1980s) In the first half of the 1980s, attempts were made to ‘normalize’ the ad hoc approach to complex technological innovation, within the planning framework, by instituting ‘S&T projects’ for the development of complex production technology as a continuous and integral part of the economic plan, instead of these projects being specially initiated case by case. As mentioned previously (Part 1, Chapter 2), the management of these planned projects relied on officers from the industrial ministries, who served as the major coordinators with the support of the R&D institutes. Project participants were drawn from different units, because of the high degree of institutional segmentation which had developed. The project coordination had to correlate the tasks of the various participants. Thus this elaboration of planning practice in the 1980s did not alter the basic features of the institutional structure for development work, except that these projects were granted continuity. The serious institutional limitations of planned project management (Part 1, Chapter 2) remained: 1 The capacity of the administrative authorities was limited: administrative power was unable to deal with the innumerable different needs of various industrial users (interviews, Ministry of the Machinery Industry and Ministry of the Electronics Industry).

MACHINERY INDUSTRY DEVELOPMENT

133

2 There was a loss of efficiency due to the arm’s length collaboration among participants. This was caused by the ‘outside’ coordination of a large number of participants (Ou Wen 1991). One consequence was long lead times for the introduction of new technology. 3 There were significant deficiencies in the dissemination of the resulting technologies. The planning approach was found useful in acquiring some important industrial technologies, and in putting them into first use,26 but it proved far from efficient in achieving the widespread adoption of new technologies (Interviews with the Ministry of the Machinery Industry), despite the higher level of internal coherency achieved by the Chinese planning system. Summary: institutional structure and technological learning In the light of the growing appreciation that the institutional context has a most important bearing on the efficiency of technological learning (for example, Lundvall, in Lundvall (ed.) 1992:1), this background chapter has focused on the institutional settings and operation of the Chinese R&D and innovation system for machinery technology in the pre-reform era. Prior to the current reforms, the Chinese machinery technology R&D and innovation system embraced several different kinds of institutional settings. While the decentralization of planning control in the Chinese system meant that significant resources were devoted to the R&D institutes affiliated to local governments, the centrally affiliated R&D institutes played the pre-ponderating role in initiating technological changes in the industry from the 1950s to the 1970s. The superiority of the central over the locally affiliated institutes stemmed largely from the product-oriented specialization which was the organizing principle for not only the R&D institutes, but also key productive enterprises. Coordination from the central planning administration was also predominantly organized according to product categories. Such product specialization was much less developed in the local systems. The centrally affiliated machinery technology R&D institutes consisted of two groups. Apart from the sixtyfour well-known central institutes, which were entirely centrally commissioned and financed, there were a larger number of institutes which were organizationally attached to individual productive enterprises but partly funded and commissioned by the centre, and functioned under direct central coordination. The creation of the latter kind of R&D institutes was a unique institutional innovation of the Chinese planning regime. Such institutes substantially supplemented the role of the centrally affiliated institutes in providing the machinery technology innovation system with its product-specialized foundation. These R&D institutes functioned as technical assistants in the process of innovation planning and as the active agents in systematizing product standards. This precise specialization, both in organizational terms (the setting up of R&D institutes), and in technological terms (by standardizing the type and size of each product) substantially reduced the cost of planning coordination, and enabled the Chinese machinery industry to make adaptive innovations as early as the late 1950s. In accordance with this institutional structure of product-oriented specialization, the adaptive innovations made by the Chinese machinery industry from the 1950s to the 1970s focused primarily on product diversification. This institutional structure had two striking disadvantages in relation to technological learning. One is that the system made a total and rigid separation between the functions of creating a new technology and of implementing the newly-created technology in industrial production. The other is that the system suffered from a serious lack of motivation to change (a low incentive structure). The planning authority was the only source of innovation initiatives, and the channels for communication between the functional units were maintained by planning coordination. These two aspects explain why the Chinese machinery industry achieved an impressive quantitative capacity but poor quality performance. The functional disjuncture and

134

MACHINERY TECHNOLOGY INSTITUTES

the lack of incentives prevented the system, in general, from achieving greater dynamism in technological learning. The industry thus fell into a condition of being adept at the massive production of general-purpose products but weak in the development of more user-specific technology and insensitive to cost and profit considerations.27 The institutional structure described above was the starting-point for the current reforms. The R&D institutes which had in principle been developed to conform to the central coordination system had virtually no justification in a market-oriented economic regime. They embodied institutional legacies from the past, particularly the separation from their industrial users and narrow product specialization. But most of the learning accumulated from the past, when they had the function of generating industrial production technology, was also embodied in these institutes. Therefore the transformation of industrial technology R&D institutes has been one of the most crucial issues for policy makers and policy researchers, and is of critical importance for the success or failure of market reform. The following steps in this study will examine how the R&D institutes in the Chinese machinery industry sought ways of moving themselves, with their institutional legacy and learning accumulation intact, into new roles in the first decade of the reform programme.

15 GENERAL TRENDS IN THE TRANSFORMATION OF THE MACHINERY INDUSTRY The extent and direction of the market reform of government-run R&D institutes

This chapter aims to examine general trends in the transformation of the R&D institutions in the machinery industry, based on statistics officially used for monitoring the reform of the scientific and technological management system in China since 1985. The statistical data shows the shares of institute income derived from government funds and from market earnings (the latter is recorded in the statistics as ‘horizontal earnings’). The figure for horizontal earnings is used here as an indicator of the extent to which the institute’s activities have been ‘marketized’. The statistics also show the composition of the market earnings, under several headings: ‘technology development’, ‘technology transfer’, ‘technological consultancy and technological services’, ‘trial production’ and ‘other production and sales’. This data will be used to identify the nature of the contractual relationships between the R&D institutes and their users on the technology market during the current reforms. Income structure: government funds versus market earnings We begin by comparing the extent of market orientation induced by the reform policy in the governmentrun R&D institutes for machinery technology to the figures for government-run industrial R&D institutes as a whole. The government-run R&D institutes for machinery technology comprise the institutes which are entirely centrally commissioned and financed and those which are affiliated to local governments (groups I and III). It does not include the ‘B’ (group II) institutes described in Chapter 14, which were affiliated with host enterprises prior to the reforms and received only part of the funding and direction from the central planning system, because such institutes have been incorporated into their host enterprises since the reform and are no longer affiliated to the government in any sense. The technological development activities carried on in what were formerly B institutes are now recorded as activities of the host industrial enterprises. Table 15.1 shows the income structure of the government-run R&D institutes in the machinery industry in 1993. Note that the data should be considered as illustrative rather than precise, since the official statistics utilize rather rough definitions, and the statistics available are not restricted to the narrowly defined machinery industry. Two calculations are provided in the table to nullify the possible effects of vague definitions in the statistics. Calculation I is for institutes whose main work falls roughly in the areas of ISIC 382, 383, 384 or 385. Calculation II is limited to those in the field of ISIC 382 (general machinery manufacturing), which is the largest part of the ISIC 38 industry in China. Electronic and telecommunication equipment is excluded from both calculations. The two calculations produce roughly the same results, showing that by 1993 government funds were only a small proportion of the institutes’ income

136

MACHINERY TECHNOLOGY INSTITUTES

(12–13 per cent). Market earnings were more than 80 per cent of the total, indicating that the governmentrun R&D institutes in the machinery industry were primarily market-orientated by that time. Table 15.1 Income structure of government-run R&D institutes in the machinery industry (1993) Overall income

Income structure (billion yuan)

Government funds

Market earnings

Other sources

Calculation I 100% Calculation II 100%

4.14 (billion yuan) 13% 2.60 (billion yuan) 12%

0.52 83% 0.32 82%

3.44 4% 2.14 5%

0.18 0.14

Source: Databook of Statistics on Science and Technology 1993:14–15, 53–54. Notes: 1 All the income data is calculated on a gross value basis. 2 Calculation I covers those R&D institutes engaged in the ‘general machinery manufacturing industry’ (roughly ISIC 382), ‘transportation equipment’ (roughly ISIC 384), ‘electric equipment and machinery’ (roughly ISIC 383 but excluding electronic and telecommunication equipment), and ‘instruments, meters and other measuring equipment’ (roughly ISIC 385). There were a total of 504 such institutes belonging to the central ministry or to the local bureaux for the machinery industry at provincial or municipal levels. 3 Calculation II covers only those institutes engaged in the ‘machinery manufacturing industry’ (roughly ISIC 382). There were a total of 335 such institutes belonging to the central ministry or to local bureaux of the machinery industry at provincial or municipal levels. 4 The ‘other sources’ of income are defined as support, donations and credits from domestic or international societies or individuals.

If we compare the figures in Table 15.1 with those in Table 15.2, which covers government-run industrial technology R&D institutes for all branches of industry, it can be seen that the government-run R&D institutes in the machinery industry had, on average, gone further in moving into the market than the R&D institutes for industrial technology as a whole. The institutes in the machinery industry received 4 to 5 percentage points less of their income from government funds, and their market earnings were 4 to 5 percent higher. Despite the differences, Tables 15.1 and 15.2 demonstrate that, by the first half of the 1990s, all industrial technology R&D institutes had become significantly market-orientated. There is no comparable data for the income structure in 1985, when the reforms began, but it is certain that it contrasted strongly with the picture in 1993. One figure is available for income structures in 1985, but it is for all government-run R&D institutes, of which only about half were for industrial technology. In that year, the overall income of all government-run R&D institutes was 7.5 billion yuan, of which government funds accounted for 70 per cent, market earnings for 29 per cent; and bank loans for 4 per cent (SSTC and NCSTD 1990:50). We can assume that at least 50 per cent of the income of industrial technology R&D institutes came from government funds. The fact Table 15.2 Income structure of government-run R&D institutes for industrial technology in all industries (1993) Overall income

Income structure (billion yuan)

Government funds

Market earnings

Other sources

Calculation I 100% Calculation II

13.57 (billion yuan) 17% 9.43 (billion yuan)

2.33 78% 1.81

10.59 5% 7.15

0.66 0.48

MACHINERY TECHNOLOGY INSTITUTES TRANSFORMATION

Overall income

Income structure (billion yuan)

Government funds

Market earnings

Other sources

100%

19%

76%

137

5%

Source: Databook of Statistics on Science and Technology 1993:13, 52. Notes: 1 All figures in the table are calculated at gross value. 2 Calculation I covers the entire industry, including the machinery industry (mining is included but transportation and post and telecommunications are not). This calculation embraces all of the 1,804 independent (governmentrun) industrial R&D institutes affiliated to central industrial ministries or belonging to the local industrial bureaux at provincial or municipal levels. 3 Calculation II covers the entire industry except for the machinery industry, resulted from subtracting the data covered in calculation I of Table 15 from the results of calculation I in this table. 4 The ‘other sources’ are defined as support, donations and credit from domestic or international societies or individuals.

that 29 per cent of their income came from market earnings by 1985, when the S&T reform programme officially began, is an indicator of the extent of pragmatic market reform which had accompanied the relaxation of planning control over enterprises prior to 1985. The government-run R&D institutes for the machinery industry can be further broken down into those affiliated to the central government and those affiliated at the local or provincial levels. Table 15.3 shows that, by 1993, only 8 per cent of the income of the 64 R&D institutes which were affiliated to the central Ministry of the Machinery Industry came from government funds, 84 per cent came from market earnings and 8 per cent from ‘other sources’. This is a dramatic change, compared to the income structure before the reforms began, when 57 per cent of their income came from government funds and the remainder from market earnings. The tables show that government funds did not diminish much in monetary terms (although the value of yuan fell significantly because of annual inflation rates of 10 to 20 per cent per year over this period). The change in the institutes’ income structure was largely due to the phenomenal increase in market earnings. Table 15.3 Income structure of R&D institutes affiliated to the central Ministry for the Machinery Industry (1993) Income composition (billion yuan)

Overall income Government funds

Market earnings

Other sources

1993

1.791 (billion yuan) 100% 0.256 (billion yuan) 100%

0.147 8% 0.146 57%

1984

1.502 84% 0.110 43%

0.142 8% n.a. n.a.

Sources: RISA 1994: Appendix Table 1; RISA 1990: Table 1; Databank of Statistics on Science and Technology 1993: 50. Notes: 1 All figures in the table are calculated at gross value. 2 Data in this table is for institutes affiliated to the central Ministry of the Machinery Industry, which includes almost all institutes affiliated to the central government in some way. 3 The 1993 data covers 64 institutes, three more than those listed in Table 14.12. The additional three institutes specialized in information, planning and standardization, respectively.

138

MACHINERY TECHNOLOGY INSTITUTES

Overall income

Income composition (billion yuan)

Government funds Market earnings Other sources 4 The 1984 data covers only 52 of the 64 institutes included in the 1993 figures. These 52 institutes are classified as ‘technology development’ institutes, which were the targets of the reform policy requirement to reduce government funding. 5 The ‘other sources’ are defined as support, donations and credit from domestic or international societies or individuals.

From Table 15.1, which covers both locally and centrally affiliated R&D institutes and Table 15.3, which refers only to the centrally affiliated institutes, it can be seen that there were no significant differences between these groups of government-run machinery technology institutes. By the first half of the 1990s, both the locally and centrally affiliated institutes had dramatically reversed their dependence on government funds. For both groups, market earnings accounted for more than 80 per cent of total income (Tables 15.1 and 15.3). However the institutes affiliated to the Central Ministry for the Machinery Industry received just 8 per cent of their income from government funds, while the aggregate figure for all government-run machinery technology institutes was 12 to 13 per cent. This was offset by the higher proportion of the income of centrally affiliated institutes which came from ‘other sources’, part of it from technological support programmes provided by international organizations such as the UNDP and World Bank. If we compare the market earnings of the centrally affiliated R&D institutes for the machinery industry in Table 15.3 (84 per cent of total income) with the figure for government-run industrial technology institutes as a whole (78 per cent, see Table 15.2), it is again evident that the centrally affiliated R&D institutes in the machinery industry have gone further in shifting to a market orientation. Yet clearly all industrial technology R&D institutes had largely shifted to a market orientation by the first half of the 1990s. Characteristics of various components of market earnings In the official Chinese statistics which are used for monitoring the progress of the reforms, but have only been published to a limited extent, the market earnings of R&D institutes are divided into five items: 1) earnings from technology development; 2) earnings from technology transfer; 3) earnings from technological consultancy and technological services; 4) earnings from trial production; and 5) earnings from other production and services. Table 15.4 summarizes the definitions of these components of market earnings which are applied in the official statistics. These various types of contracts have differing characteristics with respect to a) the type of output in which the technological know-how is embodied, b) the duration of the contracts, c) how much technological novelty is involved, d) the degree of user-interaction and e) how complete the delivered output is. The relationship between the five income items in Table 15.4 and these five aspects of contract characteristics are summarized in a matrix form in Table 15.5. The type of output in which the technological know-how is embodied The technological know-how which an institute sells may be embodied primarily in the physical product, as in sales of the output from trial production

MACHINERY TECHNOLOGY INSTITUTES TRANSFORMATION

139

Table 15.4 Definition of items of market earnings Items of market earnings

Definition and illustration

Item 1 Technology development

‘Institute earnings from research and development work, under contract for various social entities, aimed at generating a new technology, a new product, a new process, a new material, or a resulting system.’ (SSTC 1994) Item 2 ‘Institute earnings from transferring a patent right or the right Technology transfer to exploit a patent, for licensing the implementation of a patent technology, or for transferring a non-patented technology.’ (SSTC 1994) One typical example is income from the sale of standard blueprints (Interview with Qin and Dong). Item 3 ‘Institute earnings from employing its manpower, physical Technological consultancy and technological services installations and accumulated technology, under contract to a user, to provide services for a designated technological project whether in feasibility studies, technological forecasting, technological surveying, technology assessment or other services to provide technological solutions with respect to a special issue.’ (SSTC 1994) Typical examples are income from providing testing, designs and feasibility studies for engineering projects, and the provision of training (Interview with Qin and Dong). Item 4 ‘Institute earnings from the sale of output from the trial Trial production production of a new product at the prototype stage of development.’ (SSTC 1994) Item 5 Earnings from ‘other production and sales’ includes a) Other production and sales earnings ‘from the batch production of conventional products in the institute’s workshops, farms, etc.’ and b) ‘from other non-technological activities.’ (SSTC 1994) These two activities are separated in the statistics, but are combined for the purpose of this study. The latter group of earnings comes mainly from the institute using its assets for profitable uses, such as the hire of guest houses, meeting rooms and vehicles. (SSTC 1994) Sources: Databank of Statistics on Science and Technology 1993:429: ‘Specification of Key Indicators’; Interview with Mr Qin Yunke and Mrs Dong Lijun, Beijing, Sept. 1994.

and other production and sales (items 4 and 5, particularly 5a, the ‘batch production of conventional products’). Technological know-how may also be entirely or significantly codified and transferred in the form of technical documents, such as blueprints, process and material specifications or reports on the institute’s technical analysis. Earning items 1), 2) and 3) of Table 15.4 are of this kind. In the case of technology development (item 1), the technological output is likely to be provided in both physical products and technical documents. It is therefore reasonable to regard the sum of items 1), 2) and 3) as an approximation of income from engineering services, while the sum of items 4) and 5) can reasonably be regarded as an approximation of income from finished machine products. The provision of engineering services disseminates engineering knowledge, as a basis to enable the users to produce final products. The provision of finished machine products transfers the useful value of the physical product. Both kinds of earnings have resulted from reorienting the technological activities conducted in these R&D institutes to follow market signals.

140

MACHINERY TECHNOLOGY INSTITUTES

The duration of the contracts Two kinds of contractual relationships can be differentiated: period contracts (or ‘enduring contracts’) which last for some time on the one hand, and spot contracts which are made through one-off deals on the other hand. This dichotomy is of course an abstraction from the reality, where there is no clear-cut boundary between the two types. In general terms, the contracts for earnings falling under items 1), 2) and 3) of Table 15.4 are likely to be period contracts, though the length of the period concerned will vary, while spot contracts are more likely for earnings items 4) and 5). In the case of a period or enduring contract, it is necessary for the producer to obtain the user’s responses and input to the contracted activity. This differentiates the first three earnings items from items 4) and 5). The tacitness of technological know-how, and the need to tailor engineering services to a particular user’s special conditions, account for much of the difference between the first three earnings items and the latter two. Tailoring services for a user naturally requires close communication with the user. The degree of technological novelty involved A higher degree of technological novelty is associated with greater uncertainty about both technological and market viability. Intensive and continuous inputs of technological and user information are therefore needed. Contracts for technology development (earnings item 1) entail, by definition, the highest degree of technological novelty. It is evident that period contracts will be most suitable for such projects. Technology transfer and technological consultancy and technological services (items 2 and 3) can be considered primarily as applications of accumulated technological assets, that is, of more mature technological knowledge, but both, particularly the latter, may involve a degree of technological novelty in relation to user-specific requirements. There is no compelling reason to think that ‘trial production’ and ‘other production and sales’ (items 4 and 5) would always involve a lesser degree of technological novelty than items 1, 2 and 3. Some of the products sold are technologically new, or at least represent technological novelty in relation to the technological level of the Chinese market. But it is evident that trial production generally entails a higher degree of technological novelty than ‘other production and sales’. The degree of user interactions The intensity of user interactions, largely in the form of communication between users and producers, is a major factor in the choice between period contracts and spot contracts. As has been mentioned, the tacitness and uncertainty of technology tend to require enduring interaction between the supplier and users, and the enduring interaction needs to be preserved by an enduring contract. It is also useful to note that sales of general purpose products do not generally involve a great deal of enduring interaction with the user. Special-purpose technology is more likely to require an enduring contract, to enable user-specific requirements to be incorporated in the technology developed. Thus one would expect a transaction for technological consultancy to require more intensive communication with the user than the sale of a product from trial production, even though the technological novelty involved in trial production may be greater. How complete the delivered output is ‘Completeness’ refers here both to the inclusion of any required complementary elements along with the core technology, and to the maturity of the technology that is delivered to the user. The degree of

MACHINERY TECHNOLOGY INSTITUTES TRANSFORMATION

141

completion of output is not reflected in the statistics, but a qualitative assessment of the completeness of outcomes delivered under the various earnings items is nevertheless informative for an understanding of the characteristics of these contracts. In the early days of the market reform, the incompleteness of the technology sold through the technology market was recognized as an important reason for the ineffectiveness of these transactions (Part 1: Chapter 3). Since the early 1990s there has been a trend for technology suppliers to deliver more completed outcomes for all kinds of contract, implying that the interface between suppliers and users is becoming more clearly defined. Table 15.5 Characteristics of various items of market earnings Market contract

Technological novelty

1

Technology development

high or moderate high

2

Technology transfer Technological consultancy & technological services Trial production

moderate to low

3

4 5

Other production and sales

moderate

Intensity of userinteraction

Duration of contract

Technological output embodied in

Degree of output completion

(long) period contract

both software and hardware product blueprints

high and rising

moderate

(shorter) period contract high or moderate (medium) period contract

high or moderate low

spot contract

moderate or low

spot contract

low

high and rising technological documents

high and rising

physical products physical commodities

high high

From this discussion, summarized in Table 15.5, it can be seen that contracts for items 1, 2 and 3 have certain common features which contrast with items 4 and 5. The first three items involve a significant amount of engineering services and have stronger effects in the dissemination of engineering know-how, whereas items 4 and 5 represent mainly finished machine products and non-technological services. Perhaps the most important distinction is in the higher intensity of interactions with users required to fulfil contracts for the first three items. User information is a part of the input required to put such contracts into effect, and user-specificity to some extent accounts for the relatively higher degree of novelty involved in these three items. Period contracts are more likely to be chosen for all three, to foster continuous information inputs from the users. The latter two items normally demand only loose interactions with the users, and can be transacted anonymously, so that period contracts are not very necessary. Composition of market earnings: machinery technology R&D institutes affiliated to the central Ministry Table 15.6 shows the composition of market earnings in 1993 for the R&D institutes affiliated to the central Ministry of the Machinery Industry and for two other groups of institutes for comparative purposes. The first is the centrally affiliated institutes in ‘other industries’, that is, R&D institutes engaged in the ferrous

142

MACHINERY TECHNOLOGY INSTITUTES

metal, non-ferrous metal, chemicals, petro-chemicals, textiles, light industry, coal mining, and oil and gas industries, and governed by the relevant central ministries. The second is centrally or locally affiliated Table 15.6 Composition of market earnings of government-run R&D institutes, for various industrial groups (1993) Group of industrial technology R&D institutes

Overall market earnings (billion yuan)

Composition of market earnings (%)

1

3

4

2

5

Technology Technology development transfer

Technologic Trial al production consultancy & technologica l services

Other production and sales

1

1.502

38

12

15

22

13

2.729

22

6

17

26

29

9.090

16

4

12

18

50

2

3

Machinery industry, centrally affiliated Other industries, centrally affiliated All industry except centrally affiliated machinery

Sources: RISA 1994, Table 4, Appendix Table 1, 3, 19; Databook of Statistics on Science and Technology 1993:54, 68, 69, 70. Notes: 1 All amounts and calculations are based on the gross value of contracts. 2 The centrally affiliated R&D institutes for the machinery industry in this table are those that belong to the central Ministry of the Machinery Industry, comprising the 61 ‘group I’ institutes shown in Table 14.12, plus three institutes specializing in information, planning and standardization. 3 The centrally affiliated R&D institutes in ‘other industries’ are those in the ferrous metal, non-ferrous metal, chemicals, petro-chemicals, textiles, light industry, coal mining, and oil and gas industries, and which are under the direction of the relevant central ministries. 4 The institutes for ‘all industry’ excepting centrally affiliated R&D institutes for the machinery industry comprise all institutes which belong to the central industrial ministries or to local industrial bureaux at provincial and municipal levels (1,804 institutes), minus those defined in row 1 of the table (64 institutes). This group totals 1,740 institutes (1,804–64) in 1993.

institutes for all of these branches of industry, but excluding the centrally affiliated R&D institutes for the machinery industry.1 The centrally affiliated institutes for the machinery industry received 38 per cent of their total market income in 1993 from technology development (item 1). Technology transfer (item 2) contributed 12 per cent, technological consultancy and technological services (item 3) contributed 15 per cent, the sale of products

MACHINERY TECHNOLOGY INSTITUTES TRANSFORMATION

143

from trial production (item 4) accounted for 22 per cent and 12 per cent came from other production and sales (item 5). Thus items 1, 2 and 3, which could be grouped together as ‘engineering services’, as explained above, contributed 65 per cent of the total market income of centrally affiliated machinery technology R&D institutes. This was significantly higher than for the centrally affiliated institutes for ‘other industries’, which earned 22 per cent of their market income from technology development, 6 per cent from technology transfer, and 17 per cent from technological consultancy and technological services, so that engineering services combined contributed only 45 per cent of the total market earnings. For the group of all industrial technology R&D institutes affiliated to central or local governments, but excluding the centrally affiliated institutes for the machinery industry, the share of market income coming from engineering services was just 32 per cent, made up of 16 per cent, 4 per cent, and 12 per cent for items 1, 2 and 3 respectively. For institutes of this group, finished products and other services dominated the market earnings, accounting for an average of 68 per cent of the total market income. It is also striking that the centrally affiliated institutes for the machinery institute, which employed one seventh of the R&D scientists and engineers,2 captured 28 per cent of all income earned under technology development contracts and 33 per cent of all income from technology transfer. Unfortunately it has not been possible to derive data for the institutes for the machinery industry which are affiliated to local government. If these figures were available it would be possible to derive total figures for institutes for the machinery industry affiliated to the government at any level, and to compare them with figures for institutes for all industry excepting those in the machinery industry affiliated at any level. Nevertheless, there is plentiful evidence that the income composition of the machinery technology R&D institutes affiliated to local government, shown in Table 14.12 as group III institutes, differs from the income composition of centrally affiliated machinery technology R&D institutes. It can be concluded that the market-oriented transformation currently taking place in the centrally affiliated R&D institutes for the machinery industry is characterized by a movement towards engineering services, as indicated by the predominance of technology development, technology transfer, and technological consultancy and technological services in their market earnings. This contrasts with the figures for the remaining industrial technology R&D institutes, for which the market-oriented transformation has meant a shift towards the generation and dissemination of finished products and other services. One critical implication of the difference is that greater engagement in engineering services, for the centrally affiliated institutes of the machinery industry, means more intensive communication with users, and therefore more reliance on contracts which create ongoing relationships, for longer or shorter periods. The development of market earnings: machinery technology R&D institutes affiliated to the central Ministry We have seen that, by 1993, the centrally affiliated R&D institutes for the machinery industry earned a large proportion of their market income (65 per cent) from engineering services. Table 15.7 presents historical data tracing the evolutionary development of this income structure from the beginning of the reforms. This shows that the market income of these institutes was not so clearly dominated by engineering services during the early stages of the reform. In fact, it was biased more to finished products and other services (items 4 and 5) which accounted for 56 per cent of total market income in

144

MACHINERY TECHNOLOGY INSTITUTES

Table 15.7 Development of the structure of market earnings for centrally affiliated R&D institutes for the machinery industry, 1984–1994 Year

Total market earnings (million yuan)

Item 1 (%)

Item 2 (%)

Item 3 (%)

Engineerin Item 4 (%) g services (sum of items 1, 2 and 3) (%)

Item 5 (%)

Finished products and other services (sum of items 4 and 5) (%)

1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994

110.2 191.5 243.4 289.8 332.1 422.4 517.9 684.3 948.4 1502.1 1167.0

n.a. n.a. n.a. n.a. n.a. n.a. 12 31 34 38 29

n.a. n.a. n.a. n.a. n.a. n.a. 18 11 10 12 17

n.a. n.a. n.a. n.a. n.a. n.a. 28 14 18 15 24

44 44 50 49 54 59 59 56 62 65 70

09 13 14 14 13 17 18 16 13 13 10

56 56 51 50 46 40 40 44 38 36 30

47 43 37 36 33 23 22 27 25 23 20

Sources: RISA 1990: Table I–1, II–1–3; RISA 1995: Appendix Table 1, 3; Other editions of the RISA report, including those for 1989, 1991, 1992, 1993 and 1994. Notes: 1 All figures are based on the gross value of contracts. 2 Data for 1984–1989 covers 52 ‘technology development’ R&D institutes, i.e. institutes engaged primarily in technology development (as distinct from basic and applied research, to use the ‘Frascati’ terms). Data for 1990–1994 covers all 64 R&D institutes affiliated to the central Ministry for the Machinery Industry. 3 The items of market earnings are defined in Table 15.4 above.

1984–1985. The proportion of market income derived from engineering services increased gradually over the following ten years, with only small fluctuations. This steady expansion in engineering services reached a breakthrough point in the late 1980s when engineering services began to surpass finished products as the major source of market income. By 1993–1994, engineering services made up two thirds of the total face value of horizontal income contracts. This shift did not entail a reduction in the monetary value of contracts for finished products and other services (which actually increased five-fold). But there was a ten-fold expansion in total market earnings between 1984 and 1994, and the growth took place disproportionately in engineering services, which grew by a factor of 16. It is not possible to deduce trends in the composition of income from engineering services, because of the lack of disaggregated data for items 1, 2 and 3 prior to 1990.3 If, as we have seen, the centrally affiliated machinery technology institutes won a large share of the engineering services contracts, and the proportion of their market income deriving from engineering services grew steadily over several years, it is reasonable to suggest that there must be some powerful reason behind these developments. The following chapters will examine some cases in detail in the hope of clarifying these reasons.

MACHINERY TECHNOLOGY INSTITUTES TRANSFORMATION

145

Summary This chapter has examined the extent of the market orientation of machinery technology R&D institutes. Official statistics on the institutes’ income structure for 1993 show that the sixty-four R&D institutes for the machinery industry affiliated to the central ministry earned 84 per cent of their income from market contracts, and the proportion is altered very little if institutes affiliated to local government, and those in ISIC sectors 383 (excluding electronics and telecommunication equipment) 384 and 385 are included. All groups derived between 82 and 84 per cent of their income from market sources. If all ‘industrial’ research institutes are included (i.e., for all ISIC 3 industries plus mining), the proportion of income deriving from the market is still 78 per cent. Thus the R&D institutes for machinery technology, and the entire system of industrial technology R&D institutes can be said to have moved decisively away from the heavy dependence on government funds which characterized it ten years ago. A qualitative examination of the characteristics of the various kinds of market earnings suggests the collective term of ‘engineering services’ to combine three items which are separated in official statistics (technology development, technology transfer and technological consultancy and technological services) but which have significant common characteristics. In particular, transactions for these three items focus primarily on the development and dissemination of engineering know-how, and entail an explicit enduring contract, for a longer or shorter period, between the institute and a named user. The ongoing contractual relationship is required to foster the input of information from the user. The remaining items of market earnings (trial production, and other production and sales) are considered together in this study as the sale of ‘finished machine products and other services’. In transactions for these items, it is primarily the utility value of particular machines which is disseminated. Since no close interaction between supplier and user is required, such transactions can be made by anonymous contract and spot transactions. The structure of institutes’ market income was then analysed on the basis of this distinction between engineering services and finished products. It was discovered that the centrally affiliated machinery technology group earns a strikingly high proportion of its market income from engineering services (65 to 70 per cent in 1993–1994). No other group of institutes has a comparable reliance on income from engineering services. Historical data shows that the reliance of the centrally affiliated machinery technology group on income from engineering services has increased steadily, from an initial reliance around 1984 on the sale of finished machine products to parity around 1988–1989, when engineering services became equally important, and more recently to the domination of market income by engineering services. The findings in this chapter provide a foundation for the following empirical and analytical chapters. In the first place, these findings justify the decision to concentrate on the centrally affiliated R&D institutes of the machinery industry. As they have gone through the transformations required by the market reform and opening to international technological exchange, these institutes have maintained stronger capabilities in the area of engineering services than other groups. The function of engineering services is in effect to contribute to producers’ abilities to produce machinery products. This raises further questions: how and why did this group develop the ability to provide engineering services in the technology market when other groups of institutes were moving more to producing finished products? How are the various items of engineering services interrelated, and what is the relationship between engineering services and the production of finished products? What has changed in the field of engineering services, in comparison to the pre-reform approach when these institutes were assigned a similar function, but within a supply structure for machinery technology which excluded foreign suppliers and guided user-supplier relationships through administrative coordination rather than market contracts?

16 THE TRANSFORMATION OF THE ‘PRODUCT TECHNOLOGY’ R&D INSTITUTES

This chapter will provide more detailed empirical observations regarding the transformation of the R&D institutes for the machinery industry that were previously assigned to product technology development. This covers most of the R&D institute establishments depicted in Table 14.12, except for a few centrally affiliated institutes which were previously assigned to develop manufacturing technologies. The latter group shall be examined in the following chapter. A case study approach has been adopted, covering cases in four sectors: machine tools, fossil-fuelled electricity generation equipment, electrical cables and wire, and small internal combustion engines.1 Some basic statistics for these sectors are provided in Table 16.1. Both R&D institutes and the related enterprises in the first two sectors were visited, but in the latter two sectors only R&D institutes were visited because there are very many productive enterprises using the technological output of these institutes. The purpose of the case studies is to identify the directions which the transformation has taken and the factors which influenced the transformation. The results will be illustrative of the course of the current reform, rather than exhaustive. Three factors will be given special attention: 1 The technological complexity of the institute’s activities. The R&D intensity required in the development of its products is one indicator of technological complexity, and the comprehensiveness of the systems and products it develops is another. 2 The population of manufacturing enterprises in the sector. This ranges from very few enterprises in the case of electricity generation equipment to large numbers of enterprises in the small internal combustion engine sector and the electric cable and wire sector. The number of enterprises in a certain sector, and their level of development, provide an approximation of the size and quality of the technology market an appointed institute may find in the sector. It indicates the characteristics of the demand side of that segment of the technology market. 3 Technology imports in the sector. The opening to foreign technology has radically changed the technology supply structure, since the institutes in the survey were previously exclusive suppliers. Information on technology imports indicates how great the effect of this on an institute has been, and why. These factors are examined retrospectively over a period of about ten years, which is long enough to identify evolutionary paths in the transformation. An attempt has been made to identify interactions between changes in each institute and changes in the environment in which the institute operates.

‘PRODUCT TECHNOLOGY’ INSTITUTES

147

Table 16.1 Basic statistics of selected industrial sectors (1993) Industrial sector

Number of manufacturing firms

Employment (thousands)

Fixed capital at purchase price (billion yuan)

Gross sales (billion yuan)

Value added (billion yuan)

Exports (billions of US dollars)

Machine tools

618 (for modular machine tools only, 50) 24 (for power plant turbines only, 3) 245 (approx. another 1,000 for parts and components production) 2,306

398.4

9.32

13.36

5.74

0.219

54.2

2.33

2.62

0.66

0.125a

319.9

8.82

21.84

5.23

0.053b

403.6

13.01

33.65

9.78

0.259

110,284

19,918.1

397.38

928.25

279.80

14.590

Turbines

Internal combustion engines

Electric cable and wire Total for entire machinery industry

Source: China Machinery Industry Yearbook 1994, I.17, VI–10, 13, 15, 18. Notes: 1 Figures in the table, except those marked with a or b, cover the manufacturing firms specified in the second column without considering the subgroups or additional enterprises mentioned in brackets. The information in brackets refers to the sub-branches from which our case-study institutes are drawn. 2 The export figure for turbine products (a) covers all power generation equipment, and not just turbines. Other figures in this row are for turbines only. 3 The export figure for internal combustion engines (b) is for diesel engines only. 4 The total for the entire machinery industry covers all ISIC 38 industries, except the electronic and telecommunications industry. 5 Except for those specified as being given at purchase price, all figures are in 1993 yuan.

The following five sections will examine cases of the transformation of R&D institutes in these sectors. Most of the institutes described in the first four sections are entirely centrally affiliated and commissioned, i.e., A institutes of group I, except that the first case study is of an institute based in a manufacturing enterprise and partly centrally commissioned, i.e., a B institute (also categorized as a group II institute). The fifth section below will consider the transformation of two institutes affiliated to local government, i.e., group III institutes, and the last section summarizes these observations. The categorization of groups I, II and III institutes, and of the A and B institutes is described in Table 14.12. The transformation of ‘B’ institutes In the 1980s there were about 138 R&D institutes for the machinery industry which were organizationally and administratively based in a host enterprise, but partly centrally commissioned and financed. Almost all of these B or ‘group II’ institutes were located in a leading enterprise, and specialized in the development of a particular variety of machine product. This group constituted the institutional and technical substructure

148

MACHINERY TECHNOLOGY INSTITUTES

for innovation planning under the Chinese centrally planned regime. Since twenty of these institutes were assigned to work on the various varieties of machine tools, the machine tool sector was chosen as typifying the way in which the B institutes, which had been an arm of the central government serving its need to manage planned innovation, were transformed to become integral parts of their host enterprises. Two B institutes for machine tools were visited: the Milling Machine Tool Research Institute, within the Beijing No. 1 Machine Tool Plant, and the Grinding Machine Tool Research Institute, within the Shanghai Machine Tool Works. Since the two cases proved to be very similar, only the first is described below.

CASE TEXT 16.1 THE BEIJING NO. 1 MACHINE TOOL PLANT AND ITS MILLING MACHINE TOOL RESEARCH INSTITUTE Background The Plant was formed in 1949 on the basis of a few small repair shops and received substantial investments during the first five year plan (1952–1957). It has been one of the eighteen key machine tool manufacturers in China. This Plant specialized in milling machine tools. Its Milling Machine Tool Research Institute, established in 1959, was one of forty B R&D institutes in the ‘machine tools and tools’ sector of the machinery industry (see Figure 14.1 and the section of ‘R&D institutes: expansion and development’ in Chapter 14). The Bureau of Machine Tools and Tools of the Ministry of the Machinery Industry supervised these institutes until the start of the current reforms. It received technical guidance from an A institute, the Beijing Research Institute for Machine Tools (T–A 1 in Case Text 14.2). The strongly hierarchical relationship between the B institutes and A institutes in this sector has been discussed in Chapter 14. The Plant The Plant now has 7,000 employees, generated gross sales of about 500 million yuan in 1993 (approx 80 million US dollars at the official exchange rate) of which 6 million US dollars were exported. Its products include various conventional and CNC milling machines, notably piano-milling machines, piano-milling and boring machines, single and double column milling machines, CNC vertical milling machines, CNC pianomilling machines and vertical and horizontal machining centres. At present, conventional milling machines are still the major product, with only a few dozen CNC machines being produced per year. Since the reform, the Plant has upgraded its capacity from traditional milling machines to CNC milling machines through technology imports. Although attempts were made as early as 1958, domestically developed CNC milling machines were never produced commercially. The investments required for technological upgrading came from the Plant’s own revenue, supplemented by government subsidies. One remarkable example of upgrading through technology imports is the importation of technology for CNC piano-milling machines from Waldrich Coburg, Germany, in 1984. The import was based on a ‘cooperative manufacturing agreement’ under which the Plant produces machines under the German brand name. Hundreds of technicians from the Plant were trained in the German company, and the first two machines to be made were produced under the inspection of German experts. It was not just the design which was imported, but also technologies for manufacturing, testing, quality insurance, and management. The Plant considers such an extensive import agreement to have been necessary for a radical transition. The enterprise has also considerably improved its own plant by purchasing machining, testing and designing equipment on the international market. This internationally procured equipment also serves as a source of information for innovation. For instance, flexible manufacturing centres are being developed on the basis of experience with an imported system which was originally procured for the Plant’s own manufacturing processes. The economic reforms have increased access to investment funds. Since 1986 the Plant has been run under a ‘contractual responsibility’ system, which gives it control over approximately two thirds of its profits. This has enabled the Plant for the first time to mobilize a significant proportion of its revenue for re-investment in

‘PRODUCT TECHNOLOGY’ INSTITUTES

production facilities and technology. Expenditure of this kind totalled 60 million yuan in the period 1986–1990, and 120 million yuan is planned for 1991–1995, compared to about 9 million yuan in 1981–1985. The CNC machines and flexible manufacturing systems now installed in the Plant are worth about half of the total value of its machining facilities. The Plant’s current strategy relates to a shift in demand which became evident in 1993 and 1994. The Plant’s customers, mainly producers of automobiles and other transportation equipment, are now demanding milling machines of greater precision and sophistication. In 1994, sales of conventional machines were sluggish, while not all orders for CNC machines could be met. At the same time, huge numbers of foreign machine tools, largely CNC machines, are being imported. This means that the Plant is now confronted with international competition, although most of its users are still domestic. It perceives long delivery times as the major bottleneck preventing it competing more effectively in this market. The Plant has responded with both computerization and institutional change, with the main reliance apparently on computerization. To shorten the design time, the Plant has invested 2 million US dollars in computer-aided design facilities (CAD), and is now attempting to establish a computer-aided integrated manufacturing system (CIMS). The Plant is a member of the national CIMS project.2 Another response has been an ‘internal specialization’ programme, under which management autonomy and responsibility has been decentralized to internally specialized departments. The Plant management did not seem very hopeful about the effects of the corporatization of state-owned enterprises, which is now being implemented and which is intended to improve the management of state-owned assets. There were questions about the membership of the board of management under a corporate structure, doubts about the ability of likely candidates to work effectively within a corporate structure, and concerns about retired workers. The Plant still takes a large part of the responsibility for its 3,000 retired workers, with the Beijing Municipal Government also contributing. Reform of the social security system has only just begun. This less enthusiastic response to the corporatization programme, on the part of an enterprise which has shown itself willing to undertake both technological and institutional change to overcome its disadvantages in these areas, is noteworthy. The Research Institute The Milling Machine Tool Research Institute of the Beijing No. 1 Machine Tool Plant was founded in 1959. As a B institute it was charged with the development of milling machine technology under the centrally planned innovation system, while belonging to and being located in the Plant. Since around 1985 it has received neither directions nor funds from the Ministry, and is fully owned by the Plant. The Institute has 170 staff, of whom 30 are senior engineers and about 60 are engineers. A little over a quarter of the engineers are electronics experts. The Institute has divisions for mechanical design, electric and electronic design, standardization, information, and testing. The Institute has played a major role in the acquisition of imported technology for the technological upgrading of the Plant. When the German piano-milling machine technology was imported, for example, the Institute supported the acquisition with translation and testing, and in making minor modifications. The translation and interpretation of imported blueprints and technical documents clarified or corrected the Institute engineers’ understanding of the new product model, which they had more or less known about from literature. The transmission system of the imported machines was recognized as being particularly different from their previous conceptions. The testing and analyses which the Institute performed enhanced their engineering knowhow with respect to some principles underlying the imported technology. The testing indicated the need to rectify their own machining devices to meet the manufacturing specifications set by the German licenceholder. The modifications which the Institute made to the imported designs in response to the requirements of domestic users, though still minor, were said to be of the same quality as those which were originally imported. The modifications extended the learning already achieved through providing translation and testing services, and constitute the foundation for the more sophisticated involvement which the Institute is now attempting. Another role of the Institute in the modernization of the Plant has been coordinating contractual research and testing by external sources. For instance, some testing on the rigidity and dynamic properties of mechanical

149

150

MACHINERY TECHNOLOGY INSTITUTES

structures was carried out at Tiangjing University and some special casting material was developed at the Research Institute for Iron and Steel Technology of the Ministry of the Metallurgic Industry. The ongoing Computer Integrated Manufacturing System (CIMS) project is assisted by experts from South-East University and the United Nations. In such cases, the Institute decided what contractual technological services were required, chose the contractual suppliers, and set the requirements for suppliers in the light of the needs of the Plant’s development. It was interesting to note that the Institute did not mention receiving significant help from any of the R&D institutes in the machinery industry. It was explained that these institutes could not help because of the rapid change in machine tool technology. In most cases, the Institute used the contractual services by combining the output with its own development and design work. In the CIMS project, although there is a heavy reliance on external expertise, it is the Plant experts who will take the final responsibility for the real implementation of the CIMS system. The Institute can therefore be regarded as the intermediate agent to acquire outside support for the Plant in pursuing technological change. The roles of the Institute as described above point to a radical change in its functions. Until the current reforms, the Institute was assigned quite different functions under central planning coordination and served not only the host enterprise but also other milling machine manufacturers. The Institute’s former functions, according to the reports of the interviewees, are listed below. Most of these functions have now been shifted to other bodies, reduced, or cancelled since the market reform.

1

Compiling and revising standards. Prior to the reforms, this involved mainly the compilation of type and size standards for milling machine tools, under the coordination of the Beijing Research Institute for Machine Tools (Institute T–A 1 of Case Text 14.2). Additional tasks included explaining the standards, assigning particular product types to be manufactured by particular milling machine producers, and monitoring the implementation of the standards. Since the reform, the previous national standard system has been replaced by the introduction of international standards, whose use is seen as important if domestic producers are to compete in the market. Responsibility for the standards has now passed to the Association of Milling Machine Producers. Individual enterprises are in addition encouraged to define their own standards for special and more advanced machines. The Institute is now formulating a plant standard for CNC piano-milling machines on behalf of its host, the Beijing No. 1 Machine Tool Plant. 2 Planning. Prior to the reform, the Institute was responsible for compiling an innovation and production plan for milling machines, specifying output, types, and producers, once every five years. Such planning has now ceased. 3 Organizing joint design efforts. Under the planning approach, the Institute coordinated the work of milling machine producers who were involved in the joint design of milling machine tools. This work has also stopped. 4 Providing testing services to milling machine producers. This work has been much reduced since the reform, and is now performed on a paid basis. 5 Information services. The Institute was formerly responsible for gathering information in relation to milling machines and disseminating it through its bulletin. The Association of Milling Machine Producers has now officially taken over the responsibility, and the scope of the information has widened to include market information.

Producer associations, a new institutional arrangement devised in response to the changing government-enterprise relationship, have largely taken over the inter-enterprise functions formerly assigned to the B institutes. The Association of Milling Machine Producers was established in 1986,

‘PRODUCT TECHNOLOGY’ INSTITUTES

151

together with Associations of Lathe Producers, Grinding Machine Producers, and others. These Associations are themselves members of an umbrella association: the Association of Machine Tools Producers. Standardization and information, which are functions beneficial to producers, are now presumed to be the responsibility of the producer associations and not of the government. Associations are also considered responsible for fostering communications between producers and the government, and among producers. The Association of Milling Machine Producers has convened annual producers’ conferences, where the Chief Managers and Chief Engineers of the milling machine manufacturers can exchange their experiences under the reform programme. The Association has also organized machine tool product fairs on several occasions, in partnership with other associations of the machine tool industry. The Milling Machine Research Institute is now the host of the Association of Milling Machine Producers, and provides manpower and money to the Association. It also hosts the Association’s Expert Committee for Standardization. These two titles of the Institute are indicative of the way new institutional structures are being developed on existing foundations. But the institutional restructuring is thus far only partial. The activities of the Association are not yet initiated actively by producers. Plant managers commented that the Association was in fact still semi-official, with the Ministry involved indirectly in encouraging and promoting it. Sources: Interviews with Mr Jin Yuling and Mr Liu Dezhong, at the Beijing No. 1 Machine Tool Plant, Beijing October 1994; Introduction to the Beijing No. 1 Machine Tool Plant, provided by the Plant; Materials on the reform of the Plant, mimeo, provided by the Plant; People’s Daily, Overseas Edition, 30 November, 1995.

To conclude, the B institutes, which made up a unique part of the technological system of the machinery industry under the Chinese planned economy, have been fully integrated into their host enterprises since the start of the current reform, and are now consolidated as part of the strategic operations of their host enterprises. This conclusion may be supported by the evidence in Case Text 16.1, which concerns the Milling Machine Tool Research Institute, by very similar findings at the Grinding Machine Tool Research Institute, and by other incidental observations in the course of this study. This constitutes a basic difference between the transformation of this group of institutes and the transformation of all other institutes of the system. The integration of B institutes with the host enterprises has preserved a great deal of accumulated technological capability for a number of leading enterprises. As they have been integrated with their host enterprises, the functions of these institutes have changed from domestically generating product diversification for a number of producers in the sector concerned (as seen in Case Text 16.1, and in Chapter 14) to supporting the acquisition of imported technology specifically by their host enterprises. Such technology imports have been the main source of technological change since the late 1970s. The B institutes’ previous experience has made them the main technological strength of their host enterprises. They have played a critical role through contributing to the interpretation and testing of imported technology and to adaptations for local requirements. They also served their host enterprises by coordinating technological service contracts with external sources, and obtaining other support when necessary. The technology imports have been crucial in accelerating technological change of the enterprises to which the B institutes are attached. However there is still a great deal to be done in mastering and integrating imported technology more adequately, especially with respect to the transformation of the host enterprises. For instance, the main challenge for the host plant of the Milling Machine Tool Research Institute, as shown in Case Text 16.1, is to reduce the time taken to develop and produce new designs. This is due to profound technological and institutional handicaps, as the Plant has recognized.

152

MACHINERY TECHNOLOGY INSTITUTES

The case study has also revealed a change in the relationships among producers of similar products. Before the current reform, this relationship was characterized by the homogeneous sharing of various producer-specific technological information. The B institutes were assigned extensive functions in support of this information sharing. The introduction of market competition seems to have revealed the commercial importance of producer-specific information, which is now largely protected by the holders. The joint activities of producers have shifted away from competition sensitive information to issues of common concern to producers, such as market information, marketing activities, price coordination and standardization. Producers’ associations have been promoted to replace the government in coordinating activities to further their common interests, and the B institutes, which were previously the basis of government coordination, have played a temporary role in furthering this transformation. The transformation of an ‘A’ institute in a sector with rapid technological change In the category of government-run machinery technology R&D institutes, fifty-three institutes were entirely affiliated to and financed by the central government and charged with the development of product technology. These are the group A institutes as specified in Table 14.12. These A institutes, together with another eight institutes which were also entirely affiliated to and financed by the central government but were charged with the development of manufacturing technology, comprise the group I institutes. This section and the two which follow will consider the transformation of product technology institutes in the A group in selected sub-sectors characterized by rapid technological change, by a concentrated enterprise structure, and by numerous producers, respectively. This examination will focus on those A institutes whose assigned function was to develop machines, whether simple or more complex. It will not consider the other A institutes shown in Table 14.13 which were charged with the development of other products such as components and materials or with developing the environmental and application technology of machine products. The case of the Dalian Modular Machine Tool Research Institute exemplifies the transformation of A institutes in a sector experiencing rapid technological change, the machine tool sector. Seven of the fiftythree A institutes were assigned to the machine tool sector, and the case examined below is typical for most of these. Although the research institute in question is not affiliated to a manufacturer, it will be helpful first to consider the type of manufacturing enterprises in this sector and how they have responded to the reforms, by examining the case of the Dalian Machine Tool Works, the leading manufacturer of modular machine tools in China.

CASE TEXT 16.2 THE DALIAN MACHINE TOOL WORKS Background Founded in 1948, the Dalian Machine Tool Works has been the leading manufacturer of modular machine tools. It now has 6,600 employees, of whom 600 are engineers. Its main products are modular machine tools and transfer lines, various types of lathes, and flexible manufacturing systems. Sales in 1993 were about 300 million, with a small amount of exports. In 1993 the works produced 1,800 universal lathes, 250 conventional (i.e., rigid) modular machines, and about 50 CNC lathes. In the late 1950s and early 1960s a close relationship, based on a division of labour, developed between the Works and the Dalian Modular Machine Tool Research Institute, which will be described in Case Text 16.3. The Institute prepared designs, and the Works produced them. Designer teams from the Institute often stayed in

‘PRODUCT TECHNOLOGY’ INSTITUTES

the Works to deal with any problems with the designs. However in the period 1970–1973, during the Cultural Revolution, technicians from the Works were organized to do design work alongside engineers from the Institute and so learn the skills required. The Institute engineers were then driven away. This unpleasant episode remains something of an impediment to the possibility of the Institute and the Works merging later on in the present reform. Driving the Institute engineers away did not create great problems for manufacturing at the Works, because the standardized design approach transferred the necessary technological information, and the modular machines required at that time were not too complex. Technological upgrading during the reform has been based on technology imports, which have been partly subsidized by the government. The Works achieved the ability to produce flexible manufacturing systems of the FM 200 type by importing the technology from KTM, in the UK. The government paid for slightly less than half of the total expenditure for these imports, with the Works paying for the rest. The Works may now produce the imported models themselves, but the market is still too small. There was no mention of any modifications being made to this imported flexible manufacturing technology. From these indications, it appears that the Works is at the stage of being capable of imitating advanced imported technology, but not yet able to modify it. Some more conventional modular machine tool technology was also imported, from Huller Hille in Germany, and has been intensively applied in the Works’ own manufacturing processes. The Works has also improved its means for machining, designing, and testing by procuring precision machines, instruments, and computer-aided designing systems through international procurement. International standards were adopted in 1985, and some German firm standards began to be adopted in 1990 with the import of modular machine tool technology. The introduction of computer-aided design technology started in the mid-1980s, with support from the Ministry. Designs for the heads of modular machine tools are now drawn using computer-aided technology. The Works has drawn on both outside assistance and its own efforts to support the technology acquisition process. The Dalian Modular Machine Tool Research Institute assisted in two ways: by contributing to the translation and interpretation of imported designs and technological documents and by providing information relating to the Institute’s role as a centre for standardization and quality inspection for the modular machine tool sector. The Works’ own design and R&D departments, with a staff of 250, have a capability in (conventional) modular machine tool design. They face difficulties because of weaknesses in the application of computer technology, and in the scientific experiments which are increasingly needed to support engineering design work in the development of machine tools. These weaknesses have hampered the Works in keeping up with the rapid change in this sector. To fill the gap, the Works once attempted to rely on contractual services for support in solving key technical problems. That was regarded as having failed, with the comment that ‘the time horizon for contractual services is too short, compared to the period needed to strengthen the Works’ own capability.’ The Works then turned to relying on contractual services for training support to their internal acquisition of key techniques. Young engineers from the Works are sent to some universities, particularly those studying areas of interest to the Works such as vibration analysis, modular machine tool CAD, casting CAD, flexible manufacturing systems, and spindle Computer-Aided Production Planning (CAPP). It was said that universities, rather than R&D institutes, are chosen for the training contracts because they are not so profitseeking as R&D institutes. The Works seems willing to merge with the Dalian Modular Machine Tool Research Institute, which would be in accordance with government policy favouring the merging of R&D institutes with productive enterprises. The Works recognized that the Research Institute has advantages in the abilities to develop special purpose machines and auxiliary machinery and control devices. These advantages are maintained by the superior manpower of the Institute, especially in the fields of computers and automation. Designs made by the Institute were said to be not very workable from the point of view of the manufacturer. Current strategies

153

154

MACHINERY TECHNOLOGY INSTITUTES

The technology imports of the 1980s were regarded by the Works as less successful than hoped. The assimilation time, which in the case of the flexible manufacturing system FM 200 was almost ten years, was too long. Although the very long assimilation time in this particular case was partly because of the choice of an inappropriate technology, which was too sophisticated to have large numbers of domestic users in the mid-1980s. The Works learnt that they suffer from serious weaknesses which have prevented them keeping up with the fast pace of innovation in the machine tool industry. These weaknesses are now becoming a real threat to the Works’ survival, at a time when producers in the automobile industry are replacing their manufacturing equipment and producers in some other industries seem likely to follow. Two strategies have been formulated. One is the further expansion of international cooperation, with the goal of ‘gaining access to the dynamics of technological development’. This strategy is thought of as the first choice as far as the external sourcing of technology is concerned. It signals a shift from discrete licensing for technology transfer to developing all-round cooperation with multinationals. One step of this strategy is the joint bid which the Works prepared, with Honsberg of Germany, for the development of one transfer line at the No. 1 Automobile Factory. The bid was successful. The development project will embrace both joint design and joint manufacturing. Another strategy is internal restructuring. Until 1994 the Works retained the very centralized management mode which it had inherited from the past. Centralized internal management was compatible with the production of a single and standardized product under the centrally planned regime, but the Works has now recognized that this heritage has become a major impediment, preventing them responding quickly and effectively to changes. Any issue relating to the Works, or to its 6,000 employees or dozens of sub-units, had to be resolved at meetings attended by nine chief and deputy-chief managers. This left the chief managers little opportunity to give time and attention to strategically important decisions. A scheme for internal restructuring is about to be put into action, with internal management being broken down into a central department and several business departments. Each business department will operate as a profit-making centre, specializing in a certain product category. The central department will retain power in investment decision, i.e., the internal allocation of resources, and will include the R&D department. This implies that the development of advanced machine tools is seen as having the same strategic importance as internal investment. The internal restructuring will establish a new mode of internal management, which the managers of the Works call ‘internal specialization’. The major difficulty they perceive in this internal restructuring is the lack of experience in internal cost and profit management. Surprisingly, the reform policy has not seriously addressed either the issue of internal restructuring or the techniques of internal cost/profit management, although enterprise managers are apparently making pragmatic experiments. The Works managers did not have positive expectations of progress in external specialization in the near future. Previous attempts at specialization, largely external, had failed. Their assessment of the government programme to corporatize state-owned enterprises was as pessimistic as that of the managers from the Beijing No. 1 Machine Tool Plant. One particular concern was the perceived lack of adequately experienced persons to be the state representatives on the corporate board, to ensure the efficient utilization of state assets. Sources: Interview with Mr Wang Keng, Dalian Machine Tool Works, Dalian, October 1994; Introductions to the Dalian Machine Tool Works, provided by the Works.

CASE TEXT 16.3 THE DALIAN MODULAR MACHINE TOOL RESEARCH INSTITUTE, OF THE MINISTRY OF THE MACHINERY INDUSTRY Background The Institute was founded in 1956, along the lines of the Soviet archetype of a ‘design bureau’ for modular machine tools. It was designated by the Ministry of the Machinery Industry to specialize in the development of modular machine tools to meet the requirements of the newly established automobile and tractor industry in

‘PRODUCT TECHNOLOGY’ INSTITUTES

that time. Since then the Institute has been the only one in China specializing in modular machine tool technology. It now has a staff of about 1,400, including 200 senior engineers. The Institute’s R&D departments employ about 800 people, and the remainder work in various affiliated workshops. The Institute’s major technological tasks related to the design of modular machine tools. The basic design was originally imported from the Soviet Union. Designs made in the Institute were then produced in a related factory. Initially there was only one modular machine tool factory in China, the Dalian Machine Tool Works. In the second half of the 1960s, with the construction of the No. 2 Automobile Factory, the number of modular machine tool producers was increased to ten. There are now 25 firms producing complete machines, 15 making parts, and 10 producing auxiliaries. The Institute’s involvement in the construction of the No. 2 Automobile Factory illustrates its working methods in the old system. The construction of the automobile factory required some 800 modular machine tools, which were to be produced in ten different plants. Ten designer teams from the Institute were sent to the plants, doing design work ‘on the manufacturing site’ for about three years until the manufacture of the products they had designed had become routine. Doing design work on the manufacturing site achieved several aims: the design was adapted as well as possible to the particular manufacturing circumstances in a plant, the plant technicians were trained and gained skills in working with designers, and design problems which appeared during trial manufacturing could be solved. These on-site services were provided free of charge. As part of the nationwide standardization effort, the Institute has also coordinated the compilation of the ‘type and size’ standards for modular machine tools since the 1960s. This work was based on imported Soviet standards, with some structural and processing norms simplified. Since then the standards for modular machine tools have become more systematic, and parts of the machines have been made more interchangeable. The standardization created a foundation making it possible to design and manufacture the huge number of modular machine tools that were required, not only for the No. 2 Automobile Factory, but also for other domestic needs including consumer goods such as bicycles and sewing machines, which have been produced in greater numbers since the 1970s. Technological upgrading From being a traditional modular machine tool designer, the Institute has upgraded its technological mastery in some areas of advanced manufacturing technology during the market reform. The Institute drew on several sources for its technological upgrading: direct technology imports, learning from providing support to importing enterprises, and from international cooperation. A notable instance of direct technology import was the FM 100 horizontal machining centre which the Institute imported from FMT in the UK (The Dalian Machine Tool Works imported the UK FM 200 machine). The Institute is now licensed to manufacture the FM 100, but demand has been limited. Where productive enterprises imported required technologies themselves, the Institute found a complementary role in providing translation, interpretation and additional adaptation services. A typical case of this role was the transformation of imported German Huller Hille design standards for modular machine units. The Institute translated, modified the imported technology for domestic users, and thereupon developed their own series of standards for modular units, called the ‘I’ series, as a result of absorbing the German technology. The Institute now maintains three series of standards for modular machine tools: the original one developed in the 1960s– 1970s, the German one and the ‘I’ series. International cooperative partnerships have been developed mainly as commercial relationships. By providing technological support in marketing, training, repairing, and user engineering to multinationals such as Allen-Bradley of the United States, REXROTH of Germany, and the Japanese OMRON, in their business in the Chinese market, the Institute has acquired not only related technology but also international commercial connections. The Institute’s modest success in exporting its products, including ‘multipoint lubricating units’, began through one such partnership. The investments for technological upgrading have come mainly from the government, through the channels of ‘capital construction’ and ‘project’ funds. Capital construction funds were used to improve the Institute’s facilities for testing and trial production during the late 1970s and early 1980s. The investment enabled the Institute to produce sample machines to their own designs in the Institute’s workshops, which now operate as a

155

156

MACHINERY TECHNOLOGY INSTITUTES

trial production base. Project funds from the ‘key S&T projects of the five year plan’ were important throughout the 1980s. R&D activities financed from ‘key project’ funds concentrated on advanced manufacturing technology, such as flexible manufacturing, industrial robots, CAD, and auxiliary machinery for advanced manufacturing systems. These are now the main areas in which the Institute has an advantage. Trends in contractual activities The Institute was identified under the reform policy as a ‘technology development’ type (see Part 1, Chapter 3). Government grants for daily operations were stopped in 1984. As Table 16.2 shows, the Institute’s income shifted in 1991 from an initial reliance on providing engineering services (combined technology development, technology transfer, and technological consultancy and technological services) to more reliance on trial production, with engineering services continuing to be of some importance in most years. The users for these different market activities differ. In broad terms, technology transfer, technological consultancy and technological services are provided mainly to smaller producers of modular machine tools and transfer lines. Trial production is a business selling the products made for final machine tool users, mainly from the automobile and transportation equipment industry. The customers for contractual technology development services are more diverse. They may be producers of modular machine tools or the final user of the modular machine tools. This income structure shows that the Institute’s relationship with modular machine producers is moving from a pure supplier-user link in which the Institute supplies product technology to something more like rivalry, with the Institute and machine producers competing in the production of some machines. Table 16.2 Change in income structure 1984–1994: Dalian Modular Machine Tool Research Institute (million yuan)

Source: Unpublished data provided by Mrs Zhang Meiguang, Research Institute for System Analysis, of the Ministry of the Machinery Industry, 1995. Note: Amounts are not corrected for inflation. From 1989 on, the total income includes ‘other income’ (see Table 15.1) which includes support, donations and credits from domestic or international societies or individuals.

