WORLD BANK TECHNICAL PAPER NO. 384 WTP3sLI Work in progress for public discussion N ov. CItching Up to Leadership lnI V/ ( // i(t ////(//(t) -S//)/)(/'/ l|// /////(it//S I/ J(//k/// i t (i / { Slt 1e, 44 x UI Ik. / l / \ S/////Åt/ t/4. RECENT WORLD BANK TECHNICAL PAPERS No. 310 Elder and Cooley, editors, Sustainable Settlement and Development of the Onchocerciasis Control Programme Area: Proceedings of a Ministerial Meeting No. 311 Webster, Riopelle and Chidzero, World Bank Lending for Small Enterprises 1989-1993 No. 312 Benoit, Project Finance at the World Bank: An Overview of Policies and Instruments No. 313 Kapur, Airport Infrastructure: The Emerging Role of the Private Sector No. 314 Vald6s and Schaeffer in collaboration with Ramos, Surveillance of Agricultural Price and Trade Policies: A Handbook for Ecuador No. 316 Schware and Kimberley, Information Technology and National Trade Facilitation: Making the Most of Global Trade No. 317 Schware and Kimberley, Information Technology and National Trade Facilitation: Guide to Best Practice No. 318 Taylor, Boukambou, Dahniya, Ouayogode, Ayling, Abdi Noor, and Toure, Strengthening National Agricul- tural Research Systems in the Humid and Sub-humid Zones of West and Central Afica: A Framework for Action No. 320 Srivastava, Lambert, and Vietmeyer, Medicinal Plants: An Expanding Role in Development No. 321 Srivastava, Smith, and Forno, Biodiversity and Agriculture: Implications for Conservation and Development No. 322 Peters, The Ecology and Management of Non-Timber Forest Resources No. 323 Pannier, editor, Corporate Governance of Public Enterprises in Transitional Economies No. 324 Cabraal, Cosgrove-Davies, and Schaeffer, Best Practices for Photovoltaic Household Electrification Programs No. 325 Bacon, Besant-Jones, and Heidarian, Estimating Construction Costs and Schedules: Experience with Power Generation Projects in Developing Countries No. 326 Colletta, Balachander, and Liang, The Condition of Young Children in Sub-Saharan Africa: The Convergence of Health, Nutrition, and Early Education No. 327 Vald6s and Schaeffer in collaboration with Martin, Surveillance of Agricultural Price and Trade Policies: A Handbook for Paraguay No. 328 De Geyndt, Social Development and Absolute Poverty in Asia and Latin America No. 329 Mohan, editor, Bibliography of Publications: Technical Department, Africa Region, July 1987 to April 1996 No. 330 Echeverria, Trigo, and Byerlee, Institutional Change and Effective Financing of Agricultural Research in Latin America No. 331 Sharma, Damhaug, Gilgan-Hunt, Grey, Okaru, and Rothberg, African Water Resources: Challenges and Opportunities for Sustainable Development No. 332 Pohl, Djankov, and Anderson, Restructuring Large Industrial Firms in Central and Eastern Europe An Empirical Analysis No. 333 Jha, Ranson, and Bobadilla, Measuring the Burden of Disease and the Cost-Effectiveness of Health Interventions: A Case Study in Guinea No. 334 Mosse and Sontheimer, Performance Monitoring Indicators Handbook No. 335 Kirmani and Le Moigne, Fostering Riparian Cooperation in International River Basins: The World Bank at Its Best in Development Diplomacy No. 336 Francis, with Akinwumi, Ngwu, Nkom, Odihi, Olomajeye, Okunmadewa, and Shehu, State, Community, and Local Development in Nigeria No. 337 Kerf and Smith, Privatizing Afica's Infrastructure: Promise and Change No.-338 Young, Measuring Economic Benefits for Water Investments and Policies No. 339 Andrews and Rashid, The Financing of Pension Systems in Central and Eastern Europe: An Overview of Major Trends and Their Determinants, 1990-1993 No. 340 Rutkowski, Changes in the Wage Structure during Economic Transition in Central and Eastern Europe No. 341 Goldstein, Preker, Adeyi, and Chellaraj, Trends in Health Status, Services, and Finance: The Transition in Central and Eastern Europe, Volume I No. 342 Webster and Fidler, editors, Le secteur informel et les institutions de microfinancement en Afique de l'Ouest No. 343 Kottelat and Whitten, Freshwater Biodiversity in Asia, with Special Reference to Fish No. 344 Klugman and Schieber with Heleniak and Hon, A Survey of Health Reform in Central Asia (List continues on the inside back cover) WORLD BANK TECHNICAL PAPER NO. 384 Finance, Private Sector, and Infrastructure Network Catching Up to Leadership The Role of Technology-Support Institutions in Japan's Casting Sector Sakura Kojima Yoshitaka Okada The World Bank Washington, D.C. Copyright @ 1997 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing November 1997 Technical Papers are published to communicate the results of the Bank's work to the development community with the least possible delay. The typescript of this paper therefore has not been prepared in accordance with the proce- dures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. Some sources cited in this paper may be informal documents that are not readily available. The findings, interpretations, and conclusions expressed in this paper are entirely those of the author(s) and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to members of its Board of Executive Directors or the countries they represent. The World Bank does not guarantee the accuracy of the data in- cluded in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such bound- aries. The material in this publication is copyrighted. Requests for permission to reproduce portions of it should be sent to the Office of the Publisher at the address shown in the copyright notice above. The World Bank encourages dissem- ination of its work and will normally give permission promptly and, when the reproduction is for noncommercial purposes, without asking a fee. Permission to copy portions for classroom use is granted through the Copyright Clearance Center, Inc., Suite 910, 222 Rosewood Drive, Danvers, Massachusetts 01923, U.S.A. The complete backlist of publications from the World Bank is shown in the annual Index of Publications, which con- tains an alphabetical title list with full ordering information. The latest edition is available free of charge from the Dis- tribution Unit, Office of the Publisher, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from Publications, The World Bank, 66, avenue d'Iena, 75116 Paris, France. Cover photos: "Melting," traditional foundry factory in 1850s, provided by Chuo Malleable Iron Co., Ltd. "Casting line for auto parts," recent foundry factory, provided by Automobile Foundry Co., Ltd. ISSN: 0253-7494 Sakura Kojima is a lecturer in the Department of Liberal Arts at Tokai University. Yoshitaka Okada is a professor in the Department of Comparative Culture at Sophia University. Library of Congress Cataloging-in-Publication Data Kojima, Sakura. Catching up to leadership : the role of technology-support institutions in Japan's casting sector / Sakura Kojima, Yoshitaka Okada. p. cm. - (World Bank technical paper; no. 384) Includes bibliographical references. ISBN 0-8213-4054-9 1. Founding-Technological innovations-Japan. I. Okada, Yoshitaka, 1949- . 11. Title. III. Series. TS229.5.J3K65 1997 338.4'76712'0952-dc2l 97-30561 CIP Contents List of Tables and Figures....................................zv Foreword .................................................v Abstract-Japan's Casting Sector .................. ...........vu Acknowledgments........................................vm Executive Summary........................................1 Introduction.............................................2 Historical Background..............................................2 Main Issues and Hypotheses.........................................2 The Postwar Development of Japan's Casting Industry..................................5 Production Trends.................................................5 Technological Development: Institutions and Policies.............................................5 Industry Overview.........................................9 Production Trends from 1981 to 1990 ................... ................9 Industrial Organization............................................10 The Technology-Support System..............................12 Overview.......................................................12 Survey of the Casting Industry and Technology-Support System- Methodology and Sampling.........................................13 Research Findings on Ten Sample TSIs.................................15 Evaluation of Technology-Support System..............................18 Evaluation of Technology Policies....................................20 Case Studies: Firm and TSI Interaction in Technological Change..................................22 Comparison of Keiretsu Core Firms and Independents ....................22 Comparison of Keiretsu and Non-keiretsu Subcontractors ..................23 Other Case Studies................................................24 Conclusions and Lessons for Developing Countries..................................25 Appendix A.............................................27 Appendix B.............................................28 Appendix C................................................29 Notes................................................. 30 References.............................................. 31 iii List of Tables and Figures Tables 2.1 Production Trend of Casting Industry (1950-1979) ....................................... 6 3.1 Production Trends in Cast Iron (1980-1990) ................................................... 9 3.2 Cast Iron Production Share by End User ........................................................9 3.3 Labor Productivity by Subsector ..................................................................... 10 3.4a Share of In-House Production .......................................................................... 11 3.4b Production Share by Type of Firm .................................................................. 11 3.5 Share of Establishments by Size ...................................................................... 11 4.1 Monthly Testing and Analysis Requests ....................................................... 13 4.2 Size Distribution of 18 Purposively Sampled Firms ......................................13 4.3 Purposively Sampled Firms Classified by Technological Dynamism ........14 4.4 Distribution of 50 Returned Questionnaires and 500 Selected Firms..........14 4.5 Technological Capacity by Firm Size ................................................................17 4.6 R&D Levels by Type of Firm ............................................................................. 17 4.7 Competition and R&D....................................................................................... 17 4.8 Firm Size and Frequency of TSI Use ................................................................ 17 4.9 Firm Size and Share of Staff Training ............................................................. 18 4.10 Evaluation of Major TSIs by Firm Size and Degree of Appreciation of TSI Support .......................................................................18 4.11 Policy Menus Used by Firms ............................................................................. 19 5.1 Average Number of R&D Staffs by Firm Size .................................................22 Figures 3.1 Production Trends by Subsector ...................................................................... 9 3.2 Labor Productivity by Subsector ................................................................... 10 3.3 Number of Establishments (Cast Iron) by Size, 1990 ....................................10 iv Foreword Successful industrialization depends critically on tech- institutions (TSIs) are appropriate, and what roles can nological development. Continuous improvements in they play? Empirical study has generally lagged behind technology enable firms, industries, and national econo- the new conceptual understanding of ITD. Although mies to enhance productivity and build competitiveness there have been a number of firm-, industry-, and coun- in new areas. Advances in the economic theory of inno- try-level case-studies, there has been a lack of compara- vation, together with accumulated experience of policy tive analysis taking account of variation across sectors makers and practitioners, have shed light on the process with different core technologies, industrial structures, of industrial technology development (ITD) and the po- and public institutions. In view of this gap, the Asia Tech- tential role of supporting policies and institutions. Tra- nical Department designed and led a multi-country, ditional linear theories of innovation, including both multi-sector study titled "Institutional and Policy Priori- "science-push" and "demand-pull" models, have been ties for Industrial Technology Development". The project replaced by an understanding of a more deliberate pro- assembled research teams in seven economies - China, cess of progressive accumulation of technological capa- India, Japan, Korea, Taiwan (China), Mexico, and Hun- bilities by firms and industries. Government's ability to gary - to study how firms in six industrial sectors have drive ITD through "public goods" investments in basic expanded their technological know-how, obtained sup- scientific research is now seen as limited, and important port from external institutional, market, and policy-in- efforts to reform isolated and ineffective public research duced sources, and interacted with TSIs. The study establishments have recently been mounted, particularly involved both historical research drawing on secondary in developing countries. It is also recognized that indus- sources and the administration of surveys (through in- trial firms must go beyond "learning by doing" to invest terviews and by mail) of firms and TSIs using the same in acquiring technological knowledge and mastery, questionnaires in each location. Its results provide a rich drawing on stocks of technologies from abroad and from base of quantitative and contextual information, both em- external sources in their own countries. Beyond their pirically detailed and broadly comparative, on the policy individual efforts in technological learning, moreover, and institutional aspects of ITD. Financial support for firms are stimulated and supported in important ways this effort came from Japanese Government trust fund, by the environment in which they operate. This envi- Canada's International Development Research Centre ronment includes the policies and economic institutions (IDRC), Taiwan(China's) IECDF and the World Bank which both shape firms' incentives to invest in techno- itself. logical learning, and provide complementary assets - The following report, "Technology Support Institu- technical skills, information, technical and financial sup- tions and Japan's Casting Sector" by Sakura Kojima and port - which make such investments more productive. Yoshitaka Okada, offers an excellent example of the Deriving lessons for policy making in developing study's results. It illustrates how Japan used technology countries remains a challenge. What are effective tech- policy and institutions to improve industrial capabilities nology policies? What sorts of technology supporting and eventually become a world leader in castings, which v has been critical to the related success of Japan's machin- groups and associations, and public TSIs, which have ery and automotive industries. The study suggests that changed over time and have served different firm popu- even in a technologically mature sector, such as casting lations in distinct ways. The report provides examples and foundry, a variety of private and public institutions of successful interaction between firms and TSIs, offer- were important in technology diffusion. Professors ing important lessons for policy making aimed at catch- Kojima and Okada highlight the complementary rela- ing-up technologically and moving forward in mature tionship between private institutions, including business industries. Carl Dahlman Director, World Development Report The World Bank vi Abstract-Japan's Casting Sector Japan's casting and foundry sector has been integral to training, and trouble-shooting, much more than R&D. the