Global Taiwan : Building Competitive Strengths in a New International Economy 9781315704050, 9780765616166

Citation preview

GLOBAL TAIWAN

Publication of this book was generously supported by a grant from the Chiang Ching-Kuo Foundation for International Scholarly Exchange. The opinions, findings, and conclusions or recommendations expressed in this book are those of the authors and do not necessarily reflect the views of the Foundation.

GLOBAL

TAIWAN BU I LD I NG COMPETITI VE STRENGTHS IN A NEW INTERNATIONAL ECONOMY

SUZANNEBERGER AN D RICHARD K. LESTER, EDITORS

AN EAST G ATE BOOK ROUTLEDGE

Routledge Taylor & Francis Group

LONDON AND NEW YORK

} An East Gate Book First published 2005 by M.E. Sharpe Published 2015 by Routledge 2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN 711 Third Avenue, New York, NY 10017, USA Routledge is an imprint of the Taylor & Francis Group, an informa business Copyright © 2005 Taylor & Francis. All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Notices No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use of operation of any methods, products, instructions or ideas contained in the material herein. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation without intent to infringe. Library of Congress Cataloging-in-Publication Data

Global Taiwan : building competitive strengths in a new international economy / edited by Suzanne Berger and Richard K. Lester. p. cm. “An east gate book.” Includes bibliographical references and index. ISBN 0-7656-1616-5 (hardcover : alk. paper); ISBN 0-7656-1617-3 (pbk.: alk. paper) 1. Taiwan—Economic conditions—1975– 2. Industrial policy—Taiwan. 3. Industries— Technological innovations—Economic aspects—Taiwan. 4. Globalization—Economic aspects—Taiwan. 5. Competition, International. 6. Taiwan—Foreign economic relations. I. Berger, Suzanne. II. Lester, Richard K. (Richard Keith), 1954– HC430.5.G56 2005 337.5124’9—dc22

2004017166 ISBN 13: 9780765616173 (pbk) ISBN 13: 9780765616166 (hbk)

Contents List of Tables and Figures Preface

1. Globalization and the Future of the Taiwan Miracle Suzanne Berger and Richard K. Lester 2. Industry Co-Evolution: A Comparison of Taiwan and North American Electronics Contract Manufacturers Timothy J. Sturgeon and Ji-Ren Lee

vii xi

3

33

3. Leading, Following, or Cooked Goose? Explaining Innovation Successes and Failures in Taiwan’s Electronics Industry Douglas B. Fuller, Akintunde I. Akinwande, and Charles G. Sodini 76 4. A Tale of Two Sectors: Diverging Paths in Taiwan’s Automotive Industry Edward Cunningham, Teresa Lynch, and Eric Thun

97

5. Moving Along the Electronics Value Chain: Taiwan in the Global Economy Douglas B. Fuller

137

6. From NAFTA to China? Production Shifts and Their Implications for Taiwanese Firms Marcos Ancelovici and Sara Jane McCaffrey

166

7. Innovation and the Limits of State Power: Integrated Circuit Design and Software in Taiwan Dan Breznitz

194 v

vi

8. Cross-Straits Integration and Industrial Catch-Up: How Vulnerable Is the Taiwan Miracle to an Ascendant Mainland? Edward S. Steinfeld

228

Notes Bibliography The Authors Index

281 297 321 323

List of Tables and Figures Tables 1 2 3 4 2.1 2.2 2.3

2.4 2.5 3.1 3.2 3.3 4.1 4.2 4.3 5.1 5.2 6.1 6.2 6.3 6.4

IPC Interviews: Taiwanese-Owned Firms by Sector IPC Interviews: Firms in All Countries by Sector IPC Interviews: Country of Ownership IPC Interviews: Country of Plant IT Hardware Produced by Taiwanese Firms, 2002 Top Five Buyers of IT Hardware from Taiwan, 1998–2003 Comparison of Top Five to Top 100 EMS Firms; Revenues, Employment, Facilities, and Location, 1999 and 2002; Compound Annual Growth Rate (CAGR) 1999–2002; and Top Five Share of Top 100, 2002 Taiwanese ODM Firms Overseas and Chinese Production Volume Ratio by Major Product Category, 2001 and 2002 Solectron’s Geographic Net Sales, Nine Months Ended May 31, 2003, and 2004 Foundry Market Share of the Three Largest Pure-Play Foundries DRAM Technology Sources Technology Sources for Taiwan’s AMLCD Firms Exports of Taiwanese Assemblers to Their Partners’ Plants East Asian Countries’ Contributions to Japan’s Auto Parts Imports Automotive Parts, Import and Export Growth (and Composition), 2001–3 The Concentration of Taiwan’s IT Industry in 2000 Taiwan’s MNC Research Centers Taiwanese FDI in Mexico, 1995–2002 FDI in the Maquiladora Sector Textile and Apparel Industry, 1999–2002 Mexican Electronics Industry, 1994–2001 Top PC Vendors and ECMs in 2002

xii xiii xx xx 47 51

57 66 67 84 88 90 107 109 109 151 162 172 172 173 175 vii

viii

LIST OF TABLES AND FIGURES

6.5 Largest Dollar Amounts of U.S. High-Tech Imports by Country of Origin, 2000–2002 6.6 Electronics Firms Moving Out of Mexico 6.7 Textile Firms Leaving Mexico 6.8 Shipping Trends at One Large U.S. Apparel Firm 7.1 Top Ten Taiwanese Design Houses by Ranked Sales 7.2 Software Sales and Exports of Indigenous Firms in Ireland, India, and Israel, 1991–2000 8.1 Input Sourcing by Taiwan-Invested Enterprises in China, 1995–1998 8.2 Firm Size of Higher-Technology Enterprises in China, 1995–2000 8.3 Main Inhibitors of Export Growth 8.4 Sources of Enterprise Financing

176 178 182 184 201 214 241 245 248 252

Figures 2.1 Comparison of Dyadic Supplier Archetypes with the Turnkey Supplier: General vs. Specific Assets 2.2 Co-Evolution with Competing vs. Shared Supply Bases 2.3 Revenues at the Top Twenty EMS Firms, 1999 and 2002 2.4 Product Mix for the Largest Five EMS Firms, 2001 2.5 Comparison of Typical ODM and EMS Firms: Value Chain Scope, Product/Customer Scope, and Geographic Scope 2.6 Variation in the Thickness of the Interfirm Link: General vs. Specific Assets in the Context of Co-Design and Build-to-Print 2.7 Revenue Comparison of Largest Five ODM and EMS Firms: 1993–2003 2.8 Return on Sales Comparison of Largest Five ODM and EMS Electronics Contract Manufacturers: 1993–2003 4.1 Taiwan Auto Vehicle Production, 1989–2002 4.2 Total U.S. Imports of Taiwanese Auto Parts, 1989–2002 5.1 Foundry and IDM Production Share in Taiwan’s IC Industry 5.2 Stan Shih’s Smile Curve 6.1 U.S. Imports of Computers and Computer Parts, Mexico vs. China 6.2 U.S. Imports of Televisions and Radios, Mexico vs. China 6.3 Growth Rates of Apparel and Household Good Imports to the United States from Mexico and Selected Caribbean Basin Initiative Countries

42 45 56 59 62

64 69 71 103 115 143 150 179 179

181

LIST OF TABLES AND FIGURES

6.4 Apparel and Household Textile Goods Imports to the United States, Mexico vs. China 6.5 Annual Growth Rate in U.S. Imports of Apparel and Household Goods, China vs. Mexico 6.6 U.S. Imports from Mexico: Apparel, Electronics, and Autos 7.1 Market Relations and Communication Environment of the Taiwanese IC Design Houses 8.1 Economic Growth in Taiwan 8.2 Economic Growth in Mainland China 8.3 Taiwan’s FDI into Mainland China, 2002 8.4 Total Foreign Capital Flows into China, 1995–2002 8.5 Comparison of Estimated Realized Taiwanese FDI into Mainland China, 1991–2002 8.6a FDI into PRC, 1993 8.6b FDI into PRC, 2001 8.7 Taiwan FDI into PRC by Industry, 1992–2002 8.8 PRC FDI by Province, 1985–2001 8.9 Taiwan FDI into PRC by Province, 1992–2002

ix

181 182 191 204 230 231 233 233 234 236 237 238 239 240

This page intentionally left blank

Preface Global Taiwan is the first report from the field of a four-year-long study at the MIT Industrial Performance Center (IPC) on the impact of globalization on industry in advanced and developing economies. In 1999, a team of researchers from social science, engineering, and management departments at MIT set out to explore, at ground level, the ways in which companies in the United States, East Asia, and Europe are responding to changes in the international economy as barriers come down to the movement of capital, goods, and services across national frontiers. With stories about the outsourcing of jobs from high-wage to low-wage economies appearing on the front pages of newspapers on a daily basis, globalization has become a burning issue in all advanced societies. Everywhere there are high levels of public concern about the consequences for employment, innovation, and growth. The basic objectives of our research were to contribute to public debates about globalization and societal well-being by analyzing the underlying forces driving the transformation of advanced economies from the end of the 1980s to the present and the responses to them by companies competing in world markets. From the outset, we knew that research in Taiwan would be a critical part of our investigation. The dilemmas facing this society exemplify the opportunities and dangers of the new international economic order for both advanced and emerging economies. Taiwan represents an extraordinary story of economic development and democratization. Taiwan in the 1950s ranked among the world’s poor and backward agricultural societies. Today it is a powerhouse of high-technology industry and services, and its industrialists control vast webs of economic activities outside their own borders. Since the opening of China twenty-five years ago, Taiwan’s historic and cultural proximity to the Mainland has provided entrepreneurs with unparalleled opportunities for economic expansion, far beyond the possibilities that a market of twenty-three million Taiwanese might have offered. Advances in digitization over the past decade have made it possible to reorganize activities once carried out inside vertically integrated companies into global value chains, enabling Taiwanese companies with xi

xii

PREFACE

Table 1 IPC Interviews: Taiwanese-Owned Firms by Sector Sector Electronics Software Auto/Auto parts Textile/Apparel Other/government, academia Other/industry Total

Number 85 49 34 28 65 25 286

% 30 17 12 10 23 9 100

highly specialized capabilities to enter information technology (IT) businesses and to capture powerful market positions. Over the same period of spectacular growth, however, the expansion of the Taiwanese economy has come to be more and more dependent on the transfer of activities to Mainland China and on the establishment of new businesses there, rather than within Taiwan. The relocation of Taiwanese companies to Mainland China thus threatens to hollow out the domestic economy with a shift, first, of low-skill and low-wage manufacturing labor, then of more skilled jobs and technical design work, and eventually, of research, development, and product definition along with well-paid jobs. In the 286 interviews we carried out in Taiwanese companies over the period 1999–2003, we were able to track this process and to follow the destruction and creation of jobs within the home society in parallel with the establishment of manufacturing facilities, sales offices, and R&D centers abroad (see Table 1). Today Taiwan faces the challenge of maintaining and enhancing employment opportunities and capabilities for innovation and growth on its own territory even as its companies relocate many of their operations to China. At the same time, Taiwanese firms are now confronting tough new competition in their core businesses both from powerful, innovative companies like the Western global contract manufacturers and, on the other side, from rapidly advancing Chinese rivals. These challenges are familiar to policymakers, industrialists, and scholars throughout the industrialized world. Analyzing Industrial Performance and Globalization Our research on globalization builds on a longstanding collaboration between engineers and social scientists at MIT to study the sources of productivity and competitiveness in contemporary advanced countries.

PREFACE

xiii

Table 2 IPC Interviews: Firms in All Countries by Sector Sector Electronics Textile/Apparel Auto/Auto parts Software Other/government, academia Other/industry Total

Number 194 159 85 54 69 68 629

% 31 25 14 9 11 11 100

In Made in America (1989), an interdisciplinary group of MIT researchers studied the American industrial economy in a period during which productivity growth and innovation stagnated, and new challengers like Germany and Japan seemed poised to overtake the United States across a wide set of sectors. In Made by Hong Kong (1997), an Industrial Performance Center team investigated the situation of Hong Kong industries pulled between, on the one hand, the opportunities for reproducing on a massive scale in China a model of low-cost, low-tech, high-quality production that had been pioneered in Hong Kong and, on the other hand, the possibilities of investing more at home in research, development, and education and in building Hong Kong’s capabilities for hightech products, brands, and manufacturing-linked services that could command higher margins than the old businesses. Today, in our new research on globalization, we have used approaches and methodologies already developed in our work on the United States and Hong Kong. We started, first, from macro-level theories about the effects of globalization on convergence and diversity in national economies. Next, we moved to the firm level and carried out 629 long, semistructured interviews with managers in firms competing in similar product markets. Finally, we developed a set of explanations from the ground level up to account for the strategic choices of firms operating in a global economy (see Table 2). Virtually everyone agrees on the changes that have driven the transformation of the international economy over the past twenty years. New information and transportation technologies have pushed down the costs of moving information, goods, and services across distances, and producers everywhere now find themselves in global competition. International trade agreements have lowered the barriers to cross-border movement of goods, services, and capital. Liberalization of financial markets in all advanced industrial countries has swelled the flows of

xiv

PREFACE

portfolio lending, trading in foreign exchange and in new financial instruments, and long-term foreign direct investment. The end of the Soviet bloc in 1989 and the opening of China a decade earlier created vast new spaces for production and great potential for new consumer markets. Firms from countries that were on the margins of the advanced industrial world of the 1980s, such as South Korea, have today become major players in the international economy. The impact of new information technology on globalization has been particularly far-reaching. Advances in information-processing power and speed and the standardization of communication protocols allow firms to exchange complex and voluminous data seamlessly so that products that once had to be made in-house in order to achieve conformance and high quality can now be manufactured by independent suppliers and subcontractors. The digitization of the interfaces between stages in the sequence of operations along the way from design through manufacturing allows firms to focus on functions that they do especially well and that create high added value for them. A lead company can purchase from suppliers around the world functions like manufacturing that it no longer sees as its own competitive strength. Levels of performance that once required organizing functions within vertically integrated enterprises can now be achieved through the coordination of supply chains with multiple autonomous firms located at the nodes between defining and designing a product and the final customer. The pressures of the international economy mean increases in competition as well as increases in the fluctuations of capital markets and demand. This volatility heightens both the opportunities for gain in world markets and the risks involved in large-scale and long-term investments. Financial market crises, like those that took place in rapid succession in Mexico (1994), Asia (1997), Russia (1998), and Argentina (2002), have a multitude of domestic as well as international sources, but the openness and deregulation of financial markets and the rapidity with which funds can be transferred in and out of societies are key contributing factors. For firms, this translates into greater risks and higher costs of capital, and hence into greater reluctance to sink long-term and heavy investment in brick-and-mortar facilities in cyclical industries prone to oversupply, like semiconductors, where new plants cost close to US$3 billion. In such an environment, using outsourcing or contract manufacturing allows lead firms to reduce capital requirements and risk. This has created an opening for a vast new business for contract manufacturers and

PREFACE

xv

global suppliers who specialize in high-quality production and thrive on managing risks by spreading the costs of their investments across a wide customer base. The result has been the transformation of some lead firms into organizations focused on a smaller set of core competencies and, simultaneously, the mushrooming of a new population of strong and independent subcontractors: contract manufacturers like Solectron and Flextronics; “original design manufacturers” (ODMs) like the Taiwanese electronics firms Hon Hai and Quanta; apparel contractors like Kellwood, a U.S. firm, or Li and Fung, a Hong Kong trading company; and auto-partsmakers like Delphi, Magna, and Visteon. In a world in which it is possible to buy high-quality manufacturing in the market from global suppliers or original equipment manufacturer (OEM) or ODM companies and it is no longer necessary to possess such capabilities in-house, new companies can enter the world market far more easily, since they need not spring forth fully capable of undertaking all the operations required to transform their innovations into goods and services in the hands of consumers. The changes in technology, markets, and the international economy have thus led to great changes in the organization and location of production over the past fifteen years. When two MIT studies, Made in America and The Machine that Changed the World, looked at the best industrial practices at the end of the 1980s, certain common features became apparent among companies around the world. Many of these features were embodied most clearly by a group of leading Japanese manufacturing companies. The lessons learned were that close interaction between a firm’s core activities and those of lead customers and suppliers was essential to efficient, highquality manufacturing, just-in-time delivery, rapid learning, and the commercialization of innovation. The most successful firms were those, big or small, that integrated their core functions, either by locating in physical proximity, or by tight coordination under common direction. Bringing research, development, design, manufacturing, and marketing close enough together so that people involved in these activities could interact with each other daily—on cross-functional teams or otherwise—appeared to make a big difference for developing new things, designing them for quality and lean manufacturing, and getting them to market quickly. In such an environment, vertically integrated companies had enormous advantages, since they were able to extend their range of control over all the steps between innovation and the final customer, and because owning the

xvi

PREFACE

steps was usually the only way to ensure that there was a tight enough degree of integration between them for quality, efficiency, and just-intime delivery. Fifteen years later, the world of production seems to have turned on its head. We see a fragmentation and splitting off of large parts of the production system, more marked in some sectors and in some countries than others, but important enough overall to have had major effects on employment and the location of production. This fragmentation creates a new division of labor between brand or lead firms; subcontractors specializing in functions like manufacturing, logistics, testing, or design, which were once encapsulated within the lead firms and have now been spun out; and assemblers, distributors, and retailers. An economy that once was largely organized within the brick-and-mortar walls of vertically integrated companies today looks more and more like a network of supply chains linking firms that each perform a few core functions. Activities that in most of the firms we visited in the mid-1980s were being carried out in close geographic proximity are today relocated around the world by companies engaged in a continuous search for new production sites with cheaper costs, for added workers with scarce skills, and for access to new markets. Some of the world’s most successful firms today, like Dell, have radically separated product definition, design, and marketing from manufacturing. Quanta, a Taiwanese ODM notebook manufacturer, succeeds with no links to final customers, no brand, and little involvement with product definition. Alongside this world of fragmented enterprises, there remain other modes of organizing production. Some firms, like Sony, Toshiba, NEC, Matsushita, and Fujitsu in Japan, and Siemens, Philips, and ST Microelectronics in Europe, remain vertically integrated giants that retain a high proportion of the activities needed to make their products in-house. Some firms heavily concentrate their core activities in close geographic proximity to others in the same industry. Key parts of the innovative activities of advanced industrial countries cluster in local geographic areas. In the United States, examples include Silicon Valley, the life sciences cluster around MIT and Harvard in the Boston area, and some 200 or so other localities where economic life centers around various stages of innovation and production of particular products (Porter 2001). Outside of the United States, we also find strong clustered economies in, for example, the industrial districts of northern Italy and the science parks of Taiwan.

PREFACE

xvii

But how can we know if the different organizational forms in today’s snapshot of the system of production are all equally viable? Have we captured in our static picture some firms as they take off and others as they enter into decline? Can the diversity we see today be resilient over the long term? How well do these various kinds of economic organization fare under globalization? Do different industries have distinct patterns, for example, clustering in biotech and software; modularity and fragmentation in electronics? Do all firms in the same sector making the same products have to converge on the same set of best practices because of global competition? Or do the basic variations have to do with the home countries in which the firms originate? Would a French or a Japanese firm see a different set of possibilities in responding to new international challenges than a U.S. firm might identify? Here is where our puzzle begins. Two broad ways of thinking about these questions have thus far emerged in social-science literatures. Convergence theories and national models theories represent strong opposing perspectives on the consequences of globalization. They start from very different assumptions about how societies are organized, and they come to very different predictions about the likely impact of globalization on the trajectory of advanced industrial countries. Convergence theory (made popular in books like Kenichi Ohmae’s The Borderless World, 1990) has at the core the idea that when national controls over borders disappear and capital, goods, and services move freely, the central institutions of economic life in all countries will eventually converge on common structures and practices. As firms making the same products compete across world markets, they will be forced onto the same technological and organizational trajectories. Systems of production and distribution will become increasingly similar around the globe. In contrast, the “national models” approach to globalization starts from the idea that there are diverse types of capitalism in the world, each with institutions that link firms to society and to government in ways that produce enduring differences in economic performance. As firms from different countries enter into international competition, they try to capitalize on their special strengths and to acquire others through purchase or outsourcing (Hall and Soskice 2001). Globalization in this perspective is predicted to lead to competition that preserves and even reinforces national specificities. For all their differences, however, both the convergence model and the national varieties of capitalism model

xviii

PREFACE

focus mainly on explaining how societies in the aggregate, at the macro level, function and change. These theories have been our reference points in designing our own study of globalization. But as we have tried to interpret the data from hundreds of interviews in companies in North America, East Asia, and Europe, we have come to see the limitations of approaches that focus on macro-level institutions and processes of society and seek to deduce from them a set of behaviors that individual actors are likely to follow. The issue is not only a difference in explanatory power between deductive analytic approaches to understanding society on the one hand and empirical analysis that starts from the relationships discovered in data collected in interviews, observations, and from official and company statistics, on the other. The main difficulty is that approaches that start from society as a whole and move by deduction to conclusions about how individual parts of the system will react under common pressures tend to find great sameness in the responses of actors of a broadly similar kind. This is hardly surprising, because in such theories the actors are only thinly described: for example, firms as rational maximizers of profit. If we start with the most minimal and elegant assumptions about firm behavior, then most of the action in the globalization story is likely to take place in the external forces pressing down on the firm, or in the changes in the societies in which the firms are embedded, as the firms respond to changes in the institutional resources and incentives of their environment. In contrast, if we start from firms understood as actors with legacies built up out of previous experiences and strongly shaped by the particular societies in which they were born, if we conceive these legacies as resources or lenses—resources for developing new strategies and implementations, lenses for identifying familiar and new aspects of problems and seeing novel options—then we are likely to discover a far greater diversity in the behaviors of firms than any that we might have deduced from their contexts. In our comparative studies of companies in Taiwan and in other industrial countries, we start from a micro-level analysis and with the evolution of firms operating in the same industrial sectors and producing similar goods and services. We track over time their choices about how to structure themselves: about what to produce, with whom to produce it, and where to produce it. We call this third kind of approach “open pathways” to suggest the different ways in which firms combine the

PREFACE

xix

legacies of their origins and past experiences with new resources that they are now able to acquire in the open world economy. What we have discovered is the breadth of choices available to companies as they structure their resources and reach out to acquire new capabilities in the global economy. In contrast to the determinism of the convergence hypothesis and the stability of the varieties of national capitalism approach, we see a range of possible solutions to the problem of doing well under the constraint of great new pressures to adapt rapidly to international markets. The Industrial Performance Center Globalization Study In order to explore these strategic responses to globalization, a team of MIT researchers carried out 629 interviews in companies in the United States, Japan, China, Taiwan, Hong Kong, Italy, France, Germany, Mexico, and Romania. Almost half of the interviews were in Taiwan and Mainland China (see Tables 3 and 4). The selection of firms was concentrated in electronics, software, automobiles and auto parts, and textiles and apparel, along with a number of interviews in other industries and in government and academia. These four target industries represent a range from high-tech sectors experiencing rapid and continuous technological innovation to lower-tech sectors with slower technological turnover and heavy competitive pressures from lower-wage societies. Akintunde (Tayo) Akinwande and Charles G. Sodini (MIT Department of Electrical Engineering and Computer Science and Microsystems Technology Laboratories), Douglas B. Fuller (MIT Department of Political Science), and Timothy J. Sturgeon (MIT Industrial Performance Center) carried out the research on the electronics sector. Sturgeon, Eric Thun (Department of Politics, Princeton University), Edward Cunningham (MIT Department of Political Science), and Teresa Lynch (MIT Department of Political Science) conducted the work on the automobile and auto parts industry; Dan Breznitz (MIT Department of Political Science) was responsible for the software sector interviews. Marcos Ancelovici, Suzanne Berger, Teresa Lynch and Sara Jane McCaffrey (MIT Department of Political Science) investigated the textile and apparel industries. Edward S. Steinfeld (MIT Department of Political Science) directed the group’s work on China. The entire team met twice a month over the life of the project to frame the study, design the interviews, and analyze the results. We are collectively the authors of Global

xx

PREFACE

Table 3 IPC Interviews: Country of Ownership Country Taiwan United States Japan Italy China Germany France Canada Other Europe* Other Asia** Total

Number 286 94 62 43 34 32 29 11 13 25 629

% 45 15 10 7 5 5 5 2 2 4 100

* Romania, Scotland, Sweden, United Kingdom, Netherlands ** India, Indonesia, Malaysia, Philippines, Singapore, South Korea, Thailand Table 4 IPC Interviews: Country of Plant Country Taiwan United States China Japan Italy France Germany Mexico Canada Other Asia* Other Europe** Total

Number 262 97 69 57 38 26 16 12 8 25 19 629

% 42 15 11 9 6 4 3 2 1 4 3 100

* Thailand, Malaysia, Philippines, South Korea, Singapore, Indonesia ** Scotland and Romania

Taiwan. We gratefully acknowledge the contribution to the project of JiRen Lee of National Taiwan University. We learned much from him during his stay at the IPC in 2000 and his co-authorship added an important dimension to the work. The group also benefited from the participation in our meetings of Andrew B. Bernard (Dartmouth); Donald Lessard (MIT Sloan School), Richard Locke (MIT Sloan School and Political Science), Michael Piore

PREFACE

xxi

(MIT Economics and Political Science), and graduate students Brian Hanson, Llewelyn Hughes, and Georgeta Vidican. Brad Buschur was present on all fronts with assistance for the project. In Taiwan, Jesse Lan and James Wang offered excellent research assistance. To no one, however, does the project owe more than to Anita Kafka, who from beginning to end provided extraordinary administrative support and wise advice. For the funds that supported the overall Globalization Study across its American, Asian, and European sites, we are grateful to the Alfred P. Sloan Foundation, the Volkswagen Foundation, Fujitsu Research Institute, the Chinese National Federation of Industries, and the Ministry of Economic Affairs (Taiwan). Our conviction that studying Taiwan would provide special insights into more general dilemmas of globalization was shared by a group of Taiwanese civil servants, academics, and private-sector actors, who generously and courageously supported our research. Among the many who helped us, we are particularly grateful to George Yang, former state minister for science and technology, who at every stage encouraged our efforts. We wish to acknowledge as well the support and wise counsel of Jung-Chiou Hwang, director general of the Department of Industrial Technology at the Ministry of Economic Affairs. We took to heart the frank advice of Professor Po Chih Chen, who met with us at various early stages of the work, and of Taiwanese academic colleagues with whom we met on July 8, 2000, at the Workshop on Networked Production and Globalization at the Academia Sinica. Despite their best efforts to guide us on our way, we are certain that our work still reflects the deficiencies of newcomers to the study of Taiwan. For these faults we alone are responsible. Our hope is that the Chinese proverb, “He sees well who sees from afar” may have enough truth to justify the endeavor.

This page intentionally left blank

GLOBAL TAIWAN

This page intentionally left blank

1 Globalization and the Future of the Taiwan Miracle Suzanne Berger and Richard K. Lester

The Lessons of Taiwan and Globalization This is a book that presents research by a new generation of scholars drawn to study Taiwan by its extraordinary record of democratization and economic growth over the past fifty years and by the challenges it faces in a time of globalization. Taiwan’s rise stands as an example for developing countries of the possibilities of moving from authoritarian to democratic rule at the same time as a society pursues rapid economic growth. Scholars and policymakers still debate the “lessons of Taiwan” and search in its experiences for validation of competing theories of state and market in developing countries. Today Taiwan has come to have yet another emblematic significance. It is widely regarded as a critical case of the prospects and dangers of globalization for societies with advanced industries and high wages. Taiwan’s relationship with China shows in microcosm the opportunities and vulnerabilities of all developed countries in a world in which capital, goods, and services flow freely across borders and supply chains stretch seamlessly between high- and low-wage economies. The potential and the threat can be seen clearly in the evolution of Taiwan’s information technology (IT) industry—the heart of the Taiwanese economic miracle. By 2002, Taiwanese investors in the Mainland controlled two-thirds of China’s total IT output, and their share was rapidly increasing (Cavey 2003, p. 13; Ernst 2002). These investments provide an enormous boost to the competitiveness of Taiwanese firms by lowering their costs and by allowing an expansion that would be unimaginable in the limited space of a small island with twenty-three million people. The embedding of Taiwanese firms in the fast-growing Chinese market offers vast prospects for the future. But, at the same 3

4 CHAPTER 1

time, manufacturing in Taiwan proper is at risk of being hollowed out; by 2002 about a half of the total IT output of Taiwanese firms was produced on the Mainland (EIU 2003c, p. 38). And the Taiwanese, like Americans, Europeans, and Japanese, ask what will remain? and how, in a world of open borders, can rewarding jobs, innovative activities, and a good society be created at home? Taiwan’s Development In the 1950s Taiwan was a poor society living on foreign aid and agriculture with a per capita income of US$200 a year. From the 1960s on, an export boom based on labor-intensive consumer goods like clothing, toys, lighters, and plastic products helped raise annual growth rates to 10 percent. The first overseas contracts to source electronic components made in Taiwan came in the late 1960s. At the end of the 1980s, Taiwan was emerging as a major player in electronics manufacturing (Ernst 2000, p. 238). By 2001, Taiwanese companies manufactured 70 percent of all personal computer (PC) motherboards, 55 percent of all laptops, 56 percent of all liquid crystal display (LCD) monitors, and 51 percent of all color display tube monitors (Bout 2003, pp. 1–3). Taiwan pioneered the “pure-play” water fabrication plant in the 1980s, and now Taiwan earns over 70 percent of the world’s semiconductor foundry revenues. The sales of the Taiwanese electrical and electronics industry in 2002 totaled almost US$88 billion, with the semiconductor industry adding another US$21.4 billion (MoEA 2003). Taiwan’s per capita GDP in 2003 was US$12,465 (EIU 2003c). How Taiwan achieved this is a story that has been told many times and in many different and contradictory ways (Amsden 1985; Amsden and Chu, 2003; Aoki et al. 1997; Chu 1989; Fei et al. 1979; Gereffi 1996; Hatch and Yamamura 1996; Mathews and Cho 2000; Rodrik 1994; Saxenian 2001; Saxenian and Hsu 2001; Schive 1990; Wade 1990; World Bank 1993; Wu 2005). Explaining Taiwanese success has come to be a major battlefield in the struggle between neoclassical market-based theories of economic development and revisionist theories that attribute the lion’s share of credit to state interventions. Virtually every explanation of the Taiwanese growth experience is contested by another account, and these academic controversies continue to rage. Even among the revisionists there are very different emphases. Some scholars (Wade 1990; Amsden 1985; Amsden and Chu 2003) argue that the state’s relative

GLOBALIZATION AND THE FUTURE

5

impermeability to private interests allowed it to play a major role in identifying objectives for industrial policy, implementing them with credit and trade policies, monitoring firm performance in order to limit rentseeking, and coordinating private and public expectations around growth targets. Rodrik (1994) agrees on the importance of state capabilities, but stresses their impact in raising the level of private investment, rather than in targeting winners. Another variant of the revisionist view (Mathews and Cho 2000) focuses on the sophisticated institutional capabilities that enabled East Asian countries to promote a “national system of economic learning” by leveraging resources in a process of start-up and diffusion in high-tech industries. Both neoclassical and revisionist explanations are challenged by recent scholarship (Wu 2005) that demonstrates the subordination of economics to politics and of the economic policymaking agencies to politicians. The penetration of politics into the bureaucracy, Wu argues, resulted in paralyzing conflicts, and he finds scant evidence of independence or monitoring capabilities. Yet another axis of debate has to do with assigning weights to the importance of foreign influence for Taiwanese development, for instance the role of U.S. foreign nonmilitary aid, which in the first two postwar decades accounted for more than a third of Taiwan’s total investment (Roy 2003, p. 99), and the transfer of technology by foreign multinationals, particularly the Japanese (Schive 1990; Hatch and Yamamura 1996). Firm size and scale are also controversial in the analysis of the Taiwan miracle. Amsden and Chu (2003) emphasize that the critical factor for “second-mover” firms in a latecomer economy like Taiwan is to combine technology acquisition with scale in domestic enterprises, and they interpret Taiwan’s experience as one in which private and public sectors upgraded and expanded. These conclusions on scale are contested by Wu (2005), who presents evidence for the decisive role of small and medium enterprises in triggering export-led growth.1 Amsden and Chu are also skeptical about the significance of foreign investors for Taiwan’s upgrading, but other scholars point to the historical record of the role of multinationals in technology transfer and to the Taiwan government’s current energetic campaign to attract foreign multinational corporations (MNCs), with hefty incentives for locating R&D centers in Taiwan to reinforce innovative activities on the island, even as manufacturing moves out.2 In this book, we do not join these battles, but we do recognize the critical importance of the legacy of the first thirty postwar years in Taiwan,

6 CHAPTER 1

when problems of economic reconstruction and development were inextricably tied up with political legitimation and the survival of the regime. At that time, the Taiwanese government had to control and pacify its own hostile population even as it poured the lion’s share of state resources into military preparations for war with an unyielding adversary across the straits. As in many developing countries today, the Taiwanese rulers had to consider economic policies mainly as a function of their potential for exacerbating or mitigating domestic and foreign security problems. The policies of those years have left deep traces in Taiwanese social and economic institutions. Most important, they consolidated an industrial structure in which the main players have been small and medium-sized companies, and even the larger firms have been distinctly smaller than foreign counterparts. Measured in terms of revenues, the disparity is even greater. Consider even Taiwan Semiconductor Manufacturing (TSMC) and United Microelectronics (UMC), Taiwan’s highly successful semiconductor foundry firms. Although they fabricate about a fifth of the chips in the world, it was only in 2003 that TSMC made it into the ranks of the world’s top 10 semiconductor manufacturers (in 10th place).3 In Taiwan, as in South Korea, Singapore, and Japan, the state bureaucracy did play a major role in identifying, encouraging, protecting, and financing leading sectors of the economy. The principal difference between Taiwan and the others is that elsewhere the main actors in the state–industry dialogue were very large companies (domestic, in the case of South Korea and Japan; foreign, in the case of Singapore) that received the majority of funding and protection. In Taiwan, large firms controlled a significant part of the economy, but they were not the engines of its transformation. During the first three decades of Kuomintang governments, the regime kept large firms, public and private, on a tight leash, fearing the emergence of powerful groups in civil society that might contest the central power.4 This ruled out industrial policies like those of Korea or Japan, which would have made large firms the privileged agents of the state’s push toward modernity. The large state-owned enterprises and private companies in Taiwan were mainly confined to the domestic market and were rewarded with monopoly rents. Banking and credit policies that were originally determined for political reasons launched the economy along a trajectory that developed considerable momentum. Thus, even when political anxieties about the emergence of power centers in

GLOBALIZATION AND THE FUTURE

7

society eased, the industrial transformation did not slow, and it was small and medium-sized companies which continued to drive economic modernization and export-led growth. State policies from the 1970s on cleared obstacles to the expansion of these firms. Many of the resources needed by small and medium-sized companies to fuel their ascent up the ladder of technological development were made available by the state. But—with a few very important exceptions, like the semiconductor foundries—the tremendous leaps and bounds forward in exporting and growth that Taiwan made from the 1970s on were not the work of firms that the state had picked and funded. They were not ships that the state bureaucrats could steer along a course of their own design. Indeed, the intense political competition and conflict within the ranks of economic policy agencies imply that bureaucratic capabilities for steering, monitoring, or correcting the course of the multitude of private actors were quite limited.5 Because of its pursuit of a model of growth based on small and medium-sized companies, Taiwan faced from the beginning the dual problems of competing in world markets with much larger rivals and dealing with customers whose scale dwarfed the Taiwanese. Taiwan’s experiences in positioning itself in global production systems in ways that attenuated these weaknesses and dependencies offer perhaps the most valuable of its lessons for other developing countries. The emergent industries of developing countries do in fact have more opportunities to enter global production systems today because the modularization of production creates the possibility of focusing on a single competence and linking up with other companies in value chains to produce final goods. If in the past a company had to master a broad range of the activities involved in the overall process from product design to sale to final customer, today it is possible, even optimal, to narrow the range of functions carried out within the walls of the firm. This makes it easier for new firms—from developing or developed countries—to enter the market. But the other side of the coin of the new opportunities for entry into an international economy built on networked production are new dependencies and an immediate confrontation with the best-in-class producers from around the world. With the liberalization of trade, the new entrants of developing countries are projected into a world in which they face much larger and more powerful customers, suppliers, and competitors. These are challenges Taiwanese firms have had to master all

8 CHAPTER 1

along. Today, the lessons of the past in building strength in the face of scale provide the Taiwanese with capabilities that can be mobilized in an international economy very different from that of the economic takeoff in the 1970s. New Competitors, New Terms of Competition: Challenges for Taiwan Our own research starts from the fact of Taiwanese success in the 1990s. We ask here whether, in the new global economy, the performance of Taiwanese companies can be maintained and expanded with the same business strategies and public policies as in the past. Can they still work today for Taiwan? Can they work for other countries seeking to move up the ladder? If changes are required to sustain economic performance in an open international economy, what resources are available to be mobilized in the private and public sectors in Taiwan? As digitization and modularization make it possible to reorganize production that once took place in vertically integrated firms into value chains linking independent players, how are Taiwanese firms adjusting in order to thrive in the new environment? As a massive relocation of manufacturing and services moves plants and jobs to low-wage countries like China and India, what will remain at home in societies like Taiwan? As trade liberalization confronts Taiwanese firms with larger and more powerful customers, suppliers, and competitors, how can they cope? How can Taiwanese firms defend and expand the positions they have captured in international production networks in the face, on one side, of rapid innovative advances by powerful lead firms and brands and, on the other side, of the rising capabilities of firms in emerging economies? This last dilemma appears across the board in the interviews we have conducted in Taiwan. For example, in a labor-intensive industry like auto parts for the after-sales market, the Taiwanese have made enormous gains in the American market, as Cunningham, Lynch, and Thun relate in chapter 4 of this volume. But the Taiwanese parts manufacturers still find themselves lightweight and vulnerable in their relations with U.S. distributors and brands even as they observe over their shoulders the advance of lower-cost Chinese producers. In the electronics industry, as Fuller describes in chapter 5 of this volume, Taiwanese firms carved out what appeared to be an unbeatable lead in original equipment manufacturer (OEM) and original design

GLOBALIZATION AND THE FUTURE

9

manufacturer (ODM) markets and by 2002 ranked as the world’s fourthlargest producer of IT hardware (excluding semiconductors and Internet appliances). Today they see themselves in competition not only with very large Western electronics manufacturing services (EMS) firms that have emerged over the past ten years as formidable rivals of the Taiwanese in contract manufacturing but also with Chinese manufacturers, who have rising technical capabilities, low costs, and privileged access to a growing market of consumers. As Sturgeon and Lee explain in chapter 2 of this volume, electronics contract manufacturers like Solectron, Flextronics, Jabil, and Sanmina have acquired a set of capabilities that allow them to provide services along an increasingly broad range of functions, from product design to manufacturing, logistics, and service. The EMS companies have expanded globally by buying up production capacity around the globe from lead firms seeking to shed their manufacturing or by building new facilities. They now are able to produce to the highest technical standards in proximity to their major customers virtually anywhere in the world. Increasingly, as Sturgeon and Lee show, the EMS companies are developing design services for their customers, thus moving into terrain once securely held by Taiwanese ODM firms. At the same time, the Taiwanese face new competition from China, where a new set of domestic firms is beginning to be able to match Taiwanese performance in some segments of IT and consumer electronics industries. In part this is a problem that the Taiwanese themselves have created with their massive investments in Mainland China. Figures on Taiwanese investment in the Mainland are notoriously underestimated, since in order to avoid government restrictions, many projects are funded from Caribbean tax havens or through Hong Kong or Southeast Asia. Mainland statistics showing US$31.2 billion of Taiwanese investment in China over the decade 1992–2002 represent the low end of evaluations (Cavey 2003, p. 12). The Taiwanese today benefit from these investments, but they are simultaneously transferring technological and managerial skills to others, without being able fully to capture and appropriate the stream of benefits of the spillover of these transfers. During the glory years of the PC industry, Taiwanese firms like Acer, HonHai, Quanta, ASUSTeK, Gigabyte, First International Computer, and Mitac dominated large shares of the world market across a wide range of products. Taiwan Semiconductor and UMC pioneered the wafer fabrication foundry business model that allowed chip design firms to focus on product definition and marketing by transferring manufacturing

10 CHAPTER 1

to Taiwan. Taiwanese producers, who had marked out a distinctive and profitable terrain between the high-wage, high-tech companies of advanced economies and the low-wage, low-tech companies of the East, now find themselves squeezed from both sides. Taiwan’s dilemma today is rapidly becoming the major issue for firms in all developing societies as they face both the formidable innovative capabilities and market power of established producers and the rising capabilities of emerging producers within the vast labor reserves of India and China. From our research in Taiwan, two broad, interrelated issues emerge as critical for Taiwan as it faces the pressures and opportunities of globalization: • Product/market strategies: Competing on price versus competing on innovation, differentiation, and brand • Locational strategies: Staying in Taiwan versus moving to China versus moving elsewhere. We discuss each of these issues in the following sections. New Product and Market Strategies The commodification of manufacturing that we describe in this volume as an outcome of technological advances in codifying the interface between functions in the production system poses a particular challenge for Taiwanese firms. Manufacturing capabilities have been the core asset of Taiwan. As Michael Porter has summarized, Taiwan’s strengths are in manufacturing, not services, in quick technology absorption, competition on price, R&D for process improvements, and low-cost inputs (Porter 2001). As the Taiwanese face more and tougher competition today in manufacturing, they have two broad options going forward. In virtually every interview we conducted with Taiwanese entrepreneurs and managers, we found them considering, on one hand, strategies that center on competing on price and lowering costs, and, on the other hand, strategies that use innovation, differentiation, or branding in order to build barriers to entry and to escape competitive pressures. While these two routes may converge in part, they branch out in very different directions, and, as the interviews show, present real and diverging choices for the reorganization of firms’ central functions, firms’ levels of investment in R&D, and the location of company operations, sourcing, and human resources.

GLOBALIZATION AND THE FUTURE 11

Competing on Cost The progressive opening of opportunities to invest in China has created a vast new array of possibilities for Taiwanese firms. Like the Hong Kong industrialists who seized on the opening of China after 1979 as a chance to preserve and expand a pattern of economic growth based on low-cost manufacturing, Taiwanese businessmen have found a solution in China to the rising wages and rents at home. Gaining access to the Chinese market has been a protracted process for Taiwanese companies, since the legal barriers to cross-straits investment have only gradually and partially come down. In the face of large military buildups on the Mainland and strong pressures for “reunification,” Taiwan’s governments have sought to contain the amounts of Taiwanese investment on the Mainland and to restrict the kinds of technologies transferred (Yang and Hung 2003). Successive Taiwanese governments resisted pressures from business for direct transportation and communication links with the Mainland and made strong efforts to regulate the amounts of investment in China and the kinds of technology that can be exported from Taiwan to China. Many Taiwanese businesses circumvented the regulations by investing in China through Hong Kong, other Southeast Asian countries, or funds channeled through the Cayman Islands and other opaque tax havens. Nonetheless, the legal barriers remain very constraining, since they require expensive and time-consuming detours.6 Firms skirting the edge of the law also open themselves up to a variety of pressures and sanctions at home. The “go South” (i.e., to Southeast Asia, not to China) and “less haste, be patient” policies of the 1970s and 1980s have yielded over time to a willingness to establish some direct links and to accept higher levels of technologically more advanced transfers. This was expressed in the “active opening, effective management” policy adopted in 2001 on the recommendations of a conference with broad participation of members of the political, academic, and business communities.7 Today, although many restrictions still remain on the books, the scale of Taiwanese investment in China is enormous. Figures from the Taiwanese and Chinese governments on Taiwanese investment in Mainland projects between 1992 and 2002 range from US$26.4 billion to US$31.2 billion, but some unofficial estimates are as high as US$100 billion (Cavey 2003). In June 2004, Hu Sheng-Cheng, chairman of the Council for Economic Planning and Development (CEPD), estimated that 75 percent of Taiwanese firms’

12 CHAPTER 1

overseas foreign direct investment goes to China (Hille 2004, p. 2 ). The government is currently considering proposals to allow firms to raise funds for investment in China on the Taipei exchange, which would reduce the incentive for detouring the flow of capital through third countries. Whatever the hopes for the future of using Mainland production sites to tap the vast potential market of Chinese consumers, the reality is that today, reducing labor costs still appears to be the main driver of the move to China. Other factors matter as well—market access, finding more workers and engineers with scarce technical skills, acquiring land for expansion—and these will be discussed in subsequent chapters of this book. But, for any Taiwanese firm focusing on a low-cost strategy, a shift of some or all of its operations to China or to other low-wage countries is an invariable component. Some managers, expressing various nationalist or other personal motivations, are reluctant to go to China, and seek out other low-wage countries like Vietnam or the Philippines. In one of the large, integrated textile companies in which we interviewed, for example, the owners were hesitant about political risks in China and so they set up new plants in Swaziland, where they employ 5,000 workers, while still leaving 350 employed in the Taiwan plant. In order to ensure fast deliveries to American customers, this company also has plants in Mexico. The manager calculates that monthly wages in Swaziland are US$140–160 a month—somewhat higher than the US$100–120 a month it would pay in China or the US$80 it would pay in Vietnam, but still much better than the US$500–600 a month it pays in Mexico. Exactly what the wage differential is between China and Taiwan is not easy to determine. In fact, even for the same set of skills, estimates are all over the board, ranging from ten to one to practically the same (Cavey 2003). One manager told us that Chinese engineers may earn one-third of what Taiwanese engineers do, but that the Taiwanese are three times as productive. Others we interviewed talked about the high costs of sending Taiwanese managers to supervise and train in Mainland sites and how this narrowed the cost advantages of the move. In some areas like software, demand may have driven the salaries and benefits of Chinese engineers up close to Taiwanese levels. Still, on average, most firms can count on significant savings for labor if they move some operations to China. They can also expect to reduce their costs of land, water, and construction. In China, as elsewhere, local governments

GLOBALIZATION AND THE FUTURE 13

eager for jobs provide sites on easy terms. The costs of compliance with environmental regulations may also fall when firms shift production abroad. The most serious problem of relying on a strategy of lowering costs is that this is a game that many others can, and do, play. And as they do, the margins for OEM manufacturers shrink. Steinfeld, in chapter 8 of this volume, describes the intense competitive pressures in China as manufacturers concentrate on driving down costs across a range of products with very low barriers to entry. In a world in which everyone is chasing after the same markets—air conditioners or television sets or mobile phones—prices plummet, margins collapse, and soon no one is making profits and the search for yet another product in relatively easy reach begins all over again. The Taiwanese are heavily involved in this scramble to reduce costs and prices through moving operations to China, but the ultimate consequence may be very costly. Competing on Innovation, Brand, and Differentiation Taiwan’s OEM and ODM companies have always faced the challenge of defending their profits in relationships with lead firms that are larger and more powerful. This effect is exacerbated by high dependence on a small number of customers. Sturgeon and Lee in chapter 2 of this volume estimate that five firms—HP/Compaq, Dell, Sony, Apple, and IBM—accounted for 63 percent of all IT hardware made by Taiwanese firms in 2002. In order to reduce dependence and to escape narrowing margins and cutthroat competition, businesses need to offer unique or difficult-to-replace products and services that raise the entry bars for would-be rivals. Brand names are one way to establish such barriers, but thus far Taiwanese firms have not been greatly successful in branding. Many of the brands that have been developed by Taiwanese firms are mainly sold in Taiwan—a market that is too small to offer a solid base for growth. In clothing, for example, 90 percent of clothing with Taiwanese brand names is sold on the local market, evidence of very weak brand penetration outside Taiwan.8 There are promising cases of companies trying to build brands on the Mainland, like Giant, the number-one bicycle brand in China, or BenQ, a mobile phone, PC-peripheral, and notebook manufacturer that split out of Acer and today has about 13,000 employees and close to US$3 billion in sales.9 After two years of developing a brand, in 2003 it

14 CHAPTER 1

already ranked seventh in China in brand recognition after other longestablished Taiwanese firms. BenQ’s revenues enabled it to plan to spend US$59 million on promoting its branded products in 2003. But fundamental issues remain: first, can an OEM or ODM firm develop branding without doing great, even fatal, damage to relationships with its old lead firm customers? Acer’s difficult experiences in trying to build a brand for its PCs while maintaining a parallel OEM business making PCs for other brands demonstrated how fierce the resistance of the lead customers can be to competition from their suppliers and how real the dangers of destroying the relationship. The lesson many of the ODM and OEM firms have drawn is that branding is too risky for their main business. Another difficulty has to do with Taiwan’s distance from its customer markets. As many of the managers we interviewed observed, it is hard, perhaps impossible, to build a brand unless the firm has a deep understanding of the market that develops only when it lives, breathes, innovates, and manufactures there. These managers doubted that Taiwanese firms would ever understand Western consumers well enough to build brands in the United States and Europe. Many firms pin hopes of developing brands on China, a society with which they believe their own cultural affinities are closer and where the physical presence of their business community—some 400,000 Taiwanese (2 percent of the island’s population) work in the Shanghai region alone (Cavey 2003)—provides opportunities for deep learning. Whether common language and traditions are a wide enough bridge to span the profoundly different experiences of these two societies over the past century can be questioned. As one manager said ruefully, “Taiwan and PRC often seem like West and East Germany: one country divided by one language.” Though the Taiwanese believe their common culture provides leverage in China that Westerners do not have, the specificities of China often baffle the Taiwanese as well. Building a brand, moreover, is under any circumstances an expensive affair, and in China the size and regional segmentation of the market make it even more costly. The expenses of protecting the brand by fighting piracy are also very high, as the experiences of, for example, Gold Lion, a Hong Kong firm with a very successful Chinese brand name, show. Creating barriers to entry by competitors by branding is likely to be an option available only to a very limited number of Taiwanese firms with deep pockets and patient capital. The other main route to differentiating products and services lies in

GLOBALIZATION AND THE FUTURE 15

innovation. Taiwan’s expenditures on R&D (2 percent of GDP in 1999) fall below U.S. and Japanese levels (Cavey 2003, p. 23). The government has announced plans to increase spending on R&D over the next few years and it is also offering incentives to foreign multinationals to set up R&D centers in Taiwan. As Fuller reports in chapter 5 of this volume, there has been a good response from foreign firms. But it is still too early to assess whether these centers will establish the kinds of relationships with domestic firms that could produce big spillovers from activities conducted within the foreign-invested labs into the local economy. If we measure innovative capabilities by the amount of patenting, and use the number of patents granted in the United States as the metric, Taiwan looks very strong. Taiwan and Israel are the only two emerging economies to close the gap with the G7 in terms of the patent per capita ratio (with Taiwan next after the United States and Japan).10 South Korea has started to close the gap, but only very recently, and is still far behind; in Europe, only Finland has the same growth trajectory. However, the quality of Taiwanese patents, especially scientific ones, is low. Taiwanese patents are concentrated in manufacturing and process technologies. Public laboratories like the Industrial Technology Research Institute (ITRI) continue to be major sources of patents, but university laboratories do not show up as significant drivers of innovation. Although the apparent contribution of academic research to patents is low, it is possible that the large number of patents registered by individuals represents academic researchers filing under their own names.11 Even if this is the case, this picture still suggests that Taiwan needs to do more to capitalize on the strengths of its universities by bridging the gaps between university laboratories and the private sector more effectively. Patenting is of course not the only and not always the best reflection of innovative potential. As Fuller argues in chapter 5 of this volume, firms with product generations that turn over rapidly may not file patents, since by the time they are granted the product is likely to be out of date. The great strengths of Taiwan’s systems firms are not likely to be reflected in patent statistics, since they often do not find it worthwhile to invest the time and money in filings.12 Still, to the extent that Taiwan’s advantages lie in rapid adaptation of existing products and processes, rather than in radical discontinuous innovation, Taiwanese firms appear more vulnerable to the competition of new rivals, and in particular to the rising capabilities of Chinese firms. One official at ITRI despaired: “We are an OEM country. R&D in Taiwan is not exactly Bell Labs or what it

16 CHAPTER 1

is in the United States IT industry. It’s R&D for manufacturing and process, and that kind of technology is easy to move to China.” Said another: “If we don’t concentrate on R&D, all of our companies will move to China.” Even managers told us that we should not think of Taiwanese companies as R&D based. Many claimed that only ITRI does R&D; as for the rest of them, as one put it, “we are just trying to make money.” They had various explanations of why there is so little genuinely innovative activity in the private sector: the risk-averse character of venture capital, which funds “mezzanine” projects, rather than true start-ups, and the stock bonus system, which rewards short-term earnings, were two that were very frequently cited. As one manager concluded, “This is a culture of fast food and fast dreams. No one wants to wait.” Aside from branding and innovation, there are a variety of other strategies that the companies we visited were implementing. For companies that are good at incremental innovation and “fast followership,” diversification often seems a promising move: at a minimum, it is a way of distributing eggs and risks in a number of baskets, and, at best, it is a way of offering “one-stop shopping” to the customers. One manager expressed his objective with this strategy: “We want to go from being a single tree to being the whole forest.” But as Sturgeon and Lee argue in chapter 2 of this volume, the capabilities that Taiwan OEM and ODM firms have developed in detailed design and production quality and speed are “not easily adaptable to a wide range of electronics products beyond personal computers.” Diversification thus often seems simply to move firms into nearby product niches that are soon themselves under heavy competitive pressures. One issue that came up repeatedly in the interviews with the managers was how to preserve the technology gap with China even as Taiwanese companies partner with domestic Chinese firms. The vice president of a firm making switches, local area network (LAN) products, and other communication products told us that partnering with the Chinese is essential to keeping ahead of them. This is a firm that has 500 engineers in a R&D center at Hsinchu, 100 in a product design center in Shanghai, another eighty in a joint venture with a Chinese state-owned enterprise in Shanghai, and other R&D engineers at three sites in California. The firm sees its Taiwanese center as the main driver of product development. But, the vice president explained, an “in-control strategy” of working with the Chinese is vital because it is the only way to learn their

GLOBALIZATION AND THE FUTURE 17

weaknesses, gauge their general performance level and technical abilities, and thus stay in the driver’s seat.13 But the Taiwanese can hardly count on one-way learning in such relationships. More generally, the question is: How, in a world of fragmented production, can Taiwanese firms get the parts of the production chain that yield high margins and are relatively less vulnerable to competition? With the assets of Taiwan as highly concentrated as they are in the PC supply chain and, more generally, in the hardware side of electronics, the fast-advancing capabilities of Chinese firms (with or without Taiwanese partners in these areas) and the tough new competition from Western electronics manufacturing services (EMS) companies are forcing the Taiwanese to face a momentous set of choices. In retrospect, the present time may be seen, like the shift from import substitution to export orientation at the end of the 1950s, or the drive into electronics in the 1980s, as a critical turn in the road. Exactly what the features of the new road map will be can no longer be traced out by tracking the path of more advanced countries and following in their footsteps. Taiwan has caught up with the others and arrived with them at the frontier. Its dilemma is, however, that strengths that have brought it so far may not be as well adapted to the next part of the journey. For the stretch ahead, Taiwan needs capabilities that move more of its economy out of the grinding competition over costs. In such competition, victories may be short-lived and costly for the future well-being of its own society. Location and Relocation Integral to decisions about product and market strategy are decisions about location. For the Taiwanese firms we studied, the three critical choices revolved around how much activity to locate in China, how much to keep in Taiwan, and how much to locate elsewhere in the world. Moving to China For a Taiwanese business, locating some part of the company’s activities on the Mainland seems to offer enormous pay-offs.14 At the same time, no decision is more fraught with incalculable risks than the choice to invest in China. The most serious risks are political and have to do with the unsettled and potentially explosive relations between Mainland China and Taiwan, a territory that the PRC considers an integral part of

18 CHAPTER 1

China but that many Taiwanese regard as their own independent and sovereign state. Even when a Taiwanese firm has factored the costs of all the remaining barriers to exchange between the two societies into a calculation about the benefits and disadvantages of locating on the Mainland, it cannot take into account the dangers of a flare-up of hostilities, let alone the impact of a potential war. Sometimes anxieties about these risks take the form of “urban legends,” repeated from one ear to the next, as, for example, a current one about an impending collapse of China after the 2008 Olympics. In the category of risks that are easier to verify are the sudden swerves in policy that occur under authoritarian regimes. Despite its entry into the World Trade Organization (WTO), China retains considerable leeway in structuring access to its markets as seen, for example, in the recent ruling requiring that wireless products sold in China incorporate Chinese encryption standards purchased from specified Chinese firms. As The Economist noted, “Dell, Intel, Sony and others may have to choose between sharing technology or curtailing shipments to China” (The Economist 2003). Among Taiwanese entrepreneurs, along with greater or lesser aversion to political risk there are also greater or lesser degrees of Taiwanese pride and nationalism and admiration for, or antagonism to, the PRC. These factors are immeasurable but very significant influences on decisions about locating in China. Beyond the political issues, there are more calculable advantages and disadvantages to building a strategy based on China. As suggested above, reducing costs of labor, land, water, construction, and energy; reducing exposure to expensive environmental regulations; and lowering tax liability in Taiwan are weighty reasons that figure in most accounts managers provide of their Mainland investments. There are also savings on sales into the Chinese market without the extra taxes that are levied on imports. In the case of the semiconductor “fabs” (fabrication plants) that Taiwanese are now constructing or planning to put up on the Mainland, even when wages and transportation costs and time to market are not factors that weigh heavily in favor of Chinese investments, these taxes matter.15 More generally, building on the strong historic, linguistic, and cultural ties between the Chinese in Taiwan and those on the Mainland, the Taiwanese have now acquired a stock of experience in managing Mainland operations that is second to none in the world. As China’s consumer market expands, exploiting these capabilities and deepening

GLOBALIZATION AND THE FUTURE 19

Taiwanese firms’ access to Chinese markets are strong cards for Taiwan. In the Economist Intelligence Unit (EIU) March 2003 survey of Asian senior executives, a third of them pointed to China’s growing domestic market as a prime reason for expanding there, and, for each of the Taiwanese companies we interviewed, the growing Chinese market was mentioned as part of the relocation equation (Cavey 2003, p. 20). Many Taiwanese OEM and ODM firms believe they need production facilities in China in order to work with customers who have already located plants there, whether the customers are selling in the Chinese domestic market or for export. If the customers’ demands were simply for lower procurement prices, then China—or Vietnam, Indonesia, the Philippines, or Burma—might all equally well serve the purpose. If transportation costs and delivery times were sensitive, then choices among these low-labor markets might be constrained. But beyond lowering price, in many industries there is a strong demand to locate near the final producer, and with large swathes of industry moving to China, the Taiwanese OEM and ODM firms feel they have to follow, and they in turn pressure their suppliers to follow them. In industries where co-location of component producers and contract manufacturers is critical for rapid response, the whole cluster needs to move in order to maintain the advantage. When Taiwanese notebook computer makers shifted production to China, the entire supply base moved along with them. As Sturgeon and Lee point out in chapter 2 of this volume, by 2001, ODM production of most notebook computers had left Taiwan. Another example, from among many in our interviews, of pressure from lead firms combined with cluster effects comes from a bicycle component firm that employed 600 workers and eighty engineers in Taiwan. All of their 300 subcontractors were Taiwanese, most of them located close to the bicycle-parts-maker’s plants. But, by 1995, many of the bicycle assemblers were moving to China for cheap labor. The president told us he had to follow them, because to make components, “you need the network; you need the cluster. . . . You need to get firms together. It’s easier to communicate with Taiwanese businessmen. So we went together with them to China. It’s a matter of survival! You can’t stay away!” By 2000, this company had a plant in Shanghai with 1,000 workers and twenty expatriate Taiwanese managers making low-end bikes to sell in China. Two other major motivations for locating in China are the ability to hire more workers and engineers than are available in Taiwan and to

20 CHAPTER 1

hire people with skills different from the technical personnel educated in Taiwanese institutions. The managers we interviewed talked about an absolute shortage of workers and engineers in Taiwan across the range of industries. For lower-tech industries like textiles and garments, managers explained that Taiwanese no longer want to work in such lowprestige, dirty jobs; for high-tech industries, the complaint was that Taiwanese universities are not turning out sufficient quantities of the needed people, or simply that Taiwan has too small a population to produce sufficient numbers of workers. Beyond numbers, the managers noted an advantage in hiring Mainland Chinese—they are thought to be particularly strong in software, in comparison to Taiwanese engineers, who are thought to be good at hardware. A number of companies were planning a division of labor, with their Taiwanese design and product development centers doing hardware and their Chinese R&D centers concentrating on software. As to the downside of operating in China, the themes that surfaced in the interviews were familiar ones. Intellectual property (IP) violations continue to be a major issue, and as Taiwanese companies invest more in innovation and product development, the problem becomes ever more costly. Software applications offer especially easy targets for piracy, and one firm estimated that only 5 percent of the software used in the Mainland has actually been paid for. Even for hardware, there are major IP violations; one manager from a network-adapter-maker guessed that there were twice as many counterfeits of his product on the market as genuine articles. The need for bribes as a regular part of doing business was another point that was regularly raised, with a number of the managers calculating that, when the cost of bribes are added in, the apparent benefits of low-wage labor are significantly reduced. The capabilities of Chinese employees—both workers and engineers— came up in the interviews both on the positive and the negative sides of the balance sheet. Some of the limitations are explained by our respondents as motivational: after decades of Chinese communism, people are not accustomed to assuming personal responsibility for outcomes. The reluctance to take initiative is also exacerbated by resentment of their Taiwanese “brothers,” who are seen as too well-fed, contemptuous, clannish, and having poor understanding of local realities. Though Taiwanese managers identify these negative images as problems, they see no clear solution for eradicating them. In any event, they see themselves in a strong position in the Chinese labor market. Within their own plants,

GLOBALIZATION AND THE FUTURE 21

they remain unchallenged by unions or any other form of workers’ collective action, so there is limited incentive for addressing these issues. Perhaps the most significant limitation of the Chinese managers that Taiwanese firms hire is lack of experience with the West and hence an inability to anticipate or even understand the demands of customers. For example, a company that sells credit-card processing services in China pointed out that credit cards have been in use in China for fewer than three years. In contrast to the Taiwanese employees, who have a long personal experience in this market, the Chinese employees have no idea how to think about new applications for this market. This is a company that, like others, reckons it needs to keep all product development and innovation in Taiwan for the long haul. Staying in Taiwan At the heart of the discussions we had in the Taiwanese firms we visited on the island and on the Mainland were questions about which functions they were going to move abroad—to China; to other low-wage countries, like Vietnam or the Philippines; or to high-tech clusters in the United States, Europe, and Japan—and which functions they were going to keep in Taiwan. Over the long haul, we believe, the industrial activities that remain in Taiwan will be there because they require proximity to sticky assets that cannot readily be transported out of Taiwan. Jobs associated with such activities may be good or bad ones: gardeners hold jobs that are likely to stick in Taiwan, since the gardens cannot be moved, but it is not likely that gardeners will earn good pay. The most important of the sticky assets, therefore, are ones that bind high-value-added functions in global production chains to Taiwanese locations, for these support businesses that can compete in global markets and provide good jobs. China and other low-cost labor production sites will certainly improve their capabilities in high-value-added functions like marketing, R&D, design, and product definition over the coming years. Today, however, the engineering and management skills required for these remain in very short supply in China. There are few successful high-tech companies in China, and their numbers are not growing rapidly. The great strength of the Chinese manufacturing miracle is companies competing on cost in relatively mature products, as Steinfeld, in chapter 8 of this volume, argues. Nearly 85 percent of the value of Chinese high-tech exports still comes from foreign-owned companies, virtually all of which

22 CHAPTER 1

rely on some division of labor between R&D and design centers in Taiwan, Hong Kong, Japan, or some Western country, and manufacturing in China.16 As long as Taiwan can continue to deepen its innovative R&D capabilities on the island and do so at an accelerating pace, companies will find it essential to keep important research, product design, and development centers at home even if they also set up R&D facilities in China. Whether Taiwan will make the public and private investments in research, education, and training needed to attach innovative activities to local soil is a question that lies beyond our predictive range. But this, we believe, will be the decisive factor determining the complementarities and the division of labor between Taiwan and China. Will any manufacturing remain in Taiwan? Is Hong Kong, with only 5 percent of its GDP and less than 10 percent of its workforce remaining in manufacturing, the harbinger of Taiwan’s future? We think not. We see critical differences between the trajectories of Hong Kong and Taiwan in the years after the effective opening of China to foreign investment. In Hong Kong in the twenty years after 1979, few concerted efforts were made to provide incentives for the development of high-tech industries.17 Levels of public and private investment in research were near the bottom of the international charts for developed countries. After China’s opening, Hong Kong manufacturers, who had begun to be alarmed by rising labor costs and who had been considering upgrading their products, realized they could solve many of their potential competitive difficulties by transferring low-end manufacturing to the Mainland. They gradually, over two decades, upgraded the capabilities of Mainland plants. Eventually, the Hong Kong government and the private sector did attempt to stimulate the development of a strong and broad-range indigenous high-tech industry that might have attached some manufacturing activities more durably to Hong Kong territory, but these efforts were too little and too late. In contrast to the Hong Kong story, Taiwan manufacturing over the years has already transformed itself several times: from import substitution to export orientation, from light consumer products in the 1960s and 1970s to electronics in the 1980s and higher-tech in the 1990s. The legacy of adaptation has become a vital part of the capabilities of firms today. As one senior executive put it wryly: “If you want to build something new where there was nothing before, go to the United States. If you want to build something to last for twenty years, go to Japan. If you want to work successfully under always changing conditions and

GLOBALIZATION AND THE FUTURE 23

changing regulatory systems, go to Taiwan.” Even as much manufacturing has moved out from the mid-1990s on, industrialists have continued to invest on the island. Investment in manufacturing increased by 18.1 percent per year in the period 1996–2000, and manufacturing output increased 6 percent per year over the same period (Cavey 2003, p. 15). Although manufacturing was relatively less important to the economy than it was in the 1970s, in 2001 it still represented 26.4 percent of GDP. In Taiwan, this legacy of successive generations of industrial upgrading with a redeployment of resources to progressively more complex activities represents the hope for consolidating some manufacturing employment as well as high-end R&D and product development. To exploit the potential of this legacy requires a capacity to make the valuable products of the future, particularly ones that embed both innovative services and design in the same artifact. This will require, for example, moving from making “plain vanilla” cell phones, where margins are extremely thin, to producing a phone that is a fashion statement and that also bundles a wide variety of services. It takes the ability to work with some high-end customers in designing virtually customized products that will be produced in limited numbers. In other cases, for example in biotech, it will require continuous feedback and interaction between the later stages of product development and the first runs of commercial production. In all of these cases and many others we have identified in the interviews, the vehicles for sustaining manufacturing activities are ones that facilitate communication and action across porous organizational boundaries between marketing, design, product specification, and early-phase manufacturing. These vehicles may be joint product development; close physical proximity and easy interchange between lead customers, lead suppliers, and innovators in R&D labs; pilot plants where customers and manufacturers meet over the first runs of new products; or joint ventures. These mechanisms for promoting the incorporation of new ideas and new demands into new products will surely not all require that manufacturing capabilities be physically located within Taiwan’s territory. But the high-paying industrial jobs that do remain are most likely to be those that leverage the island’s capabilities for pulling innovation into production. One critical question here is whether Taiwan can do this across a broad enough range of sectors to ward off excessive dependence on a particular industry and vulnerability to its cycles. Taiwan’s strengths

24 CHAPTER 1

today are highly concentrated in electronics and, within electronics, in PC-related products. The great expansion of this sector worldwide over the past decade has been a tremendous boost to Taiwan’s growth, but it has also been a source of fragility, for Taiwan’s economy remains extremely sensitive to fluctuations in the IT sector. The sharp downturn in U.S. spending in this area starting in 2000 produced a collapse of manufacturing investment in Taiwan, which was down 34 percent in 2001, and of GDP, which was down 2.2 percent that same year (Cavey 2003). One of the most promising discoveries we made through our interviews is that a certain broadening of Taiwan’s industrial base seems to be occurring. Within electronics, repeated interviews in the same firms— even over the span of this research—found an extension of capabilities into new areas like LCD production and the shift to larger glass substrates. Two of the sectors we focused on in the research, software and automotive parts, have not traditionally been strong performers in the Taiwan economy. Here we identified positive new trends, as Breznitz explains in chapter 7 of this volume on software, and Cunningham, Lynch, and Thun detail in chapter 4 of this volume on after-market auto parts suppliers. To produce a dynamic population of innovative firms creating new products and processes, the broadening of Taiwan’s capabilities beyond electronics need not focus only on sectors defined as high-tech. In fact, even in traditional sectors like toys, bicycles, textiles, and clothing, which have been hard-hit by competition from low-wage producers, we found cases of very creative companies expanding their design and development operations in Taiwan even as they relocated some of their manufacturing. There is room for much more to be done, not only in building new industries like biotech, but also in galvanizing the possibilities for bringing new technologies into older industries. Moves Beyond Taiwan and China For Taiwanese firms, the choice between locations is a critical dimension of corporate strategy. Many of the electronics and software companies we interviewed have some operations in the United States and other developed countries, like Japan, Switzerland, and the United Kingdom. These facilities range from “listening posts” with five to ten employees in high-tech clusters like Silicon Valley, to full-blown R&D laboratories, like Trend Micro’s Utah and San José offices, to assembly plants doing final configuration, like Quanta’s in the United States, the

GLOBALIZATION AND THE FUTURE 25

Netherlands, and the Czech Republic. Running these small-scale R&D operations abroad is very expensive, and a few of the firms report pulling back from them and re-establishing the activity in Taiwan. Overall, however, most of them seem solidly attached to foreign soil, because the companies feel they need proximity to cutting-edge research and product development. The locations of leading research clusters do not change very quickly and so the decisions of a Taiwan firm about whether or not to locate in such clusters abroad do not produce much moving about. It is quite a different matter when the motivating factor for foreign location is proximity to final customers. As discussed above, many of the final customers for Taiwanese firms are now operating in China, and they are pressuring their Taiwanese suppliers to join them there. But aside from China-based relocation, we found other important trends toward a redistribution of Taiwan’s economic activities outside of the island. Changes in transportation costs and the recent availability of good infrastructure in countries on the periphery of the advanced economies are leading to major shifts. A number of the companies we interviewed were in the midst of moving final assembly plants producing goods for the European market from core West European countries like France and the Netherlands to new facilities in Hungary, the Czech Republic, Poland, and Romania. With the expansion of the European Union (EU) farther east, low-cost production sites have moved into the immediate vicinity of large consumer markets and are now, in all significant respects, close to the final customer. Potentially the most important shift we have discovered is from production in North American Free Trade Agreement (NAFTA) countries to production in Asia. Ancelovici and McCaffrey in chapter 6 of this volume remind us of the strong predictions advanced by academics and practitioners that globalization would result in the formation of solid regional blocs, with the majority of world commerce and production taking place within, rather than between, Europe, North America, and Asia. According to this hypothesis, companies would be forced to locate production facilities in each regional bloc if they wanted to sell in consumer markets in that part of the world. Trade agreements and the structure of transportation costs were identified as drivers of this regionalization effect, along with growing pressures from retailers for rapid replenishment, short product cycles, and slim inventories. For example, Harvard researchers working on textile and garment industries

26 CHAPTER 1

analyzed the impact of the concentration of retail trade on industrial location and argued persuasively at the time that most manufacturers servicing U.S. customers would be forced to locate in regions next to the United States, especially in Mexico, in order to satisfy demands for quick response and lower costs.18 The Taiwanese companies we visited, particularly in the early phases of our research, frequently mentioned the demands of American customers for locating in Latin America, preferably in Mexico, in order to get products to them rapidly. But locating in a non-Chinese environment poses enormous challenges for Taiwanese companies, and from our researchers’ interviews in Taiwanese plants in Mexico and from reports of activities in Nicaragua and Central America, it is clear that most of these efforts are not going well. Beyond all the predictable difficulties, there are problems the Taiwanese never envisaged, for example, labor shortages. Mexican workers, unlike Mainland Chinese, are not willing to live in dormitories, and so factories need to recruit workers from the local market, or not at all. Once a number of Taiwanese companies moved into a Mexican locality, they drained the local labor market. During our visits to Taiwanese plants in Mexico, we found production equipment standing idle for lack of workers. One Taiwanese manager recounted with surprise and disappointment the rejection of his request to Mexican authorities to be allowed to import Chinese workers. Wages rose more quickly in Mexico than had been predicted, and it seems that Mexicans are having difficulty developing the capabilities for producing the more diversified, short-batch goods that are demanded from nearby suppliers. But is it really vital for Taiwanese to be able to operate in all regions of the world? Is location near North American and European markets required in order to work with customers in those regions? For anyone addressing these questions at the end of the 1990s, the answer would have been a clear affirmative, since at that time it seemed that only geographic propinquity could satisfy the demands of American and European retailers and lead firms for suppliers to produce goods on ever-shorter lead times and at ever-lower costs. Over the past three years, however, there has been a dramatic shift, which is documented and analyzed by Ancelovici and McCaffrey in chapter 6 of this volume. Firms of all nationalities that had established operations in Mexico—whether through affiliates of their own company, or with outsourcing partners— are leaving in droves and re-establishing their activities in Asia, usually in China.

GLOBALIZATION AND THE FUTURE 27

There are, of course, factors specific to the Mexican situation that may have prompted many of these moves: the fast-rising wages, the disappointment many of our managers expressed in the difficulties of getting high-quality, flexible production out of Mexican plants and into the United States, and a sluggish response of Mexican authorities to solving basic issues of security. Not everything is leaving Mexico: production of large, bulky goods like TV sets and cars with high transport costs is quite likely to stay. Some firms have moved only parts of the production operations out of Mexico and are upgrading the remaining pieces. But still, the trend of shifting production to China is a major one and not confined to Mexico. For example, in our interviews in Europe, we also found managers wondering whether the plants they had moved to Romania would remain there or whether there might not be in the near future yet another move from Romania to China (or further east in Europe to Ukraine? Belarus? Moldova?). There are a number of factors prompting these shifts: rising labor costs in countries on the periphery of advanced industrial nations, the falling cost of air transport,19 and the capability of Asian suppliers to provide from China the full range of services from design to logistics. The result is that globalization is looking ever more like globalization, with a worldwide international division of labor, and less like regionalization, with trade between regional blocs, as one might have imagined in the 1990s. These new trends reopen for Taiwanese managers the question about where in the world they need to locate operations. If, after all, a lot of NAFTA and EU production is shifting from those regions to Asia, why shouldn’t the Taiwanese concentrate their efforts in China? There the stock of past experiences and linguistic and cultural familiarity play to Taiwan’s advantage. The mediocre results of much of Taiwan’s activities outside of Asia also might suggest abandoning these efforts and focusing on China. We think any such conclusion would be a serious mistake. First, a certain number of products are simply too heavy to make transport by air feasible in the foreseeable future, and these products and the components suppliers associated with them need to remain in or close to the major consumer markets. Second, multiple locations are a way of spreading risk.20 Whether it is SARS, politics, earthquakes, or cross-straits tensions, putting everything into China would be too dangerous, and customers are not likely to wish to negotiate large contracts with suppliers working in single locations. Finally, working in foreign locations is the single most important way

28 CHAPTER 1

that Taiwanese firms can acquire the kinds of knowledge required for advances in R&D, product definition, design, and marketing and branding: In short, for acquiring the capabilities associated with high-valueadded production. Among the vehicles for such learning might be R&D labs in the United States and Europe, design teams stationed in large consumer markets, pilot plants abroad, sponsored research in major U.S. and European university laboratories, some manufacturing in these large markets, and a broad effort to provide skilled Taiwanese workers with opportunities for long stays abroad. If one of the key objectives of foreign location is learning, then Taiwanese firms might consider redistributing their assets abroad. Some of the low-end manufacturing that is or might have been done in NAFTA countries or central East Europe could be transferred to China. But at the same time there should also be a far greater investment than there is today in embedding Taiwanese managers and researchers within the societies of the world’s largest consumer markets: the United States, Europe, and Japan. The interviews revealed a number of examples of Taiwanese firms that are making such investments, but there are too few of them and they are doing too little. Such operations are very expensive and their returns usually take a long time to be realized. For small and medium-sized enterprises these costs may be prohibitive. Here we see a positive role for the government in facilitating and financing some part of such investments. This demand was explicitly voiced in a number of the interviews, for example, by the after-market auto parts suppliers whose small scale makes it hard for them to coordinate their approaches to the U.S. market. New Public Policies for Competitive Advantage In the intensifying struggle for global markets, it is Taiwan’s firms that are on the front lines. It is their strategies, and the skill and determination with which they implement them, that will directly affect the economic well-being of Taiwan’s citizens. The impressive record compiled by these companies over the last thirty years provides grounds for optimism. Over that period, Taiwan moved from being a satellite location for labor-intensive manufacturing of low-cost, light consumer products to the center of a new system of production for technology-intensive products—a system that, if not unique to Taiwan, has attributes as highly developed there as anywhere in the world. These characteristics include a high degree of flexibility, a well-honed capacity for rapidly absorbing

GLOBALIZATION AND THE FUTURE 29

new technology and bringing new products to market, strengths in process improvement, and continued cost competitiveness, reinforced in recent years by a strong ability to access low-cost inputs beyond the country’s borders. With the massive investment of Taiwanese human and material resources on the Mainland, Taiwan has built productive capabilities there that extend and complement its own domestic strengths. Taiwan’s companies have achieved all this despite the disadvantages of having to compete against often much larger rivals and having to serve often much more powerful customers. Looking ahead, however, new business strategies will be needed in order to sustain Taiwan’s economic performance in the face of both rapid industrial advances in China and other emerging economies and the intense pressures exerted by lead firms in the West and Japan as both rivals and customers. Although each of the strategies we have discussed—lowering costs and prices, branding, diversification, relocation—will create profitable opportunities for individual companies, the key to future prosperity for Taiwan is the emergence across the economy as a whole of the capacity continually to introduce innovative new products and services on the island, and to diversify into a wider range of high-value-added activities and production networks. This would make the economy less vulnerable to the volatility characteristic of particular high-technology sectors. For Taiwan’s firms, which, especially in the electronics and IT sectors, have already achieved leading positions in design and technology integration, this will require a further transformation of their capabilities to become true technological innovators. What role might public policies play in nurturing these capabilities for innovation? What resources might the public sector mobilize in support of these changes in private industry? Are these resources adequate to the task? And, if not, how can they be enhanced? We have argued that Taiwan’s economic future hinges on its ability to deepen its innovative capabilities at home, and that increased public investment in R&D, education and training, and infrastructure deployment will be needed to attach innovative activities more firmly to local soil. But how should these resources be developed and mobilized? Just as private firms must ask whether the successful strategies and practices of the past will suffice in the new global economy, so too must public authorities consider the options available to them. One such option is to maintain and extend the quasi-governmental laboratory model that has served Taiwan so well in the electronics

30 CHAPTER 1

industry. As Fuller and Breznitz both report in chapters 5 and 7 in this volume, ITRI and its dedicated laboratories, including the Electronic Research Service Organization (ERSO) and the Computer and Communications Laboratory (CCL), have been key channels for the introduction of foreign technology as well as the major focus of R&D activity for the industry. At various stages in the development of the industry, ITRI and its laboratories have identified the most critical components, products, and processes required for future growth and competitiveness; led the effort to develop them or acquired and adapted the relevant technology from elsewhere; and successfully developed and implemented technology transfer mechanisms to local firms. This technical support has been a major contributor to the ability of Taiwan’s firms to move rapidly and successfully into new product niches. ITRI stands as one of the pre-eminent examples around the world of public research institutes playing a strategic role in industrial development. But evidence from the chapters in this volume also points to some of the limitations of the ITRI model when it comes to the twin goals of diversifying and deepening Taiwan’s innovative capabilities. As Breznitz’s discussion of Taiwan’s software sector in chapter 7 of this volume makes clear, under certain circumstances large public research institutes can become obstacles to industrial development if their missions place them in competition with private firms for resources and markets. But even where a clear division of labor in R&D is preserved— as has generally been true of ITRI itself—the setting of targets and priorities by public-sector officials is less likely to be effective in sectors where innovation outcomes depend on close knowledge or interpretation of customer needs. If, as seems probable, the frontier of innovation in IT continues to advance into user industries, with major productivity and performance advances flowing from new IT applications in such sectors as health care and life sciences, business and financial services, and traditional manufacturing industries, as well as in the home, Taiwan’s IT firms need to be actively involved in the development not only of new parts, components, and products, but also of new ways of using these technologies. In this emerging innovation environment, the traditional division of labor in R&D, with Taiwan’s small and medium-sized firms often content to let ITRI take the initiative, is unlikely to be sufficient. The ITRI model, with its mastery of the strategies of fast emulation and adaptation, will also inevitably lose some of its potency as Taiwan continues to approach the technological forefront in the information

GLOBALIZATION AND THE FUTURE 31

industries. Here the key resource will be an enhanced capability for fundamental research in science and engineering at Taiwan’s universities, not only in the core fields of microelectronics, computer science, and communications, but also in other fundamental fields that increasingly are permeated by information science and technology, including material science, molecular biology, nuclear science, and robotics. In the ITRI model there is effectively only one engine of innovation. But today Taiwan needs many engines—a multipolar innovation structure —reflecting the many sources of knowledge that drive innovation in modern, high-technology economies. A major objective of Taiwan’s public policies for technology and innovation must be to expand the number of active centers of innovation to include firms large and small, industries old and new, and universities as well as public research institutes. To that end, Taiwan’s policies for innovation should include: • Intensified efforts to develop critical masses of leading researchers in university departments and laboratories capable of making fundamental advances in science and engineering while simultaneously developing the next generation of technical personnel • Systematic encouragement of cross-disciplinary collaborations in research universities • Closer direct links between universities and industry. Several alternative models to consider include not only the approach taken in the United States, exemplified by the industrial relationships of universities such as MIT and Stanford, but also other successful models such as the ERATO program in Japan, designed to strengthen collaboration between universities, industry, and government in basic research • Exploring and developing new patterns of technology usage through the creation of new communities and networks linking Taiwan’s microelectronics, computer, and communications sectors with its traditional industries—both in manufacturing and services. Finally, if local innovators are to develop new products and services at home, and if foreign investors and innovators are to bring new ideas from elsewhere for development in Taiwan, they must be assured that intellectual property rights will be fully respected and enforced. Protecting intellectual property is a vital consideration in corporate decisions about where to locate innovative activities, and the issue of

32 CHAPTER 1

intellectual property protection remains an important and contentious one in Taiwan’s commercial relations with the United States, even after many years of negotiations. In our interviews with Japanese companies, too, intellectual property concerns were frequently mentioned as a barrier to collaborations in Taiwan. For Taiwan to succeed in becoming an international center for R&D and a destination for foreign investment in research and innovation, more must be done to strengthen the laws and regulations governing intellectual property and the agencies responsible for enforcing these laws. Above all, action is needed to build public understanding of the links between the protection of ideas and innovation and the future of creative economic activity in Taiwan. We see all these actions as necessary to strengthen the existing system of innovation in Taiwan. But—as at other critical moments in Taiwan’s economic history when people made the changes they thought necessary to preserve existing pathways and capabilities—these new solutions may in the end turn out to be transformative.