The Institute’s current situation is uncertain. It is becoming difficult for it to maintain its position as an important supplier of technology for machine tool producers, which was the purpose of government investments until the late 1980s. Producers are turning to foreign technology on a large scale, as illustrated by the Dalian Machine Tool Works’ choice of foreign suppliers, with whom the Institute cannot compete. The remaining possibility which the Institute has identified is to become

‘PRODUCT TECHNOLOGY’ INSTITUTES

157

an influential supplier of advanced machining systems, given that the reform policy does not leave room for such institutes to survive on government funds and that the Institute is unwilling to merge with the Dalian Machine Tool Works. In comparison with ordinary industrial producers of machine tools, the Institute’s main advantage, as a commercial supplier of advanced manufacturing systems, is its better mastery of flexible manufacturing technology, including auxiliary machinery and automatic devices. This mastery resulted mainly from government investment throughout the 1980s. Well-trained and experienced S&T manpower supports its advantage in this area. No domestic competitor has such comprehensive mastery of the technology for transfer line machinery and some flexible manufacturing systems. The mastery is critical at a time when the structure and operation of machine tools has been fundamentally altered along with the incorporation of electronic technology. The technological mastery of the Institute is illustrated by its successful development of the ZHSUX78 transfer line, for finishing the conical surfaces of inlet and outlet valve seats and valve stem guide holes in cylinder heads. This transfer line was especially developed for the manufacture of engines for automobile and tractor. Ordinary industrial producers in this sector have lagged behind in such sophisticated machinery. However the Institute faces serious disadvantages. As a research institute, it has no experience with commercial ventures. The management of their trial production facilities was said to resemble the management of laboratory experiments. The internal departments have until recently been organized with a specialization by technological field, rather than by business tasks. In short, the Institute is still operating like an institute. There are also general ‘external’ difficulties, such as deficiencies in the financial institutions through which the Institute raises the funds necessary for market-oriented transformation. Current strategies Since the only viable future for the Institute is to become a commercial supplier of advanced manufacturing technology, the Institute has recently decided to reorganize its internal departments. Several R&D departments will merge with manufacturing facilities to form a number of profit-making departments for products such as transfer-line machine tools, electric devices and auxiliary machinery. The manufacturing facilities involved in these mergers have been acquired by the Institute since the 1980s, through various equity-related investments in some local machinery firms. The Institute headquarters will make decisions only on strategic issues. A central laboratory is to be created out of the previous R&D department for flexible manufacturing systems. This laboratory will be under the leadership of the centre, as the technological instrument for the Institute’s strategy. These changes will result in a multi-department, or M-form, corporation resembling the pioneer transformation of the Automation Research Institute of the Ministry of the Metallurgical Industry almost seven years earlier (see Part 1 Case Text 6.1). The multi-departmental corporation model seems to be favourable for exploring the commercial potential of diverse technological strengths, with strategic considerations remaining under central supervision. Another strategic intention of the Institute, complementary to its M-form restructuring, is to form links with a foreign partner similar to those established by the Dalian Machine Tool Works. It is not yet clear how this strategy is to be achieved, but it is clear that finding a foreign partner would change the Institute’s future, by enabling it to move into the ‘first world’ of the market, i.e., the more sophisticated flexible manufacturing technology required by the booming Chinese automobile industry. Otherwise, the Institute will be largely confined to the ‘second world’, producing moderately sophisticated technology for both large and small scale manufacturers. Sources: Interviews with Prof. Lian Zengbiao, and Prof. Pan Guishang, at the Dalian Modular Machine Tool Research Institute, October 1994; Introductory material provided by the Institute; Introduction to R&D achievements and manufacturing technology, mimeo provided by the Institute, October 1994.

The above example of the Dalian Modular Machine Tools Research Institute is typical of the transformation of the A institutes in the machine tools and tools sector, in that none of the seven institutes has merged with an existing enterprise. This is partly because of differences in ownership, since these

158

MACHINERY TECHNOLOGY INSTITUTES

institutes are owned by the central Ministry while the enterprises of the industry have all been transferred since the reform to the management of local governments. Another reason is that the managers in the Ministry felt that the strengths of these institutes could be better utilized, in coping with radical technological change, if they were not merged into a particular productive enterprise, because the enterprises in this sector are not yet strong enough to properly absorb the capabilities of these institutes (interviews with Mr Yu Chengting and Mr He Wenli, Beijing, October 1994). The Ministry therefore does not intend to promote such mergers. Direction and characteristics of the transformation in an area with rapid technology change The direction of institute transformation, as demonstrated by the case of the Dalian Modular Machine Tool Research Institute, is moving these institutes towards becoming independent suppliers of mixed engineering services and finished machine products. This direction of transformation is typical of the central R&D institutes in the machine tools and tools sector. From market earning data covering all the institutes in the machine tools and tools sector (Zhang Meiguang 1995),3 it appears that a number of institutes in the sector, such as the Beijing Research Institute for Machine Tools, the Guangzhou Research Institute for Machine Tools, and the Suzhou Research Institute for Electric Spark Machines (see Case Text 14.2), show a similar market earnings structure, characterized by a mixture of engineering services and trial production. For all the institutes mentioned, income from trial production, consisting largely of finished machine products, grew rapidly in the 1990s until it outstripped income from engineering services. Thus there is a clear difference, within the group of ‘centrally affiliated’ product technology R&D institutes, between the path of institutional transformation taken by the R&D institutes in the machine tools and tools sector and the paths taken in many other sectors. In these other sectors, the institutes have moved primarily to become suppliers of engineering services, as we will see in the following cases. For them, engineering services, consisting of varying mixes of ‘technology transfer’, ‘technological consultancy and technological services’, and ‘technology development’, have been the major technological activities, and the production of finished machines has not been a significant source of their income. Two factors may explain the distinct structure of marketing activities for the machine tools technology institutes: the rapidity of the technological change in this sector and the limited size of the domestic technological market. The market for engineering services in the area of machine tools technology is small. There are fewer small and medium-size producers of machine tools, and it is these smaller domestic producers which constitute the market, especially for engineering services, since the monopolistic position of the central institutes in providing product technology has come to an end, and leading manufacturers in the sectors concerned have begun to rely on foreign suppliers and their own in-house R&D. For modular machine tools there are approximately fifty small and medium-size producers, compared to roughly one thousand such producers in each of electric cable and internal combustion engine sectors, which will be examined below. One response which the institutes might make to the smallness of the engineering services market would be to enter engineering niches outside their specialized area so as to capitalize on their advantages, as the institutes which will be introduced in the following sections have done. This was not possible for the institutes in the machine tools sector, partly because their engineering accumulations were too narrow and have become rather obsolete due to the rapid technological change in this sector. The other possibility then is to engage in manufacturing some finished machine products, which entails more substantial investment in manufacturing-related assets which they did not possess and engaging in an activity in which they had little advantage. However the opportunities for this movement were rather attractive, both in the booming

‘PRODUCT TECHNOLOGY’ INSTITUTES

159

automobile industry whose requirements for machine tools are huge and in the recently increasing demand for auxiliary machinery. The weakness in the provision of auxiliary machinery in the Chinese machinery industry was a heritage from the previous institutional arrangement in which auxiliary machinery received little attention, since technological institutes and industrial enterprises were organized along the lines of recognized principle products. The shortage became even more pressing as auxiliary machinery became more complicated, and demanded more specific inputs, to support the rapid development of CNC machine tools. Thus far, both the provision of engineering services and the production of machine products by the institutes in the machine tools sector are characterized by a moderate degree of technological sophistication, falling roughly between the sophistication of foreign suppliers and leading domestic manufacturers. The engineering services, exemplified by the Dalian Institute’s modular unit design, based on the ‘I’ series standards (see Case Text 16.3), are relatively advanced, but still ‘conventional’. In many cases one could identify an imported technology underlying the engineering services, but this often incorporated an adaptation building on the technological accumulations of the institute. The engineering services provided by the institutes in the machine tools sector, and by institutes in many other sectors of the machinery industry as well, can therefore be seen as serving the function of promoting smaller domestic producers through disseminating appropriate technology to them. The sophistication of finished machine products is at a similar level. The transfer lines developed in the Dalian Institute and used for manufacturing the conical surfaces of automobile cylinder heads are reported to have found a market niche for lines more advanced than those produced by competing domestic suppliers, although their accuracy and productivity is still modest and conventional. The intermediate position which the institutes have won since the 1980s, as compared to their enterprise counterparts, is built on the fact that they have been better able to absorb flexible manufacturing technology and related auxiliary machinery, automatic devices and computer technology than these enterprises. Their better-trained manpower enabled them to learn imported technology more effectively. The leaner structure of an R&D institute makes transformation easier than in an industrial enterprise, and fundamental institutional restructuring has proved crucial in coping with the changes in technology and in the economic environment in which they operate. But both the R&D institutes and the leading industrial enterprises are all lagging far behind in their field. Given their weakness, and the rapid technological change in the sector, it is not surprising that both the institutes and the manufacturers are seeking foreign cooperation to enhance their technological upgrading. More investment is also being sought from government programmes, for instance funds for the establishment of ‘Engineering Centres’ (see Part 1). Changes in internal organization These moderate achievements in marketing engineering services and finished machine products have required continuous efforts from the institutes in internal organizational restructuring. For approximately the first ten years most institutes retained a department structure based on technological fields, as the case of the Dalian Modular Machine Tool Research Institute illustrates. During this time an internal contractual responsibility system was used and while the institutes were intensely searching for the market niches in which their core business could be formulated. Having identified these niches, some further institutional restructuring becomes possible and inevitable. The Dalian Institute, for instance, has recently decided to reform its internal work organization by moving to an M-form corporation model. Institute departments would, in the M-form model, be specialized by business tasks rather than by technological field, with decision-making power reallocated between two

160

MACHINERY TECHNOLOGY INSTITUTES

levels: the centre reserving the power to make decisions relating to strategically important developments, which for the Dalian Institute means those on more advanced flexible manufacturing technology, while the departments would be delegated the authority to make decisions in areas of relatively conventional engineering and manufacturing in which the institute already had an established and tested competitiveness. This move to an M-form structure, justified by strategic and evolutionary considerations, may be seen as an indicator of the general trends in changes in the internal organization of these institutes. Other opportunities, such as tapping new sources of funds or attracting an international partner, would seem inadequate to alter the trend in internal restructuring, as demonstrated by the case institute (interview with Profs. Lian Zengbiao, and Pan Guishang). While the several key institutes in this sector have moved to disseminate relatively advanced appropriate technology for smaller producers, and to some extent to supply machine products to fill some inherited gaps, it remains to be seen whether and how the machine tools sector in China can really successfully meet the great challenge it currently faces. The basic obstacle that has frustrated the sector in effectively adapting to heightened competition and rapid technological progress has not been dealt with. This obstacle is primarily institutional, and lies in the productive enterprises rather than the research institutes. The vertically integrated firm structure and highly centralized enterprise management (see Chapter 14), which were suited to the centrally planned system and slowly changing technologies, has prevented the enterprises in this sector benefiting from the enormous investment in technology imports since the late 1970s (see the discussion of exports and imports in Chapter 14). Management in the industry has perceived that ‘the old structure can no longer support an industry based on new technology.’4

The transformation of an ‘A’ institute in a sector with a concentrated enterprise structure This section examines the transformation of the Shanghai Electric Equipment Research Institute, which works in a sector with a concentrated enterprise structure, the power plant equipment sector. The sector has long been a top priority for industrial development in China, as described in Chapter 14 (see Case Text 14.5). As a result, the capacity of locally produced cutting-edge equipment rose to 300 MW by the mid-1970s. Great efforts continued in the 1980s to improve the domestically developed 300 MW equipment by drawing on imported technology. Because of the concentrated enterprise structure, industrial enterprises in this sector have been technologically stronger, and R&D institutes have not had such an exclusive role in the development of product technology as in other sectors. The sector also has a relatively high intensity of R&D activities performed in both productive enterprises and R&D institutes. Intensive R&D is indispensable for the development of electricity generation products. Experiments in aero-dynamics and thermo-dynamics, for instance, are necessary when a new model of electric generator is being developed. We visited the Shanghai Electric Equipment Research Institute and two equipment manufacturers, the Shanghai Turbine Works and the Shanghai Electric Machinery Manufacturing Works, along with a newly created corporation, the Shanghai Electric Corporation. The transformation of the research institute is therefore interpreted in the framework of the relationship between a producer of technology and its users.

CASE TEXT 16.4

‘PRODUCT TECHNOLOGY’ INSTITUTES

THE SHANGHAI POWER EQUIPMENT RESEARCH INSTITUTE, OF THE MINISTRY OF THE MACHINERY INDUSTRY Background: Institutional arrangement The Shanghai Power Equipment Research Institute The Shanghai Power Equipment Research Institute was founded in 1958, dismantled as an independent institute in 1969–1979 (when it was assimilated in several enterprises), and re-established in 1979. The Institute now has 800 staff, of whom about 600 are engineers or senior engineers. It is located in Shanghai, the biggest of the three production bases for power plant equipment in China. Except for the ten-year period in which it was split up and attached to several productive enterprises, the Institute has been the Ministry of the Machinery Industry’s research centre for power plant equipment and plant systems engineering. The Ministry’s Power Generation Equipment Testing and Inspection Centre and the secretariat of the Chinese Power Engineering Society are now based in the Institute. The objective assigned to the Institute when it was re-established in 1979 was the development of power plant systems engineering, whereas its original objective had been the development of turbines and boilers. Domestically developed 300 MW generation equipment had been put into operation in 1974, but serious reliability problems had emerged in operation, to the extent that it was necessary to suspend production in 1976. Clearly, better systems engineering was crucial. The Shanghai Turbine Works There are three major manufacturing plants in Shanghai, for three major elements of power plant equipment: turbines, boilers and electric generators. Only the first of these is covered here, because turbine manufacturing is the most complex of the three and the development of turbines is most closely related to the fields in which the Institute is engaged. The Shanghai Turbine Works, founded in 1953, has 7,500 employees, of whom 900 are engineers and another 300 are senior engineers. Annual sales in 1993 were 700 million yuan. Its major products are 125 MW steam turbines (domestically developed in the late 1960s), 300 MW steam turbines (both imported and domestically developed, but production of the domestically developed model stopped in 1994) and 310 MW nuclear steam turbines (domestically developed, evaluated by Westinghouse, and now in use in the Qinshan Nuclear Power Station in Zhejiang Province, China). Demand is high. By the end of 1994 the Works had thirtythree orders for 300 MW turbines waiting to be filled. The Works supplies one third of the domestic market. Some of their products have been exported, notably 300 MW turbines for fossil-fuelled power plants, going mainly to Asian countries. The Shanghai Electric Corporation The Shanghai Electric Corporation was established in 1985 at the initiative of the Shanghai Municipal Government. Its original purpose was to coordinate the local manufacture of power plant equipment which was being carried out separately by a number of equipment producers, so as to enhance their competitiveness in both the domestic and international markets. The Corporation’s character changed in the 1990s as it was assigned new powers. It now officially manages the state-owned assets invested in the Corporation’s subordinate companies, which include the three principal manufacturers of turbines, boilers and electric generators. The total assets are valued at about six billion yuan. However the Corporation’s power to determine resource allocation is restricted, since the subordinate companies are still largely independent ‘legal persons’. The reason is simple: the sons in this case had already been born before their mother arrived on the scene. The Corporation at present concentrates on exports, as well as domestic competition. It is thought that the Corporation can only be strengthened through joint efforts by the component enterprises to compete successfully in the marketplace. The subordinate enterprises accept that this structure is good for their individual development. The Corporation’s main business at present is exporting to Pakistan and the Philippines. The Shanghai Turbine Works’ turbine exports, which were mentioned above, are organized by the Corporation. The acquisition of foreign technology: from licensed transfer to joint ventures (1980–1994)

161

162

MACHINERY TECHNOLOGY INSTITUTES

The most important event for the sector during the reform period (1980–1994) has been the import of the technology to design and manufacture 300 and 600 MW fossil-fuelled power plant equipment. The agreements, with Westinghouse for turbines and generators, and with ABB (Asia Brown-Bowell) for boilers, were signed in 1980. In 1994 another agreement was signed between the Shanghai Electric Corporation and Westinghouse to establish a joint venture to produce power plant equipment, with Westinghouse holding 30 per cent of the equity of the venture. This raises questions concerning the rationale for turning first to the licensed transfer of technology, and later to a joint venture. Why was it decided in 1980 to turn to imports, when 300 MW power plants had been developed domestically and were in use? Why did the Corporation decide on a joint venture in 1994, if the assimilation of imported technology was going well, and there is a good market for the equipment? The 1980s technology imports The decision to import technology in 1980 was due to the low operational reliability of the domestically developed technology (Zhou Jian’nan (ch.ed.) 1990: Vol. A, 22). Accidents were especially frequent (the first 300 MW turbine was involved in seven accidents in its first year of operation, for instance). The reasons for this include the lack of systematic scientific data for design purposes, inferior manufacturing quality, and a weakness in systems engineering (interview with Prof. Shen Tianxi; Zhou Jian’nan (ch.ed.) 1990: Vol. A, 38– 39). The perception of these reasons provided the motive for re-establishing the Shanghai Electric Equipment Research Institute in 1978–1979. Experts from the Institute commented that the lack of systematic scientific data for design purposes illustrates a typical shortcoming of the Chinese centrally planned system in comparison with the Soviet approach: its failure to systematically incorporate scientific testing as a stage in technology development. The 1980 technology imports were therefore comprehensive (Zhou Jian’nan (ch. ed.) 1990: Vol. A, 33–35), covering technologies not only for design, manufacturing, quality insurance, the management of production, and plant systems engineering, but also for various auxiliary equipment and instruments. It took the forms of technology transfer licensing and cooperative manufacturing. The Institute contributed to the ‘intermediate’ assimilation of imported technology in two particular areas: equipment design techniques, especially for the design of turbines, which are the most complicated part of a power plant system, and the technology for plant systems engineering (interview with Prof. Shen Tianxi). The first of these was transmitted to the turbine manufacturer, and the latter was transmitted mainly to the plant design institutes affiliated to the Ministry of the Water Conservation and Electric Power Industry, which uses power generation equipment. The Institute’s extensive experience in scientific experiments and in dealing with plant accidents enabled it to play a special role as technological intermediary. The Shanghai Turbine Works, on the other hand, devoted their in-house efforts to assimilating most of the imported design and manufacturing technology. Some modifications were made, based on the imported technology (interview with Mr Wu Ye; introduction to Shanghai Turbine Works). This is exemplified by the development of a 905 mm long blade to be installed at the last stage of the turbine rotor, which has improved the heat efficiency of the turbine by 1 per cent. The 905 mm blade technology has been accounted for as a technological asset, as part of the Works’ investment in the new joint venture. It is interesting that, while manufacturers have considered the centrally affiliated R&D institutes in the machine tool sector to be rather ineffective in assimilating imported technology, the Shanghai Power Equipment Research Institute is recognized by the Works as among their most important external supporters. The Works has relied on two groups of external supporters, domestic and foreign. The domestic supporters are, in addition to the Shanghai Power Equipment Research Institute, the Shanghai Jiaotong University and a few of the centrally affiliated industrial R&D institutes in industries other than the machinery industry. These domestic supporters have helped mainly with scientific experiments and calculations. The foreign supporter is Westinghouse, which has provided ‘clarification, elaboration, and confirmation’ in the development of the long blade by the Works (Interview with Mr Wu Ye). In parallel with the acquisition and adaptation of imported 300 MW technology, the domestically developed model was improved throughout the 1980s and early 1990s. Its development became increasingly intertwined with the imported model, since the comprehensive technological imports provided an opportunity for all-round

‘PRODUCT TECHNOLOGY’ INSTITUTES

technological assimilation. It was eventually decided to stop production of the domestically developed model, when the joint venture with Westinghouse was agreed, but this may have been partly due to pricing considerations. The imported model could be sold for a much higher price, perhaps in part because of the foreign brand name. The improvement of the domestically developed model was coordinated under the planning approach as a ‘key S&T project of the five year plan’, with the Institute being heavily committed to contribute to the improvement work (interview with Prof. Shen Tianxi). This way of dealing with the organizational separation of the Institute from industrial manufacturers continued until the late 1980s. The 1994 joint venture The 1994 decision on a joint venture with Westinghouse means that, in practical terms, the Research Institute will have no direct connection with the development of leading edge electricity generation technology. The Institute however was circumspect in its comments, maintaining that the technological disadvantage of the Chinese power plant equipment sector had been significantly narrowed during the previous fifteen years. The explanation given by the Works for the decision to enter the joint venture started with a distinction between three aspects of the technological gap: in design technology, in manufacturing technology, and in management techniques (interview with Mr Wu Ye). It was said that the gap in design technology had been largely, but not completely, narrowed, that some progress had been made in manufacturing technology, and that the gap in management techniques remains wide. The joint venture was expected to accelerate the pace of development in all three aspects. The gap in manufacturing technology was most frequently cited to illustrate the benefits to be expected from the joint venture (interview with Mr Wu Ye; Yuan Baoshan 1994). It was said that under the planning approach as this was applied to the Key S&T Projects of the five year plan, the first priority was usually to develop the product, and the manufacturing technology to produce it was often overlooked. The Works themselves had to produce a lot of the complementary manufacturing technology, much of it being down-grading substitutions. Manufacturing technology to maintain efficiency and quality was particularly seriously neglected, and this was the area that saw the most dramatic improvements from the imports of the 1980s. The introduction of the Ingersoll (USA) side entry rotor slotting milling machine, which was especially developed for manufacturing rotor blade slots, was said to have been crucial for improving manufacturing speed and accuracy. Manufacturing technology in the Works had begun to improve as never before, in combination with the expansion of international contacts through technology imports. The Works expects the joint venture to place it in continuous and intimate communication with the American venture partner in all aspects of technology: for product design, for manufacturing, and for management. The ultimate purpose is to learn the Westinghouse approach to technological innovation and the skills required to compete commercially. This is thought to require close and long-term cooperation: static transfers or the intermittent exchange of technology are thought to be inadequate. With this goal in mind, a domestic partner such as the Shanghai Power Equipment Research Institute might be considered helpful in learning product design, but not for manufacturing technologies and management techniques. Transformation of the Shanghai Power Equipment Research Institute Trends in contractual activities The Shanghai Power Equipment Research Institute had heavier government-imposed obligations than many other A institutes, because of its role in the planned project to improve 300 MW power generation equipment. This project was a major part of the Institute’s work throughout the 1980s, not only in monetary terms, but also in terms of the Institute’s resource allocation. During this period about 60 per cent of the researchers were assigned to the ‘vertical’, or government-planned and funded, projects (interview with Prof. Shen Tianxi). This was reflected in the income structure of the Institute, as shown in Table 16.3. In the 1980s government funds provided a significant but declining part of the Institute’s income, and the market earnings were composed of different items in different years in no apparent pattern, implying that marketing was a rather random activity. This altered in the 1990s, when government funds began to be outweighed by market earnings, with earnings from ‘technological consultancy and technological services’ growing strongly and consistently, while other

163

164

MACHINERY TECHNOLOGY INSTITUTES

market activities such as ‘technological development’, ‘trial production’, and ‘other production and sales’ grew more moderately and erratically. By 1994, technological consultancy and technological services provided 70 per cent of the Institute’s total income, and the total income had grown to almost ten times that in 1984, while government funds had fallen in absolute terms. Table 16.3 Change in income structure 1984–1994: Shanghai Power Equipment Research Institute (million yuan)

Source: Mrs Zhang Meiguang, 1995. Note: Amounts are not corrected for inflation. From 1989 on, the total income includes ‘other income’ (see Table 15.1) which includes support, donations and credits from domestic or international societies or individuals. The income from technology consultancy and technological services derives from two major activities. One is small power plant engineering, delivered in a highly completed or packaged form. There are an enormous number of users, mainly industrial enterprises with high electricity consumption, such as producers of fertilizers, gasoline and petro-chemicals, who need their own small power plants. In 1993, the gross value of this business was 200 million yuan. The other activity under this heading is quality inspection services. In addition to being the Power Generation Equipment Testing and Inspection Centre, the Institute is developing its activities as a commercial inspector for power plant construction, to exploit its strength in electricity generation equipment and systems engineering. The customers for this business are mainly the ‘first world’ users, that is the large capacity power stations. As international investment increases, this kind of service may flourish. Furthermore, a joint venture with a foreign company has recently been created to underwrite power plant safety insurance, implying that the Institute’s consultancy and service activities might soon expand into the field of power plant safety insurance. Trial production activities have thus far concentrated on testing equipment and auxiliary equipment for the main generators. Both kinds of business have resulted from the assimilation of imported technology, and both are weaknesses in China’s plant power equipment industry. The products selected for trial production have a high intensity of technological know-how and are demanded in small quantities. Contractual technology development is mainly commissioned by big power plant equipment producers such as the Shanghai Turbine Works, and is a relatively small item. Technology development, for both the market and for government-funded projects, now accounts for about 10 per cent of the Institute’s income.

‘PRODUCT TECHNOLOGY’ INSTITUTES

Current strategies The Institute has not yet formulated an explicit strategy for its further development, nor determined what internal restructuring may be required. They are still at the stage of trying various options, and, as the Director said, ‘the future situation is still not very clear’. Research departments organized by technological specialization have been retained, but an internal responsibility system has been established, basically to manage the relationship between research departments and the Institute centre. Significant autonomy is delegated to the research departments in contractual activities, with the centre holding final authority in internal allocation and re-allocation. It was commented that the internal responsibility system is useful at present since the Institute, which is unable by itself to raise enough funds, has to rely on research departments to capture various contracting opportunities. As they respond to external opportunities, research departments are in fact moving beyond their previous specializations, leading the Institute pragmatically in the direction of becoming an independent engineering service company, despite the lack of an agreed strategy. This would not be the Institute’s optimal choice if the alternatives could actually be realized. It was said that the most profitable needs in the market do not require the most technologically sophisticated skills. Joining a big commercial company is seen as necessary to preserve the Institute’s technological sophistication, which will mean aligning its own development with the company’s long-term strategy. It is particularly notable that this institute has adopted such a pragmatic perspective, since it was greatly frustrated by the experience of being split up and merged with several factories during the Cultural Revolution. They have wisely recognized that the situation has changed. Two alternatives have been discussed, but none has been agreed on. One is to join the Shanghai Electric Corporation as the Corporation’s central laboratory. The Institute, the Shanghai Turbine Works, the Ministry of Machinery Industry, and the Corporation all admitted that this would be a reasonable arrangement (interviews with Prof. Shen Tianxi, Mr Wu Ye, Dr Zheng Hong, and Mr Zhu Sendi). But the Corporation, pointing to its financial shortfall, was unenthusiastic. The Ministry has not actively promoted a merger, partly because of the difference in ownership: the Corporation is under the supervision of the Shanghai Municipal Government, while the Institute belongs to the Ministry. The other alternative was to be a host-country research centre for Asia Brown-Bowell (ABB). ABB was said to recognize the Institute’s strengths in highly qualified manpower and its experience in computer programming, materials, plant systems, and accident analysis (interview with Prof. Shen Tianxi). These are attractive advantages for a multinational seeking long-term penetration of the Chinese market. But the Institute was hesitant about accepting complete foreign control, and ABB has withdrawn. It seems that a continuation of the market-led transformation is the most probable fate for the Institute, if no substantial action is taken in the near future by either the Shanghai Electric Corporation or the Ministry of the Machinery Industry. This market-led transformation is tending to lead the Institute away from the central arena of the development of large power plant systems, allowing it to lag behind. If this continues, the Chinese electric equipment industry may lose one of its best R&D bases. Although the engineering services it has provided as an independent institute have yielded great social benefits, services of this kind could probably develop alongside more sophisticated activities if the Institute could find a way of maintaining a position as a central R&D institution working on the frontier of capacity development for China’s power plant technology. Transformation of the Shanghai Turbine Works The Works is now concentrating on preparations for linking up with Westinghouse. Leaner staffing is required by the joint venture, which is planned to have 4,000 employees, while the Works had 7,500 employees by the end of 1994 (and 9,000 three years earlier). Several measures have been taken, such as encouraging employees to leave and take earlier retirement and restructuring internal supporting services as independent entities. The Works has established a training centre for employees who are out of work because of the adjustments. Another strategic move is reducing the Works’ vertical integration so as to concentrate on precision processing and maintain high quality. Domestic and international procurement now accounts for about 60 per cent of direct input into manufacturing in terms of value, up from the previous level of about 15 per cent (Zhang Jianming 1994; interview with Mr Wu Ye). Apart from normal market procurement, subcontracting for

165

166

MACHINERY TECHNOLOGY INSTITUTES

parts is being developed but is still rather limited. The Works seems to prefer to create equity-involved local joint ventures for several reasons: it gets better control of quality and delivery times, and can transfer its own redundant labour. This approach contrasts with the internal specialization approach adopted by the other firms in the machine tool industry which have been examined above. Two explanations can be suggested for the difference. The first is the Works’ location near Shanghai, in the most developed area of rural industry in China, which means that there are plentiful subcontracting resources. The other is that the joint venture with a foreign company includes a rather tightly defined internal structure for the venture. Sources: Interviews with Prof. Shen Tianxi, Mr Wu Ye, Dr Zeng Hong, Prof. Chen Binmo and Mr Zhou Zhang, October 1994; Introduction to Shanghai Power Equipment Research Institute, provided by the Institute; Introduction to Shanghai Turbine Works, provided by the Works; Yuan Baoshan: Historical Development of Turbine Manufacturing technology in Shanghai Turbine Works (shanghai qilunji chang qilunji zhizao fazhang shi), fadian shebei (Power Plant Equipment), No. 8, 1994:12–15; Zhang Jianming: Modern Management of Procurement (xiandai caigou guangli), fadian shebei (Power Plant Equipment), No. 8, 1994:15–18; Zhou Jian’nan (ch.ed.) 1990: Vol 1.

Two approaches to transformation for R&D institutes in sectors with a concentrated enterprise structure The example of the Shanghai Power Equipment Research Institute is illustrative of the transformation of several key R&D institutes among the fifty-three centrally affiliated product technology R&D institutes in the machinery industry, and the difficulties they have faced. These institutes were previously assigned to serve sectors with very few producer enterprises. The institutional separation between product development and product manufacturing in these sectors was almost entirely a result of the centrally planned macroeconomic management. The most noteworthy of these R&D institutes, in addition to the case examined here, are the Changchun Automobile Research Institute, the Luoyang Tractor Research Institute, the Luoyang Mining Machinery Research Institute, the Xi’an Heavy (Metallurgical) Machinery Research Institute, and the Lanzhou Oil Machinery Research Institute. They have all been working with complicated machinery systems at varying levels of technological sophistication. Under the old system, intensive administrative coordination was responsible for bringing the machinery systems developed in these institutes into production. This involved huge follow-up investments in manufacturing, in continuous technological improvements, and in inter-institutional communications. Two approaches have emerged for the transformation of these institutes during the reforms. One is merging with a key manufacturer in the relevant sector. This was the target set by the 1987 reform policy and was first realized in the merger of the Changchun Automobile Research Institute into the No. One Automobile Factory in 1980,5 followed by a few mergers in the 1990s covering the Tractor, the Mining Machinery, and the Metallurgical Machinery Institutes. As has been shown in Part 1, institutional barriers were the most serious obstacles to merging. Among the diverse institutional barriers were the limited abilities of enterprises to absorb an institute, the differences in working cultures, and the disparity in ownership (being supervised by a central Ministry or by local governments, for instance). Intervention from higher level administration proved necessary to overcome the institutional barriers, especially since the labour and equity markets were not well developed. That intervention in fact underwrote the realized mergers just mentioned. This section examines the other approach, that is the market-led transformation of institutes previously specializing in sectors with concentrated firm structures. We have seen that market-led transformation is the most likely future for the Shanghai Power Equipment Research Institute. Another case which appears parallel is the Lanzhou Oil Machinery Research Institute. While this Institute recognizes that merging offers

‘PRODUCT TECHNOLOGY’ INSTITUTES

167

advantages in preserving its technological accumulations, through placing them within the framework of the long-run strategy of a large company, apparent opportunities to merge have appeared and disappeared during the 1980s and early 1990s, leaving the Institute even farther from realizing a merger. Before turning to address the direction and characteristics of the market-led transformation, it is worth drawing some lessons from the case of the Shanghai Power Equipment Research Institute concerning the timing of a merger, in cases where a merger is necessary. First, the planning approach which was retained in a major project to improve domestically developed 300 MW equipment can be seen to have caused the long delay before the Shanghai Institute searched for a better transformation strategy. Because it was occupied with meeting the target of the planned project, the Institute did not commit itself to the learning that is required by market reform until the 1990s, when some of its major users had formed their own technological links with multinational companies. By the 1990s, the Institute was regarded in the sector as being fairly competent in product design technology, but inferior in manufacturing technology as compared to the foreign companies, and almost incapable in management skills. It was skills in these three areas that the major domestic producer expected to obtain by linking up with a multinational company. As the disparity in the technological trajectories followed by the Institute and by its industrial producers widened, the opportunity to merge with one of them was missed, perhaps irreversibly. The resulting transitional cost has been high, for the Institute and for the sector as a whole. The second lesson is that the very complex institutional barriers have made the relevant ownership agents hesitate to intervene. So this case proves once again the limitations of market mechanisms in recognizing significant values that may be realized in the future, however effective the market mechanisms are in signalling current needs. The inability of policy coordination to act promptly to preserve assets of long-term importance was responsible for the missed opportunity, because such external institutional barriers cannot be overcome by the institute concerned, even if it has properly evaluated the situation. Direction and characteristics of the market-led transformation The market-led transformation is leading the Institute to become predominantly a supplier of engineering services, complemented to a much less extent by ‘trial production’ and ‘other production and sales’, as shown in the case of the Shanghai Power Equipment Research Institute. The engineering services are largely in the field of ‘technological consultancy and technological services’, particularly in small power plant engineering, and the services are provided to industrial enterprises with high electricity consumption. Other transactions under the heading of ‘technological consultancy and technological services’, include quality inspection services for big power station construction. These have recently expanded, indicating that these services are becoming more diverse. This market-led transformation entails a radical shift in the Institute’s user group, from key producers of large capacity equipment to users of both small and large power stations, a shift not surprising given the disparity between the learning experiences of the Institute during the reform period, and the learning of the key producers and given the very high transaction costs involved in selling their technological services to these key manufacturers. A similar shift in user group may have occurred at the Lanzhou Oil Machinery Institute, which is one of a very few cases that are close to the situation of the Shanghai example.6 The change in the Institute’s user group and the engineering fields in which it is active was made possible by its accumulated engineering capability. However the services to which it has shifted are reported to require only a moderate degree of technological sophistication. As a result, the Institute’s scientific and engineering resources are seriously under-utilized, a negative effect which is seemingly more significant for the institutes examined in this section than for institutes in many other sectors such as the machine tools

168

MACHINERY TECHNOLOGY INSTITUTES

sector. Only 10 per cent of the scientific manpower in the Shanghai Institute is now engaged in R&D work, which is commissioned by big equipment producers or from some limited government funds. Changes in internal organization Having delayed its entry to the technology market, the Shanghai Institute offers a clear case showing one stage of the internal reorganization necessary to adapt to external contractual relationships, a stage which in other institutes was generally completed some time ago. This is a stage at which the internal contractual responsibility system is introduced to replace previous centralized internal management. The internal contractual responsibility system is found to be a useful method since it prompts the research departments to actively explore contracting opportunities at a time when the Institute centre cannot, by itself, raise adequate funds for its operations. Research departments are still organized by technological specialization, but the contracts they are making are leading them to diversify their technological activities. There has been no radical programme to further restructure the decentralized internal contractual responsibility system because the Institute has still not identified what its core business will be in the marketplace. The transformation of centrally affiliated institutes in sectors with numerous producers This section examines the transformation of centrally affiliated R&D institutes in sectors employing largely conventional technology and with a large number of small and medium-size producers. Two cases were chosen for study: the Shanghai Electric Cable Research Institute and the Shanghai Internal Combustion Engine Research Institute. The sectors the two institutes are specialized to serve, i.e., the electric cable and wire sector and the internal combustion engine sector, both have very decentralized firm structures. Each of these sectors has seen the entry of more than a thousand small producers before and during the reform period. Most of the enterprises in these sectors are locally initiated and sited in small cities or towns. The technologies currently in use by the small producers in the sectors are rather conventional, reflected in the current domestic market which is still dominated by relatively low technology products. Yet highly sophisticated technologies for these sectors do exist and are being further developed. These institutes are therefore faced with the dilemma of keeping pace with advances in their field on the one hand while serving the small users appropriately on the other. The two cases should be considered as typical of the transformation of a sizable number of institutes in the group of centrally affiliated R&D institutes previously assigned to the development of product technology in sectors such as small electric machinery and apparatus, small agricultural machinery, conventional instruments, and various small electric-driven or manual tools. R&D institutes of this type account for about half of the fifty-three central product technology institutes. They share similar characteristics as regards the firm structure of the sectors they serve, and in the technology which underpins the services these institutes are committed to develop.

CASE TEXT 16.5 THE SHANGHAI ELECTRIC CABLE RESEARCH INSTITUTE, OF THE MINISTRY OF THE MACHINERY INDUSTRY Background: The Institute and the industry

‘PRODUCT TECHNOLOGY’ INSTITUTES

Founded in 1957, the Shanghai Electric Cable Research Institute has been the only institute engaged in research and development of cables, wires and cable-making machinery, and in cable plant engineering. Its work covers conducting, insulating and sheathing materials, cable and wire products and processing and testing technology for cable and wire production. It is affiliated to the Bureau of Electric Equipment of the Ministry of the Machinery Industry. The Institute’s engineering design department is licensed as an ‘A grade’ plant engineering design unit (specified for electric cable plant), an unusual designation for a few A institutes which combine the functions of product development and plant engineering. The Institute currently serves as the National Testing and Inspection Centre for Electric Wires and Cables on behalf of the State Bureau of Technical Supervision, as the Wire and Cable Certification Station on behalf of the National Import and Export Commodity Inspection Bureau, and as the representative of the Technical Committees for TC7, TC20, TC55, TC58 and the sub-committees for SC20A, SC20B, SC20C, SC18A and AC46C products in the Chinese National Committee of the IEC (International Electrotechnical Commission). The Institute has 800 staff, of whom 400 are engineers or senior engineers, 300 are technicians and 100 are workers (introduction to the Shanghai Electric Cable Research Institute). In the 1960s and 1970s, being the only technology supplier for the electric cable and wire sector, the Institute contributed to develop a wide range of cable and wire products which together met more than 90 per cent of domestic requirements. Notable among these achievements was the development of various aluminum materials as a substitute for copper, with which China is poorly endowed (Zhou Jian’nan (ch.ed.) 1990: Vol. 3, 70–71). This achievement established the Institute as one of China’s main research sites for aluminum materials. In addition, the Institute had long been playing a role similar to many other institutes in the standardization, testing, information, and planning for the industry, known as ‘organizing the technological work of the industry’. All of these services were provided free of charge. The electric cable sector has always had a decentralized firm structure, and it has become even more dispersed during the current reform because of the rapid development of small enterprises. In 1993 the sector encompassed about 1,300 producers, of which 1,000 were ‘township and village enterprises’ which produced about half of the total 27 billion yuan output in that year (China Machinery Industry Yearbook 1994: VI–28, VI–43), while in 1977 and 1985 the output of the industry was just 2.9 billion and 4.7 billion yuan, respectively (Zhou Jian’nan (ch.ed.), 1990: Vol. 3, 68). There are an enormous number of new and small entrants, since entry barriers in both monetary and human capital terms are not very high for conventional products, while prices set under the centrally planned regime provided relatively high profitability until the 1980s. As raw materials become expensive and the number of producers increase, competition is becoming intense. There is an increasing demand for better production technology but most small producers in the sector lack experience with technological change, nor are they wealthy enough to rely on foreign suppliers. Competition in providing technological services for the small producers comes from a small number of domestic equipment manufacturers whose capabilities are limited compared with the Institute (interview with Mr Gao Qingguo). There is a huge market, mainly for low-technology products. Demand for technologically sophisticated high quality products is still limited, usually raised from government procurement. This is partly because of the rather loose regulations in areas such as safety standards, which put less pressure on producers to improve technology. Until recently, the development of some special purpose cables was supported by government funds. Technology import and upgrading during the reform Technology imports have been important for the technological upgrading of the industry since the market reform began. The importation of technology has involved a great deal of duplication. Comparable technologies in areas such as the production of irradiation cross-linked cable have been imported dozens of times, to the extent that there has been excess productive capacity. It is reported that about one third of the domestic market for production equipment for electric cables and wire has been captured by foreign suppliers. The Institute did not receive direct investments for technology imports. Indirect investment came from government funds for the assimilation of imported technology and contractual fees from users who had imported technology and needed support for local adaptation. The Institute also gained access to foreign

169

170

MACHINERY TECHNOLOGY INSTITUTES

technology through providing user services to some multinationals (Introduction to the Shanghai Electric Cable Research Institute; interviews with Mr Gao Qingguo and Mr Yu Yunlong). Several service centres attached to the Institute have been set up, such as the Nokia-Maillefer Cable Machinery Service Centre, Hipotronics Cable Testing System Service Centre, and SICME Enamelling Machine and Testing Device Service Centre (with Siegma, Italy). The major service of the Institute seems to be providing maintenance for imported manufacturing and testing facilities (interview with Mr Yu Yunlong). The Institute’s main roles in relation to the technology imports seem to have been in interpretation and adaptation for imported cable and wire product and processing technology. Substantial adaptations were made in the imported production lines, mainly to use local materials, given that material inputs are a major factor in cable production, and to lower the level of automation or the processing speed in accordance with the low skill level of the labour used (interview with Mr Gao Qingguo and Mr Yu Yunlong). Lower degrees of automation are possible for conventional products in this sector, in which relatively moderate precision is required. The importance of material know-how in the sector, and the experience in using local materials, enabled the Institute to be fairly successful in localizing imported technologies through integrating imported elements with domestically developed techniques. The Institute has been disseminating relatively advanced technologies to meet increasing demands for the many small producers. Its ability to meet these demands results from the adaptation and combination of technology imports. Trends in contractual activities Table 16.4 Change in income structure 1984–1994: Shanghai Electric Cable Research Institute (million yuan)

Source: Zhang Meiguang 1995. Note: Amounts are not corrected for inflation. From 1989 on, the total income includes ‘other income’ (see Table 15.1) which includes support, donations and credits from domestic or international societies or individuals. Table 16.4 shows the market earnings of the Institute for the years up to 1993 for which statistics are available. Throughout this period market earnings have been dominated by combined ‘technology transfer’, ‘technological consultancy and technological services’, and ‘technology development’ (collectively defined as ‘engineering services’). For a few years in the 1980s the Institute’s income from ‘technology development’ was significant. That coincided with the Institute’s role in adapting imported technologies, an activity which

‘PRODUCT TECHNOLOGY’ INSTITUTES

presumably involves a certain amount of ‘technology development’. This was financed by both the government and technology importers, as has been mentioned. Since 1988 ‘technology transfer’ and ‘technological consultancy and technological services’ have increased rapidly, accounting for 80 to 90 per cent of their total market earnings. These two items of earnings have a close relation to the Institute’s growing role in disseminating relatively advanced technologies to small domestic producers. In absolute terms, the earnings from these items (‘technology transfer’ and ‘technological consultancy and technological services’) jumped from 0.47 million yuan in 1984 to 6 million, 10 million, and 19 million in 1987, 1990, and 1993 respectively.

Two undertakings typical of the Institute’s work under these headings were the development of aluminum and aluminum alloy cable technology and of oxygen-free copper cable technology. The first was mainly developed by the Institute itself, beginning in the 1970s with improvements in the 1980s. The second is basically the result of the assimilation of imported technology, with some additional innovations such as the continuous upward drawing of copper rod. Both technologies are used in cable manufacturing and supplied as highly completed or turn-key packages including material composition, processing equipment, testing devices and plant engineering. A client may contract for a complete production line costing two to three million yuan. Such sales are counted as ‘technological consultancy and services’ in the statistics. Some clients, mainly the larger and more capable producers, also buy just some software or hardware parts of the packages, such as material composition or testing techniques. In that case the contract would often be recorded as ‘technology transfer’ if it did not involve a significant amount of user-specific work.7 Such pieces of package might cost much less, say about one tenth of the price for a turn-key production line. Sales of various other elements of technological know-how which the institute has accumulated were often recorded as ‘technology transfer’, but none of these has been significant as a source of income (interviews with Mr Gao Qingguo and Mr Yu Yunlong; Introduction to the Shanghai Electric Cable Research Institute). The aluminum and copper technologies described above are characterized by a moderate degree of automation, medium processing capacity, relatively high quality and, because the wages of the Institute’s engineering manpower are low, a price which is less than half of the international level for the same processing capacity. These technologies are competitive on the domestic market, where the Institute has a two thirds market share. More than sixty production lines have been designed and built since the 1980s. These technologies are also likely to be competitive in some niche of the international market, which the Institute is about to enter. Current strategies The Institute’s current strategy is focused on expanding their marketing activities into selected manufacturing. This can be observed in the 1994 income figures (Table 16.4) in which the share of ‘trial production’, which is an approximation of selected manufacturing, has grown remarkably, from a negligible amount in 1993 to a level comparable to the sum coming from ‘technology transfer’ and ‘technological consultancy and technological services’ which was previously the pivotal income source. The Institute strategy is to extend its market activities from cable plant engineering alone to a combination of engineering and selected manufacturing. This strategy is based on the perception that the plant engineering alone offers insufficient scope for the Institute to capitalize on its technological strengths. The conventional nature of the current engineering services, though still required by the numerous small cable producers, makes little call on most of the Institute’s more advanced techniques and know-how. By engaging in selected higher technology manufacturing, the Institute hopes to position itself for the longer term, since the technologies selected for manufacture are those likely to be in large demand in the next generation, when the Institute will be ready to provide engineering services for other manufacturers as they begin to manufacture these technologies. Today’s cutting-edge manufacturing is expected to become tomorrow’s engineering services. This seems to be one justification for investing more in R&D than the current engineering service market would justify, although the investments are still rather limited. They have been managed through redistribution of the

171

172

MACHINERY TECHNOLOGY INSTITUTES

Institute’s current earnings (interviews with Mr Gao Qingguo and Mr Yu Yunlong). With this strategy, the Institute expects to move towards transforming itself into a ‘technology-intensive corporation engaged in appropriate advanced cable technologies’. Examples of the ‘selected manufacturing’ (interviews with Mr Gao Qingguo and Mr Yu Yunlong) include 1) some special fibre optical cables and communication cables; 2) some new materials such as ingredients for insulating compounds; 3) selected equipment or devices such as high voltage accessories; and 4) steel stays used for building suspension bridges. The suspension bridge stays are derived from electrical cable technology and are produced in a joint venture with three local companies. Being the first in China to produce this material, the venture now holds 80 per cent of the domestic market. The stays produced have been used in two big suspension bridges now standing across the Huangpu River in central Shanghai. (Interview with Mr Du Xueguo; Introduction to the Shanghai Electric Cable Research Institute; People’s Daily, Overseas Edition, 3 September 1995). Internal restructuring has been regarded as indispensable and is now the first priority for the Institute management. The goal is to turn the Institute into a multi-divisional corporation so that its institutional structure corresponds to the key businesses developed or being developed in both engineering services and selected manufacturing. The management expects the departments of the Institute, previously specialized by technological functions, to undergo a fundamental reorganization on the lines of key businesses. Although a detailed plan seems not to have been worked out, this will definitely entail splitting up and reintegrating technological activities and manpower among the departments and between the centre and the departments. Some technological fields may be eliminated and others enhanced or created due to the reorganization. In late 1994 to early 1995, when the survey took place, the most serious difficulty for the necessary internal restructuring was felt to be the weak authority of the Institute’s centre (interview with Mr Gao Qingguo). Even if the concept of a ‘technology-intensive corporation’ does not involve a greater concentration of authority for central management, the restructuring process itself does need some central power. There is a tension here, arising from the possible loss of freedom which, under the internal contractual responsibility system, has been enjoyed by departments and by individuals or small research teams. This provides another example of the ways in which market mechanisms alone, in this case the internal contractual responsibility system which is an internal market within the Institute, may not be able to ensure the optimal allocation of Institute resources. It remains to be seen how the Institute will overcome this difficulty. Sources: Interviews with Mr Gao Qingguo, October 1994, and Mr Yu Yunlong and Mr Du Xueguo, April 1995; Introduction to the Shanghai Electric Cable Research Institute, provided by the Institute; Statistics on the income structure of R&D institutes, Zhang Meiguang, 1995; Zhou Jian’nan (ch.ed.) 1990, Chapter 2:45–87.

CASE TEXT 16.6 THE SHANGHAI INTERNAL COMBUSTION ENGINE RESEARCH INSTITUTE, OF THE MINISTRY OF THE MACHINERY INDUSTRY Background: the Institute and the industry Founded in 1958, the Shanghai Internal Combustion Engine Research Institute has been the key institute in China engaged in the development of internal combustion engines of between 2 and 600 KW (3–800 horsepower) and the related technologies for the materials, components, testing and standardization of internal combustion engines. The Institute is affiliated to the Bureau of Agricultural Machinery of the Ministry of the Machinery Industry (see Chapter 14, ‘Centrally affiliated R&D institutes’). It now hosts the National Testing and Inspection Centre for Internal Combustion Engines, functioning on behalf of the State Bureau of Technical Inspection. The Institute is also represented on the TC70 Technical Committee of the Chinese National Committee of the International Standards Organization (ISO), and hosts the secretariat of the Chinese National Committee of CIMAC (International Council on Combustion Engines). The Institute now has a staff of 850, of

‘PRODUCT TECHNOLOGY’ INSTITUTES

whom 540 are engineers or senior engineers (introduction to the Shanghai Internal Combustion Engine Research Institute). In the 1960s and 1970s, the Institute contributed greatly to the expansion of production capacity and product diversification in the sector. It developed about one hundred different engine models, some 70 to 80 per cent of the models produced in these years (interview with Mr Wang Zhiqi). Most of these engines were used as power units equipped in agricultural machinery for drainage, irrigation and transportation, or installed in small electrical generators. Because of policies which encouraged local and rural industrial development in the periods 1958–1960, the 1970s, and under the current reforms, the industry has more than 200 finished product manufacturers. Most of these are small firms, with a few dozen firms which are larger and relatively capable in product diversification. There are more than 1,000 parts and components producers, which are even smaller than the finished engine producers (interview with Mr Wang Zhiqi). It is reported that in 1984 more than 40 per cent of engine components and parts were manufactured by parts and components manufacturers (Jing Xiaocun (ch.ed.) 1988: Part II, Chapters 2 and 3). This sector accordingly seems to be the only one in the Chinese machinery sectors that has developed an enterprise structure with some degree of subcontracting relationships. The small engine and components producers, except for the largest and most competent, constitute the market served by the Institute. Technology import and technological upgrading during the reform Since the reforms began, technological imports have been pervasive. Technologies and designs for a large number of engine models have been imported, and the production of these models now accounts for about one third of the market for finished engines. The Institute obtained technology directly from the English firm Ricardo, under agreements for ‘consultancy and technical services’ (see note to Table 14.12). It also learned about foreign technology through providing translation and interpretation services for other importing enterprises, and through commercial relationships such as providing repair and maintenance services for foreign companies. According to our interviewees (Mr Wang Zhiqi and Mr Weng Zuliang), the chief things the Institute learnt from the imported technology were the concepts and techniques regarding non-repair operation time, noise reduction, configuring the product for its final use, and the need to give more attention to real production conditions so that the designs generated can be properly produced. An engine is now considered in the design process ‘more as a system’ (Weng Zuliang), implying that the interrelatedness that is intrinsic to a complex mechanical product such as the internal combustion engine is now seen to include user requirements and manufacturing factors. Environmental factors such as exhaust emission have been considered, but not significantly incorporated into designs yet, since environmental regulations are still loose, and producers are unable to invest in such technology. The Institute’s design work has always involved rather intense testing and R&D, which seems to have been further enhanced since coming into contact with the new concepts and techniques (Introduction to the Shanghai Internal Combustion Engine Research Institute). Trends in contractual activities As the data in Table 16.5 demonstrates, the market earnings of the Institute, like those of the Shanghai Electric Cable Research Institute, are dominated by a combination of ‘technology transfer’ and ‘technological consultancy and technological services’ in most of the years covered by the statistics. But there are differences between the two institutes. The total earnings of the Engine Institute are much lower than those of the Cable Institute, and the composition of its income from contractual activities is different. In the first place, for the internal Combustion Engine Institute, average earnings from ‘technology transfer’ are some 40 per cent higher than earnings from ‘technological consultancy and technological services’, whereas for the Electric Cable Institute the average incomes from these two sources are roughly comparable. Moreover, in the years in which the Engine Research Institute did earn a substantial part of its income from technological consultancy and technological services, this seems to relate to the provision of some engineering services for automobile producers in addition to its regular customers among engine manufacturers (see below). Second, the Internal Combustion Engine Institute has not had consistent income from contractual ‘technology development’

173

174

MACHINERY TECHNOLOGY INSTITUTES

comparable to that enjoyed by the Shanghai Electric Cable Research Institute in the mid-1980s, except for a few extraordinary years which also seem to correspond to some ‘derived’ opportunity, such as work for a foreign company (interview with Mr Wang Zhiqi). The fact that ‘technology transfer’ accounts for a higher proportion of its income than ‘technological consultancy and technological services’ or ‘technology development’ resulted largely from the circumscribed demands of the small and inexperienced users of its engineering services, a problem aggravated for the Engine Institute by the separation of product engineering from plant engineering (see Chapter 14). Note that the Shanghai Electric Cable Research Institute was not specialized only in product technology development. It was also, and unusually, licensed as a ‘plant design unit’. Table 16.5 Change in income structure 1984–1994: Shanghai Internal Combustion Engine Research Institute (million yuan)

Source: Zhang Meiguang 1995. Note: Amounts are not corrected for inflation. From 1989 on, the total income includes ‘other income’ (see Table 15.1) which includes support, donations and credits from domestic or international societies or individuals.

The circumscribed demands may be illustrated through an outline of the way in which the Institute deals with clients for its engineering services (interview with Mr Weng Zuliang). For each client, contractual negotiations start with the definition of contract goals, centring on two elements. One is the use to which the engine to be designed will be put: where it would be installed and in what kind of equipment? The other is the manufacturing conditions: what machining and processing facilities would be employed to manufacture the designed engine? Both elements require two-way communications, in which the Institute introduces information about recent domestic and international developments as a frame of reference, so that the client can formulate an investment strategy which will in turn be the basis for the contract. An experienced client with substantial investment capital may require a great deal of customization of the engine design, in which case the project would very probably go through the following phases: experimentation and concept design; evaluation of the concept design; detailed design; prototype assembly and testing; revision of the

‘PRODUCT TECHNOLOGY’ INSTITUTES

175

design if necessary, and so on. The Institute has recently carried out one such customization project for an American company. Such a project would be recorded as ‘technology development’. Most domestic clients required no specific customization, so that contracts with these clients would be recorded as ‘technology transfer’, i.e. providing designs from the blueprint archives. Some domestic clients did require specific services, often minor, which would be arranged under contracts for ‘technological consultancy and technological services’, i.e. providing archive blueprints with some tailoring to specific user requirements. (This can be compared with the Cable Institute, where many plant engineering services are recorded as ‘technological consultancy and technological services’ with a much higher profit per transaction.) Interviewees pointed to the poor manufacturing facilities and inferior management among engine and parts producers, although improvements have been taking place. Similar reasons may have restricted the ability of importers of foreign technology to assimilate it. It was said that some importers did not even study the technical documents carefully, so they also did not require the Institute’s help in interpreting them. To support inexperienced small producers, the Institute has had to renew its ‘off-the-shelf designs largely from its own investment funds. One such design is the model S195 engine, which is used in small vehicles and tractors, and is mass-produced. It is apparent that the inadequate return from low-level users’ requirements has threatened the Institute’s ability to continue to contribute to upgrading engine technology for small producers. Current strategy Since the late 1980s the Institute has been attempting to extend its activities to the trial production of selected products in the areas of special purpose machining equipment, testing equipment and new materials such as ceramics (interview with Mr Wang Zhiqi). Batch production, which falls under ‘other production’, has clearly increased since the 1990s. In addition, the Institute has been seeking to expand its engineering services. Recently its customers have occasionally included manufacturers of products such as cars and motorbikes. These steps towards diversification are basically defensive measures, developed in response to the low-grade and less-expensive demands currently made by small engine manufacturers. It is hoped that a certain level of engineering ability in the area of internal combustion engines can be maintained, through investing the additional earnings from its expanded activities in engine R&D and testing. Expenditure on R&D and testing is increasing (interviews with Mr Wang Zhiqi and Mr Weng Zuliang). No overall internal restructuring scheme has been formulated, partly because the Institute does not really need to do so. Its core business will remain the development of small and medium-size internal combustion engine technology. It is expected that about half of the research staff will continue to work in this core field, no matter what kinds of additional activities may be attempted. The structure envisaged for the Institute was likened to that of the Ricardo company, with a rather large and solidly-based central body devoted to engine engineering while derivative activities are organized as affiliated companies. Sources: Interviews with Mr Wang Zhiqi, October 1994, Mr Weng Zuliang, Mr Tu Zhuzong, April 1995; Introduction to the Shanghai Internal Combustion Engine Research Institute, provided by the Institute; Jing Xiaocun (ch.ed.) 1988: Chapters 2 and 3.

This section has examined two cases, the Shanghai Electric Cable Research Institute, and the Shanghai Internal Combustion Engine Research Institute, as exemplifying the transformations of the biggest group among the centrally affiliated machinery product technology R&D institutes. This group of institutes serve tens of thousands of machinery producers, promoting production technology in various areas of ‘small’ machinery—electric equipment and apparatus, agricultural machinery, conventional instruments, and electric-driven and manual tools. Such technologies are not as spectacular as the machine tools and ‘big’ machinery which were examined in the previous sections, but the roles played by the institutes of this group are crucial in the machinery industry, in which small producers constitute a large part. The small producers have built up a great variety of productive means which are vital to the economy, and the competitiveness

176

MACHINERY TECHNOLOGY INSTITUTES

of the Chinese machinery industry as expressed in export marketing has very largely stemmed from small rather than big machinery. The cases studied here shed a light on how the institutes of this group are transforming themselves in response to the market reform, and how they have remained the mainstay of the small producers. Some specific policy implications may be derived for sustaining the institutes of this group in their services to the small producers. Direction and characteristics of the transformation to supplying engineering services in the technology market The two cases illustrate the general direction of transformations of R&D institutes in this group, towards becoming suppliers of product and plant engineering for small domestic machinery producers. Contracts were typically for ‘technology transfer’, that is, offering off-the-shelf designs, or for ‘technological consultancy and technological services’, where the design required some degree of modification for a specific user. This differentiates this group, ambiguously but still recognizably, from the machine tools institutes whose transformation has led them to supply mixed engineering services and finished machine products, and from institutes working with large machinery systems such as power plant equipment whose transformation, if they could not merge with a key manufacturer in the sector, has led them to selling engineering services to new clients, mostly small users from other sectors. The group of institutes examined here have continued to disseminate relatively advanced production technology for their sectors, with their user-group shifting to some extent to concentrate more on smaller producers, leaving aside a minority of the bigger and wealthier producers which have turned to foreign sources. The many producers recognize and appreciate the value of these institutes’ continuous presence as a source of production technology. Their independence from the users of their engineering services has been reinforced by the presence of the new producers who have entered many ‘small’ machinery sectors during the reform, and by market intermediation, which has made it easier for the institutes to diversify their contractual activities. This group thus offers a chance to see clearly the characteristics of engineering services, and what has changed, where the providers have continued to provide the same services through the technology market which they previously provided through administrative coordination. One change which has resulted from market intermediation is that the institutes seek to provide engineering services in more completed forms. In both the cases examined, it is notable that entire production lines or entire product systems have now had to be developed, embracing material composition, processing equipment, testing devices and, in some cases, plant engineering. The segmentation of various assigned technological functions under the previous institutional arrangement is therefore being overcome. One explanation for this movement may be the need to reduce transaction costs, which is the topic of Chapter 18. Furthermore, market mechanisms have obliged engineering services to meet users’ requirements. As a result, the focus of the institutes has shifted significantly to real demands rather than achieving a particular technological advance. Besides, factors in relation to the efficiency of the users of their engineering services are becoming sensitive, such as non-repair time, noise reduction and product configuration. These are now encompassed in the engineering design process, because in a market distribution system product quality matters to both producer and designer. This is a fundamental departure from the design approach taken before when producer-specific factors were not clearly conveyed to the engineering design process. The greatly increased sensitivity to user needs is one significant characteristic of what might be called a shift in ‘technological trajectory’, induced by different incentive structures. We will return to this topic in Chapter 17.