developmental success of a range of machine-based Support to small firms came through private "institu- industries. Although casting technology is mature -- tions", including business networks known as keiretsu widely commercially available and with limited scale and industry cooperatives, as well as an elaborate sys- economies -- institutional and policy support was vital ter of public institutes at the local and regional level, to the Japanese casting industry's rapid gains in quality particularly in areas with traditional clusters of casting and productivity. Rapid import and mastery of techno- firms. Small keiretsu-affiliated firms used logical know-how in the early post-war era was essen- technology-support institutions (TSIs) to help them tial for the upgrading of the sector. Government achieve the high quality standards demanded by cus- encouragement, along with controls on foreign exchange tomers. Larger independent producers relied more and technology licensing, led private industry to coop- heavily on public TSIs to compensate for their lack of erate through business associations in licensing and ab- access to keiretsu networks' technical information and sorbing major advances in casting technology, thereby resources. making the technology transfer process rapid, efficient The role of formal TSIs has changed as industry has and inexpensive. developed its capabilities. The early role of industry as- The presence of a wide variety of institutions has sociations in coordinating technology transfer and ab- enabled different types of firms to obtain external tech- sorption diminished with time. Public TSIs at the nical support. In particular, support for small and national and regional level have faced the need to shift medium-scale foundries was critical to the rapid diffu- their activities towards research to stay ahead of their sion of improved techniques, including quality control, technologically sophisticated clients, the most advanced throughout the industry. Even today, casting firms de- of whom had become world leaders in the casting mand diffusion-related services such as information, industry's core techniques. vii Acknowledgements This paper is part of the Japanese Project including seven their extensive cooperation in interviews and detailed cases of industries. It is one of the seven-country compara- answers to my many questions, and to fifty firms which tive studies of the World Bank's project on "Policy and In- kindly replied to my mailed questionnaire. stitutional Priorities for Industrial Technology Special thanks go to the following people: Development," organized and led by Mr. Melvin Masatada Kawabata, Director of the Mie Prefectural Goldman, a senior World Bank technology development Metal Research Institute; Yasuhiko Kondo Ph.D., Di- economist. rector of the Governent Industrial Research Institute, The Japanese team, consisting of nine members, includ- Nagoya; Isamu Taki, President of Tokyo International ing myself, was organized under the guidance of Professor Foundry Engineering Consultant Company; Gohzo Yoshitaka Okada, Sophia University, a distinguished Namekawa, manager of the Materials Process Technol- scholar, to systemize our case studies. I found the discus- ogy Center; Noboru Hatano, Director of the Technical sions in our group proved both informative and exciting. Department, Japanese Association of Casting Technol- Mr. Fumio Nishikawa, the former Chairman, Toray Re- ogy; Noboru Yamanaka, Manager, Casting Section, search Center, an advisor to the International Project-Team, Saitama Prefectural Casting and Machinery Research provided much kind help and introduced me to Mr. Tsuneo Institute; Tadashi Yoshiwara, a consultant to the Horie, President, Taihei Kinzoku Kogyo Co. Ltd., who gave Saitama Prefectural Research Institute. Thanks also to me valuable advice. I was invited to join the project by Pro- June Streitmun-Setoguchi for editing this manuscript. fessor Ichiro Inukai, International University of Japan, an Messrs Bennet Minton and Greg Felker's editorial as- advisor to the Japanese team. sistance is also appreciated. I am also deeply grateful to the representatives of the Any responsibility for the content of this paper, eighteen firms and ten technology-support institutions for however, is entirely mine. Sakura Koftma viii Executive Summary Japan's high quality, highly productive casting sector TSIs. Business associations and cooperatives have helped has been key to its powerful competitiveness in a range diffuse technology and provided technical support and of machine-based industries. Only forty years ago, the information services. Government policies and incen- sector was technologically far behind the leading indus- tives have both encouraged and relied on private sector trialized countries. A crucial part of the industry's drive cooperation. for technological proficiency was the role of technology- TSIs' effectiveness depends on their own dynamism supporting institutions (TSIs) and government policies. and their client firms'investments in technology. Because By encouraging cooperation within industry and estab- of competition, small and medium-size suppliers must lishing TSIs to address a variety of needs, Japan was be technology-oriented. Keiretsu relationships, for ex- able to acquire technological capability rapidly and ample, are not necessarily fixed, and keiretsu subcontrac- cost-effectively. The TSIs include national research in- tors can fall out of the group if their R&D efforts wane. stitutes, prefectural research institutes (PRIs), industry Public TSIs complement small firms' keiretsu linkages. associations, inter-company networks, industrial coop- Larger independent casting firms seek public TSI sup- eratives, consulting firms and universities. They have port to keep abreast of their keiretsu-affiliated competi- evolved along with the industry, though for most of the tion and to compensate for their lack of access to the large post-war period, institutions and policies focused on industrial groups' laboratories and other technological diffusing improved casting techniques and materials. resources. Imported technology contributed greatly to industrial TSIs, long focused on disseminating imported tech- performance in the 1950s and 1960s, and TSIs helped nology and improving production techniques, now face exploit it. changes that require innovation and more basic R&D. The casting sector supplies semifinished goods to Dynamic TSIs are seeking new support roles and greater machine industries, and its precision and quality deter- expertise. Some of their support functions, such as prob- mines that of a spectrum of final products. The industry lem-solving and laboratory services, have declined in is composed mainly of small and medium-size subcon- importance, while information services and R&D, have tractors having fewer than 300 employees. Institutional assumed increased importance. TSIs thus must reorga- support for small and independent producers, particu- nize to meet changing times: R&D-oriented TSIs have to larly from PRIs, has been especially important. Collabo- strengthen their know-how and strategic research, while ration occurs in many ways, such as keiretsu and non-R&D-oriented TSIs, such as industry associations, subcontracting. Inter-company links complement formal need to focus on information analysis and distribution. 1 Introduction Historical Background United States, for example. The myriad smaller firms seeking subcontracted work, and their very low wages, Many scholars hold that Japan's machine industries be- promoted subcontracting. Larger assembly firms facing came the world's strongest competitors because they are severe international competition transferred technical built on small- and medium-sized suppliers and know-how to subcontractors to raise quality, which was semifinished-goods subcontractors with high technologi- as imperative as cost reduction. cal capability, especially in production control know- At the end of the 1950s, under pressure of trade lb- how. The casting, forging, metal stamping, and powder eralization, the Japanese government began promoting metallurgy industries are important suppliers to a range factory modernization in small and medium firms to of machine industries.' The quality of their parts strongly prevent their bankruptcy (Small and Medium Enterprise affects the quality of final products. Most casting firms Modernization Promotion Law, 1963). The policy contrib- classified as subcontractors are small- to medium-sized uted to mechanization and automation and vastly im- in terms of capital and employees. proved the productivity of small and medium firms. A After World War II, Japan lagged far behind the west variety of TSIs also contributed to raising the technologi- technologically and did not compete internationally. But cal level of these firms. with imported technology, Japan gradually closed the gap and entered an era of long-term, high-speed eco- Main Issues and Hypotheses nomic growth. Japan is often cited as a "Gerschenkron model", that is, a latecomer enjoying the accumulation This analysis focuses on the institutional factors contrib- of innovations developed in advanced countries.2 Tech- uting to the improvement of the technological capabil- nological progress based on imported technology, ties of Japanese firms, using the casting industry as a case coupled with high levels of savings and investment, ex- study because of its important role in the machine indus- plains much of her rapid industrial growth.' The casting tries. Its purpose is to analyze how TSIs and government industry was no exception to this pattern. Major inno- policies improved the technology of firms and the indus- vations were imported, mainly during the 1950s and try as a whole. Japan's institutional experiences may have through the 1960s, and introduced by larger firms who important implications for developing nations currently then improved, adapted, applied and disseminated them pursuing rapid industrialization. to small and medium firms. As a starting point, three major hypotheses are of- As often noted, Japanese industry is characterized fered: by intercompany links based on a socially defined divi- sion of labor. The subcontracting system originated in Hypothesis 1: The technology-support system evolves military industries and the inter-firm keiretsu4 structure with the industry's development cycle. Industry and peculiar to Japanese corporations developed along with technology follow a "logistics curve" from infancy high postwar economic growth. Larger firms had an in- through growth and maturity to decline. The effective- centive-cost reduction-to contract out work. In-house ness and role of TSIs and related policies follow the curve. parts production is very low compared to that in the The importance of government policy, for example, 2 has declined as the private sector has grown. Institutions tories from university, national and local research insti- and policies played different roles in the 1950s than they tutes, and the former (captive TSIs) have outstripped do today. Although my scope concentrates on the the latter (non-captive TSIs) in total R&D expenditures, present, historical perspective is important. numbers of qualified researchers, and high-tech facili- The casting industry and its postwar development ties. Large firms that have acquired technological capa- includes: bility tend to establish links with universities and * the 1950s, an era of massive technological import academic societies in and out of Japan and with foreign from overseas, firms. * the 1960s, a period of technological adaptation and Small and medium subcontractors receive techno- dissemination, and logical support from parent corporations or their main * the 1970s and onward, when R&D advanced rapidly buyers, and technical advice and know-how from firms and the importance of imported technology declined. that supply their equipment and facilities. Today, Japanese firms and research institutions are Hypothesis 3 The keiretsu or subcontracting system moving from R&D involving minor improvements and influences the interaction of firms with TSIs. practical applications to basic technological innova- The keiretsu and subcontracting system are the most tion. In the field of casting, classified as a mature tech- important channels of technological transfer. Now tech- nology,s however, there will be little room for nology developed by the parent firm's research labora- innovative technology. tory is transferred to subcontractors. In return for free Hypothesis 2: A firm's technological capacity, which technology transfer, member firms obey standards of is strongly influenced by its size, determines its R&D quality, cost reduction and timely delivery. Hence, firms sources and use of TSIs. with keiretsu relationships do not need outside TSI sup- Leadership for technological innovation in Japan port as much as do independent firms, who rely on TSIs has migrated to larger firms and their captive labora- for R&D. 3  The Postwar Development of Japan's Casting Industry Production Trends in the United States between 1923 and 1926; spheroi- dal graphite cast iron, developed in the United King- The casting industry has been profoundly affected by dom between 1947 and 1948; and high-grade cast iron trends in the machine industries and by the overall such as acicular and compacted vermicular (CV) economy's boom-and-bust cycles. Prosperity during the graphite cast iron. The melting of cast iron improved Korean War in 1950-1953 and later periods helped Japan greatly with the invention of hot-blast and water- recover from World War II. The industry's chief custom- cooled cupolas. The low-frequency induction furnace ers were industrial, electrical, and transportation machin- eventually replaced the cupola because of its better ery manufacturers. Since the 1960s, the automobile pollution control. These technologies were imported industry has been the most important customer for cast full-scale into Japan in the 1950s. metal products. With the rapid growth of the automo- Major molding techniques included shell molding, bile and other machine industries, casting production developed by a German scientist in 1944; CO, gas hard- grew 7 percent annually from 1950 to 1979 (Table 2.1). ening, developed in England in 1948; and self-curing Just before the first oil shock in 1973, total car produc- molds, such as the fluid sand (FS) process developed tion exceeded 7 million, crude steel reached 120 million in the former Soviet Union and its adaptations, such tons, and casting output recorded almost 8 million tons. as the N process, developed in Japan. The vacuum cast- Although temporarily slowed by the oil shock, postwar ing or V process was developed in Japan in 1971, and casting grew sevenfold between 1950 and 1979. molding R&D was an important area in the 1950s and Casting industry employment has declined since the 1960s. early 1960s, when labor shortages became a serious prob- Mechanization of the production process began in lem (Table 2.1). It fell from 0.37 percent of all industries the 1950s and later moved rapidly into automation. In in 1960 to 0.19 percent in 1979. Productivity, in contrast, the beginning, the shortage of casting machine manu- improved from 17 tons per worker to 66 tons, thanks to facturers in Japan necessitated the import of equipment the rapid automation and mechanization of small and such as molding, shot-blast, and sand-control ma- large firms. The casting industry's contribution to Japan's chines. Labor shortages in the 1960s spurred automa- GNP in the 1960s and 1970s remained high, estimated to tion, as did enactment of the Antipollution Basic Law be 1.41 percent at the highest (1960) and 0.73 percent at in 1967, which led to the import or development of the lowest (1978).6 pollution control facilities. The number of casting ma- chine manufacturers grew as mechanization increased. Technological Development: The period from 1958 to the first oil shock in 1973 Institutions and Policies saw a sevenfold increase in all imported technologies, including casting.7 Basic technologies were generally Innovations in casting have evolved in three areas- imported first by large firms having relatively high materials, molding, and production process control R&D capacity. These companies then applied and im- (mechanization, automation, and computerization). proved their imports and disseminated new techniques New materials included meehanite metal, developed to other firms, including small-scale subcontractors. 