2 Industry Co-Evolution A Comparison of Taiwan and North American Electronics Contract Manufacturers Timothy J. Sturgeon and Ji-Ren Lee

Introduction The geographic scale and intensity of economic integration has been increasing, if somewhat unsteadily, for many hundreds of years. Without geographic integration, there is autarky and isolation, and so markets and the scale of production remain relatively small, and firms stay largely unaware of and unresponsive to economic activities taking place elsewhere. Global-scale economic integration has two levels, the first rather shallow, the second deeper. The first level is market integration, where firms sell to customers in very distant locations by responding to orders received for a specific good that has been advertised at a specific price. This may put the selling firm in direct competition with firms that are far away. Similarly, a firm selling locally might experience competition from imports. So, with the geographic expansion of markets, global integration is driven forward because firms become aware of and respond to customers and competitors that are distant. The second level of global integration is operations. This assumes a foreign direct investment and a geographically fragmented value chain with different activities scattered across various locations, and some degree of explicit operational control to bind these activities together. For a complex set of reasons that we will explore later, lead firms in the electronics industry have been outsourcing more, that is, turning more to suppliers and less to subsidiaries as they set up global operations. Because geographic fragmentation has increasingly been associated with organizational fragmentation, sales and competition exist not only at the level of markets, but at the level of operations as well, as when suppliers cultivate distant 33

34 CHAPTER 2

customers and vie with one another for business. This chapter explores the dynamic features of global outsourcing through an analysis of two sets of supplier firms in the electronics industry, contract manufacturers based in Taiwan and in North America. The electronics industry is extremely global in an operational sense. Companies send, receive, and act on signals from subsidiaries and suppliers in far-flung places on a daily and sometimes hourly basis. While this has been true for ten to twenty years for activities that deal solely with intangible products, such as currency trading, airline reservations, and voice communications, it is a more recent phenomenon in industries that produce complex tangible goods, such as apparel, motor vehicles, and electronics. In some ways, the low-cost, high-bandwidth voice and data communications network that has grown up to support the global integration in intangible products has laid the groundwork for the operational integration of tangible products, especially as the network has been extended to places with very low labor costs. This infrastructure may in turn lay the groundwork for the next wave of global outsourcing, the organizational and geographic fragmentation of service and professional activities. For this reason, we feel it is important to have an intellectual framework in place that will allow us to better characterize and understand how global-scale production systems evolve over time. Strategic Outsourcing and Industry Co-Evolution In this section we develop a co-evolutionary framework at the industry level by highlighting the interplay between increased strategic outsourcing and the rise of shared supplier networks. Recent literature (e.g., McKelvey 1997; Lewin and Volberda 1999; Lewin, Long, and Carroll 1999) has advanced co-evolution as a framework for analyzing the dynamics within, and the interplay between, firm adaptation and environmental selection. Simply put, firm learning—micro-level evolution—both conditions and is conditioned by the matrix of organizations, institutions, and more general social forces within which the firm is situated–that is, macro-level evolution. While the literature on firm co-evolution usefully situates firm learning within its broader competitive and institutional contexts, it fails to elaborate the co-evolution of lead firms with either individual suppliers or their broader supplier networks. More recently, Nishiguchi (2001) has promoted co-evolution

INDUSTRY CO-EVOLUTION

35

as an ideal framework for the study of interorganizational relations, especially outsourcing. In Nishiguchi’s view, lead, or “convening,” firms co-evolve with their suppliers as they oscillate between exploitative and symbiotic interfirm relationships. Competition, the third leg supporting the evolutionary conceptual stool, is left out of his framework because firms in the same value chain are not portrayed as competing with each other but with discrete value chains convened by other lead firms. Supplier networks are conceptualized, for analytic purposes, as separate and not shared or overlapping. For example, Toyota and its largely symbiotic supplier network competes against General Motors and its largely exploited supplier network. Network evolution occurs as disturbances both internal and external to the symbiotic or exploitative relationships push lead firms and suppliers alike to adapt and innovate. Co-evolution occurs within each network in a process that might be called value chain co-evolution. This chapter explores a case where increased strategic outsourcing by groups of lead firms has, over time, led to the rise of a shared supplier network, one that can be accessed by the industry as a whole, even by lead firms that compete head-to-head in final product markets. Our case is the outsourcing of electronics production and the concomitant rise of a pool of highly competent contract manufacturers based in the United States and Taiwan. In this case, interfirm relations tend to be neither exploitative nor symbiotic in Nishiguchi’s sense, but flexible, modular, and self-reliant. We agree that firm competencies are powerfully conditioned by the complementary competencies that may or may not reside in the value chains in which the firms participate, and that the development (or loss) of firm-level competencies exerts a weak effect on the competency of the entire chain. However, we observe that, under certain conditions, groups of lead firms can interact with broadly shared supply bases to drive the evolution of entire industries in a process we call industry co-evolution. The emergence of shared supply networks can be understood as a coevolutionary process encompassing series of strategic choices and interactions between lead firms and their suppliers that occur in the context of a dynamic environment. While such a conceptualization is in keeping with the line of research that adopts a co-evolutionary view of interorganizational relations (e.g., Nishiguchi 2001; Madhok 2000), our approach postulates an industrywide virtuous cycle between lead firm strategic outsourcing and the development of supplier competencies.

36 CHAPTER 2

Suppliers often increase the quality and scope of their services in response to lead firm requests. Once new supplier competencies are in place, they can be used as a basis to develop relationships with other lead firms, and can influence future lead firm decision making regarding strategic outsourcing if, and this is the crucial element, lead firms and suppliers are not locked into dyadic relationships because of asset specificity, equity ties, geographic propinquity, or other factors. Our conceptualization of the process of industry co-evolution is based on the dynamics of firm capabilities and the characteristics of interfirm linkages. By contrasting our case with the co-evolution literature that centers on how micro- and macro-level factors enable or limit the firmlevel decisions (e.g., McKelvey 1997; Lewin and Volberda 1999), our intent is not to replace micro-level theory or reject the influence of macrolevel forces, but to shift the focus to the meso-level processes that operate at the industry level. We view the development of the firm in both its immediate institutional milieu and in the context of its industry, that is, the sets of interlinked and perhaps competing value chains that serve up specific products to consumers. In this way, we look beyond the boundaries of the firm to understand the dynamics of firm capabilities, but not so far beyond as to lose sight of the ways in which the decisions made within the firm might in turn impact what the external environment has to offer. The Proposition of Transaction Value: A Dynamic View of Transactions Prior to elaborating our model of industry co-evolution, one premise on the nature of interfirm collaborations between upstream and downstream firms has to be established. The purpose of engaging in interfirm collaboration along an industry value chain is to combine two sets of complementary but dissimilar resource configurations in a mutually beneficial manner. The mutual benefits of collaboration include not only those resulting from ex ante resource complementarity but also those that might be generated due to ex post investments in relation-specific assets by either party (Bensaou and Anderson 1999; Dyer and Singh 1998) and through interorganizational learning over time, which in turn may help both firms upgrade their competences (Lee and Chen 2000). Although there might be risks of increased costs associated with relation-specific investments due to the partner’s possible opportunistic behavior

INDUSTRY CO-EVOLUTION

37

(Williamson 1975, 1985), the perceived future value generated from interdependent complementarities might, in practice, overwhelm such risks (Celly, Spekman, and Kamauff 1999; Heide and Miner 1992). In other words, basing interorganizational decisions on transaction costs alone could undermine the realization of collaborative benefits and hence the transaction value of interfirm collaborations (Dyer 1997; Madhok and Tallman 1998; Zajac and Olsen 1993). Concurring with this line of argument, we assert that the ultimate goal of interfirm collaboration is to maximize the overall value, not merely limit the costs, of interfirm linkages. Such a value-creation proposition is particularly valid for analyzing interorganizational initiatives from a co-evolutionary perspective because a certain portion of collaboration value can only be realized through the dynamic process of interaction among economic actors (Madhok 2000, p. 280). Specific patterns of interfirm (vertical) relations come from the cumulative strategic choices made by both lead firms and suppliers. Such choices are not random. Each firm bases its decisions on its existing stock of competence, which Levinthal and Myatt (1994) refer to as a firm’s initial conditions of co-evolution. At the same time, prior actions may constrain a firm’s subsequent decision making (Arthur 1989; Teece, Pisano, and Shuen 1997). A lead firm will outsource as a way to enhance the value-creation opportunities derived from combining its own competence set with those of its partners. Suppliers do not respond to these lead firm initiatives automatically, but within the context of the demands placed on them by other existing customers and their overall strategic goals. We regard strategic outsourcing as a value-enhancing initiative that seeks to utilize the linkages between two sets of complementary competencies, as well as increase profits by leveraging lowcost inputs. In other words, lead firm strategic choices about competence scope have to be complementary with those of suppliers before any value can be achieved through collaboration. Value Chain Co-Evolution from the Supplier’s View In the past decade large vertically integrated firms in a range of industries have adopted strategic outsourcing measures to remain competitive, especially in highly contested and fast-moving markets (Quinn and Hilmer 1994; Fine 1998). By focusing on the competence areas that are essential to, or have the potential for, building competitive advantage,

38 CHAPTER 2

especially in product innovation, marketing, and other activities related to brand development, lead firms have increasingly come to rely upon specialized suppliers for providing best-in-class production services to quickly reap value from innovations while spreading risk in volatile markets (Venkatesan 1992). The electronics industry is an exemplary case. Firms that develop, market, and sell electronics hardware products, from computers to communications switches to mobile-phone handsets, have turned to contract manufacturers for production and even postarchitectural design services.1 While it has been frequently noted that widespread outsourcing results in the emergence of a deverticalized industrial landscape (Hitt, Keats, and DeMarie 1999), little attention has been paid to the evolution of the supply base that has grown up to supply newly deverticalized lead firms (but see Herrigel and Wittke, forthcoming). This lacuna has led to the erroneous characterization of the deverticalization process as one where industries are evolving toward smaller, highly specialized firms, each of which has shed its “noncore” activities to focus on one or a few “core” competencies. The deverticalization trend looks very different from the supplier’s perspective. To meet the growing demand for fullservice, or “turnkey,” outsourcing solutions (Sturgeon 1997, 2002), suppliers have in many cases had to add entirely new competence areas, increasing their scope of activities while improving quality, delivery, and cost performance. Increased outsourcing has also, in many instances, vastly increased the scale of suppliers’ operations. Thus, increased outsourcing has led to a deepening of competence scope and an increase in scale at supplier firms. As supply bases come to be comprised of large, highly capable suppliers, the prospects for increased outsourcing are improved. In this way, suppliers and lead firms co-evolve in a recursive cycle of outsourcing and increasing supply-base capability and scale, which makes the prospects for additional outsourcing more attractive, not just to the lead firms that drove the upgrading of the supply base in the first instance, but for those lead firms just beginning to seriously consider large-scale strategic outsourcing. The full range of supplier characteristics has been overlooked in the realm of theory as well. The central focus of the transaction costs framework (Williamson 1975, 1985) is on whether lead firms choose to make or to buy a particular product, process, or service. Make or buy decisions do depend on the characteristics of suppliers, which are portrayed in the form of a dyad, with suppliers that do not invest in specific assets

INDUSTRY CO-EVOLUTION

39

on one hand (what we call a commodity supplier) and, on the other hand, those suppliers that do depend on specific assets (what we call a captive supplier). The key analytic variable is the degree of asset specificity embodied in the interfirm relationship. Suppliers are portrayed, largely, as either investing in specific assets or not. When asset specificity is present, the organizational field is suddenly reduced and lead firms become locked into using particular suppliers, creating the risk of opportunism and high transaction costs. Critics of the transaction costs framework argue that there are more variables than transaction costs that matter (such as transaction value, see above), or that there are other solutions (e.g., trust) to problems of opportunism than vertical integration, when the long view of repeated interaction is taken into consideration (Granovetter 1985; Thorelli 1986; Johanson and Mattsson 1987; Powell 1987; Jarillo 1988; Lorenz 1988). While we agree with these critics, our view is that there are other important reasons to avoid investing in specific assets, especially in volatile markets. Even where interorganizational trust can be developed, market volatility can disrupt long-term relationships and make supplier investment in assets specific to any single lead firm’s products an unwise proposition since both absolute and relative market positions can change with breathtaking rapidity. In many industries, lead firms and suppliers have, over time, been forced to develop mechanisms to manage their relationships with pragmatism in the face of such volatility (Helper, MacDuffie, and Sabel 2000; Herrigel and Wittke, forthcoming). A key factor facilitating the growth of outsourcing is the lead firm’s desire to achieve greater flexibility and lower risk by reducing fixed investments, especially in manufacturing facilities. Such considerations are especially important when technological change is rapid and market volatility is high, contingencies that make effective capacity planning and efficient capacity utilization of in-house or externalized specific fixed assets extremely difficult. Lead firms use suppliers, then, to ramp output up and down according to changes in demand. For their part, suppliers are able to achieve relatively stable demand profiles, high-capacity utilization rates, and low costs by pooling demand from a large number of customers. This is provided that process technology depends not on specific assets, but on generic assets (assets that are widely applicable to the entire customer pool), and that suppliers have the capacity to capture complex requirements from multiple customers and utilize these data in a format usable by their generic processes. Value chains with

40 CHAPTER 2

such “plug-and-play” characteristics have been referred to as “modular value chains” (Baldwin and Clark 2000; Schilling and Steensma 2001; Sturgeon 2002; Gereffi, Humphrey, and Sturgeon forthcoming). Because complementary competencies form the basis for much interfirm collaboration, co-evolutionary processes are driven by each firm’s managerial discretion and the performance feedback of existing interorganizational relations. But conditions external to the bilateral relationship matter as well. For example, frequent technological changes increase the cost of investments required to continuously upgrade existing capabilities, and competitive pressure in end markets may force lead firms to seek advantages in critical competency areas, the complexity and cost of which furthers the drive toward specialization in particular stages of the value chain. Moreover, when rapid technological change combines with competitive pressure, time-to-market becomes a key survival (not success) factor (Curry and Kenney 1999). Yoffie (1997) highlights the critical nature of both technological change and competitive pressure in driving many electronics-related industries toward a horizontal industry configuration, where competition occurs among firms that have become specialized in specific stages of the value chain. These two external factors then serve as an impetus for lead firms to focus on specific areas of expertise and rely on suppliers for others. In sum, the lead firm’s response to a changing environment induces suppliers to enhance their competence scope, which in turn reinforces the drift toward horizontal industry structure. The lead firm’s increasing reliance on the competence of its suppliers will often be balanced by an effort to exert some degree of control over interfirm transactions. Specifically, any conflict of interest resulting from the supplier’s aggressive expansion of competence scope into areas deemed core to the lead firm will weaken collaborative activities. This potential damage to future business can in turn become a powerful feedback to the supplier’s competence development. In addition to the use of asymmetric competence levels to exercise external control, a lead firm can implement either multiple sourcing (e.g., McMillan 1990; Porter 1980) or parallel sourcing (e.g., Richardson 1993) to enhance its bargaining power over suppliers. In the face of a lead firm’s desire to maintain both flexibility and control, a supplier is often forced to upgrade its manufacturing competence to ensure its competitiveness. There are risks associated with continuous investments in manufacturing, especially if they entail

INDUSTRY CO-EVOLUTION

41

relation-specific investments and lead to long-term captive supply relationships. But modular products and processes (Baldwin and Clark 1997; Sanchez 1996) provide suppliers with opportunities to reduce asset specificity while maintaining their capability to offer flexible manufacturing services. Having put such generic capabilities in place, a supplier can further expand its customer base and serve new application areas, as long as the processes are applicable to those areas. In other words, given the right mix of capabilities, a supplier can reduce its risk through structuring a multiple-customer and/or multiple-business profile. Figure 2.1 compares the standard supplier archetypes, which we refer to as the commodity supplier and the captive supplier, with the type of supplier having a strong multiple-customer and/or multiple-business profile, which we refer to as the turnkey supplier. Commodity suppliers depend on generalized assets (and often produce standard products), and because they connect to customers largely via arms-length market transactions, only a very thin customer interface is needed, such as a telephone ordering system or mailed, faxed, or online order forms. As price is the key factor structuring outsourcing decisions, and since thin linkages and standardized products make supplier switching easy, commodity suppliers are often found in what Nishiguchi (2001) would refer to as exploitative supplier networks. Captive suppliers depend on dedicated assets, and in extreme form serve only one customer and therefore have a single, extremely thick customer interface optimized for high levels of collaboration. There is general agreement in the literature that parties to transactions in networks comprised of captive suppliers are protected from opportunism by reliance on long-term relationships, trust, equity ties, or some other form of thick linkage (Richardson 1972; Thorelli 1986; Johanson and Mattsson 1987; Lorenz 1988, 1992; Jarillo 1988; Bradach and Eccles 1989; Powell 1987, 1990). In Nishiguchi’s framework, captive suppliers tend to be found within symbiotic supplier networks. In our view, this dyadic view of suppliers is misleading. In practice, many suppliers seek to limit interdependence with customers and diversify their business profiles, while at the same time building up generalized assets and thick customer interfaces relative to the commodity supplier archetype. We refer to such suppliers as turnkey suppliers (Sturgeon 1997, 2000) because of their relatively independent stance toward their customers, their high level of competence, and their leveraging of generalized assets to serve multiple customers and/or multiple businesses.

42 CHAPTER 2

Figure 2.1 Comparison of Dyadic Supplier Archetypes with the Turnkey Supplier: General vs. Specific Assets Customer interfaces

General Assets (shared resources)

Specific Assets (dedicated resources)

The commodity supplier

The captive supplier

The turn-key supplier

The term is meant to highlight the relatively limited need for interaction and instruction prior to collaboration. A core competence of turnkey suppliers is the ability to use their generalized assets to serve the very specific needs of their customers, which are understood and monitored via a customer interface that is relatively thin compared to the captive supplier and relatively thick in comparison to the commodity supplier. The customer interface at a turnkey supplier might be comprised, for example, of customer-specific program and product management teams. Turnkey suppliers represent more than a point on a continuum between commodity and captive suppliers. Their existence can change the character of entire industries by creating supply bases that are shared by a wide range of lead firms while allowing a great deal of nonprice data to flow across the interfirm linkage. A multiple-customer/multiple-business customer profile combined with a core of generalized assets not only reduces a supplier’s risk of fixed investment underutilization but also enables industrywide learning opportunities. Competence building through learning from partners has been indicated as one of the strategic goals of interorganizational collaborations in general (e.g., Hamel 1991; Lei and Slocum 1992; Khanna, Gulati, and Nohria 1998; Madhok 2000), and vertical relations in particular (e.g., Bettis, Bradley, and Hamel 1992; Lee and Chen 2000). Depending upon the strategic intent (Hamel 1991) and absorptive capabilities (Cohen and Levinthal 1990) of respective partners, both explicit and embedded knowledge can flow through collaborative initiatives and

INDUSTRY CO-EVOLUTION

43

be internalized by partner firms. In the case of upstream/downstream alliances, turnkey suppliers can acquire, among other things, market information, design concepts, technical specification, quality standards, and process parameters through working with customers with a higher level of competence. Learning can flow from suppliers to lead firms as well. Two of the key pitfalls of strategic outsourcing are the creation of new competitors via supplier upgrading and the loss of critical resources through suppliers that might be shared with competitors. There is substantial research that highlights the potential threat to the buying firm from suppliers’ aggressive learning and subsequent entry into the buyer’s business (e.g., Bettis, Bradley, and Hamel 1992; Makides and Berg 1988). However, we argue that the common benefits that accrue to each partner by developing a shared supply network will be no less, and often greater, than the benefits the lead firm or supplier could realize through bilateral relationships (Khanna, Gulati, and Nohria 1998). The spread of knowledge about standards and general best practices, for example, can help push entire industries toward greater efficiency. Similarly, because core assets are generic, turnkey suppliers can leverage knowledge gained from one customer to better serve others with less risk of compromising core intellectual property, typically product design and pricing data. The risks of intellectual property leakage through a shared supply base are further reduced when product life cycles are very short, as they have become in a broad range of sectors, because the criticality of product data is ephemeral. This it not to say that such risks are not present and do not influence the decision making of lead firms. Intellectual property, customer data, and market knowledge can and do leak though shared suppliers, and suppliers themselves can seek to capture this knowledge with the intention of eventually competing with their customers. The importance of trust as a governing mechanism is reduced, but not by any means eliminated, in modular value chains. Judgments about trust have affected the co-evolution of the two groups of contract manufacturers we analyze in section two of this chapter, for example. Beyond the industry- and firm-level factors underlying co-evolutionary processes, different institutional arrangements in which firms are embedded will affect strategic choices concerning competence scope and interfirm linkages. Institutional factors serve both as facilitating and restricting forces on a firm’s strategic and organizational adaptation (Lewin,

44 CHAPTER 2

Long, and Carroll 1999). Among other factors, geographic proximity matters for facilitating resource exchanges and the fulfillment of customization needs (Porter 1998). Further, the differential capacities of capital markets could limit a firm’s strategic options in constructing new interorganizational relationships through acquisitions, geographic expansion, or both. Industry Co-Evolution The key micro-level features that allow turnkey suppliers to become accessible to a broad customer base are the relatively high level of codification in interfirm transactions and the generic character of the products and services provided to lead firms. At the macro level, the system requires widely accepted standards to describe product features and components, and process technology that resists the buildup of specific assets and thus remains effectively generic. At the meso level, which is the level we would like to stress here, the emergence of significant numbers of turnkey suppliers can create capacity-pooling effects, in which lead firms can easily connect to—and disconnect from—the production and service capacities that reside externally in the supply base. The existence of high-bandwidth interfirm linkage mechanisms allows for the buildup of specific assets to the degree that interorganizational collaboration and learning can take place (creating transaction value), while the generic nature of the core products and services provided puts limits on the buildup of asset specificity and keeps the threat of opportunism from hindering interfirm collaboration because the barriers to exit are relatively low for both parties. Pooling effects and low barriers to relationship exit are created by the generic nature of the products and services provided and the multiple-customer/multiple-business profile of turnkey suppliers. Within modular value chains the forces of symbiosis and exploitation are so entwined that the basic distinction is no longer meaningful. Furthermore, competition becomes part of the analysis, because lead firms can compete head to head while sharing the same set of suppliers and, in some cases, even the same facilities and production lines. We argue that shared supply bases, as they come into existence, exert profound effects on entry conditions, the nature of subsequent competition, and future outsourcing decisions. These effects are distinct from those that can be expected to arise from cases

INDUSTRY CO-EVOLUTION

45

Figure 2.2 Co-Evolution with Competing vs. Shared Supply Bases A) Value chain co-evolution: discrete, competing, supply bases connected to specific lead firms Competition

Lead firm A

Lead firm B

Supply base A

Supply base B

Co-evolution

B) Industry co-evolution: A shared, co-evolving, supply-base connected to multiple lead firms Co-evolution and Competition

Lead firm A

Lead firm” n”

Co-evolution

Shared supply base

where buyer-supplier production networks are discrete and evolve along classically exploitative or symbiotic lines. Shared supply bases, as in the modular production network, tend to generate a self-reinforcing dynamic—a classic network effect—because pooling effects create large external economies of scale and scope and powerful learning effects. These learning effects induce an increasing number of lead firms to tap the network, which in turn further enhances the competence, scale, and scope of the turnkey supply base and induces more lead firms to participate. Thus a historical process is unleashed, where the development of an external supply base through outsourcing encourages further outsourcing, and so on. Figure 2.2 compares an industry with discrete, competing supply bases to an industry with a shared supply base.

46 CHAPTER 2

The Co-Evolution of Electronics Contract Manufacturing in Taiwan and the United States To provide a real-world example of the process of industry co-evolution, and to demonstrate how the approach can be used to understand the dynamics of competition within a specific industry, we now turn to a comparison of electronics contract manufacturers that have emerged from two different institutional environments. These are “original design manufacturers” (ODMs) based largely in Taiwan, and “electronic manufacturing service” (EMS) firms based largely in North America. These two sets of contract manufacturers have played a vital role in creating new opportunities for tightly integrated yet outsourced global production systems and hence in altering the dynamics of competition, organization, and location in electronics and computer-related industries. We make our comparison of ODM and EMS firms through a brief account of their origins, development, and current position in the industry. We argue that different starting points and strategic focuses have led each group of firms to vary in specific ways along three dimensions of competence: value chain scope, product/customer scope, and geographic scope. We then discuss some recent trends that have worked to intensify competition between ODM and EMS firms. ODM Firms Beginning in the early 1990s, the term ODM came into common usage in Taiwan to indicate a specific type of contract manufacturing service where suppliers provide design services along with volume production. The term was used to differentiate this service from manufacturing-only services, which had long been referred to as “original equipment manufacturing,” or “OEM.”2 The emergence of ODM firms was enabled by the high degree of modularity and standardization in personal computer (PC) product architecture and driven by very short product life cycles, which generate a need for a constant stream of design activities to bring new products to market. These conditions helped to create a market for the iterative, postarchitectural design services that ODM firms provide (Lee and Chen 2000). Riding on surging demand for PCs and related products, ODM firms have become a major presence in the design and production of a wide range of PC-related products, including desktop PCs, notebook PCs, motherboards, and peripherals. Table 2.1 shows the

INDUSTRY CO-EVOLUTION

47

Table 2.1 IT Hardware Produced by Taiwanese Firms, 2002 Production value 2002 (US$ Product millions) Notebook PCs 13,922 Desktop PCs 6,933 LCD Monitors 5,646 Motherboards 5,636 CRT Monitors 4,544 Optical Disk Drives 3,146 Servers 1,324 Total 41,151

Production volume 2002 (units x 1,000) 18,199 24,740 18,254 86,511 42,910 79,409 1,485 271,508

Estimated share Average of global value per production unit (US$) volume, 2002 (%) 764.99 61 280.23 23 309.30 61 65.15 75 105.90 51 39.62 45 891.58 30

Source: Market Information Center, 2002 IT Industry Yearbook, July: Market Information Center, “Report on the Competitiveness of Taiwan IT Industries,” April, 2003.

value, volume, and share of global manufacturing of various PC and peripheral devices produced by Taiwanese companies in 2002, the majority of which were supplied under ODM-type contracts, according to the Market Intelligence Center (MIC), a Taipei-based market research organization founded—and still in part funded—by the Taiwan government. In the course of our field research interviews, several of the largest ODM firms stated that approximately one-third of their business was still derived from OEM contracts, one-third from pure ODM, where fully designed products were offered to customers, and one-third from contracts where they collaborated with their lead firm customers on product design.3 In addition, Taiwanese firms produce a high share of the world output of many electronic subsystems that find their way into or are connected to PCs, such as analog modems (60.2 percent), cable modems (65.6 percent), DSL modems (51 percent), network interface cards (64.3 percent), simple local area network hubs (81.2 percent), switched power supplies (77 percent), scanners (91.2 percent), as well as computer keyboards and mice (MIC 2003). A complete understanding of ODM contract manufacturing requires a broader view of Taiwan’s electronics industry, which includes a very strong components sector. In 2002, Taiwanese firms, including industry leaders Taiwan Semiconductor Manufacturing and United Microelectronics, accounted for 72.5 percent of the world’s integrated circuit (IC) foundry services. Taiwan firms also provided 32 percent of the world’s

48 CHAPTER 2

IC packaging services and 38 percent of IC testing services (MIC 2003). These service providers, essentially contract manufacturers for the IC sector, have in turn enabled the development of one of the few clusters of IC design companies, known as “fabless IC design houses,” that have developed outside the United States (the others are in Israel and Canada) (Arensman 2003).4 While there are many different kinds of PC components, subsystems, and peripheral devices that Taiwanese firms are not active in producing or designing (namely software, printers, hard-disk drives, and higher-value semiconductors, such as microprocessors and memory), the sum of the capabilities in Taiwan makes for a powerful, agile supply base for the design, manufacture, and delivery of PCs and related products. Working in close geographic proximity, mostly along the Taipei-Hsinchu corridor in Taiwan, this supply base grew to constitute an extremely efficient system that allowed ODMs to respond very rapidly to orders from lead firms (Dedrick and Kraemer 1998). Notebook computers, for example, which generally have a high enough valueto-weight ratio to make air shipment viable, can be shipped to customers within two to three days of incoming orders. This powerful productive engine has developed, almost in its entirety, in response to orders from lead firms based in other societies. At the same time, the development of contract manufacturing in Taiwan and elsewhere has provided lead firms with an increasing range of sourcing options. This process of co-evolution has meant that Taiwan’s electronics industry has been able to develop without a significant cadre of local lead firms. From the late 1970s to the present day, sourcing from Taiwan has expanded from computer monitors to various components and subsystems, to complete desktop and notebook PC and peripheral systems. Firms from the American PC industry have played an especially important role. In the early 1980s IBM began sourcing PC monitors from television and television tube producers in Taiwan, including Tatung and Chung Hua. As the demand for PCs expanded rapidly and the open architecture of IBM-compatible PCs became firmly established in 1984 with the IBM model AT, some entrepreneurial firms in Taiwan, such as Acer and Mitac, recognized the opportunities and moved aggressively to develop the capability to design PCs and peripheral devices based on the emerging standard. IBM’s modular system architecture relied on a central processing unit (CPU) supplied by Intel and an operating system from Microsoft, and because the contracts famously did not block Intel and Microsoft from selling to IBM’s competitors, a bevy of new entrants,

INDUSTRY CO-EVOLUTION

49

intense price competition, and a series of boom and bust cycles soon followed. These conditions caused contract manufacturing to become a popular strategy for lead firms in the PC industry seeking to cut costs and limit investments in fixed capital. In the late 1980s, a set of firms that had been focused on the design and manufacture of handheld calculators entered the PC ODM field. These firms, which included Quanta, Compal, and Inventec, eventually became the dominant notebook computer ODMs, in part because the design and assembly competencies that drove miniaturization in calculators were well suited to notebook computers, where small size, low weight, and efficient power consumption are key factors for success in the marketplace. In addition, calculators, while much simpler, are similar to PCs in that they are built around a standard CPU. The modular system architecture of the IBM PC also stimulated specialized components firms in Taiwan, such as Hon Hai (Foxconn), Delta, and ASUSTek, to develop components and subsystems compatible with the new standard, such as connectors, power supplies, and motherboards. (Of these three firms, Hon Hai and ASUSTeK have recently won large OEM orders for complete PC and other systems.) These moves were part of a larger worldwide trend in the computer industry away from vertical integration and toward a horizontally organized industry comprised of specialized subindustries. As Baldwin and Clark (2000) show, the modularity of computer architecture can be traced back to the introduction of the System 360 mainframe computer in the 1970s, and to several influential texts that promoted modularity in computer system design. Modularity effectively “marketized” various computer elements and enabled the emergence of a wide range of subindustries to provide components and subsystems such as memory, storage, and displays. The openness of IBM PC architecture and the popularity of the machines served to accelerate this trend dramatically. The modular system architecture of PCs, and the dominant role of the CPU and software operating system in setting system architecture, along with intense competition and short product life cycles, created the conditions for the emergence of a set of firms to specialize in the iterative, postarchitectural portions of product design, including the boardlevel operating system (BIOS), which determines how the machine handles the input and output from its main board to the other elements of the systems, such as storage and displays; and industrial design, which determines the physical appearance of the product. However, because

50 CHAPTER 2

most functionality resides in chipsets and software, things that ODMs do not design, control over the innovative trajectory of the industry has continued to reside in other firms and other societies. Which companies are the largest buyers of electronics hardware from Taiwan? Table 2.2 shows the largest five buyers of information technology (IT) hardware from Taiwanese firms in 2003, and the value of these firms’ purchases back to 1998. IT hardware includes finished products of all kinds, but not components. The top five buyers are all PC companies, but the top ten list of American buyers includes non-PC companies such as Cisco (communications switches), Motorola (mobile-phone handsets), Viewsonic (monitors), Solectron (a U.S.-based contract manufacturer that sources some of its PC motherboards from Taiwanese firms), and Intel (which also sources motherboards in Taiwan). This broadening product mix shows that the long-standing efforts by Taiwan contract manufacturers to move beyond the PC industry is beginning to bear fruit. Aside from Sony, the most important Japanese firms sourcing IT hardware—mostly PCs—from Taiwan ODMs are, alphabetically, Fujitsu, Hitachi, NEC, and Toshiba. The most important European firms, again in alphabetical order, are Actebis (a distributor that assembles and configures PCs), Fujitsu/Siemens (PCs), Nokia, Philips, and Siemens (mobilephone handsets). By comparing the value of foreign buyers’ purchases in 2002 to the total value of IT hardware produced by Taiwanese firms that year, we can estimate that contract manufacturing accounted for 91 percent of the value of all IT hardware produced by Taiwanese firms, leaving only 9 percent accounted for by the sales of own-brand products, such as Acer’s line of branded PCs. Taiwan’s electronics industry, then, can be said to be a clear example of “supplier-oriented industrial upgrading” (Sturgeon and Lester 2004). Taiwan’s ODMs are highly dependent on a small number of buyers. The top five buyers listed in Table 2.2 accounted for 70 percent of all the IT hardware produced by Taiwanese firms in 2002, a figure that is estimated to have increased to 71 percent in 2003 (authors’ calculations, based on MIC 2003). The story of lead firm–ODM co-evolution that we have outlined here reveals a powerful dynamic of outsourcing, upgrading, and subsequent outsourcing. We have described the enabling role of open standards and modular product architecture in PCs, the intense competition and rapid product life cycles that drove lead firms to spread risk and lower costs through outsourcing, as well as the entrepreneurial agility that many

INDUSTRY CO-EVOLUTION

51

Table 2.2 Top Five Buyers of IT Hardware from Taiwan, 1998–2003 (U.S.$ billions)

HP/Compaq Dell Sony Apple IBM Top 5 Others Total

1998 7.1 1.9 0.0 0.8 2.1 11.9 3.7 15.6

1999 10.0 2.8 0.0 1.0 3.0 16.8 4.2 21.0

2000 14.5 4.5 0.0 2.5 3.5 25.0 7.6 32.6

2001 14.4 5.5 0.0 2.5 3.0 25.4 8.2 33.6

2002 15.0 6.5 3.5 2.8 2.5 30.3 13.0 43.3

2003* 15.0 7.0 4.0 3.0 2.5 31.5 13.0 44.5

2003 share (%) 34 16 9 7 6 71 29 100

Source: Authors’ calculations from MIC 2003. Note: IT hardware includes finished products, but not components. * MIC estimate.

firms in Taiwan displayed in shifting to export production by recognizing and seizing new opportunities to specialize in narrow segments of the value chain. But there were two other important factors that have not yet been discussed. The first is the Taiwanese government, which helped by licensing, refining, and disseminating foreign technology and encouraging, and in some cases underwriting, the entry of local firms into promising market areas, especially the area of IT-related products. The second is the role of Japanese technology partners, who provided critical technology that nevertheless came with little support as well as restrictions that inhibited Taiwan firms from building on the technology to develop independent product development strategies. These sorts of licensing agreements continue to be important today, for example in Taiwan’s emerging flat-panel display industry (see Fuller, Akinwande, and Sodini in chapter 3 of this volume). How has this development path positioned Taiwan’s electronics industry within the global IT industry? In 2002, Taiwanese ODM firms supplied approximately 31 percent of all the PCs sold worldwide, but since these machines tended to be models within the low- to mediumprice range, this output only accounted for about 13 percent of the value of sales.5 From these data, we can see that there are large swaths of the worldwide electronics industry and even the PC industry—especially higher-value segments—where Taiwanese firms have very little presence. Nevertheless, it is abundantly clear that Taiwanese electronics firms

52 CHAPTER 2

have become an indispensable part of the global electronics industry, a remarkable achievement for a small island that had an economy based largely on agriculture only thirty years ago. The concentration of ODM firms in the PC industry, and its dependence on several extremely large buyers such as Dell and HewlettPackard/Compaq, comes with several potential threats and challenges to future prosperity. First, the constant erosion of PC selling prices has placed a great deal of pressure on ODMs to lower costs. This pressure was greatly heightened in 1997 when Compaq decided to develop a line of PCs that could retail for less than US$1,000. Dell and other lead firms now market desktop computer systems for less than US$500. The degree of price erosion in PCs can be determined by comparing the production volume to value in the three largest product categories shown in Table 2.1. According to MIC, the total volume of notebook computers, desktop computers, and motherboards produced by Taiwanese firms grew by 29.8 percent, 8.4 percent, and 7.4 percent, respectively, from 2001 to 2002, while the value of these products grew only by 13.1 percent, 1.6 percent, and –0.2 percent, respectively (MIC 2003). Intense price pressure has triggered a huge capacity expansion in Mainland China, where operating costs are lower. The second risk comes from the fact that spending on PCs accounts for a shrinking portion of overall IT spending worldwide. After exploding onto the scene in 1984 with widespread adoption of the IBM model AT, the PC’s share of total worldwide IT6 spending peaked in 1996 at 23 percent. Since then, the applications for IT products and services have continued to proliferate with the advent of the Internet, wireless and digital data communications infrastructure equipment, and the like. By 2003, sales of PCs had dropped to just 18 percent of the value of world IT spending (International Data Corporation 2004). In a category that included data-processing hardware, data communications, and computer peripherals, PCs accounted for about 47 percent of total spending in 2001, a figure that was expected to decline to 44 percent in 2003 (International Data Corporation 2003b). These risks highlight an urgent need for Taiwanese ODMs to diversify their product and customer scope beyond PC-related products, and major ODM firms have sought to do so, with some success. As already mentioned, ASUSTeK, a leader in motherboard design and production, has successfully diversified into notebook computers, VGA cards, optical disk drives, and the manufacturing of the PS2 video game console

INDUSTRY CO-EVOLUTION

53

for Sony. Revenues have increased dramatically, while motherboard sales have dropped to only one-third of total sales. ODM firms such as Hon Hai (Foxconn), Quanta, Compal, and ASUSTeK have allocated significant resources to the design and production of mobile-phone handsets. Mobile-phone handsets provide a logical migration path for ODMs because, like personal computers, they have rapid postarchitectural design cycles based on standard chipsets, and like notebook computers require industrial design expertise related to miniaturization. However, because neither detailed design competencies nor the PC-specific supply base in Taiwan are easily adaptable to a wide range of electronics products beyond PCs, product and customer diversification has been slow and the ODM firms remain quite narrowly focused on PCs, mobile phones, and similar products. Still, except for the near monopolies held by Intel and Microsoft in CPUs and operating software, severe downward pressure on prices and margins is felt at all levels of the PC value chain: lead firms, contract manufacturers, and component suppliers alike. In addition, because of the high degree of design modularity in PCs, software, microprocessor, BIOS, and industrial design can be carried out with a large measure of independence, which lends itself to both the organizational (outsourcing) and geographic fragmentation that we see in the industry. In other words, various elements of the PC value chain can quite easily function within different firms located at great geographic distance from one another. This does not appear to be true for other segments of the electronics industry, especially for higher-value segments, and the geographic concentration of ODM design activities in East Asia and the lack of design and new product introduction facilities in Europe and the United States may be a disadvantage (Sturgeon and Lester 2004). And clearly, the technical and strategic barriers to outsourcing the design of extremely complex, high-value, and often nonstandard products, such as large computers, sophisticated communications switches, and medical devices, remain extremely high. The other strategy that ODMs have pursued to increase their profit margin and leverage in the industry has been to develop their own brand products. As new product categories emerge, there is a constant searching at many ODMs for opportunities to develop their own brands, as with new product categories such as “Internet appliances” that are just beginning to be established. Clearly, if managed successfully, maintaining both contract manufacturing and own-brand businesses can improve a company’s financial performance (Lee, Chen, and Tang, forthcoming).

54 CHAPTER 2

For example, ASUSTeK and Gigabyte sell their own brands of motherboards, especially in Europe, where custom-assembled, or “clone,” PCs are popular. However, ODM efforts to develop own-brand businesses have inevitably created competitive tension with their major buyers. For example, Acer generates about 30 to 40 percent of its revenues from its own PC brand, but has historically struggled with combining its branded PC and contract manufacturing businesses (Sturgeon and Lester 2004). To alleviate this tension, Acer has undergone a series of moves to separate the contract manufacturing and branded segments of its business, most recently by spinning off its contract manufacturing arm (as Wistron) and turning to other Taiwan ODMs (including the leading notebook computer ODM, Quanta) as a source for notebook computers. In 1993, Quanta surpassed Wistron as Acer’s largest supplier of notebook PCs (Tzeng and Chang 2003). The market share of Acer-brand PCs has improved. The company is currently the fifth-largest seller of PCs overall and the second-largest seller of notebook computers in Europe. A common strategy is for ODMs to focus their own brands on serving markets that are of little interest to their customers, such as Chineselanguage markets in East Asia. But recent efforts by lead firms in the PC and mobile phone industries to capture the growing market in Mainland China have made it clear that such markets are indeed of great interest to the customers of ODMs. And so, by and large, ODMs have been forced by their customers to remain focused on detailed design and production only. EMS Firms EMS firms exist in many places throughout the world, but over time the segment has come to be dominated by the largest five firms, all based in North America, namely Solectron, Flextronics, Sanmina-SCI, Celestica, and Jabil Circuit. Solectron, Flextronics,7 and Sanmina-SCI are based in Silicon Valley, California, while Celestica and Jabil are based in Toronto, Canada, and St. Petersburg, Florida, respectively. These firms have heterogeneous origins; some began as contractors to the U.S. space program in the 1970s, one was a contract assembler for mainframe computer and automotive electronics customers in the American Midwest, and still others originated as contract assemblers for the early PC and disk drive industries in Silicon Valley. All of the firms that have ultimately come to dominate the industry have placed an emphasis on quality,

INDUSTRY CO-EVOLUTION

55

and several of the founders held an early fascination with systematic quality improvement schemes and Japanese production methods (Sturgeon 1999). These firms were, in effect, repositories of leading-edge electronics manufacturing expertise in a nation that, in the late 1970s and 1980s, seemed to be falling behind in manufacturing. In the late 1980s and early 1990s, leading EMS firms grew by serving a set of lead firms increasingly focused on product innovation and decreasingly focused on manufacturing. In other words, the move toward strategic outsourcing drove rapid expansion among contract manufacturers. As in the case of the ODMs discussed above, growth in the early days of the PC industry drove much of this expansion. Solectron provided manufacturing services to a variety of early PC companies, including Apple and Osborne. SCI Systems (now part of Sanmina-SCI) was selected to provide main-board assembly for the original IBM PC in 1981, and grew to dominate the early EMS industry, as demand for IBM’s PC products mushroomed. In the latter half of the 1980s, uncertainty created by moves into new, automated production technologies (called surface mount technology, or SMT) and a devastating boom-bust cycle in the PC industry caused lead firms to become hesitant about making new investments in manufacturing capacity, and so they increased their orders from contract manufacturers, both in the United States and overseas. Pooling effects allowed contractors to better amortize rising investments in expensive SMT production equipment (Stanford Case Study 2001). But EMS firms too had been stung by the sudden drop in demand for PCs and disk drives, and so began a long and sustained effort to diversify their customer base, adding customers in industrial and medical electronics, and later, communications and data networking gear. As PC production shifted to Taiwanese firms, the EMS firms also moved up-market within the computer industry to workstations, servers, and mainframes. As late as the early 1990s, contract manufacturers were still perceived in the United States as low-cost sweatshops, or “circuit board stuffers,” with few capabilities and little utility beyond providing supplemental capacity in times of peak demand. All this changed—and changed quickly—beginning in 1991 when Solectron, then the number two EMS firm, won a prestigious national quality award (the Malcolm Baldrige Award) and the leading firm at the time, SCI Systems, broke the US$1 billion revenue mark. Both events did much to legitimate the EMS sector in the eyes of existing and potential customers, and a few pioneering lead firms, including IBM and Hewlett-Packard, moved to sell off their

56 CHAPTER 2

Figure 2.3 Revenues at the Top Twenty EMS Firms, 1999 and 2002 (in U.S.$ thousands) $16,000

$14,000

$12,000

$10,000

$8,000

$6,000

$4,000

$2,000

$0 1

2

3

4

5

6

7

8

9

10

2002

11

12

13

14

15

16

17

18

19

20

1999

Source: Electronic Business, “Top 100 Contract Manufacturers,” 2002.

in-house facilities to leading EMS firms. Thereafter the floodgates opened and business poured into the top EMS firms at a breakneck pace. By the late 1990s, lead firms moved to consolidate their proliferating global outsourcing relationships by awarding global manufacturing responsibility and selling worldwide manufacturing plants, if they had any left, to the largest and most capable EMS firms. Figure 2.3 shows that revenue growth at the top twenty EMS firms between 1999 and 2002 was largely contained within the largest five firms. The largest five EMS firms completed fifty acquisitions of competitor and customer facilities between 1995 and 2002, which, along with the establishment of many new facilities, gave them global operations. Capital for expansion was provided by huge share price increases that came with the stock market run-up in the late 1990s, and international management capabilities were drawn from the deep pool of expertise resident in the United States, Europe, Singapore, and Australia. Solectron, for example, expanded from a single campus in Silicon Valley in 1991 to an astonishing 135 locations by 2001. Seventy of these were in North America and sixty-five were located outside the continent, including sites in Western and Eastern Europe, Asia, and Brazil. Table 2.3 compares the revenues, employment, facilities, and locations of the largest

INDUSTRY CO-EVOLUTION

57

Table 2.3 Comparison of Top Five to Top 100 EMS Firms; Revenues, Employment, Facilities, and Location, 1999 and 2002; Compound Annual Growth Rate (CAGR) 1999–2002; and Top Five Share of Top 100, 2002 CAGR Share of top ’99–’02 (%) 100, 2002 (%)

1999

2002

1,808 8,391 8,624 5,297 2,400 26,520 46,029

13,615 12,261 10,168 8,272 3,729 48,045 68,149

96 13 6 16 16 22 14

20 18 15 12 5 70 100

Employment Flextronics Solectron Sanmina-SCI Celestica Jabil Circuit Top 5 Top 100

18,147 37,963 37,470 18,000 12,000 123,580 262,938

95,000 73,000 46,030 40,000 26,000 280,030 446,386

74 24 7 30 29 31 19

21 16 10 9 6 63 100

Total facilities Flextronics* Solectron** Sanmina-SCI Celestica Jabil Circuit Top 5 Top 100

75 52 64 32 21 244 719

110 135 100 44 31 420 605

47 61 16 11 14 20 –6

18 22 17 7 5 69 100

Facilities outside North America Flextronics* Solectron** Sanmina-SCI Celestica Jabil Circuit Top 5 Top 100

55 31 21 15 9 131 295

75 65 63 30 24 257 313

36 45 44 26 39 25 2

24 21 20 10 8 82 100

Revenues (US$ millions) Flextronics Solectron Sanmina-SCI Celestica Jabil Circuit Top 5 Top 100

Source: Electronic Business, “Top 100 Contract Manufacturers,” 2000 and 2003. *Flextronics facility figures are for 2000; growth rates have been adjusted accordingly. **Solectron facility figures are for 2001; growth rates have been adjusted accordingly.

five EMS firms to the largest 100 EMS firms in 1999 and 2002. The data in Table 2.3 are drawn from an annual survey conducted by Electronic Business Magazine, and are heavily biased toward firms based in the United States; very few Taiwan firms—and none of the ODM firms discussed in the previous section—are included. In 2002 the largest five firms in the

58 CHAPTER 2

survey accounted for approximately 70 percent of the revenues, 63 percent of the employment, 69 percent of the total facilities, and 82 percent of the facilities outside of North America of the largest 100 EMS firms in the survey. Unlike the ODM contractors, EMS firms have, historically, provided few design services. While this has kept their profit margins at very low levels, it has been easier for them to diversify their customer base (and increase revenues) because—unlike design—manufacturing processes have remained relatively applicable to a wide variety of electronics product categories. Still, there are a variety of new competencies that EMS firms have been struggling to develop related to their new, global operational footprint, such as global inventory management, logistics, and capacity planning. In the 1990s, while ODMs were increasing their capabilities in notebook computers, the EMS firms worked to diversify their customer and product portfolios to include a wide range of noncomputer goods, such as medical electronics for both hospital (GE Medical) and home (Johnson & Johnson and Lifescan) settings, consumer electronics, and especially data communications equipment. The rise of the Internet meant a huge boom in the market for data communications gear, including routers, bridges, switches, and hubs. Cisco Systems, based in Silicon Valley, was an early mover in this market, and captured nearly 80 percent of worldwide Internet protocol router sales. Cisco, as is the case with nearly all younger Silicon Valley firms, was and is organized according to a modular value chain model, and so had little in-house manufacturing capacity. Because of frequent engineering changes and high price points, Cisco has chosen to work closely with local contract manufacturers, mostly the large EMS firms based in Silicon Valley, such as Solectron, Sanmina-SCI, and Flextronics, on new product introduction. This model was highly influential at traditional telecommunications firms, such as Nortel, Lucent, Ericsson, and Alcatel, where managers’ intense desire to transform their organizations to participate in the Internet “revolution” led them to emulate Cisco’s strategies. These managers were also rewarded when stock prices increased with every announcement that they were going to shed fixed assets by selling plants to contract manufacturers. The product mix of the largest five firms in 2001 is shown in Figure 2.4. The figure shows that the relative shift away from computer and peripheral equipment was largely due to huge increases in orders from communications equipment firms.

INDUSTRY CO-EVOLUTION

59

Figure 2.4 Product Mix for the Largest Five EMS Firms, 2001

Other 9% Computers and peripherals 28%

Consumer 8% Military 4% Medical 6% Industrial 5%

Communications 40%

Source: Electronic Business, “Top 100 Contract Manufacturers,” 2003.

While product mix data is not yet available for 2002 at the time of this writing, it is very likely that the share of communications equipment in the product mix of these firms has decreased significantly because orders from the telecommunications firms nearly ceased in 2001 when the Internet bubble burst. Still, the product diversity of the EMS firms relative to the ODM firms remains quite high, especially when we consider that the products in the category of computers and peripherals include not only PCs, but a very large fraction of workstations and enterprise computing equipment as well. The strategies pursued by EMS firms during the 1990s came with a set of significant risks. Most serious, perhaps, was the risk of overexpansion. Capital outlays for acquisitions meant the accumulation of huge fixed costs, and in some cases, debt, as well as a need to integrate dozens of new facilities, a job that continues to absorb large amounts of management attention. The cost of acquiring customer facilities was ameliorated to some degree by deals that guaranteed the contract manufacturer a specified level of business by the lead firms selling the plants. But most such deals have now expired, and the long IT recession has meant that the largest EMS firms have found themselves with huge amounts of underutilized capacity. Acquisitions of competitors and the

60 CHAPTER 2

construction of new facilities, of course, came with no such buffer. Because acquired plants had been established and operated by a variety of firms, and many for purposes other than contract manufacturing, the cost and difficulty of integrating these plants have been very high. For a significant number of acquired plants in high-cost locations, such as the United States and Western Europe, acquisition was almost immediately followed by closure. With the drop-off in IT spending, EMS firms began incurring losses. In the cases of Solectron and Sanmina-SCI, these losses were very large indeed, amounting to about US$10 billion between 2001 and 2003. In a less severe downturn, the multiple-customer/ multiple-product stance of the EMS firms might have allowed them to shift from slower-growing or declining products and customers toward faster-growing market segments, but the IT recession of 2001–3 was both broad and deep, and capacity utilization dropped precipitously as a result. Given the recent acquisition binge the EMS firms had been on, this capacity glut came at a very bad time. Solectron, for example, had nearly 5,000 workers in France by late 2000, about half in permanent jobs and half in temporary jobs. Most of these jobs were in its Bordeaux plant, which had been acquired from IBM in 1992. By July 2003, Solectron’s employment in France had shrunk to less than 1,500, with the Bordeaux plant slated for closure later in the year. Plants in Brittany and Nord-Pas-de-Calais had been acquired from Alcatel and almost immediately closed. The remaining work from the Bordeaux plant was transferred to Solectron plants in Romania and China (Le Monde 2003). Of these sudden plant closures, coming as they did on the heels of an equally dramatic acquisition spree, Solectron’s president could only say, “Our customers, who are the ones to decide, want our production in low-wage countries . . . it’s a question of survival” (Le Monde 2003). Solectron has lost more than US$6 billion since 2001 and has had its entire management team replaced. Differences and Similarities Between ODM and EMS Firms Because of their different histories and institutional contexts, ODM and EMS firms have shown significant differences in all three dimensions of business scope that we have focused on in this chapter: customer and product scope, value chain scope, and geographic scope. ODM firms provide a wider range of value chain activities than EMS firms do, especially

INDUSTRY CO-EVOLUTION

61

in the area of postarchitectural product design and development. As already discussed, lead firms in the PC industry, such as Hewlett-Packard/ Compaq and Dell, have retained marketing, product strategy, and much of their conceptual design activities. Ongoing innovation by component producers Intel and Microsoft dictates most of the standard machine architectures that postarchitectural design must adhere to, but ODM firms have become world leaders in bringing PC products to market quickly by joining product concepts developed by PC companies with the changing product architectures dictated by the component makers. EMS firms have so far concentrated mainly on the base manufacturing processes that are common to most electronics products and subsystems, particularly circuit board and product-level assembly, and despite recent efforts, have been much slower to develop design businesses. This difference in value chain scope is related to differences in product and customer scope. EMS firms are able to leverage the same production processes across a wide range of product categories, including computers, communications equipment, consumer electronics, electronic instruments, industrial electronics, medical electronics, and electronics for military and aerospace applications; while ODM firms have largely concentrated their portfolio in PC-related product categories, an area where they have built up design capabilities. ODM and EMS firms are also different in regard to the geographic scale and scope of their production facilities. Owing in large part to their aggressive acquisition of lead firms’ manufacturing facilities and their response to customer requests to provide one-stop global manufacturing services, the largest EMS firms have many more manufacturing locations than do ODM firms. As EMS firms have set up global operations, logistics and supply-chain management have become important competencies. ODM firms, on the other hand, have their manufacturing sites geographically concentrated in Taiwan and Mainland China. The differences between ODM and EMS firms are summarized in Figure 2.5. Despite differences, both ODM and EMS suppliers have contributed to the emergence of value chain modularity in the electronics industry because they share the following key features: 1. Good product and component standardization; highly codifiable output from the design process; rapid technological change at the component, process, and final product levels; product and

62 CHAPTER 2

Figure 2.5 Comparison of Typical ODM and EMS Firms: Value Chain Scope, Product/Customer Scope, and Geographic Scope

Value Chain Scope Distribution/Service Sales/Marketing

Product/Customer Scope Product Strategy

Key Lead Firm Functions

US-Based EMS Firm

Product Definition

etc. Industrial

Product Design Manufacturing

Medical TaiwanBased ODM Firm

Logistics

Datacom Telecom Servers PCs

Geographic Scope East Asia • North Am • Eur • East Eur • South Am

2.

3. 4. 5.

market proliferation; and, during most time periods, rapid industry growth An overlapping set of customers, initially consisting of American electronics systems firms, which have ceded control over manufacturing to outside service providers A quasi-merchant, noncaptive, multiple-customer/multipleproduct stance toward customers and markets Reliance on a generic, easily transferable, widely applicable core of fixed assets Reliance on mechanisms to support a thick customer interface (relative to commodity suppliers).

The difference in value chain scope between ODM and EMS firms alters the source, but not the thickness of the interfirm link. The ODM supplier’s involvement in postarchitectural design requires a thicker linkage mechanism relative to the commodity supplier, but the standard and modular characteristics of PC design architecture cause the link between ODMs and their customers to be thin, relative to the captive supplier. Despite a sustained attempt to increase co-design activities,

INDUSTRY CO-EVOLUTION

63

asset specificity in EMS firms is generally limited to customer-specific program and project management activities that strive to give the customer the impression that EMS production is “an extension of the customer’s organization.” But the more complex, less standardized, and less modular products that EMS firms generally produce work to thicken linkages with customers, and recent fieldwork has shown a great deal of interaction around new product introduction activities such as prototyping, board layout, and process validation. For the ODMs, the link between conceptual design and iterative design tends to be relatively thin because of the unusual roles that Intel and Microsoft play in setting PC design parameters. While industry-standard manufacturing capabilities are a vital element of both the ODM and EMS firms’ competence profiles, this variation shows that suppliers can have a range of relationships with different customers and with the same customers across different contracts (Herrigel and Wittke forthcoming).8 These dynamics are illustrated by Figure 2.6, which highlights the variable thickness of the turnkey supplier’s customer interface in the context of co-design (thicker) and build-to-print (thinner). Unlike the commodity supplier, both ODM and EMS firms tend to have some resources dedicated to specific customers. At the center of both EMS and ODM firms, however, lies a core of general assets that can be quickly shifted away from any one customer and toward other existing and new customers. The specific assets that do exist in both the ODM and EMS firms, since they are much shallower than those of the captive supplier and are combined with a core of general assets, actually facilitate customer switching because they ease the process of initial collaboration. This points again to the fluid character of modular value chains. Growing Convergence and Competition During the 1980s and most of the 1990s, contract manufacturers based in Taiwan and North America co-evolved with a large measure of independence, and so experienced little direct competition with one another despite sharing some of the same customers, such as Dell, IBM, HewlettPackard, and Compaq. As already discussed, ODM firms have concentrated on the design and manufacture of motherboards and fully assembled mass-market desktop PCs and more recently the design and assembly of notebook computers, especially those models with low to

64 CHAPTER 2

Figure 2.6 Variation in the Thickness of the Interfirm Link: General vs. Specific Assets in the Context of Co-Design and Build-to-Print

General Assets (potentially shared resources)

The commodity supplier

Specific Assets (dedicated resources)

Customer interfaces

The captive supplier

Co-design

Build-to-print

(Turn-key supplier, instance A)

(Turn-key supplier, instance B)

medium price ranges. EMS firms have also produced computers and related products as their main business, but have tended to focus on higher-price-point PCs, computer servers, workstations, and large machines for enterprise computing, as well as disk arrays, printers, and other higher-value computer peripherals. There are several very recent trends that have both pushed ODM and EMS firms toward direct competition and shifted the competitive advantage, at least temporarily, in the direction of the ODM firms. The first reason is the rise in lead firm demand for design services. Lead

INDUSTRY CO-EVOLUTION

65

firms have already outsourced manufacturing, and so the outsourcing of routine design activities, as well as the full design of low-cost product lines, seems to be the next logical step. The 2000–2003 market downturn caused these cost-cutting efforts to become especially urgent. Each and every large contract manufacturer has created a design services group in response to this interest, and some have successfully completed projects, albeit mainly for small, start-up firms with little if any design competence. Flextronics’s management has perhaps been the most aggressive in this regard, and has targeted ODM design services as a key strategic growth area (Balyko 2002). Still, it is clear that the design competencies at EMS firms fall far short of those at the ODMs. There appears to be a fundamental difficulty that is inhibiting rapid expansion of design services at EMS firms, namely their broad product scope, which is an advantage in allowing manufacturing loads to be more easily balanced, but a drawback because it makes it difficult to develop deep expertise in any single product area. Design, it seems, is not a generic activity. A second factor increasing head-to-head competition and advantaging ODM firms has been the rapid expansion of manufacturing capacity in Mainland China by both sets of firms. Customers have encouraged all of their contractors to expand in China, in part because of lower costs there relative to other East Asian centers of electronics production such as Singapore, Malaysia, and Thailand; in part because of the lure of the huge potential market in China; in part to increase scale economies; and in part to take advantage of the critical mass of component manufacturers that have moved production there. But the ODM firms have a head start in this regard. The migration of Taiwan’s electronics production base to Mainland China began in the mid-1990s with components and peripherals, then spread to desktop PCs and motherboards, with the latest stage being notebook computers. This migration has established two industry clusters for electronics manufacturing, one in the Pearl River Delta near Hong Kong and the second in the Yangtze River Delta region near Shanghai. Under the tight control of their Taiwan headquarters, the ODM firms’ manufacturing subsidiaries have co-located with their component suppliers to respond to the time-to-market ramp-up and cost-reduction requirements set by lead firms. As a result, the share of the value of electronic hardware production by Taiwanese firms in Mainland China increased from 35 percent in 1999 to 37 percent in 2001, to 47 percent in 2002, while the share of production in Taiwan decreased from 47 percent in 2001 to 36

66 CHAPTER 2

Table 2.4 Taiwanese ODM Firms Overseas and Chinese Production Volume Ratio by Major Product Category, 2001 and 2002 Overseas production (%) 2001 2002 Notebook PCs Desktop PCs Motherboards LCD Monitors CRT Monitors Optical Disk Drives Servers

12 86 55 26 98 94 49

43 94 63 76 98 95 54

Chinese production (%) 2001 2002 5 48 52 23 66 92 16

38 52 62 68 72 92 24

Source: Market Information Center, 2002 IT Industry Yearbook, July 2002; Market Information Center, 2003 IT Industry Yearbook, July 2003.

percent in 2002 (MIC 2003). Table 2.4 provides evidence of the increasing concentration of the ODM firms’ overseas production in Mainland China. Some EMS firms, such as Flextronics and SCI, have had manufacturing plants in China since the early 1990s, and all of the top EMS firms have scrambled to add capacity there since the 2001 industry downturn. Flextronics now has two large industrial parks in China. Table 2.5 provides an example of the recent geographic shift to China by a leading EMS firm. The share of Solectron’s geographic net sales generated in China jumped from 10 percent in the first nine months of fiscal year 2003 (September 2002 through May 2003) to 17 percent in the corresponding period of fiscal year 2004. At the same time sales in all “high cost” locations declined from 50 percent to 42 percent. The United States was hardest hit, with its share dropping precipitously from 34 percent to 27 percent. Still, the total capacity of EMS firms in China remains far below that of the ODMs. What had been a major advantage of the EMS firms relative to the ODMs, namely their global operational footprint (Sturgeon and Lester 2004), became a liability as demand dropped off sharply for the products of plants in high-cost locations (enterprise computers and communications infrastructure equipment), and plants located regionally for fast response (such as those in Mexico and Eastern Europe) were shifted to China to reduce costs. The earlier surge in investments from Taiwan in the 1990s has clearly put the ODMs at an advantage in China.

INDUSTRY CO-EVOLUTION

67

Table 2.5 Solectron’s Geographic Net Sales, Nine Months Ended May 31, 2003, and 2004

United States Other North and Latin America Europe Malaysia China Other Asia Pacific Total Net Sales High cost geographies* Low cost geographies*

Nine-month sales (US$) 2003 2004 2,487.5 2,341.3 1,005.4 1,195.2 1,042.7 773.7 880.5 7,385.0 3,682.7 3,702.3

1,308.0 1,246.3 1,349.8 1,463.3 915.4 8,624.1 3,587.6 5,036.5

Share (%) 2003 2004 34 27 14 16 14 10 12 100 50 50

15 14 16 17 11 100 42 58

Source: Solectron Quarterly Report filed July 2, 2004. *High cost geographies are the United States and Europe. All others are considered “low cost.”

Another trend that is giving ODM firms an advantage is the recent surge in notebook computer demand for home use, driven by lower prices, increased functionality, the advent of low-cost wireless routers for the home, and display quality that has begun to rival desktop machines. The big change came in the 1999–2000 time period, when the share of notebook computer sales, which had increased incrementally from about 10 percent of total PC sales in the late 1980s to about 16 percent in 1998, began to increase more rapidly from 17 percent in 1999 to 27 percent in 2003 (International Data Corporation 2003a). The historical strength of ODM firms in notebook computers means that they are being favored as the market tilts in their direction. Finally, to the extent that Japanese firms have embraced strategic outsourcing along the lines of value chain modularity—and most have been very reluctant and slow to do so—they have chosen Taiwan contract manufacturers over their North American rivals. As Table 2.2 on page 51 shows, Sony did not begin sourcing from Taiwan ODMs until 2002, but MIC estimates that such purchases amounted to US$4 billion in 2003, a figure that places Sony as the third-largest purchaser of IT hardware from Taiwan firms. Toshiba is also a major customer, though not in the top five. While the European lead firms seem to have largely embraced American-style contract manufacturing, Japanese lead firms,

68 CHAPTER 2

after a period of questioning in the late 1990s, appear to have retrenched to their traditional strategies for vertical integration. While this variation is striking, and may have to do with the product categories in which Japanese firms are concentrated, namely small-form factor consumer electronics, where miniaturization necessitates product architectures that are highly integrated and product specific (e.g., portable audio electronics and digital cameras), vertical integration appears to be a difficult strategy to maintain in the face of heavy pressure to reduce costs. The pattern of revenue and net income growth at the largest five EMS and ODM firms tells a dramatic story of heady growth, crisis, increased competition, and reversal of fortune. In the period 1993 to 2000, EMS revenues outpaced revenue growth at the largest five ODM firms by a significant margin. The EMS firms were typically awarded larger and greater numbers of orders by lead firms. Net income was relatively low during this period—below 5 percent—but firms often chose to exchange short-term profitability for increases in market share. Notwithstanding the decline in EMS revenues in the 2001–2 period due to a sudden drop in demand for communications infrastructure equipment, between 1993 and 2002, 75 percent of the total revenues generated by the top five ODM and EMS firms combined (US$336.7 billion) were captured by the top five EMS firms, as shown in Figure 2.7. There are three reasons for this. First, their larger geographic scope allowed EMS firms to take over global manufacturing responsibility for their customers. In addition, the narrower value chain scope of the EMS firms allowed them to rapidly add customers in new product segments, such as communications infrastructure equipment, while the ODM firms grew mostly by capturing a larger share of the production within the expanding PC industry. Finally, EMS firms grew both organically and through acquisition (of both competitors and customer facilities), while ODM firms grew only organically, scaling up largely on Mainland China. In other words, the ODM firms’ focus on a single electronics subsector, PCs, may have limited their growth relative to the EMS firms, which have been able to expand into a wide spectrum of electronics subsectors, especially the burgeoning data communications and medical electronics fields. Another possible factor has to do with value chain scope. Because ODM firms have expertise in product design, lead firms may have been wary about granting them too much business as they increased their strategic outsourcing, instead choosing EMS firms which, because they do far less design work, were perceived as less of a potential threat.

INDUSTRY CO-EVOLUTION

69

Figure 2.7 Revenue Comparison of Largest Five ODM and EMS Firms: 1993–2003 (in U.S. $ millions) $60,000

$50,000

Top 5 ODM firms

Top 5 EMS firms

$40,000

$30,000

$20,000

$10,000

$0 1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

Source: Company annual reports. Note: The largest five EMS firms are Flextronics, Solectron, Sanmina-SCI, Celestica, and Jabil. The largest five ODM firms are Hon Hai, Quanta, Acer, Compal, and ASUSTeK.

Similarly, ODM firms may have fallen into a “competency trap” of sorts (Levitt and March 1988). Because of their higher degree of specific assets related to design, ODM firms may have necessarily had to move more slowly into noncomputer sectors because of the time required to build up the requisite design expertise. There are, of course, other possible causes for the growth differential between ODM and EMS firms that do not have to do with business scope. These include macro-level environmental factors like the share price increases in the United States stock markets in the 1990s, which provided capital for the largest EMS firms to acquire dozens of customer facilities and smaller competitors. ODM firms were mostly listed on the Taiwanese stock market, and this made growth through acquisition extremely difficult. Additional factors, such as the long experience of American managers with international operations and a desire on the

70 CHAPTER 2

part of American lead firms for spatial and social propinquity with their increasingly important outsourcing partners—at least on the headquarters level—may have also played a part. Ironically, perhaps, the differences that led EMS firms to outgrow ODM firms in the 1990s have worked against them since the downturn. Firms may temporarily substitute market share for profitability, but this can turn out to be a very dangerous strategy in volatile industries such as electronics. EMS firms’ diversification into noncomputer product categories, an effort to reduce demand cyclicality through diversification, placed them at the center of the largest boom-and-bust cycle the electronics industry has ever witnessed. Moreover, their strategy of growth through acquisition left them with dozens of underutilized plants in highcost locations, and as the downturn dragged on, the situation was exacerbated by customer demands to shift production to low-cost locations. As already discussed, EMS firms have been scrambling to close plants in high-cost locations, a very expensive proposition, while at the same time investing heavily in new production capacity in China. Finally, the narrow value chain scope that had enabled rapid expansion into new product areas began to work against the EMS firms as customers began to look for outsourcing partners that could quickly take over the design for low-cost segments of their product lines. Much of this business has gone to ODM firms. As a result of all of these trends, revenue growth at ODM firms has far outpaced the EMS firms in the past several years. In the period from 1993 to 2000, the revenues of the top five EMS firms grew 50 percent per year, while the revenues of the top five ODM firms grew 36 percent per year. In the period from 2000 to 2003, however, the revenue of the top five ODM firms grew 34 percent per year, while the revenue of the top five EMS firms, as a whole, declined 5 percent. Much of this decline can be attributed to Solectron and Sanmina-SCI, which together lost 90 percent of the more than US$10 billion lost by the top five EMS firms during this period. But even without this precipitous, and perhaps temporary, drop in profitability, ODM firms have consistently performed better than EMS firms. Figure 2.8 compares returns on sales for the largest five ODM and EMS firms from 1993 to 2003. During the 1990s, it was rare for a major EMS firm to exceed 5 percent returns on sales, while it was equally rare for a major ODM firm to drop below 5 percent. The systematic difference in profitability might be explained by the design services provided by

INDUSTRY CO-EVOLUTION

71

Figure 2.8 Return on Sales Comparison of Largest Five ODM and EMS Electronics Contract Manufacturers: 1993–2003 15.0%

10.0% Top 5 ODM firms Top 5 EMS firms

5.0%

0.0% 1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

-5.0%

–10.0%

–15.0%

Source: Company annual reports. Notes: For ODMs, data were collected for 1993–2003; for EMS firms, data were collected for 1995–2003. The largest five EMS firms are Flextronics, Solectron, Sanmina-SCI, Celestica, and Jabil. The largest five ODM firms are Hon Hai, Quanta, Acer, Compal, and ASUSTeK.

ODMs, which generally have a higher rate of profit than manufacturing, and the fact that at least half of EMS revenues are derived from purchasing components from contracts negotiated by their customers, a service with very little, if any, profit, since customers are well aware of the prices that EMS firms are charged by component manufacturers and distributors. This reversal of fortunes may turn out to be permanent and signal the beginning of a permanent decline of the North American EMS firms relative to the Taiwan ODMs, as outsourcing of design increases and production continues to concentrate in China, or it may not. There are two scenarios that could bolster the competitive position of the EMS firms. The first is that the EMS firms develop competitive design competencies and increase their footprint in China to match that of the ODM firms. To us, this seems unlikely, given the large head start that the ODM firms have in these two areas. The second, and perhaps more plausible, scenario is that a full-scale revival of worldwide IT spending, perhaps combined with a new “killer” application or major technological shift in production technology, will once again constrain capacity. This would

72 CHAPTER 2

work to alleviate extreme downward price pressure, and put a premium on co-location in advanced economies to support co-design and collaboration in new product introduction. It may also revive interest in regional production (e.g., in Mexico and Eastern Europe) to support customization and rapid time-to-market. If such shifts do occur, and they have as their innovative epicenter the places that have traditionally spawned them in the past (e.g., Silicon Valley), the North American EMS firms will likely still be in an excellent position to win this new business. A third scenario, and one that is perhaps the most likely, is for the continued co-existence and expansion of both sets of contract manufacturers, as the demand for manufacturing services continues to expand, and a blurring of the distinctions between the two groups of firms. The firm that is now the largest Taiwanese contract manufacturer, Hon Hai, provides few design services and thus might better be labeled an EMS firm than an ODM. If the largest EMS firm, Flextronics, is successful in developing its design services business, it may make more sense to call it an ODM firm. Only time will tell us which scenario is correct, or if some other outcome will emerge, but we can be sure that rapid innovation and volatility will continue to characterize the electronics industry for some time to come. Conclusion: Implications for Strategy and Further Research By exploring the emergence of a shared, modular supply base from a co-evolutionary perspective we have tried to provide a framework for understanding the modern industrial landscape, especially as it has come to be characterized by strategic outsourcing and vertical specialization. Questions about the dynamics and evolutionary development of interfirm vertical relationships across different value chains have implications for both the formulation of competitive strategy and for future research. First, our argument suggests that the key organizational transformation underway should be thought of not as deverticalization but as delinking production-related activities from innovation-related activities (Sturgeon 2000). In Schumpeterian competition, innovative firms that gain market share build up large productive assets specific to the product and create barriers to new entry. Building on Schumpeter’s conception of innovation (1942), Nelson and Winter (1982) show that the

INDUSTRY CO-EVOLUTION

73

dynamic process of industrial evolution tends to create larger firms and more concentrated market structures over time because successful innovations lead to higher profits and greater investments in productive capacity that put innovative firms ahead of their competitors and then protect them from new market entry. However, the emergence of modular value chains alters the relationships between innovative capacity, firm size and scope, and entry conditions. Firms that outsource a large share of their manufacturing no longer have to carry the financial, administrative, and technical burdens of fixed capital related to production, allowing them to focus on innovation and become more organizationally and geographically flexible (Sturgeon 2000, p. 16). On the other hand, the market position of dominant firms is no longer protected by large-scale firm-specific investments in plants and equipment, making market entry more feasible because existing firms are no longer buffered from competitive pressure by large amounts of in-house fixed capital. Barriers to entry are lower because competitors can tap the same modular supply base and therefore gain access to leading-edge production capacity. Thus, competition between lead firms becomes more tightly tied to product-level innovation, increasing the pressure on firms to execute in the areas of product development, brand development, and marketing. The rising costs associated with an intensified focus on innovation and marketing in turn make it less likely that lead firms will invest in in-house productive capacity in the future, speeding the shift toward modular industry organization, a shift that augers well for the continued growth of contract manufacturers of all types. Second, our research into the emergence of a shared supply base creates a fruitful arena to explore the relationship between organizational capabilities and production routines. To be capable of providing turnkey production services at competitive prices, a contract manufacturer has to specialize in a set of generic production routines that can be efficiently applied across a wide range of customers. By doing so, a supplier can effectively elevate the level of capacity utilization and hence reduce overall costs. On the other hand, a contract manufacturer may have to accommodate various degrees of customer-specific needs for differentiation purposes. Furthermore, as customers demand services at sites around the globe, suppliers have to replicate and/or integrate their routines across national borders while maintaining service integrity. The sudden increase in demand for production services in China has been to the advantage of ODMs, but as we have already suggested, an upturn

74 CHAPTER 2

could easily turn the tables back in favor of the EMS firms by increasing the value of regional production systems and the ability to tie these regional systems together in an integrated global operational footprint. Third, research on the changing patterns of industrial outsourcing offers an excellent opportunity to examine the determinants of a firm’s boundary decisions. The traditional analytical tool for explaining boundary decisions lies in transaction cost theory (Williamson 1975, 1985). This theory requires decision makers to consider the costs that may be incurred due to transaction-specific investments made by either party in order to decide whether to outsource or produce in-house. However, as even complex transactions become easier to codify, and the buyer-supplier relationship within the context of the modular value chain becomes quasimerchant in nature (Sturgeon 2000), the outsourcing firm’s concern over the costs of governing external transactions with suppliers matters less than its desire to benefit from access to capabilities that are costly to develop internally and impossible to acquire through arms-length market transactions (Barney 1999). In other words, the potential level of value that can be added by partners and potential partners becomes a critical determinant of decisions about boundaries between transacting firms and the extent of outsourcing activities (Zajac and Oslen 1993). Fourth, one of the central questions for policymakers and entrepreneurs in Taiwan is how to develop lead firms with strong global brands and a firm grasp on the innovative trajectory of the products they make and the markets they serve. Using the above discussion as a window into recent trends in global economic organization, it is easier to explain why the electronics industry in Taiwan has been highly successful in some areas, namely large-scale manufacturing and postarchitectural design, while being so unsuccessful in other areas, namely product definition, brand development, and new market creation. This is because new forms of industrial organization, such as value chain modularity, allow these sets of functions to be efficiently coordinated among separate firms and separate locations. The policy questions that arise are profound. Should Taiwan be satisfied to be a vibrant hub for component and contract manufacturing, or should it strive to develop capabilities in the realms of branding and product strategy? In field interviews, respondents from EMS and ODM contractors, as well as from their customers, repeatedly stressed the pitfalls associated with contract manufacturers moving into branded products. Competing with customers, even in small ways, risks “killing the golden goose.”

INDUSTRY CO-EVOLUTION

75

If moving into branded products is not to be a sequential step from contract manufacturing, then are there other ways that Taiwan, as a whole, can move in the direction of branding and other high-value activities? One way could be for Taiwan hardware start-ups to rely on the contract manufacturing capacity in their backyard and across the globe, instead of seeking to build up in-house capacity, just as start-ups in Silicon Valley and as the many fabless semiconductor design houses in Taiwan do. But to do this, Taiwan entrepreneurs would have to think in new ways, moving from a “manufacturing-first” to an “idea-first” approach to new firm formation. These questions are especially pertinent as the bulk of Taiwan’s electronics manufacturing base shifts to Mainland China. And for Taiwan and North America alike, the critical issue may be, not the relative competitiveness of its firms, but the rapid shift of production— and perhaps even design—employment to China.