‘PRODUCT TECHNOLOGY’ INSTITUTES

177

These changes in the character of the engineering services have demanded intensive technological learning. Direct and indirect access to technology imports has been important for the technological learning, since the imported technologies embody the missing elements in these institutes’ technological accumulations, as has been mentioned. The relatively profound technological experience built up in the past has been another factor in the institutes’ ability to make some creative adaptation. These institutes have to a greater or lesser extent been combining imported technology with domestically created technologies in a form more appropriate to local producers. Their ability in adaptation and combination in turn strengthens the competitiveness of their engineering services in a rather open technology market. Policies specifically required to sustain the transformation We have seen that the effects of market intermediation for engineering services in sectors with many small producers have been basically positive. But the observations also reveal some weaknesses of market mechanisms which deserve to be addressed, given that the importance of the institutes of this group has often been insufficiently appreciated. The most serious threat to the sustainability of the transformation is that the many inexperienced small producers make only low-grade demands. Returns from low-grade services have in many cases proved to give inadequate returns on the investment needed to produce the engineering services which are currently being marketed. Accordingly some of the more marginal institutes in this group are likely to find it difficult or impossible to continue to supply these services, because of poor cash flows. Specific rescue funds are needed for those which are really contributing to small producers but are in a dangerous financial situation. Loosely stipulated regulations for safety standards in the electric cable sector and for exhaust emission standards in the internal combustion engine sector for instance, means that there is less pressure from regulatory structures on producers to improve their production technology. Tighter environmental regulations are not only necessary for the protection of the environment, it could also be a sensitive policy measure to stimulate demand for technological upgrading and thus deepen the related technology market. The need for greater integration of product engineering and plant engineering has a policy relevance regarding this group, as demonstrated by the case of cable engineering services in which the fruitful integration was in part due to the absence of institutional hindrances. The institute in question is also licensed as a plant design unit, and this has facilitated the translation of innovative product designs in terms of plant engineering. This highlights the need to coordinate policies for the reforms of ‘R&D’ institutes and of design institutes. The two sets of reform policies are now being formulated and implemented by two separate policy institutions, as described in Chapter 14. The current separation can hardly be helpful in overcoming the barriers to an optimal integration. Changes in internal organization The two cases broaden our perspective regarding the organizational restructuring of institutes, since there are explicit differences between their restructuring programmes. The Cable Institute seeks to restructure itself towards a multi-departmental form (M-form) corporation with each department engaging in one core business field that has already been tested or is to be explored in the market place. The Internal Combustion Engine Institute does not have such a programme and there seems no need to departmentalize its engine engineering in connection with its institute development strategy. One of the basic reasons for the differences is the degree of multiplicity or interrelatedness of the core business areas covered by an institute’s external contracts. The highly interrelated nature of an engine

178

MACHINERY TECHNOLOGY INSTITUTES

system and the greater intensity of design and testing work required for engine development, which makes it necessary to keep a large part of the institutional structure under the coordination of the institute centre, very probably explains the rather centralized unified form (U-form) strategy chosen by the Engine Institute for its internal organization. This contrasts with the situation of the Cable Institute. Another difference can be observed in the strategic role that manufacturing activities are expected to play for the two institutes as they attempt to expand. For the Cable Institute, small batch manufacturing is regarded as a strategic step enabling it to design for and test the next generation of engineering services. The Engine Institute envisions manufacturing more as a means of raising income to complement the inadequate funds available for engine R&D. Thus the Cable Institute is to organize selected manufacturing activities in one or more departments, alongside the department of plant engineering, whereas the Engine Institute is placing its manufacturing in companies affiliated to the large centre. Indeed at least half of this Institute’s engineering manpower has been and will be retained for engine engineering. The transformation of machinery technology R&D institutes affiliated to local governments In this section two case studies are used to illustrate the transformation of machinery technology R&D institutes affiliated to local governments (the group III institutes in Table 14.12). There are more than 490 such institutes in the mid-1980s, supervised by local governments at the provincial and municipal levels. Unlike the institutes which were entirely or partly funded by the central Ministry, institutes in this group were not directed to specialize in developing certain kinds of machinery technology, but rather to provide general technical support to local planning administrations and local productive enterprises. However some design and testing work which was specific to the local industrial structure was carried out in these institutes. The two institutes visited are the Dalian Machinery and Electrical Research and Design Institute, which is affiliated to the Dalian Municipal Government, and the Zhejiang Mechanical and Electric Design and Research Institute, which is affiliated to the Zhejiang Provincial Government. These two were selected primarily because it was convenient to visit them, although it was considered desirable to have one sample institute affiliated to a municipal government and one to a provincial government, the two levels of local governments to which the group of ‘local affiliated R&D institutes’ affiliate. However we were told that these two institutes were known to have been successful in transforming themselves, in the sense that they had redefined their technological role and that contractual earnings were stronger (interview with Mr Zhu Sendi).

CASE TEXT 16.7 THE DALIAN MACHINERY AND ELECTRICAL RESEARCH AND DESIGN INSTITUTE (DMERDI), OF THE BUREAU OF THE MACHINERY INDUSTRY, DALIAN MUNICIPAL GOVERNMENT The Dalian Machinery and Electrical Research and Design Institute is affiliated to the Dalian Municipal Government, directly supervised by that government’s Bureau of the Machinery Industry. It was originally established in the 1950s and re-established in 1978 as part of the policy of recreating institutes which had been closed during the Cultural Revolution (See Part 1). Unlike the centrally affiliated institutes, the DMERDI did not have a specialized technological field, although it had accumulated some relative competence in areas such as heat treatment processes. Its major function was to support the diffusion of technologies to local

‘PRODUCT TECHNOLOGY’ INSTITUTES

manufacturers under the direction of the local administration. A significant part of the Institute’s manpower was allocated to duties known as ‘organizing technological work for the industry’, including standardization, (technical) planning, testing, and providing information. Before the current reform the administrative office of the Institute was large, with a staff of forty, while the Institute’s total workforce was, and still is, about 100. That large office was considered necessary for relating and communicating with the complex structure of superior organs within the planning system, although it was also said that the administrators were ‘very idle’. We were not able to form an exact conception of how the Institute worked during the planning era. Shut down in the period 1966–1978, the details of the prior period had been lost from the Institute’s memory. Changes during the reform During the reform, the Institute has changed entirely, from a government-financed technological agency to a commercial supplier of food storage and processing machinery and engineering, with coldstore engineering as the core business. A message from the central Ministry highlighting the opportunity of the cold-store business appears to have been critically important in the Institute’s 1984 decision to pioneer commercial coldstore engineering. We did not hear of any financial support from either central or local government for this shift, but the government funds previously allocated for the Institute’s daily operations were stopped in 1984. The Institute supplies fully packaged, turn-key coldstore engineering with key devices incorporated. The coldstores are provided under contractual arrangements. Eleven such stores have been constructed since 1984. Local suppliers have now captured most of the domestic market from foreign suppliers, and the Institute has one third of this domestic market. The Institute’s core business involves a moderate degree of technological sophistication. The Institute has know-how in relation to atmosphere-conditioning, which reduces the oxygen content using a catalyst reactor, an automatic humidity control device and some techniques for maintaining a slight positive atmospheric pressure in the store (interview with Mr Cai Weihua; Introduction to the DMERDI). The Institute learnt the latter techniques by imitating foreign devices, but developed the catalyzed atmosphere-conditioner itself with the help of the nearby Dalian Institute of Chemical Physics of the Chinese Academy of Sciences. Internal restructuring Radical internal restructuring took place to match the functional change. Technical work has been entirely reorganized around commercial tasks. Previous general-purpose technical departments have been replaced by the three departments defined by their business tasks: cold-store engineering, electric devices (used mainly for the cold-stores) and other food packing and processing machinery. Each of the departments incorporates necessary testing, design, engineering or manufacturing. The cold-store engineering department earns half of the Institute’s total income. In 1990 it was certified as a ‘B grade design unit’ for ‘conditioned-atmosphere coldstore engineering and auxiliary equipment’ (cf. the Shanghai Electric Cable Institute, which is clarified as an ‘A grade design unit’). Each member of the technical staff in the former technical departments was required to either relocate in one of the business departments, or to create a profitable new undertaking. The administrative staff has been cut radically, from forty to ten, and there is now one Institute Director with one Assistant, where previously there were four directors. Some of the redundant officers had personal ties with superior government organs, but the superiors seem to have tolerated the cuts. The administrative office was thus reformed to serve the Institute’s present objectives more effectively. Tasks such as standardization, testing, planning and information, which the Institute was previously assigned as part of its function in supporting the planning system, have almost entirely vanished or been reorganized. A separate Dalian Centre for Machinery Product Quality Inspection has been established elsewhere, functioning under the control of the newly-created product quality inspection system, which has taken over the testing work of the Institute. The information service, which has been placed on a paid basis, is now provided mainly for firms rather than for the Bureau. A plant design office of the Bureau of the Machinery Industry of the Municipal Government is still affiliated to the Institute, but operates rather independently from the Bureau. It provides about one third of the Institute’s income.

179

180

MACHINERY TECHNOLOGY INSTITUTES

On the other hand, as an influential body in the food machinery sector, the Institute now hosts the secretariat of the Association of Food Packing and Processing Machinery Producers. The Institute was said to be rather active in organizing the Association activities, which in 1993 included the First National Food Packing and Processing Machinery Fair (Dalian Daily, 25 September 1993). The Institute management seems to have been rather centralized throughout the transition process. Rather than delegating decision-making autonomy, its current ‘internal contractual responsibility system’ in effect stipulates profitability and related rewards for the departments, with the Institute centre holding absolute power in decision-making and resource allocation. The concentration of the core area of the Institute’s commercial activity, which the Institute entered rather early (1984), seems to have given the centre the strength to lead the transformation process forcefully. Sources: Interview with Mr Cai Weihua, October 1994; Introduction to the DMERDI, provided by the Institute; DMERDI: Detailed Stipulation for the Implementation of Internal Contractual Responsibility System, provided by the Institute; Dalian ribao (Dalian Daily), 3 May, 4 December 1992; and 25 September 1993; DMERDI: Project Specification: Conditioned Atmosphere Fresh Storage Coldstore.

CASE TEXT 16.8 THE ZHEJIANG MECHANICAL AND ELECTRIC DESIGN AND RESEARCH INSTITUTE (ZMEDRI), OF THE BUREAU OF THE MACHINERY INDUSTRY, ZHEJIANG PROVINCIAL GOVERNMENT Founded in 1958 as the Provincial Agricultural Machinery Technology Research Institute, the present Zhejiang Mechanical and Electric Design and Research Institute was renamed in 1988 after merging with the Provincial Research Institute of Mechanical Science and Technology in 1965 and with the (plant) Engineering Design Office of the Bureau of the Machinery Industry of the Provincial Government in 1985. This is an institute at the provincial level, under the direct leadership of the Bureau of the Machinery Industry of the Zhejiang Provincial Government. The Institute has a staff of 600, of whom 200 are workers and 400 are engineers or technicians, including 60 senior engineers (interview with Mrs Shi Jingxiong; Introduction to the ZMEDRI). Changes during the reform During the market reform the Institute has developed multiple functions, which are incorporated in departments for plant design, testing and manufacturing respectively. The plant design service has expanded considerably (interview with Mrs Shi Jingxiong). This has been a strategic objective, pursued by the Institute through the 1985 merger with the Engineering Design Office of the Bureau of the Machinery Industry, which led to the certification of the Institute as a ‘(plant) design unit’. The nearby No. 2 Design and Research Institute of the central Ministry of the Machinery Industry provides technical support (see Chapter 14). An international cooperative project on wind turbine generators with a Danish partner, which was sponsored by the EC and the Chinese State Science and Technology Commission, is reported to have helped the Institute to learn the concepts and methods used for plant engineering services such as feasibility studies, resource surveys and infrastructure evaluation. The users for plant design services are mainly local machinery firms, and the service is provided on a paid basis. Demand is high, since Zhejiang is one of the provinces with the most rapid industrial development. One third of the Institute’s 400 technical employees work in this field. Testing services have been deliberately enhanced (interview with Mrs Shi Jingxiong). The testing department has been licensed as part of the newly created national quality inspection system, which is coordinated by the State Bureau of Technical Quality Inspection and its provincial branches. Seven provincial centres for testing and inspecting mechanical and electric products have been set up in the Institute, most of them for small and lowvoltage electric devices such as electric motors. Technical support for this business comes from the Shanghai Scientific Research Institute for Electrical Apparatus, a centrally affiliated institute specializing in small and medium-sized electric motors (see Case Text 14.3). The inspection of such products has been tightened under

‘PRODUCT TECHNOLOGY’ INSTITUTES

181

the new national quality inspection system, especially for those devices which may cause risk to users or damage the environment. A considerable portion of the Institute’s technical manpower has been re-deployed in this area. The charge per inspection is low, but there is an enormous client base in the Zhejiang Province, where small and medium-size rural enterprises are booming. The income from this service is therefore not insignificant. Manufacturing has concentrated on some selected mechanical and electric products. The first is hydraulic cylinders, which had been developed and produced in small batches in the Institute’s trial production workshop earlier, and are now in large batch production. Annual turnover has increased from less than 1 million yuan in 1984 to 12 million in 1993. The other products manufactured include some food packing machinery, a few types of hotel electrical devices, and some spooling machines. These have resulted from various previous developments, whose production is now separately organized in a licensed company. One new product line which is expected to become a significant part of the Institute’s manufacturing activities involves equipment for resin-bound sand casting (ZMEDRI: Feasibility Study). A new company has recently been established and licensed as an NTE in the Hongzhou Development Zone, promoted by the national Torch Programme (see Part 2). The company aims to disseminate the resin-bound sand casting technology through selling integrated series of the necessary equipment, including the shake-over machine, vibrocrusher, pneumatic reclaimer, air-sand separator, sand temperature controller, continuous sand mixer and screened electric and pneumatic controllers. The market is huge, and has thus far been dominated by imports because the Chinese machinery industry has not been capable of producing such integrated series of machines, in which the individual devices must be compatible, even though no single one is very technologically complex. Encouraged by a preliminary but remarkable success indicated by earnings of 3 million yuan in 1992, the company aims to expand its capacity to the yearly output at 15–20 production lines by the mid-1990s. Another title, and a small amount of funds, came to the company in 1993 when it was designated as a Branch Productivity Centre for Machinery Technology, part of another national thrust to promote a framework of ‘productivity centres’. The products described above are all produced or intended to be produced in batch volumes. Hydraulic cylinder production has made by far the largest contribution to the Institute’s manufacturing, providing a significant part of the reserves the Institute needs to invest in transforming itself. Internal restructuring In the course of pursuing these diverse opportunities, an ‘internal contractual responsibility system’ has evolved with greater autonomy delegated to departments. For instance manufacturing, with the exception of the hydraulic cylinders produced in the Institute’s integrated workshop, is carried out by companies with the status of ‘independent legal persons’ in which the Institute is the shareholder. This entails an irreversible dispersal of central authority to individual departments, each with its own clearly delineated core business. The looser interdependence between the departments raised questions regarding the role of the Institute’s centre and the future of the Institute as an entity, at a time when the Institute’s Director was pondering what shape the Institute might take in the future. Sources: Interviews with Mrs Shi Jingxiong and Mr Lu Jianhong, June 1993; Introduction to the ZMEDRI, provided by the Institute; ZMEDRI: Product specifications, provided by the Institute; ZMEDRI: Feasibility Study on the Development of Complete Equipment for Resin-bound Sand Production, provided by the Institute.

Direction and characteristics of the transformation of R&D institutes affiliated to local governments As exemplified in the two cases, the transformation of the machinery technology institutes affiliated to local governments has led them, during the market reform, to become commercial manufacturers of selected

182

MACHINERY TECHNOLOGY INSTITUTES

machines, combined with plant and product engineering services. In addition to the cases given in this section, there is qualitative evidence from other locally affiliated institutes to confirm this analysis,8 although there are no quantitative statistics available comparable to the statistics on the centrally affiliated institutes. Nevertheless two features stand out for the locally affiliated institutes in comparison with the centrally affiliated group. First, manufactured products constitute a considerably higher, perhaps the largest part, of the output of the local institutes. If comparable statistics were available, this would be recorded in the income earned from ‘trial production’ and ‘other production and sales’, whereas the marketing activities of the centrally affiliated institutes (except for the institutes in the machine tools sector) examined in the three preceding sections are dominated by ‘technological transfer’ and ‘technological consultancy and technological services’. The products manufactured by the local institutes diverge much more widely, whereas the centrally affiliated institutes tend to be more restricted to specialized sectors, where they have expanded into manufacturing at all. The difference can largely be explained by the fact that each centrally affiliated institute was formerly assigned to developing one particular category of product, whereas the locally affiliated institutes’ functional assignments were broader. Consequently, the technological accumulation in the local institutes was more diverse and they were less bound to a particular sector. By entering selected niches with finished machine products, the transformation of the local R&D institutes may have generated a number of new entrants in some unexploited areas of the industry, where technological sophistication is not too high but which had been overlooked for development under the previous economic planning. The entry of the Dalian Institute to the coldstore sector and the steps taken by the Zhejiang Institute to fill the niche for resin-bound sand equipment provide support for this thesis. The policy relevance is that institutes could be more active in pioneering some spheres in which the Chinese machinery industry is not active if an array of policy measures could be formulated in which they were treated not as R&D institutes but as a special group of innovative small machinery firms. The second feature of the locally affiliated institutes is in the area of engineering services. The local institutes engage more in general plant engineering, as illustrated in both the Dalian Institute and the Zhejiang Institute. Local small machinery enterprises are their major users. The central institutes, in contrast, focus more on product engineering relating to their specialization, and their plant engineering, where they have developed it, was often in fact an expansion of complex product engineering. This difference may be attributable to the less rigid institutional separation established earlier between product R&D and plant engineering at local levels.9 Thus it is possible for local institutes to use their existing expertise in plant engineering to support their entry, in some niches, as manufacturers. Changes in internal organization Internal contractual responsibility systems appear to have been widely adopted by the locally affiliated institutes to provide greater incentives to respond to market opportunities. In other respects the changes in internal organization for this group of locally affiliated institutes, illustrated by the two cases, vary as widely as in the centrally affiliated institutes. The Dalian institute adopted a more centralized management model in accordance with its rather uniform core business. The Zhejiang institute had a more decentralized internal structure, with the ‘departments’ actually having the status of ‘independent legal persons’. It seems that, in line with the observations in the preceding sections, a more centralized internal organization may evolve when an institute develops a single core business, while more decentralized internal organization may arise when the development of an institute leads it into a number of fields which have little in common.

‘PRODUCT TECHNOLOGY’ INSTITUTES

183

It is reasonable to expect that larger local R&D institutes would tend to be more decentralized, since these institutes do not already have a core technological field in which they have accumulated a special strength, but do have a larger pool of engineering manpower for which they have to seek opportunities in more disparate fields. If each department develops its own core business along the lines suggested by market opportunities, the institute centre would be left as little more than a shareholder. The technological infrastructure Given that the local institutes are close to local enterprises, and were officially assigned to serve them, we were interested to see whether the market reform might have weakened the roles of a ‘technological infrastructure’ which is thought to be critical especially for small and medium-size enterprises. The limited but fairly convincing evidence observed in this section suggests that this has not occurred. For a partial review of the role of the institutes in this infrastructure, it is useful to differentiate the ‘infrastructure’ functions into two spheres: the technical support of production operations in a sector, and technological diffusion. In the first place, the two cases showed that functions such as information gathering, technical planning and product testing which fall under the heading of technical support were previously conducted chiefly to support planning coordination rather than to support industrial producers. The expansion and proliferation of the local institutes was in fact not required by the local enterprises but by the bureaucratic complications of the planning apparatus (Interview with Mr Zhu Sendi), which entailed enormous ineffectiveness as reported in the case of the Dalian institute. The support functions could not continue without fundamental restructuring. The restructuring of support functions during the reforms has been fairly effective. This observation applies not only to the local institutes but also to the B institutes (those which were centrally affiliated but based in a host enterprise) which were described earlier in this chapter. This restructuring has consisted of the creation of ‘producer associations’ to coordinate the supporting functions needed by industrial producers, and building up a ‘product quality inspection’ system to look after those supporting functions which concern consumer interests. Locally affiliated R&D institutes are taking various roles in the changed institutional framework for support functions, as exemplified in the Dalian Institute case, which now hosts the secretariat of the Association for Food Packing and Processing Machinery Producers, and in the Zhejiang Institute case, which now plays a considerable role in the testing and inspection of small electric devices, serving a mainly provincial market. The technology diffusion function has been largely put on a commercialized basis during the market reform. It is fair to argue that this function is being enhanced rather than weakened. These local institutes, which lacked their own specialized technological strengths, had little effective ability in technology diffusion before. As their expertise in technological innovation increases with their entry to some selected manufacturing during the reform period, diffusion from these local institutes, as well as from central institutes, has accelerated. Under market mechanisms, both the centrally affiliated and locally affiliated institutes are diffusing their know-how, both embodied and disembodied, more broadly (beyond their previously assigned sector or region), more intensively (as competition becomes intense) and more effectively (meeting the user’s real needs through a direct contract).

184

MACHINERY TECHNOLOGY INSTITUTES

Summary: an overview of the transformations of product technology R&D institutes in the machinery industry This chapter has surveyed empirical evidence of the transformation of machinery industry R&D institutes which were previously assigned to develop product technology. Three groups of institutes in this area have been examined: I) those which were partly centrally commissioned and based in a host enterprise, classified as the B institutes; II) those which were entirely centrally affiliated, the A institutes; and III) those affiliated to local governments. The focus in this chapter has been on the second of these groups. Since the whole innovation system for machinery technology as it was established was overwhelmingly bound to the development of product technology, this chapter has covered the largest part of this system. There remains only a small portion of the centrally affiliated institutes, devoted to manufacturing technology, which will be the subject of the next chapter. Case studies have been adopted as the chief approach. For the A institutes, cases were chosen in four sectors: machine tools, power plant equipment, electric cables and wire, and internal combustion engines, in the expectation that these would illustrate the characteristics of institute transformation and the factors influencing its course. Statistics with respect to the market earnings of individual institutes, available only for the A institutes, were used to specify the nature of the transformations which emerged from case studies in the field. Direction and characteristics of the transformation: a comparative perspective among three groups While the degree of ‘marketization’ in the different groups of product technology R&D institutes was similar, the directions of their transformations differ considerably. For the entirely centrally affiliated institutes (i.e. ‘A’ institutes, or group I institutes with functional assignment for product technology, as depicted in Table 14.12), the direction was towards becoming commercial suppliers of product and plant engineering services, along with some machine product manufacturing, except that in the machine tools sector manufacturing was the major activity and engineering services secondary. This nevertheless confirms the findings from the statistical analysis discussed in Chapter 15 (see Table 15.6) in which the centrally affiliated group as a whole was found to be characterized by a significant concentration on engineering services. The locally affiliated institutes (group III institutes in Table 14.12), in contrast, have moved primarily towards being commercial suppliers of selected machine products. This is associated with some kinds of engineering services, mainly general plant engineering. The centrally commissioned institutes based in a host enterprise (group II, also ‘B’ institutes in Table 14.12) have been integrated with their host enterprises, to become a key factor in that enterprise’s in-house R&D and design capability. This group now specifically serves the strategic objectives of the host enterprises, rather than a whole sector. If we compare the transformations of centrally affiliated institutes with those of the locally affiliated institutes the main differences, in addition to the extent to which engineering services and manufacturing are combined, is that both the engineering services and manufacturing carried out by the centrally affiliated institutes are considerably more specialized in a certain sector, while the manufacturing carried out by locally affiliated institutes is widely dispersed in any achievable niches and their engineering services are generally provided for plant construction. The difference is basically historical, resulting from the past institutional arrangement in which the centrally affiliated institutes were assigned to specialized product categories. This specialization allowed the centrally affiliated institutes to accumulate technological competence in certain fields more effectively than the local institutes.

‘PRODUCT TECHNOLOGY’ INSTITUTES

185

The institutes’ roles in technological innovation and diffusion differ correspondingly. The centrally affiliated institutes have, on the whole, become important in disseminating relatively advanced technology for the machinery industry, delivered chiefly through engineering services. They combine the imported with domestically developed elements through intensive technological learning and adaptation, and they serve the smaller domestic producers because larger and wealthier manufacturers have relied on foreign suppliers. The locally affiliated institutes, on the other hand, have become active as niche suppliers of some selected machine products chiefly through their own manufacturing. Market opportunities, more than any prior capability in specialized fields, appeared to have determined which niches they have entered. The unique group II institutes, having been integrated with their host enterprises, are playing a key role in the acquisition of imported technology by leading machinery manufacturers. In short, all the constituents of the former technological innovation system of the Chinese machinery industry have been, and are being, transformed in different ways. The sum result of these developments has been a fundamental reorientation of the system, from one charged with implementing centrally planned innovations and operating in an environment closed to international exchange, to a system which is coping with the opening to international trade and competitions, and becoming adept at combining domestic efforts with achievements developed abroad to take advantage of market opportunities. In all of these diverse transformations, the market mechanisms have functioned reasonably effectively, except in some cases where individual institutes appear to have been unable to overcome external institutional barriers on their own. The directions and characteristics of the transformations in various groups of the previously governmentrun R&D institutes for product technology in the machinery industry are summarized in Table 16.6. Factors influencing transformations: the centrally affiliated A institutes Three factors were supposed at the beginning of this chapter as likely to have an effect on the transformations: the complexity of the technology with which the institute in question deals, the number of manufacturing enterprises in the sector concerned, and technology imports in the sector. It would be useful to summarize findings under these three factors, focusing on the group I institutes, those which were entirely centrally affiliated. Of the three factors, technology imports were found to have had a significant effect in all of the cases examined. The supply structure for machinery technology has been Table 16.6 Direction and characteristics of product technology R&D institutes transformation in the machinery industry

Direction and characteristics of transformation

Roles in the changed R&D and innovation system

Group I (A institutes) Entirely centrally affiliated

Group II (B institutes) Centrally commissioned and based in a host enterprise

Group III institutes Affiliated to local governments

• engineering services dominate (with exception); • high degree of specialization bound to previous field; • serving mainly the smaller and less wealthy domestic enterprises. As major adaptors and disseminators of relatively

• Integration with the host enterprises as in-house R&D and design department; • serving the strategies of the host enterprises.

• manufacturing of machine products dominates; • specialization of manufacturing induced by market niche.

As a key element in leading As active niche suppliers of enterprises’ capabilities in selected machine products,

186

MACHINERY TECHNOLOGY INSTITUTES

Group I (A institutes) Entirely centrally affiliated

advanced product technology based on a combination of imported and domestically developed technologies

Group II (B institutes) Centrally commissioned and based in a host enterprise the acquisition of imported technology

Group III institutes Affiliated to local governments based on the active assimilation of imported technology

changed remarkably because of substantial technology imports. The differences in the transformation of institutes in the various sectors can therefore be ascribed mainly to differences in the first two factors. It is evident that institutes are more likely to be able to continue to function as suppliers of engineering services if they are in a sector with more producers, particularly producers of smaller size and moderate capability, since these constitute the major buyers in the engineering services, market. On the other hand, the more radical the technological changes involved in the sector, the harder it is for a domestic institute to compete with international suppliers in providing engineering services. These two factors together explain the differences observed in institutes’ abilities to sustain competitive engineering. At one end of the sustainability spectrum stand institutes in the electric cable and internal combustion engine sectors, where there are very many producers and the real demands for engineering services are not very complicated. Institutes in these areas have been able to continue as providers of product and plant technology. At the other end are the institutes in the power plant equipment and machine tool sectors whose positions appeared rather fragile. In the machine tool sector technological changes have been too rapid for the institutes, while in the power plant equipment sector there are too few producers to provide a stable market for the institute. The responses of these institutes differed: in the machine tool sector the institute has expanded to fill some niches in the market for finished machine products, while the institute in the power plant equipment sector was moving into a different but related segment of engineering, for small power plants. Table 16.7 summarizes the relationships between the size of the technology market in a sector, the rapidity of technological change in the sector and some characteristics of the transformation of the institutes in the sector. Note that, as already stated, those institutes previously assigned mainly to the development of materials are not considered in the summary, although in the official categorization they are regarded as ‘product technology’ institutes of the machinery industry. Table 16.7 Market size, technological change and the transformation of centrally affiliated product technology R&D institutes in the machinery industry

Size of technology market Rapidity of technology change Contractual items complementary: items 1, 2, 3 Characteristics of transformation

Machine tools sector

Electric power plant equipment sector

Electric cables sector Internal combustion engines sector

small

very small

large

large

high

moderate

moderate

moderate

dominant: item 4 complementary: items 4, 1 mixed machining system and auxiliary

dominant: item 3 complementary: item 4 small power plant engineering; have

dominant: items 3, 2 complementary: item 4 product and plant engineering,

dominant: items 2, 3

product engineering, accompanied by

‘PRODUCT TECHNOLOGY’ INSTITUTES

Machine tools sector machinery products, accompanied by product engineering

Electric power plant equipment sector moved out of large capacity power equipment engineering and plant engineering

187

Electric cables sector Internal combustion engines sector accompanied by miscellaneous manufacturing more diversification sophisticated products

Notes: Contractual items are: Item 1: technology development; Item 2: technology transfer; Item 3: technological consultancy and technological services; Item 4: trial production; Item 5: other production and sales.

Recreating the technological infrastructure The analysis of institute transformations reveals a perspective regarding some aspects of the reconstitution of the technological infrastructure in the context of macro-economic changes. We have identified several institutions that have been created during the reform to restructure the old infrastructure. These can be grouped along the lines of a differentiation made in a previous section, in which the functions assumed to be served by a technological infrastructure were divided into: 1) support to industrial producers; 2) support to users and social interests; and 3) technology diffusion or dissemination. These new institutions embrace the ‘producers’ associations’, created to coordinate the necessary support to industrial producers, the ‘product quality testing and inspection’ system, established to look after the interests of consumers, and the introduction of the technology market which has in fact been the major institution to intermediate the distribution of commercially profitable technology (both engineering services and machine products manufacturing diffuse technologies to differing extents). This restructuring of the technological infrastructure appears to have increased the specialization and effectiveness of functions which were previously assigned to a single and uniform structure. Naturally the different groups of institutes have different roles in the changed technological infrastructure. These have already been examined in the relevant sections, and are summarized in Table 16.8. Product engineering in a market context Product engineering, which had previously been assigned to the R&D institutes engaged in product technology, has been affected in some critical ways under the guidance of market mechanisms. First, the supply of product engineering has had to adapt to the real demands of users, which has led to a significant shift in technological focus. In most cases, more resources were devoted to conventional, but relatively advanced, fields. Second, there has been a widespread trend to provide complete or packaged engineering services. ‘Completing’ an engineering service implies either adding the complementary elements needed to realize the potential of a core technique, or expanding product services to combine them with plant engineering and training, or both.

188

MACHINERY TECHNOLOGY INSTITUTES

Third, the technological factors sensitive to economical operation of a product, or its ‘economy in use’ have begun to be incorporated into product and plant engineering, partly through renewing the concepts and skills of design engineering, and partly by intensifying communication with users. More intense communications were particularly important in contractual ‘technological consultancy and technological services’, which commonly involved some degree of tailoring to user requirements. Table 16.8 Functions of various groups of institutes within the reconstituted technological infrastructure for the machinery industry Function

Group I (A institutes) Entirely centrally affiliated

Support to industrial C producers new institution: producers’ associations Support to users and society C interests new institution: a system for product quality testing and inspection Diffusion or dissemination P of production technology new institution: technology market

Group II (B institutes) Centrally commissioned and based in a host enterprise

Group III institutes Affiliated to local governments

P

C

N

C

C

C

Note: The capital letters refer to the role played by each group in the changed technological infrastructure: P: principle role; C: complementary role; N: insignificant role.

These three aspects of the critical changes in product engineering all derive from a user-centred and marketbased orientation, in contrast to product engineering under the previous regime, which was plan-centred with the planned supplier being the key player. This shift in fundamental orientation required most of the 1980s to run its course, as can be seen from the fact that in almost all cases there was a gestation period before the institutes achieved a significant increase in the value of their contractual income, in the late 1980s or early 1990s. Such a fundamental transformation could be described as ‘path shifting’ from one technological trajectory to another, accompanied by internal reorganization of the institute’s work. Chapters 18 and 19 will return to these two aspects, following the study of R&D institutes for manufacturing technology in Chapter 17.

17 THE TRANSFORMATION OF THE ‘MANUFACTURING TECHNOLOGY’ R&D INSTITUTES

The manufacturing technology institutes This chapter presents some empirical observations regarding the transformations of a small group of centrally affiliated R&D institutes which were previously assigned to the development of manufacturing technologies. Three of the eight members of this group were visited in the course of this study: the Beijing Research Institute for Mechanical and Electrical Technology (BRIMET), the Beijing Research Institute for Automation in Machinery Industry (BRIAMI) and the Shanghai Materials Research Institute (SMRI). The first two will be discussed in this chapter. The third, as a materials institute, is atypical and will not be introduced here, but its income structure is presented in the appendix table at the end of the chapter. A brief description of all eight institutes of this group has already been given in Case Text 14.1. They are listed again for readers’ convenience: • • • • • • • •

Harbin Research Institute for Welding (HRIW); Shenyang Research Institute for Foundry Technology (SRIFT); Shanghai Materials Research Institute (SMRI); Wuhan Research Institute for Material Protection (WRIMP); Zhengzhou Research Institute for Machinery Engineering (ZRIME); Beijing Research Institute for Mechanical and Electrical Technology (BRIMET); Beijing Research Institute for Automation in the Machinery Industry (BRIAMI); Research Institute for Standardization in the Machinery Industry (RISMI).

The eight institutes were supervised by the Science and Technology Department of the Ministry of the Machinery Industry, and collectively comprise the Research Institute of Machinery Science and Technology (RIMST). The grouping is more or less artificial. Two of the eight specialize in materials (and material protection) and one in standardization, which do not, strictly speaking, fall under ‘manufacturing technology’. The RIMST is therefore better seen as a grouping of institutes engaged in generally applicable fields, other than ‘product technology’. The latter was defined by product category and supervised by the bureaux of the Ministry which specialized in sectoral affairs (see Chapter 14). Since the early 1980s the central Ministry has officially relaxed this functional division (interview with Mr Zhu Sendi), because, pragmatically, there is no boundary strictly dividing product technology from manufacturing technology. The system of planned investment was modified at that time so that an investment in manufacturing technology could be carried out in a product technology institute, and vice

190

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

versa. The trend to blur the distinctions between the types of institutes has accelerated during the reform as institutes reallocate their resources in response to market opportunities. All of the product and manufacturing technology R&D institutes are now extending their technological activities beyond their original assignments in some respect. However, the original functional assignment has had a significant impact on the operation and accumulation of the manufacturing technology institutes in at least two respects. First, the R&D intensity in these institutes has been considerably higher. Because manufacturing technologies such as welding, casting, forging, stamping and heat treatment involve changes in material structure and mechanical properties, intense testing and experimentation is indispensable. Scientific knowledge has been applied much more extensively, and these institutes have had to recruit the best staff and equip themselves with the best instruments. Second, the institutes of this group have had much looser and less regular links with productive enterprises. Unlike the product technology institutes who were rather uniformly integrated in the planning system through an array of institutional arrangements in which both R&D institutes and production enterprises were specialized along comparable product lines, and forceful administrative coordination was imposed to routinely harmonize (product) innovation planning with production planning (see Chapter 14, sections ‘A management procedure for product innovation’ and ‘Centrally affiliated institutes for manufacturing technology’). The manufacturing technology R&D institutes were not linked to manufacturing in the same way. They functioned largely outside of the ordinary operation of the planning system, coordinated as necessary under special and individually approved innovation projects. Contractual technology development The most prominent feature of the market earning structure of the manufacturing technology institutes is that ‘technology development’ contracts provide an extraordinarily high proportion of market earnings, in contrast to the group of product technology institutes whose market earnings come primarily from a combination of ‘technology transfer’ and ‘technological consultancy and technological services’, as discussed in Chapter 16. Table 17.1 Table 17.1 Market earnings structure of the entirely centrally affiliated R&D institutes compared with the sub-group of manufacturing technology institutes (RIMST)

RIMST 1991 1993 1994 All entirely centrally affiliated institutes 1991 1993 1994

Technology development (%)

Technology transfer (%)

Technological consultancy and technological services (%)

Trial production, other production and sales (%)

47 80 60

3

24

25

30 38 29

11

14

42

MACHINERY TECHNOLOGY INSTITUTES

Technology development (%)

Technology transfer (%)

Technological consultancy and technological services (%)

191

Trial production, other production and sales (%)

Source: Shi Jianzhong 1992: Table 8; Zhang Meiguang 1995. The data from the two sources may not be fully comparable.

compares income statistics for the manufacturing technology R&D institutes (i.e. RIMST institutes) with income statistics for the centrally affiliated R&D institutes as a whole.1 Not surprisingly, the income structure of this RIMST group is not homogeneous.2 Of the five institutes in the group which are pertinent to our focus, four show the characteristically high proportion of income from ‘technology development’. These are the Harbin Research Institute for Welding, the Zhengzhou Research Institute for Machinery Engineering, the Beijing Research Institute for Mechanical and Electrical Technology, and the Beijing Research Institute for Automation in the Machinery Industry. The income structure of the remaining institute, the Shenyang Research Institute for Foundry Technology, is becoming similar to the income structure of the product technology institutes, relying mainly on a combination of ‘technological consultancy and technological services’ and ‘trial production’. The reliance of these first four institutes on technology development work can be compared with two exceptional cases among the centrally affiliated product technology institutes. These two, the Jinan Research Institute for Foundry and Forging Machinery and the Xi’an Research Institute for Heavy Machinery, also obtained a high proportion of their income from ‘technology development’, according to the data on their income structure (Zhang Meiguang 1995) and a general introduction provided by the managers and policy analysts at the Ministry of the Machinery Industry (e.g. Mr Zhu Sendi). These two product technology institutes, together with the first four manufacturing technology institutes listed above, constitute a group of six institutes which have proved uniquely capable of marketing ‘technology development’. The question then is what contractual ‘technology development’ involves, and why these institutes have moved in this direction during the transformation? What specific roles have they been able to play, in a market-oriented system, by moving into ‘technology development? This chapter will concentrate on exploring these questions. It is remarkable that one of these six institutes, the Beijing Research Institute for Automation in the Machinery Industry, is recognized as the only one engaged in microelectronics-based manufacturing automation, and the remaining five are all in ‘conventional’ mechanical processing technologies such as welding, heat treatment, foundry and forging. The following two sections will examine the former institute and one of those engaged in more conventional technologies, the Beijing Research Institute for Mechanical and Electrical Technology, to illustrate similarities and differences between them in their transformation. Factors which were considered in the previous chapter, such as technological complexity, the number of users and technology imports will again be considered. This chapter will close with a short discussion of the characteristics of the transformations examined which are typical of the ‘manufacturing technology R&D institutes’ and of the implications of these characteristics. The transformation of an institute engaged in conventional manufacturing technology As noted above, most of the institutes previously assigned to the development of manufacturing technology were working with conventional manufacturing technologies. This section is based on a case study of the

192

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

Beijing Research Institute for Mechanical and Electrical Technology, of the Ministry of the Machinery Industry, which is engaged in forging and heat treatment technology. The contractual technology development project taken as an example of the Institute’s work is the development of a combined rolling and die forging production line. The case study is intended to highlight those characteristics of the transformation which may be typical of other institutes assigned to conventional manufacturing technologies.

CASE TEXT 17.1 THE BEIJING RESEARCH INSTITUTE FOR MECHANICAL AND ELECTRICAL TECHNOLOGY (BRIMET), OF THE MINISTRY OF THE MACHINERY INDUSTRY Background: the Institute and heat treatment and forging technology in the past The Beijing Research Institute for Mechanical and Electrical Technology of the Ministry of the Machinery Industry was founded in 1958, and given its present name in 1971. It has been the leading Chinese institute for forging, stamping and heat treatment technologies, and its work now covers related testing and control technologies, the computer-assisted design of dies and die materials, forging equipment, superplastic metal technology, surface engineering, vacuum heat treatment, turbine blade forging, fine blanking (organized in the Service and Development Centre for Fine Blanking Technology and funded under a UN development project) and CAD/CAM for forging systems (various introduction materials to the departments of the Institute). At present the Institute hosts the secretariats of several nationwide societies and committees (Introduction to the BRIMET): the Heat Treatment and Forging divisions of the Machinery Engineering Society of China, the National Technical Committee of the Chinese Standardization Commission for Heat Treatment, the National Technical Committee of the Chinese Standardization Commission for Forging Technology and the National Technical Committee of the Chinese Standardization Commission for Forging Die Technology. The Institute currently has a staff of 900, of whom three quarters are engineers or senior engineers and the remainder are workers. While most of the ‘product technology’ R&D institutes in the traditional Chinese machinery industry could work from experience to modify designs already in use to achieve some degree of product diversification, the heat treatment and forging technology institute could not undertake any change without performing scientific experiments, because these processes shape and deform metal under temperature and pressure, and the analysis of such processes requires intensive testing and experiments regarding the properties and structure of the metal, with appropriate measurement and control techniques. The interdependence of scientific knowledge and manufacturing techniques is evident in the work of this Institute, for instance in the superplasticity of metal (interview with Prof. Hai Jingtao; Introduction to the department of Metal Superplastic Technology),3 a field in which engineering research which aims to identify processing conditions not only entails the intensive application of scientific knowledge but also improves scientific understanding. Here basic research cannot be separated from engineering development. The Institute is now the biggest centre in this field in China. The Institute’s contribution to heat treatment and forging technology for ordinary machinery production from the 1950s to the 1970s was admitted to be slight, although the Institute’s research achievements were substantial. This is in accordance with other evidence of the backwardness of manufacturing technology in this industry. The basic reason is that heat treatment and forging technology seems never to have been well integrated in the institutional structure of the planned regime. It was explained that the development and application of forging and heat treatment technologies was planned, under the planning system, only rather incidentally. Institute achievements, which often resulted from the Institute’s own initiatives, were largely shelved because they did not meet ‘real’ demands. However the capabilities of the Institute were utilized for military and some special industrial projects. In all, ten such projects were initiated from the 1950s to the 1970s, often associated with important capital construction investments such as the 700 mm turbine rotor blade which was used for the domestically developed 300 MW electricity generation equipment, and a few projects

MACHINERY TECHNOLOGY INSTITUTES

for automobile parts manufacturing technology used in the construction of the No. 2 Automobile Factory. Under such projects, the Institute was required only to provide users with segmented and incomplete outcomes. For instance, in the automobile project, the Institute provided die designs, process specifications, and some single items of equipment which were installed and completed by the factory under the planning coordination. Services of this kind were not successful, but the users were also undemanding. The Institute’s engineers felt that users during this period did not care too much about quality and timing (interviews with Profs. Hai Jingtao, Huang Fuhua and Liu Caizheng). Trends in contractual activities during the reform Table 17.2 shows the Institute’s earning structure as it developed from 1984 to 1994. Prior to 1989, the Institute’s market income was minor, and was derived in roughly equal proportions from all items, whereas since 1989 contractual ‘technology development’ has dominated the marketing activities of the Institute, growing rapidly and consistently. Since 1992, technology development has been providing more than 90 per cent of the Institute’s total market earnings. This is the most prominent characteristic of its marketing structure, and one which clearly differentiates it from the ‘product technology’ institutes. It was said that the Institute is generally inclined to contract with users to provide complete production lines, embracing basic equipment (called ‘elementary’ equipment) that embodies the Institute’s core know-how, complementary technologies for testing, controlling and connecting, and installation and debugging services for the line until Table 17.2 Change in income structure 1984–1994: Beijing Research Institute for Mechanical and Electrical Technology (million yuan)

Source: Zhang Meiguang 1995. Note: Amounts are not corrected for inflation. From 1989 on, the total income includes ‘other income’ (see Table 15.1) which includes support, donations and credits from domestic or international societies or individuals.

it is turned over in use (interview with Prof. Hai Jingtao). By early 1995 approximately thirty production lines had been delivered or ordered, mainly for the automobile industry (Profs. Liu Caizheng and Huang Fuhua). Six of these were for automobile front axles manufacturing, two for automobile universal joints, six for drive shafts, seven for gears, and ten heat treatment lines. Moving to provide complete production lines was said to be partly a response to the low returns on the transfer of segmented technology such as blueprints.

193

194

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

A typical ‘technology development’ project: systems integration and user participation In order to understand what changes are involved in the commercial ‘technology development’ projects, and the conditions for successfully delivering the developments, an additional survey was conducted with the help of the Institute, with respect to the ‘combined rolling and die forging production line’ which is used for automobile front axles. As has been mentioned, six such lines have been sold.4 There are two ways of manufacturing automobile front axles: roll forging, as applied by the former Soviet automobile industry and in the Chinese No. 1 and No. 2 automobile factories, and die forging, a technology introduced in the 1980s and supplied by a Western German company. Both shape the workpieces under pressures which may exceed ten thousand tonnes per square metre. Roll forging is less precise and slower, while die forging is much more expensive because the equipment requires fine dies and complicated automation technology. The idea of combining the two methods derived from an analysis in the mid-1980s showing that roll forging could still be useful for forging the less precise parts of the workpieces if these are finely cast, while using additional die forging to achieve precision where it is strictly required. This could give greater processing speeds with lower costs, and the working pressure could be reduced to less than three thousand tonnes per square metre. The concept is creative and, so far as we know, unique to the process. The technology is used for truck rather than car production, with an annual capacity of producing 100,000 to 150, 000 axle sets (interview with Profs. Huang Fuhua and Liu Caizheng). Several events were considered decisive in the development of combined rolling and die forging production lines (interviews with Profs. Liu Caizheng and Huang Fuhua). In the early 1980s, imports from EUMUCO of Germany transmitted the concept, techniques and basic design for forging systems engineering. Forty-four items of ‘elementary equipment’, including forging manipulators which comprise the ‘bricks’ of the systems were imported. The Institute’s learning from the technology imports and from modifying some of the imported technology such as the forging manipulators provided a sound basis for the next phases. In 1988 the Institute decided to develop complete forging lines, and mobilized its capabilities to achieve the target. From 1988 to 1992, the idea of combining rolling and die forging was first embodied in elementary equipment and complementary units, which were then integrated into a system. Finally, in 1992, the Institute was able to provide the line as packaged engineering. It appears that, in comparison with planning approaches, two attributes characterize the commercial development of the forging systems: a high degree of systems integration which makes the innovation economically applicable and a high degree of user participation which provides the user information necessary for systems development. Systems integration Systems integration, in the area of ‘conventional’ manufacturing processes such as forging, does not necessarily entail unique single-item breakthroughs. It does demand creative combinations of items to produce optimal performance. The market mechanisms, as shown in this case, tend to encourage innovative efforts towards systems integration, in which some strengths in core know-how (the Institute’s profound knowledge of the forging process, and the idea of combining rolling and die forging), were developed and combined with the necessary complements to achieve real applicability. In this case the most notable of these complements were forging manipulators, which are specific to certain systems and were unavailable on the domestic market. Planning coordination gave little encouragement to such creative combinations of technologies. The case reveals a number of technical requirements which are critical to the development of integrated systems. First, a collection of ‘elementary equipment’ that embodies core know-how for a certain process and makes it feasible to adapt the systems to various users’ conditions; second, necessary complementary technologies, which have been particularly affected by the widespread application of computer and information technology in testing, controlling, connecting, and transmission functions; and third, the concept and technique of systems integration itself. All three aspects of the necessary technical means were acquired in the Institute, with a heavy reliance on imported technology. In other words, technology imports offered the technical means for a fundamental path-shifting in the development of manufacturing technology.

MACHINERY TECHNOLOGY INSTITUTES

It is interesting to note that the commercially-oriented development has led to a fundamental re-allocation of R&D resources in the Institute which has tended to alleviate weaknesses in some areas of the necessary technical means. Elementary equipment and selected complementary techniques now constitute a part of the Institute’s work. On the other hand there is no evidence that the commercialization has involved any appreciable reduction in the technological sophistication of the Institute, partly because of the nature of forging and heat treatment systems, in which core know-how on the processes themselves is critical. This has been one of the Institute’s advantages. The introduction of elementary equipment and complementary techniques enhanced the interaction between the core know-how and the technical means that translate the know-how into commercially competitive systems. User participation User participation is indispensable for system integration, since optimal performance requires that the systems be highly customized to individual users. Thus the development of the forging lines described above and all the other systems mentioned was only possible under the impetus of a real demand, which meant that the workpiece to be processed and the specific user conditions were specified. These in turn determined the details of the system design, and the functional parameters against which the system was to be tested. User input has continued to be important after the first line had been successful installed, because each line needed to be adapted for the particular user in some respects, such as the level of automation, the various features of the forging dies, the matching of the elementary processes and the systems layout. All of these differ in each case. The technical structure of such systems, consisting of a number of items of elementary equipment, has made them open to customization, in the sense of being able to be adapted to specific user’s requirements at an affordable engineering cost. The development of such systems was carried out under longer term contracts to ensure close and continuing communications with the users. This type of contract is labelled in the official statistics as (contractual) ‘technology development’. Thus market mechanisms provided the channels (through enduring contractual relations) for effective communications between developer and user which have proved indispensable for the development of more economically efficient manufacturing systems. The planning mechanisms did not provide such communication channels. Chinese automobile producers have now begun to accept domestic suppliers for such technology, but the market appears to be constrained. It was said that 70 per cent of that market, which is expected to expand to six million cars and trucks per year by the first decade of the next century, is likely to be bound to utilize supplies from foreign investors. The opportunity for domestic learning seems to have been largely contracted out. Changes in internal structure and the question of further development The Institute’s R&D departments are still organized by professional field. An internal contractual responsibility system gives R&D departments the authority to sign any contracts that the department can handle with its own resources. They are rewarded by bonuses in accordance with their earnings but the allocation of the contractual income is determined by the centre, in line with the Institute’s strategy. Two changes have however been evolved within the traditional departmental layout. The first is that almost all of the departments have diversified their activities towards some complete sets of engineering or equipment (interview with Prof. Hai Jingtao). The Institute’s workshop serves as an internal common-use manufacturing base. The important equipment for the forging systems such as manipulators and dies are manufactured there. The second change is the creation in 1988 of a central coordination mechanism for the development of the highly integrated manufacturing systems (Profs. Liu Caizheng, Huang Fuhua). One Chief Engineer and eight Deputy Chief Engineers constitute the team of the coordinating force, taking full responsibility for the contracts. The participation of departments in systems development is organized by the centre, under instructions which specify the task, timing, budget, and quality. Ninety per cent of the contractual fees received for this type of development are placed at the direct disposal of the centre. The Institute management has not considered an internal restructuring scheme to be necessary to cope with their rapid expansion in contractual ‘technology development’.

195

196

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

The main concern of the Institute is to maintain a certain strength in academic research (interview with Prof. Hai Jingtao). If possible, they would prefer to be a research centre like the Japanese machinery technology institute led by MITI (Ministry of International Trade and Industry of Japan) rather than a commercial supplier of manufacturing systems like the German EUMUCO. This inclination partly explains the Institute’s rather passive responses to market opportunities and threats, a strategy described by Prof. Hai Jingtao as, ‘Maintaining the research, while seeking (contractual) engineering projects with the tactic of “shou zhu dai tu” (standing by a stump, waiting for rabbits to dash themselves against it).’ This strategic choice raised several questions, which were put to our interviewees. First, could reform policy provide the possibility for some institutes in the industry to remain as non-commercial academic centres specializing in fundamental engineering research? One response was rather sceptical (interview with Mr Zhu Sendi), reasoning that none of the R&D institutes in the industry has the ability to do so because of the past. Besides, there is no evidence that the current policy will be revised. Another question then follows: what institutions can fill the gap in the fundamental engineering research necessary to support application engineering? The answers pointed to engineering universities as candidates for this function, but at present most of the engineering universities in China do not or can not do this as well. It is still not clear how the reform policy will coordinate such a crucial issue. The second question regarding the strategy of the Institute was about the future of its rather impressive application engineering. Is it possible to maintain the desired technological level in a field such as forging and heat treatment without the vital feedback from users, feedback that can only be acquired by engaging in the commercial supply of the manufacturing systems? A long discussion clarified two things which the Institute currently considers most important: first, that maintaining a certain scientific foundation has been critical for the Institute’s success in the development of some manufacturing systems, and second, that maintaining a certain level of inputs for the necessary fundamental studies has been the most difficult task during the reform. One reason stressed by the Institute is that it faces a huge burden of responsibilities to provide social services such as staff housing. These services absorb a large part of the Institute’s profits, although contractual income has increased considerably. The Institute did not deny the value to it of commercialized engineering services, but neither was it going to set an active goal of responding to the market. The question remains: is the current momentum of the Institute sustainable if taking a strategy of just waiting for chance? Sources: Interviews with Prof. Hai Jingtao, September and October 1994, and with Prof. Huang Fuhua, Prof. Liu Caizheng, Prof. Lu Chuluan and Mr Wang Dechen, April 1995; Introduction to the BRIMET, provided by the Institute; Introduction to the Service and Development Centre of Fine Blanking Technology, provided by the Institute; Introduction to the Department of Metal Superplastic Technology, provided by the Institute; Introduction to the Department of Forging Manipulators, provided by the Institute; Introduction to the Department of Vacuum Heat Treatment Technology, provided by the Institute; Product specification for vacuum furnaces, provided by the Institute.

Characteristics of the transformation towards becoming a commercial supplier of manufacturing systems The case study of the BRIMET reveals an income structure (Table 17.2) which is typical for a small group of ‘manufacturing technology’ R&D institute (Table 17.1), for which a very high proportion of market earnings since about the 1990s have come from contractual ‘technology development’. Much of this development is directed at the supply of complete production lines, delivered as turn-key projects. Such activities are not ‘experimental development’ according to the Frascati definition but rather the commercial development of integrated manufacturing systems. Thus the Institute’s transformation is leading it, somewhat reluctantly, towards becoming a supplier of manufacturing systems, something that was neither required nor experienced prior to the market reform.

MACHINERY TECHNOLOGY INSTITUTES

197

This study shows the importance, for a ‘conventional’ manufacturing process such as forging, of translating core technologies into integrated manufacturing systems, and of ongoing user participation under long-term contracts. Both factors are crucial to the step from engineering capability to economic value, and both are better supported by a market-oriented system than by planned coordination. Initial success in the commercial supply of manufacturing systems was based on a combination of accumulated know-how and the selective development of ‘elementary equipment’ and complementary units based on the assimilation of imported technology. The Institute studied displayed some creativity in the combination, which is no doubt another factor in the transformation. It was commented (Mr Zhu Sendi) that the only ‘manufacturing technology’ institute which had not entered the market for commercial manufacturing systems (the Shenyang Research Institute for Foundry Technology) had not done so in part because of its limitations in systems and key equipment design and in accumulated know-how regarding the manufacturing processes. The same theme emerges from the explanation of the achievement of a few ‘product technology’ institutes which have acquired similar capabilities in the commercial supply of manufacturing systems. The comment (Mr Zhu Sendi) attributed their capabilities to the institutes’ relative strengths in process know-how, incorporated with their expertise in equipment design. In short, where other factors such as access to imported technology are the same, accumulated capabilities in the design and testing of the manufacturing processes have proved to be critical for moving effectively into the supply of manufacturing systems. Internal organization This case provides additional evidence for a centralized mode of internal work organization appearing due to the transformation. No radical restructuring had taken place or was expected in the BRIMET, but vertical coordination of the work of various departments had been established since about 1988 to take responsibility for the development of integrated systems. In parallel, the existing departments had expanded into some complementary technological fields. The rather smooth transformation of the internal work organization was said to be largely because the Institute’s work concentrated on a relatively homogenous technological field, i.e., on forging and heat treatment technology (Mr Zhu Sendi), while only centralized coordination could handle the complicated work of developing forging systems utilizing the various technical specializations of the departments. Moreover, unlike the automation field (see the next section), the core know-how on ‘conventional’ processes like forging which the Institute has accumulated remains critical. Incorporating a core know-how which has already been accumulated with market-oriented development would seem to involve less serious reorganization of internal work. The transformation of an institute engaged in micro-electronics-based automation technology This section examines the transformation of the Beijing Research Institute for Automation in the Machinery Industry (BRIAMI). This institute is engaged in microelectronics-based automation technology for manufacturing systems. It is the only one of the centrally affiliated institutes which were previously assigned to ‘manufacturing technology’ which does not concentrate on ‘conventional’ manufacturing processes but on process control and systems control. It is examined here to identify similarities and differences between such an institute and the other institutes of the central manufacturing technology group.

198

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

CASE TEXT 17.2 THE BEIJING RESEARCH INSTITUTE FOR AUTOMATION IN THE MACHINERY INDUSTRY (BRIAMI), OF THE MINISTRY OF MACHINERY INDUSTRY Background: The Institute and manufacturing automation under planning coordination Research work started in an institute department in the 1950s, but the Beijing Research Institute for Automation in the Machinery Industry was only formally set up as an R&D institute in 1978. It is the central institute engaged in manufacturing automation for the machinery industry. Its establishment was prompted by an awareness of the importance of manufacturing automation, which was formally recorded in the 1978–1985 National Science and Technology Development Programme (Part I Chapter 2). At the time, however, there were still fiery debates about whether there was a real need for manufacturing automation in a country like China, with its huge labour surplus. The technological fields of the Institute cover: 1) computer application software for manufacturing automation, such as CAD/CAM, computer-assisted engineering, computer aided production management systems, decision support systems and office automation; 2) automatic control and application engineering for mechanical processes, i.e., systems and integration engineering, networking and communication, manufacturing process control and management, material flow and warehouse automation and automatic inspection; 3) industrial robots and artificial intelligence; 4) design and mask-making for application-specific integrated circuits (ASICs) for manufacturing automation; 5) hydraulic and pneumatic control technology and 6) some electro-physical apparatus and instruments (Introduction to the BRIAMI). Two workshops, one mechanical and one electronic, give the Institute some capacity in manufacturing, notably of industrial robots and programmable logic controllers. The Institute now hosts the secretariats of a number of nationwide societies and committees: the Automation Technology Division of the Machinery Engineering Society of China and National Technical Committees of the Chinese Standardization Commission for Industrial Automation Systems and of the Chinese Standardization Commission for Hydraulic and Pneumatic Technology. The Institute is also the Chinese technical representative for ISO/TC 184 and ISO/TC 131. The Institute has a staff of 1,300, of whom 800 are R&D scientists and engineers (Introduction to the BRIAMI). Manufacturing automation was one of the priorities of the planned projects known as ‘Key S&T Projects of the Five Year Plan’ during the 1980s. These projects were intended to enable the Chinese machinery industry to catch up in this area through absorbing and adapting imported technology in particular manufacturing processes. Such projects had a heavy impact on the initial development of the Institute in that period. The commitment to the Key Projects was reflected in a noticeably higher level of government funding throughout the 1980s (and even the early 1990s) (Table 17.3, below), for which the only parallel among the institutes examined in this study is the Shanghai Power Equipment Research Institute which had a similar heavy involvement in projects to improve domestically developed 300 MW power plant equipment during the 1980s. The Institute contributed to the key manufacturing automation projects in three fields (interview with Mr Li Baihuang). The first was the automation of some single machines. The Institute was assigned to develop programmable numerical control (NC) devices for bearing grinding machines produced in the Wuxi Machine Tool Works, a machine tool producer specializing in these machines. Another single machine was the industrial spraying robots which the Institute was commissioned to develop for the robot spraying production line in the No. 2 Automobile Factory. The second field to which the Institute contributed was flexible manufacturing systems (FMS), including a sheetmetal FMS for the Great Wall Switching Gear Works in Tianshui and an automated high bay warehouse system for the No. 2 Automobile Factory. The third field was plant automation. The Institute was commissioned to develop management information systems and CAD/CAM techniques for machinery producers. Later on it joined the national Computer Integrated Manufacturing Systems (CIMS) project, responsible for the application engineering of the system in the Jinan No. 1 Machine Tool Works. For these fields of development, core technologies were imported. The Institute focused substantially on

MACHINERY TECHNOLOGY INSTITUTES

application engineering, including the development and adaptation of management software, the networking of systems, and the development of complementary devices for purposes such as testing and material transmission (interview with Mr Li Baihuang). One feature of the planning coordination of these projects, according to the Institute (Mr Li Baihuang), was that it led to fragmentation. Each participating institute was involved in only some parts of a system, and the integration of these parts in a form ready for installation in a particular manufacturing enterprise was assumed to be the task of the planning administration. The segmented tasks usually stopped at the prototype stage. Such a system could not result in well-codified technical know-how or commercially-proved competitive output. Consequently the technology developed, after having being applied once in the specified plant, could generally not be applied elsewhere or further improved.5 Nevertheless the Institute’s participation in the Key S&T Projects in the 1980s left it with an accumulation of learning about commercially competitive manufacturing automation technology, based on intensive imports. The Institute was familiar with application engineering using internationally compatible components, devices, and design norms in some particular areas such as robot spraying systems and automated high bay warehousing and sheet-metal FMS, and was capable of some generic techniques such as programmable logic controllers and CAD/CAM (Mr Li Baihuang). This has given the Institute the foundations for their commercial development of selected manufacturing automation systems in the late 1980s and the 1990s. Trends in contractual activities during the reform As shown in Table 17.3, the market income of the Institute has since 1989 been significantly greater than its government funds. Like the forging and heat treatment technology institute examined in the previous section, the Beijing Automation Institute’s market earnings have been Table 17.3 Change in income structure 1984–1994: Beijing Research Institute for Automation in the Machinery Industry (million yuan)

Source: Zhang Meiguang 1995. Note: Amounts are not corrected for inflation. From 1989 on, the total income includes ‘other income’ (see Table 15.1) which includes support, donations and credits from domestic or international societies or individuals.

dominated by contracts for ‘technology development’ in every year from 1989 to 1994. In most of these years 80 to 90 per cent of its market earnings came from this item, which increased particularly dramatically between 1990 and 1991. Its ‘technology development’ contracts are for the

199

200

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

development of automation systems. During these years more industrial manufacturers showed an interest in automation technology, and more users began to require suppliers to provide them with integrated systems. Some user enterprises, such as the Jinan No. 1 Machine Tool Works in the CIMS project, themselves took over the direction of systems integration after the government had withdrawn its coordination, but this was more an exception than the rule, according to the Institute Director (Mr Li Baihuang). It was explained that the trend to demand for integrated systems was partly because automation systems have become more complicated. We will turn to analyse some details of the contractual development of automation systems soon, based on selected typical projects as in the previous case study. Another feature of the income structure is that the Beijing Automation Institute, in contrast to the forging and heat treatment technology institute, earns a significant part of its income from ‘technological consultancy and technological services’, with an odd fluctuation in 1990 when the Institute earned nothing from this item, and a tripling from 1993 to 1994 when it reached a level almost comparable to the income from ‘technology development’. This income relates to the Institute’s consultancy work for imported automation lines. This includes debugging installed systems and providing advice and training for the importing enterprises. In some cases the Institute has been contracted directly by the foreign suppliers to provide support services (Mr Li Baihuang). The extraordinary variations in the ‘technological consultancy and technological services’ may therefore be related largely to variations in import levels year by year. Typical technology development projects: application engineering As in the previous case, a detailed study was made with the help of the Institute to ascertain some basic features of the commercial development of manufacturing automation technology. Three commercialized automation projects were selected: automated high bay warehouses, automatic robot spraying lines, and operations automation engineering. The first of these, the Automated High Bay Warehouse, was established as a planned project in the first half of the 1980s and was conducted by a group of experts originally working on industrial computers. After the first warehouse was installed in the No. 2 Automobile Factory the technology was provided to other users on a contract basis, while being continuously renewed using newly available components and computers. The testing unit and mechanical parts have been further developed, and the communication function of the warehouse was expanded in 1992 to include networking with the plant system. The Institute now has about one third of the higher end of the domestic market, i.e., of those warehouses in which networking with the workshop and plant is practicable. By 1995 it could supply four to five warehouses per year with a total contract value of about ten million yuan. Demand has exceeded the Institute’s capacity, and potential orders worth about the same amount have had to be rejected. In all, thirty warehouses have been finished or contracted. Competitors are mainly domestic suppliers, although a few Japanese companies are coming in. The users are from sectors such as the automobile, textile, airport, and storage industries (interview with Prof. Li Fanhai). The second project, the Automatic Robot Spraying Line, was financed as a planned project in the second half of the 1980s and conducted by an expert group previously working on the generic engineering of industrial robots. The project outcomes were soon pushed to commercial production by the Institute. There had been a number of other planned robot projects under the Key S&T Projects of the 1980s, including welding, assembling, stamping and loading robots, but the Institute’s Robot Spraying Line was said to be the only one which is commercially procurable on the technology market. The robots used in the spraying lines are produced in the Institute’s own workshop. The commercial success of this line is considered to be due to the fact that the technology is somewhat simpler, to easy access to and reliance on internationally available control computers, the careful development of complementary equipment, some novelty in systems design which resulted in an identifiable brand profile in the domestic market, and active marketing. The Institute now has almost all of the domestic market, with total contracts reaching forty million yuan by 1994. The automobile industry is the major user (interview with Prof. Gao Shiyi).