5 Catching Up to Leadership Business associations of casting firms were set up in the solving, and advisory services. 1950s and 1960s for importing, adapting, and dissemi- For spheroidal graphite cast iron, Toshiba, one of nating technology. Local public (prefectural) research Japan's largest electrical equipment producers, obtained institutes also helped disseminate information and tech- a patent from international Nickel, a U.S. company, in niques to small firms. And industrial policies were es- 1952 and proceeded to initiate production and adapt the tablished favoring dissemination. technology. After the patent expired, the Japan Ductile Cast Iron Association was set up to disseminate the lat- Institutions est cast iron technology and engage in R&D. An associa- tion member firm later used centrifugal casting to apply As the casting industry sought to acquire foreign tech- spheroidal graphite cast iron to iron pipes. nologies in the 1950s, a pattern emerged of establishing CO2 gas hardening, which uses water glass rather business associations, often under the leadership of a few than conventional clay as a binder, was introduced into large companies, as exclusive licensees of foreign tech- Japan in 1952 by Hitachi Ltd. Its engineers developed the nologies. This strengthened the weak bargaining posi- N process, in which sand is hardened naturally without tion of Japanese firms as licensees, reducing the overall gas blowing. The FS method was imported by Nippon cost of technology. The Japanese government's frequent Casting Co., Ltd., in 1968. It was reintroduced in an im- intervention to prevent excessive competition among proved form by the Japan High Grade Casting Iron As- Japanese firms helped implement and support the pro- sociation and then disseminated to association members. cess. Shell molding was introduced into Japan in 1956 and Meehanite metal, an innovative, high-grade cast iron disseminated by the Japan Shell Molding Association, set mainly used in industrial equipment and machine tools, up by 26 firms-a number that grew tenfold in the five was introduced into Japan in 1951 by the Mitsui Ship- years after the Association's founding. The association's building Co., Ltd., one of Japan's largest firms under a original role was as sole licensee within Japan. It subli- contract with Meehanite Worldwide, a U.S. firm. Mitsui censed to member firms, but it also worked to acquire set up an R&D subsidiary, the Japan Meehanite Metal Co. and disseminate other technologies from abroad.' Ltd., later the sole licensee for the material in Japan. Ja- The business associations set up in the 1950s did not pan Meehanite Metal sublicensed the patent to 28 firms, limit their work to information dissemination; they also and provided consulting, overseas information, problem- did R&D under the direction of technical committees Table 2.1 Production Trend of Casting Industry (1950-1F979) Total productionTotal production Productivity GNP Working population (ton) (100 mAt. yen) Employees (100 med. yen) (10,000) Year (A) (B) (C) (A)/(C) (B)/(C) (D) (E) (B)/(D) (%) (Q/I(E)(% 1950 974707 351 n.a. n.a n.a 40347 n.a 0.869953156 n.a 1955 1 348898 824 n.a n.a n.a 88646 4194 0.929539968 n.a 1960 2869861 2292 165740 17.31543985 1382888.862 162070 4511 1.414203739 0.37 1965 3717332 3111 151270 24.57415218 2056587.559 320699 4787 0.970068507 0.32 1970 6976458 9652 166796 41.8262908 5786709.513 735031 5153 1.313141895 0.32 1973 7774933 12089 142976 54.37928743 8455265.219 1130899 5289 1.068972561 0.27 1975 5537694 12005 123476 44.848343 9722537.173 1487980 5323 0.806798478 0.23 1976 5977839 13266 117026 51.08128963 11335942.44 1672946 5378 0.792972397 0.22 1977 6241557 14493 110039 56.72131699 13170784.9 1862092 5452 0.77831815 0.2 1978 6273458 15010 107191 58.52597699 14003041.3 2050463 5532 0.73202979 0.19 1979 6971014 16894 106226 65.62436692 15903827.69 2217218 5596 0.761945826 0.19 Source: Kenji Chijiiwa, eTrends in the Japanese Casting Industry (1950-1980)," Sohgoh mono, Vol. 21, No. 2, (1980), p.a3. 6 The Postwar Development of Japan's Casting Industry composed of large companies. Cooperation among com- est capital loans. Priorities then shifted to promoting a few peting firms was fostered by the severe constraints on strategic industries such as synthetic fibers, petrochemi- capital and foreign exchange. cals, electronics and machinery, to which the government Foreign exchange constraints left technological im- gave tax and tariff reductions or exemptions, low-interest port under the strict control of the Japanese government. financing, and import permits. This gave the government great leverage over firms' ac- With the liberalization of trade and private sector de- quisition and dissemination of technology; companies velopment in the 1960s, the significance of government were under tremendous pressure to implement the lat- industrial policy gradually declined. By 1970, the Indus- est advances, since taking a lead in importing technol- trial Structure Council of the Ministry of International ogy increased later profit margins and market shares. As Trade and Industry (MITI) began advising the government the Japan Shell Molding Association demonstrated, busi- on the "maximum use of the market mechanism," a turn- ness associations reduced excessive competition among ing point in Japan's industrial policy. Since then, the role firms eager to implement the technological advances. of government agencies such as MITI has changed from The existence of a few pioneering companies able to direct and interventionist to indirect and intermediary. adapt and improve imported technology was decisive in From the late 1950s, the government designated stra- Japan's postwar industrial development. Their absorp- tegic industries as engines of growth, granting incentives tive capacity, and the efficient institutional arrangements to key industries for limited periods. The machinery in- among firms and associations were key to casting devel- dustry, for example, was designated strategic in 1956, fol- opment in the 1950s and 1960s. lowed by electronics in 1957 and aircraft in 1958. Consultants such as Japan Meehanite Metal Co. Ltd. Promotional policy measures included tariff and tax in- also helped disseminate-technology. However, prefec- centives, such as accelerated facilities depreciation and tural research institutes and keiretsu relationships were reduced corporate taxes, and financial incentives, such as more important. These two routes supported small and long-term, low-interest loans and export promotion incen- medium companies. Later chapters focus on the pecu- tives. liarly Japanese institution of prefectural research insti- MITI established departments to supervise each stra- tutes. tegic industry. Most MITI policies were implemented smoothly because industry associations were consulted Policies beforehand. The National Casting Industry Council was set up in 1945 as a pressure group promoting the casting Industrial policy immediately after World War II was industry, and two years later MITI established a Casting interventionist and regulatory, reflecting the war and Forging Department. In 1961, the casting industry was economy. Policies and priorities have since changed, designated for full promotion under the 1956 Temporary consistent with industrial development and environmen- Measures for Promoting the Machinery Industry and, in tal imperatives. During reconstruction, the priority was 1963, was designated a target industry under the Small to rebuild the nation's economic and industrial infra- and Medium Firms Basic Law of 1963.9 structure and promote basic industries such as coal, iron, Since the early 1960s, the promotion of small and steel, and shipbuilding. These industries were treated medium firms has been a central goal of industrial policy. preferentially in materials allocation, import quotas, low Large companies, especially in the machinery industry, interest capital financing, etc. appreciated the government's policy because their depen- Once the economy had recovered, policy priorities dence on subcontractors. Under international pressure to in the first half of the 1950s turned toward streamlining liberalize trade and foreign exchange at the end of the overall industries through tax incentives and low-inter- 1950s, the government focused on mechanizing and au- 7 Catching Up to Leadership tomating small and medium firms to make manufactur- Act and Foreign Capital Regulation of 1950. Prosperity ers internationally competitive. Policies included im- in the late 1950s spurred deregulation of imports. Fre- proving their technology, training employees, quent government intervention in technological trans- streamlining management, restructuring small business actions between Japanese and overseas firms included to avoid excessive competition and strengthen coopera- advice against domestic competition that would raise tion, and eliminating business practices that weakened royalty payments. With MITI's guidance, Japanese the bargaining position of small firms. firms introduced overseas technology rather cheaply.0 The Small and Medium Enterprise Modernization The government also provided guidance-again to Promotion Law granted tax incentives for capital invest- minimize competition-to ensure that business associa- ment and low interest rate financing from the Small tions became sole licensees for key technologies." Business Finance Corporation. The idea of quality con- In the 1950s and 1960s, the government created in- trol, introduced from the United States and dissemi- centives for technological upgrading by exempting nated by the government, became deeply rooted in equipment from tariffs, allowing tax deductions for ex- small and medium firms. Labor shortages also acceler- perimental research expenses, providing R&D subsidies, ated small-business automation, resulting by the 1970s R&D consignments, and low-interest R&D financing. in increased productivity, improved quality, and re- Subsidies for R&D started in 1950 for mining and manu- duced costs, and a strengthening Japan's international facturing. In the casting industry, TSIs, including busi- competitiveness. ness and academic associations, national and regional Technology policies constituted an important ele- public research institutes, and universities, received ei- ment of industrial policy. In the early 1950s, technologi- ther subsidized or consigned R&D. In 1960 the Materi- cal imports were vital to filling technological gaps, but als Process Technology Center was set up under MITI they were restrained by the Foreign Exchange Control mainly to promote R&D in the casting industry. 8 Industry Overview Production Trends from 1981 to 1990 Figure 3.1 Production Trends by Subsector (millions of tons) 9 Total casting production in 1990 exceeded 8 million tons and amounted to 2,467 billion 8 yen, or 0.56 percent of Japan's GNP. Japan Totalcasting now ranks third in the world, after the U.S. and the People's Republic of China. Almost 6 all products are consumed domestically. S The casting industry consists of seven subsectors-cast iron, cast iron pipes and fit- tings, malleable iron casting, steel casting, 3t.ers copper-based alloy casting, die casting, and 2 ... I _ precision casting. Cast iron is the most impor- 1980 1981 1982 1983 1984 1985 986 1987 1988 1989 1990 tant, accounting for two thirds of the Source: Annual Statistcs of MateHal Process Industres industry's total value.2 Casting production- cast iron in particular-increased steeply from Table 3.1 Production Trends in Cast Iron (11980-90) 1987, reflecting the longest period of postwar (tons) prosperity (Figure 3.1). Production grew at an Year Cast iron Others Total casting annual average of 10 percent in the 1950s, and 1980 4,637,493 2,712,595 7,350,088 1981 4,247,527 2,605,077 6,852,604 nearly as fast in the 1960s (Table 2.1). With the 1982 3,926,470 2,584,122 6,510,592 1983 3,804,479 2,516,882 6,321,361 first oil shock, production fell sharply in the 1984 4,284,966 2,552,147 6,837,113 early 1970s and declined slightly through 1985 4,406,415 2,580,842 6,987,257 1986 4,162,473 2,318,420 6,480,893 most of the 1980s. Cast iron production fol- 1987 4,278,717 2,341,314 6,620,031 1988 4,943,521 2,448,404 7,391,925 lowed a similar trend over the decades. Pro- 1989 5,218,477 2,573,114 7,791,591 duction began to recover after 1987 (Table 3.1). 1990 5,49S,911 2,701,871 8,197,782 The value of cast iron production exceeded Note: Others include pipes and fittings, malleable steel casting, light metal alloy, copper base alloy, die casting, and precision. 1,000 billion yen in 1990. Source: Annual Statistics of Material Process Industries, (August 1991), pp. 20-21. The automobile industry consumed 54 percent of casting industry output in 1990 Table 3.2 Cast Iron Production Share by End User (Table 3.2). Its consumption has grown on (percent) average 9.7 percent since 1965. Output for in- End user 1990 1965 dustrial machinery-the second largest cus- Industrial machinery 17.1 31.1 Other industrial machinery 14.1 37.5 tomer-fell from 31 percent in 1965 to 17 Automobile industry 54.3 12.5 percent in 1990. Other end-users are other in- Transportation machinery 5.1 5.0 peren ~er Electrical machinery 2.6 3.6 dustrial machinery at 14 percent, transporta- Miscellaneous 6.5 10.1 tion machinery at 5 percent, and electrical Note: Other industry machinery includes machinery for mining, road building and construction, metal machinery at 2.6 percent (Table 3.2). working machinery and machine tools, textile machinery and agriculture and fishery. Transportation machinery includes railways, ships and harbour installations. Miscellaneous rolls, ingot molds and mold boards, pipes and fittings, bearings, valves and cocks, household articles, and others. Source: Annual Statistics of Materal Process Industries, (August 1991), pp. 50-54. 9 Catching Up to Leadership Table 3.3 Labor Productivity by Subsector (ton/worker) Pipes and Steel Die Precision With its negative "3-1" image as "dirty, de- Year Cast iron fittings Malleable casting casting casting manding and dangerous," casting has had dif- 1981 86.8 175.5 62.1 53.4 35.7 3.0 1982 85.3 184.9 62.4 50.9 34.8 3.1 1983 85.6 159.7 66.3 48.0 37.0 3.6 highly skilled workers. Employment stood at 1984 94.8 154.5 73.2 56.3 36.8 3.8 1985 98.2 173.2 76.6 58.4 39.4 3.7 87,558 in 1990, down 12.5 percent from 1981, 1986 97.4 153.1 75.3 54.3 40.7 3.6 although total production tonnage increased 1987 107.0 162.0 80.3 58.4 43.9 3.8 1988 120.6 152.0 86.9 65.0 46.1 4.0 19.6 percent in that period, reflecting improved 1989 127.6 133.3 96.4 68.9 49.4 4.1 1990 131.7 169.8 93.6 72.4 52.9 4.4 productivity (Table 3.3 and Figure 3.2). Cast Source: Annual Statistics of Material Process industries, p. 16. iron productivity improved 4.7 percent annu- ally. The number of casting factories decreased from 1,775 in 1981 to 1,513 in 1990.13 Cast iron factories, almost half of the total, also fell from Figure 3.2 Labor Productivity by Subsector 971 to 739 over the period. Most other (tons per worker) subsectors, such as steel and nonferrous metal 200 - 180 - Pipes and fittings casting follow the pattern, although the num- 160 -'- * ber of die casting and precision casting facto- 14 ries both increased 15.8 percent. 