3 Leading, Following, or Cooked Goose? Explaining Innovation Successes and Failures in Taiwan’s Electronics Industry Douglas B. Fuller, Akintunde I. Akinwande, and Charles G. Sodini

This chapter evaluates the ability of Taiwanese firms to become technological innovators as the Taiwanese economy shifts from a period of catch-up to the process of developing new technologies befitting its now high level of economic development. Rather than ask the older question of how can Taiwan develop, this chapter asks in which technologies can Taiwanese firms succeed in becoming innovators and in which technologies are their prospects of attaining innovation quite limited? From an analysis of the attempts of Taiwanese firms to compete in three high-technology product segments of the electronics industry—complementary metal-oxide-semiconductor (CMOS) logic fabrication, dynamic random access memory (DRAM), and active matrix liquid crystal displays (AMLCDs)—this chapter argues that Taiwan can become an innovator in those product areas sharing the following characteristics: a high level of granularity in the production chain, high-volume production, manufacturing-based outputs, and no requirement for large amounts of patient capital. Product areas sharing these characteristics are able to utilize the strengths of the Taiwanese economy and, drawing upon these strengths, greatly increase the likelihood that firms can overcome the technological and economic barriers to innovation. Volume production means the opposite of craft or small-batch artisanal production made famous in German and Italian industrial districts. Patient capital is capital willing and able to accept substandard returns for sustained periods in the hope of long-term gains. The term granularity used in this chapter is quite similar to the term 76

LEADING, FOLLOWING, OR COOKED GOOSE? 77

modularity frequently used to describe similar phenomena, so clarification of the distinction between the two is in order. The basic idea of modularity is that processes (production chains in this case) can be broken down into parts (modules) with clearly defined (a) functions and (b) interfaces between the different modules. These clearly defined interfaces simply hand over processes from one function (module) to the next. Granularity refers to the degree to which processes (production chains in this case) have been decomposed into distinct functional parts (modules). For example, if we take a value chain that has five functions and another that has ten functions, then the tenfunction value chain is more granular. Thus, granularity is a measurement of how many modules compose a production chain. These product characteristics depend on certain underlying strengths in Taiwan’s economy. The precise relationship between these characteristics and Taiwan’s economic structure will be discussed in the conclusion of this chapter. The dependent variable, innovation, is defined as the ability to create the next generation of product or process technology. Absorption of the latest generation of technology from others is not innovation. Cases of successful absorption of technology are examples of classic technological followership, rather than innovation. While government initiatives and foreign technology transfers may have been necessary to foster innovation in Taiwan (Hong 1997, Mathews and Cho 2000), the comparison of the three product segments that we are analyzing shows that foreign technology transfer and government initiatives are not sufficient to bring about successful innovation. All of these product segments originally received some technology from abroad. Thus, technology transfer may have been necessary, but the nature of the technology transfer in terms of the motivation of the foreign counterparts suggests that the need for technology transfer has been a relatively low hurdle. The foreign firms transferring the technology were interested in unloading maturing technologies faced with declining profit margins on to the Taiwanese. Receiving these maturing technologies was the easy part, the hard part was using these technology transfers as a base from which to innovate. The case studies detailed below demonstrate that some segments were able to build from these humble origins and some segments were not. Borrowing Akamatsu’s imagery of the flying geese pattern of East Asian technological development, this chapter attempts to assess in

78 CHAPTER 3

what segments of the electronics industry Taiwanese firms have moved from the position of technological catch-up (the follower goose position), to one of technological innovation (the lead goose position). Many scholars have taken issue with the assumption, shared by the product cycle theories of Akamatsu and Vernon (Vernon 1966, Akamatsu 1962), that the follower countries will naturally be able to grasp those technologies that the lead countries no longer find profitable to produce. Here we raise another critical issue that generally has been ignored in debates on development in East Asia, at least until the financial crisis brought a torrent of criticism down on every aspect of economic policymaking in Asia. It is the problem of the cooked goose, that is, the problem of wasting resources on segments in which one may not even be able to perform as a follower goose, let alone move into the lead goose position. Comparing Taiwanese efforts in CMOS logic, DRAM, and AMLCD technologies, this chapter argues that Taiwanese firms have succeeded in becoming globally recognized innovators in CMOS logic, are emerging as classic followers in AMLCDs, and are in a precarious position in the DRAM segment, a potential cooked goose. What has differentiated the successful case of innovation in CMOS logic from the other two examples? The Taiwanese were able to take advantage of the technological trends in this segment to create a whole new business model, one suitably adapted to emerging technological trends and to the characteristics of Taiwan’s own industrial structure. Necessary, but not sufficient, factors to account for Taiwanese firms’ success in innovation in CMOS include an influx of Taiwanese returnees from the United States, the state’s role in actively training personnel for the industry, the state’s role as capitalist for the principal Taiwanese ventures in this field, and creation of the necessary infrastructure for the industry by both state and private actors (Hong 1997, Hsu 1997, Mathews and Cho 2000). Conversely, why do Taiwanese DRAM firms face such a precarious future? Principally, the Taiwanese economic structure is ill suited to meet the scale requirements needed in this segment. The general lesson to be drawn from these case studies is that Taiwan can be successful in certain product segments. First, we will take a detailed look at how each case matches up against the determining product characteristics. Then, we will discuss how the Taiwanese economy supports and facilitates firms making products with the requisite characteristics and punishes those making products without these characteristics.

LEADING, FOLLOWING, OR COOKED GOOSE? 79

The Lead Goose: The Success of CMOS Logic Fabrication Starting in the late 1960s, the Taiwanese government realized that microelectronics would be an important factor in electronics in the coming decades and that further growth of Taiwan’s electronics industry would depend on developing this segment. After canvassing various foreign experts and organizing a team of Taiwanese Americans involved in the semiconductor industry in the United States to serve as advisers, the Taiwanese government signed an agreement with RCA of the United States to transfer CMOS technology. RCA was willing to transfer CMOS technology for reasons similar to those of later technology transfer partners with which the Taiwanese would deal. Competitive pressure eroding profit margins, the high capital demands necessary for production, and the idea that the transferring firm could best use its own resources to invest in R&D in product areas with potentially higher rates of return were all factors in the decision to transfer the technology to the Taiwanese. The Taiwanese were not, however, simply passive recipients in the technology transfer of CMOS logic technology to the government laboratory. Electronics Research Service Organization (ERSO) and its parent organization, the Industrial Technology Research Institute (ITRI), the Ministry of Economic Affairs (MOEA), the Science and Technology Advisory Group, high ranking party and government officials such as Premier Sun and Minister Without Portfolio Kuo-ting Li, the most prominent Taiwanese technical universities, and several state or Nationalist Party–owned banks all cooperated to promote the industry. ERSO/ITRI, with the help of other government agencies, served as a locus of R&D support for new firms and was the organization from which many of the new Taiwanese integrated circuit (IC) companies were spun off, including United Microelectronics Corporation (UMC), Winbond,1 and Taiwan Semiconductor Manufacturing Corporation (TSMC). The Hsinchu Science-Based Industrial Park (HSBIP) was created in 1980 and related industrial services were developed by ERSO. Before these government agencies were spun off as private firms, such as Taiwan Mask, clustering was fostered to create the requisite infrastructure needed for a worldclass IC industry. Furthermore, state and local firms upgraded and expanded Taiwan’s human capital by luring back Taiwanese from the U.S. IC industry and providing engineering training in the major technical universities as well as further postgraduate training within ERSO/ITRI. These policies to

80 CHAPTER 3

promote Taiwanese industry were necessary to ensure that private firms could innovate on their own and leave behind technological dependency on foreign licensing. Without this effort led by the state to build up industrial and human capital resources, private firms would not even have entered the IC industry, as the formidable technological barriers were too intimidating to the generally small firms that composed the private sector of Taiwan’s economy (Hong 1997, Hsu 1997, Mathews and Cho 2000). Furthermore, these same human capital and industrial infrastructure factors were equally necessary if any Taiwanese firms were going to compete in the liquid crystal display (LCD) industry. What separates the story of success in CMOS logic from the other segments of the Taiwanese electronics industry examined here is the presence of an innovative business model. When the Taiwanese first started to create private IC firms in the early 1980s, the IC industry was dominated by integrated device manufacturers (IDMs) that vertically integrated the production of ICs from the design segment to fabrication through to the test and assembly stages. Led by a returnee from the United States, Morris Chang, TSMC experimented with a new type of IC firm organization that took advantage of several emerging technological opportunities. Chang conceived of a firm that would focus on fabrication of the ICs, leaving the design and the test and assembly stages to other firms. While many IDMs had served as foundries for chip designers as a sideline to their own businesses, this new model was called the pure-play foundry model because it would focus solely on being a fabrication foundry for its customers and undertake no other activities. This decoupling of the IC design and fabrication segments had previously been demonstrated by some collaborative projects between universities and the government’s Defense Advanced Research Projects Agency (DARPA) in the United States, and led to the formation of MOSIS, a prototyping and small-volume production service for integrated circuit developers which essentially provided technical proof of the concept of a digital interface of design and fabrication. However, no one had proven that this concept would work in actual IC production. The decoupling of the IC design and fabrication stages was made possible by the ability to codify knowledge of device characteristics in computer models using computer aided design (CAD) technology. TSMC actually may have overanticipated this trend by trying to use this model in the late 1980s, whereas complete computer modeling of the interaction was not possible until about 1993–94. Nevertheless, TSMC certainly

LEADING, FOLLOWING, OR COOKED GOOSE? 81

reaped the advantages of being the first mover in this field, as the technological trends justified this new type of fabrication-only firm, the pureplay foundry, and it outstripped its Taiwanese rivals to become the largest IC producer in Taiwan in 1993. TSMC’s sales revenues were US$475 million in 1993 and grew to roughly US$5.3 billion in 2000, an increase of over 1,100 percent (China Economic News Service 2001a). TSMC invented the pure-play foundry model, and it is still the largest pureplay foundry today. One example of how difficult it was to manage a fabless design house before the emergence of TSMC and its pure-play foundry model was the operational tightrope act that Crystal, one of the early American fabless design houses, had to carry out to ensure the fabrication of its chips. Crystal had to coordinate production done in the fabs of seven different IDMs in order to have access to the best foundry service for a particular product and to ensure that the fabless firm was not too dependent on a single IDM firm. The logistical nightmare that Crystal had to endure in a world without pure-play foundries points to the great opportunity for those firms that eventually figured out how to be pure providers of foundry services. The pure-play foundry model succeeded not only because the trend of decoupling led to the emergence of a large number of fabless design houses in search of fabrication services, but also due to the ability of pure-play foundries to capitalize on the benefits of focus. Pure-play foundries devoted their energies to increasing the sophistication of their wafer production rather than having their energies and R&D budgets split between product innovation and fabrication process innovation. The pure-play foundry model was also ideal for firms that were not at the technological frontier because they could learn from their customers primarily through customer feedback. While this type of learning continues to this day, the feedback mechanism and more direct demands and technology transfers were critical in the early stages to be able to catch up with the process technology of advanced firms. For example, VLSI Technology transferred specifications for 1.2-micron technology to help TSMC upgrade to that level of process technology (Mathews and Cho 2000, p. 172). The Taiwanese foundries, with TSMC in the lead, soon became world leaders in being able to produce multiple products via multiple processes within a single fab, and achieve extremely high production yields. Today, there are credible reports that the leading Taiwanese foundries

82 CHAPTER 3

are among the industry leaders in process technology (Cataldo 2001). The yearly benchmarking done by the quasigovernmental ITRI placed Taiwan, represented by Taiwan’s foundries, at the technological frontier, along with the United States and Japan, in CMOS logic fabrication at 0.13-micron process technology (TSIA 2001). TSMC has also been cooperating with other leading IC firms, including members of the Crolles2 alliance and NEC to align their next generation of process technology, 0.09-micron technology (Lepedus 2002). UMC recently acquired the Holy Grail of the pure-play foundry business by receiving central processing unit (CPU) orders from one of the CPU IDM giants, AMD, marking the first time that these complex products have been outsourced to the foundries by the IDMs producing them. However, it must be acknowledged that the Taiwanese foundries recently have also been able to take advantage of the trend toward embodying greater amounts of process technology in the capital equipment to keep up with the technological frontier. Could the Taiwanese have advanced as far in process innovation without resorting to this new model? The evidence suggests that it is unlikely. The Taiwanese firms that have continued upon the IDM path have not experienced the same level of revenue growth as the foundries. More importantly, they do not appear to be leaders in either process or product innovation. Leaving aside the DRAM case to be discussed below, the firms that decided to pursue the IDM model fall into three categories: failed or small firms;2 firms that were converted to the pure-play foundry model once that model proved so successful (UMC, the Acer subsidiary in its various permutations, and Holtek); and several firms struggling to simultaneously design higher-value-added products and invest in production facilities, such as Macronix (MXIC) and Winbond, which have also engaged in DRAM fabrication. A recent announcement suggests that even MXIC may relinquish the IDM strategy by spinning off its design functions (Digitimes 2002). The Taiwanese pure-play foundries are not only the acknowledged manufacturing leaders in terms of process technology, yield, and running multiple products and multiple processes in the same fabs. They also continue to maintain their first-mover advantage in the critical manufacturing logistics of the foundry business. The newest tactic to propel forward the process frontier is to use the most advanced process technology to fabricate customers’ chips, for example, customers have been encouraged to have their chips produced on the newest

LEADING, FOLLOWING, OR COOKED GOOSE? 83

process lines of 0.13 micron rather than on lines using the still advanced, but not cutting-edge, 0.18-micron process.3 It was not the invention of a new model itself that determined success. Rather, the new model of the pure-play foundry was able to capitalize on the product characteristics that play to Taiwan’s strengths. The model avoided the problem of the need for lots of patient capital by not trying to do everything (the expensive IDM route) and, conversely, benefited from focus through granularizing the production chain and picking one part of the chain in which to specialize. The problem of entering an industry segment requiring lots of patient capital will be amply demonstrated by the DRAM case presented below. The success of Taiwan’s leading foundries using relatively limited capital to focus on a particular segment of the IC production chain begs the question of why other nations did not also adopt this strategy. First, the strategy had tremendous risk when Morris Chang adopted it in the late 1980s because it was not clear if the digital interface between fabrication and design would work for commercial production. By taking this risk, Taiwan gained first-mover advantage over potential foundry rivals and maintains its process technology lead to this day. Secondly, the other major IC players (the European Union, the United States, Japan, and Korea) all had developed successful IDMs by the mid-1990s when it was becoming clear that the pure-play foundry model worked. Thus, they had demonstrated their own route to success. In the U.S. IC industry, granularization had already begun to emerge, but the United States then and now has the lead in IC design technology, so it has made sense for the U.S. technologists to establish fabless design houses to pursue their comparative advantage on the design side (Macher, Mowery, and Hodges 1999, pp. 253, 268). Aspiring IC players, including Singapore, China, Malaysia, and even Korea (once the DRAM model appeared to go sour with the failure of LG Electronics) have attempted to mimic the Taiwanese pure-play model since the latter half of the 1990s, but even the relatively successful Chartered of Singapore remains well behind the Taiwanese foundries in terms of market share and technology (TSIA 2001, p. 5), as Table 3.1 shows. The Cooked Goose: The Precarious State of DRAMS Starting with the agreement between Acer and Texas Instruments in 1989, Taiwanese firms began to enter the DRAM market.4 While the

84 CHAPTER 3

Table 3.1 Foundry Market Share of the Three Largest Pure-Play Foundries

Firm TSMC UMC Chartered (Singapore)

2001 revenue (US$ billions)

2002 projected revenue (US$ billions)

3.705 1.898 0.463

4.9 2.185 0.490

2001 worldwide foundry market share (%) 54.17 27.75 6.77

Source: IC Insights, at www.icinsights.com.

new entrants benefited from the same background conditions that helped the CMOS logic segment in terms of the same human capital resources and the continued support of ITRI and the Taiwanese state in general in facilitating the deepening of the infrastructure and R&D efforts, the DRAM ventures never went beyond passive acquisition of technology from foreign partners. The foreign firms transferring the DRAM technology were not motivated in ways that were fundamentally different from the motivations of the firms involved in the CMOS transfer, except perhaps that the competitive pressures were higher and the margins were being eroded faster in DRAM (Macher, Mowery, and Hodges 1999, p. 254). To this extent one might have expected a more rapid exit by the advanced firms and more generous terms for technology transfer. Was the DRAM sector handicapped by a relative lack of spin-off activity? While it is true that the state did not embark on any large-scale spin-offs of firms after 1995, most of the DRAM entrants continued to maintain R&D relationships with ERSO, and the fate of the last state spin-off, Vanguard, a DRAM firm (see below), does not suggest that the spin-off of more firms would have been the answer. Indeed, one of the reasons for the state’s retreat from largescale projects was that both industry executives and technology policy experts in Taiwan argued that the government’s role as R&D facilitator and infrastructure-builder was no longer needed on such a large scale, given the maturation of the industry in Taiwan. Furthermore, Taiwanese conglomerates, such as the Formosa Plastics Group and Walsin Lihwa, founded some of the new DRAM ventures.5 It had been precisely this type of firm that had been reluctant to invest in the earlier stages of the Taiwanese industry’s development, thus necessitating the state’s role as venture capitalist.

LEADING, FOLLOWING, OR COOKED GOOSE? 85

Given the similarity in the necessary background factors that supported both CMOS logic technology and DRAM technology development, what was the primary difference that accounted for the inability of the DRAM firms to advance the technology on their own? According to the leading firms in the DRAM business interviewed by the Industrial Performance Center (IPC) team, a DRAM firm typically needs to capture 15 to 20 percent of the world market to be able to generate enough profits to meet the capital costs necessary to maintain production capacity and to fund the R&D costs to develop the next generation of DRAM. The need for such a large scale or volume of manufacturing (here scale refers to a firm’s total production volume rather than to the scale of individual plants) derives from the fact that DRAM has such low per unit profit margins. All five of the leading DRAM producers interviewed by the IPC team (representing 70 percent of total DRAM production in 2000) agreed with the necessary scale requirements of 15 to 20 percent of the world market. Furthermore, several smaller players interviewed concurred with this assessment despite their own current small size of operations. Thus, to succeed in becoming a leader in the DRAM business as it is structured today, a new entrant would need copious capital to build the necessary production capacity, buy the current generation of DRAM technology, and fund R&D. In contrast, foundries need only meet the capital expenses to build the production facilities. While theoretically a firm with enough capital could invest the requisite resources to build plants, buy technology, and start its own R&D efforts simultaneously, expenditure of such a vast quantity of resources would be extremely imprudent risk management. Historically, the successful new entrants into DRAM did not succeed by marshaling all their investment at once. In fact, no new DRAM entry has ever built up to this enormous scale overnight. Instead, the previous new entrants, from Japan and later from Korea, relied on access to patient capital that was willing to forgo returns in the short term while the firms ramped up to the necessary scale of production to sustain plant capacity and the R&D necessary for the next generation of DRAM. The hope was that large profits would be made when the firms achieved large volume production of their own proprietary cutting-edge DRAMs. Both Korean and Japanese firms carried much higher debt-to-equity burdens than their Taiwanese counterparts during their respective pushes to become serious DRAM contenders. From 1980 to 1997, Korean manufacturing firms’ debt-to-equity ratio

86 CHAPTER 3

varied between 300 and 500 percent, except in 1989, when it dipped slightly below 300 percent. The Japanese firms maintained debt-toequity ratios between 350 and 250 percent until 1990 and were always at least 100 percent higher than U.S. ratios. Taiwanese debt-to-equity ratios were never higher than 175 percent and hovered around 100 percent after 1989, precisely the time Taiwan began to develop a significant IC industry (OECD 2001, p. 159). The systemic reasons for Taiwanese “impatient capital” will be addressed in a later section. Japanese and Korean capital spending on the IC industry correlates with the timing of their increased share of the world semiconductor market. Significantly, as the Japanese debt-to-equity ratio declined in the 1990s, its capital spending on the IC industry and market share also declined.6 No Taiwanese firm has achieved anything close to the volume of production considered necessary to sustain DRAM innovation. In 2000, the four purely DRAM producers (Nanya, Powerchip, Promos, and Vanguard) together only had US$2.37 billion in revenues and none individually had more than US$700 million in revenues (TSIA 2001, p. 3). In contrast, Samsung Electronics, one of the DRAM leaders, had US$27.23 billion in revenues in 2000, of which US$10.35 billion, 38 percent of revenue, was generated by the DRAM-heavy IC division (Samsung Electronics 2000). With the constraints of small volume, the Taiwanese DRAM manufacturers have never been able to generate sufficient profits to fund new fabs and technological development simultaneously. They have therefore had to depend on repeated infusions of technology from foreign firms for each successive generation of DRAM. These constraints hold, despite the fact that, by some accounts, the Taiwanese are the low-cost manufacturers in this segment. With a small production volume, even a slight edge in costs will still not generate sufficient revenue to ramp up to the necessary volumes and conduct R&D.7 Of the seven firms that have entered the DRAM market in Taiwan (Quasel, Powerchip, Nanya, Acer, Vanguard, Mosel Vitelic [including its Promos joint venture (JV) with Infineon], and Winbond), three (Quasel, Acer Semiconductor, and Vanguard)8 have exited the market and the other four are dependent on foreign licensing to be able to compete at the current product generation. The typical sequence of events that has led Taiwanese DRAM firms to exit or to remain in a state of continued technological dependence is as follows. The firm enters the market by building a fab and licensing DRAM technology from a foreign firm. The firm begins producing DRAM, but finds that with licensing

LEADING, FOLLOWING, OR COOKED GOOSE? 87

fees and low margins there is simply not enough capital to ramp up to the volume needed to support both the continued capital for plant maintenance and the R&D to become an independent innovator. In some cases, they tire of the low profits with no sign of improvement and decide to close shop. This occurred in the case of Quasel and TSMC’s conversion of its Vanguard unit to foundry production. In other cases, the technologically dependent Taiwanese firm loses its technology partner when the latter decides to exit DRAM. In these cases, the firm has to decide either to abandon the DRAM model, as in the case of Acer selling its operations to TSMC for conversion to a foundry after its partner Texas Instruments exited the DRAM market, or to find a new technology source, as in the cases of Nanya and Winbond. In the latter cases, even with a new technology partner, the old problem remains of an inability to afford the costs of both ramping up volume and paying licensing fees. Nanya has lost its technology source twice, first replacing Oki with IBM and then IBM with Infineon when Oki and IBM decided to discontinue their DRAM technology development. Winbond lost its technology source when Toshiba decided to exit DRAM, and turned to Infineon. Powerchip seemed likely to lose its technology source when Mitsubishi appeared to be on the verge of exiting the DRAM market. However, Mitsubishi has recently agreed to give Powerchip the next generation of technology, and has entered into an alliance with Elpida, which in turn is an alliance between the two largest remaining Japanese DRAM firms, Hitachi and NEC. Promos has just reached an agreement for the transfer of the next generation of technology from Elpida after Promos’s falling out with previous technology source and Promos shareholder Infineon. Out of the seven Taiwanese DRAM firms and the nine different foreign technology transfer agreements, not one Taiwanese firm has been able to create a new product generation on its own, rather than paying for technology from abroad (see Table 3.2). The only way to solve the precarious dependence on foreign technology sources other than exiting the market appears to be for the Taiwanese firm to become a subsidiary of a large overseas competitor. In this way, the firm is at least guaranteed its source of technology although the firm essentially has given up on becoming a source of innovation in its own right. While the Infineon-Mosel Vitelic JV, Promos, appears to have come to an end due to disputes between Infineon and Mosel Vitelic over managerial control, other firms are employing the JV route to resolve their technological dependence. Nanya and Infineon have agreed to form

88 CHAPTER 3

Table 3.2 DRAM Technology Sources Firm Nanya Powerchip TI-Acer Vanguard Promos Winbond

DRAM Technology Source (in chronological order) Oki, IBM, Infineon Mitsubishi, Elpida Texas Instruments ITRI, Mitsubishi Infineon, Elpida Toshiba, Infineon

a fifty/fifty JV for a new 300mm fab in Taiwan using Infineon’s technology. The proposed 300-mm venture between Mitsubishi, its Japanese partners, and Powerchip appears to place Powerchip on a similar trajectory. The cost of building fabs has increased over time. Fabs cost approximately US$20 million in the early 1970s when the standard process technology linewidth was 3.0 microns. The cost went up to US$100 million in the early 1980s, when the linewidth was about 1.0 micron. In the early 1990s, fabs cost about US$300 million at 0.7 micron and soared to over US$1.2 billion at 0.35 micron in the late 1990s. Fabs are projected to cost US$12 billion in the later years of the present decade (Macher, Mowery, and Hodges 1999, p. 152). The increasing cost of fabs suggests the possibility that the fab cost was the determining factor in the success of the foundry segment and the failure of DRAM. One could argue that as foundries started earlier, they enjoyed a lower cost of entry due to the cheaper price of fabs. However, most of the growth of the Taiwanese fabrication industry occurred during the latter half of the 1990s (the fabrication revenues more than tripled in U.S. dollar terms from US$4.57 billion to US$14.9 billion during this period) when eightinch fabs became the norm. By 2000, most of the fabs (twenty-two out of thirty-seven) were of roughly the same generation, eight-inch wafer fabs. In capacity terms, these eight-inch fabs contributed 72 percent of total production in 2000.9 The entry of the Taiwanese DRAM firms occurred precisely in this second half of the 1990s, when most of Taiwan’s fabs and most of Taiwan’s total IC capacity were built by foundry and DRAM firms alike. Thus, foundry firms had to contend with the same range of fab costs that DRAM firms did for the vast majority of their capacity. Could Taiwan’s lack of success in the DRAM field simply be attributable to the bad luck of entering the DRAM market at a bad time? While the worldwide DRAM market did experience mostly negative

LEADING, FOLLOWING, OR COOKED GOOSE? 89

revenue growth from the beginning of 1996 through the beginning of 1999,10 Macher, Mowry, and Hodges observe that DRAM had already become a low-margin commodity item when the Koreans entered the market in a significant way in the early 1990s (1999, pp. 253, 276). Thus, the Koreans, if perhaps not the Japanese at an earlier time, faced a commoditized DRAM market similar to the one faced by the Taiwanese. Among the Korean firms, only Samsung made much progress in DRAM in the 1980s due to its early focus on slowly building DRAM competencies. Hyundai did not enter into commercial production of an own-designed product until 1991. The Taiwanese firm Quasel offers an apt comparison to the Korean firms, as it entered the market in the 1980s, more or less at the same time as the Koreans. Quasel entered the market in 1984 and closed down after suffering two years of losses. In contrast, Hyundai started its fab operations in 1984 and suffered ten years of losses before growing into one of the biggest DRAM producers (Mathews and Cho 2000, pp. 130–31). The commoditization of DRAM since the early 1990s and the relative success of Korean entrants equipped with very patient capital suggest that the recent downturn in DRAM was not the determining factor in the inability of the Taiwanese firms to innovate in DRAM. To be fair to the Taiwanese firms that entered the DRAM market, there is no clear alternative to the current DRAM business model. In DRAM, the design and the fabrication processes are still relatively tightly linked.11 On top of this, with such low margins, this product area is not a very desirable one in which to be a fabless design house. The Follower Goose: Taiwan’s New AMLCD Firms Since 1997, a number of Taiwanese firms have begun to enter the AMLCD industry. At the time of this writing, there are six firms with fabrication facilities producing AMLCDs, and another firm with the technology to do so. All of these firms have technology transfer relationships with foreign firms, primarily Japanese ones. Three of these firms were computer manufacturers and another was a CRT manufacturer, so they had a direct stake in the new display technologies, either as downstream users of the displays or as manufacturers of competing display technologies. The other firms had no direct connection to the display industry. All but one of these firms had been concentrating on producing large-screen panels (thirteen- to eighteen-inch screens) for notebook computers and desktop PC monitors. The motivations behind the technology transfer to

90 CHAPTER 3

Table 3.3 Technology Sources for Taiwan’s AMLCD Firms Taiwanese firm AU Optronics (formed by the merger of Acer Display and Unipac)

Major foreign technology source IBM (Acer Display), Matsushita (Unipac)

Chi Mei Optoelectronics

Fujitsu

Quanta Display

Sharp

Hannstar Display

Toshiba

Chunghwa Picture Tubes

ADI (Mitsubishi JV)

Toppoly

ITRI

Prime View International

No direct transfer, but recruited personnel from David Sarnoff Laboratories of the United States

Source: IPC Interviews.

Taiwanese firms were similar to the previous cases.12 However, the AMLCD case appears to fall between the two other cases in the following sense. While the Taiwanese firms, by their own account, are not innovators, they appear safely ensconced in the role of followers, without being dependent on continual technology transfers from abroad (see Table 3.3). The AMLCD industry has been able to take advantage of the human capital built up for the electronics industry in Taiwan, and ERSO has run training and small R&D projects. However, as this industry developed in the late 1990s, there seemed to be little need for the state to set up firms directly as they had in the past. Established firms, including Walsin Lihwa, Chunghwa Picture Tubes, Acer, Quanta, UMC, and Chi Mei were willing to enter this sector as they found technology sources and saw large market opportunities. Also, returnees did not play such a critical role, as the United States did not have a large developed AMLCD industry, though there have been some exceptions, such as returnees from David Sarnoff Laboratories. The firms have not confronted the need to invest in continuous product R&D because, unlike many IC products, there is not a rapid generational turnover requiring technological innovation. For the AMLCD industry, scaling up from thirteen-inch to eighteen-inch panels does not require technological development efforts on the scale needed to develop

LEADING, FOLLOWING, OR COOKED GOOSE? 91

the next generation of DRAM, let alone the next generation of microprocessors. As one manager from one of the new Taiwanese entrants put it, “We are a firm that has made money from mass producing commodity products on low margins. These displays are no different. We will make money from producing a lot of the low-margin AMLCD panels.”13 The Taiwanese firms have rapidly expanded their share of the world market from less than 10 percent in 1999 to a projected 30 percent in 2003.14 In contrast, the Japanese have reduced investment in production capacity in order to concentrate their resources on developing high-value displays for new markets, such as the next generations of PDAs and cell phones and automobile displays. The Japanese strategy has entailed avoiding head-to-head competition with the Koreans and, more recently, the Taiwanese, in the area of large panels for computer displays. The Koreans have essentially taken their traditional approach of heavy capital investment and have conducted some R&D on higher-value-added displays, but as of now remain in the same computer-oriented markets as the Taiwanese.15 Thus far, the AMLCD industry appears to be one that does not require continuous innovation for firms to continue to generate some profit margin. The generational turnover has not been very rapid for the AMLCD panels, and increasing panel size, one of the important characteristics for competing in the notebook and desktop display markets, is relatively easy to do. Thus, the Taiwanese firms can follow the model of passive acceptance of foreign technology without worrying about needing regular innovation infusions from advanced country partners to stay afloat. The strategy taken in the AMLCD sector fits the classic model of Taiwan (and Korea) receiving technologies that Japan abandons as it creates newer products (although in this case the Koreans pushed their way into the large-screen AMLCD market before the Japanese entirely retreated to higher-value-added displays). The Product Characteristics of the Three Cases In the end, what separates the tremendous success of Taiwan’s CMOS logic pure-play foundries from the bleak outlook for Taiwan’s DRAM producers is the match or mismatch between each segment’s product characteristics and the strengths of Taiwan’s economy. There is a match between the CMOS logic pure-play foundry’s product characteristics and Taiwan’s strengths, and there is a mismatch between the DRAM

92 CHAPTER 3

segment’s characteristics and those strengths. CMOS logic pure-play firms have met the criteria of granularity and no need for copious streams of patient capital, whereas DRAM firms have not met either of these criteria. Both segments have met the criteria of being high-volume and manufacturing-based segments. If the Taiwanese had stuck with the IDM model when trying to develop CMOS logic, they would probably have faced many of the problems faced in the DRAM segment in terms of being constrained by limited resources from simultaneously developing the requisite process and design technologies as well as the necessary scale of production. Unfortunately for Taiwanese DRAM producers, there does not seem to be a good alternative model for DRAM production at this juncture. Indeed, the lesson is that the Taiwanese should not go into any segment in which they cannot leverage all of their strengths. The case of AMLCDs more closely fits the classic formulations of Vernon and Akamatsu, in which the receivers of technology can benefit simply by receiving the cast-off technology of the global technology leaders. Granted, the Koreans used their patient capital to push into both the DRAM and AMLCD sectors, forcing the Japanese and others to unload these technologies in a manner that does not resemble the passive acceptance of technology they discussed, but the general idea from Vernon and Akamatsu that advanced countries shed lower-value-added technologies to the follower countries does fit the AMLCD case. The Taiwanese have not innovated in this sector, but unlike in the DRAM sector, they do not face the pressure of a rapid generational turnover of products that would make their current production and product technology quickly obsolete. The jury is still out on the prospects for future innovation in the AMLCD segment in Taiwan because it is unclear if there are the necessary technological possibilities for granularity. One indication that there is some progress toward granularity is the move by some Taiwanese firms to concentrate on one small part of the whole AMLCD process, such as color filter production. Conclusion: The Connection Between Product Characteristics and Taiwan’s Underlying Economic Change Taiwan’s need for granular production and its inability to succeed in segments needing patient capital are two sides of the same coin. Taiwanese firms need to innovate in a manner that does not consume large

LEADING, FOLLOWING, OR COOKED GOOSE? 93

quantities of capital over long periods of time with substandard returns in hopes of long-term gains because Taiwan’s financial system has not been set up for this kind of investment. Debt-to-equity ratios for Taiwanese manufacturing firms have been at very low levels since the early 1970s, especially when compared with Korean firms through 1997, with the differential between the two being less than 100 percent in only one year, 1973 (Scitovsky 1986, p. 165; OECD 2001, p. 159). The Taiwanese debt-to-equity ratios are similar to and often lower than those of the United States (Fields 1995, p. 108), an early industrializer, in apparent contradiction to Gerschenkron’s (1962) arguments about late developers marshaling patient capital to develop. Taiwan developed so late relative to the early industrializing countries that the term late developer has been applied to Taiwan and the other East Asian newly industrialized countries (NICs). Despite this history, in a radical departure from other late developers in Northeast Asia, such as Korea and Japan, Taiwan eschewed the use of high debt-to-equity ratios to develop (OECD 2001). Japan and Korea used patient capital to scale up to a level at which further R&D needed for innovation could be supported, the classic Schumpeterian approach. Why and how was the financial system set up to keep debt-to-equity ratios at levels far lower than in the other late developers in Northeast Asia? As has been documented extensively in the literature on the political economy of Taiwan, the reasons for this system originated with the Kuomintang or Nationalist Party (KMT) regime’s minority status, fear of the influence of big business, and an historic fear of inflation. These characteristics caused the regime to discourage a strategy like Korea’s of building up big private businesses, especially ones controlled by the non-Mainlander majority, through a supply of potentially inflationary cheap credit (Cheng 1990, 1993, 2001, Gold 1986, Fields 1995, Park 2000). The KMT state did not discourage private enterprise, but the state did view large private enterprises as a threat and the financial system was the preferred method to prevent the concentration of capital in private hands (Cheng 1990, pp. 143, 155, 160; 1993, pp. 58–59, Gold 1986, pp. 70–71, Park 2000, p. 158). These historical experiences and beliefs led the KMT regime to craft financial policies emphasizing monetary stability and limited direct credit, with a slight bias toward larger enterprises. This slight bias, created by the government’s requirement that loans be backed by collateral, prevented many smaller firms from accessing the formal financial

94 CHAPTER 3

market (Fields 1995, p. 148). Institutionally, the emphasis on monetary stability was expressed by the independence of the Central Bank of China from the industrial policymaking bureaucracy, if not from the top leadership within the KMT state, and by the authority the bank had over commercial rates and economic policy in general (Fields 1995, pp. 142–43, Wade 1990, pp. 208–9). The role of subsidized export loans as part of the total loan package was smaller than in Korea. Furthermore, the subsidies themselves were smaller, as measured by the difference between curb rate and export loan rate in the 1960s and 1970s, than in Korea, and the export loans were not targeted at specific firms (Fields 1995, p. 143). Indeed, in Taiwan, the loans were always on terms of positive real interest rates, even for firms with connections to the state (Ibid.). Thus, the end result was that in Taiwan, small and medium-sized enterprises (SMEs) predominated and capital was not made more patient through massive state support to stretch the period of time before firms expected adequate returns. With liberalization, banks and the state still did not allocate lots of credit patiently as there were not any Taiwanese corporations too big to allow them to fail and requiring bailout, nor so big that they rode roughshod over the government to capture control of the banking sector (Chu 1994). Indeed, the liberalization of the late 1980s was followed by a slight decline in the debt-to-equity ratio of Taiwanese manufacturing firms (OECD 2001, p. 159). In contrast, the Korean debtto-equity ratios did not decline until the aftermath of the 1997 financial crisis. The financial liberalization of the 1980s did not change much precisely because these huge Korean chaebol were often too large to fail without serious repercussions the state did not want to confront (Woo 1991, Fields 1995, Jones 2002). Consequently, the Korean chaebol were able to utilize the patient capital to become serious competitors in the DRAM market. The Koreans had achieved this success at a tremendous financial cost, estimated at US$4 billion for all of the 1980s (Mathews and Cho 2000, p. 133). While the Taiwanese government helped to spin off companies from ERSO through bank loans, this support was a small fraction of what the Koreans were able to muster. The Taiwanese spent approximately US$200 million during the 1980s.16 In the 1990s, the Taiwanese government conducted one last major spin-off, Vanguard, and discontinued further major projects due to heavy criticism from emerging private industry firms (Fuller 2002, pp. 22–23). Thus, Taiwanese government support cannot be construed as making up for the lack of patient capital needed

LEADING, FOLLOWING, OR COOKED GOOSE? 95

to extend the time horizons for loans before lenders demanded adequate returns. As late as 1997, the scale of Taiwan’s IC industry was not large. The total sales revenue for all eleven of the Taiwanese IC manufacturers combined was not quite as large as that of the world’s fourteenth-largest IC producer, Hyundai (ITIS 1998, pp. III–1, VIII–1). Even in 2000, the total fabrication revenue of Taiwan’s sixteen fabrication firms, US$14.9 billion (TSIA 2001, p. 1), was smaller than the revenue of Samsung Electronics alone and only 43 percent larger than Samsung’s IC revenue alone (Samsung Electronics 2000). Taiwanese firms may no longer be the SME midgets they were (Fuller 2002, p. 23), but they still lack patient capital, and large technology firms are still few in number.17 Without the option of innovating by leveraging relatively patient capital, the Taiwanese have had to granularize production to create narrow segments in which they can use their limited funds and narrow focus to reap the maximum possible rewards. Thus, the inability to leverage patient capital in Taiwan has essentially dictated that Taiwanese firms that wish to be innovators seek out those segments that do not have patient capital requirements. The firms that have disregarded this constraint imposed by Taiwan’s financial system and invested in the DRAM segment have never found the patient capital to build the necessary volume to support R&D for the next generation of products as well as continued capital expenditure. Given the reality of the close link between design and production in the DRAM field, they have also been unable to granularize production in this segment to be able to avoid the need for patient capital. Consequently, they have never been able to become innovators, despite their reputation as efficient producers. Instead, they are stuck in a segment in which they spend lots of money on production facilities in return for low profit margins made even lower by the cost of licensing the required next generation of technology from abroad. The second necessary condition for success is high-volume production. We arrive at this conclusion despite the fact that all three sectors in this case study are high-volume businesses. This characteristic is not really intuitive from our case studies but derives from a general survey of Taiwan’s postwar industrialization. Despite having a SME-heavy industrial structure, the Taiwanese have often pursued niche products with success, but have not pursued craft products. Taiwan’s economy does not seem to foster powerful, independent associations of producers with a broad scope in which to govern their own affairs. Neither does Taiwan

96 CHAPTER 3

provide the critical formal and informal training institutions that have characterized craft production in Germany and Italy. Historical examinations of the rise of these craft production centers in Germany and Italy suggest that the necessary institutions developed over long periods of time and are probably more the result of the accidents of history than of particularly prescient planning (Herrigel 1996, Piore and Sabel 1984). Thus, Taiwan is unlikely to develop these types of supporting institutions and should shy away from craft production. Finally, the product characteristic that is probably the least salient for determining successful innovation in Taiwan is manufacturing-based outputs. Again, all three sectors that we have examined here share a manufacturing basis, and the evidence that this is a necessary characteristic comes from Taiwan’s historical industrialization process. However, even though institutional constraints might dictate that high-volume production is the most suitable path for Taiwanese firms, there is not such a strong justification for focusing solely on manufacturing. Nevertheless, there is some justification in terms of educational focus and what we will call “strategic heuristics.” Taiwan’s education system places a premium on engineering, and engineering has a hands-on manufacturing orientation. What we mean by strategic heuristics is that Taiwanese businessmen and engineers have spent so much time involved in manufacturing that their historically accumulated inclination, reinforced by continual successes, and their deep understanding of manufacturing lead them to continue to pursue this type of business. Their past deep experience with manufacturing also gives them an advantage in pursuing manufacturing as opposed to pure design or service segments, so this strategic heuristic is more than a stubborn inclination to stick to established practice. This characteristic is a much weaker constraint in determining success or failure in Taiwan and will likely weaken more over time. The education system is undergoing reforms that may lead to a slightly different emphasis, and the strategic heuristic may weaken as more Taiwanese firms figure out how to compete in design and service segments. Indeed, there have been some substantial successes in certain parts of the IC design house segment, such as the design of chipsets, showing that some segments that are not directly involved in manufacturing can succeed.

4 A Tale of Two Sectors Diverging Paths in Taiwan’s Automotive Industry Edward Cunningham, Teresa Lynch, and Eric Thun

“Regionally, all eyes are now on the Middle Kingdom, and there is a general feeling that nowhere else in Asia really matters any more. South Korea is maturing, Japan is stagnant and India is just too much bother, while the auto industries in almost every other Asian country only exist thanks to tariff protection. When the barriers come down, so will the fortunes of car- and parts-makers in Thailand, Taiwan and elsewhere.”1

Late Development and Late Choices Every country is concerned about its place in the international division of labor, regardless of its stage of development. States that are late on the development path seek to develop industries whose impact will be both broad and long term. They know inappropriate policies will exclude them from the dynamic sectors that power the global economy, and are concerned that domestic firms might be “cut out” of the global production network. States that developed early, on the other hand, are concerned that their leading sectors will not adjust rapidly enough to the changes in the world economy. They risk being “hollowed out” as leading firms move operations to countries where labor is less expensive and less regulated. Taiwan is unusual in that it confronts the challenges of both the developed and the developing world. Until quite recently it was a model of “late” development but now confronts the questions of adjustment that are so familiar to the advanced capitalist countries. It fears both being “cut out” and “hollowed out” simultaneously. Taiwan, by all accounts, is a star pupil in the class of late developers. Average Taiwanese real per capita income rose from 10 percent of the 97

98 CHAPTER 4

American average in the early 1950s to approximately 50 percent by the early 1990s. The value of total exports and imports as a percentage of GDP increased from 23 percent in 1952 to approximately 81 percent in 1999, with much of the later growth resulting from integration of the global electronics industry.2 By 2002 Taiwan had become the world’s fourth-largest producer of PC-related hardware, with annual sales of US$17.4 billion.3 According to some estimates, Taiwan supplies over half of the world’s many PC-related products.4 But success creates its own pressures. The steady rise of manufacturing costs has forced firms in many of Taiwan’s “traditional” sectors to relocate manufacturing facilities offshore. An economy that, only a few decades before, benefited greatly from the outward investment of higher-cost developed economies facing the challenges of adjustment, now suddenly finds itself in the same situation. It is often argued that late-developing states do not face the same challenges as their predecessors in the initial stages of industrial development—is this also true as the economy matures and problems of adjustment become increasingly important? When confronted with declining sectors, does a late-developing state have the same choices as an advanced industrialized state, or is it some way constrained by its late development? Most studies of late development focus on the first stage of the industrial development process, when the primary challenge is developing a competitive industrial base. As Robert and Jean Gilpin have noted, “[i]n the language of contemporary economics, every state, rightly or wrongly, wants to be as close as possible to the innovative end of ‘the product cycle’ where, it is believed, the highest ‘value added’ is located.”5 The critical questions in this strain of the literature focus on the most effective means of promoting the development of dynamic economic sectors, and there is little discussion of what to do with those industries that are past their prime. Although industrial decline is of little concern to most policymakers during the early stages of industrial growth, it inevitably shadows success. As an economy develops, productive resources (labor, capital, land, and technology) must either be transferred from declining sectors to emerging sectors, or new strategies must be crafted to more effectively utilize resources that remain in traditional sectors. Bruce Cumings describes the process as a product life cycle: as a technology and manufacturing process is mastered in a given country, competition intensifies and firms become increasingly focused on reducing their costs. Eventually, there is a point at which it pays to allow others to make a

A TALE OF TWO SECTORS

99

given product, or at least provide the requisite labor.6 According to such a framework, leading firms relocate manufacturing activities to locations with lower input costs, and domestic firms in the host economies begin to compete directly with the leading firms from abroad.7 Countries serving as investment “receptacles” are thus expected to come to resemble the industrial structure of the more advanced economy, much like the “geese” flying ahead of them in the presumed formation. Taiwan, in its relationship to Japan, is often pointed to as a prime example of this process, and the expectation is that Taiwan in turn should be able to move mature industries offshore in the same manner as Japan. The product-cycle theory captures a key dynamic in the maturation of industry through its focus on the outflow of investment from a mature economy, but it also obscures many of the issues facing mature latedeveloping countries. Ravenhill and Bernard, for instance, argue that neither Taiwan nor Korea developed replicas of Japanese production structures, and not simply because of the importance of local context (e.g., politics, history, or institutions). First, it is extremely difficult to develop single commodities in isolation. “Rather than an ‘ahistoric’ flow of a single commodity,” they argue, “contemporary production needs to be seen in terms of interrelated complexes of industrial activity involving networks of firms, and continuous innovation of a key range of inputs in a multitude of related industries.”8 The backward linkages in a late-developing country are rarely as rich as in the economies of their predecessors. Second, the forces of innovation and higher-value-added activities are seldom transferred to the offshore manufacturing site. In other words, it is not complete industries being transferred from one “goose” to another, but the piece of an industry that is mobile—the laborintensive segment of final assembly—and this partial diffusion of technology creates an intraregional hierarchy of production, with the value-added depth of the relocated industrial production growing weaker at each stage.9 The implication of this logic, if true, is important: weaker control of the value-added parts of a production chain means that little will be left behind as an economy matures and labor-intensive activities move offshore. A commodity chain perspective offers similar reasons for concern. In a global economy, “the network of labor and production processes whose end result is a finished commodity” is spread across nations, and different parts of the chain—supply of raw materials, product design, production, or marketing and sales—give different opportunities for profit and

100 CHAPTER 4

control.10 In a producer-driven chain, large, integrated, capital-intensive enterprises operating at high volumes play the key role in controlling forward and backward linkages. In a buyer-driven chain, large retailers and brand-name marketers use their control over final markets to control a decentralized production network. As the regional hierarchy approach suggests, the Taiwanese economy might look much like that of a developed economy, but its place in the global commodity chains might create a very different set of opportunities as individual industrial sectors mature. This chapter builds upon these approaches and examines three different segments of the automotive sector—automotive assembly, original equipment manufacturing (OEM) parts, and aftermarket parts—to analyze how mature industries in Taiwan face the pressures of adjustment. Taiwanese OEM firms, on the one hand, confront great difficulties at home. In the early years of growth, firms in this segment of the auto sector benefited from strong government support and extensive technical linkages with Japanese firms. OEM firms remained reliant on foreign partners for advanced technology, however, and the small market constrained growth. In the past decade, rising labor costs and World Trade Organization (WTO) accession, combined with the difficulty of achieving economies of scale, have forced Taiwanese auto assembly firms to begin searching for new survival strategies. As in advanced industrial nations, moving offshore is one of the most popular strategies for promoting sustained growth. Taiwanese aftermarket parts firms, on the other hand, are increasingly thriving at home. They are migrating up the global value chain by acquiring logistical, quality-control-related, and design work, and they are consolidating their control over the parts of the value chain that they occupy. The largest of these firms are dominating the U.S. market for their particular product, and are establishing service offices and even final assembly sites in the United States. Many challenges remain. Competitive pricing patterns and an inability to standardize the industry through private actions have led to a distinct collective action dilemma that precludes cooperation among manufacturers and prevents the strengthening of competitive advantage for Taiwanese auto parts firms. Regardless, the outlook for Taiwan’s auto parts industry is considerably brighter than that for OEM assembly. The sectoral contrast evident in Taiwan’s automotive industry results from at least two factors. First, initial ties of dependence are difficult to sever. In particular, capacity for innovation and degree of technological

A TALE OF TWO SECTORS

101

independence are critical because they constitute the key determinants of Taiwanese firms’ comparative advantage over the low-cost competitors in the regions to which they are relocating. Perhaps more importantly, competency in these two areas will determine Taiwanese industry’s capacity for transforming what is left at home to include more valueadded activities. Innovative capacity and technological independence, however, depend in part on how the Taiwanese firms originally acquired technology (e.g., licensing, joint ventures), and from whom such technology was acquired. Thus, for Taiwanese auto assembly firms and OEM parts producers, long-standing dependence on Japanese technology has limited firms’ presence in regional and international markets by constraining product and investment strategies available to firms. The island’s growing aftermarket auto parts firms, on the other hand, are technologically “independent” because of the use of “reverse engineering,” the prevalence of generic parts, and the generally slower pace of technological change in the aftermarket sector. They are also able to transform their cost advantage into service, quality, and managerial advantages that enable them to increase control over larger segments of the value chain. In contrast, continued dependence on Japanese technology hinders the relocation efforts of Taiwanese auto assembly firms, while offshore relocation efforts replicate rather than compensate for these weaknesses. Second, relocation strategies reflect both the nature of the global production networks that organize a sector, and the place (both current and potential) of Taiwanese firms in them. In particular, opportunities are shaped by the degree to which production within a sector can be divided both geographically and between firms into distinct, or modular, activities, and how these activities are then coordinated. In other words, it is important to understand the linkages that hold the chain together. The “buyer-driven chains” that characterize the aftermarket sector are dominated by retailers, brand-name marketers, and trading companies that command a decentralized production network.11 Success in such an environment depends upon a firm’s ability to create brand equity, establish exclusive distribution channels, create a niche demand, maintain competitive service levels, and minimize the leverage of the firms higher in the value chain. Niche approaches offer prospects for growth and profitability in such an environment. In contrast, the “producer-driven chains” that characterize automotive assembly and OEM parts manufacture are dominated by highly integrated producers who control the

102 CHAPTER 4

forward and backward linkages in manufacturing. Success in this form of network depends on the ability of a firm to produce at high volume and efficiently integrate all functions of production. Niche approaches are rarely profitable in such an environment. The Taiwanese Assembly Sector The automotive sector in Taiwan is, by many indicators, reasonably healthy. In contrast to the rapid decline in output and employment in other traditional sectors, such as textiles and garments, the automotive sector has actually grown over the past decade. Between 1989 and 1999, total automotive industry output increased from US$4.7 to US$9.1 billion and employment from 66,000 to almost 76,000.12 Annual sales of commercial and passenger vehicles increased from 100,000 in the early 1980s to almost 600,000 in the mid-1990s.13 In 2002, total auto parts sales nearly equaled total vehicle sales.14 Although vehicle sales began to decline after the 1994 peak, the assembly and supply sectors continue to make significant contributions to national output and employment, accounting for 1.3 percent of total employment in 1998, compared to 1.0 percent in the United States during the same year.15 Despite these signs of continued strength, automakers in Taiwan have found it difficult to overcome two primary problems: a small, fragmented domestic market, and their continued dependence on foreign technology. The small size of the domestic market has been a critical constraint for Taiwanese auto firms. Economies of scale have always been important in the global auto industry, and in an era in which rapid technical change has increased the cost of research and development, the trend toward consolidation has only been accentuated. Large volumes allow a firm to spread the high engineering cost of new model development across many vehicles, and a decreasing number of firms achieve the volumes that are required. In 2001, only thirteen auto companies produced more than one million vehicles each, and these accounted for 87 percent of the world’s vehicle production.16 In Taiwan, however, the consolidation trend has been muted. In 2001, only 271,704 vehicles were produced in Taiwan, and this number was divided between ten assembly companies. The production levels of the leading firms are shown below in Figure 4.1. Although four firms accounted for 88 percent of total domestic

A TALE OF TWO SECTORS

103

Figure 4.1 Taiwan Auto Vehicle Production, 1989–2002 140,000

120,000

Auto Vehicles

100,000

CMC Kuozui Ford Lio Ho Yulon San Yang

80,000

60,000

40,000

20,000

1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year

Source: Taiwan Transportation Vehicle Manufacturers Association 2003.

production in 2003—China Motor Corporation (CMC) (25 percent), Kuozui (27 percent), Yulon (18 percent), and Ford Lio Ho (18 percent)— it has still proven difficult for firms to produce more than 100,000 vehicles in any given year.17 Rather than consolidating, assemblers continue to enter an already crowded market. In January 2002 Honda announced a split from its four-decade joint venture with Taiwanese auto assembler Sanyang, and purchased an existing manufacturing plant, financed with a US$110 million investment by Honda (USA).18 Sanyang, in turn, has affiliated with Hyundai to produce low-volume variations of the Elantra. Without a domestic market large enough to support the cost of ambitious design programs, Taiwanese firms have pursued two means of acquiring technology: foreign partnership and government support. The primary means has been a heavy reliance on technical relationships with foreign firms. Each of the top assemblers is owned in part by a large foreign firm: Mitsubishi owns 21 percent of CMC (and DaimlerChrysler owns 37 percent of Mitsubishi), Nissan owns 25 percent of Yulon (and Renault owns 37 percent of Nissan), Toyota and its Hino subsidiary own 57 percent of Kuozui, and Ford owns 70 percent of Ford Lio Ho.19 Japanese dominance is clear—even Ford Lio Ho relies on Mazda-designed

104 CHAPTER 4

vehicles (and Mazda is owned by Ford)—although the web of global ownership and technical agreements is often complex. Without a domestic market large enough to justify extensive R&D expenditures, Taiwanese assembly firms were forced to rely on Japanese technology in the early stages of development, and when the Taiwanese firms have made efforts to develop independent technical capabilities, the Japanese partners have provided little help.20 In the absence of foreign support for independent technical capability, the Taiwanese government has attempted to play a supportive role. The automotive sector has long been a target of governmental concern in Taiwan.21 Beginning in the 1950s, the government sought to use high tariff barriers and aggressive localization requirements to support local firms, and as late as the mid-1980s, tariffs on imported passenger cars were as high as 65 percent and a minimum of 70 percent local content was required. As Noble has noted, “[t]hese restrictions caused Japaneseaffiliated assemblers to import knock-down kits to accumulate assembly experience and a rudimentary supply network. Next, as local firms accumulated experience, the government attempted to increase local contents requirements.”22 But the government’s policy was far from consistent. At times, officials would utilize import liberalization to impose market discipline.23 From the perspective of the Kuomintang (KMT) Party officials who controlled the central government, this approach had the beneficial side effect of limiting the power of private, Taiwanese-owned firms. At other times, the Industrial Development Bureau (IDB), affiliated with the Ministry of Economic Affairs, focused on reducing dependence on Japanese technology through the promotion and protection of autos “as a strategic industry and potential source of exports.”24 Whether the result of conflicting organizational agendas, blunt policy mechanisms, the economic constraints of a small domestic market, an early reluctance to encourage private-sector development, or the classical failures of protectionism, government intervention in the auto sector resulted in mixed success at best. Noble characterizes auto industrial policy as being “uncharacteristically weak and vacillating.”25 Yongping Wu describes it as an exemplary case of industrial policy failure.26 The modest success of government policy in the early years did not deter ambitious efforts to foster technical independence in the 1980s and 1990s. After reducing tariff and local content rates, the IDB began to subsidize the development of independent design capabilities for auto engines,

A TALE OF TWO SECTORS

105

chassis, and body panels in 1985. The IDB organized open competitions for design proposals, and the assembly firm with the winning proposal gained the cooperation and support of the IDB during the design process. The IDB, together with the Industrial Technology Research Institute (ITRI), subsidized the design process, provided facilities and human resources for the design and testing stages, and then, through negotiations with the Ministry of Finance, offered commodity tax reductions during the manufacturing and retail stages.27 The most significant tax reduction was a 9 percent reduction for domestically designed auto bodies/panels, engines, and chassis. Although these savings proved to be a powerful financial incentive for Taiwanese assembly firms, the tax was abolished in 2002 to conform to WTO compliance regulations.28 Currently the IDB partners with small and medium-sized companies, although the WTO regulatory changes have effectively minimized their value to auto assembly firms. Other projects have been coordinated by the Mechanical Industrial Research Laboratory (MIRL), a lab within ITRI, which, together with the IDB, has assisted firms in the development of a domestic engine, key components, and system integration.29 In 1990, under the instructions of the Ministry of Economic Affairs (MoEA), MIRL initiated a feasibility study for the development of a common Taiwanese-designed and manufactured 1.2-liter auto engine. A consortium between several Taiwanese firms was formed for the project, foreign firms were invited to serve as consultants, and total investment in the project amounted to US$60 million over six years. Results have been decidedly mixed—the engine that resulted is used in only one model with an annual production level of 25,000 vehicles—but this has not deterred the planning of future projects. Current R&D initiatives include the development of a hybrid engine and a variety of electronic vehicle systems, for example, GPS capabilities, rear distance measurement devices, and heads-up displays for windshields.30 Although Taiwanese assemblers have little experience with fullfledged vehicle development—the one attempt to design a vehicle from scratch was by all accounts a technological and market failure—local factors have contributed to the development of some design capabilities. With the explicit or implicit approval of their partners, Taiwanese assemblers have carved a niche in redesigning vehicles developed by their foreign partners, and several have established formal “R&D” centers to perform this work. Most of this design work focuses on modifying the

106 CHAPTER 4

interiors and exteriors of their foreign partners’ vehicles, which stems from local market conditions, where demanding Taiwanese consumers and assemblers’ attempts to differentiate themselves have resulted in a strong emphasis on vehicle styling. In the post-WTO period, this relatively limited automotive design experience has been leveraged into a significant, if somewhat niche, regional role for the top Taiwanese assemblers, as foreign partners have adopted some Taiwanese-designed vehicles for production and sale at other sites, mostly in Asia. As a result, exports from Taiwanese assemblers and their key suppliers to foreign partners’ plants increased almost 216 percent between 1999 and 2003. (See Table 4.1.) Many of these exports consist of exterior and interior parts of partners’ vehicles that were redesigned in Taiwan. For example, CMC increased its exports of car sets from about 10,000 in 1999 to 118,000 in 2003, a result of Mitsubishi’s use of Taiwanese models at its plants in Southeast Asia and increased demand from CMC’s plant in Fujian, China.31 Ford’s exports, though modest, are also concentrated in exterior parts for models designed in Taiwan. Kuozui exports exterior parts for the Taiwanesedesigned Corolla and Camry to Toyota plants, mostly in Southeast Asia; and Yulon exports interior and exterior parts for the Cefiro, mostly to Southeast Asian nations, though it has also carved a role as an important supplier to Nissan’s first-tier suppliers in Asia.32 Thus, while Taiwanese assembly firms still lag in vehicle development and exports, they are becoming ever more skilled designers and suppliers of interior and exterior parts for regional vehicles. The Taiwanese OEM Supply Sector Supply firms within the OEM sector—300 of the approximately 2,000 Taiwanese auto parts firms are dedicated primarily to the OEM market— suffer from the same constraints as the assembly firms: the domestic market is small and they are heavily dependent on Japanese technology. Noble notes that 95 percent of technology agreements signed by Taiwanese parts markers are with Japanese firms and over 90 percent of joint ventures (JVs) are with Japanese firms.33 There is a core of strong OEM supply firms and, prompted by local content requirements, some assemblers have capitalized on these local capabilities. The Varica, produced by CMC, utilizes the Taiwan-produced engine, and the Freeca has a local content rate of over 90 percent.34 Overall, however, the wide

A TALE OF TWO SECTORS

107

Table 4.1 Exports of Taiwanese Assemblers to Their Partners’ Plants (in U.S.$ millions)

Assembler Yulon Motor China Motor Kuozui Motors Ford Lio Ho Prince Motors Taiwan Isuzu Total

Exports to partners 1999 (est.) 126 86 35 16 3 8 274

Exports to partners 2003 (est.) 263 534 176 27 8 5 1,013

Percent growth 1999–2003 109 520 409 66 232 –39 216

Source: Estimated by the authors using data from Quincy Liang, “Government Moves to Boost Automakers’ Parts Exports,” China Economic News Service, Taiwan Economic News, September 2003; and Quincy Liang, “Parts Exports by Taiwan Automakers Reach Record High,” China Economic News Service, Taiwan Economic News, November 1999.

gaps in supplier capabilities have limited the local content of most assembly firms. The overall local content of Taiwanese-produced vehicles is between 50 and 60 percent.35 Most of the imported parts are from Japan and most locally sourced parts are from Japanese JVs. In the mid1990s, for example, Kuozui (Toyota’s Taiwanese JV) purchased 80 percent of its components from Japanese JVs, and another 15 percent from Taiwanese firms with technology agreements with Japanese firms.36 The dominance of Japanese firms in Taiwan’s auto sector—and indeed, throughout Asia—has had the effect of reducing Taiwanese suppliers’ exposure to European and American assemblers and has contributed to Taiwan’s limited presence in OEM supply chains outside of Asia. Only a few Taiwanese firms have broken into even lower tiers of the supply chains of European and American assemblers. Although in most years, half of all parts and components produced in Taiwan are sold in foreign markets, historically exports have been concentrated in generic parts and components, rather than higher-quality replacement parts for specific models or OEM sales. Notwithstanding policies specifically aimed at increasing exports to Japan, Taiwan’s major export market in Asia, absolute levels remain low.37 While exports to Japan now account for about 5 percent of Taiwanese components exports—compared to only 2 to 3 percent in the mid-1980s—Taiwan’s share of total Japanese parts imports actually dropped slightly during the 1990s. In 2003, Taiwan’s imports from Japan

108 CHAPTER 4

were about eight times higher than their exports to Japan.38 In the late 1990s, several Taiwanese suppliers reported that the ongoing recession in Japan had led to more outsourcing and dual sourcing, and a greater willingness on the part of Japanese firms to utilize technological capabilities in lower-cost countries such as Taiwan.39 These changes, however, appear to have benefited low-cost East Asian countries other than Taiwan. Between 1998 and 2002, the proportion of Japanese parts imports from China and Korea doubled, while the proportion originating in Taiwan actually fell by 20 percent. (See Table 4.2.) Despite the constraints on Taiwan’s OEM suppliers, some recent trends may highlight a positive shift. Historically, these firms have not had strong footholds in regional markets, in part because of modest integration of sourcing channels across East Asia. Although there were early hopes that Taiwan would emerge as a regional center for parts production, only three of Taiwan’s fifteen largest export markets were in Asia as of 1999, and a large proportion of these exports were generic parts for the aftermarket. By 2003, however, the regional landscape was beginning to change substantially for Taiwanese OEM parts firms. With the growth in automotive production in China (including at Taiwanese-owned firms) and the partial liberalization that accompanied Taiwan’s entrance into the WTO, a nascent division of labor began to emerge within East Asia. In the first six months of 2003, auto parts imports and exports were about 40 percent higher than in the same period in 2001, with export growth driven by exports to low-wage countries in East Asia such as Vietnam, Thailand, and China. By the middle of 2003, seven of Taiwan’s largest auto parts export markets were in Asia, and Asian exports accounted for 27 percent of all Taiwanese automotive parts exports compared to just 20 percent in 2001. (See Table 4.3.) Relationships with Taiwanese assembly firms have driven shifts in auto parts trade patterns. In the first half of 2003, when Taiwan’s auto parts exports were US$254 million greater than in the same period in 2002, exports from Taiwanese assemblers to their Japanese partners’ plants in Asia grew by US$186 million, the equivalent of over two-thirds of total export growth for the year. Almost 80 percent of these exports were from two assemblers—Yulon and China Motors—and their lead suppliers. Much of this growth, as mentioned earlier, can be traced to the adoption of Taiwanese-designed vehicles at the production sites of Taiwan’s Japanese partners, and the demand for automotive parts from Taiwanese-owned assembly plants in China. This early evidence suggests

A TALE OF TWO SECTORS

109

Table 4.2 East Asian Countries’ Contributions to Japan’s Auto Parts Imports Contributions (%) 1998 2002 4.5 9.1 5.3 7.2 3.1 6.2 6.9 5.5 19.8 28.0

China Thailand Republic of Korea Taiwan Total Major East Asian

Growth (%) 1998–2002 102 36 100 –20 54.5

Source: Calculated by the authors from data prepared by the Massachusetts Institute for Social and Economic Research (MISER) using data from the Japanese Ministry of Finance. Table 4.3 Automotive Parts, Import and Export Growth (and Composition), 2001–3 (Unit: NT$ 1,000) Imports Asia/Australia North America Europe Exports Asia/Australia North America Europe

2001

2002

2003

Growth

19,881,063 (86%) 633,086 (3%) 2,598,706 (11%)

20,471,637 (82%) 1,771,871 (7%) 2,760,848 (11%)

27,412,612 (84%) 1,909,715 (6%) 3,325,137 (10%)

38%

7,675,494 (20%) 16,546,015 (43%) 4,380,622 (11%)

10,056,940 (23%) 19,895,175 (45%) 5,055,642 (11%)

14,485,426 (27%) 21,469,196 (40%) 6,596,463 (12%)

202% 28% 89% 30% 51%

Source: TTVMA, 2003.

that even the limited design capabilities of Taiwanese assembly firms are being successfully leveraged into increased exports of OEM parts as these models are increasingly produced offshore by Taiwanese assemblers or their Japanese partners. Aftermarket Parts Sector Of Taiwan’s approximately 2,000 auto parts firms, only about 300 serve primarily the OEM market, while the majority is dedicated to the production of aftermarket parts. In recent years, Taiwan’s leading aftermarket

110 CHAPTER 4

firms have expanded production capabilities in Taiwan and Mainland China, have created and dominate a fast-growing aspect of the aftermarket industry in the United States, and a few have begun to enter the U.S. OEM production network on a niche basis.40 These leading firms are in the process of managing two far-reaching transitions that further strengthen their competitive advantage and improve their function in the manufacturing network. The first fundamental transition is from dependence on a small and static domestic auto market to integration into foreign markets. This strategic shift is particularly evident in the collision parts industry, as the largest Taiwanese OEM and aftermarket parts manufacturers have concentrated their efforts in the creation and domination of U.S. non-OEM parts aftermarket. As a result, several leading Taiwanese firms now manufacture close to 90 percent of U.S. aftermarket collision parts and are beginning to supply accessory parts for the U.S. OEM market.41 These firms are establishing service offices and even final assembly sites in the United States. The second transition is a movement up the value chain through increasing levels of logistical, quality-control-related, and design work, as key parts distributors shed risk and reduce the number of their suppliers. Several leading firms of the collision repair sector have successfully leveraged their firm’s or founder’s regional OEM contract manufacturing expertise and diversified their revenue by launching auto parts aftermarket product lines overseas. In OEM auto parts production, contract manufacturers develop the capability to interpret design, create sample products, source inputs, maintain product and process quality, reliably deliver the buyer’s price, and guarantee on-time distribution channels. Managing such a range of activities not only serves to enhance the ability of local firms to manufacture internationally competitive consumer goods, but also generates substantial backward linkages to the domestic economy. As in the case of the OEM assembly firms, the past is critical to future success in auto parts production. The largest Taiwanese OEM auto parts suppliers have begun to leverage such knowledge of process and quality management into opportunities in the U.S. aftermarket, serving separate bases of OEM and aftermarket customers. Significantly, these firms have created, and therefore gained access to, an expanding developed country market despite their limited integration in regional OEM production chains. In the automotive aftermarket, system coordination and profit “derive not from scale, volume, and technological advances as in producer-driven

A TALE OF TWO SECTORS

111

chains, but rather from unique combinations of high-value research, design, sales, marketing, and financial services that allow the buyers and branded merchandisers to act as strategic brokers in linking overseas factories with evolving product niches in their main consumer markets.”42 The key obstacle for Taiwanese aftermarket auto parts firms, then, is not necessarily the technical dependence of late development— an expectation of the production-hierarchy approach. Rather, as predicted by the commodity-chain approach, many Taiwanese auto parts firms have become dependent on markets in the developed world, yet lack the power of brand recognition. The expectation is therefore that even though a local industry might have strong technical capabilities, it will be cut out of the production network as soon as firms from lessdeveloped locations begin to compete on the basis of price. Aftermarket parts are often manufactured outside the OEM network by independent firms that lack access to OEM design specifications. These firms replicate auto parts using coordinate measurement machines and CAD/CAM (computer-aided design/computer-aided manufacturing) technology. To manufacture a new product, the typical aftermarket firm purchases five samples of an auto part from an OEM supplier. A CAD/ CAM automated design system effectively “reverse-engineers” a digital average of the original samples and a master is created with a fiveaxis laser cutting machine. Detail work is finished by a five-phase machining center, and the finished product is measured and tested for tensile strength, composition homogeneity, and smoothness.43 The United States serves as the major export destination of aftermarket products, accounting for 35 percent of these exports in 2002, followed by Japan at 6 percent and Indonesia at 4 percent. Major export items included sheet metal body parts, tires, bumpers, rearview mirrors, head and tail lights, limited engine parts, ignition wiring sets, brakes, and jacks. Taiwan’s import of auto parts and components in 2002 amounted to approximately US$1.8 billion, with 76 percent of these items coming from Japan. Much of these imports were drivetrain parts, reflecting continued Taiwanese dependence on foreign partners for most engines and other critical parts and components.44 Buyer-Driven Chains The auto parts aftermarket differs from OEM assembly and parts production in four primary respects. First, the aftermarket is a buyer-driven

112 CHAPTER 4

chain. The majority of profits, information, and coordination are controlled by the branded retailers of auto parts, that is, the parts distributors. In the United States, the world’s largest auto aftermarket, insurance companies also wield considerable influence. Such an environment provides a challenging framework for designing growth strategies but also provides upgrading opportunities for leading Taiwanese auto parts firms that operate advanced supply chains. Second, the relationship between designer and supplier is severed in the aftermarket, as manufacturers seek to replicate, and not redesign, auto parts. Dependence on upgrading through Japanese technology is therefore less of a concern in this industry. Third, technological change is slower, as aftermarket parts are designed to replace parts in the existing vehicle population. Fourth, the aftermarket activities of Taiwanese auto parts firms have received minimal government assistance thus far, but may now benefit greatly from targeted programs that complement attempts to excel in buyer-driven industries, as proposed below. In a recent interview, a noted U.S. repair shop owner stated: “[t]he function that the parts distributor adds is obviously one of great importance, but distributors also tell the parts manufacturers that they ‘know the United States aftermarket,’ when in fact they only know their own goal and objective—profit. We have seen this in its strongest form in the automotive aftermarket sheet metal industry.”45 Such complaints by U.S. repairers and Taiwanese manufacturers are common, and reflect trends in an industry that has become increasingly modular. Advances in the use of standards, information technology, and supply chain management have increased the ability of distribution firms to coordinate dispersed production networks.46 However, such advances have also enabled supply firms to control an increasing share of value-added activities. The aftermarket supply chain is highly modular and most of the products are generic parts. As a result, there are few ties binding the manufacturer and the distributor. However, this has begun to change as manufacturers such as Gordon Auto Body Parts Co. have taken on increased responsibilities as product clearinghouses and centers of quality control. These ties increase the distributors’ cost of switching suppliers and allow suppliers to assume more value-added responsibilities in the supply chain. As electronic interfaces linking these activities increase in sophistication, the ability of a lead supplier to manage effectively such a network of subsuppliers also creates managerial barriers to entry.47 While a Taiwanese firm that competes through access to low-cost labor is

A TALE OF TWO SECTORS

113

pursuing a fragile source of comparative advantage, the firm that has the capability to manage an increasingly sophisticated supply chain is not. The clear objective must be to combine relocation with upgrading. The former is necessary to lower labor costs, while the latter is required in order to maintain a degree of comparative advantage over lower-cost competitors. Increasing standardization in the aftermarket and technological improvements are creating opportunities for Taiwanese firms to accomplish both. Consolidation trends in the U.S. parts distributor and independent repair sectors have intensified competition and led to an increased focus on minimizing repair time. However, demand in the auto parts aftermarket is rapidly rising, volatile, and difficult to predict. While weather patterns and the existing stock of vehicles on the road may provide useful data for ensuring sufficient long-term supply, repair shops seek to minimize repair time by maintaining appropriate short-term inventory. Technological advances have allowed improved tracking of such inventory. Bar codes, computer technologies, modern distribution centers, and the promulgation of standards across firms enable lean retailers to transform the relationships between firms within the aftermarket sector. While distributors seek to maintain arm’s-length relationships with parts suppliers in an effort to encourage interfirm price competition among them, new demands on the industry—including shorter repair times and price pressures—are necessitating relationships that are highly integrated by both information technologies and standards. These new demands also provide Taiwanese aftermarket firms with the opportunity to maintain a competitive advantage and upgrade their abilities. Taiwanese firms are increasingly realizing the potential to become the key force behind efficient commodity chains. While small, local firms in Southeast Asia or Mainland China may be able to undercut a Taiwanese firm on labor costs, they are unlikely to be able to make the investments in electronic data interchange that make rapid response possible. In short, being able to forecast demand effectively, process electronic orders from buyers, plan and track production, and manufacture auto parts in a flexible fashion are all skills that provide a far more enduring form of comparative advantage for Taiwanese firms than constantly scouring the globe for the lowest cost of labor. In addition, dependency on Japanese technology poses less of a concern for Taiwanese auto parts aftermarket manufacturers.48 While value chain integration may be taking place in areas such as product quality

114 CHAPTER 4

control, tracking, and delivery, aftermarket parts are effectively “reverseengineered” OEM parts. Despite the fact that the technological capability required to produce such parts is considerable, the speed of required technological change in such an industry is slow. As a result, while a pronounced dependency on Japanese technology has hindered the growth and competitive upgrading of Taiwan’s OEM suppliers, a similar lack of independent production technology has not proven overwhelming to aftermarket parts manufacturers. Indeed, as will be examined below in the discussion of firm-level strategies, many leading auto parts firms have developed their own niche technology for parts production and complement such strengths with investments in inventory management capabilities. Foreign Solutions The U.S. auto parts aftermarket is very large and growing. Estimates vary, but a 2003 total value of approximately US$26–30 billion is cited by most organizations, with the highest estimates at US$47 billion. The U.S. collision parts market reached roughly US$11–15 billion in 2003 (larger than Taiwan’s entire automotive sector), and, until the entrance of Taiwanese firms, was entirely controlled by OEM manufacturers.49 However, Taiwanese parts manufacturers have, over the past two decades, expanded the U.S. aftermarket and now produce approximately 80 to 90 percent of non-OEM aftermarket collision parts (a market estimated at US$2.5 billion in 2003).50 As shown in Figure 4.2, Taiwanese auto parts imported into the United States have doubled over the past decade, and the potential for market growth is considerable. A recent survey of repairers concluded that repair shops utilize aftermarket parts in about 14 percent of repair jobs.51 Moreover, market statistics indicate continued growth for the aftermarket auto parts sector. The total automotive aftermarket increased by 4.6 percent in 2002, while U.S. imports of automotive parts and accessories rose 10.2 percent, to $67.0 billion, the same year.52 Significantly, the average age of cars (nine years) has increased one full year since 1993, while the number of light vehicles that are no longer under the standard three-year warranty increased from 156 million in 1998 to 170 million in 2002.53 Over 80 percent of insurance appraisals for vehicles six to nine years of age specify aftermarket parts for replacement. The figure only drops to 54 percent for vehicles three to five years of age, and even the appraisals

A TALE OF TWO SECTORS

115

Figure 4.2 Total U.S. Imports of Taiwanese Auto Parts, 1989–2002 1600 1400

Amount (US$ million)

1200 1000 800 600 400 200 0 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 Year

Source: United States Census Bureau, 2003.

for two-year vehicles specify aftermarket parts 35 percent of the time.54 The percentage of repair shops that purchase aftermarket auto parts has also increased dramatically in recent years, rising from 47 percent in 1990 to 84 percent in 2003.55 Firm Strategies As in the case of auto assemblers, Taiwanese auto parts manufacturers have sought to diversify their consumer markets by replicating production activities and strengthening home production capabilities. In addition, these firms have continually improved their position in the global production network by creating new markets and expanding the scope of their economic activities. Firm-level examples effectively convey the nature of this dual transition, as these ventures into the U.S. auto aftermarket provide opportunities for firm growth, for the development of new customers, and for upgrading product lines at domestic plants. With 2003 sales expected to grow by 20 percent, Gordon Auto Body Parts Co., a leading Taiwanese auto parts aftermarket manufacturer, has grown rapidly, producing over 1.4 million auto parts annually.56 Established in the mid-1980s, the firm was founded by an individual who had

116 CHAPTER 4

previously operated a tool-and-die-making mold manufacturer that served OEM manufacturers. Of the 350 employees, over 10 percent are dedicated to R&D, with the majority of these individuals focused on mold development. An example of a successful niche strategy, this firm never entered the domestic aftermarket in Taiwan, appreciating the small scale of the domestic market, and chose to invest all its resources into the development of the U.S. aftermarket. Currently the firm exports close to 100 percent of manufactured products, with 90 percent shipped to the United States. In an effort to raise capital for production expansion in Taiwan, the founder opened a small percentage of the firm to public investment several years ago. While Gordon maintains relationships with over 100 customers, one customer accounts for over 25 percent of sales, again revealing the consolidation of the traditional U.S. distribution network. As the consolidation of distributors continues in the United States, the firm has sought to market itself as a “one-stop shop” for all Taiwanese parts, and in the words of the current vice president, to create a “value-added channel of distribution” for U.S. distributors.57 More specifically, Gordon now performs two services: it continues to produce replacement panels and serves as the clearinghouse for over fifty supply firms that produce complementary products, such as plastic moldings, lights, fans, radiators, condensers, and the like. These secondary suppliers ship their items to the firm’s central plant, where the complementary parts are inspected for quality and vehicle fit, then bundled with the firm’s own products and shipped to the United States. To accommodate rising demand, Gordon opened a second plant in Taiwan in 1999, which includes a sales department. Because of the production cost structure—labor accounts for only 6 percent of total costs— the firm argues that there is little economic logic in moving production capacity to Mainland China. Currently, 30 percent of the workers are foreign, attractive because of their willingness to work overtime for salaries that are approximately 70 percent of those paid to Taiwanese employees.58 The firm is considering establishing a marketing presence in China to research the viability of a diversified supplier base, however, establishing manufacturing on the Mainland “is much farther down the road.”59 Another illustration of a niche strategy and value chain migration that combines technological innovation and creative market entry in the aftermarket sector, Mobiletron has developed a versatile rearview

A TALE OF TWO SECTORS

117

mirror that serves as a multimedia platform incorporating a number of functions. A four-inch thin film transistor-liquid crystal display (TFTLCD) screen built into the mirror serves as both a high-end antiglare mirror and also as an “in-car information and monitoring center.” It can display images “transmitted by a charge-coupled device (CCD) camera mounted in the front or rear of the car (or both), a DVD or VCD player, an in-car inspection camera (to monitor children in the back), or a global positioning system (GPS) device.”60 The mirror is controlled by a hands-free unit, and the display can be expanded to include such features as a wider angle of view (through wide-angle cameras), the ability to save images captured by a digital video recorder unit (advertised as being quite useful for collecting evidence pertaining to accidents), tire-pressure display (through wireless sensors in the tires), and an electronic compass. The firm’s ultimate goal is “to perfect a mirror that can serve as a multimedia platform with diversified applications that replace all in-car entertainment, security, and navigation devices.”61 The firm has aggressively sought to enter the U.S. market, and recently reported receiving significant orders from the car rental conglomerate Avis. Several other significant orders are under negotiation. The firm’s management is currently considering the purchase of a U.S. auto parts distribution channel and is in talks with CMC to include the mirror in the OEM production of several vehicle models.62 Similarly, the Genera Corporation, the distribution arm for leading Taiwanese aftermarket headlamp manufacturer TYC, was created as an attempt to stimulate production and sales of automotive lamps through overseas market channels. Recently, Genera has actively promoted its brand through a widely successful partnership with Universal Studios. Linking the launch of an accessory headlamp product line with the release of the blockbuster movie 2 Fast 2 Furious, about Los Angeles street racing, which glamorized the “rice rocket” (a highly modified Asian import car), Genera included scenes and logos from the film in its packaging to attract younger consumers. According to Genera, signature headlamps were featured on the Mitsubishi cars used in the film. Since the establishment of several distribution and sales offices in the United States in 1991, Genera has diversified into aftermarket forward replacement lighting equipment, fan assemblies, mirrors, and more recently air conditioner condensers and radiator assemblies. Sales have increased from US$2 million in 1991 to over US$120 million in 2002,

118 CHAPTER 4

supported by nearly 750,000 square feet of national distribution centers located in California, Illinois, Texas, Georgia, and New Jersey.63 Significantly, Genera recently severed its dependence on Japanese technology. The technology required to produce the reflective surfaces in the interior of Genera’s headlamps is proprietary, costly, and controlled by four firms worldwide, according to a firm representative interviewed.64 Previously, in order to build a plastic reflector to specification for a new product, Genera would “line up with all the other Taiwanese producers” and request casts to be developed and built by the sole Japanese firm with access to such technology. The wait would often take months. In 2000, Genera created a joint venture with a Canadian firm, DBM Reflex.65 DBM provides design support for OEM suppliers such as Automotive Lighting, Visteon, Valeo, Textron, and Guide Corporation, and their products are found in vehicles manufactured by BMW, Ferrari, Ford, and Volkswagen.66 The new subsidiary, DBM Reflex of Taiwan, provides Genera with technical and design independence while minimizing product development time. Genera’s parent company, TYC, provided the financial resources and land in Taiwan, and DBM provided the personnel and technology. As a result, an increasing number of Taiwanese auto lighting firms have now chosen to produce through the DBM factory in Taiwan. Several Taiwanese firms interviewed are also entering the U.S. OEM parts supply chain. Macauto Industrial Co. has proven particularly successful, having signed U.S. contracts for innovations that fundamentally reengineer the OEM design of vehicles to allow more sunlight through the creation of two additional sunroofs. The firm has already established dominance in the Taiwanese domestic sunshade market, with a market share of over 90 percent, and manufactures high-quality sunshades with a patented rolling technology.67 The shift in U.S. OEM manufacturing created by this firm has led to several U.S. OEM contracts with van and truck manufacturers. Currently the firm has signed contracts for 200,000 units of the “Panorama Roof” for use in Chevrolet’s 2004 Silverado models. Nissan has also signed a contract for 350,000 units for its 2004 Quest model.68 In addition, in 2004 Macauto secured a six-year OEM contract for the Mercedes-Benz E-Class series. The firm continues to expand its OEM relationship with BMW, and opened a branch office in South Korea following the conclusion of a contract with Hyundai Motors.69 As indicated by the firm’s president in a recent interview, such market diversification has proven incredibly important for the firm’s growth, as these two initial orders alone are nearly double the annual OEM production levels

A TALE OF TWO SECTORS

119

for the entire Taiwanese auto industry. This evolution in manufacturing and market segmentation has allowed the firm to complement its operations in the aftermarket—an industry with little technological change or room for innovation—with a transition to an industry where experience in parts production can be leveraged to manufacture parts demanding technological innovation and for a larger customer base. Leading U.S. OEM assemblers and suppliers have begun to recognize the strengthening capabilities of Taiwanese manufacturers. Frank Deiss, DaimlerChrysler AG’s director of Global Procurement and Supply, recently remarked that Taiwanese parts meet the firm’s OEM standards and will be sourced for plants in Germany, China, the United States, and Canada for Mercedes-Benz, Chrysler, Jeep, Smart, and Fuso vehicles.70 Similarly, Delphi Technology Inc., one of the largest global suppliers of auto parts, led a delegation to Taiwan in early 2004 and selected fifteen manufacturers for procurement evaluations, including Macauto Industrial and Genera’s parent company.71 Outward Foreign Investment: A Solution to Domestic Constraints? Assemblers Given the structural impediments to wholesale domestic upgrading and the positive trend in regional trade patterns, foreign investment would seem like a natural solution to the problems facing Taiwanese automotive firms, solving problems on both the demand and supply sides. On the demand side, offshore investments have the potential to alleviate the problems created by the small and saturated market at home, low volumes, and falling tariffs. In Mainland China, the site of most foreign investment by Taiwanese automotive firms, the market grew by 58 percent in 2002 and is now approximately ten times larger than the Taiwanese market. By 2003, China had surpassed South Korea to become the fourth-largest vehicle producer in the world.72 Moreover, the Chinese market will continue to have relatively higher levels of protection than Taiwan for at least the next five years.73 Similarly, the 2003 Association of Southeast Asian Nations (ASEAN) Free Trade Area (AFTA) has introduced capped tariffs at 5 percent for in-region production and the market should continue to expand.74 On the supply side, investments in China and ASEAN member nations address problems associated with

120 CHAPTER 4

rising wages in Taiwan, high land costs, and a tight domestic supply of production and technical workers. A rush for the exits was partly forestalled by the need for foreign partner approval of relocation plans— just as new components had to be sent back to Japan for testing and approval, export and investment initiatives also required partner approvals75—but the lure of an easy solution to the problems of the domestic marketplace was very real. Among assemblers, two leading Taiwanese firms, CMC and Yulon, have made major offshore investments. One began investing outside Taiwan in the mid-1990s, when it purchased 5 percent of an assembly operation in southern China.76 In late 1999, this same assembler purchased a share of its partner’s operations in the Philippines and, soon after, assumed management control.77 A year later, the same firm purchased a 25 percent share of a second Chinese assembly operation, which produced 20,000 vehicles annually.78 The second assembler has followed a similarly aggressive offshore investment strategy. In the mid-1990s, the company signed a 50–50 JV agreement with a provincial government in southern China to establish assembly operations and proceeded to build a plant with potential capacity of 150,000 vehicles per year, on a site that can be expanded to accommodate production of 300,000 vehicles. In 1997, the company began small-scale production, and output levels have increased steadily.79 The driving motivation has been obvious: new markets, and in particular large markets that continue to benefit from considerable protection. In 1994, explained a manager at one firm, executives at his company decided that only through foreign investment would the firm be able to achieve the threshold volume of 300,000 vehicles per year that it believed was necessary to be a viable independent auto firm.80 If China had been an open market, he commented, his firm would not have been able to compete initially, but he hoped that continued efforts to hinder the enforcement of WTO compliance would allow his firm to expand on the Mainland. Because foreign ownership in Mainland assembly JVs is limited to a maximum of 50 percent, the Taiwanese firm searched for a local partner that would surrender 100 percent operational control in exchange for the investment capital. Ultimately, a coastal province in the south was chosen, in a region that previously had virtually no local auto industry. The local government was aware of the high profits that cities such as Shanghai were realizing through vehicle production, and it was eager to support the Taiwanese project.