MACHINERY TECHNOLOGY INSTITUTES

The third project, the Operational Automation Engineering, was conducted by experts who were previously working on the development of PLCs (programmable logic controllers). The development was once supported as a planned project, with the intention of producing PLCs domestically. The failure to achieve that resulted in a cancellation of the plan. The failure was ascribed to the rapid technological change in this area and the small size of the domestic market (interview with Mr Zhu Sendi), although there are diverse users for operational automation. Since the 1990s, these experts have turned to working on application software for various kinds of operations automation, with a few turning on developing application-specific integrated circuits (ASIC). The income earned from this area is not significant, and the team faces competition from both NTEs and small Hong Kong companies (interview with Mrs Qiao Meirong). Application engineering The three examples demonstrate that automation systems development is differentiated from manufacturing process systems particularly by its emphasis on ‘application engineering’. Both kinds of development are recorded under the same contract item of ‘technology development’ and have similar objectives in the area of systems development. But ‘application engineering’ represents a downstream move of technical efforts, from singular and generic (i.e., general purpose) areas such as industrial computers, industrial robots and general PLCs, towards specific purpose and integrated systems such as automated warehouses, spraying robots and application-specific integrated circuits. The effect of international trade has been more significant in the development of automation systems, which has been facilitated by the fact that a growing number of basic components, control devices, and design norms used for automation systems are available on the international commodity and technology market. Therefore endogenous efforts can and must be focused on application engineering rather than general purpose techniques. This has entailed a radical re-allocation of institute resources. With generic technologies being increasingly internationally tradable, what advantages do local developers have in the application engineering of automation systems? Local advantages may derive from some appropriate feature of a system, from careful system design to incorporate basic components and norms in line with specific needs so that a local brand is developed, as in the case of the robot spraying lines. The other source of opportunities for creative local developers derives from the non-generic mechanical apparatus and complementary units which are needed for the systems. For instance stackers, store bays and synchronizing devices are rather new and are unique to particular automation systems. The ability to develop such complements in a form well suited to a system and better adapted to local operating conditions constituted an important part of the case Institute’s competitiveness. The user’s participation is evidently more intrinsic to application engineering than to the development of generic technology for automation systems. Turning to application engineering in itself implies turning to closer interactions with user demands. As in the case of forging systems, user specificities have been an indispensable input in each individual system which the Institute has contracted to develop. For example, about 40 per cent of each new contract for the high bay warehouse is devoted to user-specific design work (interview with Prof. Li Fanhai). Being close and familiar to local users gives the local developer some comparative advantage over the foreign supplier assuming a certain mastery of the technology. The productivity of engineering development and the accumulation of engineering knowledge With a sustained increase in demand for automation systems, the efficiency of application engineering began to be perceived as a bottleneck (along with other difficulties resulting particularly from a shortage of investment funds which will not be considered here). It appeared that a point had been reached at which a major change was required in the way in which engineering development was carried out for the automation systems. This was described as the need to convert engineering development from a ‘craft’ (xiao zuo fang) approach to an ‘industrial’ (gong cheng hua) approach (interviews with Profs. Li Baihuang, Li Fanhai, Gao Shiyi, Mrs Qiao Meirong). The conceptualization of the two approaches described here is based on the Institute engineers’ own understandings expressed in interviews, and from mentions of these terms in a few other cases (interviews at the Dalian Research Institute for Modular Machine Tools; interview with Mr Zhu Sendi).

201

202

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

The craft approach was understood as doing engineering development case-by-case, with each case entirely conducted by one or a few designers who are expected to be aware of all aspects of the assignment. This is the Institute’s present practice. Its drawbacks include the limited division of labour and the lack of any incentive to codify the tacit knowledge acquired from laboratory experiments, trial production, engineering design and user feedback. This has been a serious impediment to the collective learning which the Institute needs to improve its engineering capability. The inefficiency of this approach has manifested itself in the capacity limit of just five warehouses per year, a serious threat to the Institute’s competitiveness even with very cheap skilled engineering labour. The Institute would like to explore the ‘industrial’ approach, with development work organized by specialization rather than around individual engineers, and active steps to structure the engineering knowledge gained so that it can be transmitted among team members more easily rather than being entirely tacit and thus tied to the individual ‘craftsmen’. Along with the greater internal division of labour, this should create a learning dynamic for the more rapid accumulation of engineering knowledge and the capacity needed for commercial development of automation systems. This challenge, which was perceived as a need to achieve much greater efficiency in the development of automation systems, also points beyond that to the split character of the ‘knowledge institution’ for industrial technology development as it evolved in China. In the Institute’s own history as an R&D agent, the pursuit of novelty in laboratory work was highly prized and individual search was the main working style, while knowhow for application systems was considered inferior. On the other hand, neither the Institute itself nor the Chinese machinery industry as a whole had significant experience with the development of engineering knowledge based on dynamic technological change. The lack of capability in engineering knowledge creation would seem to be a serious challenge the Chinese industry must meet if it is to promote its productivity in engineering innovation. Changes in internal structure and strategy The Institute departments are still organized by professional fields, with six specializations: computer application software for manufacturing automation, automatic control and application engineering for mechanical processes, industrial robots and artificial intelligence, design and mask-making for applicationspecific integrated circuits (ASICs) for manufacturing automation, hydraulic and pneumatic control technology, and electro-physical apparatus and instruments. As in many other cases, the technological activity in each department has expanded to meet the demand for more complete outputs. Several commercially successful projects have grown from the traditional departmental structure. An internal responsibility system has been adopted to delegate significant autonomy in decision-making and the allocation of contractual income to these departments (interview with Mr Li Baihuang). Chief engineers of the Institute take some responsibility for the cross-department coordination of very big projects, but there was no indication of the sort of matrix structure we saw in the Beijing Research Institute for Mechanical and Electrical Technology, with the vertical dimension dominating both the implementation of big contracts and the allocation of revenues from them. This is partly because most of the contracts for automation systems can be managed within one department. In its plans for the future, the Institute is explicitly modelling itself on a high-tech engineering company with the objective of commercial development of automation systems for manufacturing. Several business departments are to be organized, based on proved or potential market niches such as industrial robots and high bay warehousing. In the future, joint ventures are likely to be organized by these business departments where it is possible. The Institute headquarters would then manage strategic elements of planning, marketing and technological development, along with very large projects. The proposed National Manufacturing Automation Engineering Centre also seems likely to be at least partly a responsibility of the headquarters. This centre, funded by the State Planning Commission, will bring the Institute much needed investment and is expected to provide a better basis for the Institute to develop engineering knowledge—perceived as a weak area—and to give it a strength in some application-purpose oriented components and techniques, such as application-specific integrated circuits.

MACHINERY TECHNOLOGY INSTITUTES

203

For strategic reasons, the Institute will seek to combine work in the development of application engineering and in providing engineering services. Application engineering is already provided competitively, and engineering services which are required mainly by imported systems are expected to provide an avenue for learning about more complex systems and developing competitive strengths for the next generation. Sources: Interviews with Mr Li Baihuang, October 1994, Prof. Li Fanhai, Prof. Gao Shiyi, Mrs Qiao Meirong, Beijing, April 1995; Introduction to the BRIAMI, provided by the Institute.

Characteristics of transformation centring on the application engineering of automation systems The case of the Beijing Research Institute for Automation in the Machinery Industry shows a similar direction of transformation as the previous case, leading the Institute towards becoming a commercial supplier of integrated automation systems such as automated high bay warehouses and automatic robot spraying lines. As in the previous case, market earnings are dominated by ‘technology development’. There was an initial demand for manufacturing automation from users in the automobile and various other industries seeking better quality and space savings or wishing to substitute automation for labour in heavy and polluted jobs. This drive from the demand side is largely induced by the market-oriented reform, and comes despite a general labour surplus in China. Another similarity with the previous case is that the demand for highly integrated automation systems in the technology market has involved closer user participation under an enduring contract, in contrast to planning coordination under which the fragmentation of technological efforts among various participants was the rule. However, the development of automation systems differentiates itself by a marked emphasis on application engineering, a significant downstream move in technological activities. This has resulted from the growing tradability of generic technologies for automation systems, such as basic components, control computers and basic norms for systems design. This trend, a result of expanding international trade, has been more significant in electronics-based automation technology. Adopting the generic technologies dynamically as they are developed, is one factor responsible for the success in automation systems. It allows technological capabilities to be devoted to building up, and building on, specific competitive advantages. The re-allocation of institute resources away from generic areas has therefore been more apparent in the area of automation systems. The need to keep closer pace with the development of generic technology may partly explain another difference between the automation institute and the previous case. The much larger part of their marketing activities of the automation institute for the ‘technological consultancy and technological services’ involves demand factors coming from increasing imports of automation systems. It is also a strategic move taken by the case institute aimed to learn about the latest advances. Faced with the growing tradability of generic technology, local developers are finding opportunities to utilize their creativity in systems design and some specific mechanical apparatus and complementary units needed for the automation systems such as stackers, storage bays, electro-hydraulic drives and synchronizing devices. The ability to develop such apparatus and devices to suit a system and the local conditions, and the ability in combining generic automation technologies into a system with unique features are other factors behind the significant level of competitiveness achieved in this case. They have stood the case institute in good stead in developing its own brand in some business fields. Being close and familiar to local users gives a local developer an advantage over foreign suppliers, assuming a certain degree of mastery of the technology, especially in application engineering which requires closer interactions with users’ demands. In automation systems development, as in the development of forging systems, user specific information has been an indispensable input.

204

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

Internal organization This case study provides an alternative perspective to the internal organization of a machinery technology institute during its transformation. While similar to the previous case that, until very recently, internal work in the automation institute was organized under the internal responsibility system in which department structure remained basically the same as previously assigned but with much greater autonomy, the automation institute had a much less decisive institute centre whose function was restricted in coordinating the development of the most complex systems. This was because most of the contractual developments could be managed within a department. A plan for further internal restructuring was put forward in 1995, on the basis of proved or potential market niches. It would result in a multi-department high-tech engineering company, with business departments assigned fairly wide autonomy and headquarters managing only strategic items. The more decentralized internal organization of the automation institute will thereby be consolidated in an envisaged coming period. This case raises an interesting question regarding the efficiency of engineering development, highlighting the need to shift from a ‘craft’ approach to the ‘industrial’ approach. The craft approach results in low productivity and does not provide a mechanism for the accumulation of engineering knowledge beyond individuals’ tacit knowledge and experience from individual development projects. The ‘craft’ approach to engineering development is virtually a reflection of the missing capability of the Chinese machinery technology innovation system in the generation and transmission of engineering knowledge. The change in this capability, may prove to be as crucial as any other factors which have been discussed, as far as the successful transformation of the institutes and the industry is considered. Summary: comparing the transformations of manufacturing technology and product technology institutes This chapter has provided empirical evidence of the transformation of a few R&D institutes which have an important place among the centrally affiliated institutes labelled as the group I institutes in Table 14.12. These eight institutes with four thousand scientific and engineering staff were previously officially assigned to the development of ‘manufacturing technology’, in contrast to the centrally affiliated institutes previously assigned to develop ‘product technology’, discussed in the previous chapter. While their combined size is barely significant in quantitative terms, the transformation of these institutes has taken a distinctive path, as can be seen from the extraordinary importance of contractual ‘technology development’ in their market income. Two case studies have been used to obtain a more detailed understanding, covering the Beijing Research Institute for Mechanical and Electrical Technology which deals with the rather ‘conventional’ manufacturing process of forging and heat treatment, and the Beijing Research Institute for Automation in the Machinery Industry which is engaged in microelectronics-based automation technology for manufacturing. Together, they provide a fair picture of the transformation of ‘manufacturing technology’ institutes as a whole. ‘Technology development’ work and the supply of integrated manufacturing systems It has been shown that where contractual ‘technology development’ dominates market earnings the transformation of the institutes has been leading them towards becoming commercial suppliers of integrated manufacturing systems. Contracts for ‘technology development’ are in fact for the supply of an integrated manufacturing system to be tailored to the needs of a particular user, rather than for what would be termed ‘experimental development’ according to the Frascati Manual. A total of seven to ten institutes have moved

MACHINERY TECHNOLOGY INSTITUTES

205

in or might turn to this direction, including most of the eight ‘manufacturing technology’ institutes and a few ‘product technology’ institutes engaged in the fields such as heavy machinery and forging equipment. The work of this handful of institutes which are focusing on the commercial development of manufacturing systems is characterized by a high degree of systems integration and user participation. There are strong reasons to think that these characteristics are generally relevant to the commercial development of manufacturing systems. Integrated systems are only cost-efficient if it is possible to embody a high degree of flexibility so that the systems can be set up to manufacture a particular workpiece and can be tailored to a particular user’s operating context under affordable engineering labour. Such systems must therefore incorporate certain technological features which, from our empirical observations, can be categorized including those as items of ‘elementary equipment’ which embody the core know-how of the manufacturing processes and can be combined in different systems, and complementary technologies for functions such as testing, measuring, transmission, connecting, controlling, and information processing and distribution. In addition to the elementary equipment and complementary technologies, appropriate concepts and skills for systems integration are required to unite the components into a system which can effectively exploit the core engineering know-how. It is remarkable that all these technical elements for integrated systems development have been acquired by close examination of, and contact with, imported technology. Where manufacturing technology is provided in the form of integrated manufacturing systems, the supplier must be sensitive to the factors on the user’s side which determine the manufacturing efficiency and quality of the installed systems, while the users must also be seeking efficient manufacturing systems. This perhaps explains why there were neither suppliers nor demanding users of integrated manufacturing systems before the market reform. User’s participation is required not only in the early stage of systems development, to select which systems are to be developed, but also in the stages up to and after the first commercial realization, to communicate user specificities for the design and construction of each individual system. The technological aspects and the user’s participation are intimately inter-related and reinforcing. These characteristics have somewhat special implications in automation systems, as shown in the second case in this chapter. The emphasis on systems integration has involved a rather radical re-allocation of institute innovative activities from generic technology to ‘application engineering’, basically because a growing proportion of the generic technology needed for automation systems can be purchased internationally, but application engineering to shape a system in accordance with the user’s specificity remains essential. Manufacturing technology institutes: cutting versus forming A metal work piece may be machined through various processes, such as turning, milling, grinding, boring, welding, forging, casting and foundry, pressing and melting (e.g. powder metallurgy). These processes can be roughly divided into two groups: those which remove spare material through cutting, in which machine tools are the cutting instruments, and those which shape the metal into the desired form, for instance with foundry and forging machinery. What we have discussed here under the heading of ‘manufacturing technology’ is in fact only the latter, and the discussion would not be complete without some mention of the machine tool sector. While institutes assigned to metal-forming technologies are typically being transformed into suppliers of integrated manufacturing systems, the institutes for machine tool technology, examined in the previous chapter, have been lagging behind in acquiring the mastery they would need to develop integrated manufacturing systems. As their income structure shows, these institutes supply relatively smaller and

206

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

simpler machining systems and auxiliary machinery, delivered as finished products with diverse kinds of engineering services (technology development, technology transfer and technological consultancy) as a secondary source of income. This is virtually the opposite to the picture regarding the metal-forming technology group. There are two explanations for the difference, one institutional and one technological. Institutionally, the machine tools technology institutes were classified as ‘product’ rather than ‘manufacturing’ technology institutes, so that they were much more firmly imbedded in the main framework of planning coordination. Their function in that system aimed at product diversification on the basis of rigid standardization, largely through imitative designing (i.e., ‘analogous designs’), and was limited to one product category per institute. As a result these institutes were less capable of fundamental engineering testing and experimentation, which proved to be important for the development of integrated manufacturing systems. The second explanation is that recent technological advances seem to have had a much more profound impact on the machine tool sector, where the introduction of computer-based control technology has enabled metal-cutting machines to operate to the limits allowed by their conventional structure. Therefore experience-based approaches to machine tool design have largely broken down. The two factors, reinforcing one another, have prevented the institutes in the machine tools sector from becoming commercial suppliers of integrated manufacturing systems. Many other reasons, such as the market structure for conventional product engineering and the demand for auxiliary machinery which was neglected under the planning system, have combined to support the transformation of the metal-cutting technology institutes towards being suppliers of more conventional engineering and smaller but important auxiliary machinery. Engineering services in a market context: product engineering and manufacturing systems engineering Chapter 15 has shown that the centrally affiliated machinery technology R&D institutes as a group have high earnings from the items ‘technology development’, ‘technology transfer’ and ‘technological consultancy and technological services’, which together are defined in our study as ‘engineering services’, whereas the remaining industrial technology R&D institutes have higher earnings from trial production and other production and sales. The cases studied in Chapter 16 showed that where there are many users and where technological changes are not too radical, most of the centrally affiliated product technology institutes have evolved to become disseminators of relatively advanced product technology for domestic manufacturers. Their capability in dissemination is based on the assimilation and adaptation of imported technology, often incorporated with their own experience and creativeness. With some exceptions, the dissemination is carried out under contractual forms categorized as ‘technology transfer’ and ‘technological consultancy and technological services’. In contrast, Chapter 17 has shown that the market earnings of the centrally affiliated machinery technology R&D institutes which specialize in ‘manufacturing technology’, together with a few ‘product technology’ institutes, are dominated by ‘technology development’. They are becoming commercial suppliers of relatively complicated manufacturing systems. Thus whilst the centrally affiliated machinery technology R&D institutes as a whole focus on engineering services, they can be further divided into two groups supplying different types of engineering services: those institutes previously assigned to the development of product technology tending to move towards the commercial supply of product engineering, while those previously assigned to the development of manufacturing technology tend to move towards the commercial supply of integrated manufacturing systems.

MACHINERY TECHNOLOGY INSTITUTES

207

The institutes of the machine tools sector are better seen as a third group, with a focus not on engineering services but on finished machining equipment and auxiliary machinery, complemented by product engineering. Their future is not yet clear but could be of importance to the necessary reorganization of the machine tools sector. The two kinds of engineering services, for product engineering and manufacturing systems engineering, both require relatively close interaction with the users, which can best be served by organizing them under on-going contracts for longer or shorter terms. The outcomes of both are now provided in highly completed forms, in contrast to the pre-reform approach. The major difference between the product engineering and manufacturing systems engineering is in the degree of systems complexity. We have seen, for instance, that integrated systems for manufacturing technology require not only elementary equipment but also complementary technologies and the skills and concepts for systems integration. The commercial development of manufacturing systems is therefore largely a matter of systematically matching a set of techniques to the user’s conditions to achieve optimal operating performance. This is a new approach to technological innovation for manufacturing processes which the planning system could not support. There is no clear-cut boundary between the two kinds of engineering services. From the supply side, institutes with a better mastery of the scientific foundations of the technology with which they deal, combined with design abilities, were more able to enter into the development of relatively complicated integrated systems. On the demand side, as numerous cases have shown, where the market for an institute’s technology is ‘deeper’—with users of greater technological sophistication—there is a tendency to tailor the output more closely to user conditions. The engineering required then includes elements which are newer, and are developed under contracts which last longer, and are categorized as ‘technological consultancy and technological services’ or even ‘technology development’. There is also no clear-cut choice of contractual form by individual institutes. Increasing evidence points to plural approaches pursued by a particular institute to capitalize on its technological assets, in response to the diverse opportunities in the market. A complementary combination between various types of contract is becoming more common as institutes gain more experience in strategically organizing their internal resources. This provides a foundation for further dynamic transformation. Infant sectors and policy implications In the course of the market-induced transformation of the centrally affiliated machinery technology R&D institutes, which has led them to focus primarily on selected engineering services, a number of infant sectors have begun to emerge. An infant sector is a particular field of business, with a clear domain, which is relatively new and strategically important to the upgrading of the industrial structure and industrial technological capabilities. Examples from the case studies include forging systems (Beijing Research Institute for Mechanical and Electrical Technology), spraying robots and high bay warehouses (Beijing Research Institute for Automation in the Machinery Industry), internal combustion engine engineering (Shanghai Internal Combustion Engine Research Institute), cable products and plant engineering (Shanghai Electric Cable Research Institute). Other areas also seem to indicate some promise, such as the auxiliary machinery used for flexible manufacturing systems which certain institutes in the machine tools sector are attempting to develop. The importance of such infant industries lies in the fact that most of the industrial enterprises in the Chinese machinery industry had long been rigidly assigned to pure production. It is difficult for them to learn to be adept innovators in the short term, restricted as they are by a lack of innovative experience and resources, and struggling with an inappropriate institutional heritage, particularly in enterprise structures

208

‘MANUFACTURING TECHNOLOGY’ INSTITUTES

and internal work organization. Those case studies which have included the industrial enterprise to which an institute was affiliated bear out these difficulties. In such circumstances, innovative business fields developed by R&D institutes and tested in the marketplace are especially valuable. Together with some strengths which have been developed within industrial enterprises, institute-based infant industries may well be one of the major driving forces to lift the industry to a higher level of competitiveness. This has two implications. First, reform policy must be responsive to trends in important infant sectors as these emerge from the transformation. Second, the healthy growth of newly emerging engineering fields needs great internal effort to achieve a rather profound change in innovation approach, which has been described in this chapter. This is the change from a craft approach to innovation to an ‘industrial’ approach. Policy measures which accelerate the change in the management paradigm applied to engineering projects could be crucial in ensuring the transformation which has started to continue successfully. Appendix Table 17.1: Change in income structure 1984–1994: Shanghai Material Research Institute (million yuan) Year

Total income Government Market funds earnings

Structure of market earnings

Technology Technology development transfer

Technologic Trial al production consultancy and technologica l services

Other production and sales

1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994

2.73 1.42 0.83 0.83 0.65 3.03 2.75 1.85 2.05 2.84 3.51

1.38 5.83 3.40 3.07 3.31 0 6.84 11.33 17.80 26.17 19.14

9.10 8.29 8.78 10.02 10.73 17.36 16.02 25.92 31.17 50.85 41.68

6.37 6.87 7.95 9.19 10.08 11.60 11.92 18.12 26.20 34.46 28.81

0 0.12 0.08 0.30 0.46 0.42 0.46 0.72 0.73 0.35 1.53

3.80 1.24 3.67 5.31 6.05 8.15 2.11 2.42 0.84 5.19 5.28

1.19 0.48 0.80 0.51 0.25 2.68 2.19 2.63 5.55 0.56 0.01

0 0 0 0 0 2.87 1.44 2.69 3.76 2.19 2.85

Source: Zhang Meiguang 1995. Note: Amounts are not corrected for inflation. From 1989 on, the total income includes ‘other income’ (see Table 15.1) which includes support, donations and credits from domestic or international societies or individuals.

18 TECHNOLOGICAL TRAJECTORIES

Introduction This chapter provides an interpretation, from a technological perspective, of the transformations of industrial R&D institutes and some industrial enterprises which were observed in Chapters 15 to 17. It will focus on exploring the effects and characteristics of technological change during the market reform, in comparison to the pre-reform period. The changes at the micro-level which were presented in the previous three chapters, and the evidence regarding productivity and export performance introduced in Chapter 14, have shown that technological progress in the industry has accelerated since the reform began. This acceleration of technological change in Chinese industry has been generally recognized. This chapter will argue that it can be ascribed to a fundamental shift in the direction of technological progress in China’s industrial sector. The term ‘technological trajectory’ will be used in place of ‘technological progress’, because it has been observed that the cumulative and selective nature of technological change leads to movement along established consistent paths of development. Yet our observations have shown that it is possible, though difficult, to shift from one technological trajectory to another. The first part of this chapter will elaborate on the concept of technological trajectories to provide the conceptual foundations for the argument that the Chinese machinery industry has been shifting from one technological trajectory to another. The second part enumerates some elements which were missing in the technological trajectory of the pre-reform period, so as to identify the characteristics of pre-reform technological development in the Chinese machinery industry, and their basis in the institutional framework. The third part looks at the path-shifting process which has taken place during the current reforms, and the roles of existing R&D institutes. The focus will be on those changes in the characteristics of technological change whose effects are so profound as to constitute a shift from one technological trajectory to another. We will argue that the critical sources of the learning required for such a fundamental transformation in technological trajectory were internationally tradable technology and a further development of accumulated skills in design and testing, which were acting within a new framework of institutional objectives and incentives. In exploring this path-shifting between technological trajectories, which is driven by economic transition, it will be necessary to consider the rate and direction of technological progress in the pre-reform period and the impact of the previous economic system on that technological trajectory. What kinds of changes have been generated with respect to the most critical areas of technological efforts since the current reform? How did the changed economic system affect the changes in technological efforts? What kinds of skills and

210

TECHNOLOGICAL TRAJECTORIES

techniques were required because of the changes, and from where and how were the skills and techniques acquired? Technological trajectory For the purpose of this analysis, a technological trajectory refers simply to the cumulative processes of technological change, and the underlying definition of technology itself is a set of skills and methodologies applied in a certain field of industry for producing goods and services. Examples given will include technologies for metal cutting machinery, metal-forming machinery, power generation equipment, electric cables and internal combustion engines, as applied in the sectors which the study covers. The term ‘technological trajectory’ as it is used here draws attention to two aspects of the process of technological change: the cumulativeness and the selectiveness which follows from that. These two aspects will be discussed below. It should be noted that this usage differs from the way Dosi uses the term, particularly in a frequently cited paper ‘Technological paradigm and technological trajectory’ (Dosi 1982). In this paper Dosi defines a technological trajectory as the ‘pattern of “normal” problem solving activity (i.e. of “progress”) on the ground of a technological paradigm’ (brackets in the original). Here the concept of a technological trajectory centres on the ‘technological paradigm’, drawing heavily on the concept of scientific paradigm applied by T.Kuhn in his The Structure of Scientific Revolutions (1967). A technological trajectory is conceptualized as having a strong internally-determined dynamism deriving from the paradigm of technological knowledge. The intrinsic dynamism of the technological paradigm is a heuristic which points to some general tasks in a given technological field, and not to others, thus defining the direction and trajectory of technological progress. Dosi’s conceptualization of a technological trajectory would not be helpful in understanding the selectiveness or direction of technological change in general, or for interpreting the observations obtained in our study, since technology is inherently different to science. Science addresses the laws of nature, while technology addresses the artificial patterns created by human beings in their productive activities. Technological change is thus a process in which economic and social factors are indispensably and essentially involved. Dosi attempts to add ‘efforts and technological imagination’—that is, human behaviour attributes—to his concept, so as to include ‘technological and economic dimensions and tradeoffs’, but this cannot work under a paradigm-centred conception, since this leaves little room for observing and analysing the economic, social and institutional factors that interact with the technological change. The cumulativeness of technological change Cumulativeness refers to the fact that any successful change in technology is the result of an incremental and gradual process, because change involves technological uncertainty and economic uncertainty regarding both the production process and the marketing of the product. Experiments with technical solutions and economic viability are therefore indispensable, which means that the process of innovation can proceed only gradually. Cumulativeness also derives from the nature of the learning inherent in the process of technological change. Learning here refers to individuals’ and organized economic agents’ acquisition of the knowledge necessary to master the process of technological change (David and Foray 1995). Learning feeds back into the learning process itself, leading to successive improvements in the ability to absorb and generate the new knowledge required for the process of technological progress. As David and Foray (p.89) state:

MACHINERY TECHNOLOGY INSTITUTES

211

The process of knowledge generation in science and technology is cumulative and integrative. Knowledge cannot only be considered an output; it is also the main input of any process of knowledge generation. Cumulative forms of knowledge are those in which today s advances lay the foundation for succeeding rounds of progress. A synthetic knowledge is generated through convergence or collateral integration between previously independent pieces of knowledge. Thus, for many industries advances take place in a generational process; new products or processes are not radical departures from existing modes, but rather build on and extend the knowledge and technology used in the production of the products or processes they supersede. This property gives rise to successive improvements defining a trajectory. Learning is activated by, and focuses on, definite problems, and it is conducted in a distinct working locus. ‘Learning by doing’ (Arrow 1962), which implies learning that takes place especially on the shop floor, consists of increasing production skills and knowledge that yield an economic return. ‘Learning by using’ (Rosenberg 1982:120–140), takes place at the interface between the producer and user, where the operational performance of a design can be evaluated and skills and knowledge regarding the operation of the product are built up, with economic benefits in terms of both more effective operation and design modifications. Moreover, cumulativeness in the strict sense is still relevant in the machinery industry, which is a rather traditional industry originating in the nineteenth century. The most radical event for the industry in the second half of this century has been the introduction of computer-based technology. But computer technology has not altered the set of skills and rules that have driven technological change in the industry in China, as we observed. Arora and Gambardella (1994) refer in this regard to ‘the changing technology of technological change’, implying that the introduction of computer-related technology can best be seen as an improvement in innovative measures, rather than a change in the subject and nature of technological change itself. A comparable finding, drawing attention to the continuing importance of improvements in mechanical technologies, is provided by Patel and Pavitt (1994) with regard to technological change in industry outside of China.1 Thus the machinery industry offers a good opportunity to observe what impact social and economic factors have on the direction and characteristics of technological change, by comparing the technological trajectory of the industry in China before and after the current reforms. The comparison may reasonably be understood as being induced substantially by the introduction of market reform. The selectiveness of technological change Selectiveness refers to the fact that technological change takes place in some particular directions and not in others. From observations of the transformation of R&D institutes and industrial enterprises dealt with in the previous chapter, this selectiveness is to be understood also as a function of both economic and institutional factors, and not simply as the result of the intrinsic logic of the relevant technologies. By way of comparison, it will be useful to begin with a brief examination of some related views of the selectiveness of technological change under market economies.2 The well-known evolutionary theory of technological change by Nelson and Winter (1982) explains the selectivity of technological change on the basis of market mechanisms. At the inter-enterprise level, the direction of technological change is guided towards changes offering higher profitability, because higher profitability means success and a higher rate of expansion for the firms making the more profitable choice. Within firms, this theory points to the effect of technological searching and adaptive decision-making which

212

TECHNOLOGICAL TRAJECTORIES

enable firms to change their production functions in response to market (environmental) conditions. By searching, the firm finds technological alternatives in which it can invest, and this investment changes the operating routines of the firm, improving its input and output levels and its fit to the market conditions of the time. The searching and selective survival mechanisms function simultaneously and interactively to constitute the evolutionary process of technological change in Western market economies (Nelson and Winter 1982:14–20). Another widely influential theoretical explanation is that of Mowery and Rosenberg (Mowery and Rosenberg 1979; Rosenberg 1985) in the context of the ‘market pull or supply push’ debate. They contend that two groups of factors influence or motivate autonomous economic agents in technology investments. The first relates to production conditions external to the firm, such as changes in the input price, quantity, variety or volume, and continuity of supply. The second group of factors are internal, arising largely from bottlenecks encountered in the manufacturing and product design processes. These factors point to specific directions in which change is desirable and investment is necessary. It is stressed that the two groups of factors never act separately. Successful innovation is an interactive process, in which both demand and supply forces are responded to. Thus, successful innovations typically undergo extensive modification in the development process in response to the perception of the requirements of the eventual users, on the one hand, and, on the other in response to the requirements of the producer who is interested on producing the product at the lowest possible cost. (Mowery and Rosenberg 1979) The direction of cumulative improvements in underlying technology, and the dissemination of the improvements among firms and industries, is determined by this interactive process. Both these explanations identify not only technical reasons but also socio-economic factors that are critical in shaping the direction of technological change. Technical solutions are seen to be produced in response to bottlenecks arising from internal or external ‘imbalances’, while the configuration of, and the perception and response to disturbances and imbalances are all strongly connected to socio-economic factors. Among other things, the institutional framework of market economies which embodies these ‘socioeconomic’ factors is characterized, in comparison with a centrally planned economic system, by greater sensitivity in response to cost-effective factors and users’ requirements. This is because in a market-oriented economic environment, the users of machinery technology become more sensitive to cost effective performance and there are closer interactions with the supplier of technology, through market exchange. The way in which problems and bottlenecks are identified is often critical in steering the solution in a particular direction. Von Hippel and Tyre (1995) recently made an important comment regarding problem identification. They argue that problem identification takes place ‘in the field’ and is a leading part of ‘learning by doing’ and ‘learning by using’. Specifically identified information regarding a bottleneck focuses technological efforts—diagnosis, experiments and trial and error—on a particular area. Continual incremental accumulation of technological change can hardly proceed without effective problem identification from using and doing, because in practice ‘there are no well structured problems, only ill structured problems’ (cited from Simon 1973), and ‘ill-structured problems may involve an unknown solution space’, ‘unknown or uncertain alternative solution pathways’, and ‘inexact or unknown connections between means and ends’. Solutions for ill structured problems may be very costly and inefficient, or not feasible at all (Von Hippel and Tyre 1995).3 In short, the insights that technological change is cumulative and selective are explicitly relevant when one considers technological change in the context of economic transition. Problem identification, problem

MACHINERY TECHNOLOGY INSTITUTES

213

solving and adaptive decision-making all proceed within a particular institutional arrangement. Questions such as what problems are perceived and identified with greater sensitivity, and how messages about problems and bottlenecks are communicated for problem-solving and investment planning, are relevant with respect to the concept of technological trajectories, not least when the trajectory is assumed to be shifting due to radical changes in socio-economic factors. Missing elements in the technological trajectory in the pre-reform period This section examines the technological trajectory which developed in the pre-reform period, seeking to ascertain which characteristics the technological trajectory of the period had and which institutional factors relating to the pre-reform economic system had an effect on the characteristics of the technological trajectory. Characteristics of pre-reform technological development From our fieldwork and the literature search, it would appear that the pre-reform Chinese machinery industry was seriously deficient in developing technologies of two kinds: technologies relating to effective manufacturing, or ‘production economy’, and technologies relating to effective uses of its products, or the ‘operating economy’ of the products, although certain ‘higher order’ capabilities in R&D and design were developed and enabled this industry to expand rapidly its production capacity and broaden its product diversity, as well as to generate some technologically sophisticated machinery products. Technologies relating to production economy are defined here as methods and skills developed to address problems relating to economic production, that is, manufacturing machinery products with high quality and low cost. Skills and methods for testing, manufacturing and production management are closely connected to production economy. Some of these technologies are embodied in physical equipment and some are disembodied. Some very basic techniques, such as those for quality assurance, which are imperative for high-quality manufacturing, were neglected. Technologies relating to the operating economy of the final products involve mainly product planning and designing, such as skills and methods for feasibility, reliability, systems engineering and ‘unit design’. The distinction between these two kinds of technologies is to some extent artificial, since quality assurance and testing techniques, for instance, are employed to raise production quality which in turn directly improves the operating performance of the products. The technological weakness of the Chinese industry in manufacturing and production management, relating to production economy, is especially manifest in comparison with the East Asian developing market economies such as South Korea. The East Asian developing market economies started with international contractual manufacturing, usually OEM (original equipment manufacturing), where technological learning began with learning how to manufacture effectively. Under high competitive pressure, efforts have been made to continuously improve both manufacturing and management skills to meet the high standards in quality, productivity and short delivery time imposed by international buyers (Kim and Dahlman 1992; Hobday 1995: 40–45, 70–76; Lall 1990). First the skills and methods for effective and high-quality manufacturing were acquired, and these provided a basis for the more sophisticated stages of product and process innovation. On the other hand compared with industrial developed market economies, the weakness in the Chinese trajectory may be highlighted by the lack of skills and methods that are used for finding technological solutions for the construction of cost-effective machinery products, complex flexible systems, and for designing such systems efficiently and effectively. These skills and methods are sensitive to the use value of

214

TECHNOLOGICAL TRAJECTORIES

products, as well as to the value of engineering design work. One key has been the unit design principle which reduces complex systems into units.4 The unit principle is an important approach to planning complex machinery systems and has proved indispensable in the commercial development of these systems, both in terms of a unit production mechanism consisting of a succession of ‘elementary’ machines, and in a unit approach to the design itself. Some of the leading R&D institutes in China which were assigned to the development of manufacturing technologies, such as forging technology, had mastered the technological know-how about the processes yet this did not lead to effective design engineering for the development of commercially competitive manufacturing systems. China did develop some design abilities from the late 1950s and was soon undertaking some product innovation work, much earlier than the East Asian NIEs. The fact that its design capacities stagnated points to serious deficiencies in its technological trajectory. Table 18.1 Types of imported technology Technologies imported largely for manufacturing enterprises

Technologies imported mainly for R&D institutes

product designs; manufacturing techniques and equipment; production management techniques (including quality assurance); auxiliary equipment; testing techniques and devices; standards

feasibility study methods; various skills and method for design (e.g., unit design, reliability design, CAD and finite design), systems engineering; standardization

Source: Various cases in Chapters 16 and 17.

The deficiencies of technological accumulation in the Chinese machinery industry can be substantiated by looking at the kinds of technologies which were imported following international trade liberalization. The imports reflect the pent-up demand due to deficiencies in technological accumulation. Table 18.1 lists the principal imported technologies which have been noted in the preceding chapters, differentiating between two groups of importers: manufacturing enterprises, and R&D institutes.5 It will be observed that imports of technologies relating to production economy were primarily for enterprises, while imports of technologies relating to operating economy of machinery products were for both R&D institutes and enterprises. During the 1980s, turn-key projects, which were the major form of imports in the 1950s, have become much less important. This list of technology imports however shows the overall backward situation of the Chinese machinery industry in skills and methods used for effectively manufacturing, and for conceiving products with better use value to fit users’ requirements. The deficiencies in these basic skills contrast with the general image of the industry, whose size and initial mastery of product diversification placed it ahead of its counterparts in many developing countries (see Tables 14.1 and 14.2). As has been shown in Chapter 14, the size of the Chinese machinery industry as it developed in the three decades from the 1950s is very impressive in comparison with the Indian, Brazilian and South Korean machinery industries in terms of every quantitative indicator; the later three are regarded as typical of the developing countries which have successfully entered into complex capital goods production. But in terms of productivity and competitiveness, the performance of the Chinese industry was inferior to the industries of Korea, Brazil and India. The inferior productivity and competitiveness were partly reflections of the deficiencies of the ‘technological trajectory’ of the Chinese industry under a particular institutional arrangement, the centrally planned economic system.

MACHINERY TECHNOLOGY INSTITUTES

215

Table 18.2 Characteristics of the technological trajectory in selected sectors of the machinery industry Sector

Milestones of technological change

Major goals

Machine tools

1 application of high speed cutting techniques (1950s) 2 product variety and quantity

1 efficiency of machining

2 diversification of general purpose machine tools, including the introduction of traditional precision machine tools 3 attempted but failed to introduce CNC machine tools (late 1950s and 1960s) Electricity generation equipment 2 some innovations for capacity expansion, including a turbogenerator with inner water-cooled stator and rotor, and 700 mm turbine rotor blade 3 the development of manufacturing technology, but systems engineering was weak Low-power internal combustion engines 2 expansion of industrial production capacity 3 relative neglect of the development of manufacturing technology, reliability performance, and fuel economy Electric cables 2 product variety diversification and production capacity expansion Forging technology

2 a few efforts were made for military purposes and the automobile industry Automation technology

Source: Chapters 14, 16, and 17.

3 technological catching up

1 capacity increase of single machines to 300 MW

economies of scale

1 diversification of product variety, mainly for agricultural uses

product variety and quantity expansion

1 material substitution, especially aluminium to replace copper 2 product variety and quantity expansion 1 almost no progress in the introduction of new and custom-made forging technologies

1 material input substitution

concentrated on the first implementation of a particular system in one manufacturing site (exemplified by the practice in the early 1980s, under the planned approach)

Automation technology was disseminated in the same way as forging technology. The planning approach of the 1980s was adopted for upgrading automation technology, which was rather outdated at that time

Forging technology was disseminated through the provision of standard, general-purpose forging equipment.

216

TECHNOLOGICAL TRAJECTORIES

These deficiencies in the technological trajectory prior to the reforms need to be balanced with a review of the technological achievements which were achieved during that time. We approach such a review in Table 18.2 simply through summarizing some milestones of technological change described in various case texts in Chapters 16 and 17, with additional reference to the two sources concerning the history of the Chinese machinery industry frequently cited in Chapter 14.6 Short comments are included on the major goals that seem to have motivated these achievements. The data summarized in Table 18.2 shows that technological change did take place to the Chinese machinery industry during the pre-reform period. As we have seen repeatedly in the previous chapters, it was directed chiefly at product diversification and production capacity expansion. However the goals listed indicate that the efficiency of manufacturing and better operating performance of machinery products were not entirely ignored. While product diversification and production capacity expansion were the chief goals in the sectors covered in our survey, five cases which relate to manufacturing efficiency and the operational performance of products can be highlighted: • the development of large capacity electric generation equipment (Case Texts 14.5 and 16.4) was intended to achieve economies of scale; • the dissemination of high-speed cutting techniques in the 1950s (Jing Xiaocun (ch.ed.) 1990: Vol. A, 22) which was intended to increase machining efficiency; • the development of precision machine tools (Chapter 14: Case Text 14.4) was aimed at better quality metal-cutting; • an enormous effort to improve manufacturing technologies was made particularly in the ‘key S&T projects’ of the five year plan in the 1980s (Part 1, Chapter 2); Part 3 Case Text 17.2); • efforts in material input substitution, illustrated by the case of substituting aluminium for copper in electrical cables (Case Text 16.5), made the production of electric cable economically feasible based on cheaper and locally available materials. The success of these limited efforts varied. The second and fifth cases above were relatively more successful than the first, third and fourth as discussed in the related chapters and the literature referred to there. Input material substitution and the dissemination of mature manufacturing techniques such as high-speed cutting techniques were relatively successful. Moderate success was achieved with new product developments but there was a weakness in the simultaneous development of effective manufacturing techniques, illustrated especially in the case of electric power generation equipment. And some kinds of manufacturing technology which could not be incorporated in standard equipment were basically beyond the capacity of planning coordination. The stated goals of the machinery industry under the Chinese centrally planned economy were not hostile to technological changes, in fact it was very enthusiastic. But technological change in the industry took a certain direction—it was selectively sensitive to general technology, and selectively insensitive to specific technology. General technology is technology which, once acquired, may be widely employed without appreciable adjustment for particular manufacturing process and products and specific operational conditions. More mature product and manufacturing technology and material technology are both general technologies. Materials for input substitution in machine production are of a ‘general’ kind, because the inputs can be repeatedly applied by many producers for a family of product categories, often without any special recomposition. The diversification from mature basic designs, which was the most successful type of technological change in the industry in the pre-reform era, is also of a ‘general’ kind. ‘Specific’

MACHINERY TECHNOLOGY INSTITUTES

217

technology, in contrast, is technology which has to be developed or modified specifically for a certain process operated under certain conditions with a certain target of performance. Forging systems and automated manufacturing systems are typical and extreme examples. Such systems cannot be generally applied to different work pieces in any manufacturing site without considerable modifications. Further questions arise. Why was the trajectory of technological change that developed in the pre-reform regime sensitive to general technology and insensitive to specific technology? And why did the characteristic technology development trajectory lead to unsatisfactory performance for the industry (as indicated in Chapter 14) and ultimately to the wide acceptance to the current market reform? Restricted ‘specific’ learning and its institutional basis The sensitivity of the technological trajectory to general technology was largely because the institutional structure was conducive to the development of general technology, and its insensitivity to specific technology was in part because learning to solve specific problems was blocked under that institutional structure. The Chinese machinery industry suffered particularly from institutional blockages to specific learning in the first few decades after 1949. The institutional structure of the machinery industry under the centrally planned economy was described in Chapter 14. The aspects which are relevant here are the dominance of vertical administrative coordination and two other associated factors: the separation of establishments organized on functional principles and the fact that the incentive structure centred on quantitative output. Learning to solve specific problems was blocked because it was not part of the institutional agenda of machine-producing and machine-using enterprises, which are the natural place for specific problems to be ‘identified’ and ‘structured’, the first step in problem-solving. It would be unfair to say that the Chinese engineers in the machine-producing and machine-using enterprises were entirely unaware of problems. Problems were recognized and more or less structured in the field by the technical staff most closely involved, but this was weakly motivated, and addressed poorly or not at all in the decision-making agenda.7 The problem was that the incentive structure of the enterprises was oriented to quantitative output, so that messages about problems relating to lowering production costs, improving the user’s satisfaction or increasing market penetration were not on the active agenda of the enterprises. Such matters received only ‘passive’ attention, to adopt Arrow’s terminology (1974:47–59).8 The active part of the decision-making agenda was devoted to fulfilling quantitative output goals, including those for technical solutions. Thus some peculiar or ad hoc measures tended to be taken, which usually did not lead to a steady accumulation of learning aimed at improving quality or reducing costs (See Case Text 16.4). On the other hand, decision-making in the higher levels of the planning coordination framework could only deal with general technological change, since it was separated from the enterprises where specific problems are signalled. This vertical coordinating structure proved relatively successful for general technologies such as material development and product diversification, which were in fact the major objectives set for the R&D and innovation system for the Chinese machinery industry from the 1950s. In the case of material technologies, the planning coordination mechanisms were successful because materials are homogenous, and therefore less user-specific, and there is less need for specific information feedback from and for particular users, so that a centralized initiative can more easily achieve some degree of success.9 In the case of product diversification based upon mature designs, considerable success was initially achieved under the central planning mechanism because a great deal of specific learning and problem-solving had already been accumulated and incorporated in the designs. But communication between the producer and users were still found to be indispensable when a new variety was introduced. The necessary interactions were to some

218

TECHNOLOGICAL TRAJECTORIES

extent established through sending design engineers to work temporarily in the ‘field’, in producing enterprises, to identify and solve problems based on specific information. Once the seconded R&D personnel had departed, there were no formal mechanisms for identifying and solving new problems, and improvements largely ceased. The bias of the technological trajectory of the Chinese machinery industry in favour of general technologies was to a large extent a natural consequence of the institutional structures of the planning mechanism. The results of a trajectory of technological development which was sensitive to general technology, but insensitive to specific learning, were felt in a low rate of productivity improvement and the failure to create specific competitive advantages for the industry itself, or for the Chinese economy as a whole (see Chapter 14). It was a trajectory which did have strengths in increasing production volume relying on general technology, but with little ability to create firm-specific or country-specific advantages. Since the machinery industry is an upstream industry it functions as an innovation centre for many machineusing industries in the economy (Rosenberg 1985:9–31). An economy with little momentum in the improvement of productivity and in the creation of locally specific competitiveness will benefit less from industrial investment, and is vulnerable to increasing international competition when opening to international trade. To use Porter’s terminology (Porter 1990: Chapter 10), the pre-reform Chinese economic development stagnated at the ‘factor driven’ stage for a prolonged period because it was unable to create specialized factors to upgrade the economy to higher-order competitiveness.10 One lesson to be learnt from this is that a higher level of the administrative hierarchy cannot handle horizontal interactions between the functions of design, manufacturing, marketing and the operation of final products. Continuous horizontal information flows are vital for identifying specific problems and implementing solutions. This is especially true for the machinery industry, where successful technological change is particularly dependent on intensive interactions between designer, manufacturer and user, because the machinery industry is one of the ‘specialized suppliers’, to use Pavitt’s categorization of technological trajectories (Pavitt 1984). That is, in this industry users’ requirements are central to shaping the direction of technological changes. Interfaces between manufacturing, design, marketing and maintenance carry the most important information flows in this industry. Japanese successes in this area have relied largely on remarkably intensive horizontal interactions including intra-firm cooperation along innovative ‘chains’, under tough competitive pressure (see, for instance, Aoki 1988). For other industries, such as chemicals, the critical interface seems to be between laboratory experiments and equipment development. Finally, it can be seen that the incentive structure in this industry was distorted by direct vertical administrative intervention, so that technological efforts were unavoidably directed to non-specific learning. There are widespread observations to support the thesis that an institutional framework which does not emphasize horizontal interactions leads to a machinery industry which develops with little ability to internalize innovative skills or to transmit them to the economy as a whole, resulting in a lower level of international competitiveness.11 Path-shifting and the roles of existing R&D institutes during the current reform The analysis above has shown that, in the pre-reform period, a kind of technological trajectory had developed which incorporated strong socio-economic parameters. The institutional arrangement, together with the incentive structure which it engendered, provided a societal framework in which human technological activities were guided and conducted. The dominance of vertical planning coordination in China led to a technological trajectory which was sensitive to general technological change and insensitive

MACHINERY TECHNOLOGY INSTITUTES

219

to specific technology. As a result, China developed a machinery industry which accumulated competence primarily in the design and production of general, mature technologies. The technological trajectory will necessarily shift to a new path if the basic arrangement of socio-economic institutions is altered. This happened in the Chinese system with the withdrawal of planning coordination and the delegation of decision-making autonomy to industrial enterprises, beginning in the early 1980s, and to the R&D institutes from 1985 onwards (see Part I). This led to a gradual but decisive change in the technological activities of both R&D institutes and productive enterprises, and also to changes in the institutional structures, which will be the topic of the next chapter. The reorientation of technological change during market reform: enterprises and R&D institutes To understand how the technological trajectory of the Chinese machinery industry has shifted towards a new path in the course of the current market reform, it is necessary to look at the major actors in technological change in the industry to see how the selective direction of their technological efforts has altered. Until the mid-1990s, technological change in the machinery industry relied on two main groups of actors: large manufacturing enterprises and the previous machinery technology R&D institutes. R&D institutes continued to have an important but changed role, including some characteristic functions which are unique to a transitional developing economy. We summarize the technological activities of the two groups of actors from the more detailed descriptions in Chapters 15 to 17. The emphasis here is on the R&D institutes, which are the focus of the study. The manufacturing enterprises (Case Texts 16.1, 16.2, 16.4) All manufacturing enterprises in the industry have become increasingly aware that they have to be commercially competitive and technologically innovative rather than merely passively producing a quantity of output. But only large enterprises can afford to draw on imported technology to achieve a fundamental shift in their technological trajectory. Imported technology embodying the technological achievements developed in the context of market competition can contribute the missing elements in relation to effective and highquality production and the management skills needed to achieve it. As they have become relatively sensitive to competitiveness, large enterprises have been keen to learn to be innovators. Steady efforts have been made to exploit firm-specific strengths. As a result, the rate of technological change has quickened, and its trajectory is becoming increasingly sensitive to factors related to that we call the economies of production and operation. The group I–B R&D institutes (see Table 14.12), which were based in a host enterprise but were previously centrally commissioned to undertake development work in a particular field of general technology, have become strategic assets for their host enterprises with an emphasis on firm-specific technological change. However the manufacturing enterprises have not yet achieved critical success as major producers of technological innovation. This is due to their lack of experience in specific learning and to inherited institutional barriers reflected in inappropriate internal work organization and external relationships that need to be restructured. Nor has this group been able to play a significant role in disseminating innovative capacity to small and medium-sized manufacturers through subcontracting work or providing engineering services, which is often the case in industrially advanced market economies. Thus far, the role of transmitting the path-shifting dynamic to small manufacturers has been largely played by the R&D institutes.

220

TECHNOLOGICAL TRAJECTORIES

The existing R&D institutes The existing R&D institutes have been transforming themselves to become i) suppliers of product and plant engineering; ii) developers of complex manufacturing systems; and iii) producers of selected finished machinery products. The first two of these will be considered below, to see what impact these institutional transformations may have had on the industry’s shift to a new technological trajectory. Product and plant engineering and the development of complicated manufacturing systems are varieties of ‘engineering services’—they both transmit technology to producers and so have widespread ramifications for the direction of technological change in the recipient firms. AS SUPPLIERS OF PRODUCT AND PLANT ENGINEERING SERVICES (SEE CHAPTER 16)

Most of the several dozen R&D institutes which were previously centrally affiliated and were assigned to the development of product technology in areas such as small electrical machinery and apparatus, small agricultural machinery, conventional instruments, and small electric-driven or manual tools have been transformed to provide product and plant engineering services. These are sectors in which technological change is less radical and there are many small and medium-sized manufacturers. The transformation of existing R&D institutes changed the way in which technology is supplied, and has had a significant impact on the direction of technological change in these sectors. During the reforms they have incorporated into their engineering services some technological components relating to effectively manufacturing products with a higher use value. These components include cost-benefit calculations, quality standards, non-repair time and noise reduction, all of which were observed in the case studies. These previously neglected components were incorporated through the intensive learning of design skills and methods by the R&D institutes. Small and medium-size enterprises became more risktaking, and have begun to seek technologies which promise greater cost-benefit efficiency. The development of firm-specific techniques has become more feasible now because contractual relationships facilitate direct horizontal communications. But customization of engineering services to particular individual users is still rare, mainly because small and medium-size manufacturers with inferior capabilities and meagre investment funds have not been sophisticated in contracting for these services. Nevertheless, the path of technological change in small firms has begun to alter. Besides, contractual delivery has helped the suppliers of such services to penetrate the boundaries dividing industrial and regional sectors which were set up under the centrally planned administration. Thus the abilities accumulated and learned by the group of best institutes can be widely disseminated to many other metalprocessing users. AS DEVELOPERS OF COMPLEX MANUFACTURING SYSTEMS (SEE CHAPTER 17)

Systems development has an impact on the technological trajectory of complicated manufacturing processes in sectors such as automobiles and other heavier and complex machinery. Complex manufacturing systems are product-specific and specific to the user enterprise in which they are installed. Product-specificity arises because each manufacturing system is developed to process one particular kind of machine part, such as an automobile front axle. User specificity is a matter of tailoring a system in engineering terms to the particular manufacturing conditions of individual producers. The productivity, efficiency and effectiveness of complex manufacturing are sensitive characteristics of the systems. The transformation towards becoming commercial suppliers of complex manufacturing systems is therefore a break with the former technological trajectory, in which the Chinese machinery industry used to apply only

MACHINERY TECHNOLOGY INSTITUTES

221

general-purpose machines to specific processes. Perhaps ten of the best qualified ‘manufacturing technology’ related institutes are involved in this transformation, which is still in its initial stages and is limited to manufacturing systems in a few areas. The development of the systems which are specific to products and to manufacturing conditions was able to gain ground only in the course of the market reform. Market reform generates demanding users of the systems, who Table 18.3 Changes in the technological trajectory during the reform: actors and characteristics Actor

Change in technological trajectory and its source Extent of impact

Large manufacturing enterprises (several hundred active enterprises)

• market reform motivated the development of firm-specific comparative advantages, leading to a shift in the direction of technological change for large enterprises, away from general and quantitative, production and towards quality and efficiency-sensitive production • technology imports have embodied the direction and characteristics of trajectories developed in the context of commercial competition. This supported the path-shifting in the enterprises • more than one hundred R&D institutes have been organizationally and strategically integrated into large enterprises, engaged in technology specific to their host firms, in contrast to their previous centrally directed commitment to general technology • missing elements of the technology trajectory, concerning ‘production economy’ and ‘operating economy’, have been incorporated through changes in design principle and methods. This provides a basis for the development and provision of engineering services with altered characteristics • technology imports embodying the missing elements have supported the path shifting. The domestic suppliers assimilate imported technology, and often adapt it to local users, assisted by their accumulated engineering capability • the development of firm-specific technology has been possible due to direct producer-user communication under contractual relationships,

R&D institutes as suppliers of product and plant engineering (several dozen active institutes)

• the impact has centred on the more complicated and advanced machinery which used to be produced by large enterprises • the path-shifting has not generally achieved significant success, because of the institutional restrictions faced by the enterprises • the large enterprises have thus far played an insignificant role in disseminating the impact to small and medium-sized manufacturers

• the impact has centred on sectors in which the underlying technology is more mature and simpler, and the firm structure is highly decentralized. These sectors constitute a large part of the machinery industry • the effect, in terms of shifting to a new technological trajectory, has been limited and restricted mainly by the demand side because of the many small and medium-size enterprises’ inexperience and inadequate investment in technological change

222

TECHNOLOGICAL TRAJECTORIES

Actor

Change in technological trajectory and its source Extent of impact although significant customization remains rare

R&D institutes as developers of complex manufacturing systems (about ten active institutes)

• systems of this kind incorporate a high degree of product-specificity and firm-specificity, and are sensitive to manufacturing efficiency and quality. The ability to customize specific technologies represents a break with the previous trajectory, which was tied to general purpose machines • developers realize the systems either by combining their core knowhow with imported engineering methods and basic equipment (as in the case of metal-forming systems) or by combining application engineering with internationally tradable components, parts and devices (as in manufacturing automation systems) • contractual relationships have ensured the horizontal information flows necessary for customized development of the systems

• the impact has centred on complex manufacturing processes, limited so far to metal-forming and some manufacturing automation applications, but not including metalcutting applications • the automobile industry and some other heavy machinery manufacturing has begun to benefit from the development • the developers of such systems are faced with further, and even more fundamental, transformations if the preliminary success is to be consolidated as part of the critical strengths of the industry

seek manufacturing technology of higher quality and efficiency in the face of increasing competition in the machinery product market. The contractual relationships in the technology market enhanced and secured horizontal communications between the developer and user, providing the flows of information needed for tailoring the systems to the firm and product specificities. Long-term contracting, officially recorded as ‘technology development’ contracts, has been developed to ensure enduring communications, whereas mature product and plant engineering was generally supplied under relatively short-term contracts and recorded as ‘technology transfer’ and ‘technological consultancy and technological services’. The shift in the approach to developing manufacturing systems has entailed intensive technological learning for system developers, including learning some necessary basic design principles and methods. The discussion above is summarized in Table 18.3, which lists the actors who have played a major role in changing the path of technological development since the current reform, what their impact has been, and the characteristics of the new technological trajectory. From the discussion above, it can be concluded that the Chinese machinery industry has made initial, but widespread and significant, steps towards shifting the direction and characteristics of its technological change. The technological path-shifting which has emerged is directed mainly towards increased sensitivity to manufacturing efficiency and quality, and to the use value of the manufactured products. This has been stimulated by the market reform, which have changed the active area of the decision-making agenda and the operational routines of various economic agents. The latter change has required enormous institutional restructuring, which we will discuss in the next chapter.