120 - 00on Malleale Industrial Organization 80 - 60 - -- - Steel castings While casting firms supply all machinery in- 40 -dustries, machinery assembly manufacturers 20 - rcso atns also produce their own casting products. Inte- Precision castings 0 ------------------------------ grated firmswith in-house cast-parts produc- 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 tion are usually considered automobile or Source: Annual Statistics of Material Process Industries, p. 16. machine tool industries rather than casting firms. Their in-house production was 32 per- cent of total cast iron production in 1990 (Table Figure 3.3 Number of Establishments (Cast Iron) by Size, 1990 3.4a).14 Although in-house production tends to Number of firms 400 increase during recessions, it remained stable 350 -throughout the 1980s. Shares for each type of 300 -production (integrated, specialized and 250 -semispecialized) are given in Table 3.4b. 200 -Integrated firms produce more than 20 per- 150cent of all cast iron. They tend to concentrate 100 -on mass production, farming out less efficient, 50 -diversified and small-lot orders. There remain 29 - i___j ini fields technologically and economically diffi- < 29 30-49 50-99 100-299 300-499 500 < cult to automate. Most specialized firms which Number of employees handle less efficient fields, are subcontractors Source: AnWual Statittics of Material Process Industries. to integrated firms, thus, there is a division of 10 Industry Overview Table 3.4a Share of In-House Production labor between small and medium subcontrac- Year Production (ton) In-house share 1981 4,247,527 35.3 tors and integrated assembly firms. 1982 3,926,470 35.5 Subcontractors fall roughly into one of 1983 3,804,479 35.8 1984 4,284,966 34.2 two categories-keiretsu and non-keiretsu. 1985 4,406,415 34.1 dieri- 1986 4,162,473 35.4 Non-keiretsu firms fulfill orders from diversi- 1987 4,278,717 35.7 fied customers, while keiretsu subcontractors 1988 4,943,521 34.0 1989 5,218,477 32.8 limit their customer base. Especially in auto- 1990 5,495,911 31.9 mobile parts industry, most subcontractors Source: Annual Statistics of Material Process Industries, p. 45. are influenced by a keiretsu relationship of some type. Table 3.4b Production Share (percent) by Type of Firm In 1990 casting factories numbered 1,513, Number of Value Quantity excluding microfirms employing fewer than Type factories (mu. yen) (tonnage) 20. The cast iron subsector had the most- Integrated 43 (5.8) 17,606 (20.0) 102,320 (21.9) Specialized 634 (85.8) 55,400 (63.2) 284,603 (60.9) 739.1 With fewer than 50 employees ac- Semi-specialized 62 (8.4) 14,715 (16.8) 80,351 (17.2) counted for 76 percent of their total (Table 3.5 Total 739 (100.0) 8,772 (100.0) 467,274 (100.0) and Figure 3.3). Since official statistics exclude Source: Annual Statistics of Material Process Industries, pp. 68-72. microfirms, the size of the average firm is smaller than the table indicates. By my esti- followed by Fukuoka at 5.5 billion yen, Saitama at 5.46 billion yen, mate, especially in important production ar- Hiroshima at 5.3 billion yen, and Tochigi at 4.4 billion yen. Aichi eas such as Saitama Prefecture, two-thirds of accounts for 22 percent of all production and is strongly influenced casting factories are microfirms.16 Microfirms by keiretsu, thanks to its proximity to Toyota, Japan's largest auto- produce products based on customer specifi- mobile assemblers. In contrast with Aichi, third-ranked Saitama cations and need not engage in R&D. Often has a localized industrial structure, featuring a long tradition of family-run and the bottom rung of the sub- specialized resources and skills and many small firms in the same contracting system. business fines. Saitama has more companies than any other pre- The major casting areas by production fecture. Its firms enjoy strong support from both national and pre- values are Aichi Prefecture at 20 billion yen, fectural governments. Table 3.5 Number (percent) of Establishments by Number of Employees Cast iron Malleable casting Steel casting Copper base alloy Ughtmetal alloy Die casting Size of establishment numbers numbers numbers numbers numbers numbers < 29 397 (53.8) 15 (40.5) 0 (0.0) 173 (81.2) 104 (60.1) 58 (32.9) 30-49 163 (22.1) 6 (16.2) 72 (61.0) 21 (9.9) 20 (11.6) 36 (20.5) 50-99 100 (13.5) 6 (16.2) 29 (24.6) 12 (5.6) 21 (12.1) 42 (23.9) 100-299 59 (8.0) 9 (24.3) 16 (13.6) 7 (3.3) 19 (11.0) 35 (19.9) 300-490 12 (1.6) 1 (2.7) 0 (0.0) 0 (0.0) 6 (3.5) 3 (1.7) 500-999 6 (0.8) 0 (0.0) 0 (0.0) 0 (0.0) 2 (1.2) 1 (0.6) 1000 < 2 (0.3) 0 (0.0) 1 (0.8) 0 (0.0) 1 (0.6) 1 (0.6) Total 739 (100.0) 37 (100.0) 118 (100.0) 213 (100.0) 173 (100.0) 176 (100.0) Source: Annual Statistics of Material Process Industries, p. 13. 1944,8,663. The Technology-Support System Overview tory facilities, R&D and regular technical seminars and similar meetings. Japan's R&D investment reached 3 percent of the GNP The private sector originally urged the creation of in 1990, the highest in the world. Private firms' expendi- PRIs. They were established and managed by local gov- tures exceed 80 percent of the total, considerably higher ernments, but initially assets such as land, buildings, and than any other developed country, and the large firms major facilities were contributed by leading local com- and their captive laboratories lead the way. Universities panies.18 Most PRIs have just begun to reorganize their contribute 12 percent of the total and public research missions and functions. Now that small and medium institutes 8 percent.7 firms have enhanced their technological level, PRIs have TSIs associated with casting can be classified by shifted toward research of more technologically-inten- ownership and functions (Appendix B). There are seven sive areas. They are setting up networks of local univer- ownership groups-national, semigovernmental, prefec- sities and firms in order to exchange information and tural, university, business association, corporate research create breakthroughs in techniques. laboratory, and private consulting. Governmental insti- The most important semigovernmental institutions tutes are multisectoral. The Government Industrial Re- are the Materials Process Technology Center (MPTC), search Institute in Nagoya, founded in 1952 under MITI's established under the MITI, and the Japan Foundrymen's Agency of Industrial Science and Technology, is the most Society (JFS), set up by the Ministry of Education. MPTC, prominent. Innovative casting materials have been de- which receives financial support from MITI and business veloped there but, because of the technology's maturity circles, influences casting companies. It promotes indus- the institute no longer attracts young and talented re- try-wide development and mediates between gover- searchers. In fact, one well-known governmental ment and business. The government channels consigned institute's recent abolition of its casting section is ex- R&D through MPTC to qualified firms, though MPTC pected to set a trend for prefectural organizations and itself conducts no R&D. It coordinates project teams for universities. large firms and disseminates information. The JFS is an Each of Japan's 47 prefectures has at least one tech- academic organization providing a liaison between busi- nology center. Prefectural research institutes (PRIs) num- ness circles and universities. It is affiliated with the In- ber 171 and employ almost 7,000. The proliferation of temational Committee of Foundry Technical PRIs throughout Japan has worked in favor of technol- Associations, headquartered in Switzerland. Total mem- ogy transfer. They are generally multisectoral, but a few, bership includes 3,500 scholars and corporate managers, such as those in Saitama and Mie, where casting is sev- typically from large, R&D-oriented firms. eral centuries old and of great economic importance, Business circles also have set up a number of TSIs, specialize in casting. These organizations provide local such as the Japan Association of Casting Technology small and medium firms with services, including infor- (1956), the Japan High-Grade Cast Iron Foundry Asso- mation dissemination, troubleshooting and consulting, ciation (1959), and some 10 other associations. Some have training for crafts people and engineers, product and ma- played historical roles in importing and disseminating terials testing and analysis, free use of modem labora- technology. With improved technical capability, however, 12 The Technology-Support System the importance of overseas technologies declined in the experienced, knowledgeable staff people. These firms late 1970s, and these associations reorganized to adapt to often dissolve when their principals retire. the changing environment. Most business associations now promote networking among firms. Some technical Survey of the Casting Industry committees of these associations also contribute to corpo- and Technology-Support System- rate innovation, as is shown in Chapter 5. Methodology and Sampling Some associations function mainly as interest groups rather than as providers of technological support. The Ja- The research is based on a purposive sampling involv- pan Cast Iron Foundry Association (JCIFA), founded in ing interviews with representatives of 18 firms and 10 1938, is the casting industry's most prominent pressure TSIs, plus a national survey of 50 firms that responded group, and it embraces over 90 local cast iron cooperatives. to a mailed questionnaire. The findings of these surveys Its small and medium firm members number almost 1,200. will be presented descriptively, statistically, and through JCIFA and its affiliated cooperatives help the government the presentation of instructive case studies. implement policies promoting small and medium firms. Under the Small and Medium Enterprise Modernization Purposive Sampling of 10 TSIs Promotion Law, these cooperatives helped renovate facili- ties and mediated among those firms to reduce competi- To give a pic-ure of TSIs for Japan's casting industry, the tion and encourage cooperation. sample included at least one for each of the nine catego- One industrial casting cooperative set up in 1905 in ries of ownership structure. The ten intensive survey Saitama, provides member firms with joint purchases of interviews were conducted in the main TS1 categories pig iron and other resources, transportation and shot-blast (l)-(6). In addition, brief interviews were conducted with services, low-interest loans, sales promotion to other pre- representatives of (7)-(9): fectures, joint research on used-sand recycling, training, (1) Governmental research institutes: 1 and laboratory services. Request for its laboratory services (2) Local governmental (prefectural) research has increased to more than 5,000 a month (Table 4.1). Na- institutes: 2 tionwide cooperative are the most important source of (3) Semigovernmental institutions: 1 technology information to local firms em- ploying fewer than 20. Private consulting firms, still new to Japanesebusiness,havenotgrownsubstan- Table 4.1 Monthly Testing and Analysis Requests tially, mainly because they are thought to be Type of requests Frequencies expensive and PRIs still provide free con- Chemical analysis 2,000 etechnical ad- Atomic emission spectography 300 PureMechanical testing 2,500 vice to regional firms. Some consulting Micrography 220 firms are subsidiaries, and others are owned Source: Saitama industrial Casting Cooperative. by retirees of big companies. Casting con- sulting firms are small, and usually limit Table 4.2 Size Distribution of 18 Purposively Sampled Firms their activities to problem-solving and infor- Small Medium Large mation dissemination. As casting firms be- 20-49 50-99 100-199 200-299 300-499 500 come self-sufficient in problem solving, the 2 3 3 1 2 7 Sfr tn Note: The Small and Medium Firm Basic Law defnes smal firms as employing fewer than 20. rom the r etechnological point of view, however, te separation at the size of 50 employees proves more useful. They typically have only a few, but highly Source: Firm interviews. 13 Catching Up to Leadership (4) Business associations: 1 yen in capital. The size of sampled firms is given in (5) Private consulting firms: 3 Table 4.2. (6) Captive laboratories of private firms: 2 The technological capability of each firm was evalu- (7) Academic societies: 1 ated based on nine criteria: (8) Universities: 1 (1) R&D expenditures (share of total sales); (9) Business associations (pressure (2) Number of full-time researchers; groups): 2 (3) Existence of permanent R&D department or labo- Their names are listed in Appendix B, and more details ratory; about the TSIs operations are given in Chapter 4. (4) Possession of industrial rights and properties, that is, patentable technology or utility models (appli- Purposive Sampling of 18 Firms cation patents); (5) Quality of R&D, originality or equal partnerships The 18 firms were selected based on size of firm or tech- with other firms; nical dynamism, type of firm, and regional distribution. (6) Positive attitude toward outside TSIs (member- As Hypothesis 2 holds, a firm's technological level is ships in related associations indicates a firm's de- the most important factor determining which TSIs are gree of information-orientation); useful to it. In Japan, technological capability correlates (7) Participation in academic societies. with company size. The Small and Medium Firms Ba- (8) Awards from the government or business associa- sic Law (1963) defines small and medium firms as hav- tions. ing fewer than 300 employees and less than 100 million (9) Transition from traditional process control based on human experience and skill to auto- -mated control. Based on these criteria, firms are placed in Table 4.3 Purposively Sampled Firms Classified by Technological one of three categories of technological dy- Dynamism Technological dynamism namism-most, medium, and least (Table Size most Medium Least 4.3). As Hypothesis 3 holds, keiretsu influ- Small (20-49) 2 ences whether firms use TSIs, since keiretsu Med 3 E0-99 1 provides technical support to member firms. Large 300-499 2 Thus, both independent or keiretsu firms ___________00 ______<_____3____2 _____2 _ were included. These two types were fur- is,rce pat ther divided based on size and technologi- Table 4.4 Distribution of 50 Returned Questionnaires and 500 Selected cal level: (1) keiretsu core fins (large firms Firms with high technological capacity) and (by firm size, national survey) keiretsu subcontractors (small and medium Size Returned questionnares Selected firms, firms) and (2) independent finrns and non- Small 1-19 5 20-49 13 (36.0%) 288 (57.7%) keiretsu subcontractors.19 Medium 50-99 12 Casting goods are oruses ssbothcia- 100-199 prodced by bse- 200-299 4 (50.0%) 199 (39.7%) cialized and integrated firms (assembly Large 300-499 2 500 < 5 (14.0%) 13 (2.6%) Total 50 (100.0%) 500(100.0%) their own use. As of 1990, the rate of in- a. Out of 691 firms in the membership list of the Japan Cast Iron Foundry Association 500 firms were house production was about 32 percent. selected by the proportionate stratified random sampling method. Source: National firm mail survey. Although these assembly manufacturers 14 The Technology-Support System are categorized in the automobile or machine industry national survey is intended to put case studies in a nation- rather than in casting, the production share of such inte- wide context and to confirm statistically TSIs' contribu- grated firms cannot be neglected, and a few were in- tion to the industry's technological development. cluded in the sample. Casting firms are divided into A proportionate stratified random sampling by com- integrated producers, who engage in casting for their pany size (small, medium and large) seemed more effec- own use in final-goods production, and specialized pro- tive than a simple random sampling. The population was ducers, for whom casting is their main business and who 691 firms. The size distribution is given in Table 4.4. De- sell to outside customers." Since integrated firms are tails are given in Appendix C. rarely sub-contractors, the sample was thus divided into six types: Research Findings on Ten Sample TSIs (1) Integrated and keiretsu. The purposive sampling of ten TSIs clarified the impor- (2) Integrated and independent. tant features of TSIs serving the casting industry. (3) Specialized and keiretsu core firms. (4) Specialized and keiretsu subcontractors. TSI services (5) Specialized and independent. (6) Specialized and non-keiretsu subcontractors. TSIs serve three types of functions: R&D and laboratory activities like testing and analysis; networking, training