A TALE OF TWO SECTORS

121

The relocation strategy utilized by the Taiwanese firm reproduced the domestic supply network in a new location. This was necessary both to conform to Chinese domestic content regulations and to lower costs, as importing a part increased the price by 40 percent (30 percent duty, 10 percent transportation).81 The firm “persuaded” its Taiwanese suppliers to co-locate with it, primarily by rendering future contracts in Taiwan contingent upon the suppliers investing in plants in the industrial park alongside the new China assembly plant—a common practice in the global auto industry. Headquarters believed that it was necessary to bring their own suppliers to China not only to assure quality, but also because it would allow them to launch production of other models more quickly and with less volatility. Managers at many of the supply firms made it clear that they were less than enthusiastic, but had little choice. Between 1997 and 1998, thirty Taiwanese component firms built wholly owned factories in an industrial park next to the assembly plant. By 2000, these firms were supplying the JV with 80 percent of its parts (measured by value), while the remaining 20 percent came from fiftysix Chinese firms, thirty of which serve as Shanghai Volkswagen suppliers in Shanghai. Production began in 1996 with a few hundred cars, and by 1999 production volumes had reached 5,000 vehicles per year. The objective is to achieve volumes of 200,000 vehicles per year by 2005. The operation has enjoyed considerable success, and production has increased from 2,737 units in 1997 to an estimated 61,000 in 2002.82 Between 2000 and 2002, average annual sales increased by 62 percent, and average annual revenues increased by 31 percent. Profits in 2002 were up 90 percent from the previous year. One general manager noted that the current rate of Mainland plant productivity was thirty vehicles per employee annually, compared to just one vehicle per employee in 1996. The firm now produces four models in Mainland China, and local content has increased from 50 percent in 1996 to 84 percent in 2002. In the case of one model, the local content increased from 40 percent in 1997 to a projected 93 percent in 2003.83 Although the firm produces eleven vehicle models in Taiwan, revenue from its Mainland China operations is projected to surpass that of its Taiwan operations by 2007. Suppliers For OEM suppliers in Taiwan, the limitations of a small domestic market, high factor prices (especially land and labor), and growing

122 CHAPTER 4

opportunities created by Taiwanese assembly investments lead to two forms of relocation. The first involves following Taiwanese assembly firms to the Mainland, but the beneficial impact of this approach is decidedly mixed. On the positive side, there is the possibility of stabilizing overall output by providing direct access to one of the world’s most promising automotive markets: these new investments create an opportunity for firms to maintain or increase overall production levels regardless of future conditions in the Taiwan market. They may also provide exposure to and possible entrée into European and American supply chains. The problem, however, is that these investments do not necessarily contribute to cost-cutting and upgrading objectives that are critical for survival at home. Despite labor costs on the Mainland that are one-tenth to one-third of those in Taiwan, low production volumes on the Mainland, high duties, and in some cases low productivity, mean that overall production costs are often higher than in Taiwan. While costs will decline as volumes increase and workers gain experience, other costly practices, such as importing basic raw materials, are likely to be required across the Mainland for some time. Perhaps most importantly, because these plants are dedicated to producing the same or similar parts as the home plants, they currently offer few opportunities to create divisions of labor between home and host sites, or otherwise contribute to the upgrading of product lines in Taiwan. In the worst-case scenario, as production costs in China decline, the new investments might start cannibalizing the markets of the home operations. The second type of OEM supplier investment is motivated by a nearly constant effort to upgrade the domestic operations of the mother firm. These investments often contribute to domestic restructuring by providing lower-cost sites from which to export and manufacture labor-intensive products, and in some cases, perform engineering and design tasks that are costly to perform in Taiwan, or for which domestic engineers and technicians are difficult to find. These plants also provide lower-cost export bases to third countries, such as Japan, thus extending the life or increasing the profitability of existing product lines. In contrast to assembly firms that simply duplicate their Taiwan operations in the offshore locations, smaller supply firms are also using relocation as a means of compensating for weaknesses in the home location. One company, for instance, opened a factory in Wuxi, a city just outside of Shanghai, primarily in order to access engineering talent. In Taiwan, it is not only difficult to find engineers willing to work in

A TALE OF TWO SECTORS

123

manufacturing—as one manager commented, “it is hard work and most would rather be in the information technology industry”—but they are far more expensive than on the Mainland. According to local sources, the quality of engineers in Wuxi is comparable to that of Taiwanese engineers, at a quarter of the total cost. Indeed, one half of the design work performed by the Wuxi plant is done for the home Taiwan plant. Relocation therefore not only creates new market opportunities, but also lowers costs for the home operations and allows its engineers to focus more exclusively on higher-end design work.84 This second form of investment creates strong advantages for the Taiwanese firm. The investments complement the operations at home, capitalize upon Taiwanese advantages on the Mainland (both linguistic and cultural, and their experience in establishing and running plants with low break-even points), and provide an opportunity for Taiwanese firms to gain exposure to major international assemblers with JV projects in major auto-producing regions like Shanghai.85 Although only a few wholly owned Taiwanese firms are part of the supplier network of SAIC—the largest Chinese producer of passenger vehicles and the JV partner of Volkswagen and General Motors—some Taiwanese firms have used JVs with foreign partners as an entrée into this and other supply networks on the Mainland. The connections with other foreign assemblers and global suppliers, of course, can also lead to additional export channels (i.e., to the home operations of the foreign firm) and markets. Many aftermarket parts firms have also established foreign sites, usually in Asia but also, as discussed earlier, in the United States. Investments in China and other East Asian sites are often motivated by the need to identify production environments that offer lower wages or land costs or more relaxed environmental standards. Taiwanese aftermarket firms report that the production of certain parts, including wheel hubs, wheel rims, and radiators, have almost fully moved to China.86 In addition, processes that cause the most environmental damage, for example, the burnishing and spraying of wheel rims, have also been moved to the Mainland. For lighting and other firms engaged in labor-intensive production, China provides the promise of lower-cost labor. Significantly, although many of these firms have invested in China, many continue to invest and expand in Taiwan as well. One aftermarket parts manufacturer noted that chasing low-cost labor is not without problems. He comments that, “[1]abor is cheaper in

124 CHAPTER 4

China—it’s about 10–15 percent of total costs in China, compared to 20 percent in Taiwan. . . . If a product is ‘Made in China,’ though, the United States buyers will reduce the prices, so the profits will be the same. This negates any savings that are gained from producing in China.”87 As a result, his firm chose not to invest in China. Still, he noted, the firm was negatively affected by the rise of aftermarket parts production on the Mainland. In 2000, the firm’s profit rate was about 10 percent, compared to “20 or 30 percent before the rise of low-cost producers in China . . . [however] today, because of China, even a 10 percent profit rate is not easy to achieve.”88 Thus, among Taiwan’s automotive firms, foreign investment has been used to meet a variety of objectives. For assemblers, China and other sites in East Asia provide the promise of new markets with old regulations, namely levels of market protection that characterized the environment for pre-WTO assemblers. For assemblers and OEM suppliers, new sites also provide exposure to and interactions with American and European assemblers and suppliers, thus remedying Taiwan’s dependence on relationships with Japanese partners and isolation from other major players in the global automotive industry. For OEM suppliers and aftermarket firms, China and ASEAN member nation sites also provide a production environment with lower wages, access to manual and engineering labor (especially in China), lower land costs, and environmental standards that are more relaxed than those in Taiwan. Not surprisingly, some supplier firms appear to be pursuing multiple objectives at foreign production sites. Data from a sample of Taiwanese automotive firms reveal that of the roughly one-third of firms that had engaged in some type of foreign investment by the end of the 1990s, about 40 percent had invested in multiple sites. Almost half of firms with plants in Fujian province, where Taiwanese foreign assembly is concentrated, had also invested in other sites, usually on the Mainland or in the ASEAN region.89 However, firms with a diversified foreign investment strategy seem to represent only a small portion of all Taiwanese suppliers. Of the sixty-four firms represented in the sample, only twenty-three had any foreign plants and, of these, only ten had two or more plants outside of Taiwan. It is clear that, despite increases in production capacity, especially evident in the aftermarket industry, production activities have remained in Taiwan. While foreign production plays a key role in the competitive strategies of some Taiwanese automotive firms—by the end of 2001 nearly 100 firms had invested in Mainland

A TALE OF TWO SECTORS

125

China alone—the future of a large number of suppliers is likely to depend on their continued ability to produce competitively at home.90 Future Strategies, Challenges, and the “China Effect” The pressures of globalization have created both significant challenges and opportunities for the Taiwanese auto assembly and auto parts sectors. The method by which Taiwanese assembly firms acquired their capabilities—licensing contracts and JVs with foreign firms—affects their ability to later reproduce such capabilities offshore in their drive for market expansion. While such a liability may prevent Taiwanese firms from competing directly with global assemblers in large markets, a strategy of complementarity in smaller markets provides a viable option. Alternatively, the consolidation of the global auto sector has increasingly allowed suppliers to provide logistical, quality-control, and design work to assemblers and distributors. Through partial relocation offshore, characterized by the creation of service and functional tasks to new sites, parts firms are able to diversify their customer base, upgrade global customer service, gain exposure to new distribution channels, integrate global standards into their production process, and expand activities into new sectors of the industry. Such transformation depends on the increasing technological independence and superiority of Taiwanese parts firms. This transformation has not resulted in the relocation of manufacturing capacity, especially as these new markets are characterized by high barriers to entry. Through replication offshore, both types of firms are able to diversify their customer base and reduce their dependence on an increasingly competitive domestic market. The functional dynamism of Taiwanese parts firms has proven a durable asset compared to the dependency of the assembly firms. Prospectively, the “low-volume, high-complexity” assembly model may prove to be appropriate to the smaller markets of Southeast Asia. Such expansion would be furthered by existing relationships with Japanese firms already investing in such markets. However, technological dependence will continue to retard growth. The successful evolution of parts production in Taiwan may be viewed on two levels. Regionally, success may depend on the extent of Taiwanese participation in the sourcing of established Japanese regional networks and the growing international networks on the Mainland. Globally, Taiwanese firms have succeeded in serving a rapidly expanding U.S. aftermarket, but have yet

126 CHAPTER 4

to fully transition into the considerable U.S. OEM market or enter the EU market in any systematic manner. These activities should be monitored to analyze the future success of the sector. In the near term, it is unlikely that Taiwanese firms will make inroads into the highly regulated EU market.91 And even over the long term, the ability of more than a handful of Taiwanese firms to enter the U.S. OEM market may be limited. Future activities across the Taiwanese automotive OEM sector could be shaped as much by developments in the larger East Asian automotive sector as by the domestic and foreign production strategies of individual firms. One key factor will be the evolution of sourcing patterns of Japanese firms and, in particular, the extent to which East Asian countries, including Taiwan, are further integrated into these sourcing networks. A critical issue involves the extent to which Taiwanese plants will be utilized to complement or support their Japanese partners’ plants in ASEAN member countries by, for example, increasing their supply of parts and CKDs (complete knock-downs) to these sites. Direct competition may be scuttling long-standing partnerships. For example, Honda terminated a four-decade-long JV with Taiwanese automaker Sanyang over the latter’s development of a motor scooter and its aggressive entrance into the Vietnamese and Mainland markets.92 At the same time, the recent surge in Taiwanese OEM exports to its partners’ plants, as well as CMC’s recent announcement that it will sell a 15 percent stake in its Chinese venture to Mitsubishi suggest that co-existence and cooperation are possible. Similarly, in 2003 Yulon split into Yulon-Nissan, which will focus on developing activities in the Mainland, and Yulon Motor, which will continue to manage the Taiwanese domestic market.93 The most important influence on the future direction of the Taiwanese auto industry, however, is likely to be the development of the auto industry on the Mainland, and the challenges and opportunities this will create. Crucial questions for Taiwanese auto assemblers and parts suppliers are whether and how China’s automobile capabilities will be integrated into the existing East Asian production network. The most likely possibility is that Japanese assemblers will continue to pursue regional complementarity schemes in East Asia, and American and European assemblers will use Mainland plants only to serve the China market. In such a case, Mainland production capacity would be used to serve only the local market and production capabilities currently being developed in China would be only marginally integrated into existing regional arrangements. Under these conditions, events on the

A TALE OF TWO SECTORS

127

Mainland would have only limited effect on trade and production levels in other countries in East Asia. A new operation on the Mainland, once approved by the foreign partner, has the potential to increase the power and independence of a Taiwanese assembler vis-à-vis its Japanese partner. CMC is a case in point. In December 2003, it announced a new JV in Fujian province with DaimlerChrysler for the production of Mercedes-Benz vans. This venture, which will also involve the Fujian government, is expected to involve a US$250 million investment for production facilities with a capacity of 40,000 vehicles.94 The venture has significant real and symbolic value to CMC: it provides the opportunity to produce higher-end vehicles in China and forges an alliance with one of the most respected brands in the automotive industry. At about the same time that CMC and DaimlerChrysler announced their JV, CMC announced that it would also sell a 15 percent stake in its existing Chinese venture to its Japanese partner, Mitsubishi. By selling a share of its successful Chinese operation to Mitsubishi, CMC hoped to align its Mainland objectives with those of Mitsubishi, which has long hoped to capture a large share of the Mainland market but has yet to establish independent production facilities there.95 The experience demonstrates the ability of a Taiwanese assembler to leverage its capabilities in Mainland China into ties with new partners and larger roles in the regional strategies of existing partners. But achieving the potential of relocating to Mainland China will not be without its challenges. Although the Mainland market was initially attractive because it had many similarities to the Taiwanese market of the past—it was protected by high tariffs and firms operated at relatively low volumes—it is changing rapidly, and the trend is toward a convergence with international standards. First, and most obviously, WTO accession will lead to reductions in formal tariff structures at a rapid rate: tariffs on imported components will be reduced from 28 percent to 10 percent by 2005, and tariffs on imported vehicles will be reduced from 80–100 percent to 25 percent by 2006. Local content regulations will also be eliminated, giving firms the option of importing parts rather than purchasing them locally. The multinational corporation (MNC) assemblers operating in China will most likely use suppliers operating in China (whether local or foreign) when they are competitive, but if they are not, they will import. The result should be a gradual degree of convergence with international standards, both in terms of quality and price, to an extent that was not possible in a closed system.

128 CHAPTER 4

Second, the Chinese government has continued to leverage market access in exchange for increasingly sophisticated technology, and the number of foreign players has increased dramatically since the first Taiwanese investment in 1996. After a decade of leisurely development, investment in the market began to explode at the end of the 1990s with new projects created by, among others, General Motors (1997), Honda (1997), Toyota (2000), Hyundai (2002), Nissan (2002), and Ford (2003), and each introduced a relatively recent model.96 The increase in the number of JVs has been matched, at the lower end of the market, by an increase in the number of wholly owned Chinese firms offering less expensive models for the growing number of individuals purchasing cars. The result of both of these trends has been more advanced models in the market (such as the Honda Accord, Volkswagen Passat, and Audi A6) and severe price competition. The price of the Shanghai Volkswagen Santana, for example, which dominated the Chinese market for a decade, peaked at over renminbi (RMB) 200,000, fell to RMB 158,000 in 1995, then to RMB 114,000 in 1998, and finally slipped below RMB 100,000 by 2002.97 In many respects, the primary challenge for Taiwanese firms is that a new premium is being placed on technology, even in emerging markets, and this is exactly where they are constrained by their dependence on Japanese firms. In the past, when a global auto firm built an assembly operation in an emerging market, a combination of local suppliers and imported parts were used to build a model that was long outdated in more developed markets.98 This was the approach that was taken in China during the 1980s, and it was an environment in which the Taiwanese might have excelled—the market was undemanding and weakness on the design side was widely tolerated. Increasingly, however, the assembly firms are being pressured to make frequent model changes, often launching models only several years after they have been introduced in home markets. In some cases, the foreign assembler uses key suppliers from home. Guangzhou Honda, for instance, outsourced 60 percent of the domestic content for the Accord in 2003, and 90 percent of this was to Japanese firms.99 In other cases, the foreign assembler must rely upon large and powerful global suppliers because they cooperate with these first-tier suppliers in the design process, and the suppliers provide complete modules and subsystems for a model. This makes it difficult to start production in a new area without the original supplier firm. Volkswagen, for example,

A TALE OF TWO SECTORS

129

is increasingly confronting this problem in China: they have existing suppliers in Shanghai and Changchun, but as they rapidly introduce new models there is a great deal of pressure to introduce the global supplier for each particular model. The increased reliance on foreign suppliers allows the assembler to raise the quality of production at offshore sites more quickly than in the past; rather than go through the laborious process of developing local suppliers, it simply has to wait for the global suppliers to begin production, and it is then assured of quality and service at a consistent level.100 The growing significance of global supply networks, and the potential they create for launching technically advanced models in emerging markets, puts Taiwanese auto firms at a competitive disadvantage. Taiwanese firms that relocate to Mainland China usually do not have welldeveloped design capabilities at home, let alone at their offshore facilities, and shifts to new models ultimately have to be approved in Japan.101 As Ravenhill and Bernard would predict, the conspicuous dependence of Taiwanese assembly firms on their Japanese partners for design capability relegates them one step down in the production hierarchy from the Japanese. The danger is that they will eventually be cut out of the production network. The Taiwanese assembly firms are very much at the mercy of their Japanese partners, and it is often an uncomfortable position. One senior manager of a leading Taiwanese OEM supply firm, for example, explained how his Mainland customers were intending to circumvent the Taiwanese assembly firms in a few years’ time, and establish direct partnerships with Japanese assemblers “who, after all, are the ones with the actual technology.”102 In short, the Taiwanese may be able to take advantage of protected markets in China to increase their production volumes in the short term, but as barriers to trade within China gradually weaken, these advantages will deteriorate and the Taiwanese assemblers will either have to create a new source of comparative advantage or continue to strengthen their alliances with other foreign firms. Policy Recommendations While the Taiwanese government has gradually extricated itself from the assembly sector, the auto parts sector would greatly benefit from increased public-private collaboration. The paralysis of collective action dynamics and a sustained focus on cost competition combine to dampen upgrading potential and expose Taiwanese parts firms to labor

130 CHAPTER 4

cost threats from below. As codifiable design processes, product and component standardization, and increasingly sophisticated electronic interchanges encourage value chain modularity, barriers to entry will fall for firms in developing countries, provided they can meet lead firm standards.103 Such modularity, however, also allows firms to concentrate resources in the most suitable market niche. The organizational changes that result from new information technologies and standards subvert the classic “big push” logic of late development. Late industrialization, according to this logic, must proceed simultaneously among multiple fronts because of the inherent complementarities and indivisibilities of economic processes.104 The large capital requirements of the development process and the presence of early developers thus justified an active state role. In the current environment in which economic processes are global in scope and increasingly divisible, however, it is more difficult for the late developer to leverage sufficient resources to enter simultaneously the global production network in a vertically integrated fashion. It is possible, however, to now enter as one critical node in such a network. The same features that create opportunities for developing countries to become integrated into global networks—the increasing use of information technologies and standardization—also make it easier for them to be replaced. One of the most significant challenges facing the strengthening of Taiwan’s auto parts manufacturers has taken the form of a collective action dilemma. While parts manufacturers have created an industry association, Taiwan Auto Body Parts Association (TABPA), and have succeeded without the support of assemblers, the undifferentiated nature of their products creates a “public good” dynamic. Noble examines this precise effect, arguing: “[f]or firms producing undifferentiated commodities, the aggregate capacity of the industry is a kind of public good. Decisions on pricing, output, and capacity expansion are interdependent. Each firm prefers to sell all of its output at high prices, but if one firm significantly increases output the price received by all the other firms will decline.”105 As a result, Taiwanese manufacturers have pursued an internal price race to the bottom, whereby the premium products they export to the United States are sold to U.S. distributors at minimal prices. The business associations of Taiwan have been characterized by scholars as historically rather weak. Noble argues: “ . . . Taiwan presents less fertile soil for corporatist patterns of cooperation. Industries typically

A TALE OF TWO SECTORS

131

contain a large number of producers, and business groups are smaller and less diversified than in Japan or Korea, while industry associations and related organizations are less well developed than in Japan.”106 As discussed previously, the relationship between business and government in Taiwan has been continually plagued by suspicion. Corporate and policy networks “are thin, and the relationship between business and government distant . . . (political) leaders feared that business influence would infiltrate the party and undermine political discipline and bureaucratic loyalty.”107 An exclusive focus on labor costs may leave Taiwanese firms illequipped to deal with what may prove to be the more dangerous competition. Firms that have an expertise in utilizing the information technology necessary to manage effectively the individual components of a modular supply network will continue to prove successful. For example, as MNC assemblers continue to expand into the China market, their MNC suppliers may quickly combine low labor costs with advanced technology to enter the aftermarket in China and abroad. The objective for Taiwanese firms is to take advantage of modularity, but then proceed to reinforce the linkages within the chain. The policy prescriptions of a “little-niche” world are very different than a “big-push” world. Rather than pursue grandiose industrial policies, government agencies could provide market advice (e.g., sponsoring trade fairs, facilitating relations with buyers, etc.) and work to develop and disseminate new technologies and production techniques within and between sectors. It is a more modest approach, but one that will lead to a durable comparative advantage. Meeting this challenge requires relocation combined with upgrading. Firms must take advantage of the modular nature of the auto parts value chain in order to lower costs, but they must also seek to increase the “stickiness” of the connections between the pieces of the chain by focusing on electronic linkages and standardization processes. There are four primary objectives that the government could pursue to improve the competitive advantage of auto parts manufacturers. First, the government could create a coherent process of standardization and certification throughout the industry, lending credibility and momentum to the current association of aftermarket manufacturers, TABPA. The U.S. aftermarket suffers from multiple certification programs created with multiple interests of different groups. Established in early 1986, the Auto Body Parts Association (ABPA) developed a certification program that immediately encountered funding obstacles and feuds between

132 CHAPTER 4

aftermarket and OEM manufacturers. In response, insurance companies, appreciating the potential cost savings in premiums, launched another initiative, the Certified Automotive Parts Association (CAPA), in December 1987. CAPA is a nonprofit organization that certifies, through tests and inspections, the quality of automotive parts used for collision repairs. Under CAPA, certified parts must conform to quality standards for fit, component materials, and corrosion resistance to measure their “functional equivalency” to similar parts manufactured by OEM suppliers. A participating firm’s factory and manufacturing processes are continually reviewed, while a product analysis is conducted by an independent testing laboratory. Association policies are determined by a nine-member Board of Directors representing auto-body shops, consumer groups, insurance companies, and parts distributors. A technical committee comprised of experts in the collision repair and auto-body parts industry performs regular in-depth reviews of the standards employed and periodically purchases parts in the open market. While CAPA has garnered considerable respect in the U.S auto parts industry and certifies close to 35 percent of all aftermarket parts in the United States, rival systems of private certification are emerging. In 2000, Global Validators, an automotive quality consultancy, launched yet another certification process directed at improving the quality of aftermarket crash parts. The Manufacturers’ Qualification and Validation Program (MQVP) serves a purpose similar to that of CAPA. It provides “guidelines that outline policies and quality management practices designed to ensure that aftermarket crash parts are equal in form, fit, function, performance, durability and appearance to OEM parts.”108 While some argue that competition among certification standards benefits the consumer, manufacturers must cater to numerous interest groups that detract from the creation of a single process and product standard. Multiple private standards create moving goalposts that are subject to the vagaries of special interest, and do not support a safety standard for the consumer. Second, the Taiwanese government could continue to support and strengthen a sophisticated testing center such as the Automotive Research and Testing Center (ARTC). The considerable capital costs of such an endeavor have prevented the collective provision of such a public good by the various manufacturers.109 Even in the United States, the lack of such a good is partly responsible for the diversification of standards. While CAPA has recently invested in a vehicle fit test, the capital cost of a modern comprehensive testing center is considerable.110 In

A TALE OF TWO SECTORS

133

addition, other agencies within the government, such as the Mechanical Industrial Research Laboratories of ITRI, mentioned previously, could play a critical role in the continual technical upgrading of the inspection protocol. Importantly, the budget for such a center could be ensured against the political vagaries of ministry appropriation politics, perhaps through the creation of a fee structure for the center’s services. Third, the government could capitalize on scale and strengthen its role as a clearinghouse for aftermarket parts and aim to improve efficiency of communication between key parties. Regular focus group trips between repairers abroad (final customers) and manufacturers (producers), as well as targeted conferences and trade shows would create more opportunities for upgrading and mitigate asymmetry of information. The regularity of such meetings would also lead to improved communication and more coherent discussion with distributors and insurance companies, the other critical parties in many developed markets. For example, with over thirty-five branch offices and 40,000 feepaying company members, the China External Trade Development Council (CETRA) serves as a clearinghouse for many auto parts firms in Taiwan, mostly from the OEM industry. As aftermarket auto parts firms are currently expanding, CETRA’s involvement would enable the more efficient matching of international companies and domestic firms, lending efficiency and scale to a process that is often hit or miss. CETRA recently aided several Taiwanese companies that were able to sign agreements with a Mongolian aftermarket purchaser.111 CETRA’s International Sourcing Council operates through branch offices abroad, where specialized teams aid large firms in the winning of lucrative contracts with large multinational auto assemblers. Unfortunately, participating firms are mostly from the information technology sector. However, the inclusion of auto parts firms in initiatives such as the Chain Store Project, whose teams are dedicated to managing accounts with Wal-Mart, Home Depot, and other large-scale chain firms, could create value added for these expanding firms that are seeking multiple distribution channels in export markets. Fourth, the government could seek to create collective leverage among leading aftermarket auto parts manufacturers, perhaps through the establishment of a cooperative brand for the sale of “A-class” products that sell in high volume. While many lead firms have achieved international levels of International Organization for Standardization (ISO) and Quality Systems (QS) certification, there are few channels

134 CHAPTER 4

for cooperative action among manufacturers, and price competition has led to an overall reduced market share in expansion markets, as well as the loss of negotiating leverage with distributors abroad. Manufacturers in other buyer-driven chains have faced similar challenges. Orange growers in California and dairy farmers in New England have formed cooperatives such as Sunkist or Hood, which provide clear advantages to small-scale manufacturers and seek to distribute profits more equitably between retailers/distributors, producers, and end consumers. There are several advantages to the cooperative form of brand promotion. Economies of scale, cost reductions, investments in technological upgrades, separate distribution channels, and a collective brand combine to create producer market leverage, larger funding for R&D, and political influence in legislation (e.g., in 1997 the USDA granted Sunkist US$2.1 million of its total US$90 million for export promotion even after it faced criticism for the large sums it had given to large agribusiness companies the previous year).112 All Taiwanese auto parts firms interviewed cited high costs of establishing an independent distribution channel in the United States, arguing that such channels are closely guarded by large distribution conglomerates such as Keystone. In reality, U.S. auto parts distribution channels experiencing the largest growth in 2002 were warehouse clubs and superstores.113 Significantly, since eBay Motors was launched in March 2000, “it has become the biggest conduit for used cars in the world, on track to handle more than $4 billion in car-related sales [in 2003]. They sell more cars before noon than the largest dealers sell all year.”114 In February 2004, the Toyota Motor Corporation agreed to become the site’s first exclusive automotive sponsor.115 The U.S. auto parts distribution industry is clearly in a state of dramatic change, and barriers to entry have weakened considerably as channels of distribution continue to diversify. The combined market power of a unified branding group of leading Taiwanese auto parts manufacturers, focused on highvolume products, could prove quite successful in shifting nodes of coordination and further migrating up the buyer-driven value chain. Conclusion: Niche Markets in a Modular Industry Rather than attempting to compete head-to-head with the major global suppliers, an alternative strategy for Taiwanese firms would be to integrate selectively into the global production networks. The Taiwanese

A TALE OF TWO SECTORS

135

textile/apparel and electronics sectors provide evidence that a certain degree of dependence does not necessarily preclude successful relocation.116 Firms in these sectors are not competitive in all aspects of the value chain. Western firms dominate the retail business, and continue to exert strong leverage as buyers. Given the inherent modularity of the value chains in these sectors, however, Taiwanese auto parts firms have been able to pick the activities in which they have a comparative advantage (while avoiding those in which they do not) and ruthlessly exploit these advantages. When firms in these sectors relocate, they do so as part of a global logic. They may only do a few activities, but what they do is done well. The challenge for these firms is to increase the integral links that connect them to buyers so as to avoid being cut out of the supply chain. A second approach would be to tie into the lower tiers of supply networks. While a Taiwanese firm would rarely have the design capability to be a first-tier supplier, many could become important players in lower tiers. Indeed, this is an alternative relocation strategy that many Taiwanese auto firms are employing with some success.117 In China, local protectionism led to closed supply networks in the past, but increasing competitive pressure is beginning to force these networks open. The strategy of several Taiwanese firms has been to establish competitive manufacturing operations in China based initially on exports, and to position themselves to acquire new business as first-tier suppliers are forced to lower their costs. Alliances with foreign firms are sometimes utilized as well. For example, one Taiwanese supplier established a JV with Johnson Controls, a U.S.-based global supplier, in order to gain access to the General Motors purchasing network both in China and globally.118 The third strategy is to seek to dominate the rapidly expanding auto parts aftermarket in the largest markets of the developed and developing world. Taiwanese auto parts suppliers have proven that these strategies are not mutually exclusive, and synergies may develop if a combination of approaches is adopted. Taiwanese collision parts manufacturers successfully established the U.S. non-OEM collision aftermarket. These firms have complemented their domestic activities by tapping into foreign markets and several are attempting to utilize a niche strategy to link with OEM assemblers and further migrate up the value chain. Serving the aftermarket has integrated these firms into a supply chain that has allowed them to control increasing sets of value-added activities within the aftermarket that will further support their growth.

136 CHAPTER 4

Traditional industries in a mature economy are not destined to disappear. The challenges Taiwanese auto assemblers face highlight the importance of initial technological dependence, scale, and the costs of such dependence in a changed automotive industry that demands global standards in all markets. Global patterns of consolidation and the modularization of automotive parts production, however, offer distinct opportunities for value chain upgrading and expansion into additional markets, particularly for firms willing to pursue a niche export strategy. The Taiwanese auto parts sector therefore illustrates the potential importance of government policies that could stimulate productivity and growth even in later stages of development, as maturing firms seek to expand beyond domestic market constraints. This case of Taiwanese auto assemblers and auto parts manufacturers thus both warns against unitary analysis of “mature” industries and also illuminates a far wider set of possibilities for firms competing in a global market. In particular, firms that are able to identify particular segments of the manufacturing value chain and successfully dominate such segments will create critical avenues of upgrading that in turn will better inoculate them to costbased threats from below.

5 Moving Along the Electronics Value Chain Taiwan in the Global Economy Douglas B. Fuller

Introduction Electronics has been a driving force in Taiwan’s economic development for the past two decades as Taiwan has moved from developing to developed country status.1 Taiwan began to emerge as a major assembler of electronics as early as the late 1960s (Kawakami 1996; Amsden and Chu 2003). From the mid-1980s through the end of the 1990s, the industry grew at roughly twice the overall rate of GDP growth (Hsueh et al. 2000, p. 158 for 1986 to 1990 GDP figures). In 2002 the total production value of information technology (IT) hardware was US$17.4 billion, excluding semiconductor production and Internet appliances production (IIY 2003, pp. 3-1, 3-4). This made Taiwan the world’s fourthlargest producer of IT hardware after the United States, China, and Japan. The value of semiconductor or integrated circuit (IC) production in 2002 was US$19.2 billion including design services (Ibid., p. 2-2).2 The production of Internet appliances accounted for another US$3.4 billion. This figure also includes Taiwanese-owned production facilities in other economies (Ibid., p. 3-1). In 2002, the growth rates of the IT hardware (12.1 percent) and semiconductor subsectors (23.9 percent) were well above the GDP growth rate of 3.59 percent. Semiconductor output comprised 21.6 percent of all industrial production and IT hardware accounted for another 19.5 percent. The growth in these industrial sectors has had a very large impact on Taiwan’s overall economic growth, because industry is still responsible for a large portion of Taiwanese GDP (31.1 percent).3 By any aggregate statistical measure, the electronics industry is a major force in the Taiwanese economy. 137

138 CHAPTER 5

Beyond its sheer weight and impact in the Taiwanese economy, the electronics industry has also been the focus of initiatives to create public and private organizations that have turned out to be of great importance to economic and technology policy and to the Taiwanese economy generally. A number of scholars have offered detailed accounts of the state’s efforts to promote the electronics sector through the creation and expansion of the Industrial Technology Research Institute (ITRI), a group of government-sponsored research institutes that have spun off many successful technologies. ITRI continues to be a major pillar of technology research in Taiwan, along with other agencies, such as MoEA (Ministry of Economic Affairs), STAG (Science and Technology Advisory Group), CEPD (Council for Economic Planning and Development), the Executive Yuan Development Fund, CDIB (China Development and Industrial Bank), and the Bank of Communications.4 In addition to public institutions, large private firms have emerged in this sector with the organizational capacity and scale economies to enter new product areas (Amsden and Chu 2003). Demonstrating Taiwan’s new technologygenerating prowess, three Taiwanese firms rank among the top thirty worldwide holders of U.S. high-technology utility patents in recent years (Floyd and Meyer 2002, p. 25).5 To understand the success of public and now large-scale private technology initiatives in Taiwan, it is necessary to understand the evolution of policy within the electronics sector. Four features have characterized the development of the Taiwanese electronics industry. First, the state has fostered domestic industrial competencies by linking local firms to major international firms. By creating stronger bonds with firms beyond Taiwan’s borders, Taiwan has pursued the technonationalist goal of fostering domestic technological capacities through technoglobalist means. Technonationalism is the developmentalist belief that national economies can mature and national security be assured only when domestic firms have substantial control over the generation of knowledge and technology (Keller and Samuels 2003, p. 9). The technoglobalist approach eschews national control in the belief that lower barriers to transfer and diffusion of technology to the local economy allow for more rapid economic development regardless of whether the conduits of the technology to the local economy are domestic or foreign (Ibid., p. 11). Second, the model of state-led and partially state-financed research through ITRI (hereafter referred to as the ITRI model) has been the mainspring of technological development efforts in the electronics

MOVING ALONG THE ELECTRONICS VALUE CHAIN 139

industry. Third, Taiwanese industry has emphasized manufacturing, with little emphasis on firm-based R&D and little international brand presence. Fourth, the successful Taiwanese electronics firms have been major suppliers to international lead firms (hereafter referred to as the supplier model), both in IC fabrication and design and in systems manufacturing (i.e., the manufacturing of electronics end products, such as PCs). The Taiwanese model of electronics industry development has been extraordinarily effective in increasing economic output and technological sophistication, but now is bumping up against the limits of development. The constraints of the supplier model are now apparent to both state actors and private industry. Although the supplier model in IC fabrication, the pure-play foundry model, has been remarkably successful in creating reasonably balanced interdependency between the Taiwanese foundries and their foreign customers, the Taiwanese systems manufacturers, such as Wistron (formerly Acer) and Quanta, remain in a dependent relationship with foreign brand-name firms, such as Dell and Hewlett-Packard. Furthermore, new opportunities have appeared as Taiwanese technology firms have expanded and thus become more capable of undertaking large-scale research efforts, and as China has emerged as an electronics-manufacturing base and as a pool of available technical talent. The responses to these pressures and opportunities are taking place in multiple arenas. The Taiwanese state is still very actively involved in pushing Taiwan into new technology areas even as more private initiative and leadership in technological development have occurred, particularly in IC fabrication. But the dependency of Taiwanese suppliers in systems manufacturing continues even as Taiwanese industry becomes more R&D intensive, with both foreign and domestic firms participating in R&D efforts in Taiwan. Finally, there are new areas in which Taiwan appears to be rapidly approaching the cutting edge of technology, such as wireless IC design. These judgments about the evolution of the Taiwanese electronics industry are based in large part on interviews in 1999–2003 with Taiwanese civil servants, scholars, and corporate managers that were carried out by the MIT Industrial Performance Center (IPC) Taiwan Project Team (see Tables 1–4 in the preface of this volume). This chapter examines patterns of response to global pressures in three segments of Taiwan’s electronics industry: IC fabrication, IC design,

140 CHAPTER 5

and systems manufacturing segments. The state has continued to use the ITRI model of development even as private initiatives grow. The state has expanded the repertoire of technoglobalist means to build up Taiwan’s R&D intensity by encouraging foreign multinational corporations (MNCs) as well as Taiwanese firms to set up R&D centers in Taiwan. Positive state and private results in these efforts have been counterbalanced, however, by policies with questionable efficacy to support industries that will have great difficulty becoming technologically innovative in Taiwan. The fate of private industry’s efforts to utilize China as a strategic asset has generally been consistent with the overall performance of a particular segment of the industry. Where the activities have been successful, as in the IC segments, these China initiatives have helped boost performance. Conversely, where the industry has been weak, as with the systems houses, these efforts to use China have not done much to boost competitiveness. A Tale of Two Industries: Interdependency in Integrated Circuits and Dependency in Systems Manufacturing IC fabrication and PC systems manufacturing were the two main segments of the Taiwanese electronics industry until the recent development of AMLCD (active matrix liquid crystal display) manufacturing. IC fabrication and PC systems manufacturing share the key features of ITRI-supported research: strong links to foreign lead firms and a strong manufacturing focus. But these two industry segments have achieved quite different positions in international value chains. The mainstays of IC fabrication in Taiwan, pure-play foundries, appear to have as much market power over their IC design customers as the customers have over them. In contrast, the systems firms remain in a dependent relationship with the international branded buyers of their products. The Taiwanese systems firms are squeezed on unit prices by the branded firms. As S.H. Chen put it when referring to the Taiwanese PC manufacturers: “[F]irms and nations that rely heavily on high-technology hardware manufacturing can be easily caught up in deteriorating trade terms” (S.H. Chen 2002, p. 257). Every major notebook and desktop PC manufacturer in Taiwan whom we interviewed complained of the great leverage of the foreign branded firms in the negotiation of prices. Both IC fabrication and systems firms have been the targets of state policy and have participated in research consortia, though at least in the

MOVING ALONG THE ELECTRONICS VALUE CHAIN 141

past the relationship between the state and the major IC fabrication firms was closer and included direct state investment. The major IC processing technology was transferred to industry through ITRI projects in the late 1970s, resulting in the spin-off of UMC (United Microelectronics Corporation), and in the mid-1980s, the spin-off of TSMC (Taiwan Semiconductor Manufacturing Corporation). The last major ITRI project in IC fabrication technology, the submicron project, spun off the TSMC affiliate Vanguard. This project was bitterly contested by UMC, and Vanguard never succeeded as an independent DRAM manufacturer (see the discussion in Fuller, Akinwande, and Sodini in chapter 3 of this volume). In the PC industry, foreign producers were the major manufacturers of computers in Taiwan until the late 1980s, but had become minor players by 1995 (Kawakami 1996). The government worked with private enterprise on these initiatives rather than attempting to spin off private firms from public research units. Probably for this reason, the ties between the executives of the PC firms and ITRI have not been as close as those between ITRI and the managers of the IC firms, many of whom were themselves former technocrats who had left ITRI when the new firms were spun off. Exemplifying these weaker ties in the PC segment, one of the largest desktop producers, Mitac, took part in only one of the three major government initiatives in desktop computer development. The government launched three major desktop computing initiatives in the 1980s,6 and three initiatives in the 1990s to transfer notebook technology. Both industries have relied heavily on a foreign customer base as well as technology transfer from foreign partners, particularly in the early years. Given the small domestic market in Taiwan, the computer manufacturing industry has been very heavily focused on exporting to the wealthy advanced markets of the United States, Japan, and the European Union. Despite the emergence of a large local IC design industry, foreign customers still represent approximately 60 percent of the customer base of TSMC, the leading foundry, and 40 percent of UMC’s customer base. However, UMC is unique in that it has a large number of UMC-linked design houses, such as MediaTek and SiS, as permanent customers. These firms do not represent truly independent Taiwanese design house customers of UMC, but rather a quasi integrated device manufacturer (IDM), a firm form in which the company both designs and manufactures ICs.

142 CHAPTER 5

The Success of the IC Foundry Industry As Fuller, Akinwande, and Sodini discuss in chapter 3 of this volume, TSMC jumped the gun in creating the pure-play foundry model before the issue of how to codify IC designs so that outside foundries could fabricate them had been fully resolved. The uncertainties surrounding codifiability made the foundry model a gamble at the time. But the pureplay foundry business eventually replaced the IDMs as the largest part of Taiwan’s IC production, even as Taiwan grew to become the world’s fourth-largest IC producer. The technoglobal strategic suppliers, the pureplay foundries, beat out the more technonational IDM firms that tried to combine and control all the major IC activities within their own domestic companies. The state did not attempt to limit the success of the foundry model, which it had after all helped to create, in order to favor the IDMs, but promoted both types of firms equally. Many new private firms began to enter the industry as IDMs in the late 1980s, but the state did not interfere with the gradual conversion of public (i.e., firms with significant state ownership) and private IDMs into foundries. By 1994 the technology necessary to transfer IC designs to foundries had been completely developed. Many of the current managers of TSMC’s fabs returned to Taiwan from the United States to work for TSMC at this time, as the promise of this model was fully realized (IPC interviews). The feasibility of codification, and the return of experienced engineers and managers from the United States, helped to account for the gradual increase of the foundries’ market share from roughly onethird of Taiwan’s total fabrication between 1992 and 1995 to a majority share from 1997 onward. Many IDMs, including Taiwan’s pioneering UMC, converted to the foundry model or were bought out by foundries. (See Figure 5.1.) Why did the foundry model succeed better than other electronics business models in Taiwan? The mainstream IDM approach was very difficult because it required simultaneous development of the next generation of process and product design technologies. Given Taiwan’s tight credit system, firms did not have access to the large amounts of patient capital necessary to pursue the simultaneous development of process and product technology. Between 1993 and 1997, total R&D as a percentage of sales hovered just above 5 percent compared to 10 to 15 percent of sales in large U.S. semiconductor firms.7 Priority was given to building economies of scale and acquiring new equipment, which embodied an increasing share

MOVING ALONG THE ELECTRONICS VALUE CHAIN 143

Figure 5.1 Foundry and IDM Production Share in Taiwan’s IC Industry (percentage) 100

Production Share (percentage)

90 80 70 60 50 Foundry

40

IDM

30 20 10 0 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Year

Source: 2002 Semiconductor Industry Yearbook (ITRI/IEK).

of the process technology. Capital investment averaged almost 70 percent of sales over the period 1993–97. Even this high level of capital expenditure should be viewed in the context of a rapidly growing industry with a net profit growth of 34.1 percent from 1995 to 1997 (ITIS 1998, pp. VIII–15, VIII–19). In other words, Taiwan’s semiconductor investment was sustainable because the industry was profitable. Due to the tight credit system of Taiwan, Taiwanese firms did not have the option of going heavily into debt to fund investment, the strategy of the Korean IDMs. The pure-play foundry model solved the ongoing Taiwanese dilemma of how to advance technology on small budgets by learning from customers and concentrating on advancing technology in only one area—IC fabrication. The foundries’ focus on and eventual mastering of process technology did not make them independent drivers of the IC industry. Instead, Taiwanese foundry firms supplied process technology and state-of-the-art manufacturing capacity. Outsourcing IDMs and design houses sought fabrication capacity. The result was interdependence between the foundries and their customers. In industry upturns, the foundries have the upper hand, as they have control of the scarce commodity, foundry capacity. In downturns, the IC designers and outsourcing IDMs have the upper hand, as the scarce commodity is chip orders and foundry capacity is

144 CHAPTER 5

abundant. Taiwanese foundries and American fabless design houses as well as some IDMs have tried to enter into long-term relationships to smooth out the cycle of dependency. In addition to Philips’ continued large stake in TSMC, the other major foundry in Taiwan, UMC, has sold equity to American design firms in return for dedicated fab capacity. Philips and TMSC are building a fab together in Singapore, and UMC has a joint venture with Nippon Steel in Japan and is building a fab in Singapore with Infineon. UMC had also captured the equivalent of the Holy Grail for foundries, outsourcing orders from AMD, one of the two leading microprocessor firms, with an agreement to produce AMD processors in a joint-venture facility. Foundries have wanted to secure microprocessor orders because microprocessors have the alluring combination of fat margins and massive volumes. Unfortunately for UMC, AMD ran into financial difficulties and was not able to continue its share of investment in this project. Nothing better demonstrates the difference between the state of dependency on foreign branded firms in which Taiwanese system houses find themselves and the interdependency with foreign lead IC firms that the IC foundries enjoy than UMC’s decision to prune its list of customers. UMC shaved its customer list in half from roughly 400 to roughly 200 because it felt that it was not making much money from the others, some of which were just trying to play UMC and other foundries off against each other in an attempt to drive down prices. UMC decided it could afford to do without their business (IPC interview). It is safe to say that no original equipment manufacturing/original design manufacturer (OEM/ODM) supplier in the systems industry would feel able to turn away half of its customers because they are driving down prices. That branded buyers push down prices is a fact of life for the OEM/ODM firms. For example, the average selling price for the notebook ODM firms has fallen from US$1,092 in the first quarter of 2000 to US$724 in the fourth quarter of 2002, even as total unit volumes rose from three million to five million units during this period (IIY 2003, pp. 8–22).8 Further success of the foundry model is demonstrated by the growing in-house capacity for process R&D by Taiwanese firms. The leading Taiwanese foundries cooperate with the leading IDMs in process technology research: UMC with Infineon, and TSMC with the high-profile Crolles initiative along with three IDM heavyweights: Motorola’s soonto-be-spun-off semiconductor unit, Philips, and ST Microelectronics. These initiatives involve the shift to 300mm wafers and to even smaller

MOVING ALONG THE ELECTRONICS VALUE CHAIN 145

lithography metrics, such as the move to leading-edge 90-nanometer lithography. According to one foreign customer and sometime informal consultant who has been following TSMC’s progress since its inception, the firm used to poach the required overseas Chinese talent residing in big U.S. IDMs, primarily IBM, when it needed to make the big transitions between wafer sizes, such as from six-inch wafers to eightinch wafers. With the transition to 300mm (twelve-inch) wafers, TSMC for the first time was able to do the process in-house rather than relying on bringing back talent from overseas. This transition to in-house research was enabled by the rapid increase in the company’s core R&D workforce from forty to over 200 in a couple of years, but also reflects the fact that virtually all the senior Taiwanese engineers in the United States who wished to return have already done so. The possible pool of recruits from outside Taiwan has essentially dried up (IPC interview). One factor that has facilitated this transition is that TSMC (and UMC) have been able to concentrate their relatively meager resources on semiconductor fabrication process technologies, as opposed to the whole array of semiconductor technologies that global IDMs have sought to develop. As early as 1997, TSMC and UMC were able to crack the top ten of U.S. semiconductor process utility patents despite their relatively small revenue bases and small R&D budgets compared to the top global IDMs (Floyd and Meyer 2002, p. 16). The Rise of IC Design The success of the IC foundries has been replicated in IC design. Similar to the foundry model of focusing on one activity, the fabless design houses focus solely on IC design and outsource chip making to the foundries. Unlike the foundry model, this fabless model was not a Taiwanese innovation, but rather was invented by firms in the United States. However, only with the rise of the Taiwanese foundries and the codification of design data did this fabless model become successful, even in the United States. The Taiwanese design industry was still in its infancy in the early 1990s, but developed rapidly through the decade. In 1993 the industry consisted of sixty-four fabless design companies, which were all of rather modest scale as the total sales revenue for the entire design segment was only NT$11.7 billion (New Taiwan dollars, approximately US$433 million). By 2001, there were 180 houses, and sales revenues

146 CHAPTER 5

had increased more than ten times in nominal NT dollar terms to NT$122 billion (approximately US$3.7 billion) (ITIS 2002, p. 8-6). More important, a number of Taiwanese firms have become large international players. In 2002, MediaTek and VIA were both among the world’s top ten largest fabless design firms, ranking fifth and sixth respectively, with all the other top ten firms being U.S.-based (IC Insights, February 19, 2003). The Taiwanese firms showed the ability to oust their U.S. counterparts from key markets. MediaTek is today the largest player in the market for Codec chips for optical disk players, having driven out former lead firms, such as Oak and ESS. Realtek has become the biggest supplier of Ethernet chips, driving even the mighty Intel out of the market. On the other hand, Taiwanese firms have often had to accommodate and follow foreign rivals rather than confront them in head-to-head competition. VIA tried to compete against Intel in the PC chipset market rather than work with Intel as rival Taiwanese chipset design houses, ALI and SiS, chose to do. While VIA hoped to use an intellectual property (IP) sharing agreement the firm had with National Semiconductor to protect itself against charges of patent infringement from Intel, the strategy was not successful. VIA’s sales were badly hurt by the ongoing legal battle and the company’s sales fell 28 percent in 2002 and 15 percent in 2003 (IC Insights 2003). Eventually, VIA chose the better part of valor and reached an agreement with Intel. In contrast, ALI and SiS suffered none of this financial pain as they were willing to follow Intel chipset standards. The state fostered the IC design sector, although it is no longer the lead player today. ITRI played a significant role in developing this sector. CCL (the Computer and Communications Laboratory), a former division of ERSO, licensed foreign chipset designs to local firms (Noble 1998, p. 144). The local industry also received much talent from the U.S. firm, Chips&Technologies (Dedrick and Kraemer 1998, p. 159). The top managers of the three largest Taiwanese chipset firms, ALI, VIA, and SiS, are all returnees from the United States, and two of the three are ex-employees of Taiwan’s main chipset competitor, Intel. OESL (Opto-electronics Science Laboratory), another ERSO spin-off, and ERSO worked together to help firms create chips for optical disk players, and MediaTek took part in these efforts. For the large firms, inhouse efforts have replaced much of the work done by ERSO and the other institutes of ITRI, though the state still contributes much to the push into entirely new areas, such as wireless technologies. While the overall percentage of sales devoted to R&D remained roughly constant

MOVING ALONG THE ELECTRONICS VALUE CHAIN 147

at 10 percent, sales exploded and overall R&D expenditures had risen to NT$12.2 billion by 2001, more than the entire sales revenue of the design industry in 1993. The Glass Ceiling of ODM Systems Manufacturing In contrast to the pure-play foundries, the systems manufacturers in Taiwan, originally PC manufacturers, have grown rapidly, but have become trapped in low-margin manufacturing services for large international branded firms. In this industry, there is no silicon cycle in which foundry service prices rise sharply when demand is high, allowing the foundries to reap vast profits. Instead, the cost pressures applied by branded firms just lessen rather than disappear as demand increases for electronics systems products. In contrast to the foundry model, which was an innovation that gave first-mover advantage to the Taiwanese foundries, the Taiwanese PC suppliers have organizational innovation that differentiates them from competitors. The Taiwanese firms’ dependence on the international branded firms is not entirely disadvantageous. As Dedrick and Kraemer (1998) have argued, the Koreans and the Japanese tried to go it alone without extensive cooperation with the American firms under the Wintel (i.e., Microsoft’s Windows OS and Intel’s CPU) standard, and they failed in this endeavor. In classic technonationalist fashion, the Japanese and Koreans wanted to reinvent the computer to use the core technologies they owned rather than using Intel’s CPUs (central processing units) and Microsoft’s OS (operating system). However, they were never able to come up with computer systems that won significant market share from the Wintel standard. In contrast, the Taiwanese accepted the Wintel standard and succeeded by being close followers of the lead firms while avoiding the costly and ultimately unsuccessful bets on alternative standards. Thus, dependence on the branded PC firms may actually have proven to be beneficial, with Taiwanese firms undergoing technological upgrading and industrial expansion that the technonationalist Koreans can only look at with envy. In the PC industry, the state helped local firms to become suppliers to the foreign firms that had come to Taiwan to set up manufacturing facilities. The local supplier firms, sometimes with the help of the state, were able to use their supplier role to upgrade technologically. Gradually, these local suppliers developed into full-service original equipment

148 CHAPTER 5

manufacturing (OEM)9 firms and then into original design manufacturer (ODM) firms producing complete PCs for own brand manufacturer (OBM) firms, principally American ones. State promotion of the PC industry started shortly after the promotion of the IC industry. However, initiative in this area was more equitably shared between the public and private sectors. The PC industry, narrowly conceived, has been an assembly industry and thus did not have the same high technological barriers to entry as the IC industry. Thus, private firms were more willing to invest from the beginning to catch up to the technological frontier. The dominant producers in the PC sector were at first American producers, such as IBM and Commodore. The activities of these foreign producers in Taiwan were significant because they were conducting the core manufacturing activities in the production of PCs, whereas foreign firms in the IC sector only carried out the back-end tasks of assembly and testing. In the late 1970s, foreign manufacturers made up the great bulk of PC-related production in Taiwan.10 The share of foreign computers manufactured in Taiwan gradually declined from 57 percent in 1984 to 30 percent in 1990 (Kawakami 1996, p. 6). By 1995, the share had fallen to 15 percent (Hwang 1995, p. 45). Today the foreign manufacturing segment is insignificant, and figures are no longer kept by Taiwan’s Market Intelligence Center. The decline resulted not from any efforts to drive out foreign assemblers, but rather from the U.S. MNCs’ strategy of increasing the outsourcing of production to Taiwanese firms. While government projects through ERSO were important for the PC industry, the two industry leaders, Acer and Mitac, were doing OEM for International Telephone and Telegraph in 1982. Mitac did not participate in two of the three big desktop computer projects run by ERSO. The logic behind the OEM relationship helps to explain how these firms were able to upgrade technologically outside of ERSO while they were still very small by leveraging their relationships with outsourcing partners.11 The experience of Mitac, Acer, and other firms, such as the printed circuit board– manufacturer, Compeq (called Compaq in English until lawyers from Compaq in the United States caught up with it), confirms this theory of upgrading. The intensive OEM relationships with foreign, particularly U.S. firms, and the ability of relatively small firms to enter into PC production in the early years help to explain why ERSO did not play as critical a role in the development and diffusion of technology as it did in the IC industry. The notebook initiatives undertaken by the ERSO spin-off lab, CCL (Computer

MOVING ALONG THE ELECTRONICS VALUE CHAIN 149

and Communications Laboratory) in the 1990s helped diffuse notebook technology and featured more balanced cooperation between the state and private firms and less state leadership over time (Mathews 2002). The Taiwanese OEM firms have attempted to move up the chain into design in order to lessen their dependence on foreign brand name firms. Although the Taiwanese firms have significantly enhanced their capabilities, they have not yet captured enough strength in design to establish a truly interdependent relationship with the brand name PC firms. Over the last decade, Taiwanese firms have moved from designing desktops to designing low-end notebooks to designing high-end notebooks and low-end servers. This is a considerable technological achievement that only a handful of other countries can match (the United States, some in the European Union, Japan, and Korea). Yet, despite the ability of the Taiwanese to design and produce half the world’s notebooks and the increasing quality and sophistication of these products, the branded firms have still been able to force down average prices from a little over US$1,092 to US$724 over three years (2000 through 2002) (IIY 2003, Section 8, pp. 22–29). Over the course of a decade, notebook margins have fallen from 15 percent to 6 percent (B. Li 2003). Some report that major firms are pressured to offer prices that put per unit profits at US$30 or less, equivalent to roughly a 4 percent margin, assuming a selling price of US$724. Moreover, these slim margins appear to have already taken into account the relocation of production to China where labor is cheap (Tzeng 2002). Branded firms can choose from a number of manufacturing firms, both Taiwanese ODMs and American CEMs (contract electronics manufacturers). The Taiwanese firms are well aware of their predicament. Figure 5.2 shows the “smile curve” invented by Stan Shih, Acer’s founder and chairman, to illustrate the predicament. Interviews with Taiwan’s leading desktop and notebook PC assemblers indicate that these firms regard the ODM strategy as at best an incomplete solution to capturing value, and the addition of global logistics services seems to be a bid to further enhance or at least preserve value as global customers demand more services from OEM/ODM suppliers (Lee and Chen 2000; Schive 1999b, p. 2; IPC interviews with Taiwanese PC manufacturers). Firms have also tried other strategies, such as expanding product range and cutting costs by moving production overseas, principally to China. Despite these efforts, the cost pressure on the OEM/ODM firms has been unrelenting, even as the absolute size of these firms and their shares of the world market have increased

150 CHAPTER 5

Figure 5.2 Stan Shih’s Smile Curve Value Added

• Technologies • Manufacturing • Volume

• Brand • Channel • Logistics

Software CPU DRAM LCD ASIC Monitor HDD/CDD Motherboard PC System

Components Segment by Product Line Global Competition

Distribution Segment by Country Local Competition

Source: J. Dedrick and K. Kraemer, Asia’s Computer Challenge: Threat or Opportunity for the US and the World (New York: Oxford University Press, 1998), p. 156. Reprinted with permission from Acer America Corporation.

(see Table 5.1). One reason is that the large Taiwanese OEM/ODM firms are competing with a number of large, versatile contract electronics manufacturer (CEM) firms from the United States, which operate plants around the world. The Taiwanese do not control much of the technology of design or manufacturing. Thus, they do not have any obvious advantage over any other firm capable of manufacturing a computer. In contrast, the Taiwanese foundries have developed substantial process technology as well as a performance lead over would-be rivals. Emerging Opportunities for Taiwan’s Electronics Firms: Intrafirm Capabilities and China Taiwanese electronics firms have also been exposed to new opportunities that influence their strategies and their interaction with the Taiwanese

MOVING ALONG THE ELECTRONICS VALUE CHAIN 151

Table 5.1 The Concentration of Taiwan’s IT Industry in 2000 Taiwanese production (%) Share of top firms Notebook Computers 78.75 (5 firms) Desktop Computers 85.00 (3 firms) Motherboards 54.00 (5 firms) Monitors 44.00 (5 firms)

Share per top firm 15.75 28.80 10.80 8.80

World market share (%) Per top Taiwanese Taiwan firms 65 10.24 22 6.20 88 9.00 59 5.20

Source: Various years, Chinese Economic News Service (CENS) and a Japanese electronics firm’s internal marketing report.

government. Firms have grown in size and capabilities. With this growth, they have felt less inclined to follow the state’s lead or even to choose cooperation with the state over other possibilities, such as collaboration with foreign firms. The IC industry has been the shining success of the state’s technology policy, but the IC industry has also given rise to a cacophony of private industry voices increasingly critical of state intervention in the sector. The state’s efforts came under attack relatively early in the industry’s development. UMC, the first state-sponsored IC firm, criticized the Very Large Scale Integration Project that created TSMC in 1986 because UMC viewed the project as taking resources away from itself. The next project to develop more advanced process and memory technologies, the Submicron Project, was attacked by Acer from the very beginning as a waste of government resources. Acer had previously agreed to build an advanced memory fab with TI (Texas Instruments) in 1989. As the Submicron Project drew to a close in 1993 and 1994, UMC and TSMC quarreled over the spoils, a new fab that eventually became Vanguard. ITRI’s budget was cut in half in 1994 by the Legislative Yuan in the wake of criticism of the Submicron Project as a transfer of public funds to what were deemed to be mature private enterprises. While ITRI’s funding subsequently recovered, the scope for public initiative in the area of ICs has since narrowed considerably. In the late 1990s, ERSO tried to organize a consortium, ASTRO, to carry out research on future generations of process technology. A recent study on collaborative research in Taiwan cites ASTRO as an example of continued

152 CHAPTER 5

successful state–private enterprise research collaboration (Mathews 2002), but in fact the project never took off because the technology leaders, TSMC and UMC, were not interested in joining. The other major IC fabrication firms were willing to join, but these firms were not as technologically sophisticated as TSMC and UMC. Indeed, the underlying motivation of the project was to help these less technologically sophisticated firms to upgrade. In the end, however, the project was unable to obtain large government funding because TSMC and UMC were unwilling to participate. While large private firms may accept help in entering new industries, the combination of government fiscal constraint and the reluctance of leading private firms to cooperate once they acquire independent innovation capability suggests that the trend away from public leadership in the IC industry may be replicated in other areas. The shift toward private initiative in high-technology areas has been made possible by the increasing scale and scope of Taiwanese private firms, particularly those that arose in the new technology products areas. In 1987, four of the ten largest manufacturing firms in terms of sales were state owned. By 1999, only five of the top fifty were state owned and only one of these, Chinese Petroleum, was in the top ten. In 1999, the highest-ranked high-tech firm was Acer, the second-largest manufacturing firm. In 1987 Acer had been ranked fifty-third and UMC, the largest IC firm at the time, ranked 122. Indeed, eight of the top fifty firms in 1999 were operating in the Hsinchu Science-based Industrial Park and four of the top ten largest firms were Taiwanese high-technology firms. Two subsidiaries of Philips were also in the top ten (Tianxia [Commonwealth], July 1987, August 2000). By 2002, only three of the top fifty were state owned and only one of these was in the top ten, again Chinese Petroleum. In 2002, the highest-ranked high-tech firm was Hon Hai, the second-largest manufacturing firm. In 2002, six of the top ten firms were private high-tech firms and twenty-four out of the top fifty were also high-tech firms (Tianxia [Commonwealth], May 1, 2003). One measure of the increasing ability of private Taiwanese firms to pursue independent research is to compare the number of their recently received U.S. patents to the patents received by ITRI. While ITRI received 219 U.S. utility patents in 2001, several Taiwanese firms have been more active. Hon Hai has been very aggressive in patenting its electrical connector technology and has turned the tables on its U.S. rival by suing that firm, AMP, for patent infringement, after years of being the target of lawsuits by AMP. Hon Hai has received the largest

MOVING ALONG THE ELECTRONICS VALUE CHAIN 153

number of new patents in the field of electrical connectors of any firm in the world in recent years, surpassing its U.S., Japanese, and French rivals. In 2001 it ranked in the top thirty holders of new U.S. hightechnology utility patents worldwide and third among Taiwanese firms and institutes, with 309 U.S. utility patents (Floyd and Meyer 2002, pp. 19, 25, 40). TSMC and UMC posted 589 and 629 U.S. patents, respectively, in 2001, making them the second- and first-largest U.S. patent holders in Taiwan. TSMC would be the largest Taiwanese holder of U.S. utility patents, with 691 patents, if the patents of its subsidiary Vanguard were included in TSMC’s total (Floyd and Meyer 2002, p. 40). The high rankings of TSMC, UMC, and Hon Hai are remarkable, given that much of the competition comes from very large, diverse business groups registering patents in many different areas (Floyd and Meyer 2002, p. 25). As mentioned previously, both TSMC and UMC have also entered into major international R&D initiatives as equal partners. In general, there is a bifurcation in patenting patterns between the components firms, which produce ICs and capacitors, and the systems houses. Traditionally, the systems houses were OEMs and concentrated on manufacturing. There was little awareness of the importance of patenting even as the firms moved into ODM. The generally greater importance of proprietary technology for components firms, relative to the OEM/ODM service-oriented model predominant among the systems houses, also helps to explain the greater propensity of components firms to innovate and patent. Even the IC equivalent of systems manufacturing services, the pure-play IC foundries, need to differentiate themselves from their competitors through R&D in process technology. The technology intensity of fabrication is relatively high compared to systems assembly, and much of the client base consists of fabless design companies, which do not have manufacturing expertise to share with the foundries. (Those IDMs that outsource to foundries often send out products based on mature technologies while attempting to retain cutting-edge or niche processes in-house. Consequently, IDM customers at this stage may not be ideal learning partners. For example, Analog Devices outsources to TSMC mature digitized products that can be fabricated on complementary metal-oxide-semiconductor [CMOS] lines, but retains the latest analog products in-house. Analog Devices retains cutting-edge, albeit niche, analog process technology.) Systems houses also have fewer patents due to the rapid turnover of systems products within the context of servicing client firms, which often share in the economic rents gained

154 CHAPTER 5

from technological improvements. The cost of filing patents may not be worthwhile because the product is often out of date by the time the patent is granted. Thus, the number of patents may not fully reflect the commitment to R&D by systems houses. Despite these constraints on the importance of proprietary technology for systems manufacturing, the systems firms have started to stress the importance of in-house R&D and patents. This new emphasis on R&D arose in conjunction with new government policies that encourage the establishment of R&D centers. The state subsidizes up to 50 percent of the cost of corporate R&D (though the effective write-off is closer to 20 percent, because certain R&D expenditures, such as plants and equipment, are excluded from the subsidy program, according to interview subjects). The growth of R&D has been impressive. According to the Ministry of Economic Affairs’s (MoEA’s) Web site, the number of Taiwanese firms officially registered to establish R&D centers has risen to thirty-nine. Several of the larger Taiwanese electronics systems firms have added or are in the process of adding thousands of new R&D engineers. One systems firm is building a facility in which it plans to house 7,000 R&D engineers, up from 2,000 R&D engineers currently (interview with R&D manager, November 2003). In recent years, this firm’s rate of patenting in the United States has risen from an average of less than one per annum to an estimated sixteen in 2003 (internal company document), a figure representing more than half of its total inventory of U.S. patents. The new emphasis on patenting at this firm, which started out as a notebook manufacturer, reflects its recent entry into wireless products as well as servers. This firm is encouraging patenting in these newer fields while discouraging it in the notebook segment, where patents have too low a value due to product maturity, short product life, and the patent position of leading foreign firms in the basic notebook technology. The Call of China China has presented opportunities for Taiwanese manufacturers interested in low labor costs since the Taiwanese government began to allow investment in China in 1990 (or even before, as some Taiwanese businesses illegally invested ahead of official legalization). Taiwanese electronics firms generally have relocated to China as soon as their main products have matured. The PC industry has gradually moved production to China over the past decade. The lower-value-added peripherals

MOVING ALONG THE ELECTRONICS VALUE CHAIN 155

and components, such as computer pointing devices, keyboards, and cases, were relocated first. Then, in the mid-1990s, the production of power supplies and scanners began to relocate. By the late 1990s over half the motherboard and monitor production by Taiwanese firms was carried out in China, and Taiwanese desktops started to be produced there in large numbers. In 2000, for the first time, significant numbers of notebooks were produced in China (Fuller 2002, p. 2). As late as 1999, interview subjects in the notebook segment were claiming that they had no interest in investing in manufacturing facilities in China, but these firms abruptly reversed course and all of the major notebook companies have facilities in China at this point. In the last several years, Taiwanese IC design houses have begun to set up large-scale design labs in China. And since 2001 Taiwanese foundries have also been preparing to set up shop in China. The opportunities available in China have been different for the two industry segments, and these differences can be understood in terms of the different positions they occupy in the value chain. The IC industry moved to China from a position of strength. The industry wanted to expand its opportunities, and perhaps also, in the case of the foundries, to ward off competition from local foundry start-ups. In contrast, cost pressures drove the Taiwanese systems manufacturers to set up operations in China. In the former case, establishing a base in China just adds one more strength to already formidable competitors. In the latter case, China merely offers a salve for the cost pressures brought to bear by the international branded firms, rather than a solution to the problem of being at the bottom of Stan Shih’s value curve. Information from interviews with Taiwanese firms suggests that the Taiwanese desktop and notebook PC manufacturers moved their production to China because of the continuing pressure on margins. Industry experts and some industry managers claimed that it was the branded buyers themselves who pushed for this directly, though this author suspects the branded buyers simply saw the benefits in terms of cost savings reaped by the initial experiments of certain firms and demanded that their suppliers toe the line. In other words, demands for lower prices forced the ODM firms to move to the low-cost locations in order to maintain their margins. The ability of the branded firms to impose these low prices on the ODMs was enhanced by the emergence of more and more suppliers within Taiwan and abroad (the U.S. contract manufacturers) capable of handling large volumes of quality products. The movement of production to China .

156 CHAPTER 5

indicates that the increasing design capabilities of Taiwanese firms did not bolster margins enough to stave off cost-cutting pressure to move production to China. The Taiwanese foundries were not the first to invest directly in sophisticated fabrication in China, but Taiwanese engineers from TSMC and UMC formed the initial core staff of two new foundries in China, Grace and SMIC. The emergence of these firms (first announced in 2000) prompted different initial reactions from TSMC and UMC. UMC moved quickly, albeit stealthily, given Taiwan’s export-control laws, to set up an offshore company in the Caribbean to invest in an eight-inch facility in Suzhou called Hejian. (At the time, the law forbade even the transfer of older six-inch fab technology to China.) TSMC’s chairman, Morris Chang, repeatedly and pointedly expressed great skepticism concerning China’s potential as a base for advanced IC fabrication. However, by late 2001, even TSMC began the procedures to apply for permission from the Taiwanese government to invest in a by-then-legal eight-inch wafer fab in China, and received initial permission in the autumn of 2002. Meanwhile, UMC denied that Hejian was a UMC affiliate, because to have acknowledged it as such would have been to admit that UMC flouted Taiwanese technology export control laws. Nevertheless, Hejian’s affiliation with UMC was demonstrated by two journalists from the Asian Wall Street Journal, who tracked down the company filing for Hejian and discovered that the nominal company owners were two midlevel managers still working in UMC’s offices in Hsinchu (Dean and Uimonen 2002, p. A8). The start-up foundries were the front-runners in trying to bring sophisticated fabrication technology to China, but they served more as a signal to the Taiwanese foundries that China was ready than as an immediate competitive threat. One advantage offered by China is the large number of well-educated if not thoroughly trained engineers who are somewhat less costly than their Taiwanese counterparts.12 More important than the cost savings per se is the opportunity to expand the supply of engineers without having wages spiral further out of control because of Taiwan’s tight labor market for electrical engineers. China also offers geographic risk management, which Taiwanese foundries and their customers have been concerned about ever since the 1999 earthquake in Taiwan threatened to disrupt production. Finally, the local Chinese startup firms offer opportunities, given that the consumers of ICs, the electronics manufacturing firms, are moving to China. This argument is much

MOVING ALONG THE ELECTRONICS VALUE CHAIN 157

less convincing, since lightweight, high-value chips can easily be fabbed in Taiwan and then sent to the PRC. Indeed, the local IC design industry in China has used TSMC and UMC for a number of years without difficulty. UMC has been out in front in China. Hejian’s eight-inch fab is already up and running. More important, UMC has brought a number of UMC-affiliated IC design houses, as well as other Taiwanese design houses, to Suzhou. These firms plan to train a large number of local engineers to design ICs, often in conjunction with design teams in Taiwan. This effort has enabled Suzhou to become one of the most active centers for IC design in China, even though it is not one of the seven central-government-designated national IC design bases. The UMC-led design activity is still illegal under Taiwanese law, as Taiwanese IC design is strictly prohibited from being conducted in the PRC. UMC is not the only culprit, as virtually all of Taiwan’s major IC design houses, as well as Taiwan’s IDMs, have set up design facilities in China—principally in Greater Shanghai, including Suzhou and Hangzhou. These firms come in search of human resources, and China has resources to offer in terms of new graduates as well as returnees from Silicon Valley and other centers of innovation. VIA has announced a plan to hire as many as 1,000 design engineers in China and already has three separate design centers there. They have launched a large advertising campaign modeled on “Intel Inside,” proclaiming the firm’s chipset line “China’s Chip,” which is also a pun, in that it sounds like “China’s Heart” in Chinese. Other leading Taiwanese design houses, such as ALI, Realtek, Sunplus, and Faraday, have design teams in Greater Shanghai. Finally, a number of Taiwanese have come to China to start new design companies, taking advantage of the Chinese government’s preferential policies and the pool of human resources. As with the IC fabrication segment, these investments in China seem to be augmenting the firm’s scale rather than displacing activities in Taiwan. And while it is true that the Taiwanese who leave Taiwan to found start-ups in China are resources lost to Taiwan, they often use Taiwanese foundries to fabricate their chips. Continuing State–Business Interactions Despite the growth in size and capabilities of Taiwan’s private firms, the state still has a role to play in Taiwan’s electronics industry. The state has continued to employ the ITRI model even as private initiative has expanded. The ITRI model still matters in those new technology areas

158 CHAPTER 5

where Taiwan does not have large firms with relevant capabilities, such as wireless technologies. The state also has been very active in encouraging investment both in specific industry segments, such as AMLCD display production, as well as in activities that span segments, such as R&D. Some of these segment-specific initiatives have not proven very worthwhile (as will be shown below), suggesting perhaps that as Taiwan moves closer to the technology frontier it should abandon the industrial targeting approach and stick to innovational activities, such as R&D, instead. The state is also trying to attract more foreign investment to push for R&D-intensive activities as a replacement for manufacturing. Private enterprise has voluntarily cooperated with the state because ITRI offers tremendous technical resources. The collaboration is usually with ITRI, but the resources frequently come from other state organs, such as the Industrial Development Bureau (IDB) of the MoEA. Private enterprise has also generally embraced the other government initiatives for R&D centers because they offer incentives to firms. The one major area of business discontent with government concerns their management of the cross-straits economic relationship with the PRC. Here the private sector continues to criticize the obstacles to investment in and direct trade and transportation links to China. Forging New Directions: Going Wireless The state, together with private firms, has been very successful in creating new opportunities for Taiwan in wireless technologies critical to the convergence of computer, consumer, and communications products. The Taiwanese government has identified wireless technology as one of the priority industrial segments for the current six-year development plan (2002–7), the Challenge 2008 Plan. In the area of radio frequency (RF) circuit design, Taiwan government initiatives have helped propel Taiwan toward the technology frontier in an area previously dominated by U.S. and European firms. As ODM firms try to broaden their product base away from PCs and deepen their design ability, many firms have gone into cellular phone design and production. Many firms have also begun to produce wireless local area network (WLAN) equipment. Much of the technology for these products was developed in conjunction with ITRI’s CCL. Taiwan has also benefited from the severe downturn in the U.S. telecommunications industry by bringing back veteran experts from that industry. The

MOVING ALONG THE ELECTRONICS VALUE CHAIN 159

industry has also drawn on personnel from the Chungshan Institute, the Taiwanese military research institute, where many wireless technologies were first developed for military applications. This initial military orientation helps to explain the large U.S. and European lead in these technologies, particularly RF design, and the relatively weak position of Japan. In the Taiwanese case, wireless technology research was restricted to government institutes until martial law was lifted in 1987. The results of these efforts have been very promising. The Taiwanese WLAN makers now supply 80 to 90 percent of the world market. Major Taiwanese ODMs also supply the large branded cell phone makers: for example, BenQ (Acer Group affiliate) supplies Motorola and ASUSTeK supplies Microsoft’s smart phone. BenQ has ambitions to become a global brand, but its efforts to have a brand in China have been handicapped by the licensing arrangements of the PRC government. Wireless products are helping to diversify Taiwan’s electronics industry away from its concentration on PCs, but the ultimate outcome for the ODMs is unclear. They may yet find themselves in the same low-margin trap as with PC-related products. Indeed, a buyer’s market may be developing, with sixty firms now making WLAN equipment and virtually all the major electronics systems firms entering cellular phone production. In the area of wireless IC design, Taiwan clearly has made headway in catching up to the technology frontier. Just a few years ago there were no firms in Taiwan producing the digital baseband (a critical IC needed in wireless products) for WLAN equipment or cellular phones, let alone RF chips. ERSO’s microelectronics department spun off two teams in RF circuitry. One team, led by the microelectronics center’s former director, went to MediaTek. The other team went to BenQ and was subsequently spun off as Airoha. Both of these teams were trained to do RF chips for GSM phones and WLAN equipment. There is one other firm that does RF design in Taiwan, Mu Chip, though some WLAN baseband firms claim Mu Chip is much farther behind MediaTek and Airoha. The other Taiwanese IC design houses do their RF design work outside of Taiwan. For example, ALI has a team in Irvine, California. Realtek and VIA have their teams in Europe, and Fodus and Ralink have their RF teams in Silicon Valley. ERSO did not invest any effort in developing the baseband because this product was in a digital design realm with which Taiwanese firms were already experienced. Private firms, such as Realtek and ADMtek, lead the way in developing baseband chips. The significance of these new Taiwanese-based RF firms should not

160 CHAPTER 5

be underestimated. A major IC power, Japan, has had no serious RF contenders, and now Taiwanese firms are moving into this area. The movement has been much faster than anticipated abroad. One U.S. industry expert reportedly stated in January of 2003 that there was no way that Taiwanese firms would enter the RF chip market at 802.11g (the newest WLAN standard) any time soon, but by October of 2003 Airoha was able to do so. While even Taiwanese RF chip designers still concede that the major U.S. powerhouse and one of the biggest fabless firms in the world, Broadcom, remains unchallenged due to its ability to provide a total solution ahead of everyone else and then aggressively price down, other U.S. and European firms may retreat in face of low-cost Taiwanese competition. This entrance into RF design fills the final big gap in Taiwan’s overall IC design technology, analog/mixed signal design, or as Michael Lu, CEO of Airoha, puts it, “RF is the missing diamond on this crown” (M. Clendenin, IC Insights, July 21, 2003). Targeting R&D and Bringing in the MNCs One of the Taiwanese government’s responses to the loss of manufacturing jobs to China, and, perhaps, to the dilemma of the ODM firms, has been the promotion of R&D. What is strikingly new in recent policies is the government’s active courting of foreign firms. In the past, Taiwan has not been unwelcoming to foreign firms, but neither has Taiwan courted MNCs with incentive packages to set up operations in Taiwan the way Singapore has (Fuller 2002). Amsden and Chu (2003) claim that the Taiwanese government worked to drive foreign firms out of Taiwan, but their argument depends on reading the decline of foreign electronics manufacturing in Taiwan as state driven. They ignore the state’s encouragement of foreign involvement in the two state-created flagship enterprises of Taiwan’s high-technology sector: Philips has a large stake in TSMC, and foreign design firms have been buying into UMC. Taiwanese electronics industry policy better fits the technohybrid model (Keller and Samuels 2003, p. 12) of forging global technological links to boost domestic capabilities, or “globalization for nation-building” (Fuller 2002, p. 1). The Taiwanese government’s new strategy is to lure MNCs in the hope that they will propel R&D intensity and internationalization to help Taiwan compete. The response thus far has been encouraging. As of late 2003, thirty-nine local firms and nine foreign firms (six U.S., one Japanese, and two German firms) have agreed to set up R&D centers

MOVING ALONG THE ELECTRONICS VALUE CHAIN 161

and many are already operating. Even Hon Hai, which set up a large research base in Beijing, has reconsidered and will set up an R&D base in Taipei County. The overall activities of the foreign firms are summarized in Table 5.2. Interviews with two of the foreign R&D centers (October 2003) showed how the centers’ activities dovetailed with the design activities of the local firms. The foreign firms were trying to work with local systems suppliers to integrate the foreign firms’ product offerings into the designs made by local systems firms. The foreign firms are engaged in software, broadly defined to include middleware, firmware, and reference design. The software product either enables the firm’s core product or constitutes the firm’s core offering. Significantly, both firms were very active in working with local firms, one with WLAN equipment firms and the other with cellular phones firms. One of the interviewees emphasized that the reason to come to Taiwan was the concentration of systems design, even though much of the manufacturing has already moved to China. If Taiwan did not have this concentration of systems design activity, relocating the firm’s operations to the PRC would be the obvious choice because of the cost savings. The other interviewee claimed that the attraction of Taiwan was the concentration of WLAN equipment makers, though he emphasized that the greater Asia-Pacific market was also a big draw. These interviews suggest that Taiwan can attract MNCs’ design activities by retaining the design activities of the local systems houses in Taiwan even as manufacturing shifts to China. The government’s subsidies and tax breaks for these R&D centers, along with continued opportunities to work with ITRI, help to ensure that these design activities will remain in Taiwan. Furthermore, on the basis of this limited evidence, the interactions between the MNCs and local firms do appear to enhance local capabilities. One of the MNCs works with some local firms to develop the middleware it provides to cellular phone makers. The other MNC emphasizes joint development with the more designintensive local systems firms and attempts to help them understand the next generation of products and the long-term road map. Missed Targets The Taiwanese government’s industrial policy record is impressive, particularly in the electronics sector. However, the experience of recent

162 CHAPTER 5

Table 5.2 Taiwan’s MNC Research Centers Firm North American Computer Firm

Date established No date given

Research activity Research for notebooks, desktops, mobile handheld devices, and servers. Goal to have greater research interaction with Taiwanese suppliers of the firm.