MACHINERY TECHNOLOGY INSTITUTES

223

The effect of internationally tractable technology International trade has been critical to the path-shifting process. Technology imports have had a direct effect in enabling the Chinese machinery industry to upgrade its technological basis, as would be expected. But acquiring a technology will not necessarily assist an industry to redirect its dynamics. This was the situation in China prior to the current market reform: imports of Western technology were not entirely barred but they were applied to maintain a different trajectory. It is very interesting to examine the effect of technology imports on facilitating technological path-shifting during the economic transition period. For this purpose it is useful to differentiate two effects of imported internationally tradable technology, the ‘demonstration effect’ and ‘instrument effect’. Internationally tradable technology has a demonstration effect because a given item of technology that has been developed under commercial competition also serves as an example of what can be achieved by technological accumulation along a different trajectory. There is a knowledge spillover in the sense that the tradable technology highlights the weak areas in the technological accumulation of the Chinese industry. The learning which is needed to shift to the new technological trajectory is better targeted and the learning requirements, which may already be known from the literature, are cast in the form of a ‘procedural’ knowledge demonstration. For example, the engineers of the Beijing No. 1 Machine Tool Plant reported that the details embodied in imported blueprints enabled them to clarify or correct their understanding of technological development in other countries, which they had more or less known about from literature. The transmission system used in milling machines, for instance, was found to be rather different to their previous perception (see Case Text 16.1). The instrument effect of tradable technology rests on the fact that an item of imported technology inherently embodies engineering methods, developed out of a commercially competitive trajectory, which can be adopted and employed for further development of the technology which will in turn tend to support a trajectory with similar characteristics. An import which may be quite specific to a certain purpose may contain a whole array of technological solutions to similar but distinct problems. The concepts and skills used for engineering design such as feasibility study, reliability design, unit design, and systems engineering are examples of ‘software instruments’ which have developed in the context of commercial competition and have been vital for the path-shifting, as they have been widely learned and applied by productive enterprises and in R&D institutes. Drawing upon these methods and concepts, R&D institutes have been enabled not only to embark on the trajectory of specific flexible manufacturing systems, but also to assist significantly the numerous small and medium-sized producers to shift the directions and characteristics of technological change. In a more concrete sense, the instrument effect involves offering physical apparatus needed for pathshifting. Testing and measurement devices, basic machining equipment and electronic components and devices are examples. Once the design method had been learnt, it was possible to construct forging systems based on various kinds of ‘elementary equipment’, including forging manipulators and testing techniques, with only small modifications (Case Text 17.1). In this case, the imports of soft instruments, typically design concepts and methods, and of hard instruments together provided crucial support for the local developer to focus on specific innovations in which its own accumulations of core engineering know-how could be incorporated. Similarly, the developer of manufacturing automation systems was able to emphasize application engineering to cope with specific user-firm conditions, without having to undertake original development work for most of the basic electronic components, devices and design norms for automation systems, which were already internationally tradable (Case Text 17.2). The distinction between the demonstration and instrument effects is somewhat artificial. It nevertheless helps to show how the international trading of technology can, under a favourable institutional structure,

224

TECHNOLOGICAL TRAJECTORIES

support the necessary transformation of technological trajectories. The value gained goes beyond the immediate value of the traded technology. These effects can be considered as the spillover of technological knowledge. Ed Steinmueller (1996) has argued, with the focus on the information industry, for the ‘involuntary spillover’ of technological knowledge from a commercially developed technological trajectory, and sees it as playing the role of an international ‘technological infrastructure’. This means that the spillover has supported the dissemination of commercially competitive technological knowledge and helped new entrants entering the industry.12 Our argument is similar but concentrates specifically on the period of economic transition and focusing on the machinery industry. According to our observation, the effect of involuntary spillover of commercially competitive technological knowledge which is embodied in international tradable technology has likewise been significant and critical to the path-shifting of technological change during economic transition for the machinery industry. It has provided necessary sources of technological knowledge for the Chinese industry, assisted in the acquisition of technological competence which had been seriously lacking in the industry, and guided it in moving to a higher level of international specialization from having been a producer of general purpose machines, although all these changes of the Chinese industry are still in the initial stages.13 The changes in the institutional arrangements for technology imports during the reforms have been beneficial for learning from internationally tradable technology. Industrial enterprises and R&D institutes, which have become more sensitive to quality and efficiency issues, delegated decision-making autonomy and were entitled to interact intimately with their foreign suppliers. This ensures that information on specific problems embodied in traded technology is able to be channelled to the proper locations. In contrast, the technological imports in the 1950s to 1970s were going to plant design institutes which were assigned to absorb the imported designs (Chapter 14). Since the plant design institutes were not concerned with manufacturing and product-specific details, the information coming from foreign suppliers, whether Soviet or Western, was to a large extent disregarded and ignored, in addition to the problem of weak incentives. Continuing international exposure is necessary for effective learning. This is because learning to adapt to change is itself a trial-and-error process. The early adaptive strategy for technological learning usually focused on simple imitation of imported technology. This was later modified to focus on more specific issues, and to incorporate the institutes’ own minor innovations as well as the restructuring of work organization. The modifications in the learning strategy were the result of early experience which identified limitations and solutions for successive stages. In other words, continuing exposure to the international technological infrastructure means an ongoing functioning of the demonstration and instrument effects, which will specifically highlight the goals and the means for proceeding with and deepening the learning process. Recombination: the innovative approach to path-shifting What does the process of path-shifting imply from the perspective of technological innovation? The term ‘technological recombination’ is appropriate to express some unique aspects of the process. As many researchers have observed, industrial technological innovation is often a process in which elements which have been generated and used elsewhere are combined in a particular application (Kline and Rosenberg 1986; Foray 1995; Kodama 1990). A new scientific breakthrough does not necessarily serve as an input in most innovations even in important cases. One classic example is the combination of precision machining and electronic control, a combination which generated CNC machine tools (Kodama 1990). Kodama calls this ‘technological fusion’. Another example is ‘informatics’, a term widely used to refer to innovations that

MACHINERY TECHNOLOGY INSTITUTES

225

incorporate information technology in mechanical systems. This concept of innovative combination emphasizes the importance of being able to use available technological knowledge for the creation of a new application. In both these examples, the combinations involve the creation of a concrete technology in a certain area of a particular industrial sector. But the term ‘recombination’ used here is intended to emphasize the need to make up lost ground, to correct weaknesses in some dimensions neglected in the previous trajectories in an entire industry. During the process of recombination, old and new elements of technological competence are recombined, new features of technological changes are constructed and dramatic institutional restructuring entailed, leading to what we have termed ‘path-shifting’. Accordingly, the technological recombination induced by economic transition goes far beyond what is implied by the usual term combination. Technological recombination as the term is used here, can therefore be seen to be a particular and strong case of combination. In technological recombination, it is particularly indispensable that every actor undertakes far-reaching changes through intense technological learning and radical institutional restructuring. It is of interest to ask which accumulated technological capabilities have been most useful for technological recombination, given that internationally tradable technologies have been generally accessible. Two kinds of capabilities accumulated in the past, in design and in testing, seem to have been critical. They have provided a basis upon which external sources for technological path-shifting could be better assimilated and incorporated in the development of a new trajectory. Design skills are critical because ‘the path (of a technological innovation) begins with a design and continues through development and production to marketing’ (Kline and Rosenberg 1986). Design skills not only enabled the learner to understand better the technological knowledge shown from traded technology, but also provided the means to implement the learned knowledge into new designs. Incremental innovation based upon imported technology may begin immediately with design modifications which incorporate accumulated experiences. Testing skills provided an analytic capacity to increase understanding of the externally developed technology and to assist the further development of a trajectory using the newly acquired elements of technology. There seemed to be a greater likelihood of innovative recombination where both sets of skills had been accumulated. The better adaptive performance of the centrally affiliated R&D institutes may be explained by their design and testing abilities, and those productive enterprises whose task environments had required design and testing skills could also do reasonably well in innovative recombination although in fact they rarely achieved this. The electric power turbine manufacturer which improved blades for the Westinghouse 300 MW design (see Case Text 16.5) provides an illustration of a success achieved in an manufacturing enterprise where these capabilities were accumulated more strongly than many other enterprises. Both design and testing capabilities were related to previous ‘research and development’ work that was conducted under the strategy of national self-reliance and coordinated by the central planning administration. As the learning needed for path-shifting became intensive, testing work for the assimilation and recombination of imported technology has been expanded while R&D work for original innovation seems to have been reduced. Testing has been extended to cover new parameters such as reliability, non-repair operation time and noise reduction, and to more complementary techniques that were neglected in the previous trajectory, such as those for measurement, connection and auxiliary devices. This may be illustrated by the data showing the trends in the allocation of technological resources in the existing R&D institutes. It is reported that the centrally affiliated R&D institutes in the machinery industry devoted 5.7 per cent of their expenditure and 8.0 per cent of their technical personnel to basic and applied research in 1994, down from 11.5 and 13.5 per cent respectively in 1990. ‘Experimental development’ and ‘dissemination and

226

TECHNOLOGICAL TRAJECTORIES

application’, which accounted for 73 per cent of expenditure and 67 per cent of technical personnel in 1990, increased to 86 per cent of expenditure and 77 per cent of technical personnel in 1994 (RISA 1995: appendix tables 17.1, 17.2).14 The terms ‘experimental development’ and ‘dissemination and application’ may not correspond exactly to the category of ‘design and testing’ used above, they nevertheless point to the tendency towards a higher intensity of design and more application-oriented testing that is associated with the reallocation of technological resources indicated in the statistics. Finally there has been a significant investment, mainly from the government funds, in the restructuring of engineering capabilities. This has fallen into two categories: investment for the Key S&T Projects of the Five Year Plan during the 1980s, and investment for the rehabilitation of testing and experimental bases during the late 1970s and early 1980s. The latter had enhanced the physical facilities for testing and experiments more significantly than at any period since the 1950s (Interviews with Mr Zhu Sendi 1994 and 1995). Investments from the government plan for the Key S&T Projects, though it did not meet initial expectations in strengthening technological capabilities for effective manufacturing, did lead to an improvement in some areas by setting priorities and by importing foreign technology. The investment was later able to be recombined in more effective ways as economic and S&T reforms went deeper. There was also a considerable investment in time. Over a period of some ten years, the R&D institutes and enterprises have been involved in a process of continuous and wide-reaching institutional experimentation and technological learning. Challenges ahead Embarking on shifting to a new technological trajectory has entailed intense technological learning and institutional restructuring. This has in turn imposed the need for more technological learning and further institutional restructuring because inappropriate elements in the deeper levels of the old system became visible. Here we focus on two major challenges which must be met to rebuild technological capabilities so that technological innovation can proceed continuously along a new trajectory: The need for greater capabilities in the generation and dissemination of engineering knowledge As the innovation system of the machinery industry has moved to a new path, technological activities have become more intensive and extensive not only in the existing R&D institutes but also, and increasingly, in productive enterprises. Technological activities tend to be oriented to more specific problems and to involve more frequent horizontal communications, both within an institute or an enterprise and externally. The generation and dissemination of engineering knowledge, as distinct from the production of physical products, has therefore become a problem area. Bottlenecks in knowledge dissemination have limited the efficiency of innovative efforts. One such bottleneck has emerged at the interface between individual developers of innovative systems and their own engineering teams or departments, as illustrated by the development of the automated high bay warehouse (Case Text 17.2). The problem appeared to be that the engineering knowledge acquired during systems development was not able to be codified, so that there was little basis for the development of an internal division of labour in support of improving the productivity of innovation. Another bottleneck become apparent between designers and computers, because of a lack of experience in codifying engineering design knowledge. Designers who were used to working with paper and ruler and acquiring their design know-how personally and tacitly could hardly explore the advantages of computer-aided

MACHINERY TECHNOLOGY INSTITUTES

227

techniques. Limited design productivity resulted in long delivery times and a significant loss of market share (Interview at the Beijing No. 1 Machine Tool Plant). At another interface, that between domestic manufacturers and their international partners, the need for intensified inter-action has led to increased investment in documentation and codification (Interview at the Shanghai Turbine Works). Communication bottlenecks have also occurred between engineering design and manufacturing departments, caused again by a lack of capability in the codification and coordination of engineering knowledge among these departments (Interview at the Dalian Modular Machine Tool Research Institute). The inappropriateness at deeper levels inherited from the old system derive from the fact that product diversification was at that time technically supported by rigid standards, the complexity and dynamics of technological change were much reduced, and communications were simple, routine and static. As the focus of technological activities has shifted to more local and firm-specific improvements, enterprises and institutes alike have been faced, for the first time, with the need to generate original trial engineering advances. This entails the codification of engineering knowledge15 much more intensive at lower levels than under the centrally planned system, but the capability required had unfortunately been developed in the past only in rare cases. Codification itself is a form of innovation: it promotes the systematic reiteration of experience. It also provides the necessary basis for internal and external communications which, according to our observations, are critical for the formation of firm-specific and institute-specific competitive assets. However the need for improved capabilities in generating and disseminating engineering knowledge has not been properly perceived in policy thinking. The Engineering Centre Programme, for instance (see Part 1: Chapter 6), has been designed and understood as a programme to accelerate the physical outcomes of technological change, with little attention given to enhancing the generation of engineering knowledge. The need to rebuild interactions between academics and industry The market reform seems not to have weakened the dissemination of product and production technology in the industry, as can be seen from the conclusions to Chapter 16. In other words, the fundamental transformation of industrial technology R&D institutes from government-run to market-intermediate institutions has been a qualified success16 as regards the functions previously assigned to them for mature technology diversification and dissemination. A challenge however remains in the area of scientific support for industrial technology. Under the previous institutional arrangement this was a passive area of the academic R&D agenda because technological change was oriented around general technologies. As technological activities in both industrial technology R&D institutes and enterprises have become more strongly oriented to specific issues, and technological change has accelerated, scientific support for industrial innovation would be expected to become a problem area. This can already be observed to some extent. For instance, there is inadequate scientific support relating to the methods used in testing and experimentation with technical know-how, such as vibration analysis, static and dynamic properties, mechanical structures and material development (Case Texts 16.1, 16.2 and 16.3). There is a similar lack in relation to methods for applying computer technology in design, manufacturing and plant management. A third example relates to the need to preserve generic engineering R&D in selected fields where the domestic development of specific competitiveness seem promising. For example, generic engineering R&D in robot mechanics is perceived by a developer as essential for the incorporation of robots in application engineering, but no scientific support is thought to be available as yet (Case Text 17.2, interview with Mr Zhu Sendi). The last, but not the least important, area is the codification and communication of engineering knowledge developed in specific problem-solving. This

228

TECHNOLOGICAL TRAJECTORIES

requires scientific support for the language and models that are to be used in knowledge generation (Foray 1996). The relationship between industrial innovation, academic R&D and higher engineering education is a challenge which must be met. This issue has not been seriously addressed by the various reform policies thus far, except for a strong push to encourage university academics to market their research. It would be hard to exaggerate the critical importance of rebuilding a healthy scientific community in the engineering field, with a close interaction with the development of industrial technologies. Such a scientific basis has been critical for the successes achieved in the industrially developed world since the second half of the last century, whereas under the old system the Chinese scientific community was never really linked to industrial innovation. Summary Technological change, which we have considered as a process led by human’s deliberate problem-solving in selected areas rather than merely driven by the inherent logic of the technological realm, is strongly influenced by institutional structure. The institutional structure, which embodies the patterns of division of labour and the incentive structure in a certain society, largely shapes the active and passive areas of economic agents’ decision-making agenda for technology. The result is that technological change follows different paths in different institutional frameworks, being selectively sensitive to change in particular areas and insensitive to others. Different skills and experiences are accumulated in doing and using along different paths. These characteristics are indicated by the term ‘technological trajectory’ in this chapter. We have seen that, under the centrally planned system in China, a trajectory developed in the machinery industry which was selectively sensitive to general purpose technologies and selectively insensitive to specific technologies. Following such a path, the Chinese machinery industry achieved widespread expansion based on mature modern technologies but had little drive to develop specific technologies, whether specific to certain firms or to certain products. Specific learning was hampered under the top-down approach of the central planning, by disincentives to firms to improve efficiency and quality, and by banning horizontal communications and interactions that are essential to the development of specific technologies. As a result, the Chinese machinery industry as a whole, although it grew to be huge, had poor international competitiveness. Tremendous losses of economic efficiency resulted from this trajectory. Since the introduction of market reform the direction and characteristics of technological innovation in the machinery industry in China have undergone a fundamental change. There has been a shift in the patterns of the technological activities conducted in both large productive firms and existing R&D institutes. Specific learning is being enhanced to varying degrees in the context of market competition, and is supported by increasing horizontal information flows among developers and users of technology. Consequently, large enterprises are increasingly devoting themselves to the development of firm-specific technology which is sensitive to product quality and the efficiency of their production. R&D institutes have been transformed from government-run general technology suppliers, to become commercial suppliers of improved mature technologies for the large numbers of small and medium-sized enterprises. Some of the institutes are striving to become commercial developers of selected complex manufacturing systems which are highly specific with respect to both the product that is to be manufactured and the manufacturing conditions in which the systems are to be applied. Intensive technological learning has been entailed to restructure the industry’s technological capabilities during the process of shifting the technological trajectory. International technology transfer has been an important source for this learning, bringing the Chinese industry into contact with technologies developed in

MACHINERY TECHNOLOGY INSTITUTES

229

advanced market economies under the circumstances of international commercial competition. The discussion has recognized that the demonstration and instrument effects of international trading of technology were critical in accelerating the learning the industry needed for its transformation. Skills and experience accumulated in design and testing work have proved to be most useful in providing the ability to absorb the learning effect of internationally traded technology. Consequently, learning to shift the path of technological innovation during the economic transition has been largely a matter of combining accumulated strengths with new capabilities in areas previously neglected. This is termed ‘recombination’ learning. The previous institutional arrangements left the R&D institutes with an inheritance in design and testing skills and experiences which gave them an advantage compared to enterprises. This explains the important role of R&D institutes in the recombination learning required for path-shifting. As has been mentioned, these findings about the shift in the technological trajectory of the machinery industry should not be simply extended to any other industry. And the Chinese machinery industry still has a long way to go. Some of the challenges which it faces, such as an inadequate capability in generation and dissemination of engineering knowledge and the weak relationships between academics and industry, have been indicated. These are serious challenges which need to be overcome if the transformation of the industry is to be carried through.

19 INSTITUTIONAL RESTRUCTURING Internal and external contractual relationships

This chapter attempts to explain the institutional restructuring involved in the transformation of the R&D institutes in the Chinese machinery industry. Having observed the dramatic changes in the external and internal contractual relations of the group of R&D institutes as described in Chapters 15 and 16, it is necessary to ask whether there is some explanatory coherence underpinning the changes. Are these changes simply a mass of individually specific events? The chapter begins by outlining some basic assumptions developed in transaction cost theory, which are then applied first in relation to changes in the external contractual relations of the sample institutes and then to their internal organization, which can be considered as a set of internal contractual relations. The chapter ends with some remarks elaborating the argument for the importance of getting the institutions right. Basic assumptions of the transaction cost theory The transaction cost theory is part of the new institutional economics research tradition, representing the line of research which accentuates the concept of the firm (economic organization) as a governance structure for contractual relations. We will draw here mainly on Oliver Williamson’s widely accepted work in this field,1 and particularly his 1975 and 1985 books. These provide a sufficient grounding in the theory for present purposes. Classical micro-economics sees the firm (economic organization) as a production function, assumes that pure market transactions are cost-free, and has therefore been inclined to ignore or deprecate non-market modes of organizing transactions. The new institutional economics, in contrast, begins with transactions rather than production as the basic unit of analysis, recognizes transaction costs as the equivalent of ‘friction’ in exchanges among economic agents, and treats the various forms of economic organization as ways of reducing this friction. This leads to an analysis focusing on ‘the changing character of economic organization over time within and between markets and hierarchies’ (Williamson 1985:16). Different forms of economic organization are linked to what Williamson calls the ‘governance’ of contractual relationships (Williamson 1985:15–18; 1993; 1994), and the ‘efficiency purpose’ of rational economic agents to economize on transaction costs (Williamson 1985:16–17) is regarded as the drive that induces adaptive responses and initiates changes in economic organization. Transaction cost theory in this form is an explanatory instrument applicable to institutional development under a capitalist economy in which contractual relationships are the dominant manner of social arrangement. This is evident from the title of Williamson’s 1985 book, The Economic Institutions of Capitalism. Yet the theory does apply explicitly to quasi-market and non-market modes of organization within a capitalist economy, since the framework extends transaction mechanisms to embrace both markets

INSTITUTIONAL RESTRUCTURING

231

and hierarchies. It should also be applicable to the machinery technology R&D institutes in our sample, which have come to function in the marketplace as a result of the market reform. As we will see, the direction and characteristics of the institutional restructuring of these institutes are explicable in terms of transaction cost analysis. The following paragraphs will review some of the basic assumptions of the theory and explain their implications in connection with the distinctive aspects of our sample institutes, especially as regards their assets which, because of the highly specialized assignments which these institutes had in the past, are highly specific. Asset specificity and incomplete contracts Asset specificity is a key concept in transaction cost theory (Williamson 1985: 52–54). Asset specificity increases as a result of special purpose investments, that is, investments which, once made, cannot be redeployed for another use. Any disruption in contracting relationships which are based on specific assets can seriously reduce the productive value of the assets. Contractual safeguards are therefore required to protect transactions based on specific assets, and institutional transformations follow, to support effective contractual governance mechanisms. The concept of asset specificity was developed to analyse the incomplete contracts which are observed where asset specificity creates a need for governance mechanisms to safeguard transactions. A non-specific asset, on the other hand, is from a general purpose investment which can be redeployed easily and without a significant loss of productive value if an intended transaction fails. Transactions relating to such assets can rely on pure market institutions, such as spot contracting and anonymous business deals, in which the relationship between the parties is of the ‘sharp in and sharp out’ kind. These are known as complete contracts. The asset specificity of the R&D institutes in the sample is in general very high, as a result of specific purpose investments made for the institutes over a long period in order to carry out their assigned tasks in the development of product or process technology for particular parts of the machinery industry. Of the various types of specific assets defined by Williamson (1985:55, 95–96), the assets of these institutes fall largely in the categories of scientific manpower and physical installations. Experienced scientific and technological manpower is a unique and irreplaceable asset, and one whose value may be significantly reduced if it must be applied outside the field in which it has specialized, or if the individuals are lost to the institution. Physical installations are also highly specific to the technological specialization of a particular institute, as well as specifically requiring technological manpower of a certain profession to exploit their value. For our purposes, we need to supplement the concept of asset specificity with a recognition that the specificity of the goods or services which a firm produces also has implications for the risks it runs and thus for its need to protect contracting continuity. Goods and services are highly specific if they have been developed for a specific user or to be used under specific conditions and their value will be seriously reduced if contracting for that good or service is disrupted. The customized engineering services observed in Chapter 17 are good examples. Williamson and the tradition of the transaction cost approach do not explicitly distinguish between asset specificity and good or service specificity. Williamson does notice something similar in his 1985 book (pages 52–56) where, although ‘specificity’ is formally defined only in terms of asset specificity, he cites other authors in such a way as to link the production of ‘unique or imperfectly standardized goods’ to asset specificity. Later he says that ‘the pervasive organizational ramifications of assets specificity become

232

MACHINERY TECHNOLOGY INSTITUTES

evident only in the context of incomplete contracting’ (Williamson 1988:70), implying that it is incomplete contracting, rather than asset specificity, which has a direct effect on organizational ramifications. The focus on asset specificity in the transaction cost approach may be because most of the cases which were first studied in this tradition related to company mergers, where issues involving company assets were the main concern. Although specific goods and services have to be produced using specific assets, the reverse is not always true. Specific assets may be used for the production of general purpose goods and services, which may have the effect of reducing asset specificity in contractual terms. Nevertheless it is the attributes of the good or service that immediately affect the nature of the transactions, and therefore it is the specificity of the good or service that directly creates the need to protect the specificity of the asset. The distinction between asset specificity and good or service specificity is therefore an analytic convenience, especially for our study. It makes it possible to relate the shift to producing more general goods and services to changes in internal structure and external contracting, even where the productive assets involved have not changed substantially. A shift to producing more general goods and services using largely unchanged assets is a feature of the adaptive strategies of the R&D institutes of the Chinese machinery industry in response to the radical change in their external contractual environment. Having identified the relationship between asset specificity and good or service specificity, we will use the term specificity to cover both aspects, unless otherwise indicated. Uncertainty and the frequency of transactions Two other factors—the uncertainty and the frequency of transactions—are included in the analytical framework of the transaction cost theory (Williamson 1985:56–61), because these factors influence the choice of transaction governance structures, which ‘differ in their capacity to respond effectively to disturbances’ (Williamson 1985:56). Neither of these factors has any effect where the transaction does not involve any specificity. Uncertainty refers to unclear or unpredictable aspects of a transaction contract. For Williamson, this uncertainty ‘is attributable to opportunism’ and he refers to it as behavioural uncertainty (Williamson 1985: 58). If it is defined only in these terms, uncertainty is scarcely to be considered as an independent factor, since Williamson has already included behavioural opportunism as an assumption in his model (see below). But the technological contracts of the sample institutes involve other kinds of uncertainty, since most are based on innovative activities and involve some degree of technological novelty.2 This novelty leads to uncertainties regarding both the technical viability and the market prospects of the technology. By extending the concept of transaction uncertainty to include technological and marketing uncertainty, in addition to uncertainties deriving from opportunist behaviour, the model is strengthened, particularly in dealing with the R&D institutes’ transactions for engineering services. Frequency refers to the repetitive occurrence of transactions for a specific good or service which is produced using a specialized investment. ‘For large transactions of a recurring kind…the cost of a specialized governance structure will be easier to recover’, whereas ‘where frequency is low but the needs for nuanced governance are great, the possibility of aggregating the demands of similar but independent transactions is suggested’ (Williamson 1985:60–61). Transaction frequency plays a role in shaping governance structure: the higher the transaction frequency of a specific good or service, the more likely it is that a specialized governance structure can be maintained. The lower the transaction frequency, the more likely it is that there will be some institutional evolution to aggregate or integrate the specific transaction so as to spread the governance cost into internal management and over more general transactions. It will be

INSTITUTIONAL RESTRUCTURING

233

observed that the transaction frequency corresponds to the factor ‘size of technology market’ which was employed in the survey framework (see Chapter 15). We will return to examining the effects of novelty and transaction frequency on the transaction governance structure below. In short, asset specificity is assumed in the transaction cost theory to be the most important factor, with uncertainty and frequency as subsidiary factors, and these together are said to have a critical influence on the structures which are chosen for the governance of incomplete contracts, and thus on institutional forms and transformations. In view of the subject of our study, we will use a model with three major dimensions: specificity (especially the specificity of goods and services), technological and market uncertainty and the frequency of transactions. For the R&D institutes studied, it is the specificity of the goods and services they produce and the uncertainty regarding technological and market outcomes that led to incomplete contracts once the institutes were required to work within a market context, while the frequency of transactions is a factor in choosing between specialized or more general governance structures. These three analytical dimensions correspond to the factors specified in Chapter 15 and used in ordering the observations described throughout Chapters 16 and 17. Bounded rationality and behavioural opportunism The transaction cost theory makes two assumptions about human behaviour regarding contracts. On the one hand, people and economic agents (as contractual man) are thought to seek to be rational, although their rationality is only limited (Williamson 1985:44–47). This is in accordance with what Herbert Simon has called ‘bounded rationality’. On the other hand, they are assumed to be inclined to seek ‘self-interest’ when this is possible. Williamson calls this ‘opportunism’ (1985:47–50).3 Because of the opportunistic behaviour of contractors, an incomplete contracting relationship needs safeguards to protect it from hazards; and because of limited knowledge and rationality, contracting safeguards and institutional forms can be developed only in a gradual process in which economic agents adapt in response to contracting hazards while seeking to economize on governance and production costs (Williamson 1985:48–19). This leads to one of the key notions of the theory, which Williamson calls ‘fundamental transformation’, meaning that economic organizations are subject to a continuing process of adaptive transformation, corresponding to what we have called ‘institutional learning’, and that the real process of transformation matters (Williamson 1985:61–63; Williamson 1988). The transaction cost approach focuses on micro-level institutional analysis, and asserts the primary importance of issues relating to ‘getting the institutions right’.4 Recently, economic reforms in China have led to radical changes in the institutional environment,5 so that economic organizations, to maintain themselves, must change the governance structures of their external contractual relationships and their internal work organization. These adaptations and transformations in response to the radical environmental changes are the themes of this chapter. The transaction cost approach provides a vehicle for examining the particular and actual (de facto) processes involved in institutional transformations, as distinct from the official or judicial (de jure) factors and processes (Williamson 1994). Its analytic power will be tested in the following sections. Transformations of external transaction relationships This section examines changes in the external transaction relationships of the sample R&D institutes in the Chinese machinery industry. A simple contractual schema used by Williamson is adopted to depict the implications of the basic assumptions of the transaction cost theory. This will provide a basis for breaking

234

MACHINERY TECHNOLOGY INSTITUTES

Figure 19.1 Simple contractual schema Source: Replicated from Williamson 1985:33.

down transaction relationships into three different conditions around which governance mechanisms for external contracting are structured. The transformations in the institutes’ external transaction relations are then analysed by comparing the (transaction) governance costs, investment costs and ‘bureaucratic’ costs at early and later stages of the current reform. The transaction governance structure Figure 19.1 shows Williamson’s ‘simple contractual schema’6 in which three contractual conditions are depicted by nodes A, B, and C. Contractual conditions are determined simultaneously by the pointers k and s, the first referring to the hazards that are associated with a contract, the second with the safeguards which govern it. In Figure 19.1, node A represents the contractual conditions of a transaction for a general purpose good or service. There are no hazards associated with such a transaction, because the contract for it can be complete, so no safeguards are required (k and s are both equal to 0). Node A therefore approximates to trade relations under pure market intermediation, in contrast to nodes B and C where the transactions are for a special purpose good or service. In the case of node B, although there are some risks, there are no safeguards to protect any nontrivial transaction-specific investments used in the specific production (i.e. k>0, s=0). Node C represents the contractual conditions for a special purpose good or service where, unlike B, there are safeguards to protect transaction-specific assets and contain the risk of loss of productive value (i.e. k and s are both greater than 0). This contractual schema for external transactions can be used to describe the possible shifts in the transaction governance structure and the associated institutional transformation, by linking the specificity of the good or service, the specificity of the productive assets, and the governance structure which safeguards transaction-specific assets. A change in the institutional environment because of the introduction of radical reform programmes, as experienced by the sample institutes, can be expressed in this schema as an abrupt shift in the contractual conditions from one node to another, which we will see in the following section. Similarly, adaptations by the sample institutes to relieve the high risks and governance costs posed by the

INSTITUTIONAL RESTRUCTURING

235

abrupt shifts can be described in this schema as movements between nodes. These shifts will be discussed in the following sections. Institutional restructuring is largely the product of these adaptations. Initial shift in external contractual relations: node A to node B Before the current market reform, these institutes operated at a point comparable to node A, in that their transactions entailed no risk of a loss for the institutes. In 1985, with the implementation of the ‘Decision on Scientific and Technological Management System Reform’, government control and funding were withdrawn and the institutes were thrown onto the ‘technology market’ where they were supposed to survive through selling goods and services with risk (see Part 1: Chapter 3). The element of risk meant that, even where there was no change in the type of goods and services or in an institute’s customers, in the new institutional environment the transactions had been shifted from node A to node B. The pre-reform institutional environment merits a few words here, since it relates to the characteristics of the immediate post-reform institutional environment. Under the pre-reform economic system, industrial technology R&D institutes were specialized by professional field, either for a certain type of product technology or for a certain area of manufacturing technology, and their major users, the key productive enterprises, also specialized by product (see Chapter 14). The highly rigid specialization in the institutional arrangement for industrial technology meant that much of the institutes’ assets were highly specific. While these specific assets were not subject to transaction hazards (k=0), transaction safeguards were unnecessary (s=0). There was no alternative supplier of a particular technology, or at least none which was accessible to the institutes’ clients, and there was also no need seriously to calculate any loss in the value of specific assets, which could in any case not have a significant impact on the institutes, which operated under soft budget constraints. In other words, while investment in the institutes’ assets under central planning coordination was highly specific in technological terms, it was non-specific in transaction terms.7 Once the R&D institutes and their trading partners were exposed to market transactions, these technologically specific assets also became specific in transaction terms (k>0). Any loss in the value of their specific assets was now of real significance to the institutes. Despite the significant hazards to their specific assets, when the reforms began there were as yet no safeguards to protect these assets (s=0). Safeguard mechanisms do not automatically come into being along with the radical changes in their institutional environment, since these are mechanisms that have to be developed out of interactions between the institutes and the market structure they encounter. They were impelled to move to the node B condition (k>0, s=0), although the changes were very painful and the node B condition is very unstable, with a wide mismatch between significant contractual hazards and inadequate safeguards. The instability of the contractual condition at node B was recognized by the reform programmes and to some degree compensated for through the establishment of third party mechanisms, particularly the creation and improvement of the technology market institutions (Part 1: Chapter 3). However there was widespread disappointment among the industrial technology R&D institutes with regard to transactions on the technology market (Part 1: Case Texts 3.1, 3.2, 3.3). The instability of the node B transactions experienced was largely caused by the specificity of the institutes’ goods and services. It is hardly possible to write complete contracts for such transactions, and laws and courts and technology market institutions alone do not have adequate power to safeguard them (Williamson 1985:9, 163), although legal systems do still need to be developed much further in China. Even worse, at the beginning of the technology market, the technologies sold by the sample institutes in particular and by all the industrial technology R&D institutes in general were not only ‘specific’ but often simply idiosyncratic,8 a result of the extremely rigid and direct

236

MACHINERY TECHNOLOGY INSTITUTES

central coordination of technological development. These were user-unfriendly technologies with low tradability in the marketplace. Early adaptive transformations: reducing specificity Faced with serious contracting hazards which threatened the value of their specific assets, the institutes evolved three adaptive strategies in the years following 1985. All three aimed at reducing the specific or idiosyncratic character of the good or service produced using their specific assets, but proceeding along different paths. The first strategy was to move to providing complete engineering service packages, in fact an approach of ‘lateral’ vertical integration (Williamson 1985:103) within the core area of engineering know-how. Most of the group I institutes, i.e. those which were previously entirely centrally commissioned and financed (see Table 14.12), whether specialized in product or in manufacturing technology, adopted this strategy. They added complementary elements to reduce the idiosyncratic character which had previously characterized their engineering services (see Chapters 16 and 17). That led to a general pattern for the group I institutes, during the second half of the 1980s, in which they remained more or less in the same field as before (see Tables 16.2, 16.3, 16.4, 16.5, 17.1 and 17.2) by offering more technologically complete and therefore more tradable services. The specificity of these engineering services, which continue to be their main product, is still rather high so that, in the absence of other safeguards, their transactions are still at node B (k>0, and therefore s>0 is required). The second strategy is vertical or forward integration, in which an institute expands from its engineering functions in design and testing into the immediately following stage, manufacturing. The relationship between engineering design and manufacturing is so close that Williamson calls manufacturing the ‘site’ for the core technology. Most of the group III institutes, (those previously affiliated to local governments, see Table 14.12), have relied principally on this strategy (see Case Texts 16.7 and 16.8). The idiosyncratic nature of their engineering designs was thus reduced by internalizing transactions between the design and manufacturing functions. External transactions then deal mainly with finished machinery products, which are considerably less specific. The third strategy is similar to the second, in that the goal is forward integration with manufacturing, but realized in this case through an organizational merger. The 1987 policy was aimed at encouraging such mergers (see Part 1: Chapter 4), but in that period they were achieved only in the group II institutes (Chapter 16), that is, those which were previously centrally commissioned but were already located in a host enterprise both geographically and organizationally (see Table 14.12). The second and third strategies produced external transaction governance mechanisms which approximated to the node A condition. No safeguards were then needed since the specificity of their assets has been absorbed by the changed organizational boundaries. Table 19.1 summarizes the three early adaptive strategies and their consequent governance mechanisms. Note that the institutes of groups II and III Table 19.1 Early adaptive strategies and consequent transaction governance conditions Strategy 1 2

Completing engineering package Forward integration with manufacturing

Institute group

Consequent governance

group I

node B

group III

node A

INSTITUTIONAL RESTRUCTURING

Strategy 3

Being merged into a productive enterprise

Institute group

Consequent governance

group II

node A

237

Note: group I institutes: institutes that were previously entirely centrally commissioned and financed; group II institutes: institutes that were previously partly centrally commissioned but located in a host enterprise; group III institutes: institutes that were previously affiliated to local governments.

were transformed in such a way that the governance of their external transactions shifted from node B to the relatively stable node A, but the group I institutes remained at node B. This entailed further transformations for the group I institutes, which we will discuss soon. These early-stage adaptive strategies, all of which were moving within the ‘on-site’ areas of a core technology, are peculiar. They have not received much attention in the transaction cost literature because in developed market economies there is no need to address the integration between activities which interact with one another as directly as those revealed in our sample.9 Given the peculiarity of the transition situation, learning by doing seems to have been the only way of selecting solutions. Indeed the time following the 1985 Reform Decision was an important period when the institutes, their users and their principals became more adaptive to dynamic changes by learning how the market works. But why do the adaptive strategies of these groups of institutes differ? An aggregate cost-efficiency analysis can explain the differences, beginning by defining three kinds of relevant costs as follows:10 1 The bureaucratic (managerial) cost of implementing a strategy, which is the cost that the relevant bureaucracy (central ministry, local government, or host enterprise) incurs in overcoming organizational barriers, for instance in initiating and supervising a merger. The bureaucratic cost, ß, is low when a strategy does not involve external negotiations in relation to the organizational change. ß is low for strategies 1 and 2, whereas strategy 3 (merger with a productive enterprise) may entail high bureaucratic costs except in the case of the group II institutes which were already located within a productive enterprise. 2 The investment in capital goods required to implement a strategy. The capital investment, I, may well be lower for strategies 3 and 1 than for strategy 2 (forward integration). The more complex and precise the technology which is to be integrated with manufacturing, the higher the investment in capital goods which will be required. Thus the group I institutes, which specialized in more technologically sophisticated fields than the other groups, might face much higher capital investment costs if they were to attempt to integrate their R&D and design with manufacturing. 3 The cost of governance for market transactions, M, which is the cost of safeguarding the contractual relations involved in operating a particular strategy. Because transaction risks are operationalized as a cost, M is higher for node B transactions than for those at node A. A greater frequency of transactions reduces M, and in our sample the group I institutes were generally able to sell their specific technology to more users than the group III institutes, since the group I institutes were previously assigned to provide R&D and design services to a number of producers in a certain subsector of the industry, whereas the group III institutes were not. Table 19.2 relates the M–I–ß costs to the adaptive strategy chosen by each of the three institute groups. Suppose that aggregate M–I–ß cost-efficiency is the major criterion for strategic choices in institute restructuring and that various strategic choices are open to the decision-making centres of the institutes in study. Then group I institutes, as the table shows, would in general choose strategy 1 which tends to entail

238

MACHINERY TECHNOLOGY INSTITUTES

lower aggregate M–I–ß costs, or ‘transformation costs’. Strategies 2 and 3 would be secondary, because strategy 2 could require high capital investment (I) in manufacturing, which was not the task they were previously assigned, and the complexity of the technology in which these institutes are involved could be an additional barrier to adopting strategy 2 at a time when the institutes had not accumulated much wealth and had little opportunity to raise external funds. Strategy 3 could incur very high, or prohibitive, bureaucratic costs ß because higher institutional barriers would have to be overcome. It was only possible where the institute’s supervising agency, which in this case is the central Ministry of the Machinery Industry, initiated a merger and pushed it through. Group II institutes obviously tended to choose strategy 3, by merging with their host enterprises. Group III institutes would be more inclined to choose strategy 2, which for them would entail relatively moderate investments because the technologies which they possessed were simpler and smaller. While there are exceptions to the general best strategy for each group, this analysis of the M–I–ß costefficiency of various transformation strategies does explain why certain groups have, as a whole, chosen particular strategies. Group I institutes: further transformations from node B Strategy 1, moving to provide complete engineering packages, leads to a still unstable condition, from the transaction cost point of view, in the absence of Table 19.2 M, I and ß costs and choices of adaptive strategy, by institute group Group I institutes Strategy

M governance cost

1 H 2 L 3 L Group II institutes Strategy M governance cost 1 M 2 L 3 L Group III institutes Strategy M governance cost 1 M 2 L L 3

I capital investment

ß bureaucratic cost

outcome of strategic choice

L VH L

L L VH

strategy 1

I capital investment L M L

ß bureaucratic cost L L L

outcome of strategic choice strategy 3

I capital investment L H L

ß bureaucratic cost L L M

outcome of strategic choice strategy 2

Note: group I institutes: institutes that were previously entirely centrally commissioned and financed; group II institutes: institutes that were previously partly centrally commissioned but located in a host enterprise; group III institutes: institutes that were previously affiliated to local governments; strategy 1: completing engineering service packages; strategy 2: forward integration with manufacturing; strategy 3: being merged into a productive enterprise; L: low cost;

INSTITUTIONAL RESTRUCTURING

239

Group I institutes Strategy

M governance cost M: medium cost; H: high cost; VH: very high cost.

I capital investment

ß bureaucratic cost

outcome of strategic choice

transaction safeguards (k>0, s=0). Although the group I institutes largely achieved the ability to provide more complete engineering service packages during their initial transformation in the late 1980s, so that the idiosyncratic character of their technology was much reduced, significant further transformations affecting the external transaction governance system have been necessary and continued up to the mid-1990s when the field survey of the study was being conducted. Five strategies can be identified. They are described below and summarized in Table 19.3. Strategy A, ‘private ordering’, involves the development of extensive contractual relations as a governance mechanism by building up a mutual and credible commitment between the contractual parties. Where this succeeds, the contractual condition moves to node C in Figure 19.1. This governance mechanism is in contrast to the pure legal approach in which courts have a primary administrative role in contract disputes. The transaction cost theory acknowledges the key role of private ordering in protecting long-term contracts (Williamson 1985:162–169; 192–195). Typical cases of this strategy which were observed are the Beijing Research Institute of Electrical and Mechanical Technology and the Beijing Automation Research Institute for the Machinery Industry, both relating to the commercial development of complex manufacturing systems (Case Texts 17.1 and 17.2). In both cases the development was supported by a form of sophisticated bilateral contracting which, in the Chinese official statistics, is categorized as ‘technology development’ (see Table 15.2). One feature of private ordering is the frequent refinement and adjustment of the contracts as a way of dealing with the high uncertainty involved in technological development. Continuous communications between the parties, based on their mutual and credible commitment, sustain the information flows necessary to complete the contracts over time. Our observations showed that the user’s information was a constant input over the course of long-term contracts for the customization of forging systems (Case Text 17.1) and manufacturing automation systems (Case Text 17.2). Sometimes, especially in the early stages of the development, users provided their facilities for on-the-spot testing of the systems. In exchange, the developers tended to offer flexible and favourable contract prices. Obviously intense learning, to manage complex contracts, has been involved. The governance cost, M, becomes acceptable only when the contractors have learned a great deal of marketing skills. Strategy B, forward integration with manufacturing, is a strategy which was taken in the earlier stage of transformation, applying mainly to the group III institutes. It was not practicable for group I institutes because the complexity and sophistication of the technologies with which these institutes work requires high capital investment. As time passed, and the group I institutes improved their investment ability, the investment cost became less formidable. Forward integration with manufacturing moves an institute’s external transactions to node A by internalizing the specific contracts. The Dalian Modular Machine Tool Research Institute offers an example roughly conforming to this strategy (Case Text 16.3). Under strategy C, ‘enforced merging’, an institute is merged into a large productive enterprise so as to internalize transactions for specific goods or services. The contractual conditions are then moved, approximately, from node B to node A. This is identical to strategy 3 of the earlier stage, but at that time it was hardly applicable for group I because of the very high bureaucratic (managerial) cost, ß, they would

240

MACHINERY TECHNOLOGY INSTITUTES

face to overcome formidable institutional barriers. This cost remained very high during the 1990s, but in a number of cases the central Ministry of the Machinery Industry, with the agreement of the relevant provincial or municipal administrations, was involved in the decision to merge, in order to save the valuable specific assets accumulated in the institutes. Because of the reliance on high-level administrative coordination in the absence of a developed system of property rights and courts, this strategy is called ‘merging enforced by administrative coordination’ or ‘enforced merging’. In practice after the first such merger among the group I institutes, which was between the Automobile Technology Research Institute and the No. 1 Automobile Factory (see Part 1: Case Text 4.1) and was completed as early as 1980, there were no further mergers during the 1980s. Enforced merging was revived in the 1990s and by 1995 several group I institutes had been merged in this way, including the Xi’an Research Institute for Heavy Machinery, the Harbin Power Equipment Research Institute, the Luoyang Tractor Technology Research Institute and the Beijing Copying Machine Technology Research Institute (interviews at the Ministry of the Machinery Industry, 1994–1995). The merger of the Xi’an Heavy Machinery Research Institute with the Shanghai Baoshan Iron and Steel Corporation involved negotiations between the Ministry of the Machinery Industry, which is the ‘principal’ of the institute, (i.e. the representative of the state ownership), the Ministry of the Metallurgy Industry which is the principal of the Corporation, and the municipal administrations of Shanghai and Xi’an where the Institute and the Corporation are located (with the distance between the two cities being more than one thousand kilometres) (interviews at the Ministry of the Machinery Industry 1995). Negotiations among them took a number of years, and show how high the bureaucratic cost was for just one of the several cases of merging. Strategies D and E are strategies of continuing specific transactions, as in strategy A, but spreading the governance cost of specific transactions in different ways, by raising the frequency of the transactions. Strategy D is to shift the engineering services offered to a field in which there are more users, as can be seen in the case of the Shanghai Power Equipment Research Institute (Case Text 16.4). Strategy E is to continue to work in the same field so far as possible, but with some refinement and expansion. The Shanghai Electric Cable Research Institute and the Shanghai Internal Combustion Engine Research Institute (Case Texts 16.5 and 16.6) illustrate this strategy. These two strategies, ‘field shifting’ and ‘field preserving’ respectively, move the transaction governance conditions to somewhere between nodes B and C because, while safeguards are developed, these are for shorter-term and recurrent relationships, as compared to the ‘private ordering’ mechanisms (strategy A) used for long-term contracts. Table 19.3 Adaptive strategies for further transformation, group I institutes Strategy

Sensitive variable Case and short note

Consequent governance

A: private ordering

M

node C

B: forward integration

I

C: enforced merging

ß

D: field shifting

M (N, F)

commercial development of complex manufacturing systems (Case Texts 17.1 and 17.2) manufacture of advanced transfer lines and auxiliary machinery (Case Text 16.3) merging key institutes for automobiles, tractors, heavy machinery etc. into related key enterprises (Part 1: Case Text 4.1) engineering services shifted from large capacity power equipment technology to small power plant engineering (Case Text 16.4)

node A node A

node B-C

INSTITUTIONAL RESTRUCTURING

241

Strategy

Sensitive variable Case and short note

Consequent governance

E: field preserving

M (N, F)

node B-C

continuing to provide product and plant engineering services for electric cable and internal combustion engines (Case Texts 16.5 and 16.6)

Table 19.3 summarizes the discussion above. In terms of the M–I–ß framework, all these strategies can be seen as ways of lowering transaction hazards while optimalizing the cost-efficiency of the related transformations, which is measured approximately by the aggregate M, I and ß costs. The different strategies have their primary effects on one or other of these cost components, indicated by the ‘sensitive variable’ in the second column of Table 19.3, and they involve different governance institutions, indicated in the left column of the table. Strategies A, D and E are directed at coping with the governance cost M. They lead to an improvement in the pointer s of Figure 19.1, and move the contractual conditions toward C to differing extents. Strategies B and C, on the other hand, are aimed at containing the investment cost I and bureaucratic cost ß, respectively, and both shift the contractual conditions to node A. It is useful to note that in practice these strategies are often combined by a single institute or even in relation to a single contractual project. Private ordering has been particularly pervasive. Sustainability of technological specialization Why have different institutes adopted different strategies, and how does this relate to the fact that some institutes have been able to maintain their previous technological specialization in a time of radical economic transition and some have not? The sustainability of an institute’s technological specialization relates largely to the frequency of transactions, F, and the novelty, N, of the technologies which underlie the previous specializations and current transactions. An understanding of these factors is useful in thinking about policies for institutional transformation. Figure 19.2 displays the frequency and uncertainty of transactions as two dimensions characterizing the various positions of the group I institutes whose initial transformation had left them in a node B condition in the second half of the 1980s. The movements resulting from further transformations along the lines of the various strategies are presented with arrows in the figure. Whether or not an institute is able to retain its previous technological specialization depends first of all on the frequency of transactions for that technology, i.e., on the size of the technology market in that field. In Figure 19.2, the group I institutes which had a relatively high transaction frequency are represented on the shaded right-hand side. For these institutes, strategies A and E are basic choices, enabling them to approach a node C condition by controlling M. This requires a certain market size, but this is not a very tough prerequisite since many sectors of the machinery industry have a large number of small and medium-sized producers, especially in China because of the massive entry of rural enterprises over a long period. More than thirty of the sixty or so group I institutes could be said to have a reasonably high transaction frequency. Within the areas of high frequency, the degree of technological novelty influences the choice of strategy A or E. Projects entailing relatively newer and more complex technology tend to be dealt with by private ordering (strategy A), because long-term contracting under sophisticated governance mechanisms is more necessary. The low-frequency side of the initial B territory is the area of less stability. Institutes assigned to serving sectors with a concentrated firm structure fall inevitably in this area. Strategies B, C and D were taken largely in response to the instability. Strategy C, enforced merging, moves an institute away from the weak

242

MACHINERY TECHNOLOGY INSTITUTES

Figure 19.2 Relative stability of technological specialization

point B towards A, but is only applicable under strong administrative intervention. Where strategy C failed for whatever reasons, strategy D, shifting the field in which it offers engineering services, can be the next option. This is workable if the institute’s mastery of technology is great enough to enable it to transfer its engineering capabilities into a new but related field. The case of the Shanghai Power Equipment Research Institute (Case Text 16.4) is a good illustration. Finally, strategy B, ‘forward integration’ with manufacturing, has been taken particularly by institutes which were previously assigned to various kinds of machine tools. The determining factor here is probably more the radical change in the underlying technology than the firm structure of the sub-sector. In short, where the transaction frequency is low, institutes face greater difficulties in their further transformations. Strategy B entails high capital investment costs, I, strategy C will involve high external bureaucratic intervention costs, ß, and enormous efforts have to be made under strategy D to test out new market opportunities for the institute’s engineering capabilities. Institutes whose transaction conditions are located in the low frequency area should therefore be the priority, to be monitored and guided by reform policies. It is worth noting that the uncertainty of transactions, which is related to the novelty of the technology a certain institute contracts for in the technology market, has an influence. Technological novelty is determined not only by the supply side, but also by the users’ demands (see Case Texts 16.5 and 16.6). The less sophisticated engineering services provided by the institutes of cases 16.5 and 16.6 were due not to the institutes’ inferior capabilities but to the simpler and more conventional demands of the many small and medium-size users in these sectors. ‘Higher’ strategies are always thought to be preferred if possible: to engage in engineering services is preferred to entering manufacturing, and among engineering services, the more specific services such as complex product or production engineering are preferred. More specific

INSTITUTIONAL RESTRUCTURING

243

engineering services imply higher market values, and induce further specific investments. The institutes of Case Texts 17.1 and 17.2 experienced an increase in the specificity of their engineering services in the 1990s, reflected in the dominance of ‘technology development’ in their earnings structure, whereas ‘technology transfer’ and ‘technological consultancy and services’ had been the major sources of market earnings in the second half of the 1980s. Those institutes whose engineering services were restricted by inferior demands experienced a decline in asset specificity as Case Texts 16.5 and 16.6 show. They have recently taken action to create some more sophisticated demands internally and to make the necessary investments through a re-allocation of institute resources. As already discussed in Chapter 16, policies aimed at the creation of more sophisticated demands, such as tightening environment regulations and heightening safety standards, could help in this respect. Up to the mid-1990s, most of the group I institutes had demonstrated the ability to remain in their specialized engineering fields or in related fields, functioning either independently, or as in-house departments of larger entities. The machine tools institutes (metal-cutting technology) were still struggling. The importance of preserving and transforming previous investments in engineering capability during a radical economic transition is far-reaching, and we will turn to touch upon it in the final section. Transformations of internal organization In the light of transaction cost theory, the hierarchical structure of an organization can be seen as a governance mechanism for the internal contractual relationships between the centre and the divisions.11 The divisions of an organization are seen as operating on the basis of assets for which they have contracted with the centre, and the internal organizational structure relates to both these underlying internal contractual relationships and the economic agent’s external contractual relationships. The internal contractual relations may or may not entail safeguards, and the organizational structure may be more or less centralized or decentralized. These possibilities are shown in Figure 19.3, which is analogous to the schema for external contractual relations which was presented in Figure 19.1. Figure 19.3 distinguishes three basic modes of internal organizational structure, represented as U-form (Unified form), H-form (a Holding company-like form) and M-form (Multi-divisional form). In the first, the internal organization is a centralized structure and there are no safeguards imposed by the centre over corporate assets (s=0) because the divisions are not delegated the power to dispose of corporate assets. In an H-form structure, the internal organization is decentralized, with the centre having little control over the divisions which operate the corporate assets, i.e., s=0. In an M-form structure the internal organization is decentralized but effective strategic control by the centre has been developed to protect corporate assets from abuse by opportunist behaviour on the part of the divisions, i.e., s>0 (Williamson 1975:151–152; 1985: 284–285). Defined in this way, U-form, H-form and M-form structures are analogous to node A, node B and node C transaction conditions, respectively, but in the context of internal rather than external contractual relationships. The basic assumptions of transaction cost theory introduced at the beginning of this chapter remain relevant, but it is the implications of these assumptions for internal relationships which are considered here. In the internal context, asset specificity refers to the characteristics of the assets which are contracted by the centre to the divisions or departments, and bounded rationality refers to the behaviour of the centre and divisions within the corporate body, and so on. This schema will be used to interpret the transformation of the sample institutes’ internal organization in response to the development in external contracting relations, and to explain why these internal organizational transformations differed.

244

MACHINERY TECHNOLOGY INSTITUTES

Figure 19.3 Organizational choice Source: Replicated from Williamson 1985:285.

Initial shift in internal organization: U-form to H-form The internal organization of the institutes immediately after the beginning of the current reforms in 1985 generally approximated to the H-form. This is the initial condition in the transformations which concern us here, but this H-form organization had in turn developed from something equivalent to a U-form condition which prevailed in the pre-reform era. Some characteristics of this U-form organization should be mentioned, before describing the initial H-form and its further transformations. The U-form internal organization in Chinese R&D institutes and industrial enterprises was different in origin to the U-form organizations in Western market economies,12 but by describing it in the same terminology some similar features can be identified. In both the Chinese and capitalist versions of U-form organization, the centre of the organization has absolute command of organizational assets and reserves all rights to both strategic and operational decision-making. Unlike the U-form organization which were popular in capitalist economies of the 19th and early 20th centuries and originated from private ownership for single-product manufacturing, the Chinese U-form organization was not chosen by the enterprises or institutes themselves, but was required by the unified nation-wide state ownership of these organizations.13 There was no need to safeguard the specific assets that were actually used by divisions or departments. Asset specificity was highly significant for the R&D institutes, but it was significant in technological rather than transactional terms. Following the introduction of reforms, almost all of the sample institutes adopted an H-form organization, analogous to that of ‘holding companies’ in Western market economies.14 The H-form structure was chosen by the sample institutes and in fact all the industrial technology R&D institutes, despite its inherent weaknesses in imposing effective strategic control over the specific assets contracted to divisions or departments, as a means of responding to the abrupt shift in their external contractual relations, which moved from approximately node A to node B in 1985. The ‘internal contractual responsibility system’ was in practice the means for the shift. The adoption of such a system by an institute means that it delegates both decision-making autonomy and financial profitability and liability to the departments or teams of the institute so that they may initiate trade in the external markets using the contracted institute assets. And, as a

INSTITUTIONAL RESTRUCTURING

245

result of the adoption, the institutes transformed their internal organization from approximately U-form to Hform (see Part 1: Case Texts 3.2 and 6.1; Part 3: various cases in Chapters 16and 17). The shift from U-form to H-form structures can be explained from the strength of an H-form structure, which has the ability to provide ‘high-powered incentives’ (Williamson 1985:319, 403), in contrast to the Uform in which incentive loss and information distortion were serious (Williamson 1975:122–124, 129). The strategic advantage of high-powered incentives became even more indispensable in a period when planning coordination was to be dismantled but market opportunities were not yet known, which meant that the institute centres did not have adequate resources to steer strategic transformations for the institute on a whole institute basis.15 Hence, the ‘internal contractual responsibility system’ was actively promoted by both reform policies and institute centres, and welcomed widely by the departments and individual members of institutes during the second half of the 1980s. The resulting H-form structures carried the machinery technology R&D institutes through the radical transition from operating with planning coordination to relying on market transactions. Under the intervening H-form organization, institute assets were released from the control of the institute centre, which proved to be highly conducive to various adjustments and recombinations in response to external demands. A U-form organization could not perform this role, and the M-form (see below) could in most cases not be adopted, because the skills required to operate the complicated M-form internal organization were lacking and, more crucially, because there were as yet no well-tested market opportunities. The H-form, however, did have disadvantages: it reduced significantly the power of institute centres. Among the costs which this entailed was a serious loss of the ‘systems effect’ for each institute as a whole. It was conceivable that the organizational identity of an institute could be endangered under the H-form structure, and such costs and risks have been clearly perceived by institute centres up to the mid-1990s (see various cases in Chapters 16 and 17, especially Case Texts 16.5, and 16.8). There were also cases, although very rare, in which an institute came close to bankruptcy.16 Further transformations from the ‘initial Hform’ therefore cannot be avoided. An interesting question here is how the organizational identities were, in most cases, preserved. In a pure H-form institute, what prevents the institute identity from disappearing in the diverse and diverging interests of opportunistic departments and individuals? Presumably, among other factors, there was some degree of interdependency between the divisions of the H-form institutes and a minimum level of monitoring by the institute centre. It is necessary to note that the institutional transformations of key state-owned industrial enterprises seem to have followed a number of different paths. In most of the cases we examined, the U-form structure continued to operate during the 1980s and early 1990s. The evidence suggests that big state-owned enterprises are more likely to go from U-form structures to M-form, and that they make the transition when managerial efficiency becomes a key goal and the technology employed and the market niches developed become more complex and diverse. Although industrial enterprises are not the focus of the study, this comparison is useful in showing that there is no fixed mode of institutional transformation: what is needed is not a priori rules but close analysis of actual conditions and mechanisms that function in the real processes of the transformation.17 M-form internal organization and difficulties in M-form innovation Many of the sample institutes were about to embark on M-form innovation around the mid-1990s. The Mform structure is not the only choice for further internal transformation, and it may be useful to look at the difficulties it entails, for an institute proceeding from an H-form structure, before placing the institutes’ actual strategic choices along the U-H-M spectrum. As Figure 19.3 shows, the M-form or multidivisional

246

MACHINERY TECHNOLOGY INSTITUTES

internal corporate structure is characterized by 1) decentralized internal organization in which corporate assets are contracted to operational divisions, with high-powered incentives, and 2) effective control by the corporate centre of internally contracted assets. That is, there are safeguards (s>0) to protect corporate assets from abuse due to the self-interested behaviour of divisions. The M-form is therefore located in the internal organization schema at a point corresponding to node C of the external transaction schema (Figure 19.1). The M-form differs from the H-form in enabling effective central control by assigning longterm strategic decision-making to the centre and short-term operational decisions to corporate divisions, so that it has a superior ‘systems effect’ while retaining strong incentives.18 One of the difficulties in moving from an H-form to M-form structure appeared to be suspicion and rejection by the departments and individuals of an institute. They had enjoyed a great deal of autonomy in contracting institute assets, and in retaining significantly profits from the contracting for several years. M-form innovation implies that their autonomy is restricted in the long-term interest of the institute, but this may not be the short-term interest of every individual and department. Thus the intervening application of H-form structure created considerable internal barriers to further M-form innovation. These are barriers unique to transformations in the context of economic transition, compared with the move pioneered in the United States by Du Pont in the 1920s, from basically U-form to M-form structures. In that case the corporate divisions were granted stronger incentives than they had previously had, which tended to be welcomed internally. The other difficulty came from the poverty of policy thinking with respect to institutional restructuring. The ‘internal contractual responsibility system’ was for a long time the only written policy. Pointing as it did to H-form internal organizations, it could cope only partially with the challenges and problems confronted during the market reform. The instruction to adopt an ‘internal contractual responsibility system’ was of no use to institutes facing the need for M-form innovation. The techniques for the management of an M-form institute or enterprise should have been carefully introduced and studied, and this was not done. For instance, the Dalian Machine Tool Works reported (Case Text 16.2) that it lacked the corporate internal accounting technique needed to undertake M-form innovation. These difficulties explain why little was achieved in way of M-form innovation up to the mid-1990s, in spite of the pioneering success at the Automation Research Institute of the Metallurgical Industry (see Part 1: Case Text 6.1) which was evident as early as the late 1980s and was widely publicized. Nevertheless, the importance of further transformations has been widely accepted and is a priority for institute management, and further transformations are in fact under way. We can now turn to the current trends in these transformations, again using the U-H-M spectrum. Choices for further internal transformation Chapters 16 and 17 have shown that the sample R&D institutes have in fact evolved or are evolving diverse forms of internal organization. U-form, H-form and M-form are all possible, with M-form transformation being most commonly attempted by the larger centrally affiliated institutes although its implementation is being impeded by serious barriers. Table 19.4 correlates the major business of the sample institutes in the technology market with an approximate categorization of each institute’s internal organization up to 1994– 1995. Of the eight cases covered, three are attempting to move to an M-form structure, one is U-form and one H-form, two display a mixture of M- and U-form characteristics, and one, the Shanghai Power Equipment Research Institute, remains at the earlier stage which has been marked as ‘initial H’ (see the case texts in Chapters 16 and 17 for details of the internal work organization of these institutes).

INSTITUTIONAL RESTRUCTURING

247

The course of the internal restructuring of an institute from the initial H-form seems to be influenced by two characteristics of its market niches: the number of market niches for which the institute has to organize its internal work and the extent to which the technological activities conducted for these niches are interrelated.19 The niches for an institute’s major business are ‘divergent’ if there is little interrelation, ‘related’ if the major business involves systems whose parts are closely related to each other, and ‘concentrated’ if the major business consists of very few types or even a single type of output. An approximate U-form internal organization tends to appear when the number of market niches is low and the technological activities conducted for these niches are highly interrelated. An almost ideal case is presented by the Dalian Machinery and Electrical Research and Design Institute (Case Text 16.7). Conversely, where the number of market niches is high and the interrelation of the technological activities conducted for them is low, there is a tendency to retain an approximate H-form organization. An almost ideal case of this is the Zhejiang Mechanical and Electric Design and Research Institute (Case Text 16.8). As the latter case shows, the fact that the internal structure which emerges from the second stage of internal organization is again Hform does not mean the internal restructuring required for the transformation from the initial H-form has been any less radical. Finally, something approximating to an M-form structure tends to appear when both the number of niches and the interrelatedness between them are significant. Most of the group I institutes, except the Shanghai Power equipment Research Institute which is still in the ‘initial H- form’ stage, fall in this category. Figure 19.4 depicts the alternative forms of internal organization, using the U-H-M framework, in relation to the number of market niches and their interrelatedness. Both the number and interrelatedness of the niches are generalizations based on diverse details and are marked with relative indicators, ‘high’ and ‘low’, representing the ranges within a scale rather than end-points. Other factors, such as the size of an institute (see the staff numbers given in Table 19.4), the characteristics of the technology that underpins the market niches, the history and politics of an institute, and its relationship to the agency representing state ownership, all have a greater or lesser impact Table 19.4 Further transformations of internal organization, by institute Institute

Group Approximate mode of internal organization

Dalian Modular Machine Tool Research I Institute, 800 staff (Case 16.3)

M

Shanghai Power Equipment Research Institute, 800 staff (Case 16.4)

I

(initial H)

Shanghai Electric Cable Research Institute, 1,300 staff (Case 16.5)

I

M

Shanghai Internal Combustion Engine Research Institute, 850 staff (Case 16.6)

I

U-M

Major business in the technology market Transfer lines and auxiliary machinery associated with machinery engineering services (divergent niches) Small power plant engineering; inspection services for larger plants; trial production of auxiliary equipment (niche characteristics are not yet clear) Product and plant engineering for electric cables; special cables and cable materials (divergent niches) Product engineering for internal combustion engines (related niches)

248

MACHINERY TECHNOLOGY INSTITUTES

Figure 19.4 Alternative forms of internal organization: influence of the number of niches and their interrelatedness. Institute

Group Approximate mode of internal organization

Major business in the technology market

Beijing Research Institute for Mechanical and Electrical Technology, 900 staff (Case 17.1) Beijing Research Institute for Automation in the Machinery Industry, 1,300 staff (Case 17.2) Dalian Machinery and Electrical Research and Design Institute, 100 staff (Case 16.7) Zhejiang Mechanical and Electric Design and Research Institute, 600 staff (Case 16.8)

I

U-M

I

M

III

U

III

H

Integrated forging and heat treatment systems (related niches) Integrated manufacturing automation systems (divergent niches) Food storage and processing machinery and engineering (concentrated niches) Various machine products in the areas of small electric apparatus, apparatus for processing local agricultural output, and plant and testing engineering (divergent niches)

Note: Group I institutes are those previously entirely centrally commissioned and financed. Group III institutes are those previously affiliated to local governments. Group II institutes are not included in this table, because they have been incorporated into their host enterprises (see Table 14.12).

on the actual paths of internal restructuring, as can be seen from the various cases in Chapters 16 and 17. Moreover, there are no clear boundaries in practice between H-form, U-form and M-form structures. The two cases in Figure 19.4 which are marked with asterisks illustrate mixed forms. In these cases, the interrelatedness within the core businesses of the two institutes (forging systems in one case, and the complex product of internal combustion engines in the other) is high, and the core businesses have a rather centralized or U-form organization. Both of the institutes have now developed additional market niches and adopted decentralized or M-form elements, although the way in which this has been done differs in the two cases. The figure should therefore be read as a way of organizing observations and understanding information about particular cases, rather than a map of fixed paths along which internal restructuring must proceed. Nevertheless, this use of the U-H-M framework does convincingly explain why the internal organization of an institute must change in accordance with its external contracting structure, to optimize

INSTITUTIONAL RESTRUCTURING

249

the systems effect of an institute and thus the return on its specific assets. And its explanatory value in these cases should have indicated that the number of market niches and the interrelatedness of the technological activities required for these niches are major determinants of the basic direction of internal organization restructuring. Since market reform has exposed both the institutes and industrial enterprises to ongoing changes, their internal organization must likewise become an ongoing process of dynamic adaptation. Concluding remarks: getting the institutions right This chapter has applied transaction cost theory to the process of institutional restructuring with respect to the external contractual relations and internal work organization of the machinery technology R&D institutes in the period since the market reform began. Institutional restructuring has been essential to support the re-orientation of technological activities, described in the proceeding chapter as a shift of technological trajectories, that was required because of the radical changes in the macro-economic environment. It has been shown that enduring contractual relationships, for longer or shorter terms, are important external contractual mechanisms which support the technological learning necessary for an institute to keep moving along the new trajectories. Such enduring contractual relationships were a prerequisite for the development, by some specialized institutes, of the ability to produce commercially competitive engineering services. The most remarkable example of such engineering services is complex manufacturing systems, a product that the pre-reform system was never able to produce. The analysis has also shown that internal restructuring, which has produced a variety of structures approximating to the Uform, H-form and M-form, has been driven by, and aimed at, the development of external opportunities and contracting mechanisms, which has impelled a continuing reorganization of institutes assets through vertical integration, private ordering, organizational merging and so on. In a pattern which has become a familiar theme in the course of this chapter, and indeed of the book as a whole, institutional restructuring and technological learning, the two characteristics of the process of transforming a technological innovation system in the period of economic transition, have interacted and reinforced one another. However, Part 3 has shown that the transformation of existing industrial technology R&D institutions is much more complicated than the creation of new institutions, such as the NTEs which were considered in Part 2. Institutional experiments for the transformation of the existing institutes have covered the entire spectrum of external contractual conditions (the A, B and C nodes) and of internal structures (the U, H, and M forms), depicted in the transaction cost framework. The exercise sheds a light on the long-running debate about appropriate approaches to issues of economic development and economic transition. Where other economic perspectives have identified ‘getting the prices right’ or ‘getting the property rights right’ as the key, from the perspective of the new institutional economics the emphasis is on ‘getting the institutions right’. Coase (1992) says that The value of including…institutional factors in the corpus of mainstream economics is made clear by recent events in Eastern Europe. These ex-communist countries are advised to move to a market economy, and their leaders wish to do so, but without the appropriate institutions no market economy of any significance is possible. This study, although confined to one segment of China’s industrial technology innovation system, confirms that institutional analysis, especially micro-level institutional analysis, does effectively explain the cause and effect of economic reforms. Institutional analysis deserves to be considered as a useful analytical instrument for ‘getting the institutions right’. Williamson has emphasized that detailed, ‘positive’ ways of

250

MACHINERY TECHNOLOGY INSTITUTES

looking at institutions are the major source of the explanatory power of micro-level institutional analysis. In a recent paper, he expresses this as follows: The new institutional economics has recently been invited to speak to the issues of development and reform. As it turns out, the new institutional economics offers not one but several (related) perspectives. The main divide is between the institutional environmental approach, a macroanalytic perspective concerned with the political and legal rules of the game, and the institutions of governance, a more microanalytic perspective dealing with firm and market modes of contract and organization. Although many regard the first as the more pertinent for economic development and reform, I work predominantly from the governance perspective, adopting a bottom-up rather than topdown approach to economic organization. Three propositions inform the exercise: 1 Institutions are important, and they are susceptible to analysis; 2 The action resides in the details; 3 Positive analysis (with emphasis on private ordering and de facto organization) as against normative analysis (court ordering and de jure organization) is where the new institutional economics focuses attention. (Williamson 1995) Two findings which derive from the institutional analysis in this study, and are relevant to the ongoing economic transitions, deserve further consideration. Asset specificity and transition costs The analysis has shown that an economic agent, such as one of the industrial technology R&D institutes studied here, which has been freed from planning coordination has to pass through unusual and fragile intermediate stages of institutional structure, both externally and internally. The intermediate external position was indicated where transactions with outsiders had been shifted to the marketplace but there was no protection for the specific assets on which the transactions were based. And the intermediate internal position was required where institute assets were contracted to departments and small teams to test market opportunities. This was done in response to an abruptly changed institutional environment, at a time when the institute found it did not have effective mechanisms of internal management which were applicable in the new situation. In these contractual settings, the specific assets of an organization were subject to serious losses. According to the terms of transaction cost theory, these institutes were in an H-B crisis, a crisis that occurs where specific assets are the basis for external and internal contracts, but are operated with no safeguard mechanisms. The finding of the H-B crisis has implications relevant to the nature of transition costs, that is, the costs arising from the destructive effect on the strengths of an economy because of radical economic transition. Transition costs can arise especially from the loss of specific assets which have resulted from past investments in technology. Technological investment relates to the creation of ‘higher order’ factors of an economy (Porter 1990). Any significant loss of these factors may lead to a down-grading in the competitive potential which the economy can subsequently achieve. The study has shown that, in the context of planning-to-market economic reform, the shift in the rules of economic transactions by which transactionally non-specific assets become highly specific is responsible for the H-B crisis. This crisis is inescapable, and the higher the specificity of the technological accumulation, the greater the risk that the HB crisis will endanger the productive value of the accumulation during such transitions.