Regional considerations are important because of the and information services; and problem solving and con- differences in industrial organization among geographic sulting. R&D is the most crucial and sophisticated. Five areas. The survey covered eight prefectures. As of 1990, of the 10 TSIs mainly provided research and laboratory Aichi Prefecture ranked first, accounting for a quarter of services through national and local research institutes, all production, followed by Fukuoka, Saitama, universities, and private laboratories. Hiroshima and Tochigi. Aichi is strongly influenced by Most business associations function as coordinators the keiretsu structure. Third-ranked Saitama and seventh- or organizers and do not engage in R&D, but the value ranked Mie exhibit a unique industrial organization of these TSIs should not be underestimated, since net- which might be called "localized", which features a long working and information exchange frequently lead to in- tradition of specialized production based on local re- novation (see Chapter 5). Most associations have a sources and skills and a concentration of small and me- technical committee composed of large R&D-oriented dium firms in the same business. Localized casting firms members who meet regularly and study pertinent tech- enjoy an "agglomeration advantage" and plentiful sup- nology and industry strategies. Because individual firms port from central and local governments, whose research maintain strong proprietary restrictions on new technol- institutes support those firms technologically. Such pat- ogy, these committees exchange only certain information. tern is also found in the garment and textile industries. But they do give members "seeds" for R&D. An overview of the 18 firms sampled is given in The third function, problem solving and consulting, Appendix A. lends important assistance to firms in quality control and productivity improvement. National Survey of 50 Firms TSIfeatures The national survey consisted of questionnaires mailed to 500 firms in which 50 answered. While case studies in Human resources. The dynamism of an R&D-oriented TSI Chapter 5 are based on the purposive firm survey, the greatly depends on its staff. Most experienced and capable 15 Catching Up to Leadership researchers and engineers are older and the number of un- over five years. Its director was discerning about the cost dergraduates majors in metallurgy and related subjects and benefit of R&D. has declined. The shortage of younger researchers is a serious problem facing the casting industry and TSIs. The Sensitivity to client needs. TSIs other than private cap- captive laboratories of large firms still attract graduates tive laboratories do not try to meet client needs or bring who are systematically trained in-house. Thus, R&D is in more clients. They prefer informal, daily channels for moving to captive private intitutes, and the shortage of understanding client needs. Formal channels, if used, R&D staff is becoming a constraint to non-captive TSI. tend to be seminars or workshops. For business associa- Training. R&D-oriented TSIs also emphasize training tions, the board of directors is the most important chan- for their own employees. They prefer a personalized way nel. The number of members may indicate whether a TSI of training or on-the-job training to formal or outside train- satisfies client needs, but research showed that most ing. Given staff limitations, researchers at non-R&D TSIs firms joined business associations for no clear reason. must concentrate on routine work and do not get much Perhaps their greatest motive may have been to collect professional development. Thus, their functions will be information on other companies' activities. In today's reduced to coordinating intercompany networks with strained economy, however, firms are reviewing mem- little opportunity for training. berships costs, and some are proposing merger of asso- Promotion. Like most Japanese organizations, TSIs ciations to avoid overlapping functions. have used seniority as a promotion criterion. Even con- Relationship to outside organizations. Most R&D-on- sulting firms do so. It mitigates jealousy and competi- ented TSIs belong to academic associations like the tion and encourages cooperation, but it also discourages Foundryman's Society, where large firms and universi- talented young researchers. Again, however, the chang- ties can engage in information and social exchange. Pre- ing management environment has started to erode se- fectural institutes maintain connections with local niority, and R&D-oriented TSIs have begun favoring universities and also with national research institutes. achievements such as the number of research papers published, techniques patented, and contributions to Case studies of dynamic TSIs firm revenue. Labor turnover, headhunting and spin-off The senior- "Dynamism" denotes the flexibility to change with times. ity system traditionally depressed turnover. Employees In order to break out of conventional R&D and become rarely quit before retirement, headhunting is still rare. truly innovative, TSIs must strengthen their R&D capac- Starting one's own firm occurs only after retirement. ity. The two cases that follow illustrate dynamic TSIs. Cost consciousness. Since some TSIs are non-profit Case 1. Case 1 is an R&D oriented institute founded organizations and some are publicly financed, i.e., by the in 1985 as an affiliate of a large firm. It is soon to be spun government or prefecture, they do not pay much atten- off as an independent entity. Cited above, it has displayed tion to cost control compared to private institutes. And the strongest cost-consciousness, achieved important in- consulting services are perceived to be inexpensive or novations and enjoyed a tripling of its revenues. Case l's even free by the client company. Eight of the 10 TSIs an- R&D staff is positively involved in academic societies swered that their charges were unrelated to market and cultivates its connections with other TSIs. The insti- prices. Their lack of awareness may resulted in stagnant tute takes its R&D staff training very seriously, using ex- or long-term declining revenues. TSIs with a member- ternal training schools, encouraging participation in ship system are financially vulnerable to a decline in in-house and outside seminars, and promoting univer- members. The only exception, a private institute that is sity study. It has attracted younger researchers, whose one of the most dynamic TSIs, has had its revenue triple ranks have doubled to 54 within the last five years. 16 The Technology-Support System Case 1 is sensitive to the needs of its clients (medium and large firms). Its client base has Table 4.5 Technology Capacity by Firm Size Average number Average R&D ratio risen from 35 to 50 over five years, and it ex- Size group Numbers of employees to total sales pects to reach 80 customers in the near future. 1-19 S 16 0.33 The institute prefers formal channels in gaug- 20-49 13 35 0.53 50-99 12 69 0.49 ing client needs, such as staging presentations 100-199 9 142 0.78 m 200-299 4 223 1.43 of innovations at seminars. It participates 300 < 7 2,610 2.04 meetings of academic societies. It emphasizes Total s0 patents and published papers in promotions Source: Appendix C. rather than seniority. Case 1 differs radically from other TSIs surveyed. Case 2. The Mie Prefectural Research Insti- tute serving a prominent casting area also has proven an exception, given the decline in the need for most PRIs. R&D, laboratory services, Table 4.6 R&D Levels by Type of Firm technical consulting, and training are its main services. With the increased emphasis on inno- Type of firm Number of employees share (%) staff share vative and basic technologies, the Mie PRI re- Keiretsu Core 21 298 1.076 11.4 focused on R&D, especially in advanced fields. Independent 13 297 1.054 10.3 Keiretsu Subcontractor 7 51 0.571 8.6 With the serious threat to the casting industry Non-keiretsu Subcontractor 9 70 0.111 7.7 posed by labor shortages, Mie PRI has empha- Total 50 sized training services, providing primary, sec- Source: National firm mail survey. ondary, and advanced courses for engineers which are highly evaluated by client compa- nies. Since 1991, it has conducted advanced R&D with firms. Together they have worked to overcome the low fracture resistance that is a co mn rbemi cmoit atras,u- Degree of competition Average share of R&D to total sales(% common problem in composite materials, us- Very Severe 1.07 (N=1 18) ing innovative techniques such as fiber-rem- Severe 0.85 (N= 6) forced ceramics. Government organizations Moderate 0.53 (N=1 1) Little or None 0.25 (N= 4) support the effort by providing the required Source: National firm mail survey. fibers. Some PRIs have tried to rejuvenate them- selves by focusing on basic R&D, but have failed because of serious R&D manpower short- age. the Mie PRI has attracted younger re- searchers thanks to its proximity to a college Table 4.8 Firm Size and Frequency of TS1 Use with a reputation in metallurgy and casting. Size Average number of TSIs used The Mie PRI's director generously encourages 1-49 2.8 50-99 3.8 his R&D staff to concentrate on research, which 100-199 4.4 200-299 6.3 is a great incentive to work hard. Since casting 300499 s.5 is still important to the Mie economy, local gov- 500 < 5.6 emnment provides full financial support for the rage SRI. Source: National firm mail survey. 17 Catching Up to Leadership Mie's systematic surveys of the prefecture's casting by their parent firm. As the parent firm faces competi- firms enable it to grasp their needs and to offer them tion, keiretsu core and subcontractors are pushed to im- carefully conceived services. Many other PRIs are in a prove technology, cut costs, raise quality, and quicken dilemma if they were to concentrate on basic R&D, they delivery. Now, the parent firm selects more R&D-ori- would become superfluous to regional firms. Mie avoids ented subcontractors. Keiretsu subcontractors generally the quandary by making its regular need surveys. are more R&D-oriented than their non-keiretsu counter- parts, as Table 4.6 shows. Keiretsu subcontractors did Evaluation of Technology-Support System surpass their non-keiretsu counterparts in average R&D share (Table 4.6) by five times-0.57 percent for keiretsu The national survey of 50 firms indicate which type and and 0.11 percent for non-keiretsu. size of firms use which TSIs, how they use them, and how the TSIs are useful (for their profiles, see Appendix C). Competition and technological dynamism Technological dynamism Despite its dwindling number of firms, casting is the scene of fierce competition. All of the keiretsu subcontractors re- The national survey questionnaire allowed firms to rank ported that parent firms were demanding sweeping cost their own technological capacity on a scale from 1 (low- reductions and strict deadlines, by having them compete est) to 5 (highest). Nine graded themselves 5, 20 gave 4, with each other. In response, firms who perceive strong 20 said 3, and one said 2. Although the answers lack ob- competition have increased their R&D investment (see jectivity, they may be instructive. Table 4.7). It is appropriate to examine how competition Firm size and R&D. Nineteen of the 50 firms conducted affects firms' use of TSIs. no R&D.2' Firms with more than 200 employees conducted some R&D, while 56 percent of the small firms, i.e., those Dynamism and the use o TSIs employing fewer than 50, did none. R&D expenditures and staff size relate directly to the number of technology An R&D-oriented attitude makes firms more informa- patents.2 The correlation coefficient between staff size and tion-oriented, spurring them to diversify their sources. the ratio of R&D expenditures to total sales is as high as Thus, firms that feel competition use TSIs more fre- 0.84 (Appendix C). The correlation can be also confirmed quently. R&D orientation is correlated with firm size. in Table 4.5. R&D-oriented firms are naturally information-oriented. Type offirm and R&D. Twenty-eight firms have some Thus, the larger in size the firm, the more information- sort of keiretsu relationship. Of them, 21 are keiretsu core oriented it becomes, as Table 4.8 indicates. Keiretsu affili- firms, and seven are keiretsu subcontractors. Thirteen are ation appears to affect TSI use. Among firms with the independent firms and nine are non-keiretsu subcontrac- same level of technological capability, independent firms tors. The relationship between firm type and R&D capac- use on average more than five TSIs compared to 3.8 by ity is not clear. keiretsu core firms. The effect of the keiretsu on subcontractor's R&D The pattern is reversed for subcontractors. Keiretsu efforts reveals a paradox.2 The keiretsu relationship subcontractors use 3.7 TSIs; use 2.9 non-keiretsu firms. guarantees long and stable transactions for subcontrac- Keiretsu subcontractors are generally more R&D-or- tors, so they might tend to become less vigorous in ented. For keiretsu core firms, the parent firm and related R&D. On the other hand, some keiretsu subcontractors other keiretsu core affiliates are the most important source work to raise their R&D capacity in response to requests of technology and information. Independent firms, ex- 18 The Technology-Support System cluded from keiretsu groups, must search for diversified pared to only 29 percent of keiretsu subcontractors. information channels. Keiretsu core firms tend to depend more on outside trai- This analysis suggests that both the factors of tech- ing institutions, including parent companies', and less nological capacity and the keiretsu relationship, influence on in-house training than independent firms. Keiretsu TSI use. Technological capacity may be more crucial of firms, including subcontractors, make greater use of the two factors: the less R&D-oriented the firm, the lower more sources of training. the frequency of TSI use. Evaluation of TSIs Firm size and the use of training institutions Casting firms' most important technological sources are, The size of a firm correlates positively with the frequency in order, customer or parent firms, PRIs and business as- of staff training. Almost half of surveyed small firms used sociations and suppliers (Table 4.10). Thirty firms out of no training; all large firms did. Firms employing more 46 used customer or parent firms and 28 of 30 firms said than 300 trained in-house and through outside institu- that customer or parent firms were very useful. Twenty tions. In-house training was more often used than out- of 28 firms which used PRIs evaluated PRIs as very use- side ones (Table 4.9). Keiretsu relationships also function ful or useful. Business associations were regarded simi- as important sources of training for member firms. Some larly. Sixteen of 24 regarded supplier firms favorably. 52 percent of keiretsu core firms, for example, depended Casting firms perceived universities, academic asso- on parent firms for training, while 31 percent of the in- ciations and national institutes, which concentrate on dependent firms used customers. basic technology, as less important sources of technology. Keiretsu subcontractors depended on customer or This is mainly because that Japanese firms prefer devel- parent firms for 43 percent of their staff training, while opment and applied R&D to basic R&D. Besides, they only 22 percent of non-keiretsu subcontractors did. Two- see the serious staff shortages in these institutions. Fif- thirds of non-keiretsu firms used no training at all, com- teen of 46 firms used universities, but only 10 favored Table 4.9 Firm Size and Use of Staff Training (percent) Firms with Outside In-house Both types Size no training training or OJT of training 1-49 s0 39 28 17 50-99 25 75 50 50 100-199 22 56 56 34 200-299 0 50 75 25 300 < 0 71 86 86 Table 4.1 0 Evaluation of Major TSIs by Firm Size and Degree of Appreciation of TSI Support TSIs Parent/consumer % PRIs % Business associations % Size Firmss Rate of use Appreciation rate Rate of use Appreciation rate Rate of use Appreciation rate <450 1t 53 47 73 53 60 33 50-299 24 75 71 58 42 79 50 300 < 7 57 57 43 29 57 43 Total 46 65 61 61 43 70 43 a. Number of firms using at least one TS8 Four out of the 50 sampled firms did not use any TSL "Appreciation Rate" means the percentage of all TSI-user firms perceiving TSIs favorably. Source: National firm mail survey. 