Japanese IT Conglomerate

2002 for new product Research for information development; 2003 for applications and information IC design center technology products. Design of graphics chipsets and LSI ICs.

North American Telecommunications Firm

No date given

Switching gear

North American IT Conglomerate

2002

The firm will set up two research centers. One will concentrate on notebooks, desktops, servers, and mobile handheld devices. The other will be an “Asia LowCost Design Center.”

North American Software Firm

2003

The firm will work to help local service firms provide better service utilizing the firm’s software products.

European Avionics Firm

2002

The firm will do avionics technology certification and will design avionics core modules, such as the frequency modulated continuous wave (FMCW)/RF system and digital radio transceiver module.

European Chemical Firm

2002; 2003

This firm first established a “guiding manufacturing” research center to do research on the manufacturing process. The second center will do research on CVD (chemical vapor deposition) for opto-electronics, particularly Taiwan’s white laser and ultraviolet laser technology.

Source: This list does not include the two centers interviewed and mentioned in the paper. In order to better protect the request for anonymity, the names of the firms in the list have also been left out. This information was obtained from a local Taiwanese thinktank with close ties to the Ministry of Economic Affairs.

MOVING ALONG THE ELECTRONICS VALUE CHAIN 163

years suggests that industrial targeting should be limited the closer Taiwan gets to the technological frontier. The evidence from two large subsectors of electronics, DRAMs and AMLCDs, raise red warning flags. Taiwanese DRAM firms have yet to prove that they can generate technology and profitability, and the successful business model for DRAM, massive-scale economies built upon patient capital, is at odds with the impatient capital that characterizes the Taiwanese financial system (Fuller et al. 2003). In the mid-1990s, Taiwanese government officials and industry managers still thought that DRAM was necessary both for process technology learning reasons and as a critical product within PCs (IPC interviews), but the Taiwanese DRAM firms have never been able to innovate and escape their dependence on foreign technology providers (Fuller et al. 2003). After more than ten years of effort, the DRAM firms are still desperately dependent on alliances with foreign firms to receive the necessary technology. Therefore, it would seem that the state should continue its preferential treatment of foundries but discontinue preferential treatment for DRAM firms. A more pressing concern is the recent entry into AMLCDs. Most of the AMLCD firms are relatively new, but profitability and sustainability remain questionable, even though the basic AMLCD technology has already been transferred from Japan to Taiwan. Taiwanese firms have successfully entered this market, if control of the basic proprietary technology is the yardstick of success. At this point only one firm appears to be consistently profitable, AU Optronics. This sector demands large capital investments and thus far the returns on this investment have been very small. The components manufacturers for AMLCDs, such as manufacturers of color filters, enjoy higher returns (D. Toh, “Flat Panel Displays,” Global Equity Research Asia, Lehman Brothers, May 30, 2003) though it is questionable whether these sectors would have flourished without entering into actual panel production. Now that seven Taiwanese firms have made large-scale investments in AMLCD production, it is hard to pull the plug on the incentives for this industry. One prominent Taiwanese scholar of technology policy has argued that Taiwan should not have gone into this capital-intensive, low-margin sector, but now that the investments have been made, it is too late to reverse course (IPC interview). In any case, the government has no intention of turning back, as evidenced by the “Two Trillion, Twin Star” program of incentives, a prominent part of the government’s current six-year development plan. The two trillion refers to the goal that both the IC industry

164 CHAPTER 5

and the AMLCD industry, the “twin stars,” should generate over a trillion NT dollars in sales revenue by 2008. Conclusion Taiwan’s electronics industry continues to reflect the strengths and some of the weaknesses of the past even as it confronts the new challenges of China’s rise as a manufacturing center and potential as a technology center as well as a readjustment of government–business relations in Taiwan. The IC sector is growing stronger even as it reduces its reliance on the state and even as serious competition emerges in China. Despite the misgivings and the foot-dragging of the Taiwanese government, the IC sector has learned how to use China for its own benefit. At the same time, the systems manufacturers remain stuck in a less desirable position in which they sell manufacturing and design services to international branded firms that reap most of the economic value. This predicament persists, despite growing intrafirm design competencies and the efforts of the government to help the sector expand into new product areas, particularly wireless. The best hope for this sector is that continued investment in R&D by both public and private sectors will pave the way for greater rewards in the international division of labor. Based on the preceding analysis, the state might consider some of the following potentially promising directions for further upgrading. First, the state should rechannel tax benefits and subsidies away from underperforming sectors that currently are sucking up valuable resources that could be used elsewhere. The DRAM sector should have its stategranted privileges withdrawn and these firms should be encouraged to become foundries, either independently or under the control of existing foundries. Second, while the AMLCD sector is still young, the government should consider playing a role it has not yet played in the electronics sector, the role of the rationalizer. The state could provide financial incentives to the large and relatively healthy firms, such as AU Optronics and ChiMei, to take control of the weaker players in order to create the scale economies necessary to compete in this sector. Whether or not rationalization is undertaken, the AMLCD sector should be kept on a short leash financially, as this capital-intensive sector could drain massive amounts of public resources if supported indefinitely. The government should continue to provide incentives to both private and foreign R&D centers, but raise the threshold level of the technical criteria for

MOVING ALONG THE ELECTRONICS VALUE CHAIN 165

receiving subsidies as more centers are established in order to encourage further upgrading. Finally, the state should explore more ways of fostering successful brands in Taiwan. This area of traditional weakness may soon show progress due to efforts by the Taiwanese government to set up industrial design programs. Further efforts to develop the capabilities associated with branded end-products should be attempted, although the fact that there is currently only one internationally branded electronics firm from outside the triad of the EU, Japan, and North America (Samsung) suggests how daunting a challenge this may prove.

6 From NAFTA to China? Production Shifts and Their Implications for Taiwanese Firms Marcos Ancelovici and Sara Jane McCaffrey

Introduction At the end of the 1990s, Hong Ho Precision Textile’s long-term outlook for production at home looked bleak. Like many Taiwan-based manufacturing companies, the firm saw opportunity in Mexico’s low labor costs, duty-free access to the U.S. market, and proximity to the U.S. border. In 1999 Hong Ho launched an ambitious program to build textile and garment production facilities in the southeastern state of Yucatan, with plans to invest US$60 million and hire 8,000 Mexican workers (EIU 1999, Bow 1999). Early returns were promising, and demand from U.S. customers strong. At the same time that Hong Ho announced a US$10 million investment in a Mainland Chinese textile factory, the firm continued to plan an expansion of production in Mexico (Taiwan Economic News 2003). But by April 2004, Hong Ho’s Mexico strategy had stalled: prices of NAFTA yarn proved too expensive and local subcontractors too scarce and unreliable. Hong Ho halted hiring at 1,500 workers and abandoned its circular knit operation (Just-Style 2004). Similarly, electronics manufacturers who once viewed Mexico as a land of opportunity also had second thoughts: Hong Kong–based VTech closed its Guadalajara plant, leaving more than 4,000 Mexicans jobless, and relocated production to Guangdong, China. Overall, some estimates claim that since 2001, Mexico has lost about 400,000 jobs to China (Luhnow 2004). Hong Ho, VTech, and the many investors that fed the maquiladora industry had bet on the development of the North American regional economy in the 1990s. In fact, during that decade, a consensus of business 166

FROM NAFTA TO CHINA?

167

managers and academics believed that NAFTA would be one of three important trading blocks in an increasingly regionalized world. Mexico’s experience in the past three years suggests a different reality: as one company after another has picked up stakes in Mexico and moved production out of the Americas, the assumption that trade within regions would dominate international exchange seems challenged. What happened to the importance of access to end markets and fast turn? Why are foreign multinational firms relocating parts of their production from Mexico to China? What implications does this production shift have for Taiwanese firms? The commonly held assumption that regional trading blocks would dominate international trade seemed to dictate a particular locational strategy: Taiwanese firms hoping to compete in lucrative U.S. or European markets would need a production presence within NAFTA countries or the European Union (EU). Once NAFTA went into effect in 1994, Mexico served as the platform from which North American, European, and Asian multinational firms competed for a share of the U.S. consumer market. Manufacturers were drawn by Mexico’s low labor costs, proximity to U.S. borders, and duty-free access to U.S. markets. In addition, companies that supplied large multinationals in other regions were often cajoled by their lead firms to co-locate in Mexico. As the Hong Ho example illustrates, Mexico proved to be a difficult production environment for Taiwanese firms. Many managers underestimated cultural barriers and were dismayed by wage inflation. Others complained of basic deficiencies in workers’ skills and in infrastructure. But did Taiwanese firms really need to invest in Mexico in order to access the U.S. market and be successful? Scholars upholding the idea of a world divided into regional trade blocks would answer affirmatively. However, as we show in this chapter, since 2000 most leading multinationals in electronics and textile and apparel have moved part of their operations out of Mexico, and very often to China. Although some of these firms have retained operations in Mexico and tried to upgrade them, lower-end mass production seems to be well on its way to China. Chinese producers have even gained U.S. market share in television sets—bulky, difficult-to-transport products that have long been a stronghold of Tijuana, Mexico (see Figure 6.2 on p. 179). In fact, the deregionalization trend extends beyond North America: firms recently established in Eastern Europe have also considered relocating part of their operations to China.

168 CHAPTER 6

In this chapter, we assess the implications of the recent production shift from Mexico to Asia for theories of regionalization and globalization. We draw from several data sources, including U.S. import data, trade publications and other secondary literature, and our own interviews with firm managers. First, we discuss the assumptions behind the regionalization and globalization theses of the 1990s. Second, we lay out evidence of the significant transfer of production from Mexico to China that contradicts the predictions of regionalization in two important industries that are considered emblematic of globalization: textile and apparel and electronics. Textile and apparel goods contain very high labor content, which pushes production toward lower-wage locations, but are also regulated by a complex and highly protective international trade regime. Electronics production, in contrast, is more capital intensive but is also highly standardized. Standardization has facilitated an intensive process of deverticalization in the industry. Because the possibility of shifting operations to China is higher in textile/apparel and electronics production, these two sectors are critical cases. Third, we explore the importance of such a production shift for theories of regionalization and globalization. We find that proponents of regional economies have often overestimated the advantages of privileged market access and just-in-time delivery, and underestimated the critical importance of capable local suppliers. Finally, we suggest a few conditions under which proximity, which may be a temporary advantage, can be leveraged into industrial upgrading, and discuss the impact of the recent production shift for multinational producers. Regionalization, Globalization, and NAFTA Though both regionalization and globalization suggest the growing importance of cross-border trade, scholars have reached no consensus on the relationship between the two phenomena. Some authors see regionalization and regional trade blocs as part of globalization, insofar as they contribute to the erosion of national borders and may work as a stepping-stone to global free trade (Wei and Frankel 1996). Others, in contrast, present regionalization as a defensive reaction to globalization, “an effort to regain some measure of political control over the processes of globalization that have impinged on national policy” (Katzenstein 2003, p. 105). A third group of scholars stresses that regionalization is neither a first pass at globalization nor a reaction to it,

FROM NAFTA TO CHINA?

169

but a distinct process. Much empirical work emphasizes that the bulk of international trade takes place within regions, particularly North America, Europe, and Asia (Zysman 1996a). In this chapter, we are interested not so much in explaining the origins of regionalization—which all authors agree are often political—as in discussing what fosters its demise or transformation. Although the world economy is still structured around trade blocs, the production shift from Mexico to China challenges the notion that regionalization is, and will remain, the driver behind the restructuring of international production and distribution networks. Regional producers have an obvious and long-standing advantage over far-flung competitors in transportation costs, which is particularly important for makers of bulky or delicate goods. But scholars have cited two explanations for the recent growth and consolidation of regional economic integration. The first, rooted in politics, emphasized the privileged access to end markets. Free trade areas, like NAFTA, and common markets, like the EU, are designed to favor members by eliminating costly barriers to trade, including tariffs. Thus, NAFTA favored regional sourcing and boosted Mexican exports to the United States. As Mexican producers’ share of the U.S. market spiked in the late 1990s, several observers called for increased attention to the regional production network: “The emergence of an integrated North American regional economy in the latter half of the 1990s constitutes a qualitative change in the dynamics of the apparel industry that requires new forms of analysis” (Gereffi et al. 2002, p. 7). A second argument for regional organization of production hinges on strategies for the distribution of goods. According to Abernathy et al. (2004), the advent of “lean retailing” in the late 1980s changed the way clothes and other goods are sold in the United States. Rather than placing large orders for merchandise at the beginning of each season, retailers began to hold small inventories, track sales with bar code scanning, and require quick replenishment of only those goods that attracted customers. An ever-increasing range of products enhances customers’ choices, but also multiplies the cost and difficulty of maintaining inventories. As a result, Abernathy et al. argue, supplier responsiveness is central to lean retailing, as lead firms demand the supply of more and more products on a replenishment basis (2004, p. 25). Being close by, as Mexican exporters are to the U.S. border, facilitates quick delivery of re-orders. The fashion component of textile and apparel goods—where much of the value in some segments lies—depends on fast product cycles

170 CHAPTER 6

and quick time to market. In such cases, large retailers and brand-named merchandisers coordinate the functions that constitute the commodity chain across countries. Rapid innovation in some electronic product segments, like PCs, also requires quick response to market demands. In other words, the slow boat from China may arrive with merchandise that is already out of style or obsolete. With quick response a key determinant of success in the higher-value segments of the industry, Abernathy et al. (2004, p. 43) suggest, the proximity of regional producers to rich country ports is an important and enduring advantage: Greater regionalization of textile and apparel production is a natural outgrowth of the competitive forces [created by lean retailing]. . . . Instead of a single international market for apparel and textiles, three regionally based models anchored in the United States, Europe, and Japan may better reflect the realities of post-2005 globalization.

Though these scholars concentrate on textile/apparel, they believe “similar developments can be expected to emerge across other consumer product industries where replenishment is of growing importance” (Abernathy et al. 2004, pp. 44–45). One such industry is electronics: products can be fragile and/or bulky, holding large inventories is expensive, and customizing products is an increasing trend. In fact, as Chase (2003, p. 46) suggests, geography may “naturally” structure production-sharing within regions: Multinationals tend to focus on region-specific sourcing, manufacturing, and marketing because often there are diseconomies of scale in global production networks. Locations closer to the corporate parent offer a number of benefits: lower transport costs in the movement of goods between home and host countries; easier coordination with components suppliers for firms that maintain low inventories and rely on just-in-time delivery systems; and shorter lead times when firms must adapt quickly to changes in demand or consumer tastes. As a result, production sharing primarily crosses borders between neighboring (or nearby) countries.

Regional producers, according to these theories of regionalization, should beat their distant competitors on tariff/nontariff barrier costs, transportation costs, and time to market. Based on these assumptions, arguments for the continued prevalence of regional economic integration

FROM NAFTA TO CHINA?

171

predict a continued preference of U.S. buyers for suppliers located in Mexico, particularly in higher-value-added product markets. In the early to mid-1990s, business managers throughout the world found the regionalization logic compelling. Just as U.S. firms moved production south of the border to take advantage of low wages, Asian and European firms built factories in Mexico to benefit from NAFTA countries’ privileged access to the U.S. market. Gaining this edge was particularly important to firms in Taiwan. Taiwanese industry, which had been built on low-wage manufacturing, was under threat from lower-cost competition on the Mainland and in Southeast Asia. Capitalrich Taiwanese firms invested in Mexican production sites to skirt tariffs and be nearer to end markets. In fact, Taiwan has been a major source of foreign direct investment in Mexico throughout the NAFTA period (see Table 6.1). For example, in dollar terms, Taiwan was the third-largest provider of foreign direct investment (FDI) in Mexico’s textile and apparel sector between 1999 and 2002 (Ferreira 2003, p. 20; see also Table 6.2). Some Taiwanese investors moved to Mexico on their own, attracted by the promise of government subsidies as well as NAFTA markets. Other producers, whose livelihoods depended on their relationships with brand-name multinationals, were encouraged to invest by their customers who had already moved some production into the region.1 Did the promise of NAFTA live up to investors’ expectations? In the next sections, we will track developments in the two industrial sectors expected to benefit most from NAFTA: electronics and textile and apparel. The Electronics Industry in Mexico Up to the 1970s, Mexico had a large and strong consumer electronics industry: “In 1965, an observer of the global economy easily could have concluded that Mexico, and not Taiwan or South Korea, would become one of the most successful developing countries in establishing a competitive, indigenous electronics and related parts industry” (Lowe and Kenney 1999, p. 1428). Mexico failed to sustain its industry and still suffers from structural limitations that other developing countries like Taiwan were able to surpass. Nonetheless, during the last ten years the electronics industry in Mexico has been going through an extraordinary wave of expansion and transformation.

172 CHAPTER 6

Table 6.1 Taiwanese FDI in Mexico, 1995–2002 (in US$ million) 1995 0.6

1996 2.6

1997 7.2

1998 27.2

1999 19.8

2000 6.6

2001 21.6

2002 6.6

Source: OECD (www.oecd.org), November 8, 2004. Table 6.2 Foreign Direct Investment in the Maquiladora Sector Textile and Apparel Industry, 1999–2002 (in thousands of dollars) (selected countries)

Country 1999 United States 249,725 South Korea 6,348 China 626 Taiwan 16,755 Total 327,888

2000 301,193 –13,396 971 4,459 359,461

2001 2002 117,196 124,541 7,937 2,002 97 231 8.883 2,947 172,594 141,095

% Share of total, 1999–2002 1999–2002 794,655 79.4 2,891 0.3 1,925 0.2 33,045 3.3 1,001,039 100

Source: Secretaría de Economía data, published in Ferreira (2003), p. 29. Used with permission of Textiles Intelligence Ltd. www.textilesintelligence.com.

NAFTA and the Transformation of the Electronics Industry The Mexican electronics industry was shaped by two waves of investment and development. The first, led mainly by U.S. original equipment manufacturers (OEMs) such as IBM, Hewlett-Packard, and Motorola, began in the 1970s. A second wave began in the mid-1990s, after NAFTA came into effect in January 1994. This time, the firms investing were U.S. and Asian OEMs, like Acer, as well as an increasing number of electronics contract manufacturers (ECMs),2 such as Solectron, Flextronics, and Sanmina-SCI. These firms are located primarily in four states: the border states of Baja California (Tijuana), Chihuahua (Ciudad Juarez), and Tamaulipas (Reynosa); and in the western state of Jalisco (Guadalajara). This second wave of investment had a dramatic effect on the Mexican electronics industry. The number of people employed shot up, from 192,000 in 1994 to 382,000 in 2000—a figure that constitutes 9.3 percent of the total Mexican workforce employed in manufacturing industries (see Table 6.3). Similarly, annual FDI grew from US$257 million

FROM NAFTA TO CHINA?

173

Table 6.3 Mexican Electronics Industry, 1994–2001

1994 1995 1996 1997 1998 1999 2000 2001

Employment (thousands)

FDI (U.S.$ millions)

Exports’ value (U.S.$ millions)

Productivity index (1993 = 100)

192 197 226 271 305 331 382 360

257 573 571 655 661 1,535 821 271

10,791 13,181 21,097 26,404 30,464 36,548 46,248 42,978

111.3 106.8 114.2 125.7 132.8 134.2 139.5 N/A

Source: Secretaría de Economía, “Programa para la Competitividad de la Industria Electrónica y de Alta Tecnología,” 2002.

in 1994 to US$821 million in 2000. At the same time the total value of electronics exports increased from US$10,791 million in 1994 to US$46,248 million in 2000; productivity grew by 39.5 percent. Though the first wave of investment was essentially motivated by low wages, this second wave was also driven by increased access to the U.S. market as a result of NAFTA and of transformations in the organization of production that made geographical proximity even more important than in the past. Indeed, NAFTA brought tariffs for Mexican exports to the United States down from 3.94 percent to 0.27 percent. Producing in and exporting from Mexico became a way for Asian and European producers to bypass U.S. trade barriers or, in the case of U.S. companies, to lower production costs to face Asian competition. Thus, Mexico became “a space where American and Asian electronics industries increasingly compete[d] for a share of the U.S. market” (Dussel Peters 2000, p. 76). This pivotal role allowed Mexico to compete with Asian countries in the battle for the location of production and assembly processes. In addition to NAFTA, the 1994 peso devaluation nearly halved production costs and thereby made Mexico all the more attractive as a place to relocate production (Lindquist 2000). More important in the long run, changes in management strategies and the organization of production that started in the late 1980s had a significant impact on investment patterns. “Just-in-time” delivery forced firms to move manufacturing closer to end markets in order to meet rapid demand changes and turnarounds. For example, in 1999, IBM’s

174 CHAPTER 6

mobile computing division moved its notebook assembly operations out of Asia and decided to increase its Guadalajara operations. By 2000, 65 percent of all IBM portable PCs were assembled in the capital city of the state of Jalisco. The production of flat-panel monitors underwent a similar trend. In 1999, IBM pressured its suppliers to set up operations in North America. Ken Czarnecki, director of OEM operations for the IBM group, explained that “the transportation cost and time delay in shipping panels back to Asia for repair or alteration was becoming unacceptable. We needed much faster turnaround” (quoted in Robertson 1999). Turnaround time is why most repair centers are actually not even in Mexico but in the United States. Thus, in 1999 Dell moved its assembly operations for its consumer notebooks from Asian ECMs back to the United States. A spokesperson for Dell explained that the firm needed “much quicker turn-around in notebook assembly to cut inventory, shorten delivery times, and react faster to market changes” (quoted in Robertson 1999). The industry also went through a deep reorganization. Indeed, digitization and standardization of interfaces between business functions facilitated the modularization of production, the fragmentation of the value chain, and the disintegration of vertically integrated firms. When business functions no longer needed to be located in the same place, or in the same company, cross-national production networks emerged. Different value chain functions began to be executed across national boundaries. As a result, brand-name electronics firms increasingly outsourced their manufacturing operations to ECMs in order to concentrate on “core competencies,” such as product definition, design, and marketing, and to track (and respond to) market changes (Borrus and Zysman 1997, Sturgeon 1998, 2002, Lüthje 2002). This shift to modular and cross-national production networks created a new market for ECMs. U.S. brand-name firms (e.g., Dell, HP, IBM, and Apple) and their complementary ECMs regained dominance of the electronics industry (see Table 6.4). In response to the demands of brandname firms, ECMs developed a global presence in the 1990s. The top five ECMs (Solectron, Flextronics International, Sanmina-SCI, Celestica, and Jabil Circuit) grew at an average annual rate of 43 percent per year between 1995 and 2003, essentially through the acquisition of competitors and entire plants from OEMs as well as the expansion of existing manufacturing facilities throughout the world (Sturgeon 2002, Lüthje 2002). Most firms that invested in Mexican manufacturing plants during

FROM NAFTA TO CHINA?

175

Table 6.4 Top PC Vendors and ECMs in 2002

1 2 3 4 5

Top five PC vendors in the United States Dell Hewlett-Packard Gateway IBM Apple

Top five ECMs in the world Solectron Flextronics Sanmina-SCI Celestica Jabil Circuit

Sources: Olenick 2002, Sturgeon 2002.

the 1990s were leading North American–based ECMs that already had a strong presence in Asia, while most Asian ECMs concentrated on Chinese locations. In Mexico in 1995, the leading U.S. firm, Solectron, was the first ECM to build a plant in Guadalajara, Jalisco. In 1998, Lucent Technology sold its Monterrey manufacturing facility to Toronto-based Celestica, an ECM that had spun off from IBM in 1996. Before it was acquired by Solectron in 2000, NatSteel (then a Singapore state-owned firm) had also made huge investments in Guadalajara and set up several manufacturing facilities. With its concentration of electronics firms, Guadalajara came to be known as the “Silicon Valley of Latin America.” In 1999 the region produced more than 60 percent of all computers manufactured in Mexico. In 2002 the Guadalajara electronics cluster included thirteen OEMs (such as IBM, HP, Intel, Kodak, Siemens, and Hitachi), thirteen ECMs (such as Solectron, Sanmina-SCI, Flextronics, and Jabil Circuit), one original design manufacturer (ODM), sixteen design centers, and more than 380 specialized suppliers (CADELEC 2003).3 Thus, at the end of the 1990s, one could say that NAFTA and changes in the organization of production had deepened the economic integration of North America. Therefore, theories that claimed globalization would entail the consolidation of regional trade blocs (North America, Europe, and Asia), rather than the emergence of an integrated, single world market, appeared to be founded. Moreover, a similar trend seemed to take place in Europe, as the same largest ECMs opened facilities in Eastern Europe at the same time.4 However, since 2000–2001, a new trend is emerging. Market access no longer seems to be a crucial factor in location decisions. At the same time, the meaning of geographical proximity is changing, as being close

176 CHAPTER 6

Table 6.5 Largest Dollar Amounts of U.S. High-Tech Imports by Country of Origin, 2000–2002 2000 Country of origin 1 2 3 4 5 6 7 8 9 10

2002

Imports (U.S.$ billions)

Japan Mexico EU China Taiwan South Korea Canada Malaysia Singapore Philippines

48 37 28 26 22 21 21 20 15 9

Total U.S. HighTech Imports

Country of origin

China Mexico Japan EU Malaysia Taiwan South Korea Singapore Canada Philippines

$271.346b

2000–2002

Imports (U.S.$ billions) 35 34 29 26 19 16 15 11 10 7

Country of origin China Mexico Japan EU Malaysia Taiwan South Korea Singapore Canada Philippines

$219.975b

% +34.6 –8.1 –39.6 –7.1 –5.0 –27.3 –28.6 –26.7 –52.4 –22.2

–19

Source: American Electronics Association (AEA), “Tech Trade Update 2003,” www.aeanet.org. Note: Percentages are based on our calculations. Other leading countries include Brazil, Costa Rica, Hong Kong, Hungary, Indonesia, Israel, Switzerland, and Thailand.

to key suppliers seems to have eclipsed being close to end-user markets. While Mexico struggles to adapt, China has emerged as Mexico’s main challenger for U.S. market share. The Rise of China Between 2000 and 2002, more than 500 maquiladoras closed throughout Mexico. In 2001 alone, 250,000 jobs were lost and 253 factories shut down (Moody 2002). A range of macroeconomic indicators—employment, FDI, export value—clearly show a decline after 2000 (see Table 6.3 on p. 173). A priori, this slowdown in economic activity could be the result of the U.S. recession, given that 85 percent of total Mexican electronics exports typically go to the United States. Indeed, total U.S. high-tech imports decreased by 19 percent between 2000 and 2002, from US$271.3 billion to US$219.9 billion (see Table 6.5).5 In 2003, moreover, the Guadalajara plants of the leading ECMs were operating at only 60 percent

FROM NAFTA TO CHINA?

177

capacity (Smith 2003). However, macroeconomic data indicate not only that U.S. high-tech imports are declining and that Mexico’s electronics industry is experiencing a slowdown as a result, but also that China’s share of the U.S. market keeps increasing. In 2000, U.S. high-tech imports from Mexico and China represented, respectively, US$37 billion and US$26 billion. In 2002, Mexico’s share had fallen to US$34 billion, whereas China’s had increased to US$35 billion. If the U.S. recession were the sole cause of Mexico’s woes, China’s exports would have been similarly battered. The largest segment of U.S. high-tech imports is made up of computers and computer parts. While computer imports did not fluctuate significantly, computer part imports (such as motherboards) from Mexico fell after 2001, while those from China increased dramatically (see Figure 6.1). Moreover, Sturgeon and Lee (see chapter 2 in this volume) point out the growing market share of PC-related contract manufacturing by Taiwanese original design manufacturers (ODMs) that have almost all their production in China. On the one hand, this variation suggests that the production and assembly of bulkier commodities like computers remains in Mexico to limit transportation costs and delivery time, while the production of parts easy to ship is now primarily done in China. On the other hand, however, bulky commodities like TVs are increasingly produced in China even though Mexico still dominates the U.S. TV market (see Figure 6.2). These macro data help to demonstrate the growing competitiveness and market shares of firms located in China, but they do not tell us whether these firms used to be located in Mexico in the past. In other words, these data do not demonstrate that a shift is taking place from Mexico to China and that the North American region as a trade and production bloc is being transformed. A brief look at firms that recently relocated some of their operations out of Mexico provides a useful complement (see Table 6.6). The Textile and Apparel Industry in Mexico Garment assembly is a labor-intensive process, and as a result, the textile and apparel sector seems particularly vulnerable to relocation to low-wage environments. But the fashion component of many apparel products, together with the advent of lean retailing, led many observers to argue for the continued regionalization of textile and apparel chains,

178 CHAPTER 6

Table 6.6 Electronics Firms Moving Out of Mexico Firm Flextronics

Investment PCB plant

Future Closed in 2003, shifted low-skilled production to China.

Emerson Electric

Factory in Monterrey

Closed in 2002.

Motorola

Semiconductor plant in Guadalajara

Closed in 1999.

Hewlett-Packard and Lucent Technologies

Long-time, very big investors in Guadalajara

Moved to China.

Royal Philips Electronics

Computer monitor factory in Ciudad Juarez

Moved to China in 2002. Loss of 900 Mexican jobs.

Canon

Inkjet primer factory

Moved to Thailand and Vietnam in March 2002. Loss of 700 Mexican jobs.

Sanyo

Video components

Moved to China and Indonesia in 2001. Loss of 1,884 Mexican jobs.

VTech

Circuit boards plant

Moved to China in 2001–2. Loss of 6,952 Mexican jobs.

Matsushita

Electronics plant

Seeking to relocate all or part of its operations.

Pioneer

Speaker factory in Tijuana Transferred part of operations to Shanghai in 2002.

Source: Compiled by the authors from media reports.

despite competition from much-lower-wage producers overseas. In the 1980s, as competition from low-wage producers in Asia hit the U.S. textile and apparel industry hard, U.S. managers seized on this regional production model as a survival strategy. U.S. textile managers believed that low-wage Mexican garment workers and highly sophisticated U.S. textile mills would be natural partners. A regional production and sourcing strategy was the industry’s future: in fact, NAFTA promised a “renaissance for the textile industry,” argued the former chairman of Guilford Mills in 1993 (McNamara 1993).

FROM NAFTA TO CHINA?

179

Figure 6.1 U.S. Imports of Computers and Computer Parts, Mexico vs. China $18,000,000 Computers (Mexico)

$16,000,000 $14,000,000

Computers (China)

$12,000,000 $10,000,000

Computer accessories, peripherals and parts (Mexico)

$8,000,000 $6,000,000 $4,000,000

Computer accessories, peripherals and parts (China)

$2,000,000 $0 1998

1999

2000

2001

2002

2003

Source: U.S. Census Bureau, Foreign Trade Statistics, www.census.gov/foreign-trade/ www, accessed November 2004.

Figure 6.2 U.S. Imports of Televisions and Radios, Mexico vs. China $9,000,000

Television Te levi si onreceivers, re ce iver s, VCRs video vc rs &&oother ther video equip. (Mexico)

$8,000,000 $7,000,000

Television Te levi si onreceivers, re ce iver s, VCRs video vc rs &&oother ther video equip. (China)

$6,000,000 $5,000,000

Radios, Radi os , phonographs, pho nogra phs tape , t ap e decks, and other dec ks, and ot he r stereo equip. (Mexico)

$4,000,000 $3,000,000 $2,000,000

Radios, Radi os , phonographs, pho nogra phs tape , t ap e decks, and other dec ks, and ot he r stereo equip. (China)

$1,000,000 $0 1998

1999

2000

2001

2002

2003

Source: Same as Figure 6.1.

Indeed, NAFTA brought a dramatic increase in Mexican garment production and exports. By 1995, Mexico surpassed China as the largest exporter of garments to the United States, and the number of garment producers in the maquiladora sector grew from 398 in 1994 to 1,088 in 2000. As apparel production in Mexico boomed, enthusiasm for NAFTA spurred many U.S. textile firms to invest in warehouses in Mexico, and as the decade wore on, many committed serious resources to building

180 CHAPTER 6

production facilities in the country as well. Of U.S. textile firms competing in the apparel segment, seven of the top ten companies built or invested in Mexican production facilities.6 The most ambitious investments, like those planned by Burlington, Guilford Mills, and Delta Woodside, aimed to integrate textile production and garment assembly in one firm, leveraging U.S. management, marketing, and sales talent with low-cost labor in Mexico. These firms intended to offer customers full-package production services, a “onestop-shop” for apparel brands like Liz Claiborne, Ralph Lauren, or Gap. A vertical strategy would allow the textile firms to ensure a market for their fabrics, and capture a greater share of the end product’s value added, they hoped. Unfortunately for the U.S. textile managers who bet on a regional economy and for the Mexican workers they employed, NAFTA’s early promise petered out by the end of the decade. U.S. imports of apparel made in Mexico have declined since 2000, at the same time as the number and value of garments China sends to the United States continue to grow rapidly. And Mexico is not the only recent loser—several other lower-wage producers in the Caribbean Basin region, where apparel firms enjoy proximity and favorable access to U.S. markets, have also experienced slowed or negative growth (see Figures 6.3–6.5). This downturn in intraregional imports is quite recent, and follows the 1997 depreciation of several Asian currencies, a recession in the United States in 2001, and a worldwide dip in trade after the September 11 terrorist attacks. However, the downturn in the Mexican garment sector may signal a structural change rather than a temporary development. First of all, the regional strategies of several lead firms seem to be breaking down: Burlington, Cone Mills, Dan River, and Galey and Lord have all closed or scaled back their investments in Mexico since 2001 (see Table 6.7). Critics charged that U.S. textile mills, with limited knowledge of apparel markets and even less understanding of garment construction, lacked the retail contacts and the production know-how to jump into the clothing business. However, even firms with solid experience in garment production failed to make a go of a vertical/regional production strategy.7 Also, the new competitors of U.S. textile firms have not been from within the region. Textile production in Mexico has improved in the past decade, as foreign sourcing for apparel inputs dropped from 98.2 percent in 1994 to 92.0 percent in 2002, but the expansion of domestic produc-

FROM NAFTA TO CHINA?

181

Figure 6.3 Growth Rates of Apparel and Household Good Imports to the United States from Mexico and Selected CBI Countries

15% 10% Mexico 5%

Dominican Republic Guatemala

0%

Honduras -5% •10% 1999

2000

2001

2002

Source: Office of Textiles and Apparel, U.S. Department of Commerce, “Trade Data— U.S. Imports and and Exports of Textiles and Apparel,” http://otexa.ita.doc.gov/ msrpoint.htm, accessed November 2004. Figure 6.4 Apparel and Household Textile Goods Imports to the United States, Mexico vs. China $14,000,000 $12,000,000 $10,000,000 $8,000,000 (Mexico) $6,000,000 (China) $4,000,000 $2,000,000 $0 1998

1999

2000

2001

2002

2003

Source: Same as Figure 6.1.

tion in Mexico is relatively modest and much is dependent on foreign investment (Ferreira 2003). In addition, all foreign investors’ commitments to the Mexican textile and apparel industry have declined rapidly since 2000 (see Table 6.2). Since January 2001, 325 of 1,122 clothing maquiladoras have shut down (Forero 2003).

182 CHAPTER 6

Figure 6.5 Annual Growth Rate in U.S. Imports of Apparel and Household Goods, China vs. Mexico 15% 12.72%

13.77% 11.80% 10.44%

10%

5,78% 4.27%

5%

China 0%

Mexico

-5%

-4.15%

-8.42%

-10% 1999

2000

2001

2002

Source: Same as Figure 6.3. Table 6.7 Textile Firms Leaving Mexico Firm Burlington

Investment Fabric plant, slacks factory, denim mill

Future? Closed 2001. Jobs lost: 1,100.

Galey and Lord

Purchased Dimmit Industries, a Mexican manufacturer of shorts and trousers; set up two textile plants

Closed six Dimmit sewing plants and one textile facility. Jobs lost: 3,500.

Dan River

Joint venture with Mexican apparel firm; planned to invest/send 275 looms

Withdrew February 26, 2001.

Guilford Mills

US$30 million apparel factory closed after 11 months

Announced closing on March 6, 2002; company plans to reinvent as a U.S.-based auto supplier.

Cone Mills

Large denim plant in Parras, Mexico

Needs US$18 million in financing.

Proponents of regional apparel markets, like Abernathy et al., argued that shifts in sourcing apparel to Mexico and Caribbean Basin Initiative (CBI) countries would be “very beneficial” for U.S. textile

FROM NAFTA TO CHINA?

183

interests and a potential boon to Mexican fabric producers (2004, p. 41). In fact, a regional apparel sourcing strategy depends on a healthy and competitive textile industry close to garment assemblers. The current crisis in the textile industry is evidence against the regionalism thesis. With fewer local or regional suppliers for fabric, garment factories in Mexico and the Caribbean maintained large stocks of inputs in order to pursue a quick-turn strategy, but this was expensive and risky. Importing fabric from outside the region took time and required garment assemblers to pay tariffs on the non-NAFTA textiles, thus undercutting one of the main advantages of producing apparel in the region. Do lower labor costs in Asia wipe out the tariff advantages of producing within NAFTA? According to Abernathy et al., retailers should still favor suppliers who can consistently deliver reorders quickly (2004). However, even these findings suggest that many Mexican producers have yet to reach a high level of reliability. Abernathy et al. report that lead times from Mexico can be as short as four or as long as nine weeks; lead times of Chinese factories range from seven to sixteen weeks or more. The best-managed Chinese factories can deliver goods to U.S. distributors faster than the worst Mexican factories. Air freight is far more expensive than shipping over land or sea, but China’s lower wage rates could compensate for the higher transportation costs (Watkins 2002). If, with lower labor costs and high productivity, Chinese producers can beat Mexican garment assemblers on price and almost match delivery times, retailers in search of quick reorders may look to the better-managed factory, rather than the closer one. Falling air freight costs make far-flung suppliers even more attractive. As the director of sourcing for one major U.S. apparel firm said, “If I do it right, I can bring it in by air from Asia, and I can have goods air-freighted from Asia at lower cost than trucked in from Mexico.”8 As shown in Table 6.8, this executive expects an increase in this trend. Sourcing managers’ preference for Chinese suppliers has also been influenced by fashion. In the mid-1990s, one manager said, Mexico built its reputation on replenishment, free of the unpredictability of quota, and the ability to produce large runs efficiently. In 2001, the apparel market moved away from commodity products and toward merchandise with a higher fashion content. Instead of 20,000 units of the same style, customers wanted 5,000 one way, another 5,000 cut differently, and so

184 CHAPTER 6

Table 6.8 Shipping Trends at One Large U.S. Apparel Firm

1998 2003 Future

% of merchandise transported by sea/land 88 80 70

% of merchandise transported by air 12 20 30

Source: Sourcing manager’s interview with the author, November 2003.

on. “We told our facilities there that we need much shorter runs. They included some of the biggest and best Mexican companies and they couldn’t manage the change,” said one manager. Though the Mexican plants had talented upper management, good workers, and new equipment, a lack of good midlevel management left them unable to provide consistent delivery of complicated orders.9 As Mexican plants lost orders, they were “grabbed up by Asia,” the manager said. A Shift to China? As Mexico’s main advantages as a textile and apparel producer, proximity and privileged market access, seem to be eroding, China’s competitiveness in a global garment market may be gaining steam. In 2002, China joined the WTO and took back its title as top garment exporter to the United States. According to one textile analyst, China’s advantages over Mexico include: a more skilled workforce; lower electricity and water costs; lower corporate income taxes; higher availability of cheap raw materials, particularly cotton; a more diverse and well-developed network of suppliers; and a higher degree of vertical integration in the industry. (Ferreira 2003, p. 34)

Such advantages should position China well to capitalize on the removal, as of December 31, 2004, of the import quotas that remain on almost 80 percent of textile and apparel products. In fact, many managers in the industry have become so convinced that China will be able to beat competitors after deregulation that they began to increase sourcing

FROM NAFTA TO CHINA?

185

in China as early as 2001 and 2002, in order to establish relationships with Chinese manufacturers in anticipation of a ramp-up of orders as of 2005.10 The move toward sourcing garments in Asia has already impacted U.S. textile companies. As the manager of a woolen mill (who closed his facility this year and shipped his machines to China) points out, textile sourcing follows sewing.11 Despite the attractions and prospect of China as a garment supplier, textile and apparel production within NAFTA and the Caribbean Basin will not disappear. Large retailers do not wish to become dependent on suppliers in one country, or even one region. The outbreak of SARS provides a case in point: several garment managers shifted some orders from Asia to Mexico and the CBI during the height of the 2003 health crisis in China and Hong Kong.12 Just as garment districts in New York and Los Angeles counties continue to employ upwards of 22,000 and 72,000 production workers (respectively) despite high wage and real estate costs (U.S. Bureau of Labor Statistics 2002), nimble producers with lower costs and privileged access to U.S. markets will find their niche. Only a small amount of the production in New York requires highskilled tailoring; most of the sewing is quick-turn production for reorders of midpriced merchandise from chain stores and even discount outlets.13 Some Mexican producers plan to move up-market to find a niche with high-end U.S. retailers. Others will search out small-batch, rush orders (Forero 2003). Despite the bleak overall picture for U.S. sewing plants, these strategies have worked for some garment manufacturers in the United States. Plants in Mexico could capture a greater share of reorders, but to do so will require a reputation for consistent management and reliable delivery. The Changing Nature of Regionalization Why are textile and apparel firms leaving Mexico? We need to understand the nature of this shift to China in order to identify its implications for regionalization. Analysts blame Mexico’s recent exporting woes on various culprits: the U.S. recession, rising maquiladora wages, an overvalued peso, and a poor climate for business in Mexico. We discuss each of these explanations below and then offer an alternative way to understand the recent transformation of regionalization. The U.S. recession is one obvious suspect to blame for maquiladora closings. According to Gerber and Carrillo, “a one percent decline in

186 CHAPTER 6

U.S. industrial production leads to a decline of just over 1.25 percent in maquiladora employment nation-wide. The effect, then, of the slowdown in U.S. industrial production is a eight to nine percent loss of jobs in the maquiladora industry, or about forty percent of the actual job loss” (2002, pp. 18–19). But the drop in U.S. consumer demand in 2001 did not affect all importers equally. The recession in the United States cannot explain why some firms moved to China in order to export for the U.S. market, nor why an increasing number of U.S. buyers looked beyond the region for suppliers, resulting in losses for Mexicanbased producers. In addition to the U.S. recession, many point to the rising wage rate in Mexico as a cause of the drop in exports. In 1995, the minimum wage was US$2.90 per day; by 2002, it had risen to US$4.25. Entry-level factory workers in maquiladoras make US$1.50–2.50 per hour, compared with US$0.25–1 per hour in China. Clearly, the Mexican production environment is not as “low wage” as it seemed at NAFTA’s inauguration. However, the skills and productivity of Mexican workers have improved rapidly during the NAFTA period. In electronics, producers have begun moving away from basic assembly and into midrange and higher-end products, and wages have moved up accordingly. In the textile and apparel sector, the annual value-added per worker hour rose from US$3.41 in 1994 to US$5.28 in 2002, a surge in productivity that could explain the surge in wages (statistics cited in Ferreira 2003, p. 21). Moreover, the wage differential between China and Mexico is only one factor in deciding where to locate production. As Bill Coker, director of sales and marketing for the Dallas-based ECM Elcoteq Americas, explained: If you just look at the labor side, labor is significant—about a 4-to-1 ratio in favor of China. But if you look at the amount of automation that goes into a small cell phone, there’s not a lot of hand assembly. So whether we have a [capital-intensive] surface-mount line set up in Mexico or in China, there’s not a lot of cost difference in terms of running those lines. (quoted in Maclellan 2003, p. 40)

In addition, theories of economic regionalization emphasize that the advantages of lower-wage production need to be weighed against the

FROM NAFTA TO CHINA?

187

risks of being far from markets and the costs of maintaining large inventories. According to Borrus and Zysman, cross-national production networks in electronics are more about “the emergence of locations that can deliver different mixes of technology and production at different cost-performance points” rather than about lower wages. “Countries represent particular mixes of capabilities and costs: if a country did not have capability and only had low labor cost, or had a particular capability at too high a cost, firms would seek alternate locations” (1997, p. 153). The same point can be made about textile and apparel: though China’s wages are lower than Mexico’s, they are 38 percent higher than Bangladesh’s, and 81 percent higher than Indonesia’s. Yet China’s garment industry has grown more rapidly than lower-wage competitors. In addition to rising wage costs, many commentators blame Mexico’s overvalued peso for the recent downturn in exports. With a too-strong currency, all of Mexico’s inputs are overpriced, and producers in the region are disadvantaged on world markets. Gerber and Carrillo point out that “While the peso has begun to fall very recently, over the past several years, it has increased about twenty percent in value against the U.S. dollar, and about thirty percent against East Asian currencies . . . the rising dollar cost of Mexican labor may explain another twentyfive–thirty percent of the employment loss” (2002, p. 19). True, Mexican exports surged after the devaluation of 1994 and overvaluation could help explain the shift away from Mexican imports since 2000. However, as David Birnbaum has pointed out, Mexican exports fell with the appreciation of the peso in 2001, but did not recover despite the peso’s slide after that year (from P9.3 = US$1 to P10.52 = US$1). Devaluation may not lead to an immediate improvement in sales, but should result in lower prices. “In fact,” Birnbaum argues, “the opposite appears to be the case. . . . Contrary to perceived wisdom, decreases in the value of the peso correspond closely to increases in the FOB prices as measured in U.S. dollars, while increases in the value of the peso are matched by dollar FOB price reductions” (2003, p.11). Business managers claim the Mexican regulatory environment caused the recent losses. Basic infrastructure in some areas is inadequate, and producers complain about high costs for electricity, gasoline, and security.14 Corporate taxes in Mexico average 34 percent, compared to a 10 to 15 percent rate in China. Though all these factors challenge business

188 CHAPTER 6

managers in Mexico, they have remained constant throughout the NAFTA period, even during the boom of the late 1990s. For the most part, the business climate in Mexico doesn’t help explain the recent shift of production out of the region. However, a few important regulations have changed since 2000, and may be impacting Mexico’s competitiveness in the region. As of January 2000, Mexican president Vicente Fox implemented new income and asset taxes to treat U.S. companies as if they had permanent establishments in Mexico. The ruling requires U.S. firms to pay Mexican income taxes on the share of their income derived from Mexico—a sum that is difficult to calculate—plus a 1.8 percent asset tax on machinery, equipment, and inventories. U.S. firms complain that the policy institutes a system of double taxation, as they pay corporate tax on the same earnings both in Mexico and the United States. A more onerous new burden on producers in Mexico pertains to tariffs on inputs. Under NAFTA’s Article 303, duty-free imports from nonNAFTA countries ended in 2001. This change has been particularly important for the electronics sector. Asian-owned plants in Baja California, for example, import three-quarters of all components from nonNAFTA countries, and fewer than 5 percent of all current maquiladora components are sourced from within Mexico. Because levying tariffs on inputs is highly disadvantageous to Mexican exporters, the government established the Program for Sectoral Promotion (PROSEC), which allows registered companies to import inputs at lowered tariff rates, usually from 0 to 5 percent (Ferreira 2003, p. 23). However, business owners have contended that the PROSEC application process is complicated and burdensome. Enforcing NAFTA, either through levying tariffs on inputs or excluding products with non-NAFTA inputs from duty-free export to the United States, has made Mexican producers less competitive and increased incentives to shift production to China—all the more so now that China is part of the WTO. A more powerful explanation for Mexico’s recent faltering in apparel and electronics centers on the country’s lack of domestic suppliers for these industries. Garment and electronic assemblers are in a weak position because of their limited local supply base. Multinational corporations, as either owners or customers of maquiladoras, have limited linkages to Mexican suppliers. As Dussel Peters argues, the fact that NAFTA has not encouraged an integrated supply chain within Mexico

FROM NAFTA TO CHINA?

189

is one of the main failings of the agreement. Most leading electronics firms never encouraged the development of local component and material suppliers and, even in the Guadalajara cluster, the large majority of components are imported (Dussel Peters 1999, 2000). In textile and apparel, U.S. attempts to establish textile production in Mexico faltered due to poor management and lack of commitment. A lack of embeddedness makes it easier for multinational corporations (MNCs) to leave, and difficult for Mexico to follow the “supplier-oriented industrial upgrading” model that many Asian countries adopted in the past (see Sturgeon and Lester 2004). Access to a competitive supply base underpins many location decisions. For example, as Leung, general manager of VTech’s contract manufacturing division, explained, VTech closed its 60,000-squarefoot plant in Guadalajara in 2001 while it expanded its operations in Guangdong, China, because in Mexico it “didn’t have enough infrastructure support and . . . had to get components from the United States or Hong Kong” (Serant 2002a). Similarly, Royal Philips Electronics shifted its production of PC monitors from Ciudad Juarez, Mexico, to Suzhou, China, to take advantage of its competitive Chinese supply base (Malkin 2002). In 2002, Microsoft transferred the production of the Xbox video-game console from Flextronics’ Guadalajara plant to two Chinese plants. According to Todd Holmdahl, Microsoft’s general manager of Xbox hardware, the main reason was that China is closer to their supply base (Smith 2003). Access to low-cost supplies for highvolume products is an important factor insofar as, in the electronics industry, materials and parts make up 80 percent of manufacturing costs (Black 2002). Locally produced inputs are also vital in apparel production. Large U.S. marketers continue to shave weeks off product development to take advantage of demand for the most recent (and often short-lived) fashions. Some of this is replenishment, for which inputs are predictable. But an increasing share of these quick-turn orders are new, fashion-based designs. Lead times for fabric from the United States are too long, and Mexican textile mills are few and failing (Birnbaum 2003, p. 50). The Mexican auto and auto parts industry, which boasts a much-betterdeveloped supplier network, weathered the U.S. recession far better than apparel or electronics producers, and returned to record-high exports to the United States by 2002 (see Figure 6.6).15

190 CHAPTER 6

Conclusion: What Future for Mexico? Proponents of regionalization argue that distribution dictates location, and that, for many products (like fashion-intensive garments and quickchanging high-tech electronics), assembly needs to be close to final markets. However, for any quick-turn model to work, assemblers need to be close to key suppliers as well. China’s comparative advantage is not only its lower wages, but also its development of a local, competitive supply base. Our research indicates that shipping finished goods to far-off markets may not only be cheaper, but may sometimes be quicker than sending rich country inputs for labor-intensive assembly in lowwage environments. Does this emerging model for electronics and garment production spell doom for Mexican plants? It is certainly not good news for the firms that have been least able to move forward or backward in the value chain. In electronics, leading manufacturers are not abandoning Mexico completely. Instead, even as they move high-volume manufacturing to China, many are trying to upgrade their operations in Mexico and move more complex, low-volume/high-mix products from the United States to Mexico. For example, after losing orders to Jabil factories located in China in 2002, the Guadalajara Jabil Circuit factory managed to turn around by concentrating on more complex products traditionally made in the United States, such as computer routers. In just a few months, the Guadalajara Jabil factory surpassed Jabil firms located in the United States and its technicians developed a spare-parts tracking software that became Jabil’s new global standard (Luhnow 2004). In the same vein, in 2002 Flextronics opened a Technology Center in Guadalajara to offer OEMs product analysis and test characterization services (Serant 2002b). There is thus a production shift or “rotation” at work, as products move from the United States to Mexico and then China and are replaced with other products.16 On the other hand, efforts at upgrading are limited, due to a lack of investment in R&D in Mexico compared to China.17 In textiles and apparel, many of the remaining maquiladoras have tried the same strategies of firms in their sector across the developed world: improve productivity, specialize in smaller-batch production, and try to concentrate on higher-quality, more-value-added products. The problem with this “niche” strategy is that competitors are many and niches are small. However, Mexico will retain the advantage of proximity, and U.S. retail

FROM NAFTA TO CHINA?

191

Figure 6.6 U.S. Imports from Mexico: Apparel, Electronics, and Autos 45,000,000 40,000,000 35,000,000

computers, parts, semiconductors

30,000,000

apparel and household goods

25,000,000 20,000,000

TVs, VCRs, radios, and stereos

15,000,000 10,000,000

autos, auto parts, engines, and accessories

5,000,000 1998

1999

2000

2001

2002

Source: Same as Figure 6.3.

buyers will want to maintain some suppliers in the region. For instance, in Yucatan, state government officials have begun a comprehensive effort to upgrade local capacities with credit, education, and infrastructure projects. Their goal: a quick-turn, high-quality garment production environment only one day by ship from the Mexican port of Progreso to U.S. ports on the Gulf of Mexico. After the NAFTA boom, Mexico’s electronics and textile and apparel sectors are in the process of a painful transformation. Despite production shifts to China, electronics producers seem clearly committed to maintaining some production and assembly within the lower-wage areas of the NAFTA region. U.S. apparel companies are less committed to regional sourcing, but are cautious of over-reliance on suppliers from any one region (or, in the case of China, one country). What Future for Global Producers? For global producers, including Taiwanese textile and electronics manufacturers, our findings question the assumption that consumer goods producers need a low-wage assembly site close to their end markets. Many Taiwanese textile and garment firms found Mexico to be a difficult production environment. Workers refused to live in dormitories, or to work on Sunday; one firm with a sizable investment in a completely

192 CHAPTER 6

new plant actually tried to relocate a number of machine operators from Mainland China.18 Several managers believed that the benefits of a NAFTA production site did not outweigh the costs of establishing and running factories in Mexico. Some managers remained committed to the region, and planned further investment in Central America (which also benefits from a trade agreement with the United States for garments), where labor is cheaper and workers are seen as more quiescent. Other firms have decided to concentrate further investment in Asia, particularly China. Five years ago, many in the business and academic worlds believed that NAFTA would be the main entry for MNCs into the U.S. market. The crucial barrier to the U.S. market is neither time to market nor tariffs, but may in fact be information. In production sites where managers succeed in controlling the flow of inputs and information through the supply chain, as Taiwanese firms do in Mainland China, firms have a serious comparative advantage. Many Taiwanese manufacturers, who already have production sites in Mainland China, seem not to need production sites in the North American region to compete in the U.S. market. Indeed, Mainland China remains an attractive location for high-volume, low-end production. The success of Taiwanese manufacturers there suggests that within the East Asian regional production system, Taiwanese managers may be particularly well equipped to exploit the Mainland’s comparative advantage in manufacturing. Indeed, much of Asia’s recent success cannot be solely attributed either to Taiwanese investors or to Mainland manufacturers. Rather, credit should be given to the Taiwan–China “partnerships” that manage to capture complementarities in the region.19 The current distress of Mexico can partly be explained by the failure of NAFTA to foster such partnerships. However, it does not follow that Taiwan should thus simply focus on China and take advantage of its historic and cultural ties with the Mainland. In addition to the inherent political risks of an all-China strategy, Taiwanese firms that move to the Mainland risk losing their competitive advantage. Although Mainland production allows Taiwanese firms to compete on price, a high-volume, low-cost strategy leaves firms vulnerable to innovations and new competitors (many of whom, through their training of Mainland managers, Taiwanese firms have helped spawn). As Berger and Lester explain in the introductory chapter to this volume, another option is to succeed via innovation and branding. This

FROM NAFTA TO CHINA?

193

up-market strategy requires a deep knowledge of the leading world consumer markets (the United States, Europe, and Japan). Rather than production facilities in Mexico, Taiwanese firms could consider investing in distribution and service centers in the United States, which would yield a much greater understanding of their customers. In addition, up-market strategies require high levels of expenditure on R&D in end-user markets. In this respect as well, Taiwan can learn from the failure of Mexico.

7 Innovation and the Limits of State Power Integrated Circuit Design and Software in Taiwan Dan Breznitz

Of the East Asian newly industrialized economies (NIEs), Taiwan’s story is one of the most inspiring. Taiwan is the only society in the region that closed, in many critical aspects, the gap in innovational activities with the leading Western industrial nations and Japan—the G7.1 Taiwan has also developed a vibrant industrial system of indigenous new small and medium-sized enterprises (SMEs) that is not dominated by a few huge conglomerates or subsidiaries of foreign multinational corporations (MNCs). Today most Taiwanese companies specialize in producing systems and equipment for other companies. In the “new global economy” this is a position of strength. But, the fact that Taiwan has yet to foster a significant number of companies that can develop new products and not just improve and innovate on products developed elsewhere suggests a major concern that its industry may be unable to capture the lucrative rents that come with skills that bring new products to market first.2 The Taiwanese government’s goal in the last decade has been to spur greater product innovation capabilities. This chapter explores the capabilities and limits of the Taiwanese state in achieving such objectives in two key sectors of the information technology (IT) industry, software and integrated circuit (IC) design. It presents the evolution of these two subsectors and analyzes their strengths, capabilities, and innovational systems, and the business models employed by private firms in these sectors. At a detailed micro level, it maps the institutional system in which the industry is embedded and the changing roles of the state, with the objective of understanding the industry’s change over time as well as its interactions with the financial sector, the universities, and the state. Much of the literature that attempts to explain Taiwan’s economic 194

INNOVATION AND THE LIMITS OF STATE POWER

195

growth depicts it as a society in which the state takes the lead in initial forays into new economic activities, but always with the aim of spurring private entrepreneurs and through interactions with private industry (Amsden 2001; Amsden and Chu 2003; Evans 1995; Fields 1995; Fuller 2002; Hong 1997; Mathews and Cho 2000; Wade 1990; World Bank 1993). In most of these accounts, Taiwan’s industrial and innovation systems are presented as humming machines of development, characterized by smooth and well-functioning divisions of labor between government and industry. In this division of labor, the public research institutions do most of the research and development (R&D) up to the level of a working prototype, and then they diffuse the results to industry, which concentrates on final development and integrated design. The specificity of this division of labor, in contrast with that found in other countries, stems from the deep level of intervention the state has in the technological capabilities of the industry. Public research institutions not only make decisions about which technologies industry should acquire, but also develop them up to the level of working and almost sellable prototypes before handing them over to industry. It is this division of labor that is considered responsible for Taiwan’s leading role in the global IT industry. This division of labor may, however, limit Taiwan’s ability to excel in cutting-edge innovation activities. It creates a system that supports technological absorption and excels in second-generation innovation, that is, innovation that seeks to improve the production and reliability of products based on novel technologies developed elsewhere. It is not certain whether this system can assist Taiwanese companies in developing their own innovative, original IT products. Moreover, a specific institutional system of rules, regulations, and service and financial industries has evolved around this system. The question today is whether this institutional environment can support more original R&D-based companies. Previous studies of Taiwan’s high-technology industry, and for that matter, most studies in the developmental state tradition, have explained the great success of Taiwan’s IT industrialization at both the macro and micro levels. Here, in contrast, we use a micro-institutional industrial case-study method to analyze not only the successes, but also the limits and even the failures of Taiwan’s industrial technology policy.3 It is only through such an approach that we will be able to understand how state policy shapes industrial development. It is not enough to examine how policy and state action shape industrial development at the aggregate

196 CHAPTER 7

level, or to look only at the successes. We must analyze the failures as well as the limitations of the successes. In order to advance such understanding, this chapter looks at two sectors that have been at the forefront of technological innovation in the global IT industry during the 1990s: software products and fabless chip design, particularly for telecommunications and multimedia applications. They are especially interesting cases in which to examine government– industry interactions, since each sector has its own dedicated public research agency. The fabless chip design industry is considered to be the latest and most innovative segment of the semiconductor industry to emerge in Taiwan, in part as a result of the efforts of the Industrial Technology Research Institute (ITRI) and its two dedicated semiconductor labs, the Electronic Research Service Organization (ERSO) and the Computing and Communication Lab (CCL). In software, the role of the state agency, the Institute for Information Industry (III), is, however, not as clear, and less has been written about its approach to and interaction with industry.4 This chapter maps the institutional system in which the industry is embedded as well as the changing roles of the state. The data analyzed here come from a database of 629 interviews conducted as part of the Industrial Performance Center (IPC) Globalization Study, and from official and industrial statistical sources. The interviews were conducted with founders and executives of private companies, top civil servants in all the developmental and science and technology industrial agencies, academics, and venture capitalists (VCs). Building on these sources, this chapter presents a detailed micro-level understanding of the industry’s evolution, as well as of industry–financial sector, industry– university, and industry–state interactions.5 This chapter follows in a tradition of institutional analysis that assumes that in order to understand industries, one needs to analyze regular patterns of behavior, interaction, competition, and cooperation that constrain and support certain capabilities and motivate economic actors to act in particular ways in their search for material gain (Braczyk et al. 1998; Breschi and Malerba 1997; Breznitz 2004a; Culpepper and Finegold 2001; Fligstein 1991, 1996; Hall and Soskice 2001; Kitschelt 1991; Lundvall 1992; Nelson 1993; North 1990; Powell and Dimaggio 1991; Streeck 1991; Ziegler 1995; Zysman 1996a). In order to do so, this chapter looks at how the particular policies and activities of state agencies have created an institutional environment, both in the financial system and in the industrial system, that induces

INNOVATION AND THE LIMITS OF STATE POWER

197

companies to concentrate on specific activities and favor certain business models over others. In the financial system, the analysis focuses on the kind of financing available to founders and companies in the developmental phases of their companies and the specific incentives provided by the Taiwanese system of financial regulations. In the industrial system, the chapter looks at the Taiwanese IT industry as a whole, analyzing the opportunity structure it presents to new enterprises, as well as the skills and capabilities it enhances. The research has arrived at two broad conclusions: first, the publicresearch-institution-based industrial technology policy of Taiwan has been fruitful in helping the growth of private industry when (1) these research institutions have created multiple and broad interactions with the private IT industry on issues including policy development, knowledge and information, finance, and personnel; and (2) when the public research institute saw private IT firms as their final customers. On the other hand, the public-research-institution-based approach has hampered the growth of industry when the institute has competed directly with industry for customers. Second, this chapter finds that the particular trajectories of development supported by government, although they have enabled spectacular and rapid growth, still limit industry to a fairly narrow range of activities in the global IT product chain. These activities focus only on secondgeneration innovations—designs of products based on technologies developed elsewhere and products with markets that are already well defined. For example, in the case of chip design, although the Taiwanese semiconductor industry is the third- or fourth-largest in the world, with skills and technological capabilities as competitive as any other country, no Taiwanese IC firm has yet managed to produce any cuttingedge innovative IC design. Most are not even trying. As long as the industry does not develop these capabilities, it will remain dependent on the decisions and technological development of foreign firms. The first part of this chapter reviews the literature and lays out theoretical issues involved in Taiwan’s experience of rapid IT growth. The second part presents an account of the IC design industry, its development, strengths, and limits. The third part focuses on the software industry, and argues that the almost adversarial relationship between private industry and III has played a major role in the weak growth of the sector. The chapter concludes by sketching out a few policy alternatives for these two sectors.

198 CHAPTER 7

Theoretical Review and Analysis of Taiwanese IT Industrial Growth The last few decades have seen fierce debate about the exact role of the state in economic development. The dominant view among scholars in the literature on East Asia, especially on Taiwan, can be summarized as a softer version of the developmental state theory. The adjectives given to this newly discovered category of developmental state vary. But whether it is called the flexible developmental state, the developmental network state, or the embedded autonomy industrial bureaucracy, the same basic model is advanced (Amsden 1989; Amsden and Chu 2003; Evans 1995; Mathews 2002; Mathews and Cho 2000; O’Riain 2000, 2004; Wade 1990; Wu 2005).6 This model argues that for a state to initiate industrial growth, especially in a technologically intensive industry, it needs to interact closely and evolve a division of labor with local industry. The neodevelopmental statists argue that the state’s main role is in the codesign and coordination of industrial development. In contrast to an earlier generation of statist scholarship, they do not attribute to the state the role of visionary plan development and implementation (Amsden 1989; Anchordoguy 1989; Evans et al. 1985; Johnson 1982). According to the neodevelopmental statists, the state needs to: (a) make and implement decisions from a public point of view, taking the national interest rather than private interests as a reference point; (b) be informed about the needs, abilities, and difficulties of the industry so it can tailor its policies and refrain from policy initiatives that harm the industry; (c) change its policies in tandem with the changing needs of industry; and (d) let the industry gain more and more power to decide on its future as it finds its feet. The state, in sum, needs to be able to change its own role from initiator and leader to that of a supporting actor. The theorists of the new developmental state portray the state as either successful, hence a neodevelopmental state, or failing, therefore not a neodevelopmental state. The literature does not entertain the possibility that the same state can succeed in some attempts to create hightechnology industries and fail in others. An example is Sean O’Riain’s recent study of the Irish software industry: The flexible developmental state (FDS) is defined by its ability to nurture post-Fordist networks of production and innovation, to attract international investment, and to link these local and global technology and

INNOVATION AND THE LIMITS OF STATE POWER

199

business networks together in ways that promote development. (O’Riain 2000, p. 158) (Italics added.)

This definition classifies the cases using the dependent variable; it assumes that the correct level of analysis is the national, and it sees the state as a unitary actor. This approach discounts the value of micro-level analysis of the ways in which the state can spur innovation-based industrial growth, whether post-Fordist or not. It also precludes the possibility that the same state, which in one industrial sector is an ideal example of the new “softer” developmental state, successfully nurturing spectacular high-technology industrial growth, might be failing in another. The tendency to define a state as either a neodevelopmental state or not is the result of two starting points. First, it results from using a national level of analysis, looking at aggregate industrial development, or from the analysis of a specific industry or region, which is then generalized to the national level. The second assumption is that states succeed or fail. This tendency can be attributed to the underdefined nature of the neodevelopmental state and the need to classify the cases studied using the dependent variable. In most accounts Taiwan appears as an exemplary case of a government that has managed to handle these complex and sometimes contradictory roles. Three recent accounts of Taiwan’s high-technology industrialization reached almost complete agreement on the ways in which the state and industry in Taiwan have developed the IT industry since the early 1970s (Amsden and Chu 2003; Hong 1997; Mathews and Cho 2000). While each author has a different vocabulary, each describes a similar two-phase process. First, the state’s research agencies acquire a technology from abroad, absorb it and improve it, and then spin off private companies to spur the industry. Second, after private industry has emerged, the industry and state settle on a new division of labor in which the state’s role is to locate, absorb, and infuse the industrial system with new technologies, and to assist private firms with their own advanced R&D projects. The most famously successful subsector of the industry where the state was not only the initiator but also managed a successful transition into a more supporting role is IC silicon chip fabrication (Fuller 2002; Fuller et al. 2003). Here Taiwan’s particular version of science and technology industrial policy, based on big public research institutions, created a world-leading industry. In 1973, the ITRI was founded, and in

200 CHAPTER 7

1974, ITRI’s ERSO laboratory commenced its operations. Beginning from an obsolete IC technology acquired from RCA in 1976, ITRI started to spin off companies. Of these, the most important were UMC, TSMC, Winbond, and Vanguard. One spin-off, TSMC, proved critical. Under the management of then head of ITRI, Morris Chang, TSMC set out to, and succeeded in, developing a new business model of pure-play foundries. In so doing, TSMC completely transformed the workings of the international IC industry (Fuller 2002). After this success and with the rise of the IC fabrication industry, ITRI changed course and now acts as an important channel for the introduction of foreign technology and the locus of some intensive research activities, as acknowledged in many of our interviews with private companies.7 The success of the IC fabrication subsector is impressive by any standard. However, even in this story we see not only the successes and capabilities of the Taiwanese state in spurring new innovation-based sectors, but also its limits and weaknesses. As Fuller et al. show, a major difference between Taiwan and other Asian NIEs, especially Korea and Japan, was that the state under the Kuomintang (KMT) intentionally built a financial system with a severe scarcity of patient capital, a fact that limited the ability of Taiwanese companies to compete in many IT manufacturing sectors (2003). Taiwanese companies’ average equity/debt ratio has been as low as in the United States and sometimes even lower (T.J. Cheng 1990, 1993; Fields 1995; Gold 1986). This helped the Taiwanese economy as a whole to pass through the East Asian financial crisis relatively unscathed (Park 2000). It has, however, also hampered Taiwan’s ability to innovate in IT manufacturing. Generally speaking, in the last decade Taiwanese IC fabrication companies excelled only as pure-play foundries with the integrated device manufacturer (IDM) companies losing ground both internationally and domestically (Fuller 2002). Since Taiwan cannot compete directly with the South Korean and Japanese IT manufacturing industries, well endowed as they are with patient capital, it needs to develop different strategies for successful innovation. One might be to focus on sectors with high granularity, highvolume production, and no requirement for large amounts of patient capital, as suggested by Fuller et al. Other possible strategies would be to develop a sophisticated innovation services IT industry or to take a leaf out of the U.S. IT industry’s book and learn to excel in the development of cutting-edge new products. The next two sections analyze

INNOVATION AND THE LIMITS OF STATE POWER

201

Table 7.1 Top Ten Taiwanese Design Houses Ranked by Sales (2000–2002) Design House MediaTek Inc. VIA RealTek Sunplus Technology Novatek Microelectronics Corp. ALi Corp. Elan MicroElectronics Elite Semiconductor Faraday Holtek Semiconductor Inc.

2002 rank 1 (295) 2 (252) 3 (92) 4 (86)

2001 rank 2 (154) 1 (343) 3 (73) 4 (66)

2000 rank 2 (129) 1 (309) 5 (54) 3 (63)

Main products Optical storage PC chipsets Networking Consumer

5 (67) 6 (61)

6 (42) 5 (54)

8 (42) — (31)

7 (40) 8 (34) 9 (34)

7 (36) 9 (30) — (24)

9 (39) 7 (43) — (—)

Consumer PC chipsets/ DVD player IC Consumer Memory Memory

10 (33)

8 (32)

6 (45)

Consumer

Source: Information Technology Information Service (ITIS), Semiconductor Industry Yearbook; ITRI, 2003, “Industrial Statistics.” Note: Sale figures are in parentheses (in NT$100 million)

Taiwan’s experiments with these strategies by looking at the successes, limits, and failures of Taiwan’s development policies for software and IC design/fabless IC chips. The Taiwanese IC Design and Fabless Industry Over the last ten years the Taiwanese IC design industry has had remarkable growth. From fifty-one companies in 1991, the industry grew to sixty-six companies employing 2,109 people in 1996, to 225 companies employing 11,800 people in 2002. This rapid growth made it one of the three biggest IC industries in the world, together with those of the United States and Israel, with the sales of the largest ten IC design firms alone reaching US$3 billion (ITRI 2003, various years). Looking at the largest ten companies by sales in 2000–2002 (as shown in Table 7.1), we see that the industry’s skills are quite developed and flexible, as attested to by the ability of Taiwanese IC design firms to excel in almost every segment of the semiconductor industry. None of the leading companies develops completely new products that are first to market. Furthermore, even the most innovative firms, ones with some own-brand recognition, such as VIA and ALi, rely on

202 CHAPTER 7

developing secondary innovations on the basis of technologies developed by the leading U.S. MNCs. This tends to leave many of them in vulnerable positions, where changes in technology or in the behavior of leading MNCs can rapidly weaken their market position. The story of VIA, the largest Taiwanese IC design company in terms of sales until 2002, is a case in point. VIA’s main product line is personal computer (PC) chipsets. When Intel’s Pentium 3 was the leading PC central processing unit (CPU), in part because of Intel’s decision to stick with the more expensive Rambus technology, VIA approached a global market share of almost 50 percent, while Intel’s market share shrank. With the move to Pentium 4 technology, Intel attacked VIA on multiple fronts. It did not license VIA its P4 technology, and it took VIA to court over patent infringement allegations in October 2001. As a result of these moves, all first-tier PC manufacturers such as Asus or Gigabyte refrained from using VIA’s chipsets, fearing legal action from Intel. VIA countersued Intel. The legal battle was fought in five jurisdictions. In April 2003, the companies reached a settlement that included a patent swap agreement and a licensing agreement, but by then VIA had lost substantial market share and revenues. Moreover, as it agreed to pay royalties to Intel, it is not clear whether VIA can retain the profit margins it enjoyed before the legal battle started. In 2001, VIA was the biggest Taiwanese IC design house, with about US$1 billion in sales, and controlled almost 50 percent of the market share of PC chipsets; in 2002, VIA was only the second-largest Taiwanese IC design house, with sales down to US$729 million, and a global market share of less than 25 percent. In 2003, VIA sales continued to drop to US$600 million (AFX 2001; Financial Times 2003a, 2003b, 2003c; Hille 2003; Hung 2002). The main impetus for the rapid growth of the design, or fabless, sector after 1996, was the pure-play foundries. Both major foundries, TSMC and UMC, have encouraged the development of the design sector, establishing exclusive foundry–IC design house relationships with firms they refer to as “club members,” and helping them financially. Many of the top IC design houses, such as MediaTek and NoveTek, were spun off from UMC. During the IPC study, we interviewed most of the larger design houses as well as a large sample of the newer and smaller telecommunication and image-processing fabless firms. This has enabled us to observe the overall development and structure of the IC design sector as well as to gain a deeper understanding of these two niches that have been the most innovative since the late 1990s.

INNOVATION AND THE LIMITS OF STATE POWER

203

Structure, Development, and Business Models in the Taiwanese IC Design Industry Apart from the few cases where a Taiwanese company manages to sell its own branded products to final customers, such as D-Link’s WiFi (802.11-based wireless LAN) routers and ADSL modems, or VIA’s Intel CPU chipsets, one main business model is employed by Taiwanese fabless companies. This model utilizes Taiwan’s advantage in having the local presence of both the world’s biggest and most advanced pure-play foundries and some of the largest system houses/original equipment manufacturers (OEMs) like Quanta or BenQ (companies that manufacture products for own brand manufacturer [OBM] MNCs through outsourcing contracts). Most of the Taiwanese IC design houses employ two variants of the same strategy: supplying the Taiwanese (and more recently the Chinese) system houses with the chips they need in order to provide finished products at competitive prices and of sufficient quality to OBM MNCs. The Taiwanese fabless companies rely on the proximity of the world’s biggest pure-play foundries. This geographical proximity enables them to produce cheaply and speedily large quantities of newly designed chips and to inspect and assure quality in almost real time, at a relatively low cost. The Taiwanese IC design houses we interviewed, including ones focusing on telecommunications and image processing, specialize in chips based on second-generation technologies with the aim of lowering costs and increasing reliability; they also do some process innovation. Apart from an unsuccessful attempt by ICreate, a subsidiary of Etron, to develop and market advanced chips in a joint venture with the Israeli company Zoran, no Taiwanese IC design house that we interviewed aims to develop original products, new technologies, or products that are based on cutting-edge technology. Figure 7.1 below provides a graphic representation of the main differences between the market relation and information and communication flows of the Taiwanese industry in comparison with the Israeli and U.S. industries. The Taiwanese fabless companies’ business model has two variants that are not mutually exclusive. In the first variant, the design houses custom design chips for clients using their specifications (e.g., a local OEM fulfilling specific contracts with an OBM), test and quality-assure them, and then manufacture and deliver them using a pure-play foundry

204 CHAPTER 7

Figure 7.1 Market Relations and Communication Environment of the Taiwanese IC Design Houses

Comm

End users

unica

tion

System houses (big > OEM/ODM Taiwanese subcontractors such as Quanta or BenQ)

Orders

Taiwanese IC design house

ication

American/Israeli IC design houses

(

Com

mun

on icati

ion

unicat

Comm

Orders

Ordes and

Co m

Commun

Communication

Orders

nic

atio

n

OBMs (Leading USA and Japan companies such as HP or Dell)

mu

Communica

tion

Co

d n an io s at le nic Sa mu m

Communication

Pure-play foundries (Taiwan)

for fabrication. In the second variant, the IC design house builds “standardized” chips for OEM companies and other customers, manufactures them using a pure-play foundry, and sells them either directly or through distributors. The firms we interviewed claimed that, as the market for standard chips is vastly larger, revenues using the second variant with successful products are much higher. A significant number of companies, however, use both business models, mainly because customdesigned chips have the advantage of a guaranteed customer and higher profit margins per fabricated chip. Because the first, or custom-made, business model has lower costs, and companies can secure revenues either up front or rather quickly, it has been the preferred mode of operation for new companies until recently. The case of Sunplus, consistently one of the top four companies in terms of sales since 2000, is illuminating. Sunplus was established by a group of former ITRI-ERSO engineers who left ITRI to join Silicon Integrated System (SiS) and left SiS after its restructuring in 1989. The idea of the founders was to focus on the consumer market, especially toys; thus reliability and price have always been much more important than the latest technology and features. Using their own money to seed the company, the founders immediately sought revenues to supplement their capital and started operations with custom-made application specific integrated circuit (ASIC) designs. In 1994, Sunplus manufactured

INNOVATION AND THE LIMITS OF STATE POWER

205

its first standardized chip, a micro controller for toys with LCD monitors, from which it branched out to other multimedia-related chips. Today Sunplus uses the custom-made chips in its entire gift and toy business and also sells the standardized chips to other customers. Most of Sunplus’s revenues originate now with its standard chips; however, as there is fierce competition in the multimedia mass market, a large percentage of profits comes from the smaller custom-made chip division. Sunplus realizes that, even though it does not use cutting-edge technologies, in order to be able to excel, it must keep abreast of new technologies and be able to implement them quickly once prices go down. In order to do so, Sunplus, in a way similar to many successful Taiwanese fabless companies, employs a few strategies. First, it maintains a close relationship with ITRI, especially with its CCL lab, using ITRI as a channel for new technologies. Second, Sunplus has established its own investment arm that invests in foreign and local start-ups in fields that Sunplus thinks it might need in the future. Third, Sunplus directly licenses technologies. The Taiwanese IC design houses do not rely on in-house developed cutting-edge intellectual property to give them market advantage, but rather on their ability to deliver moderately sophisticated products more speedily, cheaply, and reliably than their competitors. This is true even for companies that pass through the new governmental sponsored Small Business Innovation Research (SBIR) program and/or ITRI’s incubation center. For example, one interviewee told us: “Our game plan is to focus on the China market. The United States is far too technologically advanced for us” (IPC interview with an employee of an ITRI incubation center graduate IC company, January 2003). Another interviewee, a serial entrepreneur who is now the founder of a company that is still in ITRI’s incubation center added, “The customers we want need to have good chips but not state-of-the-art chips. They care more about reliability, mass production, and unit costs than about cutting-edge technology. They are not looking for innovations” (IPC interview, January 2003). Two other interviewees, from two of the most technologically advanced Taiwanese fabless communication companies, were even more pessimistic about the state of innovative activities in the industry: There is only one place in Taiwan that develops new technology and cutting-edge products—ITRI—I sometimes miss being there; the only reason I left is because I wanted to get rich.