INSTITUTIONAL RESTRUCTURING

251

Transition costs can be considerably reduced if specific assets are better protected during radical economic transition. Two favourable elements have been found in the study which can be said to have reduced the losses of specific assets in China. First, the reform programmes were designed to be implemented through a gradual process (for instance, government funds for industrial R&D institutes were progressively reduced over five years). Time in which to learn is one of the critical resources for overcoming the H-B crisis. Second, decision-making responsibility was delegated to the locus where specific assets could be most effectively reorganized. The delegation of decision-making responsibility to the economic agents such as industrial technoloy R&D institutes operating at the bottom level (a de facto property rights delegation), while it has been partial and vague, has been roughly appropriate to encourage the leaders of the agents to take action to protect their specific assets and engage in the necessary learning. It is the transformations in the economic agents that determine whether or not new laws and corrected prices, once imposed de jure, do in fact function. This crucial part of restructuring — the institutional restructuring of economic agents themselves—seems often to have been overlooked in accordance with the advice to ‘get the prices right’ and ‘get the property rights right’. These two basic elements of the reform policy have certainly been helpful in saving more of the specific assets than was possible in Eastern European countries, where reforms have characteristically focused on getting de jure property rights right. Although the Chinese machinery industry is still in the midst of the institutional restructuring and technological learning, it has been growing rapidly. The exports of the industry reached $45 billion (US) in 1996, accounting for 29.8 per cent of China’s total exports in that year (People’s Daily, Overseas Edition, 9 and 11 January 1997). This was a marked improvement from 1990, when machinery exports were $11.1 billion (US) and accounted for 17.9 per cent of national exports, and from 1986 figures which were just $1.68 billion (US) and 6.13 per cent respectively. And according to a newspaper report, China has recently begun to export complex manufacturing systems.20 The need for M-form innovation and its special difficulties The institutional analysis has highlighted the need for further institutional restructuring, especially M-form innovation, and has revealed the conditions and difficulties of the innovation in detail. This is a finding relevant to China where gradualist reform programmes have reached a point at which M-form innovation becomes necessary but has been delayed. The delay has to some extent been a side-effect of the gradualism of the reforms, in that property rights are still vaguely defined, the construction of a market-oriented legal system has advanced slowly, and the motivation for the directors of R&D institutes and the chief managers of enterprises who have been the decisionmakers to preserve and reorganize specific assets is weak and irregular. Serious attempts to relax the barriers to further transformation began in China by the mid-1990s. M-form innovation, that is, the development of an internal structure with effective strategic coordination by the organizational centre over decentralized departments, is necessary after an institute has achieved a successful ‘initial H-form’. The rationale for the M-form innovation is simply to protect and further increase the value of specific assets by exploiting the systems effect of the institute as a whole. This is more relevant for some organizations than for others. The systems effect of M-form structure is significant for the larger and technologically more sophisticated R&D institutes, and presumably for enterprises with similar characteristics. It is less significant when an institute or an enterprise is small and serves fewer and simpler market niches. The special difficulties for M-form innovation are due to reluctance on the part of the departments and individuals who have been empowered under the initial H-form structure. Policy thinking has also shown only a limited appreciation of the importance of organizational structure and can give little guidance so that,

252

MACHINERY TECHNOLOGY INSTITUTES

even where a particular institute has itself perceived the need for M-form innovation, its efforts are not necessarily effective. These difficulties arise from the specific historical context in China, where there is no experience in the management of an M-form corporation and the management practices learnt since the market reform began have generally been tied to H-form structure (‘internal responsibility system’). It is useful to emphasize that the transaction cost theory recognizes that institutional transformation is largely an adaptive process decided by economic agents at micro-levels in response to market forces. Institutional transformation is hence seen as part of a market process in the broader context of a markethierarchy hybrid. This leads to the first implication of the findings from the institutional analysis, that direct policy intervention in the microprocess of institutional restructuring is in general of little use and suboptimal. Policies played a role in the China’s reforms first in re-aligning the economic environment, by introducing market-oriented reforms, and then in continuously tuning the factors affecting the economic environment, as was also illustrated in Part 1. But market processes alone are not always adequate for the micro-process of institutional restructuring, especially where there are limitations to the actions which individual institutes can take, as the low transaction frequency side of the initial B territory in Figure 19.2 shows. In such cases direct intervention may temporarily be necessary to prevent a significant loss of specific assets, which might have occurred in some of the institutes included in our sample (Case Text 16.4). To conclude, the series of studies which constitute the three parts of this book have identified technological learning and institutional restructuring as the essential attributes responsible for the performance of economic development in China during the market reform, and for the effective transformation of the innovation system which supports economic development. More generally, technological learning and institutional construction should be regarded as central for any successful economic development. Recently, institutional analysis, together with technological analysis, have come to be seen as the most useful approaches for highlighting the development processes of which we know still too little (e.g. Knight 1991). The several studies collected in this book have employed these approaches, and they have indeed proved informative. Our studies demonstrated that institutional analysis and technological analysis focus observation on very endogenous processes, hence they keep the researcher close to the real world and less preoccupied by the various kinds of classic or ideological obsessions which have unfortunately often blocked our thinking, with serious consequences especially in the field of reform and development studies.

NOTES

1 INTRODUCTION TO REFORM POLICY 1 For instance, the secretariat of UNIDO, in their 1979 evaluation of industrial research and service institutes in developing countries inquired: ‘Is an (independent, government-financed) Industrial Research and Service Institute (IRSI),…a reasonable option for developing countries which have not reached a relatively advanced stage of industrialization?’ (or equally, for those that have achieved some degree of industrialization?) ‘What can government and industries do to make more effective use of existing IRSIs?’ and ‘What priority role or function should an IRSI perform?’ (UNIDO 1979:34). 2 For the Soviet Union, see Amman and Cooper 1982, Chapter 10; for the German Democratic Republic, see Bentley 1992, Chapter 2; for an earlier comparative analysis of these countries, see Poznanski 1985. 3 In the Commonwealth of Independent States (CIS), ‘demands for R&D by industrial and agricultural enterprises dropped considerably,’ and ‘national and public financing to R&D was curtailed’ (Piskunov and Saitykov 1992). In Hungary, ‘the direct links between institutes and enterprises tended to favour lower quality R&D’, ‘In a quite not short period, there would not be a strong market demand for out-mural R&D by industrial enterprises as users’ (Balazs 1992:89–97). In the case of the former German Democratic Republic, it is reported that between December 1989 and July 1990, the number of employees in industrial R&D dropped by 23 per cent because these institutions were not able to sell themselves. Another estimate was that the reduction in industrial R&D manpower was as high as 50 per cent, in the same period (Bentley 1992:155). 4 See the papers cited in note 3 and, more recently, Balazs 1993. 5 Hungary had gone further than other former Eastern centrally planned economies in introducing elements of market rules before the end of the 1980s. From the perspective of ownership and control mechanisms for industrial enterprises, see for instance, Nagaoka 1989. 6 See: OECD 1992; 1993a; 1994a; 1994b. It is interesting to note that intensive discussions between outside experts and domestic decision-makers are an important aspect of the approaches used in these studies. 7 Precise analyses regarding the situation of the planned economy in China and of R&D institutions under it have been made by a number of authors. See, for example, OECD 1977, especially Chapter 2 (The planning system, by Audrey Donnithorne), Chapter 3 (Scientific institutions, by Richard Suttmeier), Chapter 4 (The institutionalization of science, by Richard Suttmeier and Genevieve Dean), Chapter 8 (Industrial structure and technology, by Hans Heymann), and Chapter 9 (Research and technological innovation in industry, by Genevieve Dean). 8 See, for instance, Suttmeier 1974. In Chapters 4 and 5 of the book, Suttmeier analyses the Chinese mobilization of the labour force to accelerate technological innovation in the period 1958–1960, and in the Cultural Revolution. Dean (1973:187–199) provides another interesting discussion, focusing on design reform in the period 1964–1966.

254

NOTES

9 Many works have described the process of market reform in China. Probably the best of the journalists’ reviews is that published in The Economist, which outlines the key steps and factors of Chinese reform, see the article titled ‘China, the Titan Stirs’, The Economist, 28 November 1992. 10 The central feature of the planned economy was that enterprises were to be a part of an administrative framework. This implies that enterprises operated under command instruction (largely through quantitative indicators) from the administrative authority, based on ownership by the state. China’s reform, in the mid1980s, started with delegating decision-making authority for routine operations to enterprise managers, at the same time retaining ownership by the state. Since about 1993, the ownership issue has been seriously addressed, along with tax and banking reforms. The approach chosen is to reorganize firms as corporations, with investors becoming shareholders. See: ‘Decision on Some Issues for the Establishment of Socialist Market Economy’ (made by the Central Committee of Chinese Communist Party), in the People’s Daily, Overseas Edition, 17 November 1993. Many reviews have reflected on the reform measures. See, among many others, the article ‘China, Birth of a New Economy’, Business Week, 31 January 1994. 11 In 1985, the number of products to be distributed directly by the state was reduced from 123 to about 60 (Yuan Baohua (ch.ed.) 1985:188). This was further decreased to 36 in 1993 (People’s Daily, Overseas Edition, 10 August 1993). As for the real extent to which state control acts to influence the operation of enterprises, surveys indicated that 87 per cent of raw materials were distributed by the state in 1984, which was reduced to 49 per cent in 1987, and 30 per cent in 1988; 71 per cent of outputs were distributed by the state in 1984, falling to 30 per cent in 1987 and 27 per cent in 1988. Sources: for the 1984 data, Zhang Shaojie 1987:195; for the 1987 and 1988 data, SSTC and NCSTD 1989:73. Note that these figures refer only to state-owned enterprises. Non-state enterprises have never been seriously controlled by the state plan. The rapid expansion of non-state enterprises has changed the balance of enterprise ownership to the extent that, by about 1990, the non-state sector was approximately equal to the state-owned sector in terms of industrial output. The majority of the non-state sector is rural industry, largely resulting from the rural reform just mentioned. 12 Up to mid-1993, 13,000 enterprises had foreign investors, with real foreign investment amounting to more than $US40 billion. The economic importance of these enterprises has been increasing, to the point that they accounted for 25 per cent of total exports in the first half of 1993. With respect to technology imports, in the period between 1979 and 1992 about 5,000 agreements were signed, embracing various forms of technology importation, mostly in turn-key projects or incorporated with key equipment procurement, for a total cost of $US34 billion (People’s Daily, Overseas Edition, 29 September 1993). As for the degree to which China’s economy is integrated with international markets, data indicates that in 1992, China’s exports amounted to $US85 billion, and its imports to $US81 billion (People’s Daily, Overseas Edition, 20 March 1993), so that merchandise trade as a ratio of gross national product (GNP), measured at the official exchange rate, had increased from 12.8 per cent in 1980 to 38 per cent in 1992 (Financial Times, 19 November 1993). This ratio was about 6 per cent in the 1960s and early 1970s. 13 ‘County’ is a local administrative unit with an average population of about 500,000. There are slightly more than 2,000 counties in China. At the county level there are roughly 3,000 more ‘science and technology’ related establishments. They are mostly charged with the dissemination of information, especially for agricultural technology. 14 The official science and technology statistics do not cover the design institutes, so it is difficult to obtain direct data. Many sources suggest that there were several thousand units with about 300,000 staff in 1980. The design institutions later expanded in step with the high levels of capital investment during most of the 1980s (see, for example, Lu Yanlin 1992). Note that the several thousand units mentioned can be divided into two classes: those specialized in ‘plant design’, which are usually ‘independent’ from enterprises, and those engaged in product and process design, which have gradually integrated with manufacturing firms or R&D institutes, although some of them have, at the same time, been guided by senior levels of the government administration. The restructuring of plant design institutes is not included in this study. 15 ‘Locking-in’ meant not only that their professional technological work was governed by the administrative power, but also that they had managerial and technological supporting duties, to assist the administrative body.

NOTES

255

These managerial and supporting duties included: compilation of product and technology standards, formulation of sectoral and sub-sectoral development projects, and testing and examination of product quality for firms in the sector. In addition, they were required to help to organize technological exchanges and other working meetings for industrial ministries. These tasks accounted for about a quarter or more of the overall activities of R&D institutes at the ministry level (Interview notes 7:4), and of those at the local levels too (Interviews in 1993 and 1994). 16 Some outside observers noted the fact that ‘R&D institutes in the field of engineering seem to do no R, only D— i.e. their scientific activities appear to be aimed not at research to discover new knowledge or novel solutions within their specialities, but at developing practical applications of existing knowledge.’ See: OECD 1977:148– 149. 17 Only aggregate data is available. The first national general survey was conducted in 1986 and aimed to acquire data on the situation at the end of 1985. This showed that, for the R&D institutes affiliated to central ministries and commissions (622 units), more than 50 per cent of their activities in terms of expenditure were not R&D and for the institutes affiliated to local administrations above the ‘county’ level (3,946 units), about 80 per cent of their activities were not R&D. These non-R&D activities may be categorized as ‘engineering and design’, ‘dissemination and consultant services’, and ‘production activities’ (defined by the State Science and Technology Commission of China), with the composition varying for the different groups (‘White Paper’, No. 1:238). Another study indicated that, of the 4,690 independent R&D institutes in 1985, only slightly more than 600 had more than 50 per cent of their activities in areas which, according to the definition in the Frascati Manual, fall within R&D. Nearly 3,000 of these institutes had almost no activity which could be classified as R&D (Xu Zenji et al. 1987). 18 For example, in the former German Democratic Republic (GDR), the proportion of research establishments’ activities falling outside Frascatti’s definition of R&D was reported to be between 20 and 50 per cent, depending on the sampling and the time of surveys (Bentley 1992:64, 142).

2 THE PRE-‘DECISION’1 PERIOD: REFORMATION OF PLANNING PRACTICE (1978–1985) 1 ‘Decision’ here is the ‘Decision on Reform of Science and Technology Management System’, in force since 1985, which was orientated mainly toward market solutions (see Chapter 3). 2 The great impact of the science and technology revolution was the main rationale for the first initiative for the establishment of R&D institutions in 1956. The late Premier Zhou Enlai, in the most important policy statement of that time, said: ‘Science is a decisive factor in the development of defence, economy, and culture.’ To support this statement, he drew at length on various advances in technologies, including those in mechanics, automation, aircraft, materials, electronics, and atomic energy, and concluded by repeating his Soviet counterpart Bulganin’s statement (to the Soviet party’s plenum in 1955) that ‘These recent achievements bring mankind to the brink of a new revolution of science, technology, and industry. Its impact will be more far-reaching than the industrial revolution which took place due to the emergence of the steam engine and electricity’ (Zhou Enlai 1956:181–182). 3 This is reflected in an important document drafted by the State Science and Technology Commission, entitled the ‘Report on Guidelines for the Development of Science and Technology of Our Country’. The report criticized the neglect of industrial production technology under the current R&D investment strategy. It argued that more attention should be given to the assimilation and dissemination of imported technologies, as well as to applications of domestic R&D outputs. It also argued that coordination between domestic R&D and technological importation should be improved (SSTC 1981). 4 In 1988, 4,732 contracts were signed for the implementation of key S&T projects. Contracts for ‘applied research’, ‘experimental development’, and ‘design and trial manufacturing’ accounted for 34 per cent, 27 per cent, and 29 per cent, respectively, of these. The rest (about 10 per cent) of the contracts were for ‘basic research’, ‘dissemination’, and ‘small batch production’ (SSB 1990a: 315).

256

NOTES

5 The industrial enterprises’ roles were 1) as the users of the resulting equipment or processing technologies; 2) as the operators of manufacturing trials for the developed product technology; and 3) in a few cases, as designers or co-designers of products under development. 6 The first use of some technologies was also planned, in this system: the outcomes from a particular S&T project would be included as inputs to particular ‘capital construction projects’ or ‘technological renovation projects’ listed in the economic plan.

3 THE DECISION ON S&T MANAGEMENT SYSTEM REFORM (1985) 1 As a result, China has compiled internationally comparable science and technology statistics since 1986, using the definitions provided in the Frascati Manual. 2 Four types of institute were distinguished for the implementation of the grant cuts: 1) technology development type—institutes engaged primarily in technology development; 2) basic research type—institutes primarily doing basic research or doing applied research which could not have any practical value in the short term; 3) public welfare and infrastructure type—institutes engaged in R&D related to the public welfare such as medicine and health, labour protection, family planning, calamity prevention and control or environmental sciences, along with institutes engaged in technology infrastructure activities such as standard-setting, taking measurements, testing, and providing information, and institutes engaged in agricultural science and technology; and 4) the multiactivity R&D institutes, with major activities in more than one of the three areas: basic research, applied research, and experimental development. The government grant was cut only for type 1 institutes, along with lesser cuts for type 4 institutes according to the proportion of their activities which focused on ‘experimental development’. About 2,000 institutes suffered government grant cuts (Interim Stipulations for the Management of Science and Technology Grants, State Council, 23 January 1986, in ‘White Paper’ No. 1:314–315). 3 In 1991, the first year after the completion of the cuts, the income structure of the industrial technology R&D institutes was: (a) income from government (including contractual research on government-financed projects) accounted for 22 per cent, (b) income earned in the market, 61 per cent, (c) bank loans, 12 per cent and (d) other, 4 per cent. Thus the ‘market’ source dominated for the group. For comparison, the proportions of the income of the institutions for agricultural R&D from the same sources were 55 per cent, 33 per cent, 5 per cent and 6 per cent, respectively. For the institutes of the Chinese Academy of Science, the figures were 68 per cent, 21 per cent, 1 per cent and 10 per cent. And for the R&D institutes engaged in meteorology, seismology, survey and mapping, measurement, and environment protection, they were 72 per cent, 27 per cent, 2 per cent and less than 1 per cent, respectively (calculated from the data in Databook of Statistics on Science and Technology 1992:70, 74, 77). Chapter 6 will provide further analysis relying on the 1993 data. 4 Earlier, in 1981, the Law of Economic Contracts had been issued and put into effect. Under this law ‘contracts for science and technology-related transactions’ were one category of economic contracts. In 1984 the Law of Patents came into force. 5 The predecessor of the Foundation was the Natural Science Foundation, initiated and operated by the Chinese Academy of Sciences from 1982. The operational procedure of this Foundation drew heavily on the practice of the National Science Foundation of the US. A number of smaller foundations with specialized objectives in particular fields have emerged since 1985. These act as complementary sources of finance for basic and fundamental applied research in special fields, or for particular purposes (Bulletin of the National Natural Science Foundation of China No. 1, 1987:20–34). The government appropriation for the National Foundation has steadily increased, from 100 million yuan in 1986 to 170 million in 1991 (China Statistical Yearbook on Science and Technology 1992:306). 6 It is reported that more than half of the best-selling products from the Shanghai suburban area (largely the surrounding rural areas) were produced under the guidance of ‘Sunday engineers’ (second job takers) from Shanghai city (SSTC and NCSTD 1989:21).

NOTES

257

7 The performance of the ‘quasi-market’ for talent may be reflected in the fact that, in 1991, there were about 150, 000 staff working in New Technology Enterprises around China, of which about half were ‘S&T professionals’ (China Statistical Yearbook on Science and Technology 1992:309). 8 Williamson calls the latter ‘transactions with bilateral governance’. In the Chinese literature many different terms are used for this form of transaction, such as joint business or coordinative management (lianying), combined undertaking (lianhe ti), and horizontal cooperation (hengxiang hezuo). These have sometimes been translated as ‘joint-ventures’ or ‘combinations’, which is not very accurate.

4 MERGING R&D INSTITUTES INTO EXISTING ENTERPRISES (1987) 1 In many centrally planned economies the average size of enterprises was much larger (which causes other problems). The enterprise structure in China developed with smaller units (State Council 1990: Part 1 Chapter 4, and Part 8 Chapter 4). This very much hampered improvements in productivity, especially in those sectors which are sensitive to economies of scale. More serious was the problem of specialization among enterprises (State Council 1990: Part 10 Chapter 2). According to the officers from the Ministry of the Machinery Industry, an oververtical integration, i.e. a specialization in terms of final products, rather than in terms of underpinning technology (‘vertically dis-integrated specialization’), was developed by the planned economy in China. This was in part due to the principle of maximizing output (rather than value-added), and partly to the segmented resource allocation system. Both the small size and poor specialization limited the wealth of enterprises, suggesting that making them more innovative and competitive in a market environment will not simply be a matter of adding an R&D element to them, or of delegating autonomy and imposing incentives on them (Interview notes 7:3–4; and other interviews). 2 The pattern of technology imports for the electronics industry has been widely reported. For instance, for consumer electronics (such as black and white and colour TVs, video-recorders, and cassette recorders) see Liu Ying (ch.ed.) 1987:188–189; for computers, see ibid.: 175; for electronic instruments, see ibid.: 213–214; for transistors and electronic components, see ibid.: 257. The development of this sector had long been focused on military purposes. The shift to civilian production since the late 1970s also resulted in a need to import production technologies. 3 The five criteria were: 1) 70 per cent of the institute R&D activity should be needed by, or have been committed to, the enterprise or group of enterprises with which a merger is contemplated; 2) the enterprise or group of enterprises must be wealthy enough to sustain the institute; 3) the two sides must be compatible with each other in their business portfolios; 4) the two sides should be geographically adjacent; and 5) the two sides should be willing to merge (Interview notes 7:3). 4 It is reported that the merging of a big tractor technology institute into a big tractor factory had been suspended. During the first round of negotiations, all the criteria defined were fulfilled, but the ministry officers and the institute managers were unwilling to merge. About 1991 the situation changed since 1) government funds were definitely diminishing, and 2) the market for contractual research was very limited with demand coming mostly from small firms, which was not attractive to the ‘key’ institute; 3) it had proved impossible for the institute to produce final products of tractors, since the investment required was too great and it was not possible to turn to producing tractor instrumentation, because of competition in a crowded market; 4) the factory had invested heavily in their own R&D and testing, implying that the institute would definitely lose its value as an influential power in tractor technology. Source: Research Centre for System Analysis, Research Institute of Mechanical Science and Technology (RIMST): A Study on the Direction and Paths of R&D Institutes in Moving towards Self-Reliance (in Chinese) mimeo, Dec. 1992.

5 SPIN-OFF ENTERPRISES AND THE TORCH PROGRAMME (1988)

258

NOTES

1 For details of the policies see ‘White Paper’ No. 3:245–250. Mr Song Jian’s speech at the 1988 National Working Meeting of the Torch Programme explains the policies further, see ‘White Paper’ No. 3:415–419. 2 Ten areas of technology were defined: 1) micro-electronics and computer technology and products; 2) information technology and products; 3) new material technology and products; 4) new energy, energy conservation technology and products; 5) bio-technology and products; 6) space and ocean technologies and products; 7) laser technology and products; 8) products for the application of nuclear technology; 9) products with integrated mechanical and electronic technology; and 10) other new and high technology. 3 The NTEs enjoyed a tax exemption for the first three years, and then paid enterprise income tax of just 15 per cent, whereas state-owned enterprises at that time paid 55 per cent. Since 1 January 1994, the tax rate for enterprise income has been lowered to 33 per cent. People’s Daily, Overseas Edition, 18 December 1993. 4 For an example of the involvement of a Zone in loan appraisal, see Case Text 10.6 in Part 2. 5 This study, covering 178 NTEs in the Beijing Zone, showed that 86.5 per cent of their initial capital was invested by their organizational initiators. Expansion was primarily financed with bank credits and re-invested profits. The study also revealed that the local branches of many specialized banks (and banking agencies) contributed to financing the expansion of NTEs in the Beijing Zone. The biggest contributors at that time were the Industry and Commerce Bank of China, the Agriculture Bank of China, and the Foundation for the Promotion of Economic Development through Science and Technology, which was jointly sponsored by the former State Economy Commission and the Chinese Academy of Sciences. 6 In fact, the leading funds from the central government constituted only a very small part of the overall investment in Torch Programme Projects, ranging from about 0.5 per cent to 3 per cent between 1988 and 1991 (China Statistical Yearbook on Science and Technology 1992:308). On the other hand, bank loans accounted for 10 per cent (1988), 13 per cent (1989), 50 per cent (1990), and 70 per cent (1991) of the investment in ‘Torch Programme Projects’, indicating a dramatic increase between 1989 and 1990 (ibid.). The leading funds in fact acted more as a policy guideline than a source of finance. The predominant role of the banks was in fact intended under the Torch Programme. 7 For Beijing Zone, see Zhao Wenyan et al. 1989. For Wuhan Zone, see the People’s Daily, Overseas Edition, 3 May 1993. For Shanghai, see the People’s Daily, Overseas Edition, 1 November 1993, and for Shenyang, see the People’s Daily, Overseas Edition, 5 August 1993.

6 THE TRANSFORMATION OF ESTABLISHED R&D INSTITUTES (SINCE THE 1990s) 1 One commentator has said: By 1991, the cuts in government grants for R&D institutes of the ‘technological development type’ had been completed. The main question now became how to further reform this type of institute. Many institutes of this type are currently inclined to develop profitable production with intensive R&D as their main option (keji changye) when they plan their development in the eighth five year plan period (1991–1995). Examples can be seen in Jiangsu Province, and in the construction materials sector. Most R&D institutes of the ‘technological development type’ have their own testing workshop, pilot plant or trial production bases. Experience in the past few years has shown that institutes that fully exploited the potentials of these plants and facilities have attained higher levels of income, more adequate to sustain their normal operations. Moreover, in these cases, all the research, development, production, and re-training of staff were better managed in the long term. (See: Li Fang 1992) 2 See: Chinese Science News, 6 December 1992. 3 Science and Technology Daily, 28 April 1993. 4 Science and Technology Daily, 12 December 1992; 30 April 1993. 5 Science and Technology News, 23 March 1993.

7 CONCLUDING REMARKS

NOTES

259

1 For the possible reduction of barriers to merging in the machinery industry, see Zhu Sendi 1994. 2 In addition to the facts cited in Chapter 1 above, more evidence may be found in the OECD report on Science, Technology and Innovation Policies in the Federation of Russia. It is said there that there were 7,000 small enterprises ‘involved in science and science services’, and that ‘some innovative “subsidiaries” are forming a “business sphere” near scientific centres’ in Russia. This is obviously restructuring of the spin-off type. See OECD 1994b:99–101. 3 The transfer of industrial R&D from government affiliated to non-profit organizations for the provision of technological services to small enterprises is also strongly suggested by an OECD report on the transformation of industrial R&D. See OECD 1993a.

8 INTRODUCTION TO SPIN-OFF ENTERPRISES 1 The market orientation of the Decision has been discussed by many researchers. See, among others, Conroy 1992: 91–96. The author states that ‘The overall trends of the S&T reform, driven by the need to optimise the output of S&T investment in production, have since their onset implied that technology is a commodity with an exchange value.’ (ibid. 91). The market orientation of the reform design stirred up vigorous debates in China about the value of technology, for which see Baark 1988. The author stated that policy makers at the highest level accepted the idea that the price of technological commodities should be regulated by ‘free’ market forces. 2 For example, in the Commonwealth of Independent States (CIS), ‘demands for R&D by industrial and agricultural enterprises dropped considerably,’ and ‘national and public financing to R&D was curtailed.’ (Piskunov and Saitykov 1992). In Hungary, ‘the direct links between institutes and enterprises tended to favour lower quality R&D’, ‘In a quite not short period “even in the longer term”, there would not be a strong market demand for out-mural R&D by industrial enterprises as users’ (Balazs 1992). In the case of the former German Democratic Republic it is reported that, between December 1989 and July 1990, the number of employees in industrial R&D dropped by 23 per cent, because the institutes were not successful in the market. Another estimate put the reduction in industrial R&D manpower as high as 50 per cent in the same period (Bentley 1992: 155). 3 In terms of total manpower for R&D, for instance, the figure for the US (in 1989) is 943,000, for the UK 276,000 (in 1988), and 899,000 for Japan (in 1990). Note that this data is only illustrative, since there is some incompatibility between the statistics for different countries. For the US the definition is ‘research scientists and engineers’, for both the UK and Japan, it is ‘total R&D personnel’ in all fields of science and with all levels of educational background (OECD 1993b:319, 306 and 191).

9 AN OVERVIEW: THE LAUNCH OF THE TORCH PROGRAMME AND THE DEVELOPMENT OF SPIN-OFF ENTERPRISES 1 See Wang Ke, Gu Shulin and others: Research Report No. 9003, National Centre for Science and Technology for Development, 1990, mimeo; also Suttmeier (1991). 2 Ten areas of technology were defined: 1) micro-electronics and computer technology and products; 2) information technology and products; 3) new material technology and products; 4) new energy, energy conservation technology and products; 5) bio-technology and products; 6) space and ocean technologies and products; 7) laser technology and products; 8) products for the application of nuclear technology; 9) products with integrated mechanical and electronic technology; and 10) other new and high technology. 3 The establishment of Special Economic Zones began in 1979, with the purpose of attracting foreign investment in general. The first four Zones of the kind were in Shenzhen, Zhuhai, Shangtou and Shamen (Yuan Baohua (ch.ed.) 1985:385–393). By the first half of the 1990s, this kind of area open to foreign investment has expanded

260

NOTES

to a large number of coastal and inland cities. The convergence between the two kinds of zones relied on the fact that the technology exploited in the Science Park is not particularly high or new, while in several Special Economic Zones gradual increase in technology-intensive projects were pursued in order to upgrade their industrial structure in the 1990s.

10 THE INITIATION OF THE NTEs 1 By mid-1994, Legend was producing 586 PCs for the domestic market. 2 Jinghai Computer Room Technology Development Company was founded in 1983 by eight engineers from the Institute of Computer Technology, CAS. They started their business with the support of a subordinate agency of the Haidian District Administration, Beijing. This agency is responsible for the coordination of local (district) enterprises and is known as ‘The United Collectives of the District’ (qu lian she) (Wang Xiaolong (ed.) 1993: 11). Stone Company, which manufactures Chinese computerized typesetting machines, was founded in 1984 by seven engineers from the Centre for Computerization of the CAS. The initiative was supported by the authority of Sijiqing Town, Haidian District, Beijing. The Town Authority provided 20,000 yuan and some physical space for the Company, and the Company was thereby designated as ‘collective’ in ownership, with the Town Administration being its initiator (ibid. 15). 3 There are no official statistics relevant to the form. The information comes from interviews. Usually the managers of Zones could not or do not differentiate form 3 from form 1. Thus the number of NTEs of this type may well be underestimated. 4 A survey conducted in 1989, sampling 178 NTEs in the Beijing Zone, indicated that 86.5 per cent of their initial capital was invested by the initiating organizations (Zhao Wenyan et al. 1989). 5 In Shenyang the manager estimated that about fifty out of the total of 1,100 NTEs were truly ‘private’, and slightly less than 60 per cent of the 488 ‘collectives’ were initiated without any direct involvement from an initiating institute. In Wuhan the manager estimated that individual initiations accounted for slightly less than 30 per cent of the total, and that most of these which were ‘collective, without [organizational] initiator and supervising unit’ were in fact individually initiated. In Hangzhou about 8 of the 161 enterprises were ‘private’ (about 5 per cent), and another forty-five NTEs (about 28 per cent) are ‘collectively owned’, but it was not possible to get further information on the character of this latter group. In Beijing the manager confirmed that 20 per cent were individually initiated in real terms, of which half were under the category of ‘collective ownership, with a local agency as initiator and supervising unit’. The Beijing Zone deliberately created the latter category to help individual initiators. (Interview notes 1:4; 9:7; 11:3; 12:2, 3). 6 M-form (multi-divisional form) and U-form (unitary form) were originally used to describe firm structures under capitalist economies (Williamson: 1985). Qian and Xu apply them to highlight differences between Chinese central planning (M-form) on the one hand, in which economic administration was conducted largely according to a decentralized territorial principle, with self-contained regional levels playing an important role, and on the other hand the central planning of the Soviet Union (U-form) in which the top-level authority exercised dominant control. 7 For the emergence of rural industrial firms, and the property rights which are applicable, see Perkins 1994. 8 For Beijing Zone, see Zhao Wenyan et al. 1989. For Wuhan Zone, People’s Daily, Overseas Edition, 3 May 1993. For Shanghai, People’s Daily, Overseas Edition, 1 November 1993, and for Shenyang, People’s Daily, Overseas Edition, 5 August 1993. 9 The leading funds from the central government constituted only a very small part of the overall investment in Torch Programme Projects, ranging from about 0.5 per cent to 3 per cent between 1988 and 1991 (China Statistical Yearbook on Science and Technology 1992:308). The leading funds acted more as a policy guideline than as a source of finance. 10 See Hu Qihen (Vice-President, Chinese Academy of Sciences): Report to the 1989 Working Meeting of the Chinese Academy of Sciences, 5 November 1989, mimeo, and Wu Jinglian 1991.

NOTES

261

11 THE TECHNOLOGICAL ACTIVITIES OF NTEs 1 Hope Corporation was founded by a group of postgraduates from the Institute of Automation of the CAS. It produced computer software used for architectural design, as well as some telecommunications and information equipment. 2 For reports on the massive import of PCs in 1984–1985, see, for example, China Business Review, March-April 1988:30–33. 3 There is no systematic data on the initial monetary investment of NTEs. However, one survey made in 1988 reported that the total start-up capital of 13 NTEs with three-year histories had been 16 m. yuan, i.e. about 1 m. each (Yu Weidong 1988:103), equivalent to a few hundred thousand US dollars at the official exchange rate of the time. A 1990 estimate for about 400 business entities sponsored by the CAS and its subordinate institutes indicated a similar level of investment, i.e. between half a million yuan and one million yuan. But the initial investment in intellectual assets is very high, since the intensive involvement of scholar-businessmen should also be accounted for.

14 THE DEVELOPMENT OF THE MACHINERY INDUSTRY IN CHINA PRIOR TO MARKET REFORM 1 ISIC division 38 is subdivided into five major groups: fabricated metal products (ISIC 381), non-electrical machinery (ISIC 382), electrical machinery (ISIC 383), transport equipment (ISIC 384), and professional instruments (ISIC 385). Since 1988, Chinese economic statistics have been based on internationally compatible definitions. ISIC 38 (excluding ‘electronic and telecommunication equipment’ which are dealt with separately under the title of the ‘electronics industry’) thus corresponds roughly to the ‘machinery industry’ in this study. In some cases, which will be noted, the data is for the machinery industry excluding fabricated metal products (381) and the consumer durable products of ISIC 383. ‘Electronic and telecommunication equipment’ covers five subsectors. In order of output value these are 1) telecommunication and forecasting equipment (45 per cent), 2) electronic components (15–20 per cent), 3) electronic devices (10–15 per cent), 4) computers (5–10 per cent), and 5) radar (5 per cent) (see China Electronics Industry Yearbook 1993:II–1) 2 The term ‘centrally planned regime’ is used in a neutral sense, to refer to an institutional arrangement which prevailed in China for more than thirty years and which defined the ‘initial condition’ from which the current reform starts. We will not address the reasons for choosing such an institutional arrangement at that time: it is enough for our purposes to recognize its existence and seek to understand the structure and operational mechanisms of its institutions. 3 In Chinese statistics, ‘industry’ includes mining. 4 The data here includes the ‘electronic and telecommunication equipment’ industry. The importance of this industry can be gauged from the fact that its output accounted for 2 per cent of total industrial output in 1986, and 2.5 per cent in 1992. 5 Here mining is excluded and electronic and telecommunication equipment is included in manufacturing. 6 This point has been further explored elsewhere in the context of industrial structure and (technological) learning efficiency (Gu 1996b), where it is argued that the capital goods sector is highly significant for industrialization, because this sector plays a central role not only in mastering increasingly complicated means of production, but also in imposing learning intensity and sophistication on the various agents involved through user-producer interactions. It should be noted in this connection that possessing a certain capacity in the capital goods industry alone is not sufficient to attain learning dynamism in an economy. ‘Getting the institutions right’, which is the goal of the current Chinese reforms, is critical to support the effective accumulation of technological capability. 7 The elasticity of energy consumption (=annual growth rate of energy consumption/annual growth rate of national income) averaged 1.24 during 1953–1978 (SSTC 1985a:144). This indicator has fallen to an average of 0.61 in the decade 1982–1992 (China Statistical Yearbook 1993:492), an improvement which was inconceivable for

262

NOTES

8

9

10 11

12

13

14

15 16

many Chinese energy experts in the early 1980s. Apart from the improvement in the energy consumption levels of machinery products, other factors, including structural adjustments, have certainly contributed to the improvement in energy elasticity. But the energy efficiency of the Chinese economy is still low in absolute terms. The labour productivity in the sample countries around 1980 can easily be calculated from the data in Table 14.1. Gross output per worker in the Republic of Korea, Brazil, and India was higher than in China, by factors of 4.6, 4. 4, and 1.6 respectively. China’s labour productivity was some 4,000 US dollars per worker. But a number of factors reduce the validity of the comparison. The number of workers in China, for instance, includes not only production workers but also those engaged in non-production services because of the different way in which Chinese enterprises are organized. This leads to an under-estimation of the labour productivity for China. Nevertheless, it is evident that the productivity of the Chinese machinery industry is lower than in comparable countries. This study was conducted by a few researchers from the Institute of Quantitative and Technical Economics of the Chinese Academy of Social Sciences, with Jorgenson and Kuroda. The study examined Chinese economic growth from 1953 to 1990, and compared it with American and Japanese growth in the same period. See Li Jingwen et al. 1993. Szirmai and Ruoen devote substantial attention to adjusting the employment data for China, reducing enterprise employment data by 9.8 per cent to allow for the manpower engaged in non-productive services. It should be noted that factors such as the increasing price of intermediate inputs and labour may have counterbalanced improvements in comparative productivity which have in fact occurred in this period. Nevertheless, it is safe to conclude that a very wide productivity gap did exist and has not yet been significantly reduced, although in absolute terms productivity is improving. Assuming that a catchingup process by a latecomer is intended to narrow the gap between it and the leading industrial manufacturer, the comparative productivity situation is crucial for the quality of development. Narrowing the gap is the primary characteristic of successful industrialization, as in Japan and South Korea. The same study indicated that Japan had improved its comparative productivity from less than 60 per cent of the US level in 1970 to about 90 per cent by the late 1980s, and the Republic of Korea improved productivity from about 14 per cent of the US level in 1970 to almost 30 per cent by the late 1980s. The dynamic picture, if it is accurate, implies that the gap between China and her successful Asian neighbours has probably been widening, even since the reforms. The domestic supply ratio of 50 to 60 per cent is estimated in the yearbook, based on the estimation that one US dollar of imported machinery products is equivalent to about 12 Chinese yuan of domestic products (China Machinery Industry Yearbook 1994:1–22). The source does not give any further details to explain this equivalency ratio. These figures are given in SPC et al. 1992:84. Machinery technology here would appear to be defined as ISIC 381, 382, 384 and 385, plus 383 (electrical machinery) excluding electronic and communication equipment, computers, and household electronic equipment. It is reported that ‘In November of 1952, the Commission of Finance and Economy, Administrative Council of the Central People’s Government, advised that a centralized and unitary design organ should be established. All available designing manpower should be mustered in design units at various levels.’ (Jing Xiaocun (ch.ed.) 1990: Vol. B, 87). As an indication of the situation prior to 1953, it is reported that previously offices from some government planning departments were assigned to assist Soviet experts in their engineering survey and designing. The Soviet experts complained that this should be the job of design institutes and engineers, rather than administrative personnel (Yuan Baohua (ch.ed.) 1985:127). The State Capital Construction Commission was reorganized several times, and was also known as the Ministry of Capital Construction. Chinese agricultural machinery embraces a large number of products: 1) tractors, 2) internal combustion engines, 3) parts and accessories for tractors and internal combustion engines, 4) tractor-drawn farm machines, 5) harvesters, 6) draining and irrigation machines, 7) machines for crop protection (sprayers), 8) machines for

NOTES

17

18

19 20

21 22

23 24

263

forestry, husbandry and fisheries, 9) machines for the transport of agricultural products (Jing Xiaocun (ch.ed.) 1988; Cheng 1972:139). The General Institute dealt with heavy machinery, mining machines, petroleum and chemical equipment. The sectoral institutes were specialized as follows: No. 1 Institute: engineering machinery; No. 2 Institute: general purpose machinery and power generation equipment; Institute Nos 3, 4 and 5: agricultural machinery; No. 6 Institute: machine tools, forging machinery, tools, grinding materials; No. 7 Institute: electric transmission equipment, electric transformers, electric materials and devices; No. 8 Institute: electric motors and printing machinery; No. 9 Institute: automobiles; No. 10 Institute: bearings; No. 11 Institute: instruments. No serious efforts to institutionalize plant engineering have been made in most developing ‘market’ economies. One exception is South Korea, where the government encouraged local engineering services, but this was not effective (Prof. Linsu Kim, personal conversation, Maastricht, 1994). In the Korean case, plant engineering was first provided by foreign designers, and then drew on firms’ internal expertise, relying on the fact that large ‘cheabols’ have been dominant in Korean development. For the former Soviet Union, see Amann and Cooper 1982:20, 453; for India see Rahman and Sharma (eds) 1974, section II. In 1955 and 1956, the Minister of the Machinery Industry, Mr Huang Jing said that ‘the machinery industry can meet demands for independent development only if it is capable of independent design, while independent design is to be grounded in scientific experimentation and research’ (Jing Xiaocun (ch.ed.) 1990: Vol. B, 242). For an introduction to China’s national S&T programme, see ‘White Paper’ No. 1:114–131. ‘Simple re-production’ is a term used for economic management in the planned regime in China, coming from Karl Marx’s concept with same phrase, implying the maintaining of existing capacity of production. In contrast is the ‘expanded re-production’, generated by capital investment. While categorized as for ‘simple re-production’, the related funds (called as technical measures fee and new product trial manufacturing fee) were largely allocated by the planning administration, out of the site where the ‘simple re-production’ should be maintained. For details of the allocation of the ‘simple reproduction’ funds, see Jing Xiaocun (ch. ed.) 1990: Vol. B, 227– 228. The current reform formally started in 1985, but the machinery industry was selected as an ‘experimental industry’ where the testing of reform policies was to begin one year before the formal start. ‘Organizing technological works for the industry’ involved the following duties, which occupied perhaps a quarter to a third of the manpower of R&D institutes (Jing Xiaocun (ch.ed.) 1990: Vol. B. 292, and interviews):

• • • • • • • •

formulation of technology policy;Z formulation of plans for scientific and technological development; compilation of product specification series and technological standards; coordination of important R&D projects; approval of new products and appraisal of product quality; promotion of technological exchange; dissemination of technological information; and provision of other technological services.

25 In the Chinese administrative hierarchy, provincial governments are superior to the municipal governments. In 1985 there were 29 provincial administrations (excluding Taiwan), and 162 municipal administrations (SSB 1990:1, 5). 26 The first use of some technologies was also planned, in this system: the outcomes from a particular S&T project would be included as inputs to particular ‘capital construction projects’ or ‘technological renovation projects’ listed in the economic plan. 27 It is reported that the machinery industry itself suffered from being ‘overwhelmingly equipped with generalpurpose machine tools and universal equipment, with too little precision equipment and special processing equipment, testing instruments and other complementary equipment’. ‘The supply of equipment used for

264

NOTES

adopting new manufacturing techniques was for a long time unsatisfactory’ (Jing Xiaocun (ch.ed.) 1990: Vol. B, 269).

15 GENERAL TRENDS IN THE TRANSFORMATION OF THE MACHINERY INDUSTRY 1 The estimates for these groups should be read with caution, notably in relation to the electronics industry. It could be that the greater weight of engineering services in the horizontal income of the centrally affiliated institutes of the machinery industry is largely due to ‘technological convergence’, whereby a small number of mechanical technologies are widely applicable in a large number of metal-using industries. One would like to confirm this by a comparison with the newly-emerged electronics and telecommunication industry, whose technologies also have the property of being widely applicable to many other industrial operations. Unfortunately we cannot provide comparable data for the electronics industry since its data is combined with the data for the machinery industry, and it is not practicable to separate it out since the proportion contributed by the electronics industry is too small to be confidently identified from different incompatible sources. But for the same reason, the combination of data for the electronics and machinery industries cannot distort the basic trends too seriously. Despite the inadequacies of the data, the estimates given here provide reasonably strong support to the conclusion that Chinese industrial R&D institutions are still rather traditional. 2 The sixty-four centrally affiliated R&D institutes for machinery technology employed about 20,000 R&D scientists and engineers, out of a total of about 140,000 R&D scientists employed in all government-run institutes for all industries (Databook of Statistics on Science and Technology 1993:11, 13). 3 There is a remarkable trend break in 1990–1991, when the emphasis shifted from technological consultancy contracts to technological development. This may be due to the longer time taken to negotiate a new contract for item (1). It might also be due to an economic environment unfavourable to capital investments in 1990, or to contracts which were initially negotiated for consultancy services being renegotiated for technological development.

16 THE TRANSFORMATION OF THE ‘PRODUCT TECHNOLOGY’ R&D INSTITUTES 1 In the official Chinese categorization of research institutes by sectors, the machine tool institutes fall under the machine tools and tools sector, both the turbine and electrical cable and wire institutes are in the electrical equipment sector, and the internal combustion institutes fall under the agricultural machinery sector. 2 In 1995, the Beijing No. 1 Machine Tool Plant won a prize from the American Machinist Association for its achievement in the implementation of CIMS. See People’s Daily, Overseas Edition, 30 November 1995. 3 We are indebted in this paragraph to unpublished data for individual centrally affiliated institutes provided by Mrs Zhang Meiguang, of the Research Institute of System Analysis of the Ministry of the Machinery Industry. 4 This comment comes from Mr He Wenli, a manager who has long been involved in the machine tools sector, in an interview at the Ministry of the Machinery Industry, Beijing, October 1994. Chinese managers are now well aware of the backwardness of the sector, repeatedly mentioning that their productivity in producing CNC machine tools was about one tenth of the productivity of producers in Taiwan. 5 See Part 1, the section ‘Merging R&D Institutes into Existing Enterprises’ and Part 1, Case Text 4.1, which describes this first merger. 6 According to the income data for individual institutes (Zhang Meiguang 1995), the income structure of the Lanzhou Institute is also dominated by earnings from engineering services. Other sources have reported that their user group of the Lanzhou Institute has moved away from large oil equipment producers because of the institutional separation and institutional disparity between the Lanzhou Institute and its previous users (Interview with Mrs Dong Lijung).

NOTES

265

7 The boundary between technology transfer and technological consultancy and services is not clear-cut (Zhu Sendi, personal communication, December 1995). Sometimes statistics are also influenced by taxation policy. For instance, institute income from ‘technological consultancy and technological service’ has been liable to tax since 1994, which may have caused such income to be counted as technology transfer income, which is still tax free. 8 Statistics are too dispersed in various local administrations to collect, especially considering that this question is peripheral to the focus of the study. It is sufficient to make reference to some additional reports, such as the official publication ‘Selection of Typical Experiences in the Reform of the Science and Technology Management System in China (1985–1990)’ (SSTC 1991c). This contains five cases referring to locally affiliated machinery technology institutes, all of which confirm the transformation trends which have been observed for this group of institutes. These five institutes are: 1) Guangdong Agricultural Machinery Research Institute (of Guangdong Province), which is now moving towards specialization in the design and manufacture of greenhouses based on the assimilation of imported technology (ibid., 49–52); 2) Hunan Agricultural Machinery Research Institute (of Hunan Province), which now focuses on the design and manufacture (with small enterprises in various forms of joint ventures) of simpler, smaller, lighter, and cheaper machines for family-based agricultural production in response to the agricultural reforms (ibid. 52–55); 3) Shenyang Automatic Control Technology Research Institute (of Shenyang city), which is now engaged in the design and manufacture of some automatic control devices, in combination with application engineering of automation systems (ibid. 58–62); 4) Jilin Mechanical and Electrical Technology Research Institute (of Jilin Province), which is entering into the design and manufacture of some selected machines used for air drying, plastic processing, and dust extraction, in combination with plant engineering (ibid. 68–71); 5) Nanjing Research Institute for Food Packing and Processing Machines (of Nanjing city), which was originally assigned to agricultural machinery and like the Dalian Institute has been turning to focus on some lines of food packing and processing machinery since the market reform (ibid. 353–356). 9 For the institutional separation at the central level between machinery plant engineering and machinery product R&D, see the section on ‘plant design institutes and capital investment’ in Chapter 14.

17 THE TRANSFORMATION OF THE ‘MANUFACTURING TECHNOLOGY’ R&D INSTITUTES 1 No separate data was available for the sub-group of (central) ‘product technology’ R&D institutes. The comparison has therefore to be between the manufacturing technology R&D institutes and the entirely centrally affiliated R&D institutes as a whole. The former is a sub-group of the latter. 2 The description in this and the following paragraphs is based on unpublished data for individual institutes provided by Mrs Zhang Meiguang, Research Institute for System Analysis, of the Ministry of the Machinery Industry. 3 Under proper conditions, metal may attain a superplastic state in which it has lower flow resistance and a higher index of elongation. Superplasticity makes it possible to make many parts which have fine and complicated shapes with little deformation. Chinese scientists have been working in this field for twenty years, and Russians are also strong in this field (interview with Prof. Hai Jingtao; Introduction to the Department of Superplastic Technology, BRIMET). 4 The Institute identified two projects as being worthy of attention, of which only one is outlined here. The other was vacuum heat treatment technology (interview with Mr Wang Dechen; Introduction to the Department of Vacuum Heat Treatment Technology; Product Specification for Vacuum Furnaces). The Institute has captured the entire domestic market in industrial vacuum furnaces with an effective maximum dimension of 600×900×400 mm and temperatures of up to 1300° centigrade. The furnaces can be supplied in a standard design or customized to a user’s special requirements. But the vacuum furnace project is smaller and not very representative of the development of manufacturing systems examined in this chapter. 5 For a more general comment on the limited ability of the planning approach, as it was elaborated in China, to promote production technology, see Part 1: Chapter 2.

266

NOTES

18 TECHNOLOGICAL TRAJECTORIES 1 Patel and Pavitt apply bibliometric methods to US patents granted between 1963 and 1988, and group them into five technological fields: chemical, electrical, nonelectrical machinery, transport and ‘others’ (mainly relating to traditional manufacturing such as textiles and to non-manufacturing industries such as construction, medicine and agriculture). They find that, while the importance of ‘science-based’ (i.e. chemical and electrical) technology grew in the twenty-five years covered by the statistics, from 40.0 per cent to 48.3 per cent of total patenting in the US, this still leaves slightly more than half of total patenting activities in the 1980s outside the science-based sectors. Non-electrical machinery technology is the most important ‘non-science’ field, accounting for 38 per cent of the total patents granted in the 1980s. They conclude: ‘we are in danger of neglecting up to nearly 40 per cent of what is going on in technical change, especially in non-electric machinery and in small firms…the persistence of improvements in mechanical technology…suggests that new ‘paradigms’ do not destroy old ones, but complement and extend them…non-electrical machinery remains useful, indeed essential, even after organic synthesis, radio waves and the electric chip…’ (Patel and Pavitt 1994). 2 ‘Market economy’ and ‘market regime’ are used here to refer in general to economies in which cost/benefit performance is the most critical determinant of the health of productive enterprises. The term ‘centrally planned regime’, is used in a corresponding non-ideological sense. Since the aim of this analysis is to show how institutional arrangements have affected the characteristics of technological change, our interest in both cases is in the institutional arrangements and not in the political or economic justifications of this situation. Thus the analysis has implications not only for centrally planned economies, but to some extent also for economies with a similar institutional character but which are not known as ‘centrally planned economies’. 3 Von Hippel and Tyre’s study (1995) also implies that the use of immensely increased scientific knowledge and analytic methods to articulate complex technology does not intrinsically reduce the critical importance of doing and using in technological change, because of their special problem-identification function, and because of the very complexity and dynamism of human technological activity. An interesting suggestion is that conscious awareness of the mechanism of problem-identification by doing and using can itself be used to accelerate the process of problem solving. One example is the process of ‘rapid phototyping’ in computer software development, a method designed to shuttle communications repeatedly between developer and user so as to optimize a software package for the user’s real needs. 4 The unit principle has been identified by researchers as a new trend in technological innovation. Foray describes it as an approach based on technology reusability and modularity: ‘reusability and divisibility will become the most valuable properties of a product in that they facilitate recombination’ (Foray 1995). The trend reflected in Case Text 17.1 in relation to forging systems is one example. In the computer software industry, ‘elements of the development process have become more routine. Sub-routines, macros and operating systems have been used to avoid reinventing the wheel…software is written from the beginning with the intention of making it more reusable.’ (ibid.). This trend may be observed in chemical engineering in the concept of ‘unit operations’, in which any chemical process, on whatever scale, may be resolved into a coordinated series of ‘unit actions’. ‘It was a methodology that could also provide the basis for the systemic, quantitative instruction of future practitioners’ (Rosenberg and Nelson 1994). 5 The grouping is primarily for analytic convenience, since R&D institutes are the particular focus of this study, but technology imports have in practice been organized separately for the two groups. 6 These sources are Jing Xiaocun (ch.ed.) 1990 and Zhou Jian’nan (ch.ed.) 1990. 7 For instance, engineers at the No. 1 Automobile Factory recognized problems with the inlet valve location of automobile engines soon after the imported technology was put into production, but this was not addressed in the decision-making agenda until about ten years later (personal conversation with the late chief engineer of the Factory, Mr Mong Shaonong 1986). 8 The decision-making agenda of an organization is determined by the nature and purpose of the organization, which leads to the ‘activation’ of some areas of the agenda, while other areas are monitored and others remain

NOTES

9

10

11

12

13

14 15

267

passive. An individual member’s recognition of a problem may not be transmitted to the agenda of the organization if the information is not considered significant (Arrow 1974). There is one specialized material technology research institute in the ‘centrally affiliated’ group, the Shanghai Material Research Institute, founded in 1956 (see Case Text 14.1). Throughout the 1960s and 1970s this Institute took the leading role in the development and dissemination of material and material testing technologies rather effectively. Almost all of the ‘product technology’ institutes of the centrally affiliated group included a department of material development for related product, worked reasonably well (interview with Prof. Fang Xiangwei at the Shanghai Material Research Institute, May 1995). Aluminium cable, for example, came from the material development department at the Shanghai Research Institute for Electrical Cables. Similar successes in materials were also achieved in the electric power generation equipment sector (see Case Text 14.5). Porter (1990: especially Chapter 10) describes the development sequence as embracing four stages: the factordriven stage, investment-driven stage, innovation-driven stage and welfare-driven stage. The international competitiveness of an economy which stays at the factor-driven stage is based only on basic factors such as cheap semi-skilled labour and natural endowments. At higher stages, the firm-specific and nation-specific factors which have been developed serve as the basis for a higher level of international competitiveness. It is necessary to note that while this and the above paragraphs are relevant to the links between the machinery industry and industrial development in less developed economies, the question of optimal resource allocation between the capital goods sector and consumer goods sector is not considered here. A review of theoretical debates about this resource allocation problem is provided by Martin Fransman (1986). Two well known models in this area are the Feldman model, which originated from the Soviet industrialization in the 1920s, and the Mahalanobis model which was the basis for Indian industrial planning in the 1950s and 1960s. Our analysis focuses on investment efficiency in the capital goods industry, taking for granted the importance of the machinery industry in enabling an economy to endogenize innovative skills. In fact the learning efficiency of the machinery industry largely determines the role which is expected of investment in the industry, while investment by itself does not necessarily lead to innovativeness in an economy. ‘Infrastructure’ is a supporting structure to economic activities. In the traditional sense it refers to transportation and communication networks. These networks are physical installations, and thus largely geographically specific. By the term ‘technological infrastructure’ Steinmueller implies a supporting structure to the dissemination of commercially competitive technological knowledge. Focused on the information industry, he argues that technological knowledge which is embodied in products from a commercially developed trajectory highlights development bottlenecks and spontaneously conveys the methods and skills to overcome them. These methods are built upon very congruous IC devices, the common digitization principle of IC design, and a small number of interface methods, so that the knowledge spillover points to an entry strategy for new entrants. This explains why technological diffusion has never stopped and new entrants have continued to emerge in the IC and computer industries despite the significantly centralized firm structure. A substantial part of the technological infrastructure is not geographically specific, and it is also infinitely reproducible (Steinmueller 1996). It would be very interesting to compare the ‘technological infrastructure’ effect in different Chinese industries during the economic transition. The penetrating power of internationally tradable technology would appear to be much stronger in the information industry due to the very strong demonstration and instrument effect, as our limited observations suggest. In the development of metalforming systems, it continues to be necessary for the local industry to maintain and generate generic technological know-how (Case Text 17.1). In comparison with this, innovative effort in the development of manufacturing automation systems (Case Text 17.2), which relates to electronics technology, appears to have moved much further ‘downstream’. Many mechanical parts of the manufacturing automation systems remain in need of development work by local developers. The data cited covers sixty-two R&D institutes. Note that the trends indicated by the data should not be understood as exact measures because statistics for different years may not be precisely comparable. It is not necessary here to address the whole range of issues concerning codified versus tacit technological knowledge which have been debated in the literature (see, for instance, David and Foray 1995; Foray 1995; Dasgupta and David 1994). Suffice to remark that our concern is the capability of institutes and enterprises in

268

NOTES

codifying, generating and communicating engineering knowledge for the purpose of developing firm-specific innovation. Being organization-specific, the engineering knowledge generated and communicated can remain largely localized in particular innovating agents, and thus tacit in some sense to outsiders. In this aspect we differ from the literature mentioned above, which has correctly argued that codification is a generalization, but has tended to identify technological knowledge generalization entirely with scientific knowledge generalization. This could lead one to overlook the importance of a wide array of codification which is indispensable to deal with ‘lower-order’ technological knowledge which is relevant to specific subjects, localized in the innovating firm and focused on operational procedures. 16 But one problem is that engineering services are still undervalued, another is that market demand for engineering services from the inexperienced small and medium-sized manufacturers who are the institutes’ clients is shallow, which may have inhibited the need for the improvement of these services (see, for example Case Text 16.5).