19 Catching Up to Leadership by PRIs and supplier firms. In joint R&D, Table 4.11 Policy Menus Used by Firms PRIs and parent or customer firms are Policies Number of firms Evaluation followed by universities. PRIs and busi- Fiscal (tax) incentives 10 Very useful ness associations provide testing and Grants 13 Very useful Low-interest loans 5 Fairly useful Training subsidies 3 Useful facilities, however, even small firms have Governmentt procurement 0 - Standards (JIS) 4 Useful become more self-sufficient. Business as- Protectionism 0 sociations mainly provide intercompany Export incentives 0 - Miscellaneous 0 -networking services. Note: Since governmental policies are available only for small and medium firms, only nine firms re- Fifty-six percent of the firms used busi- sponded to this question. Source: National firm mail survey. ness associations, and 70 of the firms regarded them favorably. Those employing fewer than 20, which were them as technology sources. The larger the firm, the more passive toward R&D and the least information-oriented, likely it was to regard the university as a partner in re- were often excluded from information networks, except search or consigned R&D. through customer firms and local cooperatives. Because casting is no longer strategic, firms rarely Table 4.10 supports the hypothesis that a firm's tech- organize research cooperatives for special themes. In- nological capacity and sources of support are a function stead, the permanent technical committees of business of its size. Small firms, usually less technologically com- associations fulfill the role. Use of consulting and private petent, used PRIs more often and regarded them more research firms is also rare, and joint ventures with for- favorable than did large firms. Medium firms (50-299 eign firms are virtually nil. employees) said they relied on customer firms. Less than Customer requirements and competition are the 30 percent of the large firms regarded PRIs favorable, a most critical factors promoting technological improve- matter of course since PRIs serve smaller companies. ment in firms. Most R&D is demand-driven, and tech- Large finms, in general highly technologically competent, nology support accompanies ordinary transactions. For also appreciated customer firms less as a technology subcontractors with high R&D capacity, customer or source than did medium firms. Large firms viewed TSIs parent firms often share in R&D from the first stage of as one of a number of information channels. Actually, design. Close cooperation between assemblers and sub- most TSIs are technologically supported by larger firms contractors is based on mutual trust. They exchange in- through being provided lecturers, initiated joint R&D, formation and discuss improvements. Such close and offered advice to TSIs. cooperation allows subcontractors to understand cus- tomer needs. Keiretsu members cooperate in R&D more Evaluation of Technology Policies often than do loosely organized subcontractors. Firms rely upon TSIs for training, information dis- Although the government's support for the casting in- semination, problem solving, joint R&D, and testing and dustry has declined, it maintains non-industry-specific analysis. For training, firms rely on their customer the support policies for small and medium firms from which most, followed by business associations, PRIs and pre- most casting firms benefit. fectural technology centers. For information services, Firms with fewer than 300 employees and capital customer firms, business associations, and supplier or assets of less than 100 million yen are entitled to the trading firms are used most often. In problem solving, policy support which includes modernization of facili- customer firms are the major source of service, followed ties, upgrading of technology levels, management 20 The Technology-Support System streamlining, restructuring of the small-business sector, of the fifty firms surveyed by mail did not answer eliminating business practices disadvantageous to whether they took advantage of these policies. Perhaps smaller firms, promoting small firms' products, guaran- some used no technology-related policy, others were teeing fair business opportunities, and providing benefits unsure if the policies they used were technology-re- for laborers and stabilized labor relations. lated.1 In any case, 25 of the 50 used at least one policy The purposive firm survey showed that nine small measure. and medium firms received benefits. They most wel- The purposive firm survey suggested that casting comed fiscal incentives, especially depreciation discounts firms would appreciate government assistance in train- (Table 4.11). Seven of the nine took advantage of low-in- ing R&D researchers and engineers; R&D related to pol- terest loans.24 The Japanese Industrial Standard (JIS), lution prevention, energy efficiency and production originated in 1949, led to government-sponsored qual- automation; computer-control technology and software; ity control campaigns. Almost 90 percent of JIS-autho- innovations in materials such as composites; and auto- rized factories are owned by small and medium firms mation of small-lot and diversified-product processes. and receive preference in government procurement, The appreciation of the yen has severely hurt the cast- items like manhole covers, ditch covers and street lights. ing industry. Many machine and automobile assembly Three firms regarded JIS as fairly important. Four ben- companies have begun shifting production overseas, es- efited from subsidized training course, particularly for pecially to Asia, decreasing domestic demand. This, along computer. Two firms used R&D subsidies. They received with the current depression has made it vital that casting mixed reviews, as subsidies are limited and application firms become R&D-oriented and pressure the government procedures are complicated and time-consuming. Half to implement more technology-related policies. 21 Case Studies: Firm and TSI Interaction in Technological Change This chapter focuses on interaction between firms and parent and Toyota's Central Research Institute. Firm A also TSIs, taking into account the keiretsu relationship, to pro- employs other core firms, with an influx of information vide insight into how individual TSI function. Firms are provided via overseas subsidiaries to get technological divided into keiretsu and non-keiretsu for the sake of con- information. venience. The keiretsu embraces myriads of subcontractors in The first section compares two groups having the a complex pyramidal and multilayered organization. same technological level. This is followed by the presen- New technologies developed by the parent or core firms tation of other cases of how TSIs contribute to firms' in- may be transferred to other keiretsu members, including novations, based on the purposive firm survey. Table 5.1 subcontractors, although critical technology is confined shows R&D staffing compared to total staff. to certain firms. In return for the transfer of technology, member firms obey the parent firm's instructions on cost Comparison of Keiretsu Core Firms reduction, quality, and timely delivery. The parent firm and Independents contracts out less efficient production jobs to subcontrac- tors. During recessions and depression, subcontractors, Firm A is a prominent keiretsu core firm that specializes in especially those on the lowest rungs of the hierarchy, bear casting and maintains top technology. It performs tech- the severest conditions. Thus, despite the burden of R&D nological development, adaptation and basic R&D in its investment, the parent firm enjoys advantages through captive laboratory. Under the parent, an important auto the keiretsu. parts manufacturer, are nine core firms, including Firm A, Firm A belongs to a variety of business and academic all belonging to the same command group. The group associations to diversify information sources. It uses TSIs belongs to Toyota. Labor is divided among core firms, with to collect information on the R&D activities of other Firm A specializing in casting. Firm A often sends R&D firms, especially those in other keiretsu groups. The dif- staff to the parent company's laboratory to exchange in- ferent keiretsu groups themselves rarely exchange infor- formation and perform joint research; it also exchanges mation directly. Firm A depends on various outside TSIs information with other core member firms. Its most im- for technical training, in addition to its in-house program. portant sources of technology are the laboratory of the The parent firm also is an important training source. Firm employs__ _ _ _ _ _ _ A has no connection with public re- search institutes. Firm B is an independent spe- Table 5.1 Average Number of R&D Staffs by Firm Size cialized producer with R&D capacity Average Average R&D Average number o th t mem includi Size of firms employees staffs (A) of other engineers (B) (A)+(B) cmaal ota fFr .Pei Small (2 Firms) 36 1.5 1.5 3.0 ously a subcontractor whose R&D de- Medium (7 Firms) 132 3.4 6.3 9.7 pended on a few key customer firms, Large (9 Firms)t 1o529 133.8 418.3 552.1 Firm B improved its technological ca- a. Due to diversification, average figures in large firms include staffs unrelated to casting technology. y an in r in s Source: Firm interviews.and ime de iverpan im transactions to a wider range of custom- 22 Case Studies: Firm and TSI Interaction in Technological Change ers. Being a keiretsu firm guarantees stable and long-term and business associations because the customer firms transactions, but concentrating all one's transactions in lend no technical support. one or two baskets is risky during a recession. Thus, some Firm E, employing 24, is non-keiretsu. It uses PRI subcontractors, like Firm B, have worked to diversify and business associations, but local cooperatives are its their relationships with customers in order to achieve most important source. Although the least dynamic in greater independence. Compared to Firm A, Firm B ac- terms of R&D, Firm E not only benefits from the coop- quires more R&D information from outside TSIs. Its eratives technical services but from their power as pres- important sources are overseas firms and research insti- sure groups. Firm E survives not by technological tutes, and, to a lesser extent, domestic universities and improvement but by its political access to market share, academic societies. Firm B makes the most of outside TSIs achieved through participation in small and medium to increase its access to scientific "seeds" and maintains firm cooperatives. It is in Saitama Prefecture, an impor- personal connections with professors whom it seeks for tant casting regions, where it enjoys support both from joint R&D activities. central and local governments. Small firms, including Firm E, collaborated in the cooperatives during reces- Comparison of Keiretsu and Non-keiretsu sions. This unique phenomenon, partly explained by Subcontractors the Japanese tradition of mutual assistance, is most prominent in localized industrial areas like Saitama. The analysis shows that keiretsu subcontractors are more With the cooperatives, small firms with diversified R&D-oriented than their non-keiretsu counterparts; they transactions feel no need to join other business organi- use more TSIs. This polarization of subcontractors began zations. in the late 1970s, when assemblers and customer firms, Small firms generally cannot afford to train staff. facing economic upheaval and international competition, However, in the case of keiretsu subcontractors, customer began selecting more R&D-oriented subcontractors. Sub- firms provide support. More precisely, training is an contractors with lower technological capabilities obligation. Customer firms undertake to ensure high dropped out of keiretsu groups and became even weaker quality of products. But, non-keiretsu subcontractors, like without the technological support of major customers or Firm E, receive no support from customer firms. buyer firms.26 Firm F, employing 53, is an unusual keiretsu subcon- Non-keiretsu subcontractors tend to depend on pre- tractor and is a rare small firm that is R&D oriented, works fectural research institutes. Firm C, employing 32, almost to diversify its R&D sources, and makes the most of the two decades ago, had belonged to a prominent keiretsu services of TSIs, particularly PRIs. Firm F recently devel- group and received technical support from parent firms. oped innovative technology working with PRIs. Only four When its keiretsu relationship was canceled, Firm C people work in R&D, but they consulted with the indus- turned for the first time to the Saitama PRI for support. trial technology center, universities, and other TSIs on The PRI provides training, seminars, advice and free technical information and the use of moder facilities, etc. problem solving. The firm also uses a local casting co- The R&D staff later found a specialist in their technology operative. The cooperative provides marketing informa- of interest at PRI in remote Hokkaido. The researcher there tion, a stable supply of materials including iron and steel worked in joint R&D for some years, and the result was a at discount, and testing and analysis. valuable patent. Firm F shows how the will and attitude Another case of a non-keiretsu subcontractor is Firm of the company owner is crucial to the firm's success. Even D, employing 48. It produces parts and machine tools for with customer support, Firm F went further, was earnest more than 50 customer firms. Firm D depends on the PR in its staff training, and used business associations to net- 23 Catching Up to Leadership work. The non-keiretsu cases above also show that less dy- Firm H, employing 1,000, took advantage of a techni- namic firms may not exploit the potential of TSIs. The cal committee of a business association. As a keiretsu core point could bear further research. company, Firm H specializes in casting for its parent au- tomobile assembly company. The parent firm and keiretsu Other Case Studies group are its most important sources of technology. The group consists of almost 200 companies, including 40 core Firm G is an independent medium-sized company, companies that frequently discuss new ideas and research. it has 80 employees and makes manhole and ditch cov- Firm H also maintains connections with business and aca- ers. Competition is severe, both domestically and over- demic associations. It sits on technical committees of the seas, partly because the technology is mature and partly Japan Malleable kon Casting Association, Material Pro- because certain Asian nations, especially China, with cess Technology Center, Japan Foundrymen's Societ and low-wage labor have entered the market. Despite lim- the Japan Association of Casting Technology. The commit- ited R&D manpower, Firm G does its own R&D. It is tees usually consist of representatives from larger firms information-oriented and makes the most of outside with high R&D capability. Through these regular ex- TSIs, particularly PRIs. It received a government consign- changes, Firm H acquired the seed for patented technol- ment from the Small and Medium Enterprises Agency ogy. to do R&D on new ways of core making, and studied new Firm H contributes technologically to the keiretsu technology thought to be superior to conventional meth- group. Until two decades ago, however, the firm had ods in cost reduction, energy efficiency, and high produc- depended heavily on technological transfer from the tivity. Firm G's local PRI is active in assisting information parent firm. It had approached as many TSIs as possible, exchange among firms, local universities, and national including national and prefectural research institutes in research institutes and takes the initiative in bringing Kanagawa and Tokyo, to improve its R&D. As its R&D firms in different business lines together. Through the capacity improved, however, it came to rely less and less PRI's mediation, Firm G came to contract an iron-work- on them. Firm H, which upgraded its R&D capabilities, ing firm and, with its cooperation devised an innovative built a fragile technological base into a robust organi- technique involving austenite ductile iron, making man- zation enabling it to become independent of the TSIs hole covers and ditch covers lighter. Despite its limited that had supplied R&D. Its success lay in the serious- human resources, Firm G could develop new technology ness of its attitude toward R&D, the use of TSIs, and cheaply with full use of TSI assistance. the effectiveness, in turn, of TSIs assistance. 