206 CHAPTER 7

I do not know about any real R&D–based company in Taiwan, apart from ITRI of course. Everybody is interested in revenues and profits. MediaTek is the biggest model of success. The biggest problem for any R&D–based model to work is the stock options/bonus regulations. These pull all the talent into the big manufacturing companies. (IPC interviews with companies in Hsinchu Park, January 2003)

In sum, the fabless chip sector evolved around a particular business model that relied on the unique institutional features of the Taiwanese IT industrial system as a whole. The main three features are: first, the existence of large system houses and OEM companies that sell in foreign markets, hence a large demand for chips based on second-generation technologies; second, industry reliance on the close proximity of the world’s largest and most advanced pure-play foundries; and third, the industry’s innovation system, which is characterized by a division of labor between ITRI and an industry that fosters quick technology transfer and second-generation innovation, while continuously infusing the system with the most recent foreign technologies. Utilizing this business model, the IC design sector provides complementary assets to the Taiwanese OEMs and pure-play foundries. First, it supplies the pure-play foundries with a constant stream of orders in a variety of IC designs, helping them to stay profitable and to extend and maintain some of their own technological capacities. Second, the Taiwanese IC design industry supplies the system houses and OEM manufacturers with the chips they need either to profitably offer solutions to Western and Japanese OBMs, or to compete successfully with them by lowering their cost structure. The development of the IC design subsector thus enriched the Taiwanese IT industry with positive feedback, strengthening the Taiwanese position as mid-level suppliers within the global IT product chain. Each part of the Taiwanese IT hardware industry—system houses, pure-play foundries, and IC design companies—strengthens and is strengthened in turn by the existence, outputs, and demands of the others. Despite this positive outcome, the IC design industry still is unable to develop alternative business models for capturing the higher rents that come from original innovative designs. We should not, however, discount the magnitude of Taiwanese achievements in IC design. Taiwan, a small society, has developed the secondlargest industry in the world, with the six largest companies having combined sales of about US$3 billion, far surpassing the South Korean,

INNOVATION AND THE LIMITS OF STATE POWER

207

Japanese, and Western European industries. The industry is not only growing but also enhancing the overall competitiveness of the Taiwanese hardware industry. Whether or not there is a move to innovative products, the industry’s growth seems secure in the near future. The Role of Institutions and Government Policy in the Structure of the Fabless Sector What are the institutional settings that induce the Taiwanese fabless companies to employ a specific business model? And if the declared policy aim of encouraging the Taiwanese fabless companies to be more innovative and conduct more technological cutting-edge R&D has not been reached, how does the implementation of government policy matter? Financial and industrial institutions shape the economic and innovative behavior of the IC design sector. By the financial institutional system, I refer to the kind of financing available to founders and companies in the developmental phases of their companies and to the specific incentives provided by the Taiwanese system of financial regulations. Regulations that deal with public and private companies that aim to go public in Taiwan and with employees’ stock options are given special attention. By industrial system I mean the past and current industrial activities in the Taiwanese electronic industry as a whole and the incentives they provide for a particular kind of activity. There are two main financial constraints on fabless companies that wish to develop innovative products in Taiwan: one affects the beginning of a firm’s life and the other the end of its developmental stage. The first constraint is the scarcity of finance for the early stages of firms that rely on cutting-edge R&D activities. The second constraint derives from the fact that the Taiwanese stock-option system gives enormous advantages to companies that have already gone public, not only granting them stronger financial capabilities, but also giving them almost a stranglehold on hiring the best engineering graduates. This is because of Taiwan’s employee stock-option regulations. These prohibit companies, both public and private, from giving stock options to employees (one of the main attractions that a cash-starved start-up can offer to entice new employees in the United States), but allow already-public companies to give actual stocks (i.e., not an option on the stocks but the stocks themselves) to their employees at a sharply discounted rate under the “profit-sharing system,” where the stock is treated as profit sharing. Moreover, the stock

208 CHAPTER 7

bonuses are taxable only at the time of sale and only on the nominal par value (set by law at NT$10 per stock no matter what the stock’s market price is), giving the employees in reality a tax-free capital gains bonus. For example, according to TSMC’s annual reports, the average profit per employee received by TSMC employees through stock per year was more than NT$1.5 million in 1996 and reached more than NT$2.5 million in 1998. An engineer with a master’s degree earned much more than the average. A CFO of a leading IC design house calculated that after four years, an engineer with a master’s degree will get an annual bonus worth about US$300,000; a good annual salary for the same employee stands today at US$18,000 (IPC interview with IC company’s CFO, November 2003). Hence, a graduate of a good university with an advanced degree who wants to become well off needs only to join an already public company to be virtually assured of reaching that goal in a few years. Not surprisingly, new private companies find it very difficult to recruit graduates of the top engineering schools. Consequently, the regulations regarding employees’ stock options in Taiwan motivate new companies to seek the shortest and surest route to an IPO. In order to go public on the main exchange, a company needs to show a few years of constant profits, so most Taiwanese fabless companies use a business model that assures quick revenues, in contrast to one employed by innovative-product-based firms that involve incurring a few years of heavy losses during their R&D phase before sales. The Taiwanese government declares that over the last few years it has supplied a large amount of capital for R&D operations directly to private companies. However, it is not clear that what the government declares as its R&D goals and the effects of the vehicles and programs it uses to distribute this financing are aligned. For example, most of the government-sponsored critical technology development programs are constructed in a way that handicaps the development of new product innovation capabilities in the business sector. A recent example is the “System on Chip” project. It is organized in three discrete stages. In the first, the government focuses on specific end-products it thinks that the Taiwanese IT industry must be able to manufacture in the mid-future. In the second step, the universities and ITRI are asked to propose the intellectual property and technology they think are necessary for companies to develop these products, and how they are going to develop these technologies and transfer them to industry. In the third phase the universities and ITRI develop the technologies and intellectual property and transfer

INNOVATION AND THE LIMITS OF STATE POWER

209

them to industry. Hence, the industry’s role in this innovation capability development program is limited to one of working on products decided upon by governmental officials, using technology transferred to it. The responsible Minister of State has promoted this program as the latest government effort to induce innovative activities and capabilities, with a budget of US$80 million per year and eighty-five new faculty positions, but it is too soon to know if it will expand industry’s innovative product design capabilities (Interviews with senior officials in the Ministry of State for Science and Technology Policy, November 2003). Another example is the R&D tax incentive programs of one of the main governmental investment vehicles, the Industrial Development Bureau (IDB) of the Ministry of Economic Affairs (MoEA). This program gives tax incentives and other benefits only to companies that make products specified in great detail by the IDB itself. The products are listed by the IDB every two years. Therefore, the list rules out truly innovative technologies with markets that are unknown or unknowable. The program specifies that the tax incentives are given only to companies that control the whole production chain, that is, from design to manufacturing and sales. Moreover, it only promotes products that the responsible official described as “not in niche markets that are so innovative that we cannot predict the future so we cannot really help” (IPC interview, May 2001). In sum, while it is true that the state aims at generating more R&D and innovation activities in Taiwan and at sponsoring programs to achieve this goal, it is not clear whether the final result will indeed induce groundbreaking innovative capabilities in Taiwan, or just extend the old technology transfer and division of labor between industry and state to a new set of more complex technologies. There are many government programs to finance industrial R&D; in fact several of our interviewees from the major developmental agencies complained that Taiwan has too many small agencies and programs, each with its own bureaucracy, resulting in a hodgepodge and confusion. Of these programs, the three most relevant to the fabless industry are the SBIR, the Technology Development Program (TDP), and the Leading Product Development Program (LPDP), all of which are sponsored by MoEA’s Department of Industrial Technology (DOIT). Aside from SBIR, most programs assist mainly in the development of products that are based on some R&D effort, whether or not they are new to the world markets. Most of the grants to fabless companies assist them in pursuing second-generation product innovation and do not provide

210 CHAPTER 7

incentives for a riskier approach. Although DOIT has been leading the way in fostering more R&D activities in the private sector in the last three years, it started to give grants directly to private companies only in 1998, after intense industry lobbying convinced the government to transfer a small percentage of the funding that was formerly given to the public research institutions (ITRI and III).8 Both the SBIR and the TDP are staffed with ITRI personnel seconded to the programs. Indeed, according to our interviewees, about one-third of the IC companies that are funded through SBIR also have contract research relationships with ITRI. Looking at private venture capitalists (VCs) we found that, unlike their U.S. counterparts, most of them prefer to invest only in the later stages of development, and only after firms have already proven their success by becoming profitable. There are several institutional reasons for this behavior. First, by law, VCs in Taiwan must be registered companies, not limited partnerships. This limits their ability to invest in young companies in two ways. As a registered company, a VC is not allowed to invest in a venture that is not already registered, ruling out true seed funding. Also, the person who acts as a VC (in the United States and Europe called the general partner) must submit all of his or her investment decisions to the board of the VC firm, thus making his or her decisions more conservative. Second, Taiwanese VCs also are more reticent because of the sources of their capital and the education of the VCs themselves. Most VC funds have strong ties to one individual or company and operate in accordance with the sponsor’s investment strategy, which is usually more risk averse and tends to select firms with established business models that the sponsors can easily understand. This is especially evident in software, where most VC companies do not invest at all.9 The VCs themselves are drawn mostly from the financial industry or the traditional industries, hence, they do not have intimate knowledge of IT, and prefer less risky endeavors. Last, the structure and listing regulations of the Taiwanese Stock Exchange, where most of the exits take place, favor companies that are already profitable for a few years. This in turn motivates the VCs to invest in companies with a business model that is focused on achieving large sales figures without a long, loss-incurring R&D phase. Even when Taiwanese VCs do invest in young Taiwanese companies, they usually prefer companies that follow the well-trodden path of producing products based on second-generation technologies with the prospect of near-term revenues.10

INNOVATION AND THE LIMITS OF STATE POWER

211

The structure of the IT industry also motivates fabless companies to pursue business models that exploit second-generation innovation, and constrains companies that try to conduct original-product R&D. The Taiwanese system supplies virtually infinite opportunities for a small new IC design house that can offer the already defined chips needed in the production of various electronics products more cheaply and reliably, especially for those with skills in the telecommunications and image processing areas.11 In addition, the economic and management skills that are needed to run these kinds of operations are abundant. In other words, the Taiwanese industrial system is richly endowed in the “specific economic competencies” needed to succeed in second-level innovation-based design (Carlsson 1995; Carlsson and Eliason 1994). The Taiwanese system of innovation is primed to give private companies rapid and continuous access to new technologies from abroad. In the last few years this kind of technology channeling became one of ITRI’s main roles. ITRI, especially through its CCL, has become a main provider of R&D in the industrial system as well as an important channel for foreign technology. The answer of each and every fabless company interviewee to the question of whether there is any company that conducts technological cutting-edge R&D was that only ITRI and its CCL do this. A top official of CCL told us in frustration that he thinks that CCL’s own activities, instead of fostering more innovational activities in private companies, actually allow these firms not to innovate. Technology transfer from CCL is very important in Taiwan. The companies use us to train their people, take our technology, and create some products. After which they go to their customers, show them the products as a proof that they can do OEM, sign some OEM contracts, and just do the same things a couple of years down the road. Our aim is to help and spur them to routinize R&D and product innovation. However, the end result is that they do not really do it. A main reason is that they can more cheaply rely on technology transfer every time they need to upgrade their skills, and continue to operate as a pure OEM. (IPC interview, May 2001)

A company that wants to innovate faces significant obstacles, from the lack of finance, the difficulty in recruiting highly skilled personnel, and the relative scarcity of business and management skills to run a Silicon-Valley–like operation. All of these problems are augmented by the fact that if geographical proximity to customers and suppliers helps

212 CHAPTER 7

companies to sell to OEMs and ODMs, then their distance, both physical and cultural, from the United States, their main market for innovative products, makes the operation more complex and more expensive. The Taiwanese Software Industry If the story of the semiconductor and IC design/fabless sector is one of impressive growth and successful policies, the software industry offers a less sanguine tale. Official statistics are inflated, but even they show the one major failure of the Taiwanese software industry: that it is mostly domestically oriented. Official figures show a rapid growth in sales from NT$22 billion in 1991 to NT$149 billion in 2001, but exports in 2001 amounted to NT$16 billion, or a little over US$490 million, which is only 10.73 percent of total sales (III, various years). It is not clear how much of these sales should be attributed to foreign companies. In addition, the official figures include in the definition of the software industry several activities, such as Internet service providers (ISPs), that are usually not counted as software. In an attempt to estimate the size of the Taiwanese-owned software industry, we calculated the software sales figures (i.e., not including bundled and sales of other products and services) of Taiwan’s public companies, added the sales figures of the top private software companies we interviewed, and finally added III’s total budget as a way to estimate the maximum dimension of the industry. Our calculation for 2002 was about US$1 billion (NT$34 billion), which is one-fifth of the official figure. However, our rough approximation was similar to the official one in export sales (our estimate was lower, but this can be explained by the practice of Taiwanese companies of masking some of the sales of their Mainland China subsidiaries). These figures are tiny in comparison with the hardware sector, or even with the IC design sector alone (for example, MediaTek’s sales alone are more than twice as large as even the official exports figure of the entire software industry). To better understand the relative failure of the software sector, consider the sales figures of Israel, Ireland, and India, which rank among the top examples of industries in emerging economies. Looking at Table 7.2 below we can see not only that the industry is larger in all the other cases, but that more than 50 percent of sales come from exports, while Taiwanese industry exports are less than 11 percent of sales. Moreover, the annual sales revenues of two of the three largest Taiwanese software

INNOVATION AND THE LIMITS OF STATE POWER

213

product companies, Ulead and Cyberlink, are about US$30 million, which is too small to put them in the league of medium-sized IC design houses in Taiwan, or large software companies in Israel, India, or Ireland. This outcome is even more puzzling if we take into account the amount of high level software development critically needed by the hardware industry, especially in the IC design sector. Given the success of the hardware sector in the development of what is usually termed embedded software, as well as the global success of a few software companies, Taiwan evidently does not suffer from lack of software programming skills. With this in mind we can understand why a high-ranking official in MoEA mused in one of our interviews: Look where we are and where other countries like India, Israel, and Ireland are with their software industry. Nobody heard about them thirty years ago. I must admit that we [Taiwan] are not doing very well with software. (IPC interview, May 2001)

In striking contrast to the electronics and semiconductor industries and the positive role played by ITRI, the research institution responsible for software, the Institute for Information Industry (III), has no stories of ERSO-like successful spin-offs or any other initiatives spurring the creation of an industry to tell. III has very little to say about its own positive role in the development of the software industry. Indeed, III can be seen as partly responsible for the state of the industry. Unlike ITRI, III was created after some of the firms that are now the leading software companies in Taiwan had already been established, and the private sector, unlike in the case of the semiconductor industry, was active in the software sector. Thus, one might question the need to spur the creation of an industry. The agenda and the funding granted to III by the Taiwanese government were vastly different from ITRI’s. First, III was asked to promote the software industry; then, III was given the task of promoting the use of IT and software throughout Taiwan and asked to help the government with its own computerization. Finally, III was also asked to generate enough revenue to cover most of its activities. The different agendas proved to be contradictory, with the end result that III transformed itself into one of Taiwan’s biggest IT consultancies and software houses, and has been competing directly with private software firms.

101 221

135 600

70 172

110 540

1992 132 270 330 558 175 700

1993 195 356 485 835 220 800

1994 258 441 734 1,224 300 950

1995 334 513 1,085 1,755 600 1,300

1996 410 585 1,750 2,670 1,000 1,780

1997

Sources: Arora and Athreye 2002; IAEI 2002; IASH 2002; and NID 2002. Notes: Israeli data are underestimated; until 1997, data in Ireland were collected every two years.

Irish Exports Irish Sales Indian Exports Indian Sales Israeli Exports Israeli Sales

1991

Software Sales and Exports of Indigenous Firms in Ireland, India, and Israel, 1991–2000

Table 7.2

509 822 2,650 3,900 1,500 2,350

1998

713 1,150 4,000 5,700 2,000 2,950

1999

788 1,269 6,300 8,750 2,500 3,700

2000

214

INNOVATION AND THE LIMITS OF STATE POWER

215

For many years III has used its favored position within the state to capture all governmental contracts. In addition, III has been given almost all the government funds earmarked for software R&D. Furthermore, the inherent characteristics of software technology make a central research institution a less useful tool than in hardware. These three developments created a situation in which III transformed itself into an aggressively competitive company that enjoys governmental funding to cover its R&D in its efforts to compete with private industry.12 Moreover, for the most part III used its role as the government’s favored channel for foreign technology to grow a thriving business as a distributor for the biggest foreign software companies. This division, called PDD, was spun off in 1997 as Shinwave. The responses of many software firms to our questions about the role of the state in the industry’s development offer the view of III as the biggest obstacle to the industry’s growth. The response of a founder of a financial software company is typical: “The state does not have any positive role; as a matter of fact, III is our greatest competitor. They also compete unfairly. I need to sponsor my R&D from my revenues, they have all their R&D covered by the state” (IPC interview, January 2003). A founder of another software company replied: In our seventeen years we had direct conflicts with III only once so I am very lucky. However, I do not think III is good for the industry. They get government money to help the industry and nothing happens. III does not really care about the industry—they just talk and talk but do not do anything for the industry. They do not even properly do the more simple and straightforward task of consulting with the government on policy issues so policymaking is all tangled up. Even in the basic task of changing the perception of the software industry in Taiwan they do nothing. Customers think software and software companies should not be paid because they do not see software as a “real” product. (IPC interview, November 2003)

This view of III is shared not only by those we interviewed in private firms and in industry associations, but also by officials of other developmental agencies. One official of an industrial R&D agency responded to a question about the role of III in the software industry’s development, “III is a funny organization. It both competes with and tries to assist the local software industry. On a charitable estimate I would say that they compete at least as much as they assist” (IPC interview, January 29, 2003).

216 CHAPTER 7

Another, a head of a different department, gave a similar response: “I agree that III is a very problematic institution. I think that the problem of III is the confusion or the strange positioning that they are in. They try to compete and help at the same time” (IPC interview, November 4, 2003). III is well situated to play a role as a facilitator of collective action and organizer of consortia. However, even in this role, both researchers and industry leaders criticize III. For example, in the crucial case of agreeing on a standard for Chinese writing input/output and internal conversion, a basic need for the software industry’s attempt to grow, Noble claims that III’s actions fostered distrust and hampered the industry’s efforts to reach agreement until an international body, on which Taiwan had no influence, settled the issue almost a decade later (Noble 1998, pp. 123–47). This impression of the failure of state efforts is strengthened by our analysis of the software industry. As the next part of this chapter shows, the most successful segments are those in which III never intervened. Structure, Development, and Business Models in the Taiwanese Software Industry The structure of the Taiwanese software industry reveals a divide between the older companies fiercely competing with III, which focus on software applications for big organizations, and a newer cohort of companies, most of which were founded after the success of the hardware IT industry in the Hsinchu Science-Based Industrial Park. This division suggests that there are actually two Taiwanese software industries. However, the reality is a bit more complex. As was suggested earlier, it is extremely hard to gather precise sales figures on the software industry in Taiwan. Most of the bigger companies, faced with fierce competition from III, have branched out into sales and distribution of hardware and even into completely unrelated fields at times of severe need. Even publicly traded game development companies, such as Summit, earn much of their revenue from bundled sales— the sales of their products on top of sales of DVDs and other entertainment electronics. Estimating software sales in Taiwan is also complicated by the fact that PC manufacturers have been for many years some of the biggest software developers as well. Nevertheless, the picture of the industry that emerges is one in which, until a decade ago, most of the Taiwanese software companies merged

INNOVATION AND THE LIMITS OF STATE POWER

217

both IT consultancy and customized development, as well as the sales of specific software systems (such as automation or finance) that were developed as part of earlier work with customers and then packaged. Most of the PC manufacturers were also successful software producers, especially in the 1980s when each manufacturer had its own slightly different version of the Chinese input/output system, and when most of the classic PC software packages from Western companies, such as spreadsheets, were unavailable in Chinese. Because III had captured the big governmental contracts and the big global IT companies were competing directly on big projects, the industry was unable to develop big software houses specializing in customized development. An example from a founder of one of the companies that managed to survive these times is illuminating: Competing with III is like competing with Microsoft. If you compete you get killed on the spot; if you cooperate you get less money and you might be killed later, but at least you get some work. However, III always wins more contracts than what they can program themselves, so they then subcontract some of them. They will pay me less than what I would get if I could compete on the project in the free market, but I prefer to cooperate with the Microsoft of Taiwan, and not get killed by it. (IPC interview, November 2003)

Most of the firms in the industry have evolved around a particular customer niche, such as international banking or security-house trading systems. Today many of these companies realize that their future lies with Mainland China. Interestingly, it is here that they find that their former relationships as subcontractors of U.S. MNCs in Taiwan are an important asset. For example, a CEO of one of Taiwan’s oldest software companies described his decision to move to China: The funny thing is that we never wanted to go to China, but our U.S. MNC partners asked us to come, both HP and Oracle. The best example is HP. They won the whole IT systems project for a big new plant, and they asked us and three other Taiwanese companies to open a China branch and do subcontracting for them. We did that, and before we knew what happened, Oracle asked us to do subcontracting for them when they won a big Chinese state-owned company contract. I think that without the Mainland, my company would not have been able to grow at all in the last three years. Only because of China do we have a chance to survive. (IPC interview, November 2003)

218 CHAPTER 7

In addition, for the most part, the industry has not been able to develop unique products or strong enough brand names that could give their own products a chance in the market after the big software MNCs launched their own Chinese versions. Thus, when, during the 1990s, Western companies such as Microsoft offered products in Chinese, the Taiwanese companies’ products were wiped out of the market. The one sector that evolved differently throughout the 1990s was the PC games industry. Unlike other sectors of the software industry, the gaming industry had four tremendous advantages that allowed it to thrive in Taiwan. First, III was not at all interested in this sector. Second, gaming, unlike many other software products, benefits from certain cultural traits, and Chinese-born gamers were very keen on playing “Chinese” games. For example, almost every Taiwanese game company has had a strategy or fantasy role playing (FRP) game based on the classic tale of the Three Kingdoms. Third, the costs of game development were very low for many years. Most gaming companies either started by sponsoring teams of high school students who developed the games as a hobby on a costonly basis, or had a mixed development strategy of in-house and semiindependent teams. Fourth, with 7-ELEVEN and similar popular chains in Taiwan selling local games on the corner of every block, gaming companies had distribution channels that reached each and every Taiwanese on a daily basis. It is not surprising that, until 2000–2001, the gaming sector was deemed by many in both the private and public sectors to be the most successful sector of the software industry in Taiwan. All the mediumand large-sized companies, such as Softstar, Summit, Gamania (formerly known as Full Soft), Interserv, and Soft-World, had gone public on the Taiwanese Stock Exchange in the 2000–2002 period. Since then, the industry has been hard hit, and in 2003 most of the companies that we revisited had either retreated from original game development or were cutting down their development activities to the bare minimum. Apart from the worsening economic situation in Taiwan, the two changes that transformed the business environment of the Taiwanese game companies were technological: the wide diffusion of CD-RW (read and write) technology and the rapid emergence of online gaming. These two technological changes have significantly and swiftly lowered the revenues coming from the sales of PC games, with most, but not all, of the leading companies reporting at best sales of tens of thousands of copies of their new games instead of hundreds of thousands of

INNOVATION AND THE LIMITS OF STATE POWER

219

copies. These developments coincide with the maturation of the international gaming industry. The quality and technological sophistication needed to make competitive game titles have drastically raised the costs of development per title. In addition, the Korean gaming industry, for many years the poor and unsophisticated cousin of the Taiwanese industry, was the first to develop online FRP games, winning market share and financial backing and overtaking the quality and technological sophistication of the Taiwanese industry at a critical moment. At the end of 2003, all but one of the extremely successful massive-multiusers-online FRP games running in Taiwan were developed by non-Taiwanese companies. The Taiwanese software games industry, without the financial resources of the Korean and Western gaming industries, finds itself technologically backward and stuck with business models that are based on low-cost development. It has been unable for the most part to compete. Individually, gaming companies do not have the financial resources to regain the technological lead vis-à-vis the Korean industry, and as the industry is widely fragmented it has been unable to coordinate collective action. A few industry leaders approached III and asked for leadership or help, only to be rebuffed until October 2002. A CEO of one of the biggest companies recounted a tale of a failed attempt to cooperate with III in late 2002: The worst of the worst is III. A representative of the gaming industry approached the president of III, the new one who promised to change III for the better and help the industry, and asked him to help our sector. Mind you, we are the only sector that was really successful in the Taiwanese software industry. After a few talks, his answer came back from his secretary—“this is not an III core business, therefore, we are not interested.” I ask you: “What the hell is their core business?” They say their goal is to help the software industry, not to make profits, but they do not give a damn about anything but making money. (IPC interview, January 2003)

This failure by III is even more surprising because in October 2002 the Taiwanese government declared digital content as a key priority technological area and specifically targeted online games as one of the more promising digital content sectors. However, both government officials and industry leaders see the digital content initiative as a failure so far. One of the industry’s representatives in the initiative described it thus:

220 CHAPTER 7

This is the latest government fad; they finally realized intellectual property is important. So we sit there for a year, so many important people and so many people from the government, and talk and discuss for hours every week and nothing happens. It is running for over a year now and still nothing happened. Elections are coming in six months time so nothing will happen until after the elections because the civil servants are waiting to see who will be their master and they are afraid to do anything. (IPC interview, October 2003)

This view of the efforts of the digital content initiative in the area of online gaming was also expressed by the top official responsible for it, describing their efforts as follows: Nothing much is happening with online games. We basically did nothing; we do not even have a budget. We have to admit that a lot needs to be done with software and digital content. However, we [the government] still think that Taiwan’s future is more in hardware. (IPC interview, November 2003)

As of the end of 2003, most of the Taiwanese gaming software companies are fast retreating from game development. From an industry whose leading companies had operations in and sold their own published products throughout most of the greater China region, the industry is transforming itself into one whose core competency is games distribution and resale. This does not mean that two or three of the leading companies will not continue to prosper and grow and develop original games. However, the number of Taiwanese game developing companies is declining sharply with no new entrants coming to replace the exiting firms. In the last few years a cohort of software companies with new business models has emerged and succeeded. These companies are much more technologically oriented. Their products either directly deal with software technology itself, or with new applications of IT technology, for example, antivirus software, optical character recognition (OCR) applications, or systems recovery. These companies appeared after the success of the IT hardware industry in Hsinchu Park; indeed, many of them are tightly connected to the industry. There are two types of operations: (a) supplying the Taiwanese hardware industry with software technology that enables it to add features to its products, differentiating them

INNOVATION AND THE LIMITS OF STATE POWER

221

from the competition, or supplying critical software that the hardware industry would find difficult or prohibitively expensive to get abroad; (b) producing software products that are directly associated with the software industry itself, such as antivirus or application development tools. One prominent example of a company focusing on software technology itself is Trend Micro, with its corporate antiviral protection products. In 2002, Trend Micro became one of the largest software security companies in the world, with sales of US$364 million. Trend was established by Taiwanese, grew in Taiwan, is run by Taiwanese, and still conducts a large share of its activities in Taiwan, but its management felt that it needed to leave Taiwan in order to be globally successful. In 1998 Trend moved its management team and headquarters to Japan, reestablished itself as a publicly traded Japanese company, and is also traded on NASDAQ. The two most successful software companies that followed more closely a business model of alliance with Taiwan’s hardware manufacturers are Ulead and Cyberlink. In 1989, three friends who worked together at III established Ulead with financing from the Taiwanese scanner manufacturer Microtek. Ulead’s first business was to supply OCR and imageprocessing software to the then fast-growing Taiwanese scanner industry that was facing difficulties in securing key software from U.S. companies. A few years later, realizing that there is no true color imaging processing editor for the PC (Adobe was selling its Photoshop software only for the Macintosh at the time), Ulead launched its own product, called Photostyler. In 1992 Adobe bought the company that held the copyrights for the technology Ulead was using. Since then, Ulead has come out with its own product for the midrange user. Using the same OEM model, Ulead wrote software for video imaging for the video capture card industry. The Taiwanese video capture card industry never took off. Ironically, this proved to be a boon for Ulead, which started to work with foreign manufacturers. Today, apart from its image-processing products, Ulead also makes and sells video and DVD processing and authoring software. Ulead sells either directly to private users or through OEM agreements with hardware manufacturers. In 1999, Ulead became the first software company to go public on Taiwan’s stock exchange. By 2002, Ulead had sales of over US$30 million and operations in all five continents. It is striking that one of the three most successful software product companies in Taiwan has total revenues that are still so relatively small.

222 CHAPTER 7

The latest globally successful Taiwanese software company is Cyberlink. Cyberlink is also interesting as it directly employs business techniques its CEO, Alice H. Chang, learned as a top executive in Trend Micro during Trend’s rapid growth and IPO period. Cyberlink is the most successful company to be spun out directly from a Taiwanese university lab, attesting to the growing capabilities of Taiwan’s computer science academic research. Jau Huang, Alice H. Chang’s husband, the cofounder of the multimedia lab at National Taiwan University, established Cyberlink in 1994 together with four of his students.13 Encouraged by Chang, then the executive vice president of Trend Micro, the team decided to develop a software product and not a hardware product. In 1995, the company was founded using self-financing. The company finished developing its first product, a video decoder (VCD decoder), in November 1996. In 1997, Chang stepped in as CEO and embarked on a strategy of OEM sales to Taiwan’s Video Graphics Array (VGA) manufacturers, the world’s largest. These OEM agreements gave Cyberlink immediate market recognition, and the company went on to develop a complete suite of DVD products that now has about 50 percent of the world market in the DVD PC multimedia tools niche. Today, Cyberlink sells multimedia management tools and has just started to sell development tools for e-training and e-learning solutions. In 2000, Cyberlink went public on the Taiwanese Stock Exchange. In the financial year 2002, Cyberlink had sales of over US$35 million with most of its sales originating with its DVD products, manufactured through OEM agreements with PC and notebook manufacturers in the United States, Taiwan, Japan, and Europe. In short, the Taiwanese software industry today seems to have a dual structure. On the one side are the older private companies dealing with business solutions together with III, which has become the biggest business software solutions provider. On the other side are the younger and more successful companies that are much more technologically oriented, with a few of them beginning to have true global reach. Overall, the Taiwanese software industry is still oriented more to the domestic market. Only a few companies possess the necessary capabilities and skills to develop products and services for the global market. Hence, relative to the Taiwanese hardware sector or compared with software in countries such as Israel or Ireland, Taiwan’s software industry cannot be considered very successful.

INNOVATION AND THE LIMITS OF STATE POWER

223

The Role of Institutions and Government Policy in the Structure of the Software Industry While III has played an inhibiting role in the development of the Taiwanese software industry, there are other reasons as well for the particular development path taken by this sector. First, one of Taiwan’s main industries has been PC manufacturing, and each company in that field was also busily producing software for its own machines, thus preventing the development of a large standardized market that could support independent product-oriented software firms. Second, until the early 1990s, partly due to internal competition within the state apparatus, there were no standards for Chinese input/output and internal conversion (Noble 1998). This spurred many companies to produce Chinese programs of the equivalent popular English packages, all of which were practically wiped out as soon as the Western companies published a Chinese version of their package. The third reason is that both the private hardware industry and the government did not see software as an independent industry but as a service component needed by other industries. This handicapped the industry in two ways: first, software has never been on an equal footing with hardware in the competition for government budget funds and grants; and, second, software companies faced many problems selling pure software solutions for reasonable prices in the market. The software industry has also suffered acutely from the lack of capital. Until the advent of the Internet, VCs were unwilling to invest at all in pure software companies unless they were already well established. Even today, software has a more difficult time securing VC financing than does hardware, as many interviewees in the VC industry, the Taiwanese VC association, and software companies pointed out. As for public finance, for many years pure software product development did not receive any grants. Even today, software is not included in the IDB’s tax incentives scheme. For these reasons, many of the older software companies, facing virtual exclusion from big projects, have branched out into other areas of operation, some of which have nothing to do with software, such as the distribution of gift packages. Thus, unlike their counterparts in the IC design sector, many of the Taiwanese software companies have suffered from a lack of focus and an inability to concentrate enough resources, management, and capital on R&D. The one

224 CHAPTER 7

segment that proved to be nimble, innovative, and successful during the 1980s and 1990s was gaming. In our interviews, many companies complained that, apart from a brief period in the late 1990s, before the latest government budgetary cuts, they have not been able to get any grants. In addition, the Taiwanese state policy of focusing on specific segments is not appropriate to the current state of the Taiwanese software industry. Its effects are to deny many promising software companies access to R&D capital, and it hardly promotes the creation of new companies in the targeted sector either. Companies that did manage to secure a state grant complained that the bureaucratic regulations forced them to spend more money on getting the grant than the grant itself provided. Those that did get the grants in the late 1990s, though, did tell us that, as other sources of financing were virtually nil, these grants were critical in helping them to finish their first R&D projects. Starting in the late 1990s, with the rapid growth and growing sophistication of the IT hardware industry in Hsinchu Park, the competing standards issue solved, III activities dissuading new companies from working on big business applications, and the growing skill base and sophistication in computer science in Taiwan, a new cohort of companies that is much more technologically oriented has appeared. Some of these companies have already managed to become global competitors in their niches and a few more seem poised to do so. These companies and the gaming sector have had very little to do with III. In sum, in the software industry, unlike the IC design sector, the industry has not been able to transform itself into a successful exporter and a global competitor. What has been achieved has taken place in spite of, rather than because of, the government’s main agency, III. In software, because of the characteristics of the industry and the timing of III’s creation, the Taiwanese policy of reliance on big public research institutions has proved to be a hindrance to the industry’s growth. Conclusion This chapter has analyzed at the micro level two subsectors of the Taiwanese IT industry, software and IC chip design, and focused on the state’s role. Using the same policy vehicle in both sectors—a big public research institution—the Taiwanese state has had both successes and failures in its attempts to foster the growth of the IT industry. The role of

INNOVATION AND THE LIMITS OF STATE POWER

225

the Taiwanese state in the development of the IT industry, in both hardware and software, strengthens the neodevelopmental state theories of the ways in which the state can promote innovation-based industry. The analysis also supports our contention with respect to the new development state theories. It is not that the state is either a neodevelopment state or not: the story of the Taiwanese IT industry shows that the same state can be both. The divergent results in hardware and software derive from the particular ways in which the state tried to implement its vision for a new innovation-based industry. If the state’s agencies, as in the case of ITRI and the hardware industry, have as their main goal the development of private industry, and are able to manage the transformation of their own positions as part of the state–industry co-evolution process, then the state can play a key constructive role. If, however, as in the case of III, the state agencies, even when organized in a structurally similar way and managed from above by the same leadership, do not see strengthening private industry as their primary goal, and are unwilling to relinquish their leading position, then the same state that so successfully developed one industry can be one of the main causes for the stagnation of another. This chapter analyzed both the limits of success in the more successful IC design sector and the successes in the generally much less successful software industry. It showed that both are critically affected by the interaction between state and industry. Taiwan’s IC design companies both utilize, and are limited by, Taiwan’s industrial structure and innovation system. Taiwan’s software companies excel in niches that are highly related to the Taiwanese IT hardware industry and untouched by III. In this chapter I have sought to expand our understanding of both the capabilities and the limits of a “developmental state” in its attempt to create new technological industries. My microanalyses showed not only the role that the interaction between state and business played in industrial development in both software and IC design, but also how the institutional settings in which they occur shape the final outcome. Thus, the emergence of the IC design industry within a system with large sophisticated OEM companies and the world’s most advanced pure-play foundries, themselves the product of the state’s efforts, propelled the IC design industry onto a specific development track. This institutional system also provided the IC design industry with certain unique capabilities, while discouraging investment in others. The software industry’s early

226 CHAPTER 7

evolution was strongly influenced by a business environment that did not view pure software as a “real” product. A state agency was both competing with it for its best customers and preventing it from getting the state’s direct financial help. The software industry’s later history was strongly affected by the needs and opportunities created by a sophisticated IT hardware industry in Hsinchu. For Taiwanese IC design houses to become truly innovative, major changes in the Taiwanese industrial and financial system may be required. But even today, the IC design subsector is successful globally and provides the Taiwanese IT industry as a whole with some important complementary assets, helping it to compete globally. The IT hardware industry’s strength underpins the continuing growth of the IC design industry. This provides evidence of Taiwanese potential in innovative service activities.14 A powerful new motor for revenues would come into play if Taiwan could fully employ innovation-services business models throughout its industrial system. ITRI’s new mode of operation with the IC industry, where ITRI conducts advanced R&D for the industry’s use, can be considered as a model. Finally, the state might rather easily modify some of its own policies to spur more technologically cutting-edge innovational activities. A simple example would be the redefinition of R&D in its various grant schemes in a way that is more similar to Israel’s, where in order to receive a grant, a company proposes a new innovative product or technology. This would ensure a financial critical mass for new enterprises that want to employ a new-product-development business model necessitating longer R&D stages without revenues from sales. In the software sector, the need to restructure policy and state agencies is more acute. As VC financing in all its forms is lacking, this is a clear case of market failure. The state role should be to channel more funding into industrial R&D activities. With the current policy, all the state financial support goes directly into III. It is not clear whether this promotes the creation of new product-based innovative software companies. Why should a state agency be the biggest IT consultancy, bespoke software developer, and system integrator? These activities have little to do with the creation of a vibrant domestic software industry. More research would be needed to develop a plan for reorganizing III. One possibility would be to spin off III’s main divisions as private companies, cutting them off from governmental financing and removing them from a preferred position in governmental projects. Another step might

INNOVATION AND THE LIMITS OF STATE POWER

227

be the reconstruction of software policy formation. Today III both suggests policies to different government offices and then competes on the projects to implement these proposals. A better system might be to limit III activity to policy formulation. In addition, if indeed the Taiwanese state is keen on the development of a globally successful software industry with a technological edge, it might rechannel the funds now directed to III into a grant program specifically for R&D software activities that are widely defined rather than subsectorally defined, for example, for digital content only. One direction for fruitful future research would be to consider the experience of other countries. For example, one useful model for a future software-focused government agency could be Ireland’s National Software Directorate (NSD), whose sole aim is to assist private industry and which does not involve itself with actual software development. The Irish NSD, with a staff of around ten permanent employees, concentrates on policy formulation and information gathering, acts as a bridge between industry and government, and initiates and looks after several innovative policy initiatives. It is widely seen as having played a critical role in the Irish software industry’s spectacular growth in the last ten years.

8 Cross-Straits Integration and Industrial Catch-Up How Vulnerable Is the Taiwan Miracle to an Ascendant Mainland? Edward S. Steinfeld

Introduction Few changes in the global economy seem as transformative as China’s emergence as a “shop floor” for the world. China’s ascent as a manufacturing powerhouse, occurring far faster and in far more complex industries than experts had ever imagined, has brought clear advantages to consumers worldwide. A host of high-quality goods, from apparel and home appliances to the most sophisticated electronic devices, are now available at unprecedented low costs. Yet, it is not the story of consumer benefits that captures much of the world’s attention. Instead, it is the competitive challenge posed by China’s rise. Chinese firms seem poised to supplant long-established counterparts in developed markets, and entire industries appear to be moving en masse to the Chinese Mainland. Whether in Europe, the Americas, or much of East Asia, concerns abound that China’s rise is precipitating the “hollowing out” of entire national economies, an erosion of industry that leads not just to job loss, but to the loss of capabilities long understood as the foundation of economic and national strength. Nowhere, of course, are these concerns more pressing than in Taiwan. In its own economic ascent over the past thirty years, Taiwan has positioned itself as a world manufacturer, fostering generations of firms focused not on branding, design, or marketing, but rather on high-end, high-value manufacturing, usually for better-known global firms that then sell the Taiwanese-manufactured product under their own brand. Mainland China, as it emerges as a global manufacturing center, appears 228

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 229

to directly challenge Taiwan’s core area of specialization, its fundamental source of competitiveness. Making matters more complex, Taiwanese firms themselves have played a significant role in driving these changes over the past decade, making investments on the Mainland that transfer not only capital, but also key capabilities. Finally, given its historically laden, highly contentious, and fundamentally unresolved political relationship with the Mainland, Taiwan has obvious security concerns surrounding China’s economic rise. This chapter provides an alternative perspective for understanding the Mainland–Taiwan economic relationship, one that departs from zerosum assessments of cross-border competition. This perspective departs from a conventional view of national economies as repositories for clearly defined industries, and industrialization as a linear evolution along an established developmental track. This chapter argues that a profound, essentially technologically driven transformation has occurred in the organization of production worldwide. China’s emergence as a manufacturing hub is only one element of this shift. Quantum advances in the management of information, through digitization, today permit vast amounts of data to be specified, manipulated, and transferred cheaply and instantaneously. Such changes allow complex production processes to be codified, disaggregated, and modularized. Tasks that once had to be performed under the roof of a single firm, or within the confines of a single nationally coordinated economy, can now in many cases be outsourced to a multitude of highly specialized, often geographically dispersed players arrayed across complex production networks. Production steps in the past were subsumed within self-contained, internally integrated production processes. These processes had to be mastered over time through the accumulation of vast amounts of tacit knowledge. They now can often be separated into distinct, codified, modularized activities. Some of those activities, even in highly sophisticated manufacturing, have become commodified, open to entry to the lowest bidders. For other activities, however, critical knowledge remains tacit, uncodifiable, and embedded in the production process. Here, the barriers to entry remain high, and the sources of competitive advantage less transitory. Mainland China’s emergence and Taiwan’s participation in that phenomenon must be understood in the context of today’s revolution in the organization of production. To the extent we actually recognize this revolution and begin thinking about the movement of economic activities

230 CHAPTER 8

Figure 8.1 Economic Growth in Taiwan (GDP at 1996 constant prices) $12,000

Amount (NT$ billion)

$10,000

$8,000

$6,000

$4,000

$2,000

19 51 19 53 19 55 19 57 19 59 19 61 19 63 19 65 19 67 19 69 19 71 19 73 19 75 19 77 19 79 19 81 19 83 19 85 19 87 19 89 19 91 19 93 19 95 19 97 19 99 20 01 20 03

$0

Year

Source: National Statistics of the Government of the Republic of China on Taiwan, www.stat.gov.tw/bs4/nis/EP2.xls

rather than entire industries, Mainland China’s rise appears neither so threatening, nor Taiwan’s source of competitiveness so fleeting as traditional frameworks might suggest. The Evolving Cross-Straits Economic Relationship From any perspective, the rapidity with which the Taiwan–Mainland China economic relationship has evolved, and the scale that it has achieved, are breathtaking. As indicated by Figures 8.1 and 8.2, the respective economies of Taiwan and Mainland China, while at dramatically different levels of development (Taiwan’s per capita income in 2002 was US$12,916,1 while Mainland China’s was US$9402), have been moving through overlapping, slightly staggered periods of sustained economic growth, with Taiwan’s period of most rapid expansion commencing in the 1970s, and the Mainland’s in the 1990s. The combination of geographical proximity, highly divergent levels of development, and—arguably—ethnic affinity has led to clear complementarities between the two economies, complementarities that have to some extent permitted mutual growth through deepening economic integration. Taiwanese investment on the Mainland ramped up in the late 1980s, and began surging along with the Mainland’s economic boom in the mid- to late 1990s. According to Taiwanese official government figures, Mainland-bound foreign direct investment (FDI) from Taiwan, only

Year

60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 20 20 20

$0

$200,000

$400,000

$600,000

$800,000

$1,000,000

$1,200,000

$1,400,000

Source: World Bank, World Development Indicators online, http://lib.harvard.edu/e-resources/details/w/wdionlin.html

Amount (US$ million)

Figure 8.2 Economic Growth in Mainland China (GDP at constant 1995 prices)

231

232 CHAPTER 8

US$174 million in 1991, surged to nearly US$3.2 billion by 1993 (Tung 2002, 36). By 1992, China had become the largest recipient of outbound Taiwanese FDI flows. Between 1991 and 2000, Taiwan’s cumulative FDI in China was US$17.1 billion, or 39 percent of Taiwan’s total overseas direct investment during that period (Tung 2002, 36). According to the Republic of China (ROC) government’s Ministry of Economic Affairs (MoEA), of Taiwan’s total officially approved foreign investments of US$10.09 billion in 2002, 67 percent (US$6.72 billion) was targeted for Mainland China. Figure 8.3 indicates Taiwan’s investment in Mainland China as a percentage of total FDI into Mainland China. As indicated by Figure 8.4, direct investment is the main channel through which foreign capital enters the PRC economy. Official data tell only part of the story. Until 1990, the ROC government explicitly banned Taiwanese investment on the Mainland. Since that period, though the ROC government has gradually reduced the most extreme institutional strictures on such investment, official limits still remain on the size of individual projects and the type of industries that may be involved. Until recently, the ROC government prohibited investments in the PRC over US$50 million, and required case-by-case review of investments in ostensibly “sensitive” sectors, such as information technology (IT), chemicals, real estate, and infrastructure. At the same time, however, many of these policy restrictions have been—and can still be—circumvented due to broader global phenomena, namely the deregulation and expansion of cross-border financial flows, and the increasing complexity and geographic dispersion of production chains. Since the 1980s, but with increasing alacrity in recent years, Taiwanese firms have proven extraordinarily capable of raising funds through overseas markets and then directing those funds toward Mainland China–based operations, often through shell companies based in third-party countries or territories. Hong Kong has been the most frequent, but by no means only, locale through which Taiwan’s Mainlandbound funds have flowed. The boom in PRC-bound FDI from Hong Kong and other third-party countries is partly driven by the disguised funds of Taiwanese firms seeking to avoid their own government’s restrictions. As indicated by Figure 8.5, the PRC government’s estimates of Taiwanese FDI into the Mainland are higher than the ROC government’s estimates, but given the savvy of Taiwanese investors and the complexity of global financial flows, it is likely that the real level of investment exceeds the estimates of both governments.

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 233

Figure 8.3 Taiwan’s FDI into Mainland China, 2002 (U.S.$ 100 million) Singapore. 23.37. 4 $ . Taiwan. 39.71. 3% Virgin Islands. 6 1 . 1 7 , 1 2 %

Germany. 9 23. 2 %

France. 5 7 6 . 1 %

Korea, 2 7 . 2 1 , 5%

Canada, S.SB. I v i

UK. 6 . 9 6 . 2 V M a c a u . J . 5 8 . i% Austral ig, 3.31,1%

Japan , 41.90. E%

' Malaysia, 3.66 1 % Thailand. t.8B. 0%

Oifters. 5 7 . 3 0 , 1 1 %

USA. 54.24, 1 0 %

HK. 1 7 8 . 6 1 . 3 3 %

O t t w . &2.9&54. 16 %

Source: State Statistical Bureau, Statistical Yearbook of China; State Statistical Bureau, China Foreign Economic Statistical Yearbook; Ministry of Commerce, PRC, Intertrade.com.cn

Figure 8.4 Total Foreign Capital Flows into China, 1995–2002 $60

Amount (US$ billion)

$50

$40

Total FDI

$30

Foreign Loans (data until 2000) Other Foreign Investments ROC FDI (ROC Source) ROC FDI (PRC Source)

$20

$10

$0 1994

1995

1996

1997

1998

1999

Year

Source: Same as Figure 8.3.

2000

2001

2002

2003

234 CHAPTER 8

Figure 8.5 Comparison of Estimated Realized Taiwanese FDI into Mainland China, 1991–2002 4,500

4,000

Amount (US$ millions)

3,500 3,000

2,500

Taiwan Estimates Mainland China Estimates

2,000 1,500

1,000 500 0 1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Year

Source: Ministry of Foreign Trade and Economic Cooperation, PRC, 2003, and Investment Commission, Ministry of Economic Affairs, ROC, downloaded from www.moeic.gov.tw Note: 1991 figures include prior date.

The cross-straits economic relationship mirrors a pattern observable across much of East and Southeast Asia. Economic integration at the firm level—through cross-border supply chains and complex production networks—has proceeded far faster than integration at the institutional and political levels. Entrepreneurial firms have proved willing and able to push cross-border integration in direct contravention of institutional, political, and diplomatic strictures. As positive economic results have become apparent, the governments have slowly and haltingly played institutional catch-up, scrambling to place rules over cross-border, interfirm commercial relationships that have often developed a dynamic of their own. While this cross-straits commercial relationship may be unique for its depth in the context of intense hostility between the governments involved, the acceleration of Taiwanese FDI into the Mainland mirrors broader trends of global investment in China. Taiwanese firms have ramped up their investments at precisely the same time as have other overseas players in China, and for many of the same reasons. Foreign firms face similar incentives to use China as a manufacturing hub, while at the same time potentially serving China’s rapidly growing demand for capital equipment and consumer goods. Taiwan’s Mainland-bound FDI has exhibited a pattern of sectoral evolution that also mirrors that of China-bound FDI from other countries

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 235

and regions. As indicated by Figures 8.6a and b, China-bound FDI has increasingly gravitated toward manufacturing activities, though the PRC official figures are not sufficiently fine-grained to illustrate the increasing specialization of manufacturing activities that many people have observed and that will be discussed later in this chapter. Particularly striking is the marked upsurge in Taiwanese funds directed toward electronics and electric appliance manufacturing in the Mainland beginning in the late 1990s and continuing through the present (see Figure 8.7). While Taiwanese FDI in the late 1980s and early 1990s involved primarily small to medium-sized firms (SMEs) engaged in labor-intensive manufacturing, Taiwan’s Mainland-bound investments by the late 1990s increasingly involved larger firms in more capital- and technology-intensive sectors, sectors for which China could serve as both a manufacturing base and a potential market. Investments in electronics typify these trends, culminating today in efforts like those of the Grace Group to build near state-of-the-art semiconductor fabrication and foundry operations in the Shanghai region. Mainland investments by Taiwanese firms have, in terms of their scale and the sophistication of the activities involved, come to resemble the China strategies of global players like General Motors, Motorola, Philips, or Nokia. Similarly, in terms of where in China they choose to invest, Taiwanese firms have also followed patterns characteristic of the broader population of overseas firms operating in the PRC. By the late 1990s and into 2001, as indicated by Figure 8.8, FDI into the Mainland was continuing in Guangdong province—the first region to be opened up through reform in the early 1980s—but was accelerating most rapidly in the Jiangsu province– Shanghai Municipality Yangzi Delta, a region that has emerged as China’s main hub of technology and knowledge-intensive manufacturing. As illustrated by Figure 8.9, FDI by Taiwanese firms, while already established in Guangdong, began to shift markedly to Jiangsu and Shanghai by 2000, again reflecting the emergence of this region in particular—to an even greater extent than Guangdong—as a center for capital-intensive, high-tech manufacturing of advanced electronic devices (i.e., computers, laptops, digital cameras, cell phones). Taiwanese investment has not, over time, concentrated in the one region of China most ethnically similar in language and cultural terms to Taiwan itself—Fujian province. Taiwanese investors have gone not where their ethnic or familiar ties are the deepest, but rather to the same regions that—whether for policy or other reasons—appeal to investors of any nationality.

Industry: 51173.68 46%

Source: State Statistical Bureau, Statistical Yearbook of China 1994.

Construction: 3878.37 3%

Transportation, Post & Telecom: 1489.91 1%

Real Estate: 43771.15 40%

FDI into PRC, 1993 (U.S.$ million)

Food Services, Material Supply: 4606.47 4%

Figure 8.6a

Agriculture: 1191.47 1%

Other: 4736.86 4%

Scientific Research: 587.75, 1%

Tourism: 1481.53 1%

Other Sectors: 3807.65 3%

Education, Culture, Arts: 451.73, 0%

Health Care, Sports, Social Welfare: 477.48 0%

236

Electric Power: Gas, Water, 2134.22, 3%

Construction: 1822.81, 3%

FDI into PRC, 2001 (U.S.$ million)

Scientific Research: 654.29, 1%

Education, Culture, Arts, Radio, Film, and TV: 71.74, 0%

Health Care, Sports, and Social Welfare: 133.05, 0%

Other Sectors: 1425.11, 2%

Mining and Quarrying: 644.48, 1%

Agriculture: 1761.74, 3%

Other: 6659.15, 10%

Social Services: 4288.84, 6%

Banking and Insurance: 86.12, 0%

Wholesale and Retail Trade and Catering Services: 1398.06, 2%

Transportation, Storage, Post & Telecom: 883.54, 1%

Real Estate Management: 5030.61, 7%

Geological Prospecting and Water Conservancy: 13.08, 0%

Source: State Statistical Bureau, Statistical Yearbook of China 2003.

Manufacturing: 48846.86, 71%

Figure 8.6b

237

1992

1993

1994

1995

Electronic & Electric Appliances Precision Instruments Food & Beverage Processing Plastic Products Basic Metals Products

1996

Year

1997

1998

1999

2000

2001

Source: Investment Commission, Ministry of Economic Affairs, ROC 2003, data downloaded from www.moeic.gov.tw

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

3,000,000

Figure 8.7 Taiwan FDI into PRC by Industry, 1992–2002 (U.S.$ thousands)

2002

238

Source: State Statistical Bureau, Statistical Yearbook of China 2003.

g n i ijin nji be nxi lia g Be Tia He ha go nin Jilin ng ai) g i S on ao jia gh an hu an xi g M Li ng an heji An uji ng on an ei n g r o l F ia nd en ub na n xi n ei h Z J a H H u o g a ne n H cl. S In H gd an ain ua hou an et i Sh i( n an Gu H ich uiz nn Tib anx nsu ai ia u S G Yu u a a gh x ng G gs Sh G Qin ing njia an N Xi i J Province

$0

$2,000

$4,000

$6,000

$8,000

$10,000

$12,000

Figure 8.8 PRC FDI by Province, 1985–2001 (U.S.$ thousands)

1985 1993 1996 2001

239

240 CHAPTER 8

Figure 8.9 Taiwan FDI into PRC by Province, 1992–2002 3,500,000

Guangdong 3,000,000

Jiangsu Fujian Hainan

(U.S.$ thousands)

2,500,000

Hebei Northeast Region

2,000,000

1,500,000

1,000,000

500,000

0 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

Year

Source: Same as Figure 8.7.

Finally, in terms of the structuring of their investments, Taiwanese firms on the Mainland have done what most foreign firms have done— they have sought control through the maximization of ownership stakes. Given uncertainties in the Mainland’s legal environment, direct ownership over business activities becomes preferable to arms-length contractual transactions or somewhat closer-in joint–venture (JV) arrangements. Ownership at the very least accords control over personnel, and affords opportunities to station expatriates in key managerial positions. The tendency toward direct ownership is further encouraged, as Yasheng Huang (2003) points out, by the Chinese government’s favoring of foreign-invested firms over domestic private firms, and thus creates on the Chinese side an unusually high demand for FDI. Chinese firms want to become foreign owned, or in a sense, need to become foreign owned—primarily by welcoming in FDI—simply to negotiate a tough regulatory environment at home. The end result, whether for Taiwanese or other foreign firms, is that when they establish operations on

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 241

Table 8.1 Input Sourcing by Taiwan-Invested Enterprises in China, 1995–1998 (%) Raw materials 1995 1996 1997 1998 53 46 49 44

Period Taiwan Taiwan-invested enterprises 17 in China Non-Taiwan-invested 18 enterprises in China* Other countries 12

Parts and semifinished products 1995 1996 1997 1998 56 51 53 49

21

17

19

18

21

20

23

21

22

23

19

21

19

22

13

13

13

7

7

8

8

Source: Table originally from Tung, China’s Economic Leverage, p. 46 (Tung employed the sources listed below): • Ministry of Economic Affairs (ROC), Zhizaoye duiwai touzi shikuang diaocha baogao [The Investigation Report on Outward Investment of Manufacturing Industry] (Taipei: Ministry of Economic Affairs, 1997), pp. 89, 92, 95, 98. • Ministry of Economic Affairs (ROC), Zhizaoye duiwai touzi shikuang diaocha baogao [The Investigation Report on Outward Investment of Manufacturing Industry] (Taipei: Ministry of Economic Affairs, 2000), pp. 29–30. *Including Chinese enterprises and other foreign-funded enterprises.

the Mainland, the ownership stakes they carve out tend to be higher than what they would seek for similar operations located outside the Mainland. Taiwanese firms in China remain deeply integrated with the home region. As indicated by Table 8.1, data for the late 1990s show that Taiwanese firms in China remained highly and consistently dependent on Taiwan-based firms for raw materials, parts, and semifinished products. While a slight trend toward localization can be observed, at least part of that localization process involved the transplanting of Taiwanese-owned sourcing networks onto the Mainland. In other words, Taiwanese-invested firms on the Mainland had shifted slightly toward greater sourcing within the PRC, but often from other Mainland-based Taiwanese-owned firms. In summary, what can be observed are increasingly dense connections between two economies at complementary, phased stages of development, albeit two economies separated by a vast political division. Despite this political division and the resulting absence of cross-border institutions, commercial integration has proceeded dramatically, and in synchrony with Mainland China’s integration into the broader global

242 CHAPTER 8

economy. Yet, as the following sections will argue, what has occurred here is more than just phased development, the movement of sunset industries from advanced to backward economic locales. It would be a mistake to conceptualize this as an economic giant, Mainland China, aggressively climbing an established ladder of development, and in the process, directly threatening to supplant its more advanced predecessors, who happen to be clinging desperately to the rungs above. Shifting Architectures of Global Production and Mainland Chinese Competitiveness As noted earlier, technological advances in the management of information, particularly digitization, have dramatically altered the architecture of production globally. Digitization, by facilitating the management and transmission of vast amounts of information, has allowed the codification of highly sophisticated processes of production.3 Once codified, processes can be split into discrete steps— modules, in effect—and standards to ensure their connectivity can be established (on modularization, see also Baldwin and Clark 2000; Aoki and Takizawa 2002). 4 Modularization, in turn, has permitted activities that once had to be co-located and managed within the confines of a single firm to be spread out across great geographic and organizational expanses. The issue is not that any activity can be done anywhere, or that all manufacturing has become completely modularized, but rather that a number of new options for structuring activities now exist (for a comprehensive taxonomy, see Fujimoto 2002, 1999). For some processes— or parts of processes—individual steps have become completely modularized, and the rules of connectivity with upstream and downstream steps are fully codified and stable. In such cases, the module stands alone both functionally and structurally, thus allowing it to be easily “plugged in” to other modules in the overall production process (or even in multiple, seemingly unrelated, production processes). At the other extreme are processes whose component steps cannot easily be codified and disaggregated. They may be separated geographically and organizationally, but their integration into a final product requires extensive coordination and communication among the producing parties. The products of each step have no “stand-alone” capability, but instead must be carefully tailored to, coordinated with, and often co-designed

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 243

and developed with specific upstream and downstream components. This sort of “integral” production architecture may be pursued as a matter of choice by a lead firm (i.e., a vertically integrated organization), but it also may be dictated by the state of technology. 5 Taken together, these emergent architectures of production have had a deep impact on Mainland China’s development (as has China’s development on them). Chinese enterprises have skillfully exploited the opportunities of modularization, aggressively upgrading their manufacturing skills so as to meet outsourcing demands by leading global players. In some cases, Chinese firms have autonomously pushed the replacement of traditional integral architectures of production with more modularized, open forms, thus forcing the commodification—and outsourcing to China—of certain activities, regardless of the preferences of overseas lead firms. Yet, while modularization affords new opportunities, it also creates major vulnerabilities, both for Chinese producers and other entrants. Fully modularized, codified, open production architectures—virtually by definition—entail the manufacturing of standardized, nondifferentiated products. Firms focusing on such activities have little choice but to compete on the basis of low cost and high volume. Moreover, they continually run the risk of being unseated by the next lowcost entrant, particularly since fully modularized products are easily substitutable from the consumer’s perspective. That Chinese firms have mastered modularized production accounts for China’s emergence as the globe’s shop floor. It also accounts for the fact that Chinese firms across a variety of sectors today find themselves locked in mutually destructive price competition. Finally, it accounts for the fact that Chinese enterprises, for all their success, have had very little luck in capturing many of the higher-value manufacturing and overall supply chain management activities that are still dominated by overseas, and often Taiwanese, firms. The lesson is that once new entrants commence modularized production, they soon face pressure to upgrade and carve out more sustainable bases for competitiveness. There are essentially three possible responses. The modularized producer can attempt to control the supply chain by actively setting rather than passively accepting rules of connectivity in the upstream and downstream directions. 6 In other words, the producer creates the modularized product that everybody else must design around (for example, an Intel processor). Alternatively, the producer might elect

244 CHAPTER 8

to shift away from modularization, and move back toward more integral processes, ones that must be coordinated and co-designed with upstream and downstream partners in the network. Finally, as is done by many leading global players, the firm may compete by providing key services— overall product definition, branding, and marketing—that shape the entire supply chain and command the bulk of the final product’s value. All these options require innovation in some sense, a daunting challenge for even the most sophisticated commodity manufacturers. Again, the shift toward more modularized, networked forms of production globally offers both opportunities and pitfalls in this area. To the extent that a modular manufacturer is engaged in multiple supply chains— that is, by producing a stand-alone component that can be plugged into a variety of downstream products—the manufacturer’s fate ends up tied to no single final product in particular. 7 Hence, the manufacturer is free to innovate in ways that not only incrementally improve existing downstream products (“sustaining” innovation), but also in ways that unseat such products by facilitating new substitutes (“disruptive” innovation). 8 Similarly, open, modularized supply chains permit the rule makers—those determining the rules of connectivity— to shift the standards, and thus force the “rule takers” to scramble in compliance. 9 Such freedom undoubtedly contributes to the extremely rapid product cycles and dizzying pace of innovation characteristic of high-tech industry today.10 Yet, it also creates major vulnerabilities for the rule takers, the followers, the commodity producers, and all the rest of the supply chain participants that must respond to innovative lead firms. Mainland Chinese Domestic Firms: Corporate Structure and “Price Wars” Much of Chinese domestic industry today is characterized by smallscale firms competing intensely on the basis of discounting. In theory, this could be understood as a prelude to industrywide shakeouts that should eliminate smaller firms and consolidate activities into a few larger producers, presumably the sorts that might engage in industrial upgrading. Evidence of such progression, however, remains sparse. Undoubtedly, thousands of firms—particularly in the state sector—have either been liquidated or substantially restructured in the 1990s, often with significant societal ramifications (Tenev and Zhang 2002; Hu 1999). Similarly, the private sector has burgeoned, with enterprises entering

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 245

Table 8.2 Firm Size of Higher-Technology Enterprises in China, 1995–2000

Observations All firms Beijing Shanghai Tianjin Guangzhou Chengdu

2000 1,500 611.9 872 798 352 539 499

Average number of employees 1999 1998 1,424 1,312 635 686 931 985 819 956 339 353 552 568 532 572

1995 1,057 798 1,152 1,203 382 582 681

Source: Data from 2001 World Bank Survey; see Yusef, Altef, and Nabeshima 2004.

and exiting at high rates (Gregory et al. 2000). That said, there has to date been relatively little evidence of rationalization, whether in terms of coherent consolidation across given sectors or movement into highervalue proprietary activities by individual firms. Instead, a pattern of corporate organization has persisted that sets Chinese firms apart from many of their global counterparts, and certainly from the lead firms in global supply chains.11 First, and not surprisingly, given China’s relatively recent emergence, Chinese firms tend to be both newer and smaller in scale than their global counterparts. In the World Bank’s 2001 survey of 1,500 higher-technology enterprises in China, firms averaged just over 600 employees (Table 8.2), and generally had been in existence for only ten to fifteen years.12 Even China’s more famous firms—those with known brands and national, if not global, status—tend toward the smaller side. China’s premier computer and IT firm, the Legend Group, now known as Lenovo, had revenues in 2001 of approximately US$3.48 billion.13 Its nearest multinational counterpart, IBM, which in late 2004 sold its personal computer division to Lenovo, had revenues that year of US$85.9 billion.14 The Chinese stateowned petroleum company Sinopec had revenues of US$34 billion in 1998, compared with US$182.3 for Exxon/Mobil in 1997 (Nolan 2001, 166). Capital Iron and Steel had revenues of US$2.16 billion in 1998, compared with Nippon Steel’s US$21.6 billion. Haier—China’s premier home appliance manufacturer, one of the best-known Chinese brands internationally, and the fifth-largest producer of white goods (large electrical home appliances like refrigerators and washing machines) worldwide in terms of market share—is dwarfed by its global competitors: Whirlpool, Electrolux, Bosch-Siemens, and General Electric.15

246 CHAPTER 8

Second, Chinese firms, though their output often ends up either in foreign hands or in overseas markets, tend to be extremely localized in terms of their actual operations. In the World Bank’s 2001 survey, 41 percent of the manufacturing firms in the sample reported producing to specifications set by foreign firms, though, again, it should be noted that these are higher-tech firms, ones for which we should expect higher levels of foreign interaction and more sophisticated production activities.16 Twenty-one percent reported directly producing parts for foreign firms, while 25 percent reported producing final products for such customers. Indicative of China’s liberal policies toward foreign direct investment, 25 percent of all firms in the survey reported having foreign equity partners, with the foreign ownership stake on average hovering just over 50 percent. Despite all this foreign interaction, however, the firms’ upstream supply network and downstream customer base tended to be confined geographically. The 2001 survey suggested that, on average, over 50 percent of upstream suppliers were located in the respondents’ own respective cities.17 Approximately 75 percent of the supply network on average was located within China. Downstream, the survey indicated for the average Chinese firm, approximately half of the customer base is located within the firm’s own municipality.18 Approximately 15 percent of the customer base on average was reported to be overseas. Whether for upstream or downstream interactions, rather traditional means prevailed—communication was conducted primarily by phone and fax, while goods themselves moved primarily via surface transportation.19 The localized nature of Chinese commercial networks leads to a third point, the relative shallowness with which Chinese firms integrate into global supply chains. Despite high levels of foreign ownership, only 15 percent of the manufacturing firms surveyed by the World Bank in 2001 reported designing parts for foreign customers, a sign that the respondents are essentially “rule takers” in open, fully modularized production processes. Only 7 percent reported providing customers with R&D or other specialized services. The figures are surprisingly low, given that the sample specifically targeted higher-tech sectors, the very ones in which we should expect high degrees of innovation, networking, and development of firm-specific proprietary knowledge. The firms were failing not only to design for downstream customers, but also to develop deep relationships of any kind with such customers, again, a sign of open, modularized production. Sixty-nine percent of the

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 247

survey respondents reported using trading companies to handle interactions with the broader customer base, thus suggesting essentially armslength, rather than deeply enmeshed, customer relations. In terms of identifying factors inhibiting greater exports, respondents focused on the difficulties of meeting foreign product standards, the high costs of meeting such standards, and particularly the intense cost competition they face (Table 8.3). Managers apparently preferred to produce for export markets, and few claimed that targeting the domestic market offered better financial gains, but managers perceived that their firms lacked the capabilities needed to meet foreign standards in a costeffective manner. At the same time, they perceived themselves to be in an intensely cost competitive environment, with pressures bearing down from both domestic and foreign counterparts. Presumably, one reason why managers find markets so price driven is that they neither sell directly to, nor design directly for, downstream counterparts. If a firm does not sell directly, it cannot develop the sort of service-based, specifically tailored relationships or products that lead to customer loyalty. If the firm—a component producer, perhaps—does not design directly for the downstream user, it is not likely to have deep interactions with that user, and hence will likely forgo the important learning opportunities (whether in terms of product or process technology, or even marketing skills and service provision) that might come from dealing with potentially more sophisticated customers.20 Nor will the firm likely be able to exert any leverage if the user decides to source the upstream component elsewhere. That then leads to a fourth and final point regarding innovative capacity. Chinese enterprises today face great pressure to upgrade their technological capabilities, and managers—as they did in the 2001 survey —routinely report high levels of what they at least perceive to be innovative activity. The pressures are understandable. Modern production, whether for ostensibly low-end goods like textiles or high-end goods like semiconductors, virtually by definition entails the management of complex processes, sophisticated and often capital-intensive machinery, highly refined product-specific materials, high expectations for quality control on the part of customers, and generally rapid turnaround times. Simply to be involved in global production networks today, even at the relatively low end, and even just to produce modularized products that can be “thrown over the wall,” new entrants must climb exceedingly steep managerial and technological learning curves, and they must do

Beijing 6 13 15 13 11 4 4 7 16 11

All firms 9 15 15 11 12 6 3 7 17 4

Source: Data from 2001 World Bank Survey; see Yusef, Altef, and Nabeshima 2004.

Shipping and transport costs Cost of meeting foreign legal and product standards Inability to produce to clients’ standards, specifications, and schedules Inability to match prices of domestic competitors who export Inability to match prices of foreign competitors Inability to meet demands by foreign clients for product upgrades and changes in specifications Difficulty of recovering payments from abroad Relative profitability of supplying the domestic market High costs of establishing a foreign distribution network Domestic content requirements

Main Inhibitors of Export Growth (%)

Table 8.3

7 2 8 18 7

12 10 17

Shanghai 8 11

9 5 6 12 9

13 14 11

Tianjin 7 13

7 4 4 13 8

19 8 4

Guangzhou 15 18

5 2 8 22 8

13 9 9

Chengdu 8 16

248

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 249

so rapidly. While modularized, networked production has opened up new doors for new entrants, the bar to basic participation has been raised quite high. That Chinese firms are so extensively involved in production for overseas markets represents a major achievement, and indicates extremely impressive learning on their part. Such learning, however, does not necessarily lead to “innovation.” In other words, it is not at all clear that these firms are developing intellectual assets, production skills, modes of serving customers, or actual products that can be understood as in any way proprietary. Rather, they are things that can be duplicated by hundreds or thousands of other firms in their immediate environment. In the 2001 survey, nearly half of all firms reported innovations in shop floor production processes, and another 46 percent reported innovations in managerial techniques, all measures that allow for the cutting of costs. What few, if any, of the firms reported were innovations that allowed the firm to charge a higher margin rather than a lower one—in other words, innovations that would encourage customers to pay a premium. Moreover, given the prevalence of product “wars” and cutthroat discounting among the proliferation of small producers in China, it appears that nobody has discovered the sort of proprietary cost-cutting solutions that afford competitive advantage over a reasonable period of time. Instead, the cost-cutting measures get duplicated from firm to firm, and the margins continue to erode. The response to this dilemma often entails another activity that survey respondents term “innovation,” the introduction of new products or entirely new lines of business.21 But, everyone else in the business environment leaps onto the same “new” product lines as well. Commodity producers end up chasing one surplus market after another, a pattern true even for China’s more advanced branded companies. The nation’s leading television manufacturers, including Konka—facing declining profits, rising inventories, and overseas import quotas—have moved aggressively, and en masse, into mobile phone manufacturing.22 Similarly, Guangdong Galanz Enterprise Group, a major producer of lowend microwave ovens, has coped with declining profits by jumping into the air conditioner market, a market already suffering from high inventories (Lee 2002). Galanz’s strategy, not surprisingly, has been to pursue extensive price discounting. Even the most established firms cope with increasing competition by aggressively discounting and expanding sales volume on existing products, entering new product areas in which they can compete again only on the basis of discounting and razor-thin

250 CHAPTER 8

margins, or finally, by trying to export their way out of trouble by pursuing overseas markets.23 In essence, firms focus on activities with low barriers to entry. Once the cost pressures become too intense, rather than moving upward into higher-end activities or taking the time to develop proprietary skills, the firms diversify into other low-entry barrier markets. The products themselves—be they televisions, personal computers, refrigerators, or data routers—are standardized. In the local context, therefore, “innovation” becomes associated with cost reduction and flexibility, the ease and rapidity with which a firm can jump from one saturated product market to the next. In this same local context (let alone globally) neither of these strategies has proved defendable over extended periods. Drivers of the Mainland Pattern: Style, Capacity, and Policy The pattern described above stems in large part from the interaction between three factors: governmental reform style, state capacity, and industrial policy. This interaction has at once permitted the integration of Chinese firms into the global economy and substantially constrained the extent of that integration, thus making Mainland firms highly dependent on the foreign firms that ultimately coordinate global supply chains. Reform Style Since the dawn of reform, China’s approach to market transition has been characterized by informality, experimentation, and decentralization (Gregory et al. 2000). Central leaders have set the overall policy aim (economic growth) and the basic constraint (the maintenance, in the vaguest terms, of “socialism”). Local officials, then, have been granted broad leeway to engage in policy experiments, virtually all of which have involved elements of market economics. “Socialism” is maintained simply to the extent that the experiments remain informal. When experiments prove successful, the center encourages their implementation —again on informal terms—nationally. If success continues, the experiments stand to be adopted post hoc as official government policy. Finally, in some—but not all—cases, the center formalizes the outcomes with new institutional rules, many of which directly challenge the initial condition of “maintaining socialism.” Through a certain element of linguistic legerdemain, that which began as an experimental alternative to

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 251

socialism (and hence its explicitly informal status) gets legitimized as socialism itself, albeit socialism “with Chinese characteristics.” The approach has proven brilliant in many respects. Without it, China’s transition to what much of the world terms capitalism could never have proceeded smoothly.24 It also explains how private enterprise—anathema just twenty years ago in China—now constitutes the predominant ownership form in Chinese industry, and has gained official status in the Chinese constitution.25 That said, there are negative ramifications of informality and decentralization as well. Entrepreneurial firms can thrive under such circumstances, and they can engage in international commerce, but their property rights tend to remain either undefined or—to the extent they tuck themselves under the auspices of a governmental bureau or state-owned firm— inaccurately defined. Without clear property or formal title to assets, these firms—as they would in any commercialized economy—face limited financing options. Borrowing from a bank becomes virtually out of the question. Instead, they have little choice but to self-finance, a situation that may ensure hard budgets, but one that also tends to limit enterprise growth.26 In a pattern consistent with that of virtually all firms in China save for larger state-owned enterprises (SOEs), enterprises in the World Bank’s 2001 sample reported relying primarily on retained earnings as their main source of financing (Table 8.4). Firms consistently reported that upwards of 50 percent of all financing came from retained earnings. Bank loans amounted to 19 percent of total financing on average, though the figures were somewhat lower in Tianjin (15 percent) and somewhat higher in Chengdu (24.8 percent). Equity financing, not surprisingly, given governmental quota restrictions on stock market listings, was low across the board (averaging 0.6 percent across the sample). Personal loans from family and friends constituted an important source of financing, averaging 8.6 percent of total financing for firms in the sample. Limited financing options, of course, frequently lead to tight liquidity constraints. The enterprise response often involves operating on a cash basis, but that rules out transactions that in more formalized systems allow for greater enterprise expansion.27 Furthermore, rather than investing in existing business lines and developing specialized skills, cash-starved firms jump to alternative businesses simply to maintain cash flow (i.e., if low-end television production isn’t generating revenue today, then move to low-end mobile phone manufacturing). Such

Observations 1,486 1,486 1,486 1,486 1,486 1,486 1,486 1,486 1,486

All firms 51.5 0.8 3.3 18.9 1.6 8.4 0.6 8.6 6.3 Beijing 51.7 1.1 3.1 17.0 1.2 7.6 0.1 7.8 10.3

Shanghai 51.2 0.6 4.2 19.3 2.0 11.9 1.1 3.2 6.6

Tianjin 49.2 0.7 3.0 14.9 1.1 8.6 1.0 17.2 4.4

Channels of financing by percentage

Source: Data from 2001 World Bank Survey; see Yusef, Altef, and Nabeshima 2004.

Capital from retained earnings/internal funds Capital from letter of credit Capital from supplier credit Capital from bank loans Capital from other financial institutions Capital from a parent or partner company Capital from equity finance Capital from personal, family and friends Capital from other sources

Sources of Enterprise Financing

Table 8.4

Guangzhou 50.1 0.4 4.0 18.7 0.7 8.9 0.1 8.6 8.5

Chengdu 55.1 1.0 2.8 24.8 3.1 5.1 0.6 6.2 1.9

252

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 253

diversification addresses liquidity issues, but it does not encourage the development of firm-specific proprietary assets or skills. Instead, firms remain stuck in low entry barrier activities. Informality, to the extent it dilutes the firm’s legal status, also limits the firm’s geographical reach. Without legal standing, the firm must engage predominantly in trust-based transactions (both in terms of with whom it chooses to deal, and who chooses to deal with it). The surest way to ensure trust is to stay local, essentially by buying from known local suppliers (or better yet, backward integrating to ensure reliable supplies) and selling to reliable local customers (so as to ensure payment). When dealing with international markets, the main option becomes to sell to a local trading company. For foreign companies dealing with such informal organizations, the strategy of choice often entails either buying from a more formalized state trading company or actually buying equity in the local producer itself. Indeed, FDI, to the extent that it places the recipient into the special regulatory category of “foreign owned,” constitutes a formalization mechanism, one that benefits provider and recipient alike. In some cases, Chinese firms sell their assets to foreign firms at a discount, but in so doing achieve a degree of formality that permits access to credit and insulation from arbitrary governmental policy (Huang 2003).28 Like informality, governmental decentralization leaves its own mark on entrepreneurial organizations. Many local governments in the reform period have eagerly promoted economic development and, as part of that goal, have frequently promoted local entrepreneurship (Oi 1999). They have been less eager, however, to facilitate development that benefits areas beyond the locality. Early in the reform era, this reluctance manifested itself in regional trade wars and overt barriers to interprovincial trade.29 More recently, in the 1990s, given central crackdowns on overt protectionism, localities have used more subtle methods: selective enforcement of product standards, more rigorous registration and licensing requirements for outsiders, and prejudicial application of health codes, just to name a few (Institute of Industrial Economics 1998, 294; Naughton 1999, 20–21). The more positive interpretation of Chinesestyle decentralization is that localities have been permitted to build institutions that best suit local circumstances and best support local development.30 The more negative view, one not wholly irreconcilable with the first, is that institutional ingenuity today has come to mean local content and “buy local” rules for local producers (“local” defined

254 CHAPTER 8

as “within the municipality”), discriminatory regulatory enforcement for outsiders trying to enter the local market, and restrictions on enterprise mergers and acquisitions.31 Managers on a day-to-day basis may not be spending the bulk of their time on governmental matters, but it is fairly clear that administrative factors impinge upon, and often restrict, commercial activities. Similar conclusions can be drawn regarding sectoral and geographic rationalization in industry. Whereas rational mergers and acquisitions are frequently blocked through various administrative interventions, commercially irrational mergers are often imposed by local administrative fiat. Particularly in the state sector, financially sound firms have been forced, often under considerable duress, to assume ownership of insolvent organizations simply to stave off bankruptcies.32 That the acquiring firm is sometimes accorded preferential policy treatment as a sort of quid pro quo only further distorts budget constraints and incentives for productive growth.33 More generally, when firms are forced to merge with failing local neighbors or to source only from local counterparts, they are often indirectly prevented from interacting with the best, most advanced suppliers. In an era of networked production, when innovation is understood as emanating as much from interfirm learning across the supply chain as from isolated activities within the individual firm, linking up with the best upstream and downstream partners becomes a key component of upgrading. Administratively imposed restrictions on such linkages, particularly restrictions that limit the linkages to a given municipality, prevent Chinese firms from accessing not only the best global suppliers, but even the best national ones. Deprived of high-quality components and important learning opportunities, many Chinese firms are pushed only farther down the road of low-end manufacturing and cost-based competition. Moreover, when localities try to keep the firm local, the firm’s problems of small-scale and limited financial resources simply deepen.34 State Capacity By the later 1990s, the architects of Chinese reform began to tackle many of the problems discussed above. It became understood that informality and decentralization, while critical initially for achieving local acceptance of reform (and for easing the ideological problems associated with what ultimately became a wholesale adoption of market

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 255

economics), had run their useful course and were now inhibiting further growth. Efforts across a variety of areas to formalize China’s market system have risen to the top of the policy agenda. The problem, however, is that these imperatives have collided with the reality of limited state capacity in China. The capacity issue manifests itself in at least two respects: the ability of the center to coordinate policy across the government’s administrative hierarchy, and the ability of the government as a whole to regulate commercial activity in the civil sphere. The first problem, discussed in the previous section, has arguably receded in recent years. The second, however, has proven more vexing. As might occur in any developing economy, the Chinese system has experienced a dramatic increase in the complexity and density of interactions between economic actors, most of which are no longer under the direct administrative control of the state. Across the board—whether in terms of financial relationships, contracts, issues of corporate control, or intellectual property rights— demand within the civil sphere has increased for both objective rules and reliable enforcement (Peerenboom 2001). Understandably, given the rapidity of China’s economic growth, the demand for market governance has outpaced the ability of the state to provide governance-related public goods. Courts are overwhelmed with cases, judges are often inadequately trained, and enforcement mechanisms are generally weak at best (Alford 2000; Lubman 1999). It is widely recognized in China today that the rule of law is essential for sustained growth, but it is far less clear how it can be achieved or even exactly what the rule of law entails (Peerenboom 2002). Meanwhile, the absence of effective legal institutions encourages the sorts of rent-seeking behavior that further erode trust in commercial transactions and society more broadly. In the financial area, for example, we have witnessed the emergence of what some Chinese describe as a “nonpayment” economy (Zhou 1999, 6). Commercial buyers make purchases and then refuse to pay. Borrowers take out loans, and then default. Banks accept deposits, and then squander them in ill-advised lending. In each case, the victim is left with little recourse. As it is said in China, “You sue, but the court won’t accept your case; the court accepts your case, but won’t begin the trial; the court begins the trial, but won’t issue a judgment; the court issues a judgment, but then doesn’t enforce it” (qi gao bu shou li, shou li bu kai ting, kai ting bu xuan pan, xuan pan bu zhi xing) (Zhou 1999, 6).