19 INSTITUTIONAL RESTRUCTURING 1 Coase, for example, comments that ‘My discussion with researchers in this field has made it clear to me that Williamson’s influence has been immense. In a real sense, transaction cost economics, through his writing and teaching, is his creation’ (Coase 1993). 2 The higher the degree of novelty which a technological innovation involves, the higher the technological and market uncertainties that may be associated with the innovation. See Kline and Rosenberg 1986. 3 Williamson in fact has a strong definition of behavioural opportunism, defining it as ‘self-interest seeking with guile’ (Williamson 1985:47), or ‘a deep condition of self-interest seeking that contemplates guile’ (Williamson 1988:68. Italics added in both cases). The element of guile is not necessary to the definition, since it is vague and adds nothing to the meaning of ‘opportunism’, and some authors have doubted whether such a strong form of the behavioural ‘opportunism’ assumption can be supported. For instance Coase has said: ‘I was also doubtful about the validity of Williamson’s treatment of “opportunism” as a significant rationale for vertical integration (not, of course, as an explanation of other contractual arrangements)’ (Coase 1993). For our purposes it is unnecessary to consider this detail of the assumption. 4 Other approaches to economic reforms are seen as giving more attention to ‘getting the price right’ or ‘getting the property rights right’ (Williamson 1994), but this does not imply a definite contradiction between these approaches. For instance, price mechanisms and property rights can be included in the transaction cost approach as important institutional environment factors. 5 The ‘institutional environment’ refers to every aspect of the institutional arrangements external to the economic agent being considered, and under which the economic agent operates. Williamson cites Davis and North: ‘The institutional environment is the set of fundamental political, social, and legal ground rules that establish the basis for production, exchange and distribution. Rules governing elections, property rights, and the right of contract are examples’ (Davis and North 1971, cited in Williamson 1994). 6 The depiction and explanation of the schema are from Williamson (1985:33; 1993:130; 1994:184) except that the pointer s, or specificity, refers in Williamson’s framework only to the specificity of the productive assets, whereas in our case the specificity of the good or service produced is seen as a more important determinant of the transaction risk, particularly since the degree of asset specificity is much the same for all of the institutes in our sample, and is not easily changed in the short term, so that asset specificity is not expected to explain the direction and differences in the institutional transformations which are observed in these institutes. Our schema also has no pointer for price. In Williamson’s original schema, the price at node A is a competitive price in a pure market, and the relationship between prices at nodes B and C is p(B)>p(C) (Williamson 1985:32– 34). Our analysis need not deal explicitly with price, given that the goods and services provided by the sample institutes are usually new so it is convenient to suppose that the prices for them are free from both regulatory controls and the influence of a free market price.

NOTES

269

7 Strictly speaking, to consider the sample institutes as operating at node A is an inexact analogy, because Williamson’s schema is intended to depict contractual relations under a capitalist (market) system. This analogy is however useful for understanding why the starting point of market reform was the node B condition. The node B condition emerged simply because assets which had been transactionally non-specific became specific, and this was due not to a change in the technological specificity of the assets but rather to the introduction of market transactions for technology. In fact the node B condition is rare in well-developed market economies. Another way of describing these changes in the transaction costs approach would be to consider the whole of the centrally planned system, including both the institutes and the enterprises which use their technology products, as one huge U-form (unified ownership) structure (see Figure 19.3). The institutes and their users are then seen as departments of that U-form structure in which there were no transaction risks because of the unified ownership and no safeguards because no risks had to be prevented (k and s are both 0). Seen in these terms, the market reform has the effect of dissolving the U-form structure to turn the internal departments into external trading partners in the marketplace. Safeguards were then needed to replace the central coordination previously provided by the U-form ‘headquarters’. 8 Technology developed in the previous institutional framework was highly idiosyncratic not only because any given development was narrowly assigned for a specific product (see the cases in Chapter 16), but also because the complements needed to incorporate a certain development into a system were assigned to different developers (see especially the case of manufacturing technology shown in Chapter 17). Other idiosyncrasies came from the weakness of the technology, associated with poor product quality and low production efficiency, as discussed in Chapter 18. 9 As Williamson puts it: ‘Vertical integration…of two kinds is usefully distinguished. The first…involves integration of successive stages within the core technology…. The second…involves integration of peripheral or off-site activities… Most discussions of vertical integration pass over the first and focus entirely on the second.’ (Williamson 1985:105). 10 These cost variables are similar to those used by Williamson in discussing vertical integration (Williamson 1985: 90–94), with modifications in accordance with the characteristics of the sample institutes. The definitions of ß and M are similar to those used by Williamson, but asset specificity, k, which is employed by Williamson to differentiate various possibilities for vertical integration does not appear in this model because asset specificity is generally high for all the institutes considered here. It can reasonably be regarded as having the same significance for all the institutes and before and after the implementation of their integration strategies. This treatment of asset specificity k locates our sample on the right side of Williamson’s Figures 4–1 and 4–2 (Williamson 1985:91, 93). Finally, the investment in productive assets, I, is introduced in our model in place of Williamson’s ‘steady state production cost’. This is not a significant difference because the sample institutes were not previously engaged in manufacture. Therefore it is the required investment in production, in relation to various adaptive strategies, which is likely to be the sensitive variable. 11 In the transaction cost literature, an economic agent’s internal governance hierarchy is closely related first to the kind of products or services which it produces and second to how it delivers these to its customers, that is, to its external contractual relationships (Williamson 1985:22). The organizational hierarchy and market intermediation both serve as governance mechanisms for contractual relationships. In transaction cost theory, the hierarchical organization (in an agent functioning in a market environment) is seen as a way of economizing on scarce ‘computational’ capacity. Given the complexity and uncertainty of human activities, completely determined decisions (like complete contracts), are seldom possible. The cost of exhaustively specifying the decision tree in advance would be prohibitive. The internal organization permits the parties to deal with uncertainty and complexity in an adaptive, sequential process. Events are permitted to unfold and attention is devoted to the actual, rather than all possible, outcomes (Williamson 1975:21– 25). 12 In the capitalist economies, U-form industrial organization originated from single and unified private ownership. In the late 1800s the U-form enterprises evolved as large single-product manufacturers, often with multifunctional internal divisions but with completely centralized decision-making (Williamson 1975:133; 1985:280, cited from

270

13

14

15

16

17

NOTES

Chandler 1966). The pre-reform internal organization of productive enterprises and R&D institutes in the Chinese machinery industry are similar in terms of large scale, internal multifunctional structure and centralized decision-making. In the capitalist economies, there were revolutionary organizational changes affecting the U-form in the US during the first half of this century. In West Europe, the same changes took place during the 1960s and 1970s. Radical restructuring was necessary in U-form enterprises in the West because the managerial inefficiency increased as the U-form enterprises expanded. ‘The inherent weakness…became critical…when the operations of the enterprise became too complex and the problems of coordination, appraisal, and policy formulation too intricate for a small number of top officers to handle both long-run, entrepreneurial, and short-run, operational administrative activities’ (Williamson 1975: 134, cited from Chandler 1966). The pre-reform Chinese U-form organization was a peculiar kind of U-form, rigidly closed to any adjustment and insensitive to cost-efficiency factors. In fact the pre-reform organizations may be regarded as one national ‘pan Uform’. Technological standardization and the linking of firms and the related R&D institutes in an organization based on product specialization reduced the management load for the centre to make the huge U-form workable, but led to a system capable only of developing general-purpose technology (see Chapter 14, ‘institutions of the machinery industry’, and Chapter 18). The H-form organization, as has already been said, refers to an internally decentralized decision-making structure based on internally-contracted company assets, but without an internal control apparatus to govern the decentralized divisions (Williamson 1975:152). H-form and U-form structures were the two leading forms of internal corporate organization in the US until the 1920s. General Motors operated at that time like an H-form company in which the company was in a situation of never being able to get things under control (Williamson 1985:280). The internal inefficiency experienced with Hform organizations arose because the internal contracts were too incomplete. This contrasts with the U-form, which imposed excessively cumbersome controls on internal contracting relations. The development of M-form corporate organizations was driven by these inefficiencies of both the H-form and the U-form, inefficiencies which became more serious as firms grew in size and complexity (Williamson 1985:295–296). For instance, the Director of the Shanghai Electric Equipment Research Institute confirmed the strategic need for an H-form internal organization in 1994, during our field visit. The internal responsibility system was said to be ‘useful at present’ because the institute centre had not yet been able to raise enough funds, forcing it ‘to rely on research departments to capture various (external) contracting opportunities’. Note that compared with the majority of the group I institutes, this institute had delayed the shift from U-form to H-form because of a heavy commitment in conducting state projects under the five year plan in the 1980s. See Case Text 16.4. One case among the sixty-four group I institutes was the Wuhan Research Institute for Computer Peripherals. Under an H-form ‘internal contractual responsibility system’, research teams of the institute became rich and the institute as a whole almost went bankrupt in the early 1990s. Later it was reorganized by its ‘principal’. The central Ministry appointed a new Director and tightened internal regulations for the institute. ‘Opportunism’ by contracting departments and individuals in the H-form context can, as indicated in the extreme case, entirely dissipate the organization’s assets (mimeo provided by the Ministry of Machinery Industry; interviews at the Ministry, October 1994 and May 1995). In the three productive enterprises covered in our study, the U-form, rather than H-form, internal organization continued until the first half of the 1990s. A ‘management responsibility system’ for enterprises was adopted but it covered the contractual relationships between enterprises and their ‘principals’, that is the local governments or central government agencies overseeing the use of stateowned assets, rather than the enterprises’ internal contractual relationships (see Case Texts 16.1, 16.2 and 16.4). The dominance of U-form internal organization in productive enterprises during the market reform may be linked to their specialization in a single product. Market niches may have been more apparent for such enterprises. The subsequent shift from a U-form to M-form structure as managerial efficiency became a key goal and internal organization became more complicated is illustrated by the Dalian Machine Tool Works. The Works was seriously planning M-form innovation in late 1994, when we visited them (Case Text 16.2, see also 16.1).

NOTES

271

18 The broad characteristics of the M-form are as follows (Williamson 1975:135– 136, 150):1) the responsibility for operational decisions is assigned to operating divisions (disaggregation of functions); 2) the elite staff of the general office perform both advisory and auditing functions (central control mechanism for internal contracting); 3) the general office is concerned principally with strategic decisions for the overall performance of the organization, including the allocation of resources (central strategic function). The M-form structure thus displays a systems effect: the organization as a whole functions more effectively and efficiently than the sum of the parts. In the United States, the first M-form innovation appeared in the 1920s at Du Pont, which had previously been a U-form structure. M-form innovation spread widely in the American economy in the following years, because of its superiority. 19 Williamson noted that the interdependency of internal divisions affects the appropriate degree of involvement by the organizational centre: ‘The appropriate degree of involvement by the general office in the affairs of the operating divisions varies with the nature of the factor or product market interdependencies that exist within the firm and thus need to be “harmonized”’. Factor interdependency is illustrated with the example of the exchange of intermediate products between divisions, and for product market interdependency the example is a situation in which competing products are being produced by different divisions. In both cases there is a need for central control (Williamson 1975:151–152). The number of market niches should be significant for our sample because technological assets, as ‘higher-order’ assets, may be deployed in response to widely differentiated opportunities in the technology market, as the cases studied have shown. 20 The manufacturing lines which are exported are used by the Thai subsidiary of a multinational for stamping automobile parts (People’s Daily, Overseas Edition, 3 February 1997).

BIBLIOGRAPHY

Books and journal articles Amann, Ronald and Julian Cooper (eds): Industrial Innovation in the Soviet Union, New Haven and London, Yale University Press, 1982. Aoki, Masahiko: Information, Incentives, and Bargaining in the Japanese Economy, Cambridge University Press, 1988. Arora, A. and A.Gambardella: The Changing Technology of Technological Change: General and Abstract Knowledge and the Division of Innovative Labour, Research Policy 23, 1994:523–532. Arrow, Kenneth J.: The Economic Implications of Learning by Doing, Review of Economic Studies 29, 1962:155–173. Arrow, Kenneth J.: The Limits of Organization, New York, W.W.Norton & Company, 1974. Baark, Erik: The Value of Technology: A Survey of the Chinese Theoretical Debate and its Policy Implications, Research Policy 17, 1988:269–282. Balazs, K.: Transition of the Science and Technology Management System in Hungary, Science and Public Policy, Vol. 19, No. 2, April 1992:89–97. Balazs, K.: Lessons from an Economy with Limited Market Functions: R&D in Hungary in the 1980s, Research Policy, Vol. 22, No. 5–6, November 1993: 537–552. Beijing Zone, Beijing Zone Writing Group, beijing xinjishu chanye kaifa shiyanqu yanju baogao (Research Report on Beijing Experimental Zone for Development of New Technology Industries), mimeo, Beijing, 1992. Beijing Zone, Beijing xinjishu chanye kaifa shiyan qu (Beijing Experimental Zone for Development of New Technology Industries—Beijing Zone): wuzhounian tekan 1988–1993 (Special Issue of the Fifth Anniversary 1988– 1993), Beijing, 1993. Bentley, Raymond: Research and Technology in the Former German Democratic Republic, Boulder, CO, Westview Press, 1992. CAS, Chinese Academy of Sciences (CAS): zhongguo kexueyuan fazhanshi (A History of the Development of the Chinese Academy of Sciences), mimeo, 1989. Castells, Manuel and Peter Hall: Technopoles of the World, London and New York, Routledge, 1994. CCCPC, Central Committee of the Communist Party of China: zhonggong zhongyang guanyu jianuo jilai dangde ruogan lishi wenti de jueyi (Resolution of the Central Committee of the Communist Party of China on Some Historical Issues), 27 June 1981. CCCPC, Central Committee of the Communist Party of China: zhonggong zhongyang guanyu jingji tizhi gaige de juedin (Decision of the Central Committee of the Communist Party of China on the Reform of the Economic Management System), 20 October 1984. CCCPC, Central Committee of the Communist Party of China: zhonggong zhongyang guanyu kexue jishu tizhi gaige de juedin (Decision of the Central Committee of the Communist Party of China on the Reform of the Science and Technology Management System), 13 May 1985. Chen Zhaoying et al.: zhongguo gaoxin jishu chanye kaifaqu fazhan baogao (Report on the Development Zones for High and New Technology Industries), Beijing, zhongguo kexue jishu chubanshe (Science and Technology Publishing House of China), 1992. Cheng, Chu-yuan: The Machine-Building Industry in Communist China, Edinburgh, Edinburgh University Press, 1972. China Machinery Industry Yearbook: zhongguo jixie gongye nianjian, Beijing, jixie gongye chubanshe (Machinery Industry Publishing House), 1991, 1992, 1993, 1994.

BIBLIOGRAPHY

273

China Science and Technology Indicators 1994: zhingguo kexue jishu zhibiao 1994, Beijing, State Science and Technology Commission, 1995. China Statistical Yearbook (zhongguo tongji nianjian), various years. China Statistical Yearbook 1990, English version. China Statistical Yearbook on Science and Technology: zhongguo keji tongji nianjian, Beijing, State Science and Technology Commission (SSTC), various years. Chudnovsky, Daniel and Masafumi Nagao, with the collaboration of Staffan Jacobsson: Capital Goods Production in the Third World, London, Pinter, 1983. Coase, Ronald H.: The Institutional Structure of Production, The American Economic Review, September 1992: 713–719. Coase, Ronald H.: Coase on Posner on Coase, Journal of Institutional and Theoretical Economics 149, 1993:96–8; reprinted in Oliver E.Williamson and Scott E.Masten (eds): Transaction Cost Economics, Volume I and II, Aldershot, Edward Elgar, 1995:668–670. Conroy, Richard: Technological Change in China, Paris, OECD, 1992. Cooper, Charles (ed.): Science, Technology and Development, London, Frank Cass, 1973. Dalum, Bent: Export Specification, Structural Competitiveness and National Systems of Innovation, in Bergt-Åke Lundvall: National Systems of Innovation, Towards a Theory of Innovation and Interactive Learning, London, Pinter, 1992:192–208. Dasgupta, Partha and Paul David: Toward a New Economics of Science, Research Policy 23, 1994:487–521. Databook of Statistics on Science and Technology: keji tongji, State Science and Technology Commission (SSTC), various years. David, Paul and Dominique Foray: Information Distribution and the Growth of Economically Valuable Knowledge: A Rationale for Technological Infrastructure Policy, in Teubal, Foray, Justman, and Zuscovitch (eds): Technological Infrastructure Policy: An International Perspective, Dordrecht etc., Kluwer Academic Publishers, 1996:87–116. Deng Xiaoping: zai quanguo kexue dahui kaimushi shangde jianghua, 18 March 1978 (Speech at the Opening Ceremony of the National Science Conference), in White Paper No. 1:251–258. Deng Xiaoping: zai quanguo kexue jishu gongzuo huiyi shangde jianghua, 7 March 1985 (Speech at the National Working Conference of Science and Technology), in White Paper No. 1:291–292. Dong Guilan and Peng Ying: qiye fuhuaqi, lianjie zhishi, jishu yu shichang de qiaolian (Incubator—the Bridge Linking Knowledge, Technology and Market—Enterprise Incubators), Guangxi, Guangxi shifan daxue chubanshe (Guangxi Normal University Press), 1992. Dosi, Giovanni: Technological Paradigm and Technological Trajectory, Research Policy 11, 1982:147–162. Enterprise Development and Management Department of the Beijing Experimental Zone for the Development of New Technology Industries (ed.): xinjishu jiqi chanpin (gongcheng) ji (Catalogue of New Technology & Related Products and Engineering), Beijing, 1992. Enterprise Development and Management Department of the Beijing Experimental Zone for the Development of New Technology Industries (ed.): xinjishu jiqi chanpin (gongcheng) ji (Catalogue of New Technology & Related Products and Engineering), Beijing, March 1993. Ernst, Dieter and David O’Connor: Technology and Global Competition, The Challenge for Newly Industrialising Economies, Paris, OECD, 1989. Ernst, Dieter and David O’Connor: Competing in the Electronics Industry, The Experience of Newly Industrialising Economies, Paris, OECD Development Centre, 1992. Fang Yi: zai quanguo kexue dahui shangde baogao, 18 March 1978 (Report to the National Science Conference), in White Paper No. 1:259–271 Foray, Dominique: The Economics of Intellectual Property Rights and Systems of Innovation: The Persistence of the National Practices Versus the New Global Model of Innovation, in John Hagedoorn (ed.): Technological Change and the World Economy, Convergence and Divergence in Technology Strategies, Aldershot, Edward Elgar 1995: 109–133. Foray, Dominique: The Economics of Codification, mimeo, seminar presented at MERIT, Maastricht, June 1996.

274

BIBLIOGRAPHY

Fransman, Martin: Machinery and Economic Development, in Martin Fransman (ed.): Machinery and Economic Development, London, Macmillan Press, 1986:1–53. Freeman, Christopher: The Economics of Industrial Innovation, second edition, London, Pinter, 1982. Granick, David: Soviet Metal-Fabricating and Economic Development, Practice versus Policy , The University of Wisconsin Press, Madison, Milwaukee, and London, 1967. Gu, Shulin: gaojishu xiangmu kaifa yu gaojishu xiangmu pingjia (Case Analysis of Six Ventures in the Chinese Academy of Sciences), Administrative World (guanli shijie) 4, 1988. Gu, Shulin: Spin-off Enterprises in China: Channelling the Components of R&D Institutions into Innovative Businesses, UNU/INTECH Working Paper No. 16, Maastricht, 1995a. Gu, Shulin: A Review of Reform Policy For S&T System in China: From Paid Transaction for Technology to Organizational Restructuring, UNU/INTECH Working Paper No. 17, 1995b, Maastricht. Gu, Shulin: The Emergence of New Technology Enterprises in China: A study of endogenous capability building via restructuring, Journal of Development Studies, Vol. 32 No. 4, April 1996a. Gu, Shulin: Towards an Analytical Framework for National Innovation Systems, UNU/ INTECH Discussion Paper 9605, Maastricht, 1996b. Gu, Shulin and W.Edward Steinmueller: China’s National Innovation System Approach to Participating in Information Technology: The Innovative Recombination of Technological Capability, UNU/INTECH Discussion Paper 9701, Maastricht, 1997. Guo Rui (ch. ed.): shijie jidian gongye yaolan (Guide to the World’s Machinery and Electronics Industries), Beijing, China Machinery Industry Press, 1994. Hanson, Philip and Keith Pavitt: The Comparative Economics of Research Development and Innovation in East and West: a Survey, Chur, Harwood Academic Publishers, 1987. Hobday, Mike: Innovation in East Asia, The Challenge to Japan, Aldershot, Edward Elgar, 1995. Jing Xiaocun (ch. ed.): dangdai zhongguo de nongye jixie gongye (The Agricultural Machinery Industry in Contemporary China), Beijing, zhongguo shehui kexue chubanshe (Social Science Publishing House of China), 1988. Jing Xiaocun (ch. ed.): dangdai zhongguo de jixie gongye (The Machinery Industry in Contemporary China) Beijing, zhongguo shehui kexue chubanshe (Social Science Publishing House of China), Vol. A and Vol. B, 1990. Kim, Linsu: National System of Industrial Innovation: Dynamics of Capability Building in Korea, in Richard R.Nelson (ed.): National Innovation System, A Comparative Analysis, Oxford University Press, 1993:357–383. Kim, Linsu and Carl J.Dahlman: Technology Policy for Industrialization: An Integrated Framework and Korea’s Experience, Research Policy 21, 1992:437–452. Kline, Stephen J. and Nathan Rosenberg: An Overview of Innovation, in Ralph Landau and Nathan Rosenberg (eds.): The Positive Sum Strategy: Harnessing technology for economic growth, Washington DC, National Academic Press, 1986. Knight, J.B.: The Evolution of Development Economics, in V.N.Balasubramanyam and S.Lall (eds.): Current Issues in Development Economics, Basingstoke, Macmillan, 1991. Kodama, K.: Can Changes in the Techno-Economic Paradigm be Integrated through Empirical and Quantitative Study? STI Review, No. 7, 1990:101–129. Kornai, Janos: Economics of Shortage, North-Holland Publishing Company, 1980. Kuhn, Thomas: The Structure of Scientific Revolutions, University of Chicago Press, IL, 1967. Lall, Sanjaya: Building Industrial Competitiveness in Developing Countries, Paris, OECD, 1990. Lall, Sanjaya: Technological Capability and Industrialization, World Development, 20, 1992:165–186. Li Fang: fazhan keji chanye shi kaifaxing keyan yuansuo gaige de fangxiang (Developing technology-intensive businesses is the direction of reform for R&D institutes engaged in technology development), in Forum on Science and Technology in China 3, 1992:41–43. Li Fusheng et al (ed.): shiyong shukong jichuang jishu shouce (Handbook of Efficient Technology for Numerical Control Machine Tools), Beijing, Beijing Publishing House, 1993.

BIBLIOGRAPHY

275

Li Jingwen, Dale W.Jorgenson, Zheng Youjing, and Masahiro Kuroda: shengchanlu yu zhong mei ri jingji zhengzhang yanjiu (Productivity and Economic Growth in China, USA and Japan), Beijing, zhongguo shehui kexue chubanshe (Social Science Publishing House of China), 1993. Lin Chenhui: chanyehua guimohua de tiaozhan—beijingshi xinjishu chanye kaifa shiyianqu mianlin de kunnan wenti he duice (Solutions to the Difficulties and Problems Faced by the Beijing New Technology Zone), Science and Technology Review 10, 1992:8–11. Liu Chuanzhi: Establishing a Computer Industry Competitive on the World Market, Science and Technology in China: Selection from the Bulletin of the Chinese Academy of Sciences, Beijing, Science Press, Vol. 5, 1991:219–224. Liu Ying et al. (eds.): dongdai zhongguo de dianzi gongye (The Electronics Industry of Contemporary China), Beijing, zhongguo shehui kexue chubanshe (Social Science Publishing House of China), 1987. Lu Yanlin: guanyu zuohao banfa liangzheng gongzuo de jidian yijian (On the Licensing of Design Institutions), jianzhu sheji guanli (Management of Architecture and Design) 3, 1992:6–11. Lundvall, Bengt-Åke: User-Producer Relationships, National Systems of Innovation and Internationalisation, in BengtÅke Lundvall (ed.): National Systems of Innovation, London, Pinter, 1992:45–67. Luo Wei: zhongguo de kexue jishu fazhan guihua (The Science and Technology Development Programmes in China), ziran bianzhengfa tongxun (Journal of Dialectics of Nature) 4, 1983:25–31. Mowery, David C.: Industrial Research and Firm Size, Survival, and Growth in American Manufacturing, 1921–46, Journal of Economic History 43, 1983: 953–80. Mowery, David C. and Nathan Rosenberg: Technology and the Pursuit of Economic Growth, Cambridge University Press, 1989. Mowery, David C. and Nathan Rosenberg: The Influence of Market Demand upon Innovation: A Critical Review of Some Recent Empirical Studies, Research Policy 8, 1979:103–53. Nagaoka, Sadao: Reform of Ownership and Control Mechanisms in Hungary and China, Washington DC, World Bank, Industry and Energy Department Working Paper, Industry Series Paper No. 7, April 1989. Nelson, Richard R. (ed.): National Innovation Systems, a Comparative Analysis, Oxford University Press, 1993. Nelson, Richard R. and Sidney G.Winter: An Evolutionary Theory of Economic Change, Cambridge, MA, and London, The Belknap Press of Harvard University Press, 1982. North, Douglass: Economic Performance Through Time, The American Economic Review 84, 1994:359–368. OECD, Organization for Economic Co-operation and Development: Science and Technology in the People’s Republic of China, Paris, OECD, 1977. OECD, Organization for Economic Co-operation and Development: Reviews of National Science and Technology Policy: Czech and Slovak Federal Republic, Paris, OECD, 1992. OECD, Organization for Economic Co-operation and Development: Science, Technology and Innovation Policies, Hungary, Paris, OECD, 1993a. OECD, Organization for Economic Co-operation and Development: Basic Science and Technology Statistics, Paris, OECD, 1993b. OECD, Organization for Economic Co-operation and Development: Science, Technology and Innovation Policies, Federation of Russia, Volume I: Evaluation Report, Paris, OECD, 1994a. OECD, Organization for Economic Co-operation and Development: Science, Technology and Innovation Policies, Federation of Russia, Volume II: Background Report, Paris, OECD, 1994b. Ou Wen: guanyu qiwu keji gongguan guanli zhong cunzai de wenti (Problems Appearing in the Management of Key S&T projects of the Seventh Five Year Plan), keji guanli yanjiu (Studies on Management of Science and Technology) 4, 1991:5–7. Pack, Howard and Larry E.Westphal: Industrial Strategy and Technological Change, Journal of Development Economics 22, 1986:87–128. Patel, P. and K.Pavitt: The Continuing, Widespread (and Neglected) Importance of Improvements in Mechanical Technologies, Research Policy 23, 1994:533–548. Pavitt, K.: Sectoral Patterns of Technical Change: Towards a Taxonomy and a Theory, Research Policy 13, 1984: 342–373.

276

BIBLIOGRAPHY

Perkins, Dwight: Completing China’s Move to the Market, Journal of Economic Perspectives 8, 1994:23–46. Piskunov, D.I. and B.Saitykov: Transforming the Basic Structures and Operating Mechanisms of Soviet Science, Science and Public Policy 19, 1992:111–118. Porter, M.: The Competitive Advantage of Nations, London and Basingstoke, Macmillan, 1990. Poznanski, Kazimierz: The Environment for Technological Change in a Centrally Planned Economy, Washington DC, World Bank Staff Working Papers No. 718, 1985. Qian Yingyi and Xu Chenggang: Why China’s Economic Reforms Differ: The M-Form Hierarchy and Entry/Expansion of the Non-State Sector, The Economics of Transition 1, 1993:135–170. Rahman, A. and K.D.Sharma (eds): Science Policy Study, New Delhi, Somaiya Publication, 1974. RISA, Research Institute of System Analysis, Research Institute of Machinery Science and Technology (RIMST) of the Ministry of the Machinery Industry (written by Dong Lijun, Zhang Meiguang, Xie Baoxin): yuansuo taishi fenxi baogao (1984– 1988) (Report on the Trends in the Reform of R&D Institutes (1984–1988)), mimeo, Beijing, 1989. RISA, Research Institute of System Analysis, Research Institute of Machinery Science and Technology (RIMST) of the Ministry of the Machinery Industry: 1980 nian yuansuo taishi henxi baogao—jixiekou 52 ge kaifasuo (1989 Report on the Trends in the Reform of the 52 Machinery Technology Development Institutes), mimeo, Beijing, 1990. RISA, Research Institute of System Analysis, Research Institute of Machinery Science and Technology (RIMST) of the Ministry of Machinery Industry, and Department of Science and Technology of the Ministry of the Machinery Industry: 1990 nian keyan gaige taishi fenxi (1990 Report on the Trends in the Reform of the R&D Institutes), mimeo, Beijing, 1991. RISA, Research Institute of System Analysis, Research Institute of Machinery Science and Technology (RIMST) of the Ministry of Machinery Industry, and Department of Science and Technology of the Ministry of Machinery Industry (written by Zhang Meiguang, Li Chong, Xie Baoxin): 1991 nian jidianbu jixiekou 56 ge zhishu kaifaleisuo taishi fenxi baogao (1991 Report on the Trends in the Reform of the 56 Ministryaffiliated Technology Development Institutes), mimeo, Beijing, 1992. RISA, Research Institute of System Analysis, Research Institute of Machinery Science and Technology (RIMST) of the Ministry of the Machinery Industry: 1992 nian jixiebu 64 ge zhishu yuansuo kejigongzuo taishi fenxi ji kaoping jieguo baogao (1992 Report on the Trends in the Reform of the 64 Ministry-affiliated R&D Institutes), mimeo, Beijing, 1993. RISA, Research Institute of System Analysis, Research Institute of Machinery Science and Technology (RIMST) of the Ministry of Machinery Industry: 1993 jixiebu 64 ge zhishu yuansuo keji gongzuo taishi fenxi baogao (1993 Report on the Trends in the Reform of the 64 Ministry-affiliated R&D Institutes), mimeo, Beijing, 1994. RISA, Research Institute of System Analysis, Research Institute of Machinery Science and Technology (RIMST) of the Ministry of Machinery Industry: 1994 nian jixiebu 62 ge zhishu yanjiu yuansuo kejigongzuo taishi fenxi baogao (1994 Report on the Trends in the Reform of the 62 Ministry-affiliated R&D Institutes), mimeo, Beijing, 1995. Rosenberg, Nathan: Learning by Using, in N.Rosenberg: Inside the Black Box: Technology and Economics, Cambridge, Cambridge University Press, 1982. Rosenberg, Nathan: Perspectives on Technology, Armonk, New York, and London, M.E. Sharpe, 1985:9–31. Rosenberg, Nathan and L.E.Birdzell, Jr.: How the West Grew Rich, The Economic Transformation of the Industrial World, New York, Basic Books, 1986. Rosenberg, Nathan and Richard R.Nelson: American Universities and Technical Advance in Industry, Research Policy 23, 1994:323–348. Sheng Shuren: Editorial Committee with Sheng Shuren as the Chief Member: 1979–1988 nian zhongguo yingjing jishu gaizao xianyou qiye shinian jian (1979– 1988 Ten Years’ Inspection of China’s Technology Imports for the Renovation of Existing Enterprises), Bijing, zhongguo jingji chubanshe (Economic Publisher of China), 1991. Shi Jianzhong (head of research team): jixie kexue yanjiuyuan kexue jishu shili fenxi ji jinhou fazhan de zhengce jianyi (An Analysis of the Scientific and Technological Strengths of the RIMST and Policy Recommendations for Further Development), mimeo, 1992. Shirk, Susan L.: The Political Logic of Economic Reform in China, University of California Press, 1993. Simon, H.A.: The Structure of Ill Structured Problems, Artificial Intelligence 4, 1973: 181–201.

BIBLIOGRAPHY

277

SPC et al. (eds.): Department of Foreign Trade of the State Planning Commission, Department of Technology Imports and Exports of the Ministry of Foreign Trade, and the Information Centre for Technology Imports of the Ministry of the Machinery and Electronic Industry (eds): zhonghua renmin gongheguo jishu yingjin sishinian (1950–1990) (Forty Years’ Technology Imports of the People’s Republic of China), Beijing, Wenhui chubanshe (Wenhui Publishing House), 1992. SSB, Department of Science and Technology Statistics, State Statistical Bureau: zhongguo kexue jishu sishi nian (Statistics on the Science and Technology of China, 1949–1989), Beijing, zhongguo tongji chubanshe (Chinese Statistical Publishing House), 1990a. SSB, Urban Social and Economic Survey Organization of the State Statistical Bureau: zhonggou chengshi sishi nian (China, The Forty Years of Urban Development), China and Hong Kong, China Statistical Information and Consultancy Service and International Center for the Advancement of Science and Technology Limited, 1990b. SSTC, State Science and Technology Commission: guanyu woguo kexue jishu fazhan fangzhen de huibao tigang (Report Outline on the Guidelines for the Development of Science and Technology in China), 23 February 1981, in White Paper No. 1:307–313. SSTC, State Science and Technology Commission: China’s Technology Policy—“Blue Paper” of the State Science and Technology Commission: Energy Resources, Beijing, 1985a. SSTC, State Science and Technology Commission: China’s Technology Policy—Blue Paper of the State Science and Technology Commission: The Machinery Building Industry, Beijing, 1985b. SSTC, State Science and Technology Commission and the Centre for S&T for Development (eds.): woguo jishu yinjin zhanlue yu zhengce yanju (Strategy and Policy for Technology Imports), SSTC, 1987. SSTC, State Science and Technology Commission (ed.): zhongguo qiye zhiguang—qiye keji jinbu ziliao huibian (Collection of Materials on Technological Progress in Enterprises), SSTC, 1989. SSTC, State Science and Technology Commission, Department of S&T Management Restructuring (ed.): zhongguo keji tizhi gaige zhengce huibian 1985–1990 (Collection of Policy Documents on the Reform of the Science and Technology Management System in China (1985–1990), Vols. A and B, Beijing, Astronautics Press, 1991a and b. SSTC, State Science and Technology Commission, Department of S&T Management Restructuring (ed.): quanguo keji tizhi gaige dianxing jingyian xuanbian (Selection of Typical Experiences in the Reform of the Science and Technology Management System), Beijing, Astronautics Press, 1991c. SSTC and NCSTD, State Science and Technology Commission and the National Centre for Science and Technology for Development: zhongguo keji tizhi gaige taishi hengxi: 1988 (Analysis of the Reform Trends for the S&T Management System of China: 1988), Beijing, SSTC and NCSTD, 1989. SSTC and NCSTD, State Science and Technology Commission and the National Centre for Science and Technology for Development: zhongguo keji tizhi gaige taishi hengxi: 1989 (Analysis of the Reform Trends for the S&T Management System of China: 1989), Beijing, SSTC and NCSTD, 1990. State Council: guowuyuan guanyu jinyibu tuijin keji tishi gaige de yuogan guiding (Stipulations of the State Council Regarding Furthering the Reform of the S&T Management System), 20 January 1987a, in White Paper, No. 2: 319–320. State Council: guowuyuan guanyu tuijin keyan sheji danwei jinru dazhongxing gongye qiye de guiding (Stipulations of State Council on Facilitating the Entry of R&D and Design Units to Large and Medium-sized Enterprises), 20 January 1987b, in White Paper, No. 1:321–322. State Council and SPC, State Planning Commission (eds.): Office of the Leading Group of the General Industry Survey, and Institute of Economics, zhongguo gongye xianzhuang (The Current State of Chinese Industry), Beijing, renmin chubanshe (People’s Publishing House), 1990. State Council and STB: Office of the Leading Group of the National Industry Census, State Council, and Department of Statistics on Industry, Transportation and Materials, National Statistical Bureau (eds): zhongguo gongye jingji tongji ziliao 1986 (Statistical Information on China’s Industrial Economy), Beijing, tongji chubanshe (Statistical Publishing House), 1987.

278

BIBLIOGRAPHY

Steinmueller, W.Edward: Technological Infrastructure in Information Technology Industries, in Morris Teubal, Dominique Foray, Moshe Justman, Ehud Zuscovitch (eds): Technological Infrastructure Policy, Dordrecht, Kluwer, 1996. STIP, Scientech Information Institute, Ministry of Machinery Industry, PRC: shijie jidian gongye yaolan (Guide to the World’s Machinery & Electronics Industries) Beijing, China Machine Press, 1994. Suttmeier, Richard P.: Research and Revolution: Science Policy and Societal Change in China, Lexington, MA, D.C.Heath, 1974. Suttmeier, Richard P.: China’s High Technology: Programs, Problems, and Prospects, in 102nd Congress (of the U.S.) 1st Session (S. Prt. 102–21, Vol.2): China’s Economic Dilemmas in the 1990s: The Problems of Reforms, Modernization, and Independence, April 1991. Sutton, A.C., Western Technology and Soviet Economic Development, 1930 to 1945, Stanford, CA, Hoover Institution Press, 1971. Szirmai, Adam and Ren Ruoen: China’s Manufacturing Performance in Comparative Perspective 1980–1992, paper presented at the Development Economics Seminar held in the Institute of Social Studies, The Hague, 16 June 1995. Tidrick, Gene and Chen Jiyuan: China’s Industrial Reform, Oxford University Press, 1987. UNIDO internal document: ID/B/C.3/86/Add.1, Joint UNDP/UNIDO Evaluation of Industrial Research and Service Institutes, 21 November 1979. UNIDO: Handbook of Industrial Statistics 1990, Vienna, UNIDO, 1990 Von Hippel, Eric and Marcie J.Tyre: How Learning by Doing is Done: Problem Identification in Novel Process Equipment, Research Policy 24, 1995:1–12. Wang Xiaolong (ed.): Zhong guan cun dianzi yitaojie dashiji (Chronicle of Zhongguancun Electronics Street 1980– 1988), 1993. White Paper No. 1: State Science and Technology Commission (SSTC): zhongguo kexue jishu zhengce zhinan 1986, kexue jishu baipishu di’yihao (Guide to China’s Science and Technology Policy (1986), White Paper on Science and Technology No. 1), SSTC, 1987. White Paper No. 2: State Science and Technology Commission (SSTC): zhongguo kexue jishu zhengce zhinan 1987, kexue jishu baipishu di’erhao (Guide to China’s Science and Technology Policy (1987), White Paper on Science and Technology No. 2), SSTC, 1988. White Paper No. 3: State Science and Technology Commission (SSTC): zhongguo kexue jishu zhengce zhinan 1988, kexue jishu baipishu disanhao (Guide to China’s Science and Technology Policy (1988), White Paper on Science and Technology No. 3), SSTC, 1989. White Paper No. 4: State Science and Technology Commission (SSTC): zhongguo kexue jishu zhengce zhinan 1989, kexue jishu baipishu disihao (Guide to China’s Science and Technology Policy (1989), White Paper on Science and Technology No. 4), SSTC, 1990. Williamson, Oliver E.: Markets and Hierarchies: Analysis and Antitrust Implications, New York, The Free Press, 1975. Williamson, Oliver E.: The Economic Institutions of Capitalism, New York, The Free Press, 1985. Williamson, Oliver E.: The Logic of Economic Organization, Journal of Law, Economics, and Organization 4, 1988: 65–93. Williamson, Oliver E.: Transaction Cost Economics and Organizational Theory, Institutional and Corporate Change 2, 1993:107–156. Williamson, Oliver E.: The Institutions and Governance of Economic Development and Reform, Proceedings of the World Bank Annual Conference on Development Economics 1994:171–205. World Bank: China: Long-Term Issues and Options, Baltimore, Md, Johns Hopkins University Press, 1985. Wu Jinglian: cong beijing xinjishu chanye kaifaqu kan chuangxin jizhi (Innovation Mechanism in the Beijing Experimental Zone for Development of New Technology Industries), Science and Technology Review 5, 1991: 12–15. Wu Jinglian and Liu Jirei: lun jinzhengxing shichang tizhi (On a Competitive Market Institution), Beijing, zhongguo caijing chubanshe (Finance and Economy Publishing House of China), 1991.

BIBLIOGRAPHY

279

Xu Zenji et al.: kexue yanju yu jishu kaifa jigou de fenlei baogao (Working Paper on Classification of Scientific Research and Technological Development Institutes), Shenyang Branch, Chinese Academy of Sciences, December 1987, mimeo. Yan Zulin: xinjishu kaifaqu yu keji qiye fuhuaqi (New Technology Development Zones and Science & Technology Enterprise Incubators, Beijing, kexue chubanshe (Science Publishing House), 1991. Yu Weidong (ed.): xiwang de huoguang, chongguancun danzhi yitiaojie diaocha (The Spark for Hope, a Survey of Electronic Street), Beijing, renmin daxue chubanshe (People’s University Publishing House), 1988. Yuan Baohua (ch. ed.): dangdai zhongguo de jingji guanli (Economic Management in Contemporary China), Beijing, zhongguo shehui kexue chubanshe (Social Science Publishing House of China), 1985. Zhang Shaojie: jingji yunxing yu jingji tiaojie (Economic Operation and Economic Adjustment), Hongzhou, zhejiang renmin zhubanshe (Zhejiang People’s Publishing House), 1987. Zhang Yiyi: Exploring China’s Development of High-Tech Zones Through the Practice of Shenzhen Science and Industry Park, Science and Technology in China, Selection from the Bulletin of the Chinese Academy of Sciences (from the Bulletin of the CAS in 1988), Vol. 3, Beijing, Science Press, 1989. Zhao Wenyan et al.: Beijing xinjishu chanye kaifa shiyanqu fazhan zhanlue yenju baogao (Research Report on the Development Programme of the Beijing Experimental Zone for New Technology Industries), commissioned by the Beijing Zone for New Technology Industries, mimeo, March 1989. Zhao Ziyang: zhenxing jingji de yige zhanlue wenti (A Strategic Issue on Invigorating the Economy), speech at the National Conference of Awarding Science and Technology Achievements, in White Paper 1982, No. 1:280–290. Zhao Ziyang: zai quanguo kexue jishu gongzuo huiyi shangde jianghua, 6 March 1985 (Speech at the National Working Conference of Science and Technology), in White Paper, No. 1:293–297. Zhou Enlai: guanyu zhishi fenzi wenti de baogao (On the Issue of Intellectuals), 14 January 1956, in Zhou Enlai wenxuan (Selected Works of Zhou Enlai), Beijing, renmin chubanshe (People’s Publishing House), 1984. Zhou Jian’nan (ch. ed.): zhongguo dianqi gongye fazhan shi (A History of the Electrical Equipment Industry in China), Beijing, jixie gongye chubanshe (Machinery Industry Publishing House), 1990. Zhu Sendi: zhongguo jixie gongye yanjiu kaifa tixi de chongzu ji dui jingji fazhan de yingxiang (The Re-structuring of the R&D System of the Machinery Industry and Its Impact on Economic Development in China), paper presented at the UNU/INTECH workshop ‘The Restructuring of Industrial R&D Institutions in China’, Maastricht, June 29– 1 July 1994. Periodical literature Bulletin of the National Natural Science Foundation of China No. 1, 1987 Business Week China Electronics Newspaper (zhongguo dianzi bao) China Business Review Chinese Science News (zhonggou kexue bao) The Economist Electronics Commercial (dainzi shangbao) Electronics International (guoji dianzi bao) Far Eastern Economic Review Financial Times Forum on Science and Technology in China (zhongguo keji luntan) Information on Science and Technology of China (zhongguo keji xinxi) Nanhu Development Zone Newspaper (Nanhu kaifaqu zongheng), published by the Association of New and High Technology Enterprises, Shenyang Nanhu S&T Development Zone People’s Daily, Overseas Edition (renmin ribao haiwaiban) Power Plant Equipment (fadian shebei) Science and Technology Industry of China (zhongguo keji chanye)

280

BIBLIOGRAPHY

Science Research Management (keyan guanli) Science and Technology Review (keji daobao) Science and Technology Daily (keji ribao) Studies on Management of Science and Technology (keji guanli yanjiu) Wenhui Daily of Hong Kong (xianggang wenhuibao) WMEM: World Manufacturing Engineering & Market (shijie zhizao jishu yu zhuangbei shichang). A Quarterly of the China Machine Tool & Tool Builders’ Association, various issues

APPENDIX TO THE BIBLIOGRAPHY The R&D institutions visited (May to August, 1993) The Institute of Physics, of the Chinese Academy of Science (Beijing) The Institute of Chemistry, of the Chinese Academy of Sciences (Beijing) Beijing Automation Research Institute, of the Ministry of the Metallurgical Industry Wuhan Research Institute for Posts and Telecommunications, of the Ministry of Posts and Telecommunications Development Centre for Seawater Desalination and Water Treatment Technology, Number 2 Oceanographic Research Institute, of the State Oceanic Administration (Hangzhou) Zhejiang Institute of Mechanical and Electrical Engineering Design, Hangzhou, Zhejiang Province Zhejiang University (Hangzhou) The Zones and NTEs visited (May to August, 1993) Beijing Experimental Zone for the Development of New Technology Industries. This Zone is located in the capital city, Beijing. Government agencies and other service sectors are clustered there, and the intensity of universities and R&D institutes of Beijing is the highest in the country. This Zone was formally approved by the State Council in 1988, as the first Zone of the kind in the country, so it served as a model for other Zones in many respects. NTEs first emerged from autonomous initiatives taken by academics in the north-west part of the city, known as ‘Beijing Electronic Street’, beginning in about 1984. Wuhan Eastlake Development Zone for New Technology Industries and Wuhan Eastlake New-Tech Enterprise Incubator. This Zone is located in Wuhan city in central China, where a major conglomeration of both heavy and light industry and a number of R&D institutes and universities are concentrated. This city is an important base for the optical telecommunication industry and R&D in China. The Wuhan Incubator was the first and the largest one among the Centres for Scientific and Technical Entrepreneurs. Hangzhou High-Tech Industry Development Zone. This Zone is located in the coastal city of Hangzhou in south-east China. The industrial structure in the area has been very ‘light’, and industrial enterprises are generally smaller. Town and village enterprises have been growing rapidly since the late 1970s. Hangzhou is stronger in higher education: the prestigious Zhejiang University and a number of other universities are located there. This Zone is relatively small, and was founded later (in 1991). Shenyang Nanhu Science and Technology Development Zone. Shenyang is in the north-eastern part of China. The area has been one of the most important bases for heavy industry since the 1950s. These industries now face serious problems in the transformation of both their management and technology. A number of R&D institutes and universities are concentrated in the area. The Shenyang Zone is bigger in terms of NTE establishments and their total turnover.

In each Zone, two or three enterprises were visited, and aggregate data for the initiation of NTEs, as well as for the technological activities in NTEs were collected. Interview Notes (May to August, 1993)

BIBLIOGRAPHY

281

Interview Notes 1: Interview at The Management Commission, Hangzhou High-Tech Development Zone, ms. Bell, R.Martin Interview Notes 2: Interview at The Development Centre for Seawater Desalination and Water Treatment Technology, Hangzhou, ms. Bell, R.Martin Interview Notes 3: Interview at The Kangda Electrical Equipment Industry Ltd., Wuhan, ms. Bell, R.Martin Interview Notes 4: Interview with Vice-Director, Hubei Provincial Science and Technology Commission, ms. Bell, R.Martin Interview Notes 5: Interview at Wuhan Research Institute of Post and Telecommunications, ms. Bell, R.Martin Interview Notes 6: Interview at Automation Research Institute of Ministry of Metallurgical Industry, ms. Shulin Gu Interview Notes 7: Interview on the ‘Spinning-in’ of R&D Institutes to Enterprises (in the sectors of machinery building and microelectronics), ms. Shulin Gu Interview Notes 8: Interview at Zhejiang Institute of Mechanical and Electrical Engineering Design, ms. Shulin Gu Interview Notes 9: Interview at Wuhan Eastlake New-Tech Enterprise Incubator, ms. Shulin Gu Interview Notes 10: Interview at Haiyan Special Feed Mill, Zhejiang Province, ms. Shulin Gu Interview Notes 11: Interview on Classification of NTEs and their Technological Activities in the Beijing Experimental Zone, ms. Shulin Gu Interview Notes 12: Interview at Shenyang Development Zone for New Technology Industries, ms. Shulin Gu The R&D institutes visited (September to October, 1994 and April to May, 1995) Beijing Research Institute of Mechanical and Electrical Technology, the Ministry of Machinery Industry (MMI) Beijing Research Institute of Automation for Machinery Industry, MMI Dalian Modular Machine Tool Research Institute, MMI Grinding Machine Research Institute, Shanghai Machine Tool Works Machinery and Electrical Research and Design Institute of Dalian City Milling Machine Research Institute, Beijing No. 1 Machine Tool Plant Research Institute of Machinery Science and Technology, MMI (Beijing) Shanghai Power Equipment Research Institute, MMI Shanghai Internal Combustion Engine Research Institute, MMI Shanghai Electric Cable Research Institute, MMI The firms visited (September to October, 1994 and April to May, 1995) Beijing No. 1 Machine Tool Plant Dalian Machine Tool Works Shanghai Machine Tool Works Shanghai Turbine Works Shanghai Electric Machinery Manufacturing Works Shanghai Electric Corporation Interviewees (September to October, 1994 and April to May, 1995)

Mr Cai Weihua Prof. Chen Binmo Mr Chen Jun Mr Deng Zhaogui

Director, Senior Engineer, Machinery and Electrical Research and Design Institute of Dalian City, Dalian, Oct. 21, 1994. Member of S&T Committee, MMI, and Adviser to the Association of Electric Power Equipment Manufacturers, Beijing, Sept. 24, 1994. Deputy Chief Engineer, Shanghai Machine Tool Works, Shanghai, Oct. 12, 1994. Deputy Chief Engineer, Shanghai Machine Tool Works, Shanghai, Oct. 12, 1994.

282

BIBLIOGRAPHY

Mrs Dong Lijun

Senior Engineer, Research Institute for System Analysis, MMI, Beijing, Sept. 22, 1994. Mr Fang Guiru Deputy Director, Senior Engineer, Research Institute of Machinery Science and Technology, MMI, Beijing, Sept. 26, Oct. 5, 1994. Mr Fang Xiangwei Director, Senior Engineer, Shanghai Research Institute of Materials, MMI, Shanghai, April 14, 1995. Mr Gao Qingguo Director, Senior Engineer, Shanghai Electric Cable Research Institute, MMI, Shanghai, Oct. 13, 1994. Mr Gao Shiyi Deputy-Chief Engineer, Beijing Research Institute of Automation for Machinery Industry, MMI, Beijing, April 22, 1995. Prof. Hai Jingtao Director, Beijing Research Institute of Mechanical and Electrical Technology, MMI, Beijing, Sept. 28, Oct. 28, 1994. Mr He Wenli Senior Engineer, Department of Fundamental Machinery Equipment, MMI, Beijing, Sept. 27, 1994. Prof. Huang Fuhua Chief Engineer, Beijing Research Institute of Mechanical and Electrical Technology, MMI, Beijing, April 22, 1995. Mr Jin Yuling Deputy-Chief Manager, Beijing No. 1 Machine Tool Plant, Beijing, Oct. 6, 1994. Mr Li Baihuang Director, Senior Engineer, Beijing Research Institute of Automation for Machinery Industry, MMI, Beijing, Sept. 26, 1994, April 22, 1995. Prof. Li Fanhai Deputy-Chief Engineer, Beijing Research Institute of Automation for Machinery Industry, MMI, Beijing, April 22, 1995. Mr Lian Yuanjian Director, Senior Engineer, Department of Science, Technology and Quality Supervision, MMI, Beijing, Sept. 23, 1994. Mr Liang Zengbiao Chief Engineer, Dalian Modular Machine Tool Research Institute, MMI, Dalian, Oct. 20, 1994. Prof. Liu Caizheng Deputy-Chief Engineer, Beijing Research Institute of Mechanical and Electric Technology, MMI, Beijing, April 22, 1995. Mr Liu Dezhong Prof. Lu Chuluan Mr Pan Guishan Mrs Qiao Meirong Mr Qin Yunke Prof. Shen Tianxi Mr Wang Dechen Mr Wang Keng Mr Wang Zhiqi

Deputy-Chief Engineer, Beijing No. 1 Machine Tool Plant, Beijing, Oct. 6, 1994. Deputy-Chief Engineer, Beijing Research Institute of Automation for Machinery Industry, MMI, Beijing, April 22, 1995. Deputy-Chief Engineer, Dalian Modular Machine Tool Research Institute, MMI, Dalian, Oct. 20, 1994. Department Director, Senior Engineer, Beijing Research Institute of Automation for Machinery Industry, MMI, Beijing, April 22, 1995. Director, Senior Engineer, Research Institute of System Analysis, Ministry of Machinery Industry (MMI), Beijing, Sept. 22, 1994. Director, Shanghai Power Equipment Research Institute, MMI, Shanghai, Oct. 10, 1994. Senior Engineer, Beijing Research Institute of Mechanical and Electrical Technology, MMI, Beijing, April 22, 1995. Chief Engineer, Dalian Machine Tool Works, Dalian, Oct. 19, 1994. Director, Senior Engineer, Shanghai Internal Combustion Engine Research Institute, MMI, Shanghai, Oct. 14, 1994.

BIBLIOGRAPHY

Mr Weng Zuliang

283

Deputy Director, Senior Engineer, Shanghai Internal Combustion Engine Research Institute, MMI, Shanghai, April 18, 1995. Mr Wu Ye Deputy Director, Senior Engineer, Shanghai Turbine Works, Shanghai, Oct. 11, 1994. Mr Yu Chengting Deputy Director, Bureau of Fundamental Machinery Equipment, MMI, Beijing, Sept. 23, 1994. Mr Ye Dingda Deputy Head, Engineer, Instrument Division, Department of Fundamental Machinery Equipment, MMI, Beijing, Sept. 28, 1994. Mr Zhang Chaodong Deputy Director and Deputy Chief Engineer, Shanghai Electric Machinery Manufacturing Works, Shanghai, Oct. 11, 1994. Mr Zhang Gongyu Former Deputy Director, Bureau of Instruments, MMI, Beijing, Sept. 24, 1994. Dr Zeng Hong Head, Plan & Investment Division, Shanghai Electric Corporation, Shanghai, Oct. 12, 1994. Mr Zhou Zhang Director, Senior Director, Research Institute of Economics of Technology for Power Generation Equipments, MMI, Beijing, Oct. 28, 1994. Mr Zhu Sendi Director, Planning Department, MMI, Beijing, Oct. 27, 28, 1994.

INDEX

ABB (Asia Brown-Bowell) 174, 216, 222 abrasive apparatus 157, 158 academic community 52; incentives to 90; interactions between industry and 312–13; non-commercial centres 268; output 79; quality 79; R&D institutes 10; responses to Torch 72 accessories 157, 171; high voltage 233 accidents 216, 217 accumulators 161 achievements 38, 297 adaptations 200, 206, 212, 241, 319, 321; dynamic 339; imported production lines 230; local, support for 229; management software 273 administration 87, 88, 126, 152, 156; authorities 77, 174; coordination 48, 224, 240, 298; departmental 65; dysfunctional control 63; hierarchy 300; industrial 32; integral position in framework 10; intervention 225, 300, 331; limited power 16; local 85, 145, 165, 243; planning 126, 143, 150, 151, 166, 170, 168, 243, 273; provincial or municipal 328; regional 84; staff 245; ‘very idle’ 244; Zone 87, 88

aeronautical industry 83, 214 Africa industrial R&D 4 agencies/agents: charge of management 26; economic 169; financing 90; investing 87; monitoring 66; ownership 225; regulatory 19–20, 36, 51, 66; technology market 53; transactions 53 agricultural sector 65; machinery 132, 142–3, 152, 165, 227, 234–5; output 127; reform 7, 17, 48 air compressors 133 Allen-Bradley 206 alloys 155, 163, 171 aluminium 228, 231, 232, 297 Amann, R. 147, 149 Aoki, M. 300 application engineering 155, 269, 273, 307; automation systems 275–9, 280–1 appointments 21 Arora, A. 291 Arrow, K.J. 290, 299 artificial intelligence 272, 279 ARTMI (Automation Research Institute of the Ministry of the Metallurgical Industry) 23, 43–5, 46, 209, 336 ASICs (application-specific integrated circuits) 272, 276, 279 assembly 127; export-oriented 136; prototype 237; robots 276 assets 82, 91; 284

INDEX

channelled 88, 89–90; competitive, institute-specific 312; efficient utilization of 204; institute’s, organized part of 75; internally contracted 335; management of 195; manufacturing-related 211; physical 76; productive 321; re-allocation 89; rules which confirm market value of 90; R&D 51; safeguards to protect from abuse 335–6; ‘social-political’ 77; special 85, 90; specific 316–18, 319, 321, 322, 323, 332, 333, 334, 341–3; state-owned 195, 204, 215–16; strategic 302; technological 66, 76, 83, 184; transfer of 89 assets (contd.) assimilation 203, 218, 221, 231, 270, 310 Association of Food Packing and Processing Machinery Producers 245, 251 Association of Grinding Machine Producers 198 Association of Lathe Producers 198 Association of Machine Tools Producers 198 Association of Milling Machine Producers 198–9 AST (Americam computer manufacturer) 85 atmosphere-conditioning 244, 245 automatic operation systems 155, 212; small 100, 101, 107 automation systems 43, 102, 154, 162, 307; complicated 265; lower degrees of 230; micro-electronics-based 261, 271–82; technologies characterized by a moderate degree of 232 Automation Engineering Development Company 104 automobile industry 152, 195, 203–4, 205–6, 207; axles manufacturing 264; booming 210, 211–12; manufacture of engines for 209; parts 263; producers 236, 267; robot spraying production line 272; Soviet 264 autonomy 37, 74, 281, 301, 308, 334; decision-making 76, 89, 279;

285

delegated 221, 248; doing business directly with foreign companies 46–7; economic agents 292; enterprises granted more 126; management 52; restricted 336; R&D institutes 20–1; state-owned enterprises granted more 9; university 81; use of hard currency 109 auxiliary equipment/machinery 207, 209, 217; demand for 212; testing 221 Balazs, K. 6 banks 71, 104; decentralization 39; investment 88; local 90; technical and market advice to 87 bar-code scanning equipment 103 barriers 228–9; entry 110, 228–9; help to overcome 127; institutional 32, 51, 224, 225, 302, 325, 328; optimal integration 242; organizational 324; unique 336 bearings 127, 150, 171, 272 Beijing 68–9; Copying Machine Technology Research Institute 328; ‘Electronic Street’ 109; Huaxia Guigu (China-Silicon Valley) Information System Corporation Ltd 68; Kehai High-Technology Corporation 80; Legend Computer Group Corporation 76–7, 96–7, 99, 100, 104; municipal government 196; No. 1 Machine Tool Plant 150, 159, 204, 306, 311; No. 1 Machine Tool Plant Milling Machine Tool Research Institute 193–200; Research Institute for Machine Tools 157–9, 169, 170, 194, 211; Research Institute for the Electrical Apparatus Industry 163; Xinghe Electronic Co. 100; see also BRIAMI; BRIMET; also under following headings prefixed ‘Beijing’

286

INDEX

Beijing University (New-Tech) 85, 95, 97, 98, 99 Beijing Zone 73, 74, 78, 79–80, 83–5 passim, 109; computer ‘giants’ with home base in 104; technological activities 95–100; NTEs 102 Bentley, Raymond 6, 63 best practice 144 beverages 23 biotechnology/biological products 40, 92 blade technology 217, 218, 309 blueprints 183, 238, 264; designs from 238; construction 144; imported 148, 196; ‘stylized’ 143 boards and cards 104, 110; functional 95, 97, 100, 107 boilers 215; industrial 133; testing 161; three principal manufacturers 216 bonuses 30, 80, 267 boring machines 194 bottlenecks 195, 292, 293; communication 311 bounded rationality 319–20, 333 Branch Productivity Centre for Machinery Technology 248 branch structure 129–32 Brazil 129, 131, 132, 133, 136, 138, 295 BRIAMI (Beijing Research Institute for Automation in the Machinery Industry) 155, 260, 261, 271–80, 282, 327 BRIMET (Beijing Research Institute for Mechanical and Electrical Technology) 154, 155, 260, 261–71, 279, 282, 327 budgets 82; hardened constraints 91; national, funds financed from 148; soft constraints 322; suffered significantly from inflation 79 built-up jigs 157, 159 bureaucracy 324; bottom-level 89; central planning 88; complications 251 buyers: inexperience of 50; international 294;

technological market not very pleasant for 62 CAAS (Chinese Academy of Agricultural Sciences) 146 cables 163, 227–34; communication 233 CAD (computer-aided design) 195, 202, 203, 207, 262, 271, 273 Cai Weihua 244, 246 CAMS (Chinese Academy of Medicine Sciences) 146, 262, 271, 273 capabilities 312; design 309; inferior 331; missing 282; testing 309 capacity 125, 194, 223, 293, 297; demand exceeds 275; excess 229; expansion 129, 142–3, 248; power generation 172; processing 232; standardized 142; upgraded 194 capital: appropriate resources 71; ‘construction’ 10, 207; foreign 38, 73, 76, 84, 85; formation 125; human 228–9; increases in stock 133; initial 76, 82; investment 237; local administrative authorities widely involved in providing 77; mobilizing 89; public 82; risk 39; shortage of 127; see also venture capital capital goods 125, 132; complex production 295; industry highly significant for industrialization 132; investment in 324–5 capitalism 316, 333 CAPP (Computer-Aided Production Planning) 203 CAS (Chinese Academy of Sciences) 16, 35, 47, 65, 83, 98, 146, 166; Dalian Institute of Chemical Physics 244; Institute of Computer Technology 76, 96;

INDEX

Institute of Physics 68; joint initiative with Schenzhen Municipal Government 69; Physcience Optoelectronics Corporation of the Institute of Physics 79–80 cash flows 241 castings 155, 197; resin-bound sand 247–8 catalyst reactors 244 central government: ‘A’ institutes owned by 156; appropriations urgently needed 4; approval of status as ‘National Level’ 72; direct control being withdrawn 126; fostering communications between producers and 198; funding 62, 165, 177–81; institutes financed by 163–4, 200; institutes no longer affiliated to 177; investments for upgrading from 207; policy 38, 53; priorities set by 87; subsidies 194, 202 centrally planned economies 3, 52, 88, 145, 204, 295; administration 309; authority exerted on behalf of 141; capital investment under 143; ‘Chinese version’ 89; considerable success initially achieved under 299; coordination 197; development of machinery industry under 151; efficiency 4; enterprises specialized by products 156; former 62–3; innovation 169, 193, 196; institutional inheritance of the R&D system under 9– 10; instruments of 87; introduced 8; OECD studies in 6; policy measures restricted to supplier side 48; prices set 229; priority to development of ‘heavy’ industries 129; productive enterprises under 150; regular scheme required under 149; resources obtained by direct coordination by 173; stated goals 298; transplanted from Soviet Union 9 centrally planned economies (contd.)