24 Conclusions and Lessons for Developing Countries The casting sector is a microcosm of Japan's rise to glo- often assisted as initiators ard honest brokers in business bal technological leadership. Institutional and policy networks. Small firms have been served by business co- support was critical to its firms' adoption and mastery operatives, which provide bulk purchasing of inputs, of technologies. While post-war process innovations in joint marketing, training, technical consulting, and joint casting techniques and materials were available through research on specialized issues like the recycling of pro- formal channels such as licensing, TSIs made acquiring cess materials. Keiretsu networks, by internalizing costs and diffusing techniques more efficient and cost-effec- and benefits of learning, allowed large assembly firms tive. in the machine industries to collaborate with their cast- In the early post-war period, for example, the ing subcontractors in adopting and mastering improved government's attempts to control foreign exchange trans- production techniques. actions pushed the private sector to cooperate in licens- While supporting cooperative private-sector technol- ing foreign technology through business associations, ogy activities, the Japanese government targeted direct thereby strengthening local industry's bargaining posi- institutional support to small and medium-size and in- tion and ensuring that imported technology would be dependent casting firms to ensure that they were not left shared widely. Public TSIs disseminated technical infor- out of technological upgrading. Financial and material mation, conducted training, and assisted firms in orga- support was given to local or regional institutions to nizing cooperative research to absorb imported serve smaller and non-keiretsu firms. PRIs often worked technology. In localities with concentrations of indepen- with industry associations and local cooperatives to pro- dent casting firms, PRIs provided consulting and tech- vide information services, training, trouble-shooting, nical services. Only in recent years, as the in-house testing, laboratory use, and R&D. Special TSIs served technological capabilities of casting firms matured, have small firms as part of a larger policy, launched in the TSIs shifted their focus to R&D. 1950s, to modernize small and medium-size firms Government policies encouraged public and private through low-interest loans, tax incentives, training, and institutions to promote technology diffusion and tech- technical assistance. nical upgrading. Private sector institutions were particu- Public TSIs'direct support to independent and small larly important, and government programs often served firms complemented the keiretsu networks in upgrading as a catalyst for private sector cooperation for improv- the technological capabilities of the casting sector. Sur- ing production technology. In addition to pressing for vey results show that, among medium-size and larger collective technology licensing to save foreign exchange, firms, specialized independent producers lacking access the government implemented subsidies and incentives to a keiretsu group's technological resources sought pub- through business associations and semigovernmental lic TSI support. Small firms that were keiretsu subcontrac- bodies. Apart from on-licensing imported technology, tors were stimulated by their main customers' exacting industry associations disseminated information to mem- quality standards and technical assistance to become bers and arranged research, though their role in joint more technologically dynamic, and they sought PRI sup- research diminished as firms mastered the major process port more frequently than non-keiretsu firms. Since a technologies. Government and quasi-government TSIs small firm's keiretsu membership usually depended on 25 Technology-Support Institutions and Japan's Casting Sector continuous technical improvement, TSI support pro- countries. Technical assistance can be encouraged vided crucial reinforcement to the technological benefits through incentives, and government can offer technical of keiretsu networks. support to local subcontractors. Business associations can Japan has unique private sector cooperation and in- help deliver incentives and technical services to mem- stitutional support for small firms. But its institutional bers. Public TSIs can also foster private sector network- and policy support for technological upgrading offers ing for research and training. guidelines for late-industrializing economies. Most im- Special institutional and policy support should be portantly, newly industrializing countries can encourage directed toward the small and medium-size firms that cooperation to diffuse technology through public, pri- often predominate in casting and related industries. In- vate, and semigovernmental mechanisms. In casting and stitutions diffusing improved production techniques to other materials processing industries, many key tech- such firms are a key part of broader policies and incen- nologies are no longer under patent and are thus easily tives encouraging modernization. Local TSIs that sup- licensed and distributed in a quasi-public manner. An port small and medium-size firms might be established emphasis on diffusion can raise productivity and qual- where firms are concentrated and encouraged to work ity and strengthen firms' capabilities. with business associations and firm cooperatives. TSI Government should use institutions to encourage support to small firms in information services, training, private sector cooperation. Keiretsu structures cannot be testing, and trouble-shooting can complement the de- easily copied, but subcontracting networks are promi- velopment of technical cooperation through subcon- nent in the machinery and electronics sectors in most tracting. 26 Appendix A Profile of 18 Firms in Purposive Firm Survey Profile of 18 Firms in Purposive Firm Survey Firm number Size (employees) Location/prefecture Type of firm R&D (mil. yen) 1 3,511 Niigata Integrated/independent 2,100 2 2,910 Aichi Specialized/keiretsu core 280 3 2,660 Tochigi Integrated/keiretsu parent 17,000 4 1,523 Shizuoka Specialized/independent n.a 5 1,251 Ibaragi Specialized/keiretsu core 150 6 532 Kanagawa Specialized/keiretsu core 165 7 524 Saitama Integrated/independent 176 8 425 Mie Integrated/independent 1,500 9 412 Saitama Specialized/independent n.a 10 267 Saitama Specialized/keiretsu subcontractor 9 11 192 Kanagawa Specialized/keiretsu subcontractor 427 12 140 Saitama Specialized/independent 67 13 127 Aichi Specialized/keiretsu subcontractor 35 14 87 Mie Specialized/independent n.a 15 58 Saitama Specialized/keiretsu subcontractor 4.4 16 53 Niigata Specialized/keiretsu subcontractor 5.8 17 48 Saitama Specialized/keiretsu subcontractor n.a 18 24 Saitama Specialized/keiretsu subcontractor 1.6 Note: For integrated firms, R&D expenditures include noncasting R&D expenditures as well Source: Firm interviews. 27 Appendix B Major Technology-Support Institutions by Main Purpose and Ownership 1. Research Institutes and Laboratories 1.1 Governmental research institutes Government Industrial Research Institute, Nagoya (Industrial Science and Technology Agency, MITI) 1.2 Prefectural research institutes Saitama Prefectural Casting and Machinery Research Institute Mie Prefectural Metal Research Institute 1.3 Universities Kagami Memorial Laboratory for Materials Science and Technology (Waseda University) 1.4 Private captive research laboratories(a) 2. Information and Intercompany Networking Coordination 2.1 Semigovernmental institution Materials Process Technology Center (MITI) Japan Foundrymen's Society (Ministry of Education) Japan Information Center for Science and Technology 2.2 Business associations Japanese Association of Casting Technology Japan Die Casting Association Japan Light Metals Association 3. Problem Solving and Consulting 3.1 Private consulting firms Tokyo International Casting Technology Consulting Japan Meehanite Metal Associated Foundry Engineering Co. Ltd. 4. Interest Groups (Business Associations) Japan Cast Iron Foundry Association Japan Malleable Iron Casting Association Steel Casting and Forging Association Note: This project avoids mentioning the names of specific private companies and their in-house laboratories in order to protect their privacy. 28 Appendix C National Survey Data (50 Casting Firms) National Survey Data (50 Casting Firms) Number of Total sales R&D R&D returned Firm Employees R&D Staffs (mil. yen) (mil. yen) (%) Group size questionnaires number (A) (B) (B)/(A) (%) (C) (D) (D)/(C) 1-19 5 1 9 0 0.00 180 0.00 0.00 2 15 1 6.67 1,600 0.00 0.00 3 17 4 23.53 400 8.00 2.00 4 18 2 11.11 220 0.00 0.00 5 18 1 5.56 400 0.00 0.00 20-49 13 6 20 2 10.00 280 0.28 0.10 7 20 1 5.00 400 8.00 2.00 8 21 1 4.76 140 0.00 0.00 9 28 3 10.71 350 0.00 0.00 10 30 1 3.33 440 0.00 0.00 11 32 0 0.00 510 0.00 0.00 12 35 2 5.71 1,800 18.00 1.00 13 36 5 13.89 740 2.22 0.30 14 37 4 10.81 900 0.00 0.00 15 43 2 4.65 620 0.00 0.00 16 44 4 9.09 400 0.00 0.00 17 45 5 11.11 690 13.80 2.00 18 48 2 4.17 660 6.60 1.00 50-99 12 19 50 5 10.00 180 0.00 0.00 20 56 3 5.36 1,100 11.00 1.00 21 57 4 7.02 2,230 66.90 3.00 22 60 5 8.33 1,200 3.60 0.30 23 61 4 6.56 680 1.36 0.20 24 64 3 4.69 1,700 0.00 0.00 25 68 3 4.41 2,100 2.10 0.10 26 70 0 0.00 2,000 0.00 0.00 27 76 63 82.89 1,200 12.00 1.00 28 81 3 3.70 1,200 0.00 0.00 29 92 30 32.61 1,400 1.40 0.10 30 97 7 7.22 2,300 4.60 0.20 100-199 9 31 102 5 4.90 3,200 0.00 0.00 32 111 12 10.81 1,600 8.00 0.50 33 113 23 20.35 1,600 0.00 0.00 34 120 8 6.67 4,200 126.00 3.00 35 139 10 7.19 4,600 0.00 0.00 36 150 10 6.67 2,000 0.00 0.00 37 168 8 4.76 3,600 108.00 3.00 38 183 6 3.28 2,500 0.00 0.00 39 195 20 10.26 4,100 20.50 0.50 200-299 4 40 201 32 15.92 7,600 114.00 1.50 41 220 5 2.27 5,570 5.57 0.10 42 230 16 6.96 4,600 138.00 3.00 43 240 35 14.58 5,900 64.90 1.10 300 < 7 44 302 20 6.62 7,000 7.00 0.10 45 450 30 6.67 9,400 141.00 1.50 46 670 13 1.94 12,700 152.40 1.20 47 706 100 14.16 1,900 57.00 3.00 48 1,500 300 20.00 69,700 2,091.00 3.00 49 2,910 200 6.87 97,000 1,455.00 1.50 50 11,734 2,300 19.60 630,000 25,200.00 4.00 Total 50 21,783 3,323 15.26 906,790 29,848.23 3.29 Note: Firm No. 27 was excluded due to doubtful R&D staff share. Source: National firm mail survey. 29 Notes 1. The machine industries include industrial, electri- Printing Bureau, Ministry of Finance, Tokyo. cal, and transportation machinery. 14. Sokeizai Sentah (Materials Process Technology 2. Minami, Ryoshin (1992), Nihon no Keizai Hatten Center) (August, 1991), Annual Statistics ofMaterial Process (The Economic Development ofJapan), Toyohkeizai, Tokyo, Industries oflapan 1990, Materials Process Technology Cen- p.91. ter, Tokyo, p. 45. 3. Ibid., p.75 15. MIT, Annual Reports, op. cit. 4. "Keiretsu" refers to the ties among companies in a 16. My estimate is based on the 1992 membership list business group, based upon either capital/board of di- of the Japan Cast Iron Foundry Association. rector membership relationship or long-term economic 17. MITI, Overseas Public Affairs Office (1992), Issues transaction. The concept embraces the horizontal keiretsu and Trends in Industrial/Scientific Technology, Overseas (Konzem or combination) mainly organized by a promi- Public Affairs Office, Tokyo, p.40. nent commercial bank, and the vertical Keiretsu, e.g., the 18. Mie Prefectural Metal Research Institute (1990), Toyota or Nissan Group including affiliates and subcon- "The History of Mie Prefectural Metal Research Institute" tractors. In this paper, keiretsu refers to the vertical unpublished documents, Kuwana. keiretsu. 19. Independent or non-keiretsu firms diversify their 5. According to the OECD classification. transactions with more consumer firms than do keiretsu 6. Chijiiwa, Kenji (1980), "Trends in the Japanese firms. Casting Industry (1950-1980)," Sohgoh Imono, Vol. 21, No. 20. A third category, semispecialized, firms produce 12. both for in-house consumption and outside sales; in the 7. Prime Minister's Office, Science and Technology firm sample, serispecialized firms were included in the Agency (1985), Annual Report on Technology Importation, specialized firm category. Printing Bureau, Ministry of Finance, Tokyo. 21. No answer or missing on the item (R&D share) 8. Explanation for each imported technology is based is heated as a zero R&D expenditure. on the unpublished document of each TSI, like Japan 22. Yoshikai, Masanori (1985), Nihon no Sangyoh Meehanite Metal Co., Ltd., Japan Shellmolding Associa- Gijyutsu Seisaku (Japanese Industrial Technology Policy), tion, etc. Toyohkeizai, Tokyo. 9. The Japan Cast Iron Foundry Association (1991), 23. Ibid., p.167. "The Short History of The Japan Cast Iron Foundry As- 24. Three major governmental finance corporations sociation" unpublished document, Tokyo. specialized in loans confined to small and medium firms: 10. Lynn, L.H. (1982), How Japan Innovates, Westview the Smaller Business Finance Corporation, the People's Press, Boulder. Finance Corporation, and the Bank for Commerce and 11. Goto, Akira (1993), Nihon no Gijyutsu Kakushin to Industrial Cooperatives. Sangyoh Soshiki (Innovations and Industrial Organizations 25. In the questionnaire, the question is confined to in Japan), University of Tokyo Press, Tokyo. technology-related policies. It might have been easier to 12. Research in this paper is confined to the cast iron answer had it included broader policy measures. subsector. 26. Kiyonari, Tadao (1990), Chushoh Kigyo Dokuhon 13. Factories employing fewer than 20 are excluded (Readings on Small and Medium Firms), Toyohkeizai, To- from official statistics. See MITI (1991), Annual Reports on kyo, p.133. Machinery Industry and Iron and Steel Industry Statistics, 30 References Chijiiwa, Kenji (1980), "Trends in the Japanese Casting and Statistic Department, Annual and Monthly Reports Industry (1950-1980)," Sohgoh Imono, Vol. 21, No. 12. on Machinery, Iron and Steel Industry Statistics, var- Goto, Akira (1993), Nihon no Gijyutsu Kakushin to Sangyoh ous years and months, Printing Bureau, Ministry of Soshiki (Innovation and Industrial Organization in Ja- Finance, Tokyo. pan), University of Tokyo Press, Tokyo. Ministry of International Trade and Industry, Overseas Imai, Ken'ichi, et al. (1982), Naibu Soshiki no Keizaigaku Public Affairs Office(1992), Issues and Trends in Indus- (Economics of Internal Organization), Toyohkeizai, trial and Scientific Technology: Toward Techno-Globalism, Tokyo. Overseas Public Affairs Office, Tokyo. Itami, Hiroyuki, et al. (eds.), (1993), Nihon no Kigyo Ministry of International Trade and Industry, Research Sisutemu (Readings on Japanese Enterprise System), Vol. and Statistics Department (1992), Census ofManufac- 3: Jinteki Shigen (Human Resources), Yuhikaku, To- turers 1989, Printing Bureau, Ministry of Finance, kyo. Tokyo. Itami, Hiroyuki, et al. (eds.), (1993), Nihon no Kigyo Ministry of International Trade and industry, Small and Sisutemu (Readings on Japanese Enterprise System), Vol. Medium Enterprise Agency. (1992), Chusou Kigyoh 4: Kigyoh to Shijyo( Enterprise and Market ), Shisaku no Aramashi (Outline of the Small and Medium Yuhikaku, Tokyo. Enterprise Policies of the Japanese Government), Japan Itoh, Motoshige, et al. (1988), Sangyoh Seisaku no Keizai Small Business Research Institute, Tokyo. Bunseki (Economic Analysis of Industrial Policy), Uni- Ministry of International Trade and Industry, Small and versity of Tokyo Press, Tokyo. Medium Enterprise Agency (1992), Gijyutsuryoku Japan Cast Iron Foundry Association (1991)," The Short Kohjo no Tebiki (A Guide to Improving the Technological History of The Japan Cast Iron Foundry Associa- Capacity of Small and Medium Firms), Research Insti- tion," unpublished document, Tokyo. tute of International Trade and Industry, Tokyo. Kawakami, Yoshiaki (1993), Gendai Nihon no Chushou Prime Minister's Office, Science and Technology Agency Kigyoh (Small and Medium Enterprises in Modern Ja- (1985), Annual Report on Technology Importation, Print- pan), Tax and Accounting Association, Tokyo. ing Bureau, Ministry of Finance, Tokyo. Kiyonari, Tadao (1990), Chushou Kigyoh Dokuhon (Read- Prime Minister's Office, Science and Technology Agency ings on Small and Medium Firms), Toyohkeizai, Tokyo. (1992), Kagaku Gijyutsu Hakusho (White Paper on Sci- Kiyonari, Tadao (1993), Chushou Kigyoh Runessansu (Small ence and Technology 1992), Printing Bureau, Ministry and Medium Enterprise Renaissance), Yuhikaku, Tokyo. of Finance, Tokyo. Komiya, Ryutaro, et al.