256 CHAPTER 8

What results is neither utter lawlessness nor an absence of growth. Instead, there exists a meandering track of unclear rules, low levels of trust, and frequent circumnavigation of legal strictures. These conditions, as indicated earlier, all impact on the organizational structure and global competitiveness of Chinese firms. At the very least, the environment impinges on both the capacity and inclination of firms to innovate. Low state capacity also impedes restructuring of a key economic chokepoint, the bank-dominated national financial system. For the first fifteen years of reform, the state-owned banking system was employed as a quasifiscal mechanism to channel funds into state-owned firms (Lardy 1998; Steinfeld 1998). Over time, this produced a buildup of nonperforming loans (NPLs) and a condition of overall technical insolvency (for recent data, see Ma and Fung 2002). To its credit, the central government, particularly since the Asian financial crisis, has acknowledged the problem and has undertaken a series of measures to encourage the recapitalization and commercialization of the financial system (Steinfeld 2000). Yet, while it has proven able by fiat to freeze credit provision, the government in the past five years has been unable to produce modern financial regulation or achieve anything resembling healthy commercialization. The banking system remains mired in conditions of moral hazard: bank balance sheets remain awash in red ink, large borrowers have not been shut down, real loan write-offs have been deferred, and bankers have then been unleashed to lend money on “commercial” terms. Under such conditions, it is no wonder that NPLs continue to accumulate (Kynge 2002).35 In 1999 the Chinese government established four asset management companies (AMCs) to take on NPLs from the four main state banks and begin recovery efforts. Since the inception of this debt-equity swap program, distressed assets valued at approximately RMB 1.4 trillion, some 20 percent of the value of the banking system’s entire loan portfolio, have been transferred to AMCs (Ma and Fung 2002). Even after these massive transfers, though, state banks have still been reporting high levels of NPLs. Officials of the China Construction Bank reported that, as of the end of 1999, NPLs still constituted at least 30 percent of the loan portfolio (Fang 2001). In 2001, the president of the Bank of China reported NPL levels at 26 percent of outstanding loans, and this was again after some 20 percent of the bank’s total loan portfolio had been transferred to an AMC. Along similar lines, analysts at the Bank for International Settlements have recently estimated that, at the end of 2001, China’s NPLs amounted to 35 percent of GDP (Ma and

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 257

Fung 2002, 3). The overall sense is not only that the stock of NPLs is large, but also that the flow is still increasing. Under such conditions, it is also no wonder that Chinese industrial producers—particularly the newer, more dynamic firms that have no established relationship with state banks—face tight liquidity constraints. Industrial Policy The main question is about ultimate governmental aims. China throughout the 1990s has pursued institutional reforms that encouraged market deepening and a leveling of the playing field among all participants in the economy. Nonetheless, the Chinese government has abandoned neither the notion of state ownership nor the ambition of building key state firms into vertically integrated, globally competitive “national champions.” Traditional Japanese and South Korean–style industrial policy still enjoys considerable appeal in China. Industrial Policy as the Unleashing of Comparative Advantage Contemporary Chinese industrial policy in many ways has a rather schizophrenic quality. On one side, the government has pursued what it describes as a “comparative advantage” strategy of development (Chen 2002). In what amounts to be an essentially Heckscher-Ohlin-Samuelson approach, the strategy is premised on the idea that development and industrialization are natural by-products of the convergence of factor prices across nations. The country’s relative factor endowments at any particular time are taken as given (in China’s case, surplus labor and scarce capital), and then development is understood as specialization in the production and export of goods intensive in the use of the abundant factor. As long as external trade and internal markets are opened up—conditions that become central goals of this sort of industrial policy—a dynamic international division of labor should ensue. Poorer countries like China begin by exporting labor-intensive products, but as capital flows in (seeking higher marginal returns in the capital-scarce environment), the developing country supposedly spontaneously climbs the ladder of industrial sophistication and capital intensity. Over an extended period, provided that markets remain liberalized, industrialization should proceed in pace, and factor prices should ultimately equalize across countries.36

258 CHAPTER 8

Policymakers in Beijing, at least on this particular side of industrial policy, have followed the theory’s prescriptions, albeit with some modifications. Reform, since its very inception, has been promoted as a process of “opening up,” and opening up specifically to foreign trade, knowledge, and technology. China throughout the 1980s, and particularly after 1992, dramatically reduced statutory import tariff rates (Lardy 2002, 37). Since 1997, the government has also substantially expanded policy initiatives that exempt certain domestic firms and institutions from paying the import duties that formally do exist (Lardy 2002, 36). Finally, in 2001, China formally became a member of the World Trade Organization (WTO), binding itself to an accession protocol more expansive, in terms of both market access and permissible trade practices, than that faced by any other developing nation in history (Lardy 2002, 63–105). Moving from informality to formality, the guiding assumption for this face of industrial policy is that the surest way to stimulate development is to expose domestic firms to international competition. Equally important, reformers have aggressively courted FDI, pursuing what has amounted to the most liberal FDI policy of any Asian developing nation. Here, a bit of practicality has tempered slavish devotion to theory. Heckscher-Ohlin-Samuelson theories assume knowledge to be perfect. That is, as long as capital and labor are allowed to flow freely, prices should equalize across countries, and productivity should equalize across firms. The actual knowledge of how to produce is presumed to be trivial: it simply flows across the ether like a television program washing across the eyes of a passive viewer or a library book changing hands freely from borrower to borrower.37 As long as the prices are right, the firm is presumed capable of producing. Noncompetitiveness, therefore, can be attributed primarily to bad policy: government distortion of prices, excessively high wages, and illiberal trade regimes.38 Policymakers in Beijing, though, have hedged their bets on this front. Instead of waiting passively for “natural” transfers of knowledge and technology, they have chosen proactively to build a vector: foreign direct investment through industrial JVs. The idea is that for Chinese firms to upgrade, they must work hand in hand with leaders from the developed world. In exchange for their transferring technology and know-how to Chinese counterparts, outsiders are then granted privileged access to the Chinese domestic market or preferential treatment on other grounds.39 The efficacy of such policies can, of course, be debated. Some scholars,

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 259

by pointing out the discounts at which foreigners have been able to purchase industrial assets, have suggested that China has been inadequately compensated for such transfers (Huang 2003). Others have suggested that traditional Chinese industrial firms devote far too much attention to hard technology—physical assets—and in the process severely neglect the training and expertise needed to absorb those technologies (Gilboy 2003). Over two decades, China’s FDI policies—not to mention its liberal policies toward emigration—have led to a monumental scaling up of managerial expertise in the country.40 Whether in foreign firms or domestic, an essentially world-class population of managers has been created at the highest tiers of the economy. Increasingly, this population has begun to flow back and forth between employment in foreign and domestic companies, and between employment within China and outside. Of course the question is whether, particularly in domestic firms, these managers can operate in an institutional environment conducive to enterprise success.41 Skilled domestic managers now exist, but can physical assets really flow? Will commercially moribund firms—legacies of the prior era for the most part—be allowed to go under, and entrepreneurial firms be permitted to rise from their ashes? Again, at least on this side of the industrial policy ledger, major strides forward have occurred. Between 1994 and 2000, with the government’s policy of zhua da fang xiao (“grasping the large, and releasing the small”), almost 60,000 small to medium-sized SOEs have been “restructured,” a term that generally signifies outright liquidation, privatization, or transfer to employee ownership (Tenev and Zhang 2002, 30). At the same time, the private sector has been permitted to burgeon, and now constitutes the largest single ownership form in Chinese industry. In the past, “enterprise reform” in China meant measures to improve performance in existing state-owned firms. Today, “enterprise reform” has increasingly come to mean measures for eliminating poor performers. Conceptually, then, this particular guise of Chinese industrial policy— the exposure of firms to foreign competition, the encouragement of FDI and knowledge transfer, and the ruthless downsizing of poor performers—can be understood as a “creative-destruction”–centered vision of development.42 Industrialization becomes the progeny of market forces, and those forces themselves are understood as the mechanism for winnowing winners from losers. The continual composition and decomposition of constellations of assets—in other words, the rise and fall of

260 CHAPTER 8

firms—is treated as a good unto itself, one that outweighs the intrinsic value of any given firm. Innovation, the driver of development, is envisioned not as the product of a steady accumulation of tacit knowledge and internal experience within long-lived corporate organizations, the kinds that must be protected and nourished by governmental policy. Rather, innovation grows out of the maelstrom of intense interfirm competition, the continual overtaking of conservative incumbents by radical newcomers, and the wild dynamism of organizational destruction and recreation. One could argue that this vision fits the contemporary era of rapid deverticalization and modularization in global industry. To the extent one believes a true technological and organizational revolution has occurred in the late twentieth century, one would expect innovation to fall mostly within the “disruptive” rather than the “sustaining” category. Traditional, vertically integrated incumbents—the firms with the greatest stake in the products and processes of the prior era—would be the least likely to remain on the cutting edge of technology. They have accumulated knowledge and experience, but not the kind relevant for a changed world. Moreover, to the extent one believes that the production of goods and services has truly become modularized, one would anticipate within any given module rising incentives for rapid and disruptive innovation. The condition itself of being a modular producer implies that one is no longer captive to single final products downstream. One simply innovates within the module, regardless of the degree of disruption that others may suffer (or added sustainability that others might enjoy). As waves of modular innovation cascade up and down the production chain, winners in any given activity presumably emerge from communities of flexible, highly specialized entrepreneurial entrants, entrants that amongst themselves compete at the same time that they cross-pollinate. Industrial policy then becomes focused on creating environments, not firms. It focuses on creating Silicon Valleys and innovative clusters, rather than industrial conglomerates and vertically integrated giants. Moreover, to the extent supply chains overlap and individual modularized producers somehow exist at the nexus, it becomes pointless to think about unified “industries” in the traditional sense. The goal becomes not to build a national electronics, auto, or aerospace industry per se, but rather to develop capabilities that accord control and high returns in the international supply chains crisscrossing all those industries.

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 261

Industrial Policy and the Building of National Champions What makes Chinese industrial policy so difficult to comprehend is that for all its focus on market-based approaches and comparative advantage, it also has an entirely different side that embodies assumptions of heavily statist Japanese and South Korean models of the past. Policy makers in Beijing may be employing all the mechanisms associated with comparative advantage strategies, but the ultimate aim remains the creation of “national champion” firms in self-reliant, vertically integrated “pillar” industries (Nolan 2001, 16). This, after all, is what the “grasping the large” side of the zhua da fang xiao enterprise restructuring policy is all about. It is about creating exactly the type of organizations associated with the Japanese and Korean models of yore: large, vertically integrated business groups that encompass entire industries from upstream to down, operate at the cutting edge of technology, and dominate global markets from their home base in China. Yet, this is a story that involves more than just new techniques for achieving old industrial ambitions. Rather, it is a story about a government claiming as its ultimate policy aim precisely the type of firms that its most high-profile restructuring (and trade) policies militate against. In essence, the government is seeking to create the very firms that comparative advantage, not to mention global technological change, select against. Of course, as some policymakers in Beijing are inclined to admit, China’s effort to build “national champion” conglomerates must differ from earlier Japanese and South Korean efforts in a few respects. First, the Chinese economy today is much larger and more diversified than were the Japanese and South Korean systems at the height of their respective experiments with dirigiste industrial policy (Perkins 2001). In Park Chung Hee’s Korea, more than half of all manufacturing was accounted for by fewer than 200 large industrial groups (Perkins 2001). In China, one would have to turn to tens of thousands, or even hundreds of thousands, of firms to account for a similar percentage of industrial activity (Perkins 2001). The point is that simply to exert the same degree of control associated with the Korean model, Chinese policymakers would be dealing with an exponentially larger task and exponentially more complex information flows than anything experienced in the Korea of the 1970s. Second, the Chinese government, in no small part because of the re-

262 CHAPTER 8

formist legacies of decentralization and informality, operates in a less unified manner than Japan or Korea decades ago. Whether by design or default, policymakers in Beijing today implement most national policy through local agents.43 This has certainly proven true in the effort to build “national” pillar industries, a task that has been essentially farmed out to individual provinces and municipalities. Consequently, while China’s industrial planners proclaim the need for national steel, auto, or machinebuilding firms, what results is the duplication of such entities in virtually every province and large municipality. Traditional Japanese and Korean industrial policy may have been all about focusing limited national resources on the development of just a handful of industrial giants. China’s industrial policy, however, is about signaling regional governments to focus resources on what ends up being thousands of firms and tens of locally duplicated industrial structures. Third, Japanese and South Korean developmental efforts were premised on the idea that, at least in their home markets, key industrial conglomerates would be granted sweeping protection. They would be held to international standards and encouraged to compete head to head with foreign firms in foreign markets, but, on the home front, they would be showered (selectively) with subsidies and sheltered from outside competition. As signified by the terms of China’s WTO accession, though, the world today is not that of the 1960s and 1970s, in no small part because the world’s wealthiest countries—though hardly paragons of free trade—simply do not tolerate the sorts of protectionism they once did with regard to Asian developers. Nor, somewhat ironically, are they inclined to tolerate the sorts of export flows previously generated by Asian “national” firms. Whether or not Korean-style industrial policy was effective on its own terms and in its own era is a major question, but one not immediately relevant to this chapter. What is relevant, however, is the basic reality that while China may seek to build the kinds of firms associated with such models, it has at its disposal few of the policy instruments and external conditions enjoyed by industrializers decades ago. That said, even if the model were feasible on practical policy terms, it would still be difficult to reconcile on logical terms with China’s “other” industrial policy, the comparative advantage approach. Indeed, practical issues of implementation aside, how can Chinese policymakers square the circle between the highly divergent conceptions represented by each of these approaches? Decision makers may presume that to the extent

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 263

they get industrial policy “right,” the resulting “national champion” pillar industry organizations will be globally competitive and hence sustainable after WTO-mandated market liberalization takes place. Yet, that really raises the question of how the divergent premises of “comparative advantage” and “national champion” can be reconciled. After all, one view stresses the primacy of churning and market selection— creative destruction—as the driver of innovation and growth. The other stresses virtually the opposite, the degree to which innovation occurs through the evolution and sustenance of established incumbents, corporate repositories of knowledge and experience.44 One view emphasizes the market’s role as a selection mechanism, a ruthless judge of winners and losers. The other emphasizes the market’s role as an incentive mechanism, a treatment that when applied to preexisting organizations encourages efficiency. One view says that firm-level incentives are inseparable from, and indeed can be understood only as emanating from, the systemwide process of “creative destruction.” The other suggests that market incentives, by encouraging existing firms to maximize efficiency, obviate—or at least reduce the likelihood of—such destruction. Indeed, in this latter view, if selection begins spontaneously to operate—if losers start to appear, particularly on a grand scale—then something must have interfered with the proper operation of the market, be it politicization, insufficient liberalization, or “bad policy” in any of its other guises. One view, in essence, understands the firm as a by-product of the market. The other takes the firm, particularly the modern industrial conglomerate, as the lynchpin and driver of the market.45 Policymakers could try, as in China, to hedge by operating on both sets of premises simultaneously.46 In so doing, however, they frequently adopt policies that function at cross purposes. For example, the quest for a “national team” has led to persistent governmental distortions of financial markets.47 Those distortions, though, by withholding capital from China’s most dynamic, market-oriented firms—its private enterprises—limit the ability of these firms to respond to competitive pressures being induced by “comparative advantage” market liberalization measures. In essence, the distortions aimed at building the national team undercut the global (and domestic) competitiveness of a huge swath of Chinese industry. Along similar lines, policymakers encourage the development of vertically integrated pillar industry firms, but then pass on the actual developmental task indiscriminately to localities. What results is neither the verticality nor overall scale that traditional Korean-style industrial policy

264 CHAPTER 8

calls for. “National champion” firms end up in reality as little more than local or regional players. At the same time, the focus on verticality encourages localities to think not in terms of cluster economies, innovative communities, or cross-cutting supply chains—the sorts of environments from which effective “comparative advantage” competitors are likely to emerge today—but instead in terms of self-contained industrial units, units that may coexist, but not interact. Firms end up with locally focused captive supply chains, a worst of all worlds situation even if one agrees with the goal of building integrated national conglomerates. To the extent the supply chain is held captive, it should at least be permitted to extend broadly in geographic terms (so as, hopefully, to incorporate “best-in-class” suppliers nationally). Keeping it local almost guarantees that the firm will fail to access the best suppliers, and hence will fail to produce world-class products. At the other extreme, to the extent one believes that firms should focus on modular activities and then link into upstream and downstream activities on a global basis (in line with the “comparative advantage” approach), administratively enforced captive supply chains should disappear altogether. By merging essentially irreconcilable visions for industrial development, policymakers end up achieving the aims of neither. Localization and geographic duplication undermine the scale and supply chain quality conditions that might, under the theory’s own assumptions, produce globally competitive conglomerates. At the same time, the institutional distortions induced to achieve national champions (local as they may be) undercut the ability of nonstate firms to compete effectively on purely market terms. The firms shielded from creative destruction remain weak, while the distortions behind that shielding leave everybody else handicapped in the face of creative destruction. That many in the latter group have survived is testament more to their fortitude than to the brilliance of industrial policy per se. Unfortunately, such survival, achieved primarily through commodity production and cutthroat discounting, is hardly the basis for extended success in the future, whether at the enterprise or national level. China’s Emerging Multinationals and the Question of “Catch-Up” The preceding sections have considered the impact of institutional factors on the overall population of enterprises in China. There are Chinese

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 265

firms at the far end of the spectrum, however, that have emerged as important national and even global players. Such firms have managed to do exactly what their peers have not: develop powerful and clearly identifiable brands, operate beyond local and regional boundaries, and achieve significant market share in their respective product areas. Within the Chinese domestic market they have been able to match—or even surpass—the positions of powerful multinational players. Some of these firms have gone a step farther by reaching out into major global markets, whether through the export of branded products, the establishment of overseas R&D and representative offices, the acquisition of overseas subsidiaries, or even the opening of overseas manufacturing facilities. In the telecommunications equipment sector (switches and routers), at least two Chinese firms—Huawei and ZTE—have emerged as plausible competitors to global leaders like Cisco, Nortel, and Alcatel (Einhorn et al. 2002). Both Huawei and ZTE have achieved significant share in the home market, one that until just a few years ago was completely dominated by foreign producers, and a market that happens to be the second largest globally in telecom fixed equipment. Huawei produces low-cost, reasonably high-quality versions of devices sold by industry leaders like Cisco. Whether the products are “knockoffs”—mere copies—and whether they represent violations of Cisco’s intellectual property are questions open to debate. What seems clear is that while Huawei products lag somewhat behind those of industry leaders technologically, they sell for as little as 60 percent of the cost of state-of-theart devices (Einhorn et al. 2002). On at least some dimensions, the approach appears to be working. Huawei in 2001 had total sales of US$3.1 billion, and has recently won major equipment contracts with China Telecom and China Mobile, the nation’s largest telecom service providers.48 Huawei has also begun making inroads into foreign markets, achieving in 2001 a 156 percent increase in international sales. The Shenzhen-based firm has branch offices in thirty-two nations and research institutes in the United States, India, Sweden, and Russia. Overseas sales still represent only 10 percent of the firm’s business, but the company aims to expand that figure (Einhorn et al. 2002). Huawei has recently won equipment contracts in Brazil, Kenya, Ecuador, Russia, Egypt, and Germany. The company has also opened sales offices in San Jose, Northern Virginia, and Dallas.49 In terms of international research collaborations, Huawei in 2002 agreed with Microsoft to set up a joint lab in Shenzhen, and with NEC to set up

266 CHAPTER 8

a lab in Shanghai.An analogous trajectory has been followed by Haier in home appliances and Lenovo in personal computers. Haier in 2001 had total sales of RMB 60.2 billion (US$7.4 billion), and by its own estimates enjoyed 30 percent domestic market share for refrigerators, freezers, air-conditioners, and washing machines.50 The firm has a highly recognizable brand within China, and increasing brand recognition overseas as well. The company today commands half the U.S. niche market in compact refrigerators, and has also begun carving out a position producing electric wine cellars (Sprague 2002). While the U.S. market for standard home appliances is still dominated by global giants like Whirlpool, General Electric, and Maytag, Haier—in a unique move for a Chinese firm—has recently attempted to gain ground by acquiring manufacturing facilities in South Carolina for full-sized refrigerators. In another high-profile move, Haier America purchased the historic Landmark Building in downtown Manhattan as its new corporate headquarters. Haier products are today marketed under the company’s own brand in twelve of Europe’s top fifteen and eight of the United States’ top ten chain store retailers.51 Like Haier, Lenovo, China’s premier IT company, has developed a powerful domestic brand, and has matched that brand with significant market share. For personal computers, Lenovo, through the middle of 2002, accounted for 27.7 percent of the Chinese market and 12.2 percent of the Asia-Pacific (excluding Japan) market.52 By the mid-1990s, Lenovo also became one of the world’s top five suppliers of computer motherboards and add-on cards. Interestingly, the company’s initial growth was fueled by innovation in Chinese word-processing technology, namely the development and commercialization of a word-processing add-on card for IBM-compatible PCs (for an extensive description, see Lu 2000).The firm, in effect, offered an innovative and much-needed method for localizing PCs for the Chinese market. Lenovo in the late 1980s elected to tailor its add-on card to the AST PC, and became the sole distributor of AST machines in China (Lu 2000, 74). As AST became the leading brand in China and Lenovo’s revenues soared, the Chinese company’s expansion—not to mention rapid climb up the manufacturing learning curve—began in earnest. Through a series of Hong Kong acquisitions, Lenovo quickly moved into the design and manufacturing of PC motherboards, shipping a steady volume of 3,500 units per month by the fall of 1989 (Lu 2000, 74). The next step involved combining the firm’s Hong Kong–based manufacturing capabilities for

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 267

motherboards with its PC distribution network in Mainland China (developed for AST) to introduce Lenovo-branded PCs into the home market. This occurred precisely when the market for PCs in China began to boom. By the end of the 1990s, Lenovo PCs were the top brand in China, and the firm was publicly listed on the Hong Kong Stock Exchange. The firm also was pursuing a strategy of vertical integration, both backward (from motherboard and add-on card to printed circuit board manufacturing) and forward into system integration services and software. Today, Lenovo is still primarily a manufacturing company, with products ranging from PCs, palmtop computers, laptops, and cell phones, all the way to basic hardware components. By its own account, the firm aims to extend its reach into high-margin IT services and applications. The final illustration involves not a single firm, but rather an entire industry—the Chinese motorcycle sector. What is interesting here is not so much the rise of a major branded player—numerous ones exist in China today—but rather the growing footprint of China’s producers as a group both domestically and abroad. China today has hundreds of motorcycle manufacturers, and almost as many local brands. Many of these manufacturers during the 1990s—and some still today—are engaged in JVs with leading Japanese firms, companies like Yamaha, Suzuki, and Honda. The Japanese initially entered these arrangements in the early 1990s for obvious reasons. They sought access to the Chinese motorcycle market (in terms of sheer numbers of units, the largest in the world), and they found that the Chinese could manufacture a motorcycle at a fraction of the cost of Japanese companies (Leggett and Zaun 2001).The Japanese supplied the brand and design, and the Chinese produced the motorcycle. Over time, Chinese manufacturers—those with Japanese JVs, and even those without—proved increasingly able to produce motorcycles independently under their own brands. The designs are often still Japanese—essentially Chinese knockoffs—and the Chinese firms often produce under a brand that either closely resembles the Japanese competitor’s (i.e., “Yameha,” “Suzaki,” or “Honea”) or somehow directly incorporates a Japanese firm’s name (i.e., “Taizhou Yamaha” or “Nihon Yamaha,” firms that have no relationship with Yamaha in Japan, and indeed, firms that the Japanese protest should be shut down) (Leggett and Zaun 2001). The result has been that while the Chinese motorcycle market experiences rapid growth, Japanese producers are witnessing an erosion of their market share. A similar phenomenon, albeit with a more effective Japanese response,

268 CHAPTER 8

has been under way in Southeast Asia, particularly Vietnam and Indonesia (countries that rank, respectively, as the third- and fourth-largest motorcycle markets in the world). Through much of the 1990s, Japanese firms—namely Honda, Suzuki, and Yamaha—produced motorcycles in Indonesia and accounted for most of the motorcycles sold in the country (Dhume 2001). In 1999, Indonesia lifted import restrictions, and lowpriced Chinese kits representing nearly sixty different brands flooded in. Chinese motorcycles quickly came to occupy 20 percent of the market, though the Japanese have since responded effectively with lowpriced, high-quality alternatives (Dhume 2001). The same sort of competition between Japanese and Chinese producers obtains in the Vietnamese market. There again the Japanese have focused on localized JV production, while the Chinese have pushed imported kits. In 2001, of the 2.5 million kits imported to Vietnam, Chinese brands accounted for 78 percent, a nearly 30 percent increase from 1999 (Cohen 2002). Again, the Japanese responded with a heavily marketed, low-cost, high-quality product—the Honda Wave Alpha. The Vietnamese government has also complicated matters by imposing “emergency” tariffs on imported motorcycle kits, ostensibly to address traffic congestion, infrastructure development, and safety issues (Cohen 2002). Meanwhile, the intense Sino-Japanese competitive dynamic continues in this market, with the addition of political strategies being employed by each side to overcome market access problems. Several characteristics are common across these examples. First, the Mainland Chinese firms’ success to date, at least in terms of market share, has for the most part been restricted to the Chinese home market. Even within that market, the emerging Chinese branded competitors still tend to occupy the lower-end segments. Their products compete mainly on the basis of low cost, and customers tend to view them as somewhat lower in quality, technological sophistication, and style than foreign alternatives. The same, of course, can be said for overseas markets, though Chinese branded firms have for the most part yet to achieve significant market share. The point is not that these firms necessarily produce shoddy products or even direct knockoffs: they have mastered extremely complex production processes and are producing goods that are in some sense viewed as substitutes for those of leading foreign firms. Rather, the point is that the Chinese are still at the stage in which their prime source of competitive advantage, even in their home market, is still low cost.53

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 269

Second, Chinese firms have shown little ability to engage in product innovation or overall product definition, areas that still appear the domain of leading foreign firms, whether globally or even within the Chinese market. For example, in China’s mobile phone handset market, companies like Ericsson, Nokia, and Motorola still set the overall design trends and occupy the high end of the market, while Chinese entrants like Bird follow in step with lower-cost alternatives. Even Lenovo, whose early successes came through innovative customization for the Chinese home market, today appears to have ceded higher-value product innovation activities to foreign partners, a pattern that will likely persist even after the acquisition of IBM’s PC division. It has teamed with Microsoft to develop a tablet PC, incorporates IBM storage technology in its own storage products (while also producing on an OEM basis for IBM), employs IBM voice-recognition software in its pocket computers, and, of course, like most PC manufacturers, uses Intel Pentium processors. Again, the point is not that Lenovo or companies like it are backward, but rather that they are primarily commodity manufacturing enterprises. The higher-value, proprietary aspects of their manufactured hardware are usually still controlled by foreign leading firms. Companies like Lenovo, in effect, are still in the position of being design takers rather than design makers. The second issue is that when Chinese firms innovate in manufacturing, this does not directly translate into sustainable competitive advantage. Both branded and nonbranded manufacturers in China, almost regardless of the level of complexity at which they operate, have exhibited an uncanny ability to push modularization. They take manufacturing processes that in the past had been performed by foreign firms in an integral, uncodified manner, and aggressively reconstitute those processes in a codified and standardized form. The motorcycle example is particularly relevant here. Japanese assemblers, even at present, for the most part design their motorcycles to accept only customized, product-specific parts. Because Honda motorcycles take only Honda parts or Yamaha motorcycles only Yamaha parts, the leading assemblers organize production around captive supply chains, whether in Japan or in their overseas JVs. Chinese manufacturers, on the other hand, while they may copy the overall Japanese product design, shift production over to standardized, interchangeable parts. The product’s architecture becomes “open” rather than proprietary, and the manufacturing process becomes modular rather than integral (integral in the sense that each step must be customized and coordinated with the next). Whether quality is sacrificed

270 CHAPTER 8

is debatable, but what has been eminently clear is that the standardized manufacturing process is much cheaper. The opening up of production architecture may be quite advantageous to consumers. It explains why high-quality appliances, PCs, home electronics, and other goods are available today at a fraction of the cost of just a decade ago, and more often than not with a “made in China” label. This opening up is also advantageous to lead firms that control product definition, design, and marketing, but seek to outsource much of the manufacturing. Similarly, it is threatening to lead firms that resist outsourcing and instead attempt to hold onto internal supply chains even in the face of commodification.54 Yet, it is important to remember why open product architectures lower prices for customers—modularization and codification lower entry barriers for manufacturers. That is the heart of the problem for Chinese firms pushing this trend. They innovatively codify a process, but in so doing, they provide entrée for a host of competitors in their immediate environment. In its darker variant, open product architecture leads to phenomena like the countless Japanese knockoffs—bearing dubiously familiar Japanese-sounding brand names—in China’s motorbike market. In its more positive face, though, we see the production of extremely complex, high-quality standardized products: PCs, routers, motherboards, digital cameras, and cell phones. Chinese producers are able to achieve high enough quality and low enough cost to make these manufacturing activities quite unattractive to overseas lead firms. Hence, the overseas firms shed the activities. The problem, however, is that to the extent they are commodified, the activities are not terribly attractive over the long run to the Chinese manufacturers themselves. For all the reasons discussed earlier, the margins remain razor thin and the competition vicious. The problems Chinese motorcycle makers have faced in Southeast Asia are illustrative. Many Chinese producers can underprice the Japanese, and fifty some odd Chinese brands may have flooded into Indonesia in 1999. Yet, given their narrow margins and liquidity constraints, not to mention lack of experience, no Chinese firms have been able to respond to Honda’s marketing counterattack. China’s best firms have to some extent compensated by integrating markets across China and investing more seriously in the building of brands. Yet, it should be noted that even among the best firms, the focus remains on standardized manufacturing (and products), vertical integration, and diversification. In other words, much like their less famous

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 271

counterparts, they cope with narrow margins and nonproprietary products by diversifying and by integrating both forward and backward in the supply chain. For example, from its initial base in add-on card and motherboard manufacturing, Lenovo moved forward into PCs and routers, and backward into printed circuit boards and semiconductors.55 Similarly, at a different level of complexity, Chinese motorcycle manufacturers, even competitors within single cities, in many cases still try to integrate their supply chains.56 Vertical integration may be a reasonable response to various institutional features of the Chinese business environment. Yet, it is not an effective response to the technological and organizational change signified by modularization. Chinese firms modularize processes and then try to push them together through vertical integration. There is a difference, however, between a firm’s choice to integrate certain processes and the degree to which processes are somehow integral in a technological or “product architecture” sense. The distinction becomes apparent in the Lenovo case. Lenovo may try to own entire supply chains from printed circuit boards right through PCs. Yet, the technological reality remains that each one of these devices today is a standardized, stand-alone component, one that can—and often is—produced in fully substitutable form by other competitors. Lenovo may choose to integrate, but it is far from clear how vertical integration could lead to any advantage over competitors specializing in any single piece of the modularized chain. Vertical integration would make sense, though, if the firm could somehow reverse the standardization process, converting an open architecture product back into an integral architecture. In other words, if a firm like Lenovo could create a new product that depended on high degrees of coordination between customized components, vertical integration might prove an effective strategy. Alternatively, a firm like Lenovo might command value if it could control the rules of connectivity across various modularized steps. At present, however, the fact that Lenovo products still have “Intel inside,” “Microsoft inside,” and “IBM inside” suggests that the overseas firms still control the rules of connectivity, while Lenovo has been left with a focus on low-value, commodified hardware. The Issue of Catch-Up The preceding discussion still leaves open the question of catch-up— the question of whether China’s lead firms, for all their problems today,

272 CHAPTER 8

may just be in the first stages of catching and ultimately surpassing their foreign rivals. In other words, might we be witnessing today the opening stages of a situation analogous to the Japanese auto industry’s rise vis-à-vis U.S. auto companies in the 1970s? As firms like Haier or Huawei enter North American and European markets, are we seeing a replay of the Japanese and South Korean story? Is it the same story of new competitors figuring out how to produce inexpensively, introducing products overseas first into lowest-end market segments, gradually building market share, and then finally down the road becoming dominant in high-value products? In answering these questions, it is worth considering the conditions under which Japanese and South Korean industrial firms rose decades ago. In that era, industries could still in a meaningful sense be understood as separate, self-contained entities, and often self-contained in national terms. We could refer to the U.S. steel or the French auto industry, and we could contemplate whether rising industrializers like Korea would develop strength in a particular industrial sector. Moreover, in these relatively autonomous industries, product innovation occurred in incremental terms, and manufacturing processes tended to be integral. The various steps in the process, while perhaps understood in broad terms in these stable industries, were uncodified (and given the state of IT at the time, probably uncodifiable). As such, they could not organizationally be pulled apart from one another, and they tended to be particular to each firm or each firm’s captive supply chain. Challengers then, to the extent that they could amass the resources needed to enter these capital-intensive industries, could compete on the basis of process innovation, the ability to produce the same products as incumbents but at significantly lower cost (Amsden 1989). Because manufacturing processes remained uncodified and integral within the firm, shop floor innovations were truly proprietary. They were, in effect, a form of art or craftsmanship that neither incumbents nor other entrants could easily copy. There was no open recipe to follow. What that description suggests is that rising Chinese firms today are operating in a dramatically transformed era, one in which the methods of the past are not directly applicable. First, it is not whole industries that move today from developed to rising nations, but instead activities. As noted earlier, what has moved to China en masse, whether at the bequest of leading global companies or through pressures from Chinese firms themselves, are the manufacturing-intensive segments of particular

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 273

value chains. More precisely, it is the codified, commodified, nonintegral manufacturing activities that move. Competing in these areas, while hardly trivial, often does involve mastering open processes rather than developing proprietary ones. It is for that reason in part that we see so many new entrants from China in manufacturing rather than the handful of firms that entered from Japan and South Korea in previous decades. Second, when Japanese and South Korean competitors emerged, they were rising up against relatively stable incumbents, incumbents whose focus was still on manufacturing. As such, the incumbents were essentially stationary targets whose products could be substituted with lowercost alternatives. Today, the situation is quite different. In large part because of modularization, the incumbents—global lead firms—are hardly stationary, and in many cases have completely transformed themselves. Firms like Lenovo, Haier, Huawei, and Bird may be rising on the basis of their low-cost manufacturing expertise. At the same time, most lead firms—whether IBM, Electrolux, Cisco, Motorola, Dell, or many others—are moving away from manufacturing entirely. Instead, they are increasingly focusing on what may be broadly termed the “service” side of production: overall product definition, design, marketing, and supply chain management. Lenovo may manufacture and sell PCs, but its nearest U.S. counterpart, IBM, has effectively become a business services and software company. Similarly, Dell, the leading U.S. PC seller, engages in very little manufacturing and assembly, but instead has built a business on distribution. Rising firms may be capturing manufacturing activities, but the former manufacturers have increasingly specialized in high-value nonmanufacturing activities. In so doing, they, in many cases, either retain control of the supply chain’s rules of connectivity—by perhaps controlling the key operating software or state-of-the-art processor designs in electronics—or retain control of activities that truly remain integral and proprietary. Examples of the latter often involve efforts by lead firms to embed services in the higher-end products they either manufacture or outsource. The point is not so much that Chinese firms are performing poorly in the developmental process, but quite the contrary. The point is that Chinese entrants and the lead firms of developed countries are jointly responding to major transformations in the organization of production. Their joint response is better understood as complementary rather than substitutable or somehow “competitive” in head-to-head terms. That then leads to a third point about the way the terms “industry”

274 CHAPTER 8

and “national industry” are understood today. In previous decades, it made sense—with a certain degree of simplification—to conceive of industries as distinct silos. Particular nations, then, could be mapped into one or more of those silos. In the current era of modularization, certain activities within industries have been separated from one another and moved across national boundaries. Hence, in any given silo now, we may see companies or countries occupying certain activities but not others. Yet, modularization is really about much more than that. Activities within discrete industries have been split apart, and these independent, highly specialized activities now cut across multiple industries. What were once distinct industry supply chains now overlap, intersect, and interact in myriad forms. It becomes increasingly difficult to say exactly which “industry” a given firm or nation specializes in. Is a semiconductor foundry in the electronics industry, or, since its chips go into cell phones, in the telecommunications sector? Is the “fab-less” semiconductor design house mapping out chips for automobiles—along with semiconductors for a host of other applications—in the auto industry? Perhaps it also happens to be integral to aerospace, telecommunications, or home appliances. IBM today is basically a business solutions and services company. What industry, in the traditional sense, does it belong to? The specialization associated with modularization has led to the blurring of boundaries between industries and the growing interaction across them. Today, it may make more sense to think of matrices and webs of specialized activities than discrete, stand-alone industrial sectors. Among other things, such organizational change leads to the phenomenon of modularized innovation and ripple effects of such innovation across formerly unrelated industries. The “fab-less” chip design house, in its efforts to design a telecommunications application, may come up with a new capability applicable to aerospace. Then again, perhaps the new chip design will have its greatest impact on “smart” home appliances. For the chip innovator, the ultimate downstream application may be irrelevant, so long as the design gets purchased in great quantity. Yet, the downstream application certainly is not irrelevant to those who are competing in the downstream activities, particularly when the new application may lead to downstream substitutes. A firm like Microsoft may keep churning out operating software for PCs, but so too does it focus on enabling the sorts of products–palmtop computers, digital writing tablets, Web-capable mobile phones—that may undercut or otherwise

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 275

replace PCs. In the modularized world, specialized innovations lead to unpredictable outcomes. The challenge for contemporary industrial policy is not just that the pace of change is faster now than in the heyday of Japanese or South Korean industrialization. More important, the organizational mechanism of change—particularly the extent to which it is spread across ostensibly unrelated firms and “industries”—is completely new. For a nation to be strong in autos, aerospace, or telecommunications, what does it need? Software companies? Semiconductor design houses? Handset manufacturers? Steel firms? Marketing firms? Here are the risks entailed in forcing the vertical integration of industries. From a product architecture perspective, it may be impossible to determine the exact boundaries of a given industry. Yet, Chinese industrial policy, by selecting “pillar” industries, does precisely this in an artificial sense. It operates under the idea that a country can, from upstream to down, “build” a steel or auto or aerospace sector. Similarly, for various institutional reasons, individual Chinese companies may themselves elect to vertically integrate their activities. Whether through institutional default or conscious policy, they end up forcing the integration—whether under a single company roof or within a single national entity—of activities that are not in any technological sense “integral.” In effect, they push together within a given organizational boundary activities that could just as easily stand alone from one another. In so doing, as such activities are held captive within single “industry” supply chains, policymakers and corporate strategists limit the extent to which modular innovation and cross-fertilization can occur. It is not surprising, therefore, that China perceives itself, probably correctly, as lagging behind India, as well as developed countries, in industries like software. Similarly, it is not surprising that China lags in high-end semiconductor design capabilities. Lenovo may put motherboard and printed circuit board manufacturing under one organizational roof, but that does not change the fact that these are essentially modular activities. Nor does it change the fact that Lenovo’s counterparts, firms like IBM, are shedding precisely those standardized activities so as to focus on the higher-value activities that for technological reasons truly are integral in the supply chain. The point is not that single lead companies necessarily control a supply chain. Who has more economic power in the PC supply chain: Intel, Microsoft, Lenovo, or Dell? The question is not only difficult to answer, but may be irrelevant. Given the degree of specialization and narrow

276 CHAPTER 8

innovation permitted by modularization, perhaps there is value to be achieved in a range of activities. Achieving that value, however, requires that the firm not simply integrate activities, but rather focus on integral activities—the functions, products, or services that are absolutely critical to upstream and downstream counterparts, and that cannot be easily duplicated by competitors. Of course, the companies that achieve this today—be they Microsoft, Intel, or anybody else—may not be around tomorrow. Nothing guarantees that the firm setting particular standards and rules of connectivity today will last until tomorrow. Yet, from the national perspective, the role of industrial policy should be to ensure that the home environment can spawn a series of such firms. In China today, neither national “pillar” industries nor the institutional problems that independently encourage vertical integration fulfill that purpose. Instead, they isolate even the best Chinese enterprises from state-of-theart technology, reduce the likelihood that Chinese firms will set rules of connectivity globally, and end up facilitating specialization among foreign lead firms. What results is not so much catch-up as a greater division of labor that arguably widens the gap between overseas lead firms and Chinese follow-on producers. Conclusion Taiwan today undoubtedly faces extraordinary pressures from globalization, particularly given the island’s traditional focus on high-end manufacturing. Activities are moving geographically with tremendous speed, placing many economies—including Taiwan’s—in an almost perpetual state of transition. While pressures on middle- to high-income regions like Taiwan may be substantial, they are arguably even greater in places like Mainland China, where basic industrialization is still unfolding. For example, even for electronics, a sector in which Mainland China has arguably made its greatest competitive advances, the highest-value manufacturing activities have remained the domain of foreign firms. Just over a fifth of China’s total US$325 billion in exports in 2002 were classified by the PRC government as “high tech,” but most consisted of low-end components and low-margin commodity products like DVD players (Rosen 2003). Moreover, 61 percent of these “high-tech” exports were produced by foreign-owned firms (Rosen 2003). At the same time, China runs a substantial trade deficit in high-tech products, a deficit that grew from US$7.6 billion in 1997 to US$15 billion in 2002.

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 277

Even a “national champion” like Lenovo still imports the vast majority of the semiconductors used in the computers that it assembles, a situation that jibes with the input flows illustrated earlier in Table 8.1 on page 241 (Rosen 2003). Again, the point is that even in this sector, it is primarily the lowest-value activities—ones associated with basic assembly and testing—that have migrated the most rapidly to emerging firms. At the same time, higher-value activities often remain deeply embedded in production processes controlled by incumbents, incumbents that in some cases retain the activities in their home countries, and in other cases perform these activities in places like China. We can certainly say that some manufacturing activities once based in Taiwan are now being performed by Taiwanese firms in Mainland China. What we cannot say, however, is that new Chinese competitors have truly emerged. Shifting geography has, at least to date, failed to bring a comparable shift in control and capabilities. What all this suggests is that the old rulebook of industrial catch-up no longer applies. In the past in the auto industry, for example, the production process was essentially self-contained, and essentially the same for low-end and high-end products. Late entrants could, by jumping in at the low end, hope to learn the process, and master it in such a way as ultimately to compete with more developed incumbents at the high end. Today, however, because production processes have been broken apart into separable activities, that path has been shut down. Low-value, commodified activities travel to new entrants, but that shift does not bring along with it the proprietary, knowledge-based capabilities associated with higher-value activities. The shift certainly does not bring with it the ability to integrate entire supply chains and ultimately control them. Indeed, more often than not, the incumbents themselves are controlling the pace and nature of supply chain fragmentation, and they do so precisely because they continue to perform the supply chain integrator and manager function. In this sense, Taiwanese firms, particularly in core areas of excellence like high-end electronics, remain decidedly in the driver’s seat, even as they extend their geographical reach across the Chinese Mainland. Of course, Taiwan has substantial security concerns vis-à-vis Mainland China that will not be allayed by greater cross-straits commercial integration. The very fact that the political relationship between Mainland China and Taiwan remains unresolved, and that the Mainland routinely declares its willingness to go to war should Taiwan formally declare

278 CHAPTER 8

independence, means that Taiwan, regardless of its economic status, remains situated in a dangerous geopolitical situation. It would be illusory to believe that enhanced commercial integration could somehow resolve this political–military problem. That said, it would be equally illusory to believe that greater security could be achieved through impediments to cross-straits business. To date, the ROC government has proven rather ineffective in stopping Taiwanese firms from investing in the Mainland. Yet, even if the ROC government were more effective in enforcing strictures, what possibly could be achieved? In their efforts to operate on the Mainland, Taiwanese firms are following the same incentives that motivate firms from countries around the world. Regardless of what Taiwanese firms do, FDI will continue to flow into the Mainland, and enterprises from the world’s most advanced economies will continue to transfer activities to Chinese counterparts. Taiwan can exert little if any leverage over this phenomenon. However, by impeding the Mainland operations of its own firms– whether through direct impediments or more subtle “Be Patient, Go Slow” (jie ji yong ren) policies—the ROC government risks undercutting the global competitiveness of Taiwanese producers without achieving any comparable benefits on the security front. To stay competitive globally, Taiwanese firms must continue to shed low-value activities to the Mainland, activities that by definition are no longer economically viable if left in Taiwan proper. In implementing this transition, Taiwanese firms ironically end up in a situation analogous to their Mainland counterparts. One of the problems on the Mainland, and one of the reasons that Mainland firms have had trouble moving into higher-value, more knowledge-intensive activities is that the institutional environment remains undeveloped and—arguably—biased in favor of overseas firms. Institutional failings end up casting a dark shadow over firm structure. By throwing up impediments to Mainland investment that force Taiwanese firms to channel their money through third parties, hide their true ownership stakes, or otherwise operate in a less-than-transparent manner, the ROC government threatens to cast a similarly dark shadow over its own firms. Taiwanese producers will be put at a competitive disadvantage relative to other global manufacturers operating on the Mainland. In an era of tremendous transformation, Taiwanese firms, through their flexibility and entrepreneurship, have proven surprisingly capable of retaining control over the most desirable manufacturing activities, and shedding many of the others. Moreover, unlike their Mainland

CROSS-STRAITS INTEGRATION AND INDUSTRIAL CATCH-UP 279

counterparts, Taiwanese firms have in many cases remained deeply embedded in networks—both commercial and social—involving some of the most dynamic regions of the world, particularly Silicon Valley. It is in these networks that Taiwan’s true competitive advantage resides, and it is in enhancing these networks—whether through the facilitation of smoother human capital flows or deepened research connections— that the ROC government should focus its development efforts.

This page intentionally left blank

Notes

Notes to Chapter 1 1. See also the writings of Chi Schive (Hu and Schive [1998]; Schive (1999b). 2. See Fuller, chapter 5 in this volume. 3. Robert Ristelhueber, 2003. 4. See Fuller, Akinwande, Sodini, chapter 3 in this volume, and Wu (2005). 5. Wu (2005). 6. For an excellent overview of the state of transportation and financial links between Taiwan and Mainland China today, see Cavey (2003), pp. 11 ff. 7. Yang and Hung (2003), p. 686. See also EIU (2003c). In March 2002 the government announced that Taiwanese firms could set up 8-inch wafer fabrication facilities on the Mainland with depreciated equipment. 8. Interview at Chinese Textile Institute, 1999. 9. The argument here builds on interviews at BenQ and Acer and Dean (2003). 10. The argument here builds on the observations of Dan Breznitz, and is based on B.H. Hall et al. (2001). For the metric most commonly used, see Technology Assessment and Forecast Branch, U.S. Patent and Trademark Office, 2002, Calendar Year Patent Statistics. www.uspto.gov/go/taf/reports.htm. See also Porter (2001) and Fuller, chapter 5 in this volume, on the increasing ability of Taiwanese private firms to patent, in contrast to the previous pattern, where the Industrial Technology Research Institute (ITRI) received the lion’s share of U.S. patents. 11. This is Dan Breznitz’s conjecture and it was supported by a number of the interviewees. 12. See Fuller, chapter 5 in this volume. 13. This firm was interviewed in 2000 and re-interviewed in 2002. The head of the joint venture in China was also interviewed in 2002. 14. On foreign direct investment in China from Taiwan as well as other countries, see Yasheng Huang (2003). 15. On the pros and cons of Taiwan building semiconductor fabs on the Mainland, see Chyan Yang and Shui-Wan Hung (2003), pp. 681–96. 16. The Economist (2003), p. 99. This Economist article quotes Ming Zeng, professor at Insead and Cheung Kong, as finding only three or four domestic Chinese high-tech successes. 17. See analysis in Berger and Lester (1997). 18. Abernathy et al. (1999). 19. The interviews showed sharp increases over the past few years in the quantities of products shipped by air. The vice president for procurement in one of the largest U.S. women’s clothing brands told us that he’d ended all his Mexican 281

282 NOTES TO CHAPTERS 1 AND 2

outsourcing. In 1998, 12 percent of the garments sold under his labels were shipped by air; in 2003, 20 percent; and he expected that to rise quickly to 30 percent. 20. In an interview with a high-end French sportswear company, the president explained that the quality of the shirts made in China was equal to those made in Italy. “Why not make them all in China?” we asked. The president responded with mock derision, “Why, Professor Berger, and I thought you were a political scientist!”

Notes to Chapter 2 1. The market research firm Technology Forecasters estimates that outsourcing by lead firms in the electronics industry had shifted 17.2 percent of the costs of all electronic goods sold worldwide to contract manufacturers by 2003 (Technology Forecasters 2004). The rise of strategic outsourcing in the electronics industry can be seen in the growth rate of the contract manufacturing industry during the 1990s, which was consistently higher than that of the industry as a whole (The Economist 2000). 2. While in Taiwan’s electronics industry the term “OEM” is widely used to indicate manufacturing-only contracting, in other countries and industries the same term is used to refer to “lead” or “customer” firms. In the motor vehicle industry, for example, OEM, or simply “OE,” is used to refer to automakers such as Ford, Volkswagen, and Toyota. In the American electronics industry, we often find the very confusing situation of firms that have outsourced most of their manufacturing, such as Hewlett-Packard, or firms that have never had large-scale in-house manufacturing, such as Cisco, being nevertheless referred to as “OEMs.” See Sturgeon (2000) for an extended discussion. 3. This terminology has recently led to the emergence of a third term, “original brand manufacturer,” or “OBM,” which denotes a supplier that has begun to develop, manufacture, and sell its own brand of products. Because an increasing number of competencies are required to move from manufacturing to design, to sales and marketing, there has been a widespread assumption that firms and nations that begin as suppliers to lead firms in other places might upgrade their position in the global economy by moving sequentially through OEM, ODM, and OBM stages of development. While movement along this trajectory certainly offers firms from developing countries an attractive learning and upgrading path, it also can set them on a collision course with their customers, which is why the contract manufacturers that have grown the fastest have been those which have adopted a “pure-play” contract manufacturing strategy and have publicly and repeatedly renounced the aim of developing and marketing their own brand of products (Sturgeon and Lester 2004). 4. In 2003, the U.S.-based Fabless Semiconductor Association estimated that there were 22 Taiwan-based fabless semiconductor firms selling chips (Arensman 2003), and in 2004, the Taiwan Semiconductor Industry Association counted 32 IC design firms as members, and estimated the total number of firms engaged in IC design in Taiwan (which might include foreign subsidiaries and small contract design firms) to be 225 (Taiwan Semiconductor Industry Association 2004.) 5. Worldwide PC unit sales and value of sales are from International Data Corporation, a private market research firm based in Framingham, Massachusetts, as cited in two 2003 press releases (International Data Corporation, 2003a; 2003b).

NOTES TO CHAPTERS 2 AND 3 283

Taiwan ODM PC unit production and value of production figures are from Market Intelligence Center, as cited in the PowerPoint presentation “The Taiwan ICT Industry Outlook,” by Paulo S. Su (Su 2003). 6. The International Data Corporation’s definition of information technology includes hardware (servers, PCs, networking equipment, printers, etc.), software (system infrastructure software, applications, and application development and deployment), and IT services (training and education, planning, implementation, maintenance and support, and operations). There are certain categories that may not be included due to lack of coverage, or that are not in International Data Corporation’s research scope, for example, telecommunications spending, and some hardware categories, such as scanners and digital cameras, among others. 7. Flextronics International is incorporated in Singapore and performs some important global functions there, such as purchasing. Top management and most key functions, including strategic planning and IT development, are centered in San Jose, California. 8. Herrigel and Wittke (forthcoming) argue that such variations are increasing as buyers and suppliers search for an advantage in the context of increased outsourcing and building supplier competence. Intense collaboration, in the context of co-design for example, builds interdependence and moves the relationship in the direction of captivity. Commodity relationships tend to flow from customer purchases of offthe-shelf designs from ODMs, which make the relationship more marketlike and tend to increase supplier power, and from build-to-print arrangements, which tend to increase buyer power. According to Herrigel and Wittke (p. 15), both parties tend to seek to mix these linkage forms and change them over time to keep independencies and power asymmetries at reasonable levels, a process they call “sustained contingent collaboration.”

Notes to Chapter 3 This chapter was originally published in Industry and Innovation 10, no. 2, 179–96, June 2003. Reprinted with permission from Taylor & Francis Ltd. www.tandf.co.uk/ journals 1. While it is true that Winbond was not funded directly by the government like TSMC and UMC, Winbond was given ERSO’s old lab as its first base of operations and the personnel were shifted from ERSO en masse in a manner very similar to that of UMC and TSMC. 2. Aside from Macronix, the last significant non-DRAM IDM in Taiwan, Hualon, recently announced that it would sell its fab in order to convert itself into a fabless design firm. 3. IPC interview. These interviews by the authors with managers from pureplay foundries were conducted as part of MIT’s Industrial Performance Center (IPC) Globalization Project. 4. A Taiwanese firm called Quasel entered the memory market in the mid-1980s, but failed within two years. 5. Walsin Lihwa’s DRAM firm, Winbond, was originally an IDM with a diversified portfolio of products, but at present, its fabrication operations are mainly devoted to memory products, so it makes sense to place this firm in the DRAM firm category.

284 NOTES TO CHAPTERS 3 AND 4

6. Macher, Mowery, and Hodges (1999, pp. 251–52) describe the capital spending and market share trends. They repeatedly mention the “superior access to capital” that these new entrants enjoyed, but shy away from explaining this superior access, except for one brief mention of Japanese “internal sources of finance.” Given the OECD data on debt-to-equity ratios, we argue that internal sources were not the source of this massive investment, but rather that the system in these countries was set up to provide patient capital, as argued in Amsden (1989) and Woo (1991) for Korea and in Calder (1993) for Japan. 7. IPC interviews with executives in the global DRAM industry and a former high-ranking ITRI executive. 8. Vanguard has not yet exited the DRAM market, but it has announced its intention to convert to a foundry operation after winding down the current generation of DRAM production. 9. See TSIA (2001, p. 4) on distribution of fabs and TSIA (1998, Tables 2–7) for the fact that prior to 1996 there was very little eight-inch fabrication (only 130,000 wafers/year). 10. See Semiconductor Industry Association (SIA) data quoted in an international investment bank report on Taiwanese IC firms dated January 14, 2002. 11. From IPC interviews. Both Taiwanese and non-Taiwanese DRAM firms acknowledged the tight link between design and manufacturing in this product. 12. From interviews conducted by the IPC team with both Taiwanese AMLCD firms and their foreign partners during 2000–2002. 13. Ibid. 14. FPD Update, Winter 2002 (www.semi.org/web/fpdupdate.nsf/index.html). The 30 percent is in capacity terms and is equal to the capacity shares that Korea and Japan are projected to have in 2003. 15. IPC interviews. The IPC team interviewed Korean firms in 2000. 16. This estimate is based on the figures in Mathews and Cho (2000, p. 191), with a slight upward adjustment, as they gave relatively low figures for the early projects. 17. There are only ten Taiwanese-owned enterprises that could be called technology enterprises and that have over US$2 billion in revenue, and this number includes state-owned China Steel, which is included due to its IC mask operations. (Authors’ calculations based on data from Tianxia [Commonwealth] May 25, 2001.)

Notes to Chapter 4 1. EIU 2003a. 2. Chow 2002, p. 255. 3. Douglas Fuller, chapter 5 in this volume. 4. Chung 1997, p. 181. 5. Gilpin and Gilpin 1987, p. 99. 6. Cumings 1984, p. 46. 7. The industry-level version of the life-cycle model, also known as the “flying geese” model, has been attributed to the Japanese economist Akamatsu Kaname. The life cycle of individual products and the relationship to firm competitiveness was the focus of Raymond Vernon’s work in the 1970s. Also see Ravenhill and Bernard 1995, pp. 172–83.

NOTES TO CHAPTER 4

285

8. Ravenhill and Bernard 1995, p. 184. 9. Ravenhill and Bernard also note that the rapid rate of product change and more complex technologies have made reverse engineering increasingly difficult (1995, p. 177). 10. Gereffi and Applebaum 1994, p. 43. 11. Gereffi 1994, p. 99. 12. Directorate-General of Budget, “Monthly Bulletin of Earnings & Productivity Statistics,” No. 317; Accounting and Statistics, Executive Yuan. March 2000. 13. Treece 2003, p. 4. 14. Taiwan Transportation Vehicle Manufacturers Association (TTVMA) statistical website, July 2003, www.ttvma.org.tw/chineseinfo.htm. 15. Directorate-General of Budget, “Monthly Bulletin of Earnings & Productivity Statistics,” No. 317; Accounting and Statistics, Executive Yuan. March 2000; United States Department of Labor, Bureau of Labor Statistics, 2000. 16. Humphrey and Memedovic 2003, p.5. General Motors, Ford, Toyota, and Volkswagen each produced over five million vehicles. 17. China Motor Corporation, the largest firm, produced 93,000 vehicles in 2002 (TTVMA 2003). 18. Shapiro and Heath 2002, p. 58. 19. TTVMA 2003. 20. This refers to the consortium of Taiwanese firms organized (and subsidized by the government) to design a common engine, as discussed later in this chapter. According to the project director, “The Japanese did everything they could to stop us.” Quoted by Noble 1996, p. 20. 21. Part of this emphasis might have resulted from security concerns. As Wu argues, “ . . . from 1950 to 1970, the bulk of government expenditure, between 50 percent and 80 percent, [was allocated] to defense and diplomacy while 0.1 percent to 3.1 percent [was allocated] to the economy and transportation. This is overwhelming evidence of the priority of political and defense goals over economic goals” (Wu 2005, p. 20). 22. Noble 1996, p. 10. 23. Interview with retired senior CEPD official, January 20, 2003. 24. Noble 1996, p. 13. 25. Ibid. 26. Wu 2005, p. 20. 27. Interview with representatives of Mechanical Industrial Research Laboratory (MIRL), January 20, 2003; Interview with representatives of China External Trade Development Council (CETRA), January 23, 2003. 28. Interview with CETRA representatives, January 23, 2003. 29. The four objectives of MIRL, according to a director, have been: the importation of technology critical to the auto sector; the development of modular capabilities in key components and system integration to be eventually transferred to local industry; the development of enterprise software such as computer-aided design/computer-aided manufacturing (CAD/CAM), product development management (PDM) and manufacturing execution system (MES); and the development of R&D strategies. Interview with MIRL representative, January 20, 2003. 30. Interview with MIRL senior management, January 20, 2003.

286 NOTES TO CHAPTER 4

31. Liang 2003a, China Economic News Service 1999, China Economic News Service 2003b. 32. China Economic News Service 2002a, 2002b. 33. Noble 1996, citing a CMC study, p. 9. 34. Interview with project general manager and deputy general manager, CMC, January 28, 2003. 35. Veloso, Romero, and Amsden (1998) report that average local content is 50 percent. TTVMA (2000) suggests local content is about 60 percent. Trade data confirm that actual local content is somewhere between these estimates. 36. Guiheux 1999, p. 114. 37. Noble 1996, p. 17. 38. TTVMA 2003. 39. Interview with president and manager, leading Taiwanese auto suppliers, January 17, 2000, January 18, 2000. 40. Interview with president, Taiwanese auto supplier, January 24, 2003; TTVMA 2003. 41. China Economic News Service 2003c. 42. Gereffi 1994, p. 99. 43. Interview with vice president of leading aftermarket firm, January 27, 2003. 44. Noble 1996, TTVMA 2000. 45. Interview with U.S. repair shop owner and board member of numerous aftermarket associations, October 17, 2003. 46. The authors are indebted to Tim Sturgeon for this point. 47. For the distinction between the different aspects of modularity, see Fine 1998, pp. 136–37. 48. In fact, many of the leading aftermarket manufacturers do not rely on the Japanese for most of their technology. For one of the largest headlamp manufacturers in Taiwan, Japanese technology is found in none of the stamping equipment, none of the assembly equipment, none of the plating equipment, 2.8 percent of the testing equipment, and 10 percent of the tooling equipment. While approximately half of the polychrome injection molding equipment is Japanese, this is a result of past purchasing decisions and not current trends. DEPO brochure “Profile for Ming Yang Traffic Industrial Co., Ltd.” Interview, August 7, 2003. 49. Sources for all estimates shown include Automotive Aftermarket Industry Association (AAIA) 2003/2004 Aftermarket Factbook, research kindly provided by Karen Fierst, Taiwan Economic News of China Economic News Service, Alliance of American Insurers, discussions with insurance industry experts at CCC Information Services Inc., and interviews with automotive industry leaders in the United States and Taiwan. 50. Ibid. 51. Bodyshop Business, June 2003, www.bodyshopbusiness. com 52. AAIA 2003, p. 3. 53. Ibid., p. 34. 54. Bodyshop Business, June 2003. 55. Ibid. 56. Interview with vice president of Gordon Auto Body Parts, January 27, 2003. 57. Ibid. 58. Ibid.

NOTES TO CHAPTER 4

287

59. Ibid. 60. China Economic News Service 2003c. 61. Ibid. 62. China Economic News Service 2004b. 63. Interview with branch manager August 1, 2003; company Web site: www.genera.com/corp.html 64. Interview with branch manager, August 1, 2003. 65. Interview with branch manager August 1, 2003; company Web site: www.dbmreflex.qc.ca/ie5/e_i5_pr2.html 66. See company Web site: www.dbmreflex.qc.ca/fiches/SuccessStory.pdf 67. Interviews with firm president and U.S. subsidiary president, January and August, 2003. 68. Ibid. 69. China Economic News Service 2004a. 70. Li 2004. 71. Ibid. 72. Organisation Internationale des Constructeurs d’Automobiles, www.oica.net. 73. Even after full WTO reductions are implemented, Taiwan will have a lower tariff rate (17.5 percent by 2007) than China (25 percent by 2006). 74. The strong possibility of nontariff barriers even after the removal of formal tariff barriers strengthens the incentive to create a presence in these large and rapidly growing markets. 75. Interviews with the Taiwanese managers of two separate Japanese-Taiwanese auto component joint ventures, January 18, 2000. As Noble (1996) notes: “control over engines [gives] the Japanese mother firm effective veto power over any Taiwan export or investment plans,” p. 32. 76. Noble 1996. 77. Moises 2000, p.8. 78. This last investment followed a May 2000 agreement by its stockholders to earmark 20 percent of the company’s value for investment in China. Agence FrancePresse, July 13, 2000. 79. Interview with project general manager and deputy general manager, Taiwanese automotive assembler, January 28, 2003. 80. Specifically, he commented, it was only at this level of production that the firm could afford to carry out design work on new models. Interview, January 25, 2000. 81. The information in this paragraph is based on an interview with a manager in the sales department of a Taiwanese auto components firm, January 19, 2000, and an interview with a Taiwanese manager at a Taiwanese-PRC joint venture in the PRC, January 25, 2000. 82. Interview with project general manager and deputy general manager, Taiwanese automotive assembler, January 28, 2003. 83. Ibid. 84. Interview with general manager, Taiwanese auto supply firm, July 11, 2000. 85. Interestingly, the organization of supply relations is the same for both forms of supplier relocation, with both initially relying on Taiwanese inputs. In some cases, this reliance is in the form of imports from Taiwan. In other cases, firms source directly from Mainland plants of other Taiwanese firms. In these cases, offshore

288 NOTES TO CHAPTER 4

investments reproduce and reinforce supply arrangements developed in the domestic arena. 86. Interview with assistant manager, Taiwanese auto supplier, January 22, 2003; Interview with vice president, Taiwanese auto supplier, January 28, 2003; Interview with vice president, Taiwanese auto supplier, January 27, 2003. 87. Interview with vice president, Taiwanese auto supplier, January 27, 2003. 88. Ibid. 89. Although some of these other investments could be in nonautomotive sectors, like motorcycle parts, they suggest, in any case, a diversified approach to foreign production. Sample collected by authors between 1997 and 1999. 90. Interview with vice president, Taiwanese auto supplier, January 27, 2003. 91. However, the EU’s recent block exemption changes may serve to force liberalization of the European parts aftermarket. Mackintosh 2003. 92. Taipei Times 2002, p.18. 93. “Taiwan’s Yulon Nissan Sees Domestic 2004 Sales Up 7.7 Percent,” Agence France-Presse, November 2, 2003. 94. China Economic News Service 2003d; “Mercedes-Benz Van Project Approved,” December 10, 2003, www.daimlerchrysler.com 95. China Economic News Service 2003a. 96. In 2002 alone, three foreign automakers invested over US$1 billion each in new 50–50 JVs: in August, Toyota committed US$1.3 billion to a JV with First Auto Works; in September, Hyundai committed US$1.1 billion to a JV with Beijing Auto Industry Corp; in October, Nissan announced plans to invest US$1 billion in a JV with Dongfeng. 97. Thun forthcoming; ChinaOnline 2000a, 2000b. 98. It was the need to use local suppliers and the uncertainty over future viability of the market that led auto companies to traditionally assemble older models in emerging market operations. Old equipment could easily be transferred from the home country, and it was generally easier to find local suppliers in the host country for parts that were not as technologically advanced. 99. Thun forthcoming. 100. On the increasing prevalence of follow design and follow sourcing and their implications, see Humphrey and Memedovic 2003. 101. The design capabilities of Taiwanese firms vary, but the R&D divisions are invariably small in comparison to their Japanese counterparts, upon whom they must heavily lean. At one components firm, a JV with a Japanese company, the R&D section consisted of twenty-three people, and was directed by a Japanese manager. In Japan, the partner had a 900-person design group. Interview with three managers, development department, Taiwanese components firm, January 18, 2000. 102. Interview, January 28, 2003. 103. See Sturgeon and Lee 2001. 104. For the classic version of this argument, see Gerschenkron 1962, p. 11. 105. Noble 1998, p. 72. 106. Ibid., p. 10. 107. Ibid., p. 15. 108. Government Accounting Office (GAO) Report GAO-01–225, “Motor Vehicle Safety,” January 2001, pp. 7–8.

NOTES TO CHAPTERS 4 AND 5 289

109. Interview with president, Automotive Research and Testing Center, January 24, 2003. 110. Interview with Karen Fierst, consultant and former deputy executive director of CAPA, July 17, 2003. 111. Interview with CETRA, January 23, 2003. 112. Rogers 1998. 113. AAIA 2003, p. 24. 114. Fishman 2003, p. 68. 115. Ives 2004, p. 4. 116. Evidence is mounting that Mainland Chinese auto parts manufacturers, for example, are entering developed markets, despite technological dependence: “In October 2002, at an Economist conference in China, [one component manufacturer] dared to suggest that, within a few years, the Chinese parts-makers might start buying up their weaker European and US counterparts. Most of the audience was aghast. Then one delegate put up his hand. ‘It’s already started,’ he said nervously. ‘SAIC [one of the biggest Chinese firms] has set up an office in Detroit just for that purpose.’ It was like listening to an axle drop” (EIU 2003a). 117. On the divergent strategies of Taiwanese auto firms, see Lynch 2000. 118. Interview with executive vice general manager, Taiwanese components firm, July 11, 2000.

Notes to Chapter 5 1. One benchmark for developed country status is US$10,000 per capita, which Taiwan achieved in the early 1990s. 2. The semiconductor production value was calculated by converting NT$652.9 billion (New Taiwan dollars) into U.S. dollars, at an exchange rate of NT$34: US$1. 3. These figures were derived from using the same exchange rate as above and aggregate GDP and industrial production data from the Ministry of Economic Affairs Web site, www.moea.gov.tw 4. There is an informative and large literature on Taiwan’s policy toward the electronics industry, including Dedrick and Kraemer (1998), Ernst (1995), Hong (1997), Hsu (1997), Kawakami (1996), Mathews (1995), Mathews and Cho (2000), Meaney (1994), Wang (1995), and Wu (1992). 5. Floyd and Meyer rank the utility patents of the top thirty technology companies. Thus, their definition of high-technology utility patents essentially means the utility patents of firms they deem to be involved in high-technology areas. All the companies in their rankings are in the IT industry. Thus, their definition of hightechnology seems to be synonymous with IT. 6. There was a later PowerPC initiative from 1993 to 1997 to transfer technology for the Motorola-IBM alternative to the dominant Wintel standard. 7. For U.S. data, see Macher, Mowery, and Hodges (1999), pp. 245–86. For Taiwanese data, see ITIS (Industrial Technology Information Service) 1998, p. VIII–15. 8. While some of the fall in average selling price is attributable to cheaper components, the ability of the branded firms to cut prices also was a large factor (IIY 2003: 8, 23–24). This decline in price is even more alarming considering that the foreign brands were giving the Taiwanese orders for increasingly higher-end note-

290 NOTES TO CHAPTERS 5 AND 6

books as only a few branded firms, such as IBM, Sony, and Toshiba, continued to retain high-end notebook production in-house (IPC interviews). 9. Strictly speaking, OEM (original equipment manufacturing) usually refers to the branded firm in most industries. Even in the electronics industry, OEM usually means the brand firm, but in Taiwanese parlance, the local supplier of manufacturing services for a brand firm is often called an OEM firm. The derivation of this usage is most likely the fact that the local supplier and lead firm are said to be in an OEM relationship when the supplier manufactures a finished product for the brand firm. In contrast, ODM has always referred to those firms that provide manufacturing and design to brand firms. 10. Kawakami (1996), p. 3, argues, based on the Ministry of Economic Affairs’ Industrial Development Bureau Annual Report 1982–83, that in 1979 the only PC manufacturers in Taiwan were U.S. ones, aside from possible procurement from the small-component suppliers implied by Kawakami’s data (pp. 16–17). However, at least one Taiwanese firm has been involved in minicomputer Chinese-language input device production since 1974 (IPC interview). 11. Lee and Chen (2000) argue that firms can upgrade from OEM to ODM, but, given Kawakami’s evidence from the relationships between outsourcing firms and OEM firms in the early years of Taiwan’s PC industry, this argument should also be extended backward, to the initial stage when the outsourcing firms had the incentive to upgrade the manufacturing abilities of local firms to at least a minimum acceptable level. 12. According to interview subjects, the base salary in the PRC for engineers is cheaper than in Taiwan, but the social benefits that the government requires the companies to pay makes the salaries almost equivalent. The stock bonus system may work differently in both countries, but this remains to be seen.

Notes to Chapter 6 We wish to thank the Industrial Performance Center at MIT for financial and institutional support. We also thank Suzanne Berger, David Birnbaum, Doug Fuller, Gary Herrigel, Richard Lester, Teresa Lynch, Ed Steinfeld, and Tim Sturgeon, for their comments and suggestions. 1. Interviews with managers at Taiwanese textile and electronics firms conducted by the IPC Globalization Team, 1999–2003. 2. Electronics contract manufacturing is also often called electronics manufacturing services (EMS). 3. However, this picture needs to be nuanced, as in 1996 four firms (IBM, SCI Systems, Motorola, and Philips) represented more than 95 percent of all electronics exports from the state of Jalisco. See Dussel Peters 1999, p. 29. 4. In 1999, Flextronics and SCI expanded their operations in Hungary, Celestica bought the Czech manufacturing facilities of the German company Gossen-Metrawatt, and Solectron expanded its manufacturing operations in Romania (Cameron 2000). It thus seems that North American ECMs were developing a “global footprint,” rather than simply following a regional strategy. 5. According to the American Electronics Association, “high-tech imports” include computers and office equipment, consumer electronics, communications equipment, electronic components, semiconductors, industrial electronics, electromedical equipment, and photonics.

NOTES TO CHAPTERS 6 AND 7 291

6. Based on a list in Textile World. Nonapparel textile firms (e.g., carpet producers) have been omitted. 7. Tarrant Apparel Group, one of the largest and most successful U.S. privatelabel manufacturers, had built a textile plant in Mexico in the late 1990s as the linchpin of a regional sourcing strategy. By 2003, the firm took a US$22.3 million loss on its Mexican textile facilities and a US$11 million write-down on its Mexican inventory, negotiated long-term leases for its mills, and its CEO announced that it “became obvious to us that we needed to exit our Mexican initiative.” Tarrant Apparel Group Press release, www.tags.com/news/20030814.html, August 14, 2003. 8. Interview, November 12, 2003. 9. Interviews with textile and apparel managers, September 2002 through December 2003. 10. Interview with textile and apparel consultant David Birnbaum of Third Horizon Ltd., March 2002. 11. Interview with mill manager, April 12, 2002. 12. Interview with apparel firm president, April 2003. 13. Interview with Linda Dvorak, director of the Garment District Industry Council, New York. 14. According to the Mexican government, security expenses to ensure the transportation of goods across the Mexico–U.S. border can reach 1 percent of these goods’ value. Secretaría de Economía, “Programa para la Competitividad de la Industria Electrónica y de Alta Tecnología,” Mexican Government (2002), p. 33. 15. The authors thank Teresa Lynch for this insight. 16. Thanks to Tim Sturgeon for drawing our attention to this point. 17. According to the Mexican government, while there are 3,676 scientists and engineers working in R&D for 100,000 inhabitants in the United States and 2,250 in South Korea, there are only 214 in Mexico. Secretaría de Economía, “Programa para la Competitividad de la Industria Electrónica y de Alta Tecnología,” Mexican Government (2002), p. 38. 18. Interview with the MIT Globalization Team, May 2000. 19. Thanks to Ed Steinfeld for drawing our attention to this point.

Notes to Chapter 7 1. In international patent issuing, one of the most reliable proxies for industrial innovation, Taiwan moved from issuing one patent in 1973 to issuing 3,693 in 1999, reaching, as early as 1997, a rate of international patents per capita that is higher than any of the G7 countries except Japan and the United States (Hall et al. 2001; Trajtenberg 2001). 2. There is little research that supplies accurate data on which activity grants the highest rents. However, studies done of the global hard-drive industry give a striking picture of the much larger percentage of rents captured by the highest-skilled activities, compared with any other activity in the value chain (Gourevitch et al. 2000). The same picture appears in the analysis of the sales per employee figures of the Israeli versus the Irish software industries, as well as any of these two industries compared with the Indian one (Breznitz 2002, 2004b). 3. One exception is the study by Fuller et al. 2003. For examples of microinstitutional industrial case studies that try to explain both the strengths and weak-

292 NOTES TO CHAPTER 7

nesses of specific national or regional industries, see Herrigel 1994, Hollingsworth et al. 1994, and Streeck 1996. For examples of studies that concentrate on the software industry, see Mowery 1996. For studies in the statist approach and its particular variant of the developmental state, see Amsden 1989; Evans 1995; Johnson 1982; O’Riain 2000, 2004; and Wade 1990. 4. According to III figures, the software industry grew from NT$11 billion (New Taiwan dollars) in sales in 1988 to NT$149 billion in 2002—an impressive growth, if vastly smaller than the growth of the semiconductor sector. However, there are very good reasons to suspect that these figures are inflated. Adding up the figures of all the leading Taiwanese software companies does not approach this figure. In our own attempt to understand the size of the Taiwanese-owned software industry, we took the software sales figures of Taiwan’s public software companies, added the sales figures we got from the top private companies, and added on top of that the total budget of III to reach a figure that is just about US$1 billion (NT$34 billion). We suspect that the official figure includes sales of both MNCs and Taiwanese companies, represents total sales figures and not just software sales, and/or that the definition of software used is a very broad one and includes, for example, Internet service providers (ISPs). Moreover, even according to those figures, exports in 2001 were only NT$16 billion (around US$400 million). This amount is still higher, but strikingly close to our own estimate, leading us to suspect that our estimate of the total size of the domestic industry is closer to reality than the official figures. These figures, putting exports at about 10 percent of sales, indicate that Taiwanese software companies face significant difficulties in developing an internationally competitive industry (III, various years). 5. In the period 1999–2004 a team of researchers from MIT’s Industrial Performance Center (IPC) investigated globalization and industry reorganization in several sectors. Field research for the IPC Globalization Study has consisted of semistructured qualitative interviews with company personnel and individuals from government agencies, unions, and academia. By early 2004, 629 personal interviews and plant tours were conducted in Canada, Mainland China, France, Germany, Indonesia, Italy, Japan, Korea, Malaysia, Mexico, the Philippines, Romania, Singapore, Spain, Taiwan, Thailand, and the United States. Of these, 286 were conducted in Taiwan or with Taiwanese companies outside Taiwan, of which 235 were linked to the IT industry. These interviews were given under condition of confidentiality and as such are masked to protect the interviewees’ identities unless the interviewee explicitly authorized us to attribute his or her remarks. 6. There is some diversity between the different neodevelopmental state theorists, in particular between those who argue that the state should advance national champions, and those who prefer a strategy based around small- and medium-sized enterprises (SMEs). It is unclear whether these differences arise because of the different context in location, timing, and industries that the different writers researched. However, in the critical area of theorizing about the optimal role and behavior of the state in its attempt to spur the growth of high-technology industries, and in their treatment of the state for all practical purposes as a unitary actor, the authors are similar enough that we can treat them as advancing one model. 7. On the more recent policies of ITRI, see Amsden and Chu 2003; Hong 1997; and Mathews and Cho 2000. 8. The amount that DOIT channels directly to private industry started at about

NOTES TO CHAPTERS 7 AND 8 293

1 percent of the total DOIT budget in 1998 with NT$200 million, and has been growing steadily both in size and as a percentage of budget to NT$2.8 billion out of a NT$17.2 billion budget in 2003. 9. IPC interview with the Taiwanese VC association, November 2003. 10. Exit refers to the change of ownership in which the VC sells its share in a company and “exits” it. 11. The strength of Taiwan’s OEM and ODM companies is reflected in the fact that in 2001 Taiwanese companies manufactured 70 percent of global production of motherboards, 55 percent of laptops, 56 percent of LCD monitors, 51 percent of color-display-tube monitors, and 36 percent of digital still cameras (III various years; ITRI various years; MoEA various years). 12. Again, exact statistics are hard to come by; the funding III received from DOIT in 2003 was NT$760 million. DOIT statistics estimate III’s 2003 total budget as NT$3 billion, out of which a third was received from projects with other government departments and a third was received from the private market (DOIT 2003). In comparison, the combined total sales of Cyberlink and Ulead, the two leading Taiwanese software-product companies, were US$65 million, or NT$2 billion, in 2002. (Trend Micro, which is now a Japanese company, had sales of US$364 million.) In a few interviews we conducted in 2001 with III’s management team we were told that half of III’s budget originates from sources other than DOIT. 13. The story of Jau Huang, a graduate of UCLA, marks a very similar dynamic to one commonly encountered in the hardware sector, the growing stream of returning Taiwanese with graduate degrees from leading U.S. universities. 14. For a different perspective, see Fuller et al. 2003.