287

Centres (Service Centres for Scientific and Technical Entrepreneurs) 38, 71, 84, 85, 89; development of 53; widespread establishment of 85; see also Wuhan ceramics 238 Changchun 73; Automobile Research Institute 224 Changsha: Electrical Auto-apparatus Works 162; Research Institute for Electrical Automobile Apparatus 162 Chang-Fei Company 101 chemicals 83, 185; equipment 127, 132, 142, 148; fine 24, 40, 92; new 25 Chen Binmuo 163, 224 Chen Chunxian 80 Chen Jiyuan 10 Cheng Shuren 127, 132, 141, 142 Cheng Chunxian 68 Chengdu: Research Institute for Electric Welding Equipment 162; Research Institute for Tooling and Tools 158 China Computer System Engineering Corporation 97 Chinese Character processing technologies 76, 95, 96, 97– 8, 99, 102, 107, 110; compiling and publishing system 107; graphics 100 Chinese Power Engineering Society 215 Chinese Standardization Commission for Forging Die Technology (National Technical Committee) 262 Chinese Standardization Commission for Heat Treatment (National Technical Committee) 262 Chinese Standardization Commission for Hydraulic Pneumatic Technology (National Technical Committee) 272 Chinese Standardization Commission for Industrial Automation Systems (National Technical Committee) 272 Chinese State Science and Technology Commission 246 Chinese-speaking economy 108 Chudnovsky, D. 118–19, 127 CIMAC (International Council on Combustion Engines) Chinese National Committee 234 CIMS (computer-aided integrated manufacturing system) 195, 273, 274 cities: coastal 109; medium-sized 26;

288

INDEX

rivalry between 90; small 227 CNC machines: lathes 201; machine tools 140, 171; milling 194 coal mining 185 Coase, R. 340 codification 312; investment in 311 cold-store sector 244, 245, 249 colleges 86 commercialization 20, 29, 38, 39, 70; financing 71, 90 ‘commune system’ 8 communication 198; bottlenecks 311; close 183, 267; continuous 219, 267, 327; effective 267; horizontal 303, 311; infrastructure 38; interinstitutional 224; internal and external 312; intimate 219; producer and users 299; repeated 147; two-way 237 Communist countries (former) 62–3; see also Eastern Europe; Soviet Union comparative advantages 99, 277 competence 243; core 110; design 301; management 77; production 301 competition 276; commercial 306; domestic 216; excellence-based 20; fierce 23, 50, 107; foreign suppliers 23; increasing 305; international 195, 300; market 200, 301 competitive advantages 97 competitive products 93, 99; inventory of 102;

producers’ strong links with R&D institute or university 103; typical producers of 104 competitive tendering 145 competitiveness 44, 70, 104, 239, 277; computers 96; established and tested 213; higher-order 300; inferior 295; international 97, 300; locally specific 300; sensitivity to 302; serious threat to 278; significant level of 281; strengthened 241 complementary technologies/ techniques 85, 266, 310 components 93, 152; access to 85; application-purpose oriented 279; basic 110, 277; computer 96; core 110; electrical 159, 163; electronic 307; producers 235; special-purpose 155 computers 40, 73, 69, 92, 93–5, 195, 212, 311; accelerated development of applications 108–9; applications of technology 99; Chinese writing input 85; control, internationally available 276; dealers 100, 102, 108, 109; English-language based technology 107; imported 96, 104–5, 109; industrial 275, 276; integrated manufacturing technology 155; internal divisions 104; language localization or ‘Sinologization’ 107; manufacturing technology aided by 148; market for 102; producers of 100; RAM expansion 97; retail and user services 73; selling 36; services for applications 20; state-owned enterprises 76–7; technology based on 291; transnationals in 85; weaknesses in application of technology 202;

INDEX

widespread dissemination/ application of technology 35, 266; see also PCs; software conducting materials 228 construction: industrial 146; plant 144, 171; power station 226 consultancy 27, 50, 184, 187, 188, 207, 211, 220, 226, 230, 235, 238, 240, 249, 259, 260, 281, 331–2; centres 46; earnings/income from 181, 221, 232, 236; engineering services 145; sales counted as 231; work for imported automation lines 275 consumer goods 133, 206 consumption: electricity 221; energy 133 ‘Contemporary China Series’ 126 contracts 9, 21, 100, 145, 227; big 44, 279; breaches of obligations 25; direct 251; diversification of activities 240; duration 183; effectiveness, in the provision of research 63; ‘enduring’ 183; external 226, 242, 315–44; incomplete 316–18; internal 315–44; laws for 19, 20, 37; longer term 267; low payment 22; management 9; outside, earnings from 33; period 183, 185; relationships 226, 303, 305, 315–44; reliance on 188; responsibility 268; spot 183, 184; technology development 50, 187, 221, 259–61, 267; technology transfer 240; trends in activities 207–9, 219–21, 230–2, 236–8, 263– 4, 273–5; value of 26, 27, 50, 275 contractual responsibility system 195, 213, 226, 227; internal 233, 248, 250, 279, 334, 336, 343 control devices 277

289

controls rigid 50; vertical 20 cooling agents 171 Cooper, J. 147, 149 copper 228, 232, 297; oxygen-free cable technology 231 core business 213, 227, 239, 242, 244, 250; clearly delineated 248 core technology 76, 77, 110, 323, 324 ‘corner branches’ 103 corporatization programme 196 costs 46, 151, 204; bureaucratic 320, 324, 328, 329, 331; coordination 149; engineering, affordable 267; governance 318, 320, 321, 329; investment 160, 320; lower/low(est) 265, 292, 293; planning management 151; production 299, 319; simplifying structures to reduce 96; transaction 51, 226, 315–44, 324, 325; transitional 225 costs (contd.) ‘counties’ 26 craft approach 278, 281–2 credit 71; social and political 66 Cultural Revolution (1966–76) 8, 12, 126, 148, 154, 201, 243; merged factories 222; scientists and engineers denounced and dismissed 13 currency: foreign 21; hard 15, 109 customization 237, 238, 267; engineering services 303; forging systems 327 cutting 283–4, 297 Czechoslovakia 172 Dahlman, C.J. 294 Dalian: Centre for Machinery Product Quality Inspection 245; Machine Tool Works 201–5, 206, 208, 210, 336; Machinery and Electrical Research and Design Institute 145, 243–6, 249, 250, 251, 337; Municipal Government Bureau of the Machinery Industry 243–6; Research Institute for Modular Machine Tools 158, 201, 202, 203, 205–13, 278, 311, 327

290

INDEX

David, P. 290 debugging 263, 275 DEC technology 97 decentralization 65, 242; banking system 39; commitment to 91; decision-making 52, 89, 145; delegation 38; emergence of NTEs and 88–91; firm structure 228; internal structure 250; planning authority 151; stimulated 8 Decision on S&T Management System Reform (1985) 7, 9, 17–28, 29, 35–6, 62, 321 decision-making 10, 43, 279, 299, 306; absolute power in 245; adaptive 291, 293; autonomous 76, 89, 279; capital investment 143; centres of institutes 325; decentralization of 52, 89, 145; delegation of 43, 105, 301, 308, 334; operational 334; power re-allocated 213; re-centralized 43–4; shared by central and local governments and enterprises 145; strategic 334, 336 defence equipment 148 delegation 20–1, 23, 38; autonomy 37, 46–7, 248; decentralized 38; decision-making 43, 105, 301, 308, 334; responsibility and rewards 90 delivery times 311 demand 4, 104, 129, 147, 191; aggregating 318; auxiliary machinery 212; better production technology 229; circumscribed 236, 237; CNC machine tools 140; computers 36, 110; enterprises not able to meet 27; exceeds capacity 275; imports of automation systems 281; industrial 170; inferior, engineering services restricted by 332; integrated systems 274;

investment 72; less-expensive 238; local, specific 99; low grade 238, 241; manufacturing automation 280; market 99; massive, from investors 143; more complete outputs 279; networking 97; plant design services 247; ‘real’ 263, 267; shift in 195; small producers 230; sophisticated 332; strong potential 103; technologically sophisticated high quality products 229; user 277, 331 demonstration effect 306, 308 Deng Xiaoping 13 Denmark 246 Department of Science and Technology 147, 151, 152 design 78, 118, 201, 219, 263, 272, 279; abilities 45; accumulated skills 289; adaptations 170; automation systems 101; basic 132, 172, 298; blueprints 148, 149, 238; boilers 161; capabilities 309; competence 301; computers 96, 101, 107, 110; concept, evaluation of 237; creativity in 281; die 263; doing work on the manufacturing site 205; electric and electronic 196; electrical equipment sector 160–3; engineering 43, 110, 228, 241, 323; enterprises encouraged and stimulated to establish their own units 166; foreign involvement in 44; foreign, localization of 132; fossil-fuelled power plant equipment 216; good, hardware outputs based on 108; imported 34, 202; in-house R&D 165; institutes 10, 157–8, 160–3;

INDEX

key equipment 270; key personnel 44; lack of systematic scientific data for 216–17; machine tools and other tools 157–8; mature, product diversification based upon 299; modular unit 212; new 25, 148, 149, 150; novelty in 276; off-the-shelf 238, 240; operational performance of 290–1; partly modified (original) 132; preliminary 148; shortcomings in technology 172; sophisticated questions in 173; standards 206; systems 276, 281; technical 148; techniques 146; technology 218; unit 294, 306; user-specific 277; weaknesses of domestic technology 174; see also plant design Development Centre for Seawater Desalination and Water Treatment Technology 80 Development Zones for New Technology Industries see Zones dies 155, 263, 268; computer-assisted design 262; forging 261, 264–5, 266, 267 diesel engines 133 diffusion function 251 ‘dissemination and application’ 310 distribution: computer 109; market system 240; planned personnel system 21; sectoral, R&D 65; technological activities of NTEs 93 disturbances 292, 318 diversification 267–8; contractual activities 240; technological activities 227, 312; minor 15; see also product diversification diversity 8, 53, 293 division of labour 66, 278, 311 documents 196, 202, 238; complicated 149;

291

investment in 311 domestic market 44, 70, 74, 78, 96, 97, 198; computers 98, 102, 105; electronics 103; forging manipulators unavailable on 266; identifiable brand profile in 276; information technology 102, 105; local suppliers captured most of 244; production equipment for electric cables and wire 229; share of 233; still dominated by relatively low technology products 227; technologies competitive on 232; turbines 215; unified 102, 105; warehouses 275 Dosi, G. 289–90 drainage 235 duplication 229 Du Pont 336 Du Xueguo 233, 234 earnings 248, 267; contractual 33, 37, 243; ‘horizontal’ 41; market 41, 210–11, 220, 230, 236, 259, 263, 269, 273– 4, 280; redistribution 233 East Asia 294 Eastern Europe 4, 89, 142, 340, 342 EC (European Community) 246 economic growth 48; average annual rate 133; driving forces behind 89; efficiency of 8, 48; entry and 127–9; rapid 89 economic reform 110; driving forces behind 89; rural 8–9; spin-off enterprises 61 ‘economies of production and operation’ 302 economies of scale 172, 297 education: backgrounds 72–3; higher 29, 65, 83 efficiency 219, 308; centrally planned systems 4; cost-benefit 303; economic growth 48;

292

INDEX

energy 133; innovative efforts 311; investment 91; loss of 175; manufacturing 297; market 50; much greater, need to achieve 278; rational economic agents 316 electrical equipment/machinery 108, 135, 152, 157, 159– 63, 165, 209, 223; cable and wire 191, 211, 227–34, 297; ceramics standardization 150; drives 162; explosion-proof apparatus 162; furnaces 162; hotel devices 247; institutes 163–4, 173; low-voltage devices 247; materials 159; motors 133, 150, 162, 161, 171, 247; power 148; small apparatus/devices 162, 227, 235, 247, 251; spark machines 158; specific facilities for 163; traction 162 electrical equipment/machinery (contd.) Electrical Power Industry 173 electricity generating equipment 159, 169, 174, 191, 235, 263, 297; development of products 214–27; major power source for 172; transmission and distribution 161 electro-physical apparatus/instruments 279 electro-transistor devices 163 electronics 120, 140; experts 196; fabrication 108; industries 31, 32, 83, 110; integrated devices 100, 101–2; managers 31; merging 32 Electronics Street 69 ‘elementary equipment’ 263, 265, 266, 267, 270; forging systems based on 307 employment: previous, professional experience and knowledge gained in 86; second jobs 21 energy: conservation technology 148;

efficient machinery 133 engine producers 238 engineering 163; ability 238; cable plant 227–34; civilian survey institutes 141; commercialized services 269; computer 98; computer-assisted 271; consultancy services 145; contractual projects 44; corporations 43, 47, 52; courses for undergraduate and graduate students 166; design services 145; engineering cold-store 244, 245; fundamental research 268; generic 275, 312; intensive reverse 170; know-how 185; manufacturing systems 284–6; mechanical 155; metallurgical automation 44; packaged 265; plant 215, 231, 239–40, 246, 302–3; process 118; product 118, 256–7, 284–6; productivity development 277–9; sector 65; small power plant 221; surface 155; survey institutes 141, 144; trial 311–12; turn-key 141; users 206; see also application engineering ‘Engineering Centres’ 212, 312 engineering services 108, 213, 226, 230, 237, 249–50; automobile producers 236; customized 317; greater engagement in 188; idiosyncratic character 323; imported technology underlying 212; income from 183, 211; integrated plant 145; key businesses developed in 233; marketing 213; mixed 210; movement towards 187; plant 238;

INDEX

restricted by inferior demands 332; specific 331; technology market 239–41; tomorrow’s 232; transactions for 318 engineers 76, 118, 119, 201, 203; Chief 198, 268, 279; Deputy Chief 268; senior 196, 205, 214, 234, 246, 262 English language 97; computer and information technologies based on 40, 107 entrepreneurship 85, 89 entry 250, 330; barriers 228–9; growth and 127–9 environmental factors 235–6; devices which may cause risk to users or damage 247; regulations 241, 332; technology 162 equity 71, 216; local joint ventures 223; physical 35; shared 78 Ernst, D. 110, 112 EUMUCO (German company) 265, 268 exchange rates 127, 194 exhaust emission 235 expansion: capacity 129, 142–3; duplicative 144; engineering services 188; higher rate of 291; plant design institutes 144–5; production 297; quantitative 143; rapid 36; R&D institutes 151 expenditure 166, 202; R&D 65, 238, 310; technology imports 140; testing 238 experiments/experimentation 170, 173, 237, 293; ‘development’ 310; indispensable 290; institutional 310; intense 259; laboratory 209, 278; more open 50;

scientific 202, 217, 262; workshops 149 expertise 80; commercializing 70; computer 81; equipment design 270; external, heavy reliance on 197; internal software 104; technological innovation 251 experts 36, 39, 194, 197, 217, 275; business entities initiated by 70; electronics 196; Soviet 148; S&T, autonomous initiatives by 68; working on application software 276 exploitation 30 exports 76, 136–41, 343; modest success in 207; packaged technology projects 145; turbines 215, 216 externalization 10, 48; extensive 3; partial 45 ‘factor driven’ stage 300 Fang Yi 14 farmers 18 feasibility studies 144, 145, 246, 294, 306; reports 86 feedback 269, 278; information 5, 299 fees 148; contractual 229, 268 fertilizers 133, 142, 221 fibre optical cables 233 final/finished products 70, 183, 185, 210, 213, 240; assembly of 108; generation and dissemination 188; manufacturers 235; marketing 213; operating economy of 294; sophistication 212 financial institutions 53 financing 33, 38, 87; capital investment 145; commercialization 90; public 20; sources of 88 fine blanking 262

293

294

INDEX

First National Food Packing and Processing Machinery Fair (1993) 245 Five Year Plans 12, 14, 141, 146; key S&T projects 15, 20, 148, 207, 218, 219, 272, 273, 276, 297, 310 FMS (flexible manufacturing systems) 272–3 FMT (UK Company) 206 food processing/packing/storage 23, 133, 244, 247 Foray, D. 290, 308, 313 foreign companies/other issues 225; complete control 222; doing business directly with 46–7; imitating devices 244; partners 210 foreign suppliers 53, 107, 229, 267, 275, 277; capable 25; intimate interaction 308; large system initially contracted by 44; local developer advantage over 281; opening to 50; reliance on 211; sophistication of 212 foreign technologies 99, 208, 235; acquisition of 216–19; alliances 172; large-scale procurement of 15; opening to 192 forging 269, 270, 273, 298; customization of systems 327; die 155, 261, 264–5, 266, 267; important equipment for 268; knowledge of the process 266; leading institute for 262–3; machinery 157, 158; rolling 264, 265; systems based on ‘elementary equipment’ 307; user specificities 277 forming 283–4 forward integration 323, 324, 331 fossil-fuelled power plant 171, 215, 216 foundry technology/machinery 155, 157, 158, 261 4S Typesetting Systems 98 France 44, 144 Frascati, Manual 270, 282 Freeman, C. 5, 62–3 funds/funding 20, 41, 62, 195, 208, 220, 229, 238, 244, 273, 321; bottleneck investment 277; capital construction 207;

external 325; inability to raise 222, 226; inadequate 242; investment 195, 238; ‘leading’ 38, 39, 71; necessary for market-oriented transformation 209; project 79, 207; rescue 241; small amount of 248; ‘special’ 148; S&T 39; tapping new sources 213 funds/funding (contd.) Fushun Research and Design Institute for Chemical Engineering 24 Gambardella, A. 291 Gao Qingguo 229, 230, 232, 233, 234 Gao Shiyi 276, 278, 280 gas(oline) 185, 221 gear transmissions 155 General Institute of Coal Mines 22–3 ‘generic technology’ 146, 154; automation systems 280–1 Germany 44, 131, 194, 196, 202, 204, 206, 265, 268; East, ‘research companies’ 6 globalization 136 GNP (gross national product) 10, 14, 127 government see central government; local government graduators 170 grain 8 Granick, D. 151 grants 19, 207 graphic processing 95, 97–8, 99, 107 ‘Great Leap Forward’ (1958–60) 8 Grinding Machine Tool Research Institute 159, 199 grinding machines 157, 158; bearing 272 Gu, Shulin 120 Guangzhou: Research Institute for Machine Tools 158, 211; Scientific Research Institute for Electric Apparatus 162 Guilin Scientific Research Institute for Electrical Apparatus 163 Guo Rui 151 Hai Jingtao 262, 263, 264, 268, 269 Hainan 72

INDEX

Hangzhou: Research Institute for Industrial Steam Turbines 161; Zone 83, 84, 88 Harbin 173; Electric Carbon Works 163; Electric Generator Factory 161; Electrical Instrument Works 163; Power Equipment Research Institute 161, 328; Research Institute for Electrical Instruments 163; Research Institute for Electrocarbon Devices 163; Research Institute for Large Capacity Generators 161; Research Institute for Welding 155, 260 hardware 22, 100, 101, 107, 231; computer, ‘comodification’ of 110; imported technology 98; mass-produced 108; outputs based on good design 108 hazards 319, 320, 322, 329 H-B crisis 342 health care sector 65 heat treatment 155, 243, 261, 266, 269, 270, 273; leading institute for 262–3; vacuum 262 Hena Province 154 He Wenli 210 H-form (Holding company-like form) 332, 336, 337, 339; successful ‘initial’ 343; U-form to 333–5 hierarchies: administrative 65, 300; ‘market’ and 5, 315; re-building 43 Hipotronics Cable Testing System Service Centre 229 History of the Electrical Equipment Industry in China, A 126 Hobday, M. 294 ‘holding companies’ 334 Hong Kong 44, 76; computer brokers 97; Legend Technology Ltd 77; middlemen 109; small companies 276 Hongzhou Development Zone 247 Honsberg 204 Hope Corporation 107 host enterprise 152, 156, 164, 197; activities 178; institutes incorporated/ integrated into 177, 199; main technological strength 199;

295

merging with 325; served by imports 200 ‘household responsibility system’ 8 Huang Fuhua 263, 264, 265, 268, 269 Huangpu River 233 Huller Hille (German company) 202, 206 humidity control devices 244 Hungary 5–6 hydraulic cylinders/technology 247, 248, 279 hydro-electric generation equipment 161 IC chips 96, 97 ideas 77 IEC (International Electrotechnical Commission) Chinese National Committee 228 ‘imbalances’ 292 imports 136–41, 192, 248, 281; affairs relating to 30; computer 96, 97, 109; decisions on 126; designs 34; direct 206; host enterprises served by 200; industrial technology 8; intensive 273; interpretation and testing of 199–200; large-scale 31, 127; less successful than hoped 203; major form of 295; pervasive 235; plant design learned through 142; power plant equipment 172, 216; Soviet, of machinery products 151; substitution 98; upgrading through 194, 202; see also technological imports in-house activities 166; computer development 109; research institute for water-cooled electric generator technology 173; R&D 10, 52, 63, 165 incentives 4, 80, 89, 250; establishment of 36; fiscal 71; ‘high-powered’ 334; local governments 90; policies 19, 37–8; set up by Decision 36–7; strong(er) 23, 336;

296

INDEX

structures 241, 299, 301; weak 308 income 44, 207, 220, 232, 245, 274–5, 276; composition of 109, 236; consultancy 221; contractual 267, 269, 279; conventional products and services 41; engineering services 211; family 129; institute supplementing 79; market 263, 273; national 127; NTEs 73; RIMST institutes statistics 260; structures 73, 177–90; technology development 230; trial production 211, 249; uncertain 22; see also earnings ‘Incubators’ 78, 86, 87, 99–101 index plants 170 India 129, 131, 132, 133, 136, 138, 295 industrial fans 150 industrial sector 17, 31–2, 65; reforms 7, 48; R&D institutes 146 industrialization 127; ambitious plan 14; capital goods industry highly significant for 132; international market-oriented strategy for 69; ‘self-reliant’ 129 inefficiency 22–5, 29, 48 infant industry 149, 286–7 inflation 79, 80, 180 information 53, 165, 196, 228; combination of skills and 64; continuous inputs from users 185; disseminating 198; exchanges 147, 173; feedback 5; intensive and continuous inputs of 183; market 198; most important flows 300; processing 150; producer 150; production-specific 200; providing 243; recent domestic and international developments 237; sharing 200;

specifically identified 293; vanished or reorganized 245 information technology 61, 92, 93–5; development 110; domestic market unified for 102; English-language based 40, 107; incorporating computer and 108; NTEs engaged in 118; transnationals in 85; user-specific applications of 36; widespread application of 266; widespread dissemination of 35 infrastructure: communications 38; evaluation 247; high standard 73; physical 38; regulatory 37; scientific 117; social 110; technological 46, 117, 250–1, 256, 307; Zones’ investment in 39 Ingersoll 219 ‘initiator and supervising unit’ 82–5 innovation 118, 203, 231, 302, 308–10; central planning system 157, 160, 193, 196; ‘chain-linked model’ of 119, 300; commercially successful 64; complex 168–9, 174; engineering 279; extended 156; ideas 77; individuallyapproved projects 259; industrial,scientific community never really linkedto 313; institutional 36; made only atthe initiative of the planning authority 165; management procedurefor 148–51; M-form 335–6, 343–4; only source of initiatives and coordination required for 169; planning for 152; problem intrinsicto 50; process 294; productivity 311; source of information for 194–5; successful 292; technological 219,251, 311;

INDEX

uncertainty of 5, 25; weak incentives for 119; weaknesses preventing keeping up with pace of 203 innovation (contd.) inspection 247, 251; automatic 272 installations 76, 80, 100, 145; complementary technologies for 263; debugging after 101; on-the-spot 108; physical 317 Institute of Technical Economics for the Electrical Industry 159 ‘instrument effect’ 306, 308 instrumentation 108, 119, 152, 155 instruments 217, 227 insulating materials 163, 228 INTECH study 126 integrated systems 274; circuits 95; electronic and mechanical 92, 99, 102–3 intellectual property 37–8; avoiding conflicts 86; inadequacy of legal protection for 4 internal combustion engines 211, 227, 234–9, 241, 339 internal organization 270–1, 281–2, 332–9; changes in 213, 226–7, 242, 250; decentralization 332 internalization 5 international exchanges 126 international markets 70, 76, 78, 90, 194; competitiveness in 215, 232; computer 98 internationally tradable technology 306–8, 309 investment 30, 108, 207, 208; agricultural machinery 142–3; allocation and re-allocation of 48; capital 9, 39, 66, 91, 133, 141–5, 146, 171, 263, 324– 5, 327; change in priority of 33–4; decisions on 44, 126, 204; demand for 72; efficiency of 91; quity-related 209; follow-up 224; foreign 37, 73; funds 195; indirect 229; industrial 300;

297

inflow of 9; institutions’ policies for 38–9; interface functions of Zone Administrations in securing 88; internal 204; international 136; joint venture 217; Key S&T Projects 310; limited 233; made blindly 22–3; necessary 332; new 142; not necessarily very high 77; planed 259; production requiring smaller 160; public 82; relatively moderate 325; return on 144, 241; R&D 62–3; special purpose 316; specific 331; SSTC 173; state 41; substantial 193, 211; taut 143; two groups of factors influence 292 involute cam formworks 170 irrigation 235 ISIC (International Standard Industrial Classification) 125, 131–2, 136, 140,151 ISO (International Standards Organization) 43, 272; Technical Committee of the Chinese National Committee 234 Italy 229 Japan 81, 131, 206, 275; horizontal interactions 300; quality assurance 81; software exports to 81; see also MITI Jiamusi Research Institute for Explosion-Proof Electrical Apparatus 162 Jiangsu Province 158 Jin Yuling 199 Jinan: No. 1 Machine Tool Works 273, 274; Research Institute for Foundry and Forging Machinery 158, 260 Jing Xiaocun 126, 127, 132, 133, 141–50 passim, 159, 163, 172, 235, 239, 297

298

INDEX

Jinghai 69, 78 joint ventures 76, 85, 101, 216–19, 233, 279; local, equity-involved 223; software 81 Kangto 98 Ke-hai 69 key devices 244 key productive enterprises 322 Kim Linsu 143, 294 Kline, S.J. 119, 308, 309 Knight, J.B. 344 know-how 35, 76, 86, 185, 231–2, 294; accumulated 270; application systems 278; atmosphere-conditioning 244; commercializing 70; core 263, 266, 271, 307; design 311; engineering 185, 196, 307, 323; evaluating 25; material 230; output in which it is embodied 181–3; process 270; R&D institutes moving to capitalize on 29; well-codified 273 Kodama, K. 308 Kornai, J. 143 KTM (UK company) 202 Kuhn, T. 289 Kunming Scientific Research Institute for Electric Apparatus 162 laboratories 79, 100, 209, 278; central 222; in-house 153, 156; novelty 278; research 24; testing 146 labour 230; cheap(er), skilled 44, 278; increased input 133; intellectual 63; intensity 69–70; mental 18; productivity 133, 135; redundant 223; surplus 271, 280; see also manpower

labour market 225; fully-developed 21; normally developed, lack of 37 land 8 language localization 108 Lanzhou: Design and Research Institute for Diesel Powered Vehicle Engines 160; Oil Machinery Institute 226; Oil Machinery Research Institute 224, 225 large-and medium-sized enterprises 36 lathes 201 Law of Technological Contracts (1987) 19, 20 lead screws 170, 171 leadership 76, 148 learning 109–10, 118, 175–6, 226, 310; accelerated 90; accumulation of 273; ‘by doing’ 290, 293; effective 308; ‘institutional’ 319; intensive 241, 309, 327; product design 219; specific 298–300, 302 ‘leasing measure’ 9 leave 21 legislation 91 Lian Zengbiao 210, 213 Li Baihuang 271–5 passim, 278, 279, 280 licensing 71, 79, 85, 104, 140, 196, 217, 236; company 247; export and import 37; technology transfer 204, 216–17 Li Fanhai 275, 277, 278, 280 light industry 185 Li Jingwen 133 Limited Liability Corporations 78 Liu Caizheng 263, 264, 265, 268, 269 Liu Dezhong 1988 Liu Jirei 105 Liu Jiren 81 loans: appraising 39; bank 38, 77, 87, 88, 145, 179; commercial 38 local government 84, 88, 210, 225; active in spinning-off 52; autonomy 89; coordination 103;

INDEX

incentives to 90; institutes affiliated to 165, 177, 187, 193, 242–51, 323; institutes owned by 156; management of 210; participation in Zones 65; policies 39, 90; responses to Torch 72; role 85 local materials 230 lubricant oils 171 Lu Chuluan 269 Lu Jianhong 248 Lundvall, B.-A. 175 Luoyang: Mining Machinery Research Institute 224; Tractor Technology Research Institute 224, 328 machine tools 102–3, 108, 142, 150, 152, 157–9, 163–4, 193, 217, 249, 331; advanced metalcutting 158; automated 147; CNC 212, 308; exports and imports 136–40; modular 158, 201–13; precision 132, 168, 169–71, 297; transfer-line 209 machine tools (contd.) machinery 93, 103; industrial control 95, 98, 104; integrated 100, 101–2; one of the most important bases of the industry 102; R&D institutes 115–344 Machinery Engineering Society of China Heat Treatment and Forging divisions 262 Machinery Industry in Contemporary China, The 126 Machinery Engineering Society of China Automation Technology Division 272 Machinery Industry Ministry 173 magnetic materials 163 maintenance 78, 108, 127, 150, 235; imported manufacturing and testing facilities 230 Malaysia 110 management/managers 53, 132, 268; agencies in charge of 26; centralized 204, 226, 233, 245, 250; chief 71, 84, 198, 204; competence 77; computer-aided 271; contract 9;

299

deputy-chief 204; experience 76, 81; firms partly owned and run by 89–90; Incubator 86; industrial ministry 147; inferior 238; information systems 273; innovation 149; institute 52; internal 91, 204, 226; international 70; local governments 210; malpractice 86; M-form institute or enterprise 336; NTEs 91, 102; planned projects 15–16; planning 148, 150, 157; plant design 143–4; procedure for product information 148–51; procedures 108; production 217, 271, 294; regulators or monitors of 83; R&D institute 14; skills 85; Soviet system 142; state-owned assets 195; techniques 218, 219; technologies for 194; trial production facilities 209; Zone 81, 82, 90, 92, 93, 98, 100, 102 manipulators 265, 268, 307 manpower 65, 70, 76, 119, 154, 199; allocation of duties 243; best-educated 89; better-trained 212; engineering 117, 232, 242, 250; involved in spinning-off approach 118; scarce 141; scientific 117, 146, 226, 317; skilled, lack of 127; splitting up and reintegrating technological activities and 233; superior 203; technical/technological 247, 317; well-trained and experienced 208 manufacturing 23, 43, 47, 117–18, 129–32 passim, 165, 170, 246; abilities in 45; advanced systems 208;

300

INDEX

automated systems 298; automobile 34; big and heavy 102; cable 231; commercially competitive systems 294; complex systems 27, 303–7, 327, 343; computer 101; concentrated 247; conditions 237; ‘conventional’ process 270; cooperative 140, 217; cutting-edge 232; development of electronics industry centred on 110; development work in enterprises 172; direct input in terms of value 223; effective 293; efficiency of 297; engineering 323; export-oriented 136; flexible 195, 201–3 passim, 207, 209, 307; foreign involvement in 44; high-quality 294; inferior quality 217; innovation process 119; integrated systems 270; international contractual 294; key 226; large enterprises 301; leading domestic 212; mainframe machine 108; market-profitable 11, 52; need to rectify devices to meet specifications 196; numerous users 160; peripheral equipment 108; poor facilities 238; precision improved 171; productivity level 133–5; prototype 149; selected 232, 233; single-product 334; skills and methods for 294; small batch 148, 242, 302; sophisticated questions in 173; trial 148, 149, 205; turbine 215; US 135 manufacturing technology 151, 155, 194, 218; advanced 206, 207, 213; centrally affiliated institutes for 153–68;

complementary 219; computer-aided 148; computer-integrated 155; development 200; domestic, weaknesses of 174; flexible 208, 210, 212, 213; generic 147; government funding for 166; new 146; R&D institutes 258–87; shortcomings 172 market reform 82, 145, 164; development of machinery industry prior to 125–76; government-run R&D institutes, extent and direction of 177–90; institutional restructuring in 62–4; introduction of 291; way of adjusting to 79 marketing 99, 200, 309; abilities in 45; active 276; automation institute 281; bad prospects 86; engineering services 213; expanding activities 232; expertise 61; export 239; finished products 213; integration of development, production and 36, 71; prominent characteristic of 263; random 220; strategic elements of 279; technological support in 206; ‘technology department’ 260–1; uncertainty regarding 290 ‘marketization’ 37, 63 markets 50, 104–5, 108, 110; access to 70; basic rule of 3; capital 88; changes in structure and organization of 64; competition 200, 301; developed economies 324; earnings 177–90, 332; efficiency of 50; enabling a move into 210; engineering services 211, 232–3; equity 225; excessively small, difficulties arising from 36;

INDEX

expansion of 7, 44; factor in determining the size of 164; finished engines 235; firms fit to conditions of the time 291–2; hierarchies and 315; imperfections 5; increasing penetration 299; increasingly introduced elements 8; information technology 95; intermediation 240, 241; knowledge 61; low-technology products 229; mechanisms 225, 240, 241, 251, 256, 266, 267, 291; most profitable needs in 222; new opportunities 331; niche 211–13 passim, 232, 249, 250, 281, 335, 337; opportunities 45, 250, 259, 335; poor institutions 5; potential 71; provincial 251; pull debate 292; resin-bound sand casting 247–8; responding 269; R&D institutes forced to sell themselves on 4; segmented 105; share 232, 311; signals 183; small 202; transformation 226; underdevelopment of institutions 25; unprofitable 43–4, 53; see also domestic market; international markets; market reforms; technology market mask-making 279 measurement 151, 155, 262; instruments/ devices 127, 157, 307 mechanical products/technology 164, 165 medical sector 65, 92, 108; medicines 40 melting process 155 mergers 210, 225, 317, 323–5 passim; during Cultural Revolution 222; ‘enforced’ 327–8; R&D institutes into existing enterprises 29–34, 51; serious obstacles to 224 metal: corrosion 155;

301

ferrous/ non-ferrous 185; superplastic technology 262 metallurgy: automation technologies 23, 44; equipment 132, 142, 148; instrumentation 43 metal-cutting 297 M-form (multi-departmental) structure 89, 90, 209–10, 213, 242, 332, 337, 339; innovation 335–6, 343–4 micro-economics 315 micro-electronics 93, 95, 171; automation technology 261, 271–82; cheapening and availability of microprocessors 110 military products/projects 120, 160, 263 Milling Machine Institute 159 milling machines 157, 193–200; side entry rotor slotting 219 mining equipment 142 Ministry of Aerospace and the Aeronautical Industry 98 Ministry of the Electronics Industry 16, 31, 108–9, 175; institutes 97 Ministry of Foreign Trade 169 Ministry of the Machinery Industry 14, 16, 30, 31, 120, 150, 166, 175, 325; Bureaux within 151, 169–70, 194, 228, 234; Department of Science and Technology 153; duties assigned by 156; institutes affiliated to 32–4, 144, 180, 181, 328; National S&T Programme projects 147–8; Planning and Capital Construction Departments 141; Power Equipment Testing and Inspection Centre 214– 15; Science and Technology Department 258; see also BRIAMI; BRIMET; Dalian Modular; RIMST; Shanghai Ministry of the Metallurgy Industry 328; Institute for Iron and Steel 197; see also ARTMI Ministry of the Water Conservation and Electric Power Industry 217 Ministry of Water Conservancy 173 MITI (Ministry of International Trade and Industry of Japan) 268 mobile plant 160 mobility 21, 37, 54, 77;

302

INDEX

limited 129 modelling operation processes 99 modernization 141, 197 modifications 132, 196, 202, 217; considerable 298; learning strategy 308; operation and design 291; specific user 240 monopolistic position 151 motor drives 43 motorbikes 238 Motorola 85 moulding materials 155 Mowery, D. 63, 292 multinational companies 206, 225, 229; cooperation with 204 Nagao, M. 118–19, 127 Nanyang Research Institute for Explosion-Proof Electric Apparatus 162 National Construction Bank 38 National Engineering Technology Centre for Metallurgical Automation 46 National Import and Export Commodity Inspection Bureau 228 National Manufacturing Automation Engineering Centre 279 National Natural Sciences Foundation 20 National S&T Programmes 13, 147–8, 271 National Science Foundation 79 National Testing and Inspection Centre for Electric Wires and Cables 228 National Testing and Inspection Centre for Internal Combustion Engines 234 National Working Conference of Science and Technology (1985) 17 Nelson, R.R. 291, 292 neoclassic theory 6 networks: computer 97, 273, 275; scientific information 98 New-Tech see Beijing University NIEs (newly industrializing economies) 294 noise reduction 235, 240, 310 Nokia-Maillefer Cable Machinery Service Centre 229 norms: design 273, 277, 307; institutional 70; internationally compatible 90;

structural and processing 206 Northeast University (Shenyang): Alpine Software Institute Ltd 81; Open Software System Corporation Ltd 80, 81, 102, 103–4 novelty 183–4, 185, 278, 318, 329–30, 331 NTEs (New Technology Enterprises) 35, 36, 45, 247, 276; development of 72–4; engaged in mainly computer and information technology 118; establishment 46, 54; initiation 52, 75–91; licensing 37; losses 91; network for financing 39; organizational networks supporting 53; regulatory environment for 38; role 40; technological activities 92–110; see also spin-off enterprises O’Connor, D. 110, 112 OECD (Organization for Economic Co-operation and Development) 5, 6 OEM (original equipment manufacturing) 294 oil 142, 185; refinery equipment 127 OMRON (Japanese multinational) 206 on-site services 205 open-door policy 8, 70, 110 Operational Automation Engineering project 276 opportunism 318; behavioural 319–20 optical fibre 101 orders 275 organization: acceptable abilities 71; basis for restructuring 117; ‘corporation’ 91; different forms 315–16; inappropriate internal work 302; quasi-market and non-market modes of 316; see also internal organization output 65, 108, 198, 228, 249; academic 79; agricultural 127; competitive, commercially proved 273; degree of completion of 184–5;

INDEX

freedom to sell 8; growth in 129; industrial 119; levels 291; more complete, demand 279–80; only a few users for 164; physical, strong orientation to 52; quantitative, incentive structure oriented to 299; research, commercializing 79; specified 170; type in which know-how is embodied 181–3 Ou Wen 16 ownership: collective 78, 82; differences in 210; disparity in 224–5; non-private 87; officially acceptable category of 77; private 82, 84, 334; public 76, 76, 82; state 334; variety of forms 71 oxygen 244 Pakistan 216 Pan Guishang 210, 213 partnerships 44, 206, 213, 246 parts 93, 152, 263; automobile 263; computer 95, 97; producers 235, 238 Patel, P. 291 patented and proprietary technology 71 path-shifting process 289, 306; dynamic 302; fundamental 266; innovative approach to 308–10; physical apparatus needed for 307 Pavitt, K. 291, 300 PCs (personal computers) 95; dealers 100; exchange centres 53, 54; imported 109; revolution 52, 105, 110; sub-systems 76, 97 penalties 145 performance 125–76, 295; better 38, 90; energy conservation 148;

303

great concern to improve 15; operating 297; targets 16; technology market 86; unimpressive 103 petro-chemicals 185, 221 petroleum 148; refining plants 142 Philippines 216 Philips 101 photo-electric comparators 170 physics 79 pilot plants 24, 45; loans for 38; products 30 planning 88, 104, 108, 228, 243, 249, 263; administration 126; annual 147–8; ‘conventional’ 12; coordination 160, 250–1, 273, 298, 299, 301; costs 151; effectiveness 160; elaboration of practice 12–16, 174–5; existing system 87; innovation 150, 151, 152, 156, 198, 259; investment 143, 144, 145, 293; long-term 147–8; management 148, 150, 151, 157; product 152, 168–75, 294; production 150, 152, 156, 198, 259; serious deficiencies 171; strong reliance on internal consensus 147; technical 155, 170, 172; traditional system 82; vanished or reorganized 245; see also centrally planned economies plano-milling machines 194, 196 plant design 126, 236, 241; institutes 129, 141–5, 308; management of 143–4 PLCs (programmable logic controllers) 276 pneumatic control technology 279 Porter, M. 119, 300 post and telecommunications 83, 117, 118 power generation components/ equipment 43, 142, 145 power plant equipment 214–27, 240; development of 171–4; fossil-fuel 172–4; large capacity 168;

304

INDEX

small-scale 132 PPP (purchasing power parity) 135 precision 147, 223 230, 265 pre-‘decision’ period 12–16 prices 218; contract, flexible and favourable 327; coordination 200; set under centrally planned regime 229 printed circuits 97 printers 97 ‘private ordering’ mechanisms 328, 329, 331 problem solving 174, 293; first step in 299; learning 298; ‘normal’ activity 289; specific 312 procurement 15, 194–5, 223, 229 producer associations 198, 200, 251 product diversification 132–3, 297, 298; accelerating 150; designs already in use to achieve 262; domestically generating 199; innovative efforts on 166; organization of 159; relative capability in 235; steps towards 238; technically supported by rigid standards 311 production 34, 36, 71, 74, 212, 217, 224, 263–5 passim; agricultural 142; automatic systems 99, 100; batch 16, 149, 183, 238, 247; cable and wire 228, 230; capacity 171, 297; cement and glass 145; commercial 35, 39, 276; common activity of R&D 6; competence 301; complex 64, 129; conditions external to the firm 292; coordination departments 152; development of technology 4, 16; disconnection of scientific research from 86; earnings/income from 181, 187; economic, important factor in determining 13; effective and high-quality 301; electrical equipment 159; engine models 235; hydraulic cylinder 248; improvement of products already in 132;

increasing volume 300; integration of development, marketing and 36, 71; in-house 6; large-scale 24, 98, 149; ‘linking’ R&D institutes and 65; management of 147, 217; mass 238; modelling 108; new products 148, 149, 150; oil 142; organic linkage between scientific research and 18; organization of 149; other 41, 249; percentage exported 136; precision machine tools 170; re-investment in facilities 195; requiring smaller investment 160; rolling and forging 261; single and standardized product 204; skills developed to address problems relating to 293; technology-intensive 43; truck 265; turn-key 231; uncertainty regarding 290; see also trial production productivity 133–5; centres 46, 248; design 311; engineering development 277–9; enormous improvement in 13; inferior 295; innovation 311; low 281, 299; relatively high 169; very large fall in 143 profitability 23, 80; financial 334; higher, changes offering 291; relatively high 229; skills needed for 110 profits 80, 101, 204, 209, 238, 269; after taxation 9; control over 195; how to return to initiating institutes 91; seeking 203 programmed exchange equipment 104 property: rights 91; transactions 38

INDEX

prototypes 148; assembly 237; manufacturing 149; testing 237 provinces: coastal 69, 74; hinterland 143; mainland 72; see also under individual names, e.g. Hena; Jiangsu; Zhejiang publication industry 98 pumps 150 Qian Yingyi 39, 89 Qiao Meirong 276, 278, 280 qualifications 84 quality 219, 308; high 232, 293; inspection 202, 245, 247, 226; matters to both producer and designer 240; users did not care too much about 263 quality assurance/insurance 81, 194, 217, 294 quality control 43, 45 quasi-markets 21, 37 quotas 8 R&D (research and development) 6, 37; contractual 21, 27; existing, transformation of 53; government and public financing of 20; in-house 14, 43; internalized 5; ‘marketization’ of 37; production a common activity of 6; projects 14, 26; re-allocation of resources 15, 16 R&D institutions/institutes 7, 22, 29–34, 115–344; autonomy 20–1, 37, 51; channelling components into innovative businesses 59–113; forced to sell themselves on market 4; general survey (1986) 19; integration of assets 35; loans available to 38; reform indispensable to adapt to new economic regime 48–50; rehabilitation and improvement 12, 13–14; released from rigid control 36;

305

restructuring 3, 7; transformation 41–7, 54; venture capital 39 raw materials 229 real estate 76 ‘recombination’ 309, 310 recruitment: aggressive 44; centres 53, 54; life-long 21 reform 120, 133; one of the major approaches to implementing 89; dramatic changes in international trade 136; plant design institutes 145; policy xxxi–58; transformation critical for success of 117; see also economic reform; market reforms Reform Decision (1985) 324 regulations: agencies 38–9, 51; environmental 236, 241, 332; policies for establishment of 37–8; putting into effect 53; rather loose 229; safety standards 241; set up by Decision 36–7 reliability 294, 306, 310 reorganization 233, 245 repair 127, 150, 235; technological support 206 Research Institute for Machinery Manufacturing Science 147 Research Institute for Processing and Production Scheduling 147, 154 Research Institute for Tooling and Tool Science 158 resources: allocation of 216, 245; central mobilization of 127, 141; guaranteed by central planning administration 170, 173; local specificities in endowments 165; mapped out 170; re-allocation of 15, 16, 277, 280, 310, 332; scientific and engineering 226 restructuring 62–4, 80, 91; alternative approaches 46; attempt to understand the basic aspects of 65; attempts 31;

306

INDEX

dramatic 309; essence of transition 50; factors influencing 51–3; fundamental 251; institutional 7, 61, 310, 315–44; internal 45, 204, 213, 221, 223, 233, 239, 244–5, 248, 268; massive and influential 61; M-form 210; organizational 117, 213, 242; partial 199; principle directions of 51–3; profound influence on 9; research institute after merging 33–4; spinoff 40, 53, 61, 66, 68, 85, 89, 118 retailing 73, 74, 104 retirement 195–6; early 21, 223 rewards 90, 145; NTEs suited to manage 39 REXROTH (German multinational) 206 Ricardo (English firm) 235, 239 rights 91; contractual 23; property 91 RIMST (Research Institute for Machinery Science and Technology) 154, 258–9, 260 risk 303, 321; NTEs suited to manage 39; selling goods and services with 321; transaction 325 RISMI (Research Institute for Standardization in the Machinery Industry) 155 robots 103, 155, 207, 272, 275, 279; loading 276; spraying 276, 277, 280 rolling production lines 261, 265, 266 Rosenberg, N. 63, 119, 290, 292, 300, 309, 308 rules 291; institutional 65; market 63; which confirm market value of assets 90 Ruoen, R. 133–5 rural industry/enterprises 9; development 235; initiated 89–90; massive entry of 330; small- and medium-size 247

S&T (science and technology) 64–5, 154, 218, 219, 272; advanced and theoretical, highest emphasis given to 146; commission in charge of affairs 169; Development Programme for the Machinery Industry (1963–72) 148; knowledge generation 290; modern, marching towards 146; ‘projects’ 174; reforms xxxi–58, 180, 310; restructuring 63; well-trained and experienced manpower 208 safety: insurance 221; standards 229, 241, 332 sales 184, 194, 220, 226, 249; computer 95, 105; counted as ‘technological consultancy and services’ 231; electronics 102; engineering services and product, contracted 44; income from 187; information technology 95; poor 62; technological know-how recorded as ‘technology transfer’ 231–2; turbines 215 San Huan New Material R&D Incorporation 80 scholar-businessmen 86 science see S&T ‘science parks’ 38, 35, 73 scientists 118, 119 screw nuts 170 ‘sector bureaux’ 147 ‘sectoral affairs’ 152 segmentation 16, 65 self-interest 319, 336 self-reliance 136, 309 self-sufficiency 172 sellers/selling 76, 109, 321; technological market not very pleasant for 62 seniority 87 Service and Development Centre for Fine Blanking Technology 262 Service Centres for Scientific and Technical Entrepreneurs see Centres services 102; after-sales 76, 78, 109, 110; automatic operation systems 104;

INDEX

computer and information technology 110; contractual 197, 202–3; earnings from 181; engineering 145, 207, 223, 280; generation and dissemination of 188; information 198; inventory of 102; low-grade, returns from 241; more tradable 323; non-technological 185; user 183, 229; see also engineering services; technological services Shanghai 174; Baoshan Iron and Steel Corporation 328; Caohejing Zone 73; Electric Corporation 214, 215–16, 222; Electric Equipment Research Institute 214; Electric Machinery Manufacturing Works 173, 214, 239, 245, 328; Internal Combustion Engine Research Institute 227, 234–9, 242, 328; Jiaotong University 218; Machine Tool Works 159; Materials Research Institute 155; Municipal Government 163, 222; Power Equipment Research Institute 161, 173, 214–27, 272, 328, 331, 337; Research Institute for Electric Cable 163, 227–34, 236, 238, 242; Research Institute for Electric Powered Tools 162; Research Institute for Electrical Automation 163; Research Institute for Industrial Boilers 161; Scientific Research Institute for Electrical Apparatus 161, 247; Turbine Works 214–18 passim, 221–4 passim, 311 shareholders 248, 250 sheathing materials 228 Shen Tianxi 217, 218, 220, 222, 224 Sheng Shuren 141 Shenyang Zone 73, 83, 84, 101–5; Engineering University 103; Low-Voltage Switch Device Works 162; Municipal Government 103; No. 1 Automobile Factory 32; Research Institute for Accumulators 161; Research Institute for Electric Drives 162; Research Institute for Electro-Engineering Equipment 163;

307

Research Institute for Foundry Technology 155, 260, 270; Research Institute for Transformers 161; see also Northeast University Shenzhen: Municipal Government 35; Science and Industry Park 69, 73 Shi Jianzhong 147 Shi Jingxiong 246, 247, 248 SICME Enamelling Machine and Testing Device Service Centre 229 Siegma 229 Siemens 174 Simon, Herbert 293, 319 Si-tong see Stone skills 85, 86, 201, 205; basic 295; built up 291; combination of information and 64; design 289, 309; developed to address; problems relating to production 293; intensive; sectors 70; intensive study of 110; low level 230; management 85, 225, 301; manufacturing 294; M-form internal organization 335; needed for profitability 110; production 290; required to compete commercially 219; special 38; technologically sophisticated 222; testing 289, 294, 309 small- and medium-size enterprises 303, 307; large number of producers 330; non-state, technology-intensive 53; technological services 46, 53; users’ demands 331 small(er) enterprises 24, 228–9, 235, 240, 241; innovative machinery firms 249; internal combustion engine sector 191; systems development 105–8 social security system 196 social services 45, 269 socio-economic factors 292, 293, 300–1 software 61, 100, 103–4, 155, 217, 231; development 107;

308

INDEX

exports to Japan 81; ‘instruments’ 306; know-how 22; management, adaptation of 273; manufacturing automation 271, 279; noteworthy 102; specific applications 98 Song Jian 36, 70–1 South-East University 197 South Korea 69, 110, 129–3 passim, 136, 138, 294, 295 Soviet Union 89, 142, 217, 308; advice 147; archetype of a ‘design bureau’ 205; automobile industry 264; experts 148; generators 172; huge military projects 63; industrial technology imports from Bloc 8; regulations for introduction of new product 149; relied less heavily on standardization 150–1; ‘science-production associations’ 4; supplies of precision machine tools 169 Special Economic Zones 74 specialization 52, 117, 153, 156, 279; central institutes which focus on 249; development work organized by 278; excessive 166; external 204; institutes important because of 159; internal 204, 223; previous, moving beyond 222; promoting 160; research departments organized by 221; rigid 322; technological 209, 227, 270, 317, 329–32 spillover 306, 307 spin-off enterprises 35–40, 45, 51, 52, 53, 59–113 spooling machines 247 SSTC (State Science and Technology Commission) 38, 46, 69, 70; investment from 173; loan appraisal 39; ‘new and high’ technology defined by 37; ‘programming group’ 169; specified areas of ‘new and high’ technology defined by 71; support from 86; ten technological areas to be encouraged 92–3 SSTC (contd.)

staff 234; administrative 245; cuts 245; firms partly owned and run by 89–90; housing 269; occupation structure 33; salaries 79; leaner 223 stagnation 15 stamping 155; leading institute for 262–3; robots 276 standard control machines 98 standardization 146, 151, 155, 196, 200, 228, 243, 258; centre for 202; department 154; elementary and primary work required for 159; imperfect 317; internal combustion engines 234; major task in 149–50; nationwide effort 206; organizing 173; promotion of 148; reliance on 149, 156, 168; responsibility of producer associations 198; Soviet system relied less heavily on 150–1; vanished or reorganized 245 standards 150, 159, 229; compiling and revising 197; delivery time 294; design 206; enterprises encouraged to define their own 198; exhaust emission 241; German firm 202; international 110, 197, 202; ‘I’ series 212; management 148; productivity 294; products which did not fit into the relevant series of 168; professional 14; quality 294; rigid, product diversification technically supported by 311; safety 229, 332; setting 165; ‘type and size’ 206; see also ISO starting-up roles 89

INDEX

State Bureaux: Technical Quality Inspection 234, 247; Technical Supervision 228 State Capital Construction Commission 141, 169 State Council 72, 169, 173; Stipulations for Furthering the Reform of the S&T Management System 29 State Economic Reform Commission 46 state-owned enterprises: government programme to corporatize 204; granted more autonomy 9; large 104, 107, 108, 335; smaller 9, 38 State Planning Commission 46, 88, 141, 279 State Science and Technology Commission 36 Steinmueller, W.E. 120, 307 Stone (Si-Tong) 78, 98, 104 sub-contractors 44, 235 subsidiaries 83 subsidies 194, 202 substitution 24, 297, 298; down-grading 219 Sun Micro 97 supply 104, 147, 331; domestic 132, 136, 138; push debate 292; structure 192 surface engineering 262 suspension bridges 233 Suzhou Research Institute for Electric Spark Machines 158, 211 switches: low-voltage 162; standardization for 150 systematization 146 systems: development 99; integration 264–6 ‘systems effect’ 335, 336 systems engineering 217, 221, 294, 306; forging 265; weakness in 217 Szirmai, A. 133–5 Taiji Computer Corporation 97 Taiwan 69 talent 37, 70, 76, 77–8 targets 16, 44, 224; economic performance 16;

309

planning 4; reform 53 taxation 9; exemption 30; incentives 37; preferential 37; profits after 9 technical analysis 183 Technical Committees 228 ‘technical specification’ 149 technicians 201, 246; training 194, 205 technological change 99, 197; cumulative processes 289–91; directions and characteristics 307; dynamic 278, 311; enthusiasm for 298; general 299; institutions for 141–75; milestones of 297; new features constructed 309; physical outcomes 312; rapid 31, 200– 13, 276; reorientation 301–8; selectiveness 291–3; shaping the direction of 300; small producers lack experience with 229 technological services 27, 50, 187, 188, 207, 211, 220, 221, 226, 230, 235, 236, 240, 249, 259, 260, 275, 281, 331–2; contractual 197; income from 232; small- and medium-size enterprises 46, 53 technological trajectories 241, 288–314; disparity in 225 technological imports 144, 146, 174, 216–18, 219, 229, 231, 235–6, 238, 265, 310; absorbing and adapting 72; assimilation of 270; direct and indirect access to 241; effect of 306; heavy reliance on 266; institutional arrangements for 308 technology market 86; contractual relationships in 305; creation of 17–28, 62, 322; delayed entry to 226; engineering services in 239–41; institutes thrown onto 321;

310

INDEX

international 277; more ‘friendly’ to some kinds of transactions 27; performance 86; pure, inadequacy of the approach 50; rather open 241; robots 276; services provided through 240; size and quality 191 technology transfer 22, 62, 184, 187, 207, 211, 238, 249, 259, 331–2; average earnings from 236; contract recorded as 231; contracts 240; declined 27; discrete licensing for 204; earnings/ inincome from 181, 232; licensing 217; major form of 142; mechanisms to strengthen 70; one-off 103; sales of technological know-how recorded as 231–2; transactions 50 ‘technopolises’ 38 telecommunications: equipment 95, 98, 101, 108; post and 83, 117, 118 testing 118, 146, 155, 170, 173, 196, 228, 243, 246, 251, 266; accumulated skills in 289; auxiliary equipment 221; boilers 161; capabilities in 309; devices 231, 240, 307; improving facilities for 207; intense 236, 259; prototype 237; reduced since reform 198; services 165; shortcomings in technology 172; skills and methods for 294; technologies for 194, 263; vanished or reorganized 245 textiles 185 thermo-dynamics 214 Tianjing 73; Design and Research Institute for Electric Drives 162; University 197 Tianshui: Electrical Apparatus Corporation 162; Great Wall Switching Gear Works 96, 272–3;

Research Institute for Electric Drives 162 Tidrick, Gene 10 tools 193; cutting and measuring 157; other 152, 157–9, 163–4; small electric-driven or manual 227; see also machine tools Torch programme (1988) 35–40, 53, 51, 90, 247; ‘Centre’ encouraged by 78; launch of 62, 66, 68–74, 77; ‘leading funds’ 20; loan quota under 87; ‘micro-electronics’ in categorization by 95; policy instrument for the implementation of 66; preferential policies stipulated by 85 ‘township and village enterprises’ 227, 228 tractor industry 205, 238; manufacture of engines for 209 trade 73; international 136, 277, 280, 295, 300, 306; liberalization 295 training 22, 53, 78, 108; centre for employees out of work 223; computer 81; contracts 203; for importing enterprises 275; technicians 194, 205; technological support in 206 transaction cost theory 315 transactions 5, 62; agencies for 19, 53; market, efficient 25; property 38; R&D projects 26; relationships 320–32; ‘technology transfer’ 50; uncertainty 50 transformation: established industrial technology institutions 115–344; structural 22, 85 transformers 161; standardization for 150 transition 48, 53, 336; common problem 25; management centralized throughout 245; painful 85; paving the way to 54; radical 329, 332; rapid 110;

INDEX

restructuring essence of 50; smooth, sensible guidance to 63 translation 196, 206 transmission and distribution equipment 159 transportation sector 117, 118, 235; equipment 119, 129, 135, 142, 195, 207 trans-regional corporations 102 trial-and-error process 293, 308 trial production 24, 41, 45, 181, 184, 187, 207, 209, 220, 226, 260, 278; income from 249; new technology 38; products selected for 221, 238; share of 232; small batches 247; workshops 11 tribology 155 trucks 265, 267 turbines 174; blade forging 262; electric power 309; nuclear 215; rotor blade 263; principal manufacturers 216; steam 161, 215 turn-key projects 269, 295; packages 231; plant construction 142 turnover 25–6, 74, 76; computers 109 Tu Zhuzong 239 typesetting systems 95, 107; manual 98 Tyre, M.J. 292, 293 U-form (Unified form) structures 89, 242, 332, 336, 337, 339; to H-form 333–5 uncertainty 63; economic 290; frequency of transactions and 318–19; inevitable in the course of radical institutional change 90; innovation 25; transactions 50, 331; weakness or deficiency in dealing with 24; technological novelty associated with 183 United Kingdom 202, 206 United Nations 197;

311

UNDP (UN Development Programme) 181 United States 44, 131, 206, 219, 336; institutional restructuring 63; manufacturing 135 universities 70, 78, 102, 141; best graduates 146; engineering 268; R&D 166; scientists and engineers leaving 86; young engineers sent to 203 upgrading 148, 196, 206–7, 229–30, 235–6; based on imports 202; capacity 194; engine technology for small producers 238; infant industry 149 users 274, 299; capability building 108–9, 110; communication producer and 299; demands 277; did not care too much about quality and timing 263; inexperience of 25; interactions 184; needs sensitivity 241; participation 264–5, 266–7; specific conditions/ requirements 108, 184, 295 USSR see Soviet Union vacuum furnaces 269 value added 79 VAX systems 97 vehicles: power plants 160; small 238; see also automobile industry; tractor industry; trucks venture capital 39, 83, 87 Venture Investment Corporation 39, 90 vertical integration 145, 223; ‘lateral’ 323 vibration analysis 203 volume 103, 125; batch 248 Von Hippel, E. 292, 293 wages 30; low 232 Waldrich Coburg 194 Wang Dechen 269

312

INDEX

Wang Keng 205 Wang Zhiqi 234, 236, 238, 239 warehouses 278; automated high bay system 272, 273, 275, 280, 311 water-cooled generator technology 173 water pumps 133 weaknesses 241, 294, 309; preventing keeping up with pace of innovation 203; strategic control over specific assets 334; systems engineering 217 welding 155, 261; electric 133; robots 276 Weng Zuliang 235, 237, 238, 239 Westinghouse 174, 215, 216, 218, 223, 309 Williamson, Oliver E. 5, 119, 123, 315–23, 332–3, 334, 341 Winter, S.G. 291, 292 Wire and Cable Certification Station 228 wires 163, 227–34 woodworking machinery 157, 159 work stations 97 workbench fixtures 150 World Bank 133, 181 worm gears 170, 171 Wu Jinglian 105 Wu Ye 217, 218, 219, 222, 223, 224 Wuhan 73; Centre 77, 92; Chutian Optical Electronics Corporation Ltd 77, 78, 99; Eastlake New-Tech Enterprise Incubator 78, 86–7, 99–101; Institute of Optical Technology 78; Municipal Government Bureau of Finance 87; Research Institute for Material Protection 155; Research Institute for Posts and Telecommunications 101; Zone 83, 84, 99–101 Wuxi Machine Tool Works 272 Xi’an: Electrician Capacitor Works 161; Heavy (Metallurgical) Machinery Research Institute 224, 260, 328; High-Voltage Ceramics Works 163; Research Institute for Electric Furnaces 162; Research Institute for Electrical Capacitors 161; Research Institute for Electrical Ceramics 163;

Research Institute for Electro-electronic Engineering 163; Research Institute for High-Voltage Electrical Apparatus 161; Research Institute for Micro-Electric Control Machines 160, 162; Research Institute for Rectifier Devices 161 Xianda Bar Code Scanner Technology Corporation Ltd 103 Xiantan Research Institute for Electric Traction Equipment 162 Xin Tiandi 85 Xin-tong 69 Xu Chenggang 39 Xuchang Research Institute for Relay Devices 161 Yu Chengting 159, 210 Yu Yunlong 229, 230, 232, 233, 234 Yuan Baohua 141 Yuan Baoshan 219, 224 Zeng Hong 224 Zhang Jianming 223, 224 Zhang Meiguang 234, 260 Zhao Ziyang 15, 17–18 Zhejiang Province: Mechanical and Electric Design and Research Institute 145, 243, 246–8, 249, 251, 337; Qinshan Nuclear Power Station 215; University 81, 88 Zheng Hong 222 Zhengzhou: Research Institute for Abrasive, Grinding and Grinding Apparatus 158; Research Institute for Machinery Engineering 154, 155, 260 Zhou Enlai 146 Zhou Jian’nan 126, 172, 173, 216, 217, 224, 228, 234 Zhou Zhang 160, 163, 224 Zhou Zian’nan 163 Zhu Sendi 158, 163, 165, 222, 243, 251, 259, 260, 268, 270, 276, 278, 310, 312 Zones (Development Zones for New Technology Industries) 35, 38, 54, 65, 66, 79; administration 82, 87, 88, 89, 90, 93; booming 90; development of 53, 71, 72–4; expanding 46; investments in infrastructure 39;

INDEX

local governments supported the initiation of 52; managers 81, 82, 90, 92, 93, 98 widespread establishment of 85; see also Beijing; Shenyang; Wuhan Zones (contd.) Zu Chenggang 89

313