(eds.) (1984), Nihon no Shimotani, Masahiro (1993), Nihon no Keiretsu to Kigyoh Sangyohseisaku (Japanese Industrial Policy), University Gurupu (Japanese Keiretsu and Corporate Groups), of Tokyo Press, Tokyo. Yuhikaku, Tokyo. Lynn, L. H.(1982),How Japan Innovates, Westview Press, Sokeizai Sentah (Materials Process Technology Center) Boulder. (1991), Annual Statistics on Materials Process Industries Mie Prefectural Research Institute (1990), " The History of Japan 1990, Materials Process Technology Center, of Mie Prefectural Metal Research Institute," unpub- Tokyo. lished document, Kuwana. Yoshikai, Masanori (1985), Nihon no Sangyoh Gijyutsu Minami, Ryoshin (1992), Nihon no Keizai Hatten (The Eco- Seisaku (Japan's Industrial Technology Policy), nomic Development of Japan), Toyohkeizai, Tokyo. Toyohkeizai, Tokyo. Ministry of International Trade and Industry, Research 31 t Distributors of COLOMBIA GERMANY ISRAEL NEPAL SWEDEN lnosenlace Ltda. UNO-Verlag Yozmt Literature Ltd. Everest Media International Services (P) Ltd. Livradia Portugal Wennergren-Williamus AB W orld Bank Carrera6 No.51-21 PoppelsdonferAllees5 PO. Boo 56055 GPO Bun 5443 Apartado 2681, Rue Do Carmo 70-74 P0. Box 1305 Apurtado Aereo 34270 53115 Bonn 3 Yohanan Hasundlar Street Kathmandu 1200 Lisbon S-171 26 Sona Publications SZntafi de Bogold, D.C. Tel: (49 226) 040020 Tel Avis 61560 Tel: (9771)472 152 Tel: (1) 347-4082 Tel: (468)705-97-O Prices and credit terms vary from Tel: (57 1) 285-2798 Fax: (40 228) 217492 Tel: (972 3) 5285-397 Fax: (977 1) 224 431 Fan: (1) 347-0264 Faa: (46 8) 27-00-71 country to country. Consult your Foo. (57 1) 265-2796 URL: hsp:/7wwwuno-ver1ag.de Fun: (9723) 5285-397 E local distributor before placing an E-mail: unovedogol.com COTE WI VOIRE R.O.Y Intemnatisnul De LindeboorrVinOr-Publikaties Compani De Librarii Bocuresti S.A. SWITZERLAND order. Cester d'Editiov et do Difusios Alricaines GREECE P0 Boo 13056 P0. Ben 202, 7480 AE Haoksberges SIr. Lipscani 00. 26, sectorS3 Librairie Paot Service Ivstilutionsel ARGENTINA (CEDA) Papasotiriou SA. Tel Aviv 61130 Tel: (3153(574-004 Bucharest C6tes-de-Monibenon30 Olicina del Libro Intemacional 04B.F541 35, Stoumnra SIr. Tel: (972 3) 5461423 Fus: (3153) 572-9200 Tel: (40 1(613 9645 1002 Lausanne Av. Cordoba 1877 Aidjan 04 1062 Athens Fan: (9723(5461442 B-mail: lindehoo§woddnnine.nl Fax: (40 1(312 4000 Tel: (4121) 341-3229 1120 Buenos Aires Tel (225) 24 6510:24 6511 Tel: (30 1( 364-1826 E-mail: royil@selvsiion.ne.ii URL: http://www.wsddonline.nl-indebo Faa: (41 21) 341-3235 Tel:RUSSIAN FEDERATION Tel(51)85-354Fo: (2525567Fa: (0 (36-854Palestinian Authority/Middle East NEWZEALAND lsdatelstvo eVes Mir> ADECO Van Diermn Edilionn"fechniques Fax: (54 1) 815-8156 CYPRUS HAITI loden Information Services EBSCO NZ Ltd. So, Lolpachniy Pereulok Ch. dl Lacuez 41 AUSTRALIA, FIJI, PAPUA NEW GUINEA Cestertfor Applied Research Cultore Diffusion FO.B. 10502 Jerualem Private Mail Bag 99914 Moscow 101831 CH1807 Blonay SOLOMON ISLANDS, VANUATU, AND Cyprus College 5, Rue Capois Tel, (972 2) 6271219 New Maret Tel: (7095) 917 87 49 Tel: (4121(943 2673 WESTERN SAMOA 6, Diogenes Street, Engomi C.P 257 Fan: (9722(6271634 Auckland Fax: (7095) 017 9259 Faa: (4121) 943 3605 D.A.I. Boo 200 Po-a-Prince Tel: (64 9) 24-8119 48 WhitehorseRvie Nicoia Tel: (50923 9260 ITALY Pu: (649524-067 SINGAPORE, TAIWAN, THAILAND 64 pithre odTel (357 2(44-1730 Fun: (509( 23 4858 Licosa Commissionaria Sansoni SPA MYANMAR, BRUNEI Centrl Books Distribution Mitchum 3132 Victoria F ( 2 0 Vie Dca Di Colabria, 1/1 NIGERIA Ashge Puhlishing Asi Paciic PIe. Ltd. 306 Silom Road Tel: F61) 8210 7777 HONG KONG, MACAO Casette Postule 552 University Press Limied 41 Kallg Padding Rsad #04-03 Bangkok 10500 Far (61) 3 9210 7700 CZECH REPUBLIC Asia 2000 Ltd. 50125 Fir8n8e Three Crown Building Jericho Golden Wee Building Tel: (662) 235-5400 E-aiL (ervicedadirecc7m88 Notional Information Center Salvo & Circulation Departrvent Tel: (55) 645-415 Private Mail Bag 0095 Singapore 349316 Fan: (66 2(237-8321 E-mail service@Odadirect.com.au URL: hftp://www .dadirect.com.au prodejna, Konviktska 5 Seabird House, unit 1101-02 Fun: (55) 641-257 lbadln Tel: (65) 741-5166 CS - 113 57 Praguel1 22-28 Wyndharn Street, Central E-mail: licosa@fthcc.0 Tel: (234 22) 41-1356 Fax: (65) 742-9356 TRINIDAD & TOBAGO, AUSTRIA Tel: (42 2) 2422-9433 Hvng Kong URL: hHpi/www .ftbcc.Mtlicosa Fax: (234 22) 41-2056 E-mail: ashgate@asianconnect.cov AND THE CARRIBBEAN Geroid and Co. Fax (422(2422-1484 Tel: (652) 2530-1409 Syoteratics Studies Unit Wehugas 6URL: hnp:l/www.nis.cz/ Fun: (852) 2526-1107 JAMAICA NORWAY SLOVENIA 9 Welts Street Welhburggasse 26NIInoASGsolrkVetiPulsigrup Cee A-1011 Wien E-mail: sales@asia200,co.hk Ian Randle Publishem Ltd. Tel: (43 1) 512-47-31-0 DENMARK URL: hntpJA/wwr.asIa2000.com.hk 266 Old Hope Rood, Kingston 6 Book Deparvment, Posthoks 6512 Etterstad Dunajska ceas5 Trinided, West Indies Fa: (43 1) 512-47-31-29 SafundsLiteratur Tel:876-927-2085 N-0606 Oslo 1000 Lubljana Tel: (809) 662-5654 UFL: tp://www.gerold.co/at.online Rosenoems AII6 11 HUNGARY Fan: 876-077-0243 Tel: (47 22) 97-4560 Tel: (386 6113383 47; 132 1230 Fa: (809) 662-5654 DK-1l970oFrederiksberg C Euro Info Service E-mail: imil@cvlin.com Pus: (47 22( 97-4545 Fax: (386 61( 133 80 30 E-mail: tobe@trinidad.net URL: http://:wwepaerold@gveatninlii BANGLADESH Tel. (45 31) 351942 Margiszgeti Bumps Hue PAKISTA UGANDA Micro Industries Development Pus: (45 31) 357822 H-1138 Budapest JAPAN Assistance Society (MIDAS)hp://ww.sl.cbs.dk Tel: (361111061 Eastern Bok Service Mir Book Agency SOUTH AFRICA, BOTSWANA Gut Ld. Hossite 5 o et 16 IAS Fan: (36 1( 302 5035 3-13 Hongo 3-chomne, Bunrkyo-ku 65, Shahrah-e-Quaid-e-Azam Forosingle titleo: PO Boo 9997, Madhvani Building HouseECUADOR E-mail: euroinf@mi.mata.h Tokyo 113 Lahore 54000 O16ord Uniserity Press Southern Africa Ptot 16/4 Jinja Rd. Dhanmondi R/Area Libri Mundi Tel: (813)3818-0861 Tel: (92 42) 735 3601 Vasco Boulevard, Goudwood Kampala Ohaka 1209 Tel:a 128092262 Libreria Ivternacional INDIA Faa: (81 3) 3818-0864 Pus: (92 42) 576 3714 P0. Boa 12110, Ni City 7463 Tel: (256 41) 251467 Tel: (880 2)(01118 PO. Bus 17-01-3020 Allied Publishers Ltd E-mail: ordvrs@svt-ebs.co.p Cope Town Fa: (25641)251468 Fx(8028118Juan Leon Mero 851 751 Mount Road URL: ht8pJ/www.tbekkome.or.jp/-svt-ebs Oxford University Press Tel: (27 21) 595 4400 E-mail: gus@swrnougando.com BELGIUM Gox: Madras -800002 5 Bangatore Town Fax: (27 21) 595 4430 Jean De Lannoy Tel: (593 2) 521-606: (5932(544-I85 Tel (9144)852-3938 KENYA Share Fuisal E-mail: oxford@oup.cs.ea UNITED KINGDOM Av du Roi 202 Fax: (5932(504-200 Foe: (0144) 852-0640 Atrica Book Service (EA.) Ltd. PO Boo 13033 Microinfo Ltd. 16BrsesE-mail librimul@librimundi.com.ec Quaran House, Mfangano Street Karach[-75350 Forosubscription orders: P.O. Boo 3, Afton, Hampshire GU34 2PG Tel: (3 a)se56 E-mail librimu2@librimundi.com.c INDONESIA PO. Boo 45245 Tel: (92 21446307 Intemtional Subscription Service England Fax: (32 2) 538-0841 Pt. Indira Limed Nairobi Fa: (92 21) 4547640 P. Boo 41000 Tel: (44 1420) 86848 Fa:(2) 3-81EGYPT, ARAB REPUBLIC OF Jalan Borobudlur20 Tel (254 2( 223 641 E-mail: ouppak@TheOffice.not Craighall Fan: (44 1420) 89889 BRAZIL Al Ahrwn Distribution Agency PO. Boo 181 Fax: (204 2) 330 272 Johannesburg 2024 B-mull: wbank@ukminfo.demon.co.uk Publicacoes Tecnicas Intemacionais Ltda. l G Street Jakarta 10320 Pak Book Corporation Tel. (27 11) 680-1448 URL: hop://www.microinocouk Rua Peixoto Gomide, 209 Cairo TeL (6221(390-4200 KOREA, REPUBLIC OF Aziz Chambers 21, Queen' Road Fax: (27 11(880-6248 010 a ol,~Tel: (20 2) 578-6083 Fox: (62 21) 390-4289 Duejon Trading Ca. Ltd. Lahore E-mail: iss@is.co.za VENEZUELA 01409 5511(259-664 Foe: (202) 578-6833 PO. Boo 34, Youida, 706 Seoun Bldg Tel: (92 42) 636 3222 636 0885 Tecni-Ciecia Libros, SA. Tel: (55 11) 259-6644 IRAN 44-6 Youido-Dong, Yeongchengpo-Ku Faa: (92 42) 636 2328 SPAIN Centro Cuidud Comercial Tamanco B-mual: ps11 tr@pi-69 The Middle East Observer Ketab Sara Co. Publishers Senul E-mail: pbc@brpinonet.pk Mundi-Prensa Libros, SA. Emi:psmse@t.o.r41, Shorif Street Khaled Eslamboli Ave., 6th Street Tel: (62 2) 785-1631/4 Castello 37 Tel: (58 2) 959 5547 5035: 0016 URL: http.//www uol.br Cairo Doloo Alley No.8 Faa. (822)784-0315 PERU 28001 Madrid Faa: (582)9595636 CANADA Tot: (202) 393-9732 PO. Boo 15745-733 Editoral Desarmtlo SA Tel: (34 1)431-3399 Renout Publishing Co. Ltd. Pay: (20 2(393-9732 Tehran 15117 MALAYSIA Aparto 3824, Limo 1 Faa: (34 1) 575-3998 5368 Canotek Road Tel: (98 21) 8717819: 8716104 University of Malaya Cooperative Tel: (51 14) 285380 E-mail: lihreria@mundiprensa.es Univeesity Bookshop, University of Zambia Ottawa, Ontario K1J 9J3 FINLAND Fax: (98 21) 8712479 Bookshop, Limited Fax: (5114) 286626 URL: hBp://www.mundiprensa.esf Greot East Read Campus Tl(6374-65Akateeminen Kirjakauppa E-mail: ketab-sara@neda.net.ir PO. Boo 1127 P. Boo 32379 Tel: (613) 745-2665 Fax (613) 745-7660 PO.B80128 Jalan Pantai Bu PHILIPPINES Muodi-Prensa Barcelona Lusaka E-mail: order dept@renoubooks.com FIN-00101 Helsinki Koefab Publishers 59700 Koala Lumpur Internationl Bookource Center Inc. Consell dv Cent, 391 Tel: (260 1(252576 U RL: http / www .renoufbooks.com Tel (3580(1214418 P0. Boo 19575-511 Tel: (603)756-5000 1127-A Antipolo St, Barangay, Venezuela 08009 Barcelona Fun: (200 1) 253 952 Pay: (358 0) 121-4435 Tehran Pus: (60 3(755-4424 Makati Cit Tel: (34 3) 488-3492 CHINA E-mail: alkatilaus @stockmann.fi Tel (99 21) 259-3723 Tel: (63 2)0896 6581: 6005:6507 Fan: (34 3(487-7659 ZIMBABWE China Financial 6 Econumic URL: hknp://wew .akateeminen.corrV Fao: (98 21) 259-3723 MEXICO Fax: (63 2) 896 1741 E-mail: horceona@mundiprensa.es Longman Zimbabwe (Pvt.(Ltd. Publishing HueTode Road, Arennie Suish nng o FRANCE IRELAND A. San Femuando No. 37 POLAND SRI LANKA, THE MALDIVES P. Boo ST1 25 8, DoaFoSiDnJe World Bank Publications Government Supplies Agency Cal. Torilao Guerra Internatinl Publishing Service Lake House Bookshop Southerton Beijing Te:(8j 1)i30-85 66, avenue d'16na Odhg an tSolAthair 14050 Mexico, D.F Ul. Piekna 31/37 100, Sir Chiftampalann Gardiner Mawatha Harare Tel: (86 10) 6333-8257 75116 Pads 4-5 Harcour Rood Tel (525)624-2800 00-677Warzawa Colombo2 Tel: (263 4(6216617 Fa:(61)60-35Tel: (33 1) 40-69-30-56/57 Dublin 2 Pay: (52 5) 624-2922 Tvl: (48 2) 620-689 Tel: (94 1) 32105 Faa: (263 4) 621670 Fax: (33 1( 40-69-30-68 Tel: (353 1) 661-3111 E-mail: infotec@6n.netrmix Fax: (48 2) 621-7255 Fmn: (94 1) 432104 Faa: (353 1) 475-2670 UL: h85p:-/n.net.m E-mail: bo7k9%ips@ikp.at.com.pl E-mail: LHL@sr.lanka.net 0107 URL: http:/ax 7.ips1g.w)w.p2ip8/e-o2r7  RECENT WORLD BANK TECHNICAL PAPERS (continued) No. 345 Industry and Mining Division, Industry and Energy Department, A Mining Strategy for Latin America and the Caribbean No. 346 Psacharopoulos and Nguyen, The Role of Government and the Private Sector in Fighting Poverty No. 347 Stock and de Veen, Expanding Labor-based Methods for Road Works in Africa No. 348 Goldstein, Preker, Adeyi, and Chellaraj, Trends in Health Status, Services, and Finance: The Transition in Central and Eastern Europe, Volume II, Statistical Annex No. 349 Cummings, Dinar, and Olson, New Evaluation Procedures for a New Generation of Water-Related Projects No. 350 Buscaglia and Dakolias, Judicial Reform in Latin American Courts: The Experience in Argentina and Ecuador No. 351 Psacharopoulos, Morley, Fiszbein, Lee, and Wood, Poverty and Income Distribution in Latin America: The Story of the 1980s No. 352 Allison and Ringold, Labor Markets in Transition in Central and Eastern Europe, 1989-1995 No. 353 Ingco, Mitchell, and McCalla, Global Food Supply Prospects, A Background Paper Prepared for the World Food Summit, Rome, November 1996 No. 354 Subramanian, Jagannathan, and Meinzen-Dick, User Organizations for Sustainable Water Services No. 355 Lambert, Srivastava, and Vietmeyer, Medicinal Plants: Rescuing a Global Heritage No. 356 Aryeetey, Hettige, Nissanke, and Steel, Financial Market Fragmentation and Reforms in Sub-Saharan Africa No. 357 Adamolekun, de Lusignan, and Atomate, editors, Civil Service Reform in Francophone Africa: Proceedings of a Workshop Abidjan, January 23-26, 1996 No. 358 Ayres, Busia, Dinar, Hirji, Lintner, McCalla, and Robelus, Integrated Lake and Reservoir Management: World Bank Approach and Experience No. 360 Salman, The Legal Framework for Water Users' Associations: A Comparative Study No. 361 Laporte and Ringold. Trends in Education Access and Financing during the Transition in Central and Eastern Europe. No. 362 Foley, Floor, Madon, Lawali, Montagne, and Tounao, The Niger Household Energy Project: Promoting Rural Fuelwood Markets and Village Management of Natural Woodlands No. 364 Josling, Agricultural Trade Policies in the Andean Group: Issues and Options No. 365 Pratt, Le Gall, and de Haan, Investing in Pastoralism: Sustainable Natural Resource Use in Arid Africa and the Middle East No. 366 Carvalho and White, Combining the Quantitative and Qualitative Approaches to Poverty Measurement and Analysis: The Practice and the Potential No. 367 Colletta and Reinhold, Review of Early Childhood Policy and Programs in Sub-Saharan Africa No. 368 Pohl, Anderson, Claessens, and Djankov, Privatization and Restructuring in Central and Eastern Europe: Evidence and Policy Options No. 369 Costa-Pierce, From Farmers to Fishers: Developing Reservoir Aquaculture for People Displaced by Dams No. 370 Dejene, Shishira, Yanda, and Johnsen, Land Degradation in Tanzania: Perception from the Village No. 371 Essama-Nssah, Analyse d'une ripartition du niveau de vie No. 373 Onursal and Gautam, Vehicular Air Pollution: Experiences from Seven Latin American Urban Centers No. 374 Jones, Sector Investment Programs in Africa: Issues and Experiences No. 375 Francis, Milimo, Njobvo, and Tembo, Listening to Farmers: Participatory Assessment of Policy Reform in Zambia's Agriculture Sector No. 376 Tsunokawa and Hoban, Roads and the Environment: A Handbook No. 377 Walsh and Shah, Clean Fuels for Asia: Technical Options for Moving toward Unleaded Gasoline and Low-Sulfur Diesel No. 382 Barker, Tenenbaum, and Woolf, Governance and Regulation of Power Pools and System Operators: An International Comparison No. 383 Goldman, Ergas, Ralph, and Felker, Technology Institutions and Policies: Their Role in Developing Technological Capability in Industry No. 385 Rowat, Lubrano, and Porrata, Competition Policy and MERCOS UR No. 386 Dinar and Subramanian, Water Pricing Experiences: An International Perspective No. 387 Oskarsson, Berglund, Seling, Snellman, Stenback, and Fritz, A Planner's Guide for Selecting Clean-Coal Technologies for Power Plants No. 388 Sanjayan, Shen, and Jansen, Experiences with Integrated-Conservation Development Projects in Asia No. 389 International Commission on Irrigation and Drainage (ICID), Planning the Management, Operation, and Maintenance of Irrigation and Drainage Systems: A Guide for the Preparation of Strategies and Manuals 4aNQ 9"-v THE WORLD BANK I l re 1t, \\. D V\k :1 2 h 123 \\ > I)l 1)1\ k I -ni.11 l~ <-s 1 "I 14054 ISBN 0-8213-4054-9