Notes to Chapter 8 1. Statistical Yearbook of the Republic of China, 2003, at www.stat.gov.tw/bs2/ 2003YearBook.pdf 2. World Bank GNI figures, 2003, at www.worldbank.org/data/databytopic/ GNIPC.pdf 3. Roughly, the idea is that “recipe books” can be written for processes that in the past could be successfully run only with tremendous prior experience, vast amounts of accumulated tacit knowledge, and a certain degree of “art.” 4. My understanding of modularity, particularly in the electronics industry, has been influenced immeasurably by the ongoing research of Charles Sodini, Tayo Akinwande, and Douglas Fuller of the Massachusetts Institute of Technology. 5. The term “integral” comes from Fujimoto 2002. 6. Examples would be Microsoft on PC operating software and Intel on processor designs. Controlling the rules means that information can be dispensed selectively to downstream producers, customers can be locked in, and premiums can be charged on new products. 7. The challenge here becomes to move into an integral component or module, one around which upstream and downstream producers must tailor their products, but one that is not easily copied by others. Again, examples, however fleeting, would be Intel processors or Microsoft operating software. 8. The distinction between “sustaining” and “disruptive” innovation (and innovators) comes from Christensen 1997; Aoki and Takizawa 2002.

294 NOTES TO CHAPTER 8

9. Aoki and Takizawa 2002. 10. This, in a sense, is what Fine refers to as “clockspeed” (Fine 1998). 11. Similar attributes are described with great clarity in Nolan 2001. 12. The World Bank’s survey involved 1,500 “higher-tech” firms (500 firms in each of the following cities: Beijing, Shanghai, Chengdu, Guangzhou, and Tianjin). For further discussion and analysis of the survey, see: Yusuf, Altaf, and Nabeshima 2004. 13. The 2001 annual report is available at www.lenovogrp.com 14. The 2001 annual report available at www.ibm.com 15. In the U.S. market, where Haier has recently established a refrigerator manufacturing facility, Whirlpool, General Electric, Maytag, and Electrolux (Frigidaire) account for 98 percent of the refrigerators sold (Sprague 2002). 16. There were 995 manufacturing firms in the sample. 17. Of the five cities surveyed (Beijing, Shanghai, Guangzhou, Chengdu, and Tianjin), Guangzhou proved to be somewhat of an exception to this point, with respondents reporting higher levels of overseas suppliers and lower levels of intracity supply network concentration. 18. Again, Guangzhou was the exception, with respondents reporting 38 percent of the customer base within the city, and 29 percent located overseas. 19. Downstream, 67 percent of goods were moved by surface transportation, and 84 percent of all communications were conducted either face to face, by phone, or by fax. Upstream, 77 percent of goods were delivered via surface transportation, and 87 percent of communications were either face to face, by phone, or by fax. 20. A positive example would be the Taiwan Semiconductor Manufacturing Corporation (Burns 2000). 21. In the 2001 survey, 36 percent of the respondents reported the introduction of new products within existing lines, and 21 percent reported new products in entirely new lines of business. 22. Konka, China’s second-largest television producer, lost US$84.5 million in 2001 (Leggett 2001). TCL International reported a 32 percent drop in profits that same year, and like Konka, has moved into other appliance manufacturing (Stevens 2002). 23. Haier, China’s largest refrigerator and air conditioner manufacturer—and an emerging global brand—suffered a 45 percent year-on-year drop in net profit in the first half of 2002. The company’s response has involved an upping of exports of low-end air conditioners, and a further ramping up of refrigerator and air conditioner production domestically (Dow Jones International News 2002). 24. For a useful definition of what just such a transformation means, see Kornai 2000, 24–42; Steinfeld 2002. 25. A change in attitude was signaled in 1997 when the communist party recognized private firms as an important part of the Chinese economy. That change was followed up in 1999 with an amendment to the Chinese constitution officially recognizing private enterprise. Finally, in 2001, China’s president and communist party general secretary, Jiang Zemin, called for the welcoming of private entrepreneurs— capitalists, in effect—into the communist party. 26. On the ramifications of informality, see Mackenzie 2001; de Soto 2000.

NOTES TO CHAPTER 8

295

27. Or they engage in informal channels of financial intermediation, credits that tend to be high in price and small in scale (Tsai 2002). 28. The implication is that China has an inordinately high demand for FDI. 29. For alternative perspectives on the problem, see Naughton 1999; Young 2000. On interprovincial trade, see also Hu and Wang 1999. 30. For more on the regional nature of development and institutional reach, see Thun and Segal 2001. 31. Decentralization was perhaps politically necessary—and sound from a policy sense—to allow firms to start up and avoid policy rigidity. The problem is that once those firms are established and need to grow, local idiosyncratic institutions and local discrimination inhibit growth. For an extensive discussion, see Wedeman 2003. 32. A prime example involves the forced merger between the Chongqing Iron and Steel Group and the deeply troubled Chongqing Special Steel Company. Author’s interviews, Chongqing, 2000. See also Pomfret 2001. 33. For example, Shanghai’s Baoshan Iron and Steel Group, after being forced to take on a financially insolvent local producer as a subsidiary, was subsequently able to have the subsidy’s debts forgiven through a debt-equity swap with state lenders. Author’s interviews, Beijing, 1999 and 2000. 34. Some scholars have argued that Chinese private firms, rather than innovating, spend most of their energy cultivating clientelistic ties with political patrons (Wank 1999). Others have argued that clientelism has receded in recent years, as institutionalization has increased (Guthrie 1999). 35. Ma and Fung (2002), still place the number at 35 percent of outstanding loans. 36. For an alternative view, see Amsden 2001. 37. The point is critiqued in Amsden 2001, 2–8. 38. Ibid., p. 3. 39. Examples would include tariff exemptions for imported raw materials used in the processing of exported final products. 40. On emigration issues, see Zweig 1997. 41. Indeed, one of the arguments against FDI policy is that such policies provide foreign firms a privileged domestic institutional environment and thus encourage domestic assets to flow to them. 42. A view similar to that outlined in Schumpeter [1911] 1968. 43. The 1994 national tax reforms and 1999 national banking reforms both attempted to move away from this approach, but only partially successfully. 44. Origins of this view can be seen in Schumpeter [1942] 1976. Note the differences between Schumpeter’s 1942 perspective and his 1911 views cited earlier. 45. For a history of this perspective, see Chandler and Hikino 1990; Chandler, Amatori, and Hikino 1997. 46. A number of nations attempt this. See Steinfeld unpublished manuscript. 47. As evidenced by the disproportionate representation of SOEs in Chinese equity and debt markets. 48. www.huawei.com 49. www.huawei.com; Einhorn et al. 2002. 50. www.haier.com 51. www.haier.com

296 NOTES TO CHAPTER 8

52. www.lenovogrp.com 53. Moreover, since they are operating in their home market, it becomes hard to rule out explanations of success based on issues of privileged access, particularly when government agencies and firms are their primary customers. 54. Japanese motorcycle and auto manufacturers are an example, as is Philips in electronics. 55. Though the semiconductor side of the business remains for the most part an aspiration rather than a reality. 56. Author’s interviews, Chongqing 2000.

Bibliography Abernathy, Frederick H. et al. 1999. A Stitch in Time: Lean Retailing and the Transformation of Manufacturing—Lessons from the Apparel and Textile Industries. New York: Oxford University Press. Abernathy, Frederick H., John T. Dunlop, Janice H. Hammond, and David Weil. 2004. “Globalization in the Apparel and Textile Industries: What Is New and What Is Not?” In Locating Global Advantage: Industry Dynamics in the International Economy, edited by Martin Kenney and Richard Florida. Stanford, CA: Stanford University Press. AFX. 2001. “Via Sees P4 Chipset Dispute Impact Fully Reflected in Q4.” AFX News Limited, December 18. Agence France-Presse. 2003. “Taiwan’s Yulon Nissan Sees Domestic 2004 Sales Up 7.7 Percent.” November 2. Akamatsu, Kaname. 1962. “A Historical Pattern of Economic Growth in Developing Countries.” The Developing Economies 1: 3–25. Alford, William. 2000. “The More Law, the More . . . ? Measuring Legal Reform in the People’s Republic of China.” Center for Research on Economic Development and Policy Reform. Working Paper no. 59. Stanford University. http:// credpr.stanford.edu/publications/abstracts.html#58 American Electronics Association (AeA). “Tech Trade Update 2003,” at www.aeanet.org Amsden, Alice H. 1984. “Taiwan.” World Development 12, no. 5/6: 491–503. ———. 1985. “Taiwan’s State and Economic Development.” In Bringing the State Back In, edited by D. Rueschemeyer, P. Evans, and T. Skocpol. New York: Cambridge University Press. ———.1989. Asia’s Next Giant: South Korea and Late Industrialization. New York: Oxford University Press. ———. 1991. “Big Business and Urban Congestion in Taiwan: The Origins of Small Enterprise and Regionally Decentralized Industry (Respectively).” World Development 19, no. 2: 1121–35. ———. 2001. The Rise of “the Rest”: Challenges to the West from Late-Industrializing Economies. Oxford; New York: Oxford University Press. Amsden, Alice H., and Wan-wen Chu. 2003. Beyond Late Development: Taiwan’s Upgrading Policies. Cambridge, MA: MIT Press. Anchordoguy, Marie. 1989. Computers Inc.: Japan’s Challenge to IBM. Cambridge, MA: Council on East Asian Studies, Harvard University/Harvard University Press. Anderson, Christopher. 1998. “A Survey of Taiwan: In Praise of Paranoia.” The Economist, November 7, 3–18. Anson, Robin. 2003. “Will China Sweep the Board When Textile and Apparel Quotas Are Finally Eliminated?” Textile Outlook International, May–June. 297

298 BIBLIOGRAPHY

Aoki, Masahiko, Hyung-Ki Kim, and Masahiro Okuno. 1997. The Role of Government in East Asian Economic Development: Comparative Institutional Analysis. Oxford: Oxford University Press. Aoki, Masahiro, and Hirokazu Takizawa. 2002. “Modularity: Its Relevance to Industrial Architecture.” Saint-Gobain Centre for Economic Research, Paris, November 7–8. Arensman, R. 2003. “Fabless Goes Global.” Electronic Business, February. Arora, A., and S. Athreye. 2002. “The Software Industry and India’s Economic Development.” Information Economics and Policy 14: 253–73. Arrow, J.K. 1962. “Economic Welfare and the Allocation of Resources for Invention.” In The Rate and Direction of Inventive Activity: Economic and Social Factors, edited by National Bureau of Economic Research. Princeton, NJ: Princeton University Press. Arthur, W.B. 1989. “Competing Technology, Increasing Returns, and Lock-in by Historical Events.” Economic Journal 97: 116–31. Automotive Aftermarket Industry Association (AAIA). 2003. 2003/2004 Aftermarket Factbook. Bethesda, MD. Aw, Bee Yan, Xiaomin Chen, and Mark J. Roberts. 1997. “Firm-Level Evidence on Productivity Differentials, Turnover, and Exports in Taiwanese Manufacturing.” Cambridge, MA: National Bureau of Economic Research. ———. 1998. “Productivity and the Decision to Export: Micro Evidence from Taiwan and South Korea.” Cambridge, MA: National Bureau of Economic Research. ———. 2002.”Productivity, Output, and Failure: A Comparison of Taiwanese and Korean Manufacturers.” Cambridge, MA: National Bureau of Economic Research. Ayala, Jim, and Richard Lai. 1996. “China’s Consumer Market: A Huge Opportunity to Fail?” The McKinsey Quarterly, no. 3. Baldwin, Carliss Y., and Kim B. Clark. 1997. “Managing in an Age of Modularity.” Harvard Business Review 75, no. 5: 84–93. ———. 2000. Design Rules: The Power of Modularity. Cambridge, MA: MIT Press. Baljko, Jennifer. 2002. “ODMs Set to Lead Recovery in Contract Sector.” Electronic Business News, December 19. Barney, J.B. 1999. “How a Firm’s Capabilities Affect Boundary Decisions.” Sloan Management Review (spring): 137–45. Baum, Julian. 1999. “Taiwan Trade-Off.” Far Eastern Economic Review, August 26. Bensaou, M., and E. Anderson. 1999. “Buyer-Supplier Relations in Industrial Markets: When Do Buyers Risk Making Idiosyncratic Investments?” Organization Science 10, no. 4: 460–81. Berger, Suzanne, and Richard K. Lester, eds. 1997. Made by Hong Kong. Hong Kong: Oxford University Press. Bettis, R., S. Bradley, and G. Hamel. 1992. “Outsourcing and Industrial Decline.” Academy of Management Executive 6, no. 1: 7–22. Birnbaum, David. 1993. Importing Garments Through Hong Kong. Hong Kong: Third Horizon Press. ———. 2000. Birnbaum’s Global Guide to Winning the Great Garment War. Hong Kong: Third Horizon Press. ———. 2003. “Crisis in the Mexican Maquila Garment Industry: Federal Problem/ Regional Solutions.” Photocopy distributed by the author.

BIBLIOGRAPHY 299

Black, Thomas. 2002. “Adios, Mexico, Hola, China.” Bloomberg Markets, November. Boillot, Jean, and Nicolas Michelon. 2001. Chine, Hong Kong, Taiwan. Une Nouvelle Géographie Économique de L’asie. Paris: La Documentation Française. Borrus, Michael, and John Zysman. 1997. “Globalization with Borders: The Rise of Wintelism as the Future of Global Competition.” Industry and Innovation 4, no. 2: 148. Bosch, F.A.J., H.W. Volberda, and M. de Boer. 1999. “Co-Evolution of Firm Absorptive Capacity and Knowledge Environment: Organizational Forms and Combinative Capabilities.” Organization Science 10, no. 5: 561–68. Bout, Bram J. 2003. “Keeping Taiwan’s High-Tech Edge.” The McKinsey Quarterly, Special Edition, no. 2: 1–3. Bouteiller, Eric. 1994. “‘Attirer Un Oiseau Sur Une Branche.’” Economie Internationale, no. 57: 99–115. Bow, Josephine. 1999. “Made in Merida: The Next Hot Mexican Label.” Women’s Wear Daily, August 17. Braczyk, H.J., P. Cooke, and M. Heidenreich. 1998. Regional Innovation Systems: The Role of Governance in a Globalized World. London: UCL Press. Bradach, J., and R. Eccles. 1989. “Price, Authority, and Trust: From Ideal Types to Plural Form.” Annual Review of Sociology 15: 97–118. Breschi, S., and F. Malerba. 1997. “Sectoral Innovation Systems: Technological Regimes, Schumpeterian Dynamics, and Spatial Boundaries.” In System of Innovation: Technologies, Institutions and Organizations, edited by C. Edquist. London: Pinter. Breznitz, Dan. 2002. “Conceiving New Industrial Systems: The Different Emergence Paths of High-Technology Companies in Israel and Ireland.” Science Technology and the Economy Program of the Samuel Neaman Institute, Technion, Israel Institute of Technology, Working Paper Series 02–11. ———. 2004a. “Collaborative Public Space in a National Innovation System: A Case Study of the Israeli Military’s Impact on the Software Industry.” Industry and Innovation 12:1. ———. 2004b. “An Iron Cage or the Final Stage? Intensive Product R&D and the Evolution of the Israeli Software Industry.” Research Policy, revised and resubmitted. Brusoni, Stefano, and Keith Pavitt. 2002. “Modularity and the Division of Cognitive Labour within/across Organization.” Saint-Gobain Centre for Economic Research, Paris, November 7–8. Bueno de Mesquita, Bruce, David Newman, and Alvin Rabushka. 1996. Red Flag over Hong Kong. Chatham, NJ: Chatham House Publishers. Burns, Simon. 2000. “I-Sight: Everyone’s a Winner.” Far Eastern Economic Review, July. CADELEC (Cadena Productiva de la Electrónica). 2003. “Industria Electrónica.” Guadalajara, Mexico, http://www.cadelec.com.mx Calbreath, Dean, and Diane Lindquist. 2002. “The China Challenge: Maquiladoras in Mexico Are Facing Intense Competition from Far East Counterparts.” San Diego Union-Tribune, August 25. Calder, Kent E. 1993. Strategic Capitalism: Private Business and Public Purpose in Japanese Industrial Finance. Princeton, NJ: Princeton University Press. Cameron, Ian. 2000. “West Goes East.” Electronics Times, May 15.

300 BIBLIOGRAPHY

Carlsson, B. 1995. Technological Systems and Economic Performance—The Case of Factory Automation. London: Kluwer Academic Publishers. Carlsson, B., and G. Eliason. 1994. “The Nature and Importance of Economic Competence.” Industrial and Corporate Change 3: 687–711. Cataldo, Anthony. 2001. “Taiwan’s Foundry Titans Duel For .13-Micron Supremacy.” Semiconductor Business News (www.siliconstrategies.com) [cited April 27, 2001]. Cavey, Paul. 2003. “Leaping Dragon, Trailing Tigers? Taiwan, Hong Kong and the Challenge of Mainland China.” London: Economist Intelligence Unit. Celly, K.S., R.E. Spekman, and J.W. Kamauff. 1999. “Technological Uncertainty, Buyer Preferences and Supplier Assurances: An Examination of Pacific Rim Purchasing Arrangements.” Journal of International Business Studies 30, no. 2: 297–316. Chandler, Alfred D., Franco Amatori, and Takashi Hikino. 1997. “The Large Industrial Enterprise and the Dynamics of Modern Economic Growth.” In Big Business and the Wealth of Nations, edited by Alfred Dupont Chandler, Franco Amatori, and Takashi Hikino. Cambridge, New York: Cambridge University Press: 24–63. Chandler, Alfred Dupont, and Takashi Hikino. 1990. Scale and Scope: The Dynamics of Industrial Capitalism. Cambridge, MA: Belknap Press. Chase, Kerry A. 2003. “Economic Interests and Regional Trading Arrangements: The Case of NAFTA.” International Organization 57, no. 30: 146. Chen, Qingtai. 2002. “Promoting Regional Economic Cooperation in Asia Amidst Economic Globalization.” Paper presented at the Conference on Asian Economic Integration, Research Institute of Economy, Trade, and Industry, Tokyo, April 22–23. Chen, S.H. 2002. “Global Production Networks and Information Technology.” Industry and Innovation 9, no. 3: 249–67. Chen, J.S., and J.R. Lee. 1998. “Leveraging Resources across Multiple Business Activities: A Business Model for OEM Suppliers in Horizontally Configured Industries.” Paper presented at the Annual International Conference of Strategic Management Society, Orlando, Florida. Chen,Tain-Jy, and Ying-Hua Ku. 2003. “The Effect of Overseas Investment on Domestic Employment.” Taipei, Taiwan, Chung-Hua Institution for Economic Research. Paper prepared for the 14th East Asian Seminar on Economics. Cheng, Lu-Lin. 1998. “Wheeling the Cycle Up: Firms, OEM, and Chained Networks in the Development of Taiwan’s Bicycle Industry.” Tokyo: Institute of Developing Economies. Cheng, Tun-jen. 1990. “Political Regimes and Development Strategies: South Korea and Taiwan.” In Manufacturing Miracles: Paths of Industrialization in Latin America and East Asia, edited by Gary Gereffi and Donald L. Wyman. Princeton, NJ: Princeton University Press: 139–78. ———. 1993. “Guarding the Commanding Heights: The State as Banker in Taiwan.” In The Politics of Finance in Developing Countries, edited by Chung H. Lee, Stephan Haggard, and Sylvia Maxfield. Ithaca, NY: Cornell University Press: 55–92. ———. 2001. “Transforming Taiwan’s Economic Structure in the 20th Century.” The China Quarterly, no. 165. China Economic News Service. 1999. “Exports of Auto Parts Climbed 4% in ’98.” Taiwan Economic News, March 2, news.cens.com

BIBLIOGRAPHY 301

———. 2001a. “Chipmaker TSMC Revenue Hits NT$166.3 in 2000.” Taiwan Economic News, January 9, news.cens.com ———. 2001b.”Local Carmakers Seek Expansion in Foreign Markets.” Taiwan Economic News, March 14, news.cens.com ———. 2002a. “Taiwan Auto Exports Gain Horsepower.” Taiwan Economic News, April 7, news.cens.com ———. 2002b. “Exports of Auto Parts to Japan Reach Record High.” Taiwan Economic News, August, news.cens.com ———. 2003a. “China Motor to Sell Stake in Mainland Automaker to Mitsubishi.” Taiwan Economic News, November, www.cens.com ———. 2003b. “China Motor Hones Image with DaimlerChrysler Tie-up in Mainland China.” Taiwan Economic News, December, www.cens.com ———. 2003c. “Taiwan Auto Parts Suppliers Make Headway in OE Market.” Taiwan Economic News, February 17. news.cens.com ———. 2003d. “Mercedes-Benz Van Project Approved.” www.daimlerchrysler.com ———. 2004a. “Macauto Wins Orders from Mercedes-Benz, Hyundai. GM.” Taiwan Economic News, May 25, news.cens.com ———. 2004b. “Mobiletron’s Auto Parts Sales Up.” Taiwan Economic News, June 18, news.cenc.com. ChinaOnline. 2000a. “Sluggish Sales Cause Shanghai Volkswagen to Lower the Price of the Santana.” July 21, www.chinaonline.com ———. 2000b. “Chasing the Checkered Flag: China Carmakers Race for Sedan Supremacy.” October 19, www.chinaonline.com Chiu, Stephen. 1994. “The Politics of Laissez-Faire. Hong Kong’s Strategy of Industrialization in Historical Perspective.” Hong Kong: Hong Kong Institute of AsiaPacific Studies Occasional Paper Series OP40, Chinese University of Hong Kong. Chiu, Stephen W.K., K.C. Ho, Tai-Lok Lui. 1997. City-States in the Global Economy: Industrial Restructuring in Hong Kong and Singapore. Boulder, CO: Westview. Chow, Peter C.Y. 2002. Taiwan in the Global Economy: From an Agrarian Economy to an Exporter of High-Tech Products. Westport, CT: Praeger. Christensen, Clayton M. 2000. The Innovator’s Dilemma: When New Technologies Cause Great Firms to Fail. New York: HarperBusiness. Chu, Yun-han. 1989. “State Structure and Economic Adjustment of the East Asian Newly Industrializing Countries.” International Organization 43, no. 4: 647–72. ———. 1994. “The Realignment of Business-Government Relations and Regime Transition in Taiwan.” In Business and Government in Industrializing Asia, edited by A. MacIntyre. Ithaca, NY: Cornell University Press: 113–41. Chung, Chin. 1997. “Division of Labor across the Taiwan Strait: Macro Overview and Analysis of the Electronics Industry.” In The China Circle: Economics and Electronics in the PRC, Taiwan, and Hong Kong, edited by Barry Naughton. Washington, DC: Brookings Institution Press: 18–30. Clendenin, M. 2003. “Price Battle Looms as Taiwan Preps Low-cost WLAN ICs.” Semiconductor Business News, July 21. http://www.siliconstrategies.com Cohen, Margot. 2002. “Biker Wars: Chinese and Japanese Manufacturers Battle for Vietnam’s Fast-Growing Motorcycle Market.” Far Eastern Economic Review, November 7. Cohen, W., and D. Levinthal. 1990. “Absorptive Capacity: A New Perspective on Learning and Innovation.” Administrative Science Quarterly 35: 128–52.

302 BIBLIOGRAPHY

Christensen, Clayton M. 1997. Market Strategy: Learning by Doing. Boston: Harvard Business School Press. Cooper, Eugene. 1998. The Artisans and Entrepreneurs of Dongyang County. Armonk, NY: M.E. Sharpe. Culpepper, D.P., and D. Finegold. 2001. The German Skills Machine: Sustaining Comparative Advantage in a Global Economy. Oxford: Berghahn Books. Cumings, Bruce. 1984. “The Origins and Development of the Northeast Asian Political Economy: Industrial Sectors, Product Cycles, and Political Consequences.” International Organization 38 (winter). Curry, J., and M. Kenney. 1999. “Beating the Clock: Corporation Responses to Rapid Change in the PC Industry.” California Management Review 42, no. 1: 8–36. Dean, Jason. 2003. “BenQ Builds a Brand.” Far Eastern Economic Review, October 23, 38–41. Dean, J., and T. Uimonen. 2002. “Taiwan Chip Makers Set Their Sights on China.” Wall Street Journal, April 1, A8. Dedrick, J., and K. Kraemer. 1998. Asia’s Computer Challenge: Threat or Opportunity for the United States and the World. New York: Oxford University Press. de Soto, Hernando. 2000. The Mystery of Capital. New York, Basic Books. Dhayagude, Tushar, et al. 2001. “Is the IDM Model Doomed: Emergence of the Fabless-Foundry Model in the Semiconductor Industry.” In Kellogg Graduate School of Management. Evanston, IL at www.fsa.org/resources/whitepapers/ dhayagude060401.pdf Dhume, Sadanand. 2001. “Road Warriors: In Indonesia, Chinese Motorcycle Manufacturers Are Challenging the Japanese for a Slice of the Market.” Far Eastern Economic Review, December 27. Digitimes. 2002. Untitled article by Hans Wu and Jane Wang, April 30. Accessed at http://www.digitimes.com Directorate-General of Budget. 2000. “Monthly Bulletin of Earnings & Productivity Statistics,” No. 317. Accounting and Statistics, Executive Yuan, March. DOIT. 2003. “TDP Annual Budget 1993–2003.” Taipei: Ministry of Economic Affairs, Department of Industrial Technology. Doner, Richard. 1992. “Limits of State Strength: Toward an Institutionalist View of Economic Development.” World Politics 44, no. 3. Dosi, G., R.R. Nelson, and S.G. Winter. 2000. “Introduction: The Nature and Dynamics of Organizational Capabilities.” In The Nature and Dynamics of Organizational Capabilities, edited by R.R. Nelson, G. Dosi, and S.G. Winter. New York: Oxford University Press: 1–22. Dow Jones International News. 2002. “China’s Haier First Half Profit Down 45% on Low-Priced Exports.” July 29. Doz, Yves, Jose Santos, and Peter Williamson. 2001. From Global to Metanational. Boston: Harvard Business School Press. Dussel Peters, Enrique. 1999. “La Subcontratación Como Proceso De Aprendizaje: El Caso De La Electrónica En Jalisco (México) En La Década De Los Noventa.” Santiago de Chile: Comisión Económica para América Latina y el Caribe (CEPAL). ———. 2000. “El Tratado De Libre Comercio De Norteamérica Y El Desempeño De La Economía En México.” Santiago de Chile: Comisión Económica para América Latina y el Caribe (CEPAL). Dyer, J., and K. Nobeoka. 2000. “Creating and Managing a High-Performance

BIBLIOGRAPHY 303

Knowledge-Sharing Network: The Toyota Case.” Strategic Management Journal 21: 345–67. Dyer, J.H. 1997. “Effective Inter-Firm Collaboration: How Firms Minimize Transaction Costs and Maximize Transaction Value.” Strategic Management Journal 18, no. 7: 535–56. Dyer, J.H., and H. Singh. 1998. “The Relational View: Cooperative Strategy and Sources of Interorganizational Competitive Advantage.” Academy of Management Executive 23, no. 4: 600–79. East Asia Analytical Unit, Department of Foreign Affairs and Trade, Australia. 1995. Overseas Chinese Business Networks in Asia. Canberra: AGPS Press. The Economist. 2000. “Successful Manufacturers These Days Need Not Go Anywhere Near a Factory.” February 12. ———. 2003. “Technology in China: The Allure of Low Technology.” December 20, 99. Einhorn, Bruce, Ben Elgin, and Andy Reinhardt. 2002. “The Well-Heeled Upstart on Cisco’s Tail.” Business Week, October 28. Eisenhardt, K.M., and J.A. Martin. 2000. “Dynamic Capabilities: What Are They?” Strategic Management Journal 21: 1105–21. EIU. 1999. “Mexico Economy–Foreign Direct Investment Overview.” London: Economist Intelligence Unit, October 20. EIU. 2003a. “Automotive Forecast: Western Europe.” London: Economist Intelligence Unit. ———. 2003b. “Country ViewsWire Taiwan.” London: Economist Intelligence Unit. ———. 2003c. “Taiwan: Country Profile 2003.” London: Economist Intelligence Unit. Electronic Business. 2000. “Top 100 Contract Manufacturers.” September. Accessed at http://www.eb-mag.com ———. 2002. “Top 100 Contract Manufacturers.” September. Accessed at http:// www.eb-mag.com ———. 2003. “Top 100 Contract Manufacturers.” September. http://www.ebmag.com Enright, Michael, Edith E. Scott, and David Dodwell. 1997. The Hong Kong Advantage. Hong Kong: Oxford University Press. Ernst, Dieter. 1995. “New Opportunities and Challenges for Taiwan’s Electronics Industry—The Role of International Cooperation.” BRIE Working Paper 78. Berkeley, CA: Berkeley Roundtable on International Economy. ———. 2000. “Inter-Organizational Knowledge Outsourcing: What Permits Small Taiwanese Firms to Compete in the Computer Industry?” Asia Pacific Journal of Management 17: 223–55. ———. 2002. “Digital Information Systems and Global Flagship Networks—a New Divide in Industrial Organization.” Industry and Innovation 9, no. 2. ———. 2003. “Internationalisation of Innovation: Why Is Chip Design Moving to Asia?” East-West Center Working Paper, Economic Series, no. 64 (Honolulu: East-West Center). Evans, P. 1995. Embedded Autonomy: States and Industrial Transformation. Princeton, NJ: Princeton University Press. Evans, P., D. Rueschemeyer, and T. Skocpol. 1985. Bringing the State Back In. New York: Cambridge University Press. Fang, Xinghai. 2001. “Reconstruction of the Micro-Foundations of China’s Finan-

304 BIBLIOGRAPHY

cial Sector.” Paper presented at the Conference on Financial Sector Reform in China, Harvard University, September. Farrell, J., and G. Saloner. 1985. “Standardization, Compatibility, and Innovation.” Rand Journal 16: 70–83. FDP Update. 2002. “Semiconductor Equipment and Materials International.” Winter. Available from www.semi.org/web/fpdupdate.nsf/index.html Feenstra, Robert C., Deng-Shing Huang, and Gary Hamilton. 1997a. “Business Groups and Trade in East Asia: Part 1, Networked Equilibria.” Cambridge, MA: National Bureau of Economic Research. ———. 1997b. “Business Groups and Trade in East Asia: Part 2, Product Variety.” Cambridge, MA: National Bureau of Economic Research. Fei, John C. H., Gustav Ranis, and Shirley W.Y. Kuo. 1979. Growth with Equity: The Taiwan Case. World Bank Research Publication. New York: World Bank, Oxford University Press. Ferreira, Alvaro. 2003. “Textiles and Clothing in Mexico: Facing up to Growing International Competition.” Textile Outlook International, July–August: 10–34. Fields, Karl J. 1995. Enterprise and the State in Korea and Taiwan. Ithaca, NY: Cornell University Press. Financial Times. 2003a. “Mediatek to Challenge World’s No. 4 Spot in IC Design.” Financial Times Global News Wire, November 25. ———. 2003b. “Taiwanese Chip Designers Facing Lukewarm Season Earlier.” Financial Times Global News Wire, December 9. ———. 2003c. “Via, Intel Reach Patent Suit Settlement, to Jointly Face Competitors.” Financial Times Global News Wire, April 9. Fine, Charles H. 1998. Clockspeed: Winning Industry Control in the Age of Temporary Advantage. Reading, MA: Perseus Books. Fishman, Ted C. 2003. “Here Come the Discountmobiles.” New York Times, September 28. Fligstein, N. 1991. “The Structural Transformation of American Industry: An Institutional Account of the Causes of Diversification in the Largest Firms, 1919– 1979.” In The New Institutionalism in Organizational Analysis, edited by W.W. Powell and P. Dimaggio. Chicago: University of Chicago Press: 311–36. ———. 1996. “Markets as Politics: A Political-Cultural Approach to Market Institutions.” American Sociological Review 61: 656–73. Floyd, D. and P. Meyer. 2002. “2001 Global Patent Trends.” Global Equity Research Asia, Lehman Brothers, June 17. Forero, Juan. 2003. “Mexico Manufacturers Lose Business to China.” New York Times, September 3. Fu, Tsu-Tan, Cliff J. Huang, and C.A. Lovell, eds. 1999. Economic Efficiency and Productivity Growth in the Asia-Pacific Region. Northampton, MA: Edward Elgar. Fujimoto, Takahiro. 1999. The Evolution of a Manufacturing System at Toyota. Oxford: Oxford University Press. ———. 2002. “Organizational Diversity and Corporate Performance.” Paper prepared for the Conference on Innovation within Firms, Saint-Gobain Centre for Economic Research, Paris, November 7–8. Fuller, D., A. Akinwande, and C.G. Sodini. 2003. “Leading, Following or Cooked Goose: Successes and Failures in Taiwan’s Electronics Industry.” Industry and Innovation 10: 179–96.

BIBLIOGRAPHY 305

Fuller, Douglas B. 2002. “Globalization for Nation-Building: Industrial Policy for HighTechnology Products in Taiwan.” Cambridge, MA: MIT Japan Working Papers. Gallup China Survey. 1995. “The First Nationwide Study of Consumer Attitudes and Lifestyle Trends of the People’s Republic of China.” The Gallup Organization, www.gallup.com Gerber, Jim, and Jorge Carrillo. 2002. “Are Baja California’s Maquiladora Plants Competitive?” Paper prepared for San Diego Dialogue/Forum Fronterizo, The Future of Manufacturing in Baja California: The Electronics and Auto Parts Sectors. Gereffi, Gary. 1994. “The Organization of Buyer-Driven Global Commodity Chains: How United States Retailers Shape Overseas Production Networks.” In Commodity Chains and Global Capitalism, edited by Gary Gereffi and Miguel Korzeniewicz. Westport, CT: Greenwood Press: 95–122. ———. 1996. “Commodity Chains and Regional Divisions of Labor in East Asia.” Journal of Asian Business 12, no. 1. Gereffi, Gary, and Richard P. Applebaum. 1994. “Power and Profits in the Apparel Commodity Chain.” In Global Production: The Apparel Industry in the Pacific Rim, edited by Edna Bonacich et al. Philadelphia: Temple University Press: 42–62. Gereffi, Gary, John Humphrey, and Timothy Sturgeon. Forthcoming. “The Governance of Global Value Chains.” Review of International Political Economy. Gereffi, Gary, David Spencer, and Jennifer Bair, eds. 2002. Free Trade and Uneven Development: The North American Apparel Industry after NAFTA. Philadelphia: Temple University Press. Gerschenkron, Alexander. 1962. Economic Backwardness in Historical Perspective. Cambridge, MA: Belknap Press of Harvard University Press. Gilboy, George J. 2003. “Nodes without Roads: Pockets of Success, Networks of Failure in Chinese Industrial Technology Development.” Ph.D. dissertation, Massachusetts Institute of Technology. Gilpin, Robert, and Jean M. Gilpin. 1987. The Political Economy of International Relations. Princeton, NJ: Princeton University Press. Gold, Thomas B. 1986. State and Society in the Taiwan Miracle. Armonk, NY: M.E. Sharpe. Gourevitch, P. et al. 2000. “Globalization of Production: Insights from the Hard Drive Disk Industry.” World Development 28: 301–17. Government Information Office (Republic of China). 2003. Statistical Yearbook of the Republic of China. www.stat.gov.tw/bs2/2003YearBook.pdf Granovetter, M. 1985. “Economic Action and Social Structure: The Problem of Embeddedness.” American Journal of Sociology 91: 481–510. Gregory, Neil, Stoyan Tenev, and Dileep Wagle. 2000. “China’s Emerging Private Enterprises: Prospects for the New Century.” Washington, DC: International Finance Corporation. Guiheux, Gilles. 1999. “A Survey on the Current Investments of Japanese Auto Parts Industry in Asia: The Case of Toyota First Tier Suppliers.” Actes du GERPISA: International Division of Labor and Assembler-Supplier Relations, no. 25 (February). Guillen, Mauro. 2001. The Limits of Convergence. Princeton, NJ: Princeton University Press.

306 BIBLIOGRAPHY

Guthrie, Doug. 1999. Dragon in a Three-Piece Suit: The Emergence of Capitalism in China. Princeton, NJ: Princeton University Press. Hall, B.H., A.B. Jaffe, and M. Trajtenberg. 2001. “The NBER Patent Citation Data File: Lessons, Insights and Methodological Tools.” Cambridge, MA: National Bureau of Economic Research. Hall, Peter A., and David W. Soskice. 2001. Varieties of Capitalism: The Institutional Foundations of Comparative Advantage. Oxford; New York: Oxford University Press. Hamel, G. 1991. “Competition for Competence and Inter-Partner Learning within International Strategic Alliances.” Strategic Management Journal 12 (summer): 83–104. Hamilton, Gary C. 1997. “Organization and Market Processes in Taiwan’s Capitalist Economy.” In The Economic Organization of East Asian Capitalism, edited by Nicole Woolsey Biggart, Marco Orru, Gary C. Hamilton. Thousand Oaks, CA: Sage: 237–97. Hamilton, Gary, Robert Feenstra, and Misha Petrovic. Forthcoming. Emergent Economies, Divergent Paths: Business Groups and Economic Organization in South Korea and Taiwan. New York: Cambridge University Press. Hatch, Walter, and K.D. Yamamura. 1996. Asia in Japan’s Embrace: Building a Regional Production Alliance. New York: Cambridge University Press. Heide, J.B., and A.S. Miner. 1992. “The Shadow of the Future: Effects of Anticipated Interaction and Frequency of Contact on Buyer-Seller Cooperation.” Academy of Management Journal 35, no. 2: 265–91. Helper, S., J.P. MacDuffie, and C. Sabel. 2000. “Pragmatic Collaboration: Advancing Knowledge While Controlling Opportunism.” Industrial and Corporate Change 9, no. 3: 443–88. Hemmert, M. 1999. “‘Intermediate Organization’ Revisited: A Framework for the Vertical Division of Labor in Manufacturing and the Case of Japanese Assembly Industries.” Industrial and Corporate Change 8, no. 3: 487–517. Herrigel, Gary. 1994. “Industry as a Form of Order: A Comparison of the Historical Development of the Machine Tool Industry in the United States and Germany.” In Governing Capitalist Economies: Performance and Control of Economic Sectors, edited by P. Schmitter, W. Streeck, and J.R. Hollingsworth. New York: Oxford University Press: 97–128. ———. 1996. Industrial Constructions: The Sources of German Industrial Power. Cambridge: Cambridge University Press. Herrigel, Gary, and Volker Wittke. Forthcoming. “Varieties of Vertical Disintegration: The Global Trend toward Heterogeneous Supply Relations and the Reproduction of Difference in US and German Manufacturing.” In Changing Capitalisms: Internationalisation, Institutional Change and Systems of Economic Organization, edited by Eli Moen, Richard Whitley, and Glenn Morgan. Oxford: Oxford University Press. Hille, K. 2003. “Intel Deal Puts SiS under Cloud.” Financial Times, April 9, 31. ——— 2004. Taiwan Aims to Ease Mainland Investment Rules. Financial Times, July 28, p. 2. Hitt, M.A., B.W. Keats, and S.M. DeMarie. 1999. “Navigating in the New Competitive Landscape: Building Strategic Flexibility and Competitive Advantages in the 21st Century.” Academy of Management Executive 12, no. 4: 22–42.

BIBLIOGRAPHY 307

Hollingsworth, J.R., P. Schmitter, and W. Streeck, eds. 1994. Governing Capitalist Economies: Performance and Control of Economic Sectors. New York: Oxford University Press. Hong Kong Consumer Council. 1996a. “Competition Policy: The Key to Hong Kong’s Future Economic Success.” http://www.consumer.org.hk/website/ws_en/ competition_issues/competition_studies/1996fairtrade.html ———. 1996b. “How Competitive Is the Private Residential Property Market?” http:// www.consumer.org.hk/website/wrap_en2/hse9607/hse_e.htm/ Hong, Sung Gul. 1997. The Political Economy of Industrial Policy in East Asia: The Semiconductor Industry in Taiwan and South Korea. Cheltenham, UK; Northampton, MA: Edward Elgar. Howell, Thomas R., Brent L. Bartlett, William A. Noellert, and Rachel Howe. 2003.”China’s Emerging Semiconductor Industry.” Washington, DC: Dewey Ballantine LLP. Hsu, Jinn-yuh. 1997. “A Late-Industrial District? Learning Network in the Hsinchu Science-Based Industrial Park.” Ph.D. dissertation, University of California at Berkeley. Hsu, Jinn-yuh, and AnnaLee Saxenian. 2000. “The Limits of Guanxi Capitalism: Transnational Collaboration between Taiwan and the US.” Environment and Planning 32, no. 11: 1991–2005. Hsueh,L., C. Kuo, and D. Perkins, eds. 2000. Industrialization and the State: The Changing Role of the Taiwan Government in the Economy, 1945–1998. Cambridge, MA: Harvard University Press. Hu, Albert, and Adam B. Jaffe. 2001. “Patent Citations and International Knowledge Flow: The Cases of Korea and Taiwan.” Cambridge, MA: National Bureau of Economic Research. Hu, Angang. 1999. “The Greatest Challenge of the New Century: China Enters the Stage of High Unemployment” (Kuaru xinshiji de zui da tiaozhan: Wo guo jinru gaoshiye jieduan). Chinese Academy of Sciences–Tsinghua University Joint Center for Chinese Studies. Hu, Angang and Shaoguang Wang. 1999. The Political Economy of Uneven Development: The Case of China. Armonk, NY: M.E. Sharpe. Hu, Ming-wen, and Chi Schive. 1998. “The Changing Competitiveness of Taiwan’s Manufacturing Smes.” Small Business Economics 11: 315–26. Huang, Yasheng. 2003. Selling China: Foreign Direct Investment During the Reform Era. New York: Cambridge University Press. Humphrey, John, and Olga Memedovic. 2003. The Global Automotive Industry Value Chain. Vienna: United Nations Industrial Development Organization. Hung, F. 2002. “Taiwan’s Chipset Sales Plunge.” Financial Times Information, July 10. Hwang, Chin-yeong. 1995. Taiwan—The Republic of Computers. Taipei, Taiwan: Commonwealth Publishing Co. IAEI. 2002. Israel Electronics Industries Profile. Tel Aviv: Israel Association of Electronic and Information Industries. Available from www.iaei.org.il/indexsite.html IASH. 2002. Software Industry Statistical Information. Tel Aviv: Israeli Association of Software Houses. Available from www.iash.org.il/content/SoftwareInds/ StatisticalInformation.asp IC Insights. Website accessed at http://www.icinsights.com

308 BIBLIOGRAPHY

III. Various years. Zixun Gongye Nianjian (Yearbook of Information Industry). Taipei: Institute for Information Industry, Market Intelligence Center. IIY. 2003. Zixun Gongye Nianjian (Information Industry Yearbook 2003). Taipei: Market Intelligence Center. IMD. 1996. The World Competitiveness Yearbook. Lausanne: IMD. “The Information Technology 100.” 1999. Business Week, June 21. Information Technology Information Service (ITIS). Various years. Bandaoti Gongye Nianjian (The Semiconductor Industry Yearbook). Taipei: Taiwan Ministry of Economic Affairs. ———. Various years. “Industrial Statistics.” Hsinchu: Industrial Technology Research Institute—Industrial Economic & Knowledge Center. Institute of Industrial Economics. 1998. Zhongguo gongye fazhan baogao (China Industrial Development Report). Beijing: Jingji Guanli Chubanshe. International Data Corporation. 2003a. “Notebooks Record Highest Growth Ever, Boosting EMEA PC Market by 18.7% in 3Q03, IDC Says.” Press Release, October 16. ———. 2003b. “PC Shipments To Reach Record Levels in 2003 and 2004, According to IDC.” Press Release, December 11. ———. 2004. “Worldwide Quarterly PC Tracker Report,” Framingham, Massachusetts, March. Ives, Nat. 2004. “Toyota Becomes the First Exclusive Auto Sponsor on eBay.” New York Times, February 19. Jarillo, J. 1988. “On Strategic Networks.” Strategic Management Journal 9: 31–41. Johanson, J., and L. Mattsson. 1987. “Interorganizational Relations in Industrial Systems: A Network Approach Compared with the Transaction-Cost Approach.” International Studies of Management and Organization 17, no. 1: 34–48. Johnson, A. 1982. MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925–1975. Palo Alto, CA: Stanford University Press. Jones, R.S. 2002. “Economic Restructuring and Institutional Changes in Korea in Wake of the 1997 Crisis.” Paper presented at the Korea Development Institute, February 21. Just-Style. 2004. “Mexico: An Action Plan for Yucatan,” April 26, Just-Style.com Kao, Cheng-shu. 1991. “‘Personal Trust’ in the Large Businesses in Taiwan: A Traditional Foundation for Contemporary Economic Activities.” In Business Networks and Economic Development in East and Southeast Asia, edited by Gary Hamilton. Hong Kong: Centre of Asian Studies, University of Hong Kong: 59–70. Kaplinsky, Raphael. 2004. “How Does It All Add Up? Caught between a Rock and a Hard Place,” at www.soc.duke.edu/sloan_2004/Papers/Kaplinsky-Chapter% 206_20May04.pdf Katz, M.L., and C. Shapiro. 1985. “Network Externalities, Competition, and Compatibility.” American Economic Review 75: 424–40. ———. 1986. “Technology Adoption in the Presence of Network Externalities.” Journal of Political Economy 94: 822–41. Katzenstein, Peter J. 2003. “Regional States: Japan and Asia, Germany in Europe.” In The End of Diversity? Prospects for Germany and Japanese Capitalism, edited by K. Yamamura and W. Streeck. Ithaca, NY: Cornell University Press: 89–115.

BIBLIOGRAPHY 309

Kawakami, Momoko. 1996. Development of Small- and-Medium-Sized Manufacturers in Taiwan’s PC Industry. Taipei, Taiwan: Chung-hua Institute for Economic Research. Keller, William, and Richard J. Samuels. 2003. Crisis and Innovation in Asian Technology. New York: Cambridge University Press. Khanna, T., R. Gulati, and N. Nohria. 1998. “The Dynamics of Learning Alliances: Competition, Cooperation, and Relative Scope.” Strategic Management Journal 19, no. 3: 193–210. Kitschelt, H. 1991. “Industrial Governance Structures, Innovation Strategies, and the Case of Japan: Sectoral or Cross-National Comparative-Analysis.” International Organization 40: 453–93. Kornai, Janos. 2000. “What the Change of System from Socialism to Capitalism Does and Does Not Mean.” Journal of Economic Perspectives 14, no. 1: 27–42. Kuo, Chengtian. 2000. “Globalization and the Adjustment of Financial Institutions: Taiwan, Thailand, and Malaysia.” Paper presented at the Workshop on Networked Production and Globalization, Academia Sinica, Taipei, Taiwan. Kynge, James. 2002. “Creaking Economy Needs Stronger Foundations.” Financial Times, October 29. La Croix, Sumner J., Michael Plummer, and Keun Lee, eds. 1995. Emerging Patterns of East Asian Investment in China. Armonk, NY: M.E. Sharpe. Lall, Sanjaya. 1998. “Meeting the Human Capital Needs of Maturing Asian Economies.” In The Future of Asia in the World Economy, edited by Francis Harrigan, Colm Foy, and David O’Connor. Paris: OECD: 149–93. Langlois, Richard. 2003. “The Vanishing Hand: The Changing Dynamics of Industrial Capitalism.” Industrial and Corporate Change 12, no. 2: 351–85. Lardy, Nicholas R. 1998. China’s Unfinished Economic Revolution. Washington, DC: Brookings Institution. ———. 2002. Integrating China into the Global Economy. Washington, DC: Brookings Institution. Le Monde. 2003. “Solectron Prepares a Fifth Wave of Layoffs,” Paris, France, July 12, p. 15. Lee, Jane Lanhee. 2002. “China’s Price Wars Build Pressure on State Companies to Reform.”Dow Jones International News, April 17. Lee, J.R., and J.S. Chen. 2000. “Dynamic Synergy Creation with Multiple Business Activities: Toward a Competence-Based Business Model for Contract Manufacturers.” In Advances in Applied Business Strategy, Vol. 6A, edited by Ron Sanchez and Aime Heene. Stamford, CT: JAI Press: 209–28. Lee, J.R., J.S.Chen, and M.J.Tang. Forthcoming. “Exploring the Growth Strategy of Contract Electronics Manufacturers in Taiwan: A Competence-based Perspective.” In Advances in Applied Business Strategy, Volume 7A (Competence Perspective on Managing Internal Processes), Stamford, CT: JAI Press, Inc. Leggett, Karby. 2001. “Konka’s Loss Shows Effects of Price Wars.” Asian Wall Street Journal, August 30. Leggett, Karby, and Todd Zaun. 2001. “Road Warriors: Motorcycle Makers from Japan Discover Piracy Made in China.” Wall Street Journal, July 25. Lei, D. 1997. “Competence-Building, Technology Fusion and Competitive Advantage: The Key Roles of Organizational Learning and Strategic Alliances.” International Journal of Technology Management 14: 208–37.

310 BIBLIOGRAPHY

Lei, D., and J. Slocum, Jr. 1992. “Global Strategy, Competence-Building and Strategic Alliances.” California Management Review 35, no. 1: 81–97. Lemoine, Francoise, Alix de Saint Vaulry, and Mouhamadou Dramé. 1994. “Hong Kong-Chine: Un Dragon à Deux Têtes.” Economie Internationale, no. 57: 63–98. Leonard-Barton, D. 1995. Wellsprings of Knowledge: Building and Sustaining the Sources of Innovation. Boston: Harvard Business School Press. Lepedus, Mark. 2002. “How TSMC Is Expanding Itself to Serve up 90Nm Chips.” Semiconductor Business News (www.siliconstrategies.com)[cited April 15, 2002]. Levinthal, D.A., and J. March. 1993. “The Myopia of Learning.” Strategic Management Journal 14 (special issue): 95–112. Levinthal, D.A., and J. Myatt. 1994. “Co-Evolution of Capabilities and Industry: The Evolution of Mutual Fund Processing.” Strategic Management Journal 15: 45–62. Levitt, B., and J. March. 1988. “Organizational Learning.” Annual Review of Sociology 14: 319–340. Levy, David L. 1997. “Lean Production in an International Supply Chain.” Sloan Management Review (Winter). Lewin, A.Y., C.P. Long, and T.N. Carroll. 1999. “The Co-Evolution of New Organizational Forms.” Organization Science 10, no. 5: 535–50. Lewin, A.Y., and H.W. Volberda. 1999. “Prolegomena on Co-Evolution: A Framework for Research on Strategy and New Organizational Forms.” Organization Science 10, no. 5: 519–34. Li, B. 2003. “Acer Laptop Unit Faces Tough IPO.” Reuters, July 11. Li, Francis. 2004. “Auto Giants Turn to Taiwan for Parts.” Taiwan Journal, March 12. Liang, Quincy. 1999. “Parts Exports by Taiwan Automakers Reach Record High.” China Economic News Service, Taiwan Economic News, November. ———. 2003a. “China Motor Exports Mitsubishi Vans To Philippines.” China Economic News Service, Taiwan Economic News, May, www.cens.com ———. 2003b. “Government Moves to Boost Automakers’ Parts Exports.” China Economic News Service, Taiwan Economic News, September. Lin, Zhiling, and Thomas W. Robinson, eds. 1994. The Chinese and Their Future: Beijing, Taipei, and Hong Kong. Washington, DC: AEI Press. Lindquist, Diane. 2000. “Chips and Salsa.” Industry Week, November 6. Lorenz, E. 1988. “Neither Friends nor Strangers: Informal Networks of Subcontracting in French Industry.” In Making and Breaking Cooperative Relations, edited by Diego Gambetta. Oxford; New York: Basil Blackwell: 183–92. ———. 1992. “The Search for Flexibility: Subcontracting Networks in British and French Engineering.” In Pathways to Industrialization and Regional Development, edited by M. Storper and A.J. Scott. London; New York: Routledge: 122–32. Lowe, Nichola, and Martin Kenney. 1999. “Foreign Investment and the Global Geography of Production: Why the Mexican Consumer Electronics Industry Failed.” World Development 27, no. 8: 1428. Lu, Qiwen. 2000. China’s Leap into the Information Age: Innovation and Organization in the Computer Industry. Oxford; New York: Oxford University Press. Lubman, Stanley B. 1999. Bird in a Cage: Legal Reform in China after Mao. Palo Alto, CA: Stanford University Press. Luhnow, David. 2004. “Up the Food Chain: As Jobs Move East, Plants in Mexico Retool to Compete.” Wall Street Journal, March 5.

BIBLIOGRAPHY 311

Lui, T.L., and S. Chiu. 1994. “A Tale of Two Industries: The Restructuring of Hong Kong’s Garment-Making and Electronics Industries.” Environment and Planning 26: 53–70. Lundvall, B.-Å. 1992. National Systems of Innovation: Towards a Theory of Innovation and Interactive Learning. London: Pinter Publishers. Lüthje, Boy. 2002. “Electronics Contract Manufacturing: Global Production and the International Division of Labor in the Age of the Internet.” Industry and Innovation 9, no. 3: 227–47. Lynch, Teresa M. 2000. “Leveraging Capabilities: Models of Foreign Production in the Taiwanese Automotive Industry.” Cambridge, MA: MIT IPC Working Paper. Ma, Guonan and Ben S.C. Fung. 2002. “China’s Asset Management Corporations.” Bank for International Settlements Working Paper No. 115, August. Macher, Jeffrey T., David C. Mowery, and David A. Hodges. 1999. “Semiconductors.” In U.S. Industry in 2000: Studies in Competitive Performance, edited by David C. Mowery. Washington, DC: The National Academy Press: 245–86. MacIntyre, Andrew, ed. 1994. Business and Government in Industrializing Asia. Ithaca, NY: Cornell University Press. Mackenzie, Davin. 2001. “A Healthy Financial Sector Requires Enterprises That Deserve Financing.” Paper presented at the Conference on Financial Sector Reform in China, Harvard University, September. Mackintosh, James. 2003. “Block Exemption: EU Rules Will Mean Big Changes.” Financial Times, March 3. Maclellan, Andrew. 2003. “Mexico Hangs Tough as Alternative to China.” Electronic Buyer’s News, September 15. Madhok, A. 1996. “The Organization of Economic Activity: Transaction Costs, Firm Capabilities and the Nature of Governance.” Organization Science 7, no. 5: 577–90. ———. 2000. “Inter-Firm Collaboration: Contractual and Capability-Based Perspectives.” In Competence, Governance, and Entrepreneurship, edited by N. Foss and V. Mahnke. New York: Oxford University Press: 276–303. Madhok, A., and S.B. Tallman. 1998. “Resource, Transactions and Rents: Managing Value through Inter-Firm Collaborative Relationships.” Organizational Science 9: 326–39. Makides, C., and N. Berg. 1988. “Manufacturing Offshore Is Bad Business.” Harvard Business Review 66, no. 4. Malkin, Elisabeth. 2002. “Manufacturing Jobs Are Exiting Mexico.” New York Times, November 5. Market Information Center. 2002. 2002 IT Industry Yearbook. Taipei, Taiwan, July. ———. 2003. “Report on the Competitiveness of Taiwan IT Industries.” Taipei, Taiwan, April. Mathews, John A. 1995. High-Technology Industrialization in East Asia: The Case of the Semiconductor Industry in Taiwan and Korea. Taiwan: Chung-hua Institute for Economic Research. ———. 2002. “The Origins and Dynamics of Taiwan’s R&D Consortia.” Research Policy 31: 633–51. Mathews, John A. and Dong-Sung Cho. 2000. Tiger Technology: The Creation of a Semiconductor Industry in East Asia. Cambridge, UK; New York: Cambridge University Press.

312 BIBLIOGRAPHY

McKelvey, B. 1997. “Quasi-Natural Organization Science.” Organization Science 8: 352–80. McKendrick, David G., Richard F. Doner, and Stephan Haggard. 2000. From Silicon Valley to Singapore: Location and Competitive Advantage in the Hard Disk Drive Industry. Palo Alto, CA: Stanford University Press. McMillan, J. 1990. “Managing Suppliers: Incentive Systems in Japanese and United States Industry.” California Management Review (summer): 38–55. McNamara, Michael. 1993. “Getting Guilford Going.” Women’s Wear Daily, January 19. Meaney, Constance Squires. 1994. “State Policy and the Development of Taiwan’s Semiconductor Industry.” In The Role of the Sate in Taiwan’s Development, edited by David Dollar, Kenneth Sokoloff and Joel Aberbach. Armonk, NY: M.E. Sharpe. Meyer, M.H., and J.M. Utterback. 1993. “The Product Family and the Dynamics of Core Capability.” Sloan Management Review 34, no. 3: 29–47. MoEA. 2003 and various years. “Economic Statistics Annual, Taiwan Area, Republic of China.” Taipei: Ministry of Economic Affairs. Moises, Evangeline. 2000. “Enterprisers: Getting the Pulse of the Customer.” Business World, September 14, 8 [cited 2003]. Moody, John. 2002. “Firms Flee Rising Labor Costs in Mexico.” Montreal Gazette, April 2. Morawetz, David. 1981. Why the Emperor’s New Clothes Are Not Made in Colombia. New York: Oxford University Press. Mowery, D. 1996. The International Computer Software Industry: A Comparative Study of Industry Evolution and Structure. New York: Oxford University Press. Naughton, Barry. 1996. “The United States and China: Management of Economic Conflict.” Paper presented at the Conference on Domestic Factors in U.S.–China Relations, Fairbank Center, Harvard University. ———. 1997. The China Circle: Economics and Electronics in the PRC, Taiwan, and Hong Kong. Washington, DC: Brookings Institution Press. ———. 1999. “How Much Can Regional Integration Do to Unify China’s Markets?” Paper presented for Conference on Policy Reform in China, Stanford University, November. Nelson, R.R. 1993. National Innovation Systems: A Comparative Analysis. New York: Oxford University Press. Nelson, R.R., and S.G. Winter. 1982. An Evolutionary Theory of Economic Change. Cambridge, MA: Belknap. NID. 2002. Survey of the Irish Software Industry. National Informatics Directorate. Available from www.nsd.ie/htm/ssii/stat.htm Nishiguchi, T. 1994. Strategic Industrial Sourcing: The Japanese Advantage. New York: Oxford University Press. ———. 2001. “Coevolution of Interorganizational Relations.” In Knowledge Emergence: Social, Technical, and Evolutionary Dimensions of Knowledge Creation, edited by I. Nonaka and T. Nishiguchi. New York: Oxford University Press: 197– 222. Noble, Gregory W. 1996. “Trojan Horse or Boomerang: Two-Tiered Investment in the Asian Auto Complex.” Berkeley, CA: Berkeley Roundtable on the International Economy, November 1.

BIBLIOGRAPHY 313

———. 1998. Collective Action in East Asia: How Ruling Parties Shape Industrial Policy. Ithaca, NY: Cornell University Press. ———. 2000. “Conspicuous Failures and Hidden Strengths of the ITRI Model: Taiwan’s Technology Policy Toward Hard Disk Drives and CD-ROMs.Report 2000– 2002.” University of California, San Diego, Information Storage Industry Center. Nolan, Peter. 2001. China and the Global Economy: National Champions, Industrial Policy, and the Big Business Revolution. New York: Palgrave. North, Douglas. 1990. Institutions, Institutional Change and Economic Performance. New York: Cambridge University Press. O’Riain, S. 2000. “The Flexible Development State: Globalization, Information Technology, and the ‘Celtic Tiger.’” Politics and Society 28: 157–93. ———. 2004. The Politics of High Tech Growth: Developmental Network States in the Global Economy. Cambridge, UK: Cambridge University Press. OECD. 2001. “Corporate Governance in Asia: A Comparative Perspective.” Paris: OECD. Ohmae, Kenichi. 1990. The Borderless World: Power and Strategy in the Interlinked Economy. New York: HarperBusiness. Oi, Jean C. 1999. Rural China Takes Off. Berkeley, CA: University of California Press. Olenick, Doug. 2002. “Dell Computer Holds Top PC Spot.” Reed Business Information US, October 28, 12. Organisation Internationale des Constructeurs d’Automobiles, http://www.oica.net/ htdocs/Main.htm Pan, Mei-Lin. 2000. “Restructing Taiwan-Based Production Networks: The Apparel Industry.” Paper presented at the Workshop of Networked Production and Globalization, Academia Sinica. Park, Phillip Hoon. 2000. “A Reflection on the East Asian Development Model: Comparison of the South Korean and Taiwanese Experiences.” In The East Asian Development Model: Economic Growth, Institutional Failure and the Aftermath of the Crisis, edited by F. Richter. London: Macmillan Press: 141–68. Patrick, Hugh, with Larry Meissner, eds. 1991. Pacific Basin Industries in Distress. New York: Columbia University Press. Peerenboom, Randall. 2001 “Seek Truth From Facts: An Empirical Study of Enforcement of Arbitral Awards in the PRC.” (www.law.ucla.edu/erg/pubs/ arbitral_awards.pdf). ———. 2002. “Let One Hundred Flowers Bloom, One Hundred Schools Contend: Debating Rule of Law in China.” Michigan Journal of International Law 23, no. 3: 471–544. Perkins, Dwight H. 2001. “Industrial and Financial Policy in China and Vietnam.” In Rethinking the East Asian Miracle, edited by Joseph E. Stiglitz and Shahid Yusuf. Washington, DC: World Bank: 247–95. Pin, Jean-Louis. 1999. L’ouverture économique de la Chine (1978–1999): Au Profit de qui? Paris: La Documentation Française. Piore, Michael J., and Charles F. Sabel. 1984. The Second Industrial Divide: Possibilities for Prosperity. New York: Basic Books. Pomfret, John. 2001. “Legacy of Socialism Keeps China’s State Firms in Red.” Washington Post, June 20. Porter, Michael E. 1980. Competitive Strategy: Techniques for Analyzing Industries and Competitors. New York: Free Press.

314 BIBLIOGRAPHY

———. 1990. The Competitive Advantage of Nations. New York: Free Press. ———. 1998. “Clusters and the New Economics of Competition.” Harvard Business Review, November–December, 77–90. ———. 2001. “The Competitive Advantage of Taiwan.” Paper presented at the Commonwealth Speech, Taipei, Taiwan, July 31. Powell, W.W. 1987. “Hybrid Organizational Arrangements: New Form or Transitional Development?” California Management Review (fall): 67–87. ———. 1990. “Neither Market nor Hierarchy: Network Forms of Organization.” Research in Organizational Behavior 12: 295–336. Powell, W.W., and P. DiMaggio. 1991. The New Institutionalism in Organizational Analysis. Chicago: The University of Chicago Press. Prahalad, C.K., and G. Hamel. 1990. “The Core Competence of the Corporation.” Harvard Business Review: 79–91. Quinn, J.B. 1992. Intelligent Enterprise. New York: Free Press. ———. 1999. “Strategic Outsourcing: Leveraging Knowledge Capabilities.” Sloan Management Review (summer): 9–21. Quinn, J.B., and F.G. Hilmer. 1994. “Strategic Outsourcing.” Sloan Management Review 35 (summer): 43–55. Ravenhill, Mitchell, and John Bernard. 1995. “Beyond Product Cycles and Flying Geese: Regionalization, Hierarchy, and the Industrialization of East Asia.” International Organization 47: 172–83. Rawski, Thomas G. 2003. “Recent Developments in China’s Labour Economy.” Report prepared for the International Labour Office in January 2002, www.soc.duke.edu/sloan_2004/ Papers/Rawski_China_papertext_ Nov03.pdf Richardson, G. 1972. “The Organization of Industry.” The Economic Journal 84: 883–96. Richardson, J. 1993. “Parallel Sourcing and Supplier Performance in the Japanese Automobile Industry.” Strategic Management Journal 14: 339–50. Ristelhueber, R. 2003. “Taiwan Breaks Into Top 10 Chip Ranking.” Electronics Business News, January 20. Accessed at http://www.ebnonline.com, p. 8. Robertson, Jack. 1999. “North America Gaining Rep as Pc Assembly Site.” Electronic Buyer’s News, August 9. Rodrik, Dani. 1994. “Getting Interventions Right: How South Korea and Taiwan Grew Rich.” NBER Working Paper 4964. Cambridge, MA: National Bureau of Economic Research, www.nber.org/papers/w4964 Rogers, Richard T. 1998. Sunkist Case Study: A Discussion. Department of Resource Economics University of Massachusetts–Amherst. Available at www.ag.uiuc.edu/famc/program98/rogers.htm Rosen, Daniel H. 2003. “Low-tech Bed, High-tech Dreams.” China Economic Quarterly, Q4, pp. 20–27 (http://www.theceq.info/ceq2003q4_sec2_technology.pdf). Roy, Denny. 2003. Taiwan. A Political History. Ithaca, NY: Cornell University Press. Samsung Electronics. 2000. “Annual Report” (www.samsungelectronics.com). Sanchez, R. 1996. “Strategic Product Creation: Managing New Interactions of Technology, Markets, and Organizations.” European Management Journal 14, no. 2: 121–38. Sanchez, R., A. Heene, and H. Thomas. 1996. “Introduction: Towards the Theory and Practice of Competence-Based Competition.” In Dynamics of Competence-

BIBLIOGRAPHY 315

Based Competition: Theory and Practice in the New Strategic Management, edited by A. Heene, R. Sanchez, and H. Thomas. Oxford: Pergamon, Elsevier Science Ltd.:1–35. Saxenian, AnnaLee. 2001. “Taiwan’s Hsinchu Region: Imitator and Partner for Silicon Valley.” SIEPR Policy Paper No. 00–44. Palo Alto, CA: Stanford Institute for Economic Policy Research. http://siepr.stanford.edu/papers/pdf/00– 44.html. Saxenian, AnnaLee, and Jinn-yuh Hsu. 2001. “The Silicon Valley-Hsinchu Connection: Technical Communities and Industrial Upgrading.” Industrial and Corporate Change 10, no. 4: 893–920. Schilling, M.A., and K. Steensma. 2001. “The Use of Modular Organizational Forms: An Industry Level Analysis.” Academy of Management Journal 44, no. 6: 1149–68. Schive, Chi. 1990. The Foreign Factor: The Multinational Corporation’s Contribution to the Economic Modernization of the Republic of China. Stanford, CA: Hoover Institution Press. ———. 1999a. “How Did SMEs in Taiwan Survive the Crisis?” Paper presented at the 1999 Industry Economics Conference, Melbourne, Australia. ———. 1999b. “A Study on Taiwan: High-Tech Industries in the Spotlight.” Paper presented at the Geneva-Hong Kong Conference on Global Production: Specialization and Trade, Hong Kong. Schumpeter, Joseph Alois. 1968. The Theory of Economic Development: An Inquiry into Profits, Capital, Credit, Interest and Business Cycles. Cambridge: Harvard University Press, [1911]. ———. 1976. Capitalism, Socialism, and Democracy. London: Harper & Brothers [1942]. Scitovsky, Tibor. 1986. “Taiwan and South Korea, 1965–1981.” In Models of Development: A Comparative Study of Economic Growth in South Korea and Taiwan, edited by L. Lau. San Francisco: ICS Press: 145–65. Secretaría de Economía. 2002. “Programa para la Competitividad de la Industria Electrónica y de Alta Tecnología,” Mexican Government. Semiconductor Business News (SBN).http://www.siliconstrategies.com Serant, Claire. 2002a. “Vtech’s Contract Division Learns There’s No Place Like Home.” Electronic Buyer’s News, November 25. ———. 2002b. “Flextronics Opens Tech Center in Mexico.” Electronic Buyer’s News, December 10. ———. 2003. “Mexico Spins a New Orbit.” Electronic Buyer’s News, January 21. Shapiro, Don, and Gary Heath. 2002. “The Drive to Survive.” Topics, November. Shieh, G.S. 1992. Boss Island: The Subcontracting Network and MicroEntrepreneurship in Taiwan’s Development. New York: Peter Lang. Smith, Geri. 2003. “Wasting Away: Despite SARS, Mexico Is Still Losing Export Ground to China.” BusinessWeek, June 2. Solectron. 2004. Form 10Q, filed January 12 for the period ended November 28, 2003. Accessed via http://www.solectron.com Sprague, Jonathan. 2002. “China’s Manufacturing Beachhead.” Fortune, October 28. Stanford Case Study. 2001. “Solectron: From Contract Manufacturer to Global Supply Chain Integrator,” Case No. GS-24, Graduate School of Business, Stanford University.

316 BIBLIOGRAPHY

State Statistical Bureau. Various years. Statistical Yearbook of China. Beijing: China Statistical Publishing House. ———. Various years. China Foreign Economic Statistical Yearbook. Beijing: China Statistical Publishing House. Steinfeld, Edward S. 1998. Forging Reform in China: The Fate of State-Owned Industry. New York: Cambridge University Press. ———. 2000. “Free Lunch or Last Supper: China’s Debt-Equity Swaps in Context.” China Business Review, July–August. ———. 2002. “Moving Beyond Transition in China: Financial Reform and the Political Economy of Declining Growth.” Comparative Politics 34, no. 4: 379–98. ———. 2004. “Market Visions: The Interplay of Ideas and Institutions in Chinese Financial Restructuring,” Political Studies 52, no. 4, December, 643–63. Stevens, Clint. 2002. “Domestic Chinese Appliance OEMs on the Ropes.” Appliance, June 1. Streeck, W. 1991. “On the Institutional Conditions of Diversified Quality Production.” In Beyond Keynesianism: The Socio-Economics of Production and Full Employment, edited by E. Matzner and W. Streeck. Brookfield, VT: Edward Elgar: 21–61. ———. 1996. “Lean Production in the German Automobile Industry: A Test Case for Convergence Theory.” In National Diversity and Global Capitalism, edited by S. Berger and R. Dore. Ithaca, NY: Cornell University Press: 138–71. Sturgeon, Timothy J. 1997. “Turnkey Production Networks: A New American Model of Industrial Organization?” BRIE Working Paper 92a. Berkeley, CA: Berkeley Roundtable on International Economy. ———. 1998. “Network-Led Development and the Rise of Turn-Key Production Networks: Technological Change and the Outsourcing of Electronics Manufacturing.” Cambridge, MA: Industrial Performance Center Working Paper, MIT. ———. 1999. “Turn-Key Production Networks: Industrial Organization, Economic Development, and the Globalization of the Electronics Manufacturing SupplyBase.” Ph.D. Dissertation, University of California at Berkeley. ———. 2000. “Turn-Key Production Networks: The Organizational De-Linking of Production from Innovation.” In New Product Development and Production Network, edited by U. Jürgens. Berlin: Springer-Verlag: 67–84. ———. 2002. “Modular Production Networks: A New American Model of Industrial Organization.” Industrial and Corporate Change 11, no. 3: 451–96. Sturgeon, Timothy J., and Ji-Ren Lee. 2001. “Industry Co-Evolution and the Rise of a Shared Supply-base for Electronics Manufacturing.” Paper presented at the Nelson and Winter Conference, Aalborg, Denmark, June 12–15. Sturgeon, T.J., and Richard K. Lester. 2004. “The New Global Supply-Base: New Challenges for Local Suppliers in East Asia.” In Global Production Networking and Technological Change in East Asia, edited by Anjum Altaf, Kaoru Nabeshima, and Shahid Yusuf. Oxford: Oxford University Press: 35–89. Su, Paolo S. 2003. “The Taiwan ITC Industry Outlook,” Market Intelligence Center, Institute for Information Industry, Taipei, Taiwan. Accessed January 2004 at http:// mic.iii.org.tw/english/ Sung, Yun-Wing, Liu, Pak-Wai, Wong, Yue-Chim Richard, and Lau, Pui-King. 1995. The Fifth Dragon: The Emergence of the Pearl River Delta. Singapore: Addison Wesley.

BIBLIOGRAPHY 317

Tabata, Mayumi, and Dung-sheng Chen. 2000. “Inter-Organizational Co-Operation between Taiwanese IC Companies and Japanese Companies—Trust Formation and Power Dynamic.” Paper presented at the Workshop on Networked Production and Globalization, Acadamia Sinica, Taipei. Taipei Times. 2002. “Honda to Sever Ties with Sanyang.” January 11, p. 18. Taiwan Economic News. 2003. “Hong Ho to Set Up Textile Plants in Mainland China.” July 18. Taiwan Semiconductor Industry Association. 2004. “Ratio of TSIA Members to the total IC Companies [sic].” Accessed at http://www.tsia.org.tw/tsia_En/about/ MemberAssay3.htm on October 20. Taiwan Transportation Vehicle Manufacturers Association (TTVMA). 2000. www.ttvma.org/statistic.htm ———. 2003. Annual Report of Taiwan’s Automotive Industry, 1999. Technology Forecasters, 2004. Interview with Eric Miscoll, October 7. Teece, D.J., G. Pisano, and A. Shuen. 1997. “Dynamic Capabilities and Strategic Management.” Strategic Management Journal 18, no. 7: 509–33. Teles, Steven M. 1996. “Public Opinion and Interest Groups in the Making of United States China Policy.” Paper presented at the Conference on the Domestic Sources of U.S.–China Relations, Fairbank Center, Harvard University. Tenev, Stoyan and Chulin Zhang. 2002 Corporate Governance and Enterprise Reform in China. Washington, DC: World Bank/IFC. Thorelli, H. 1986. “Networks: Between Markets and Hierarchies.” Strategic Management Journal 7: 37–51. Thun, Eric. Forthcoming. Changing Lanes in China: Foreign Direct Investment, Local Governments, and Auto Sector Development. Thun, Eric, and Adam Segal. 2001. “Thinking Globally and Acting Locally: Local Governments, Industrial Sectors, and Development in China.” Politics & Society 29, no. 4: 557–88. Tianxia (Commonwealth), July 1987, August 2000, and May 1, 2003. ———. 2001. “Top 1000 Manufacturers in Taiwan,” May 25, 2001, 91–192. Toh, D. 2003 “Flat Panel Displays,” Global Equity Research Asia, Lehman Brothers, May 30. Trajtenberg, M. 2001. “Innovation in Israel 1968–1997: A Comparative Analysis Using Patent Data.” Research Policy 30: 363–89. Treece, James. 2003. “Strait Bet: Taiwan Banks on China.” Automotive News, February 17, p. 4. Tsai, Kellee S. 2002. Back-Alley Banking: Private Entrepreneurs in China. Ithaca, NY: Cornell University Press. TSIA (Taiwan Semiconductor Industry Association). Various years. “Annual Report.” Available at www.tsia.org.tw Tung, Chen-yuan. 2002. “China’s Economic Leverage and Taiwan’s Security Concerns with Respect to Cross-Strait Economic Relations.” Ph.D. dissertation, Johns Hopkins University, Baltimore, Maryland. Tzeng, D. 2002. “Moves by Some Taiwan Notebook PC Contract Makers To Cut MVA Prices May Hurt the Industry.” Digitimes, May 30. http://www.digitimes.com Tzeng, David, and Sarah Chang. 2003. “Quanta to Become Acer’s Biggest Notebook Supplier.” Digitimes, April 4. http://www.digitimes.com Ulvinen, T.L. 2000. “Are You Ready for Contract Manufacturing?” http:// ieeexplore.ieee.org/xpl/tocresult.jsp?isNumber=17861

318 BIBLIOGRAPHY

U.S. Bureau of Labor Statistics. 2002. http://www.bls.gov United States General Accounting Office. 2001. “Motor Vehicle Safety.” GAO 01–225, http://www.gao.gov/cgi-bin/getrpt?GAO-01–225. Veloso, Francisco, Jorge Soto Romero, and Alice H. Amsden. 1998. “A Comparative Assessment of the Development of the Auto Parts Industry in Taiwan and Mexico: Policy Implications for Thailand.” MIT Working Paper. Cambridge MA: Massachusetts Institute of Technology, IPC, March. Venkatesan, R. 1992. “Strategic Outsourcing: To Make or Not to Make.” Harvard Business Review, November–December, 1–11. Vernon, Raymond. 1966. “International Investments and International Trade in the Product Cycle.” Quarterly Journal of Economics 80, no. 2: 190–207. Vogel, Ezra F. 1991. The Four Little Dragons. The Spread of Industrialization in East Asia. Cambridge, MA: Harvard University Press. Wade, Robert. 1990. Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization. Princeton, NJ: Princeton University Press. Wang, Jenn-hwan. 2003. “The Making of Unknown Heroes: National System of Learning, OEM, and Taiwan’s IT Industry.” Paper presented at 15th annual Society for the Advancement of Socio-Economics, Aix-en-Provence, June 26–28. Wang, V. 1995. “Developing the Information Industry in Taiwan: Entrepreneurial State, Guerrilla Capitalists, and Accommodative Technologists,” Pacific Affairs (October). Wank, David L. 1999. Commodifying Communism: Business, Trust, and Politics in a Chinese City. New York: Cambridge University Press. Watkins, Ralph. 2002. “Mexico Versus China: Factors Affecting Export and Investment Competition.” Industry Trade and Technology Review, July. Wedeman, Andrew Hall. 2003. From Mao to Market: Rent Seeking, Local Protectionism, and Marketization in China. Cambridge Modern China Series. Cambridge, UK; New York: Cambridge University Press. Wei, Shang-Jin, and Jeffrey A. Frankel. 1996. “Can Regional Blocs Be a Stepping Stone to Global Free Trade? A Political Economy Analysis.” International Review of Economics and Finance 5, no. 4: 339–47. Williamson, Oliver E. 1975. Markets and Hierarchies. New York: Free Press. ———. 1985. The Economic Institutions of Capitalism. New York: Free Press. Winckler, Edwin A., and Susan Greenhalgh. 1998. Contending Approaches to the Political Economy of Taiwan. Armonk, NY: M.E. Sharpe. Wong, Siu-Lun. 1988. Emigrant Entrepreneurs: Shanghai Industrialists in Hong Kong. Hong Kong: Oxford University Press. Woo, Jung-en. 1991. Race to the Swift: State and Finance in Korean Industrialization. New York: Columbia University Press. World Bank. 1993. The East Asian Miracle: Economic Growth and Public Policy. New York: Oxford University Press. World Bank GNI figures. 2003. www.worldbank.org/data/databytopic/GNIPC.pdf Wu, S. 1992. “The Dynamic Cooperation Between Government and Enterprise: The Development of Taiwan’s Integrated Circuit Industry.” In Taiwan’s Enterprises in Global Perspective, edited by N.T. Wang. Armonk, NY: M.E. Sharpe. Wu, Yongping. 2005. In Search of an Explanation of SME-Led Growth: State Survival, Bureaucratic Politics and Private Enterprise in the Making of the Taiwanese Economy. Cambridge, MA: Harvard University Press.

BIBLIOGRAPHY 319

Yang, Chyan, and Shui-Wan Hung. 2003. “Taiwan’s Dilemma across the Strait.” Asian Survey 43, no. 4: 681–96. Ye, Andrew. 2003. China Economic Quarterly Q4: 19–40. Yoffie, D. 1997. Competing in the Age of Digital Convergence. Boston: Harvard Business School Press. Young, Alwyn. 2000. “The Razor’s Edge: Distortions and Incremental Reform in the People’s Republic of China.” The Quarterly Journal of Economics 115, no. 4. Yusuf, Shahid, M. Anjum Altaf, and Kaoru Nabeshima, eds. 2004. Global Production Networking and Technological Change in East Asia. Washington, DC: The World Bank. Zajac, E.J., and C.P. Olsen. 1993. “From Transaction Cost to Transactional Value Analysis: Implications for the Study of Inter-Organizational Strategies.” Journal of Management Studies 30, no. 1: 131–45. Zhou, Xiaochuan. 1999. Renewal and Regeneration (Chongjian yu zaisheng). Beijing: China Financial Press. Ziegler, Nicholas. 1995. “Institutions, Elites, and Technological Change in France and Germany.” World Politics 47: 341–72. Zweig, David. 1997. “To Return or Not to Return? Politics Versus Economics in China’s Brain Drain to the U.S.” Studies in Comparative International Development 33, no. 1: 92–125. Zysman, John. 1996a. “The Myth of a ‘Global’ Economy: Enduring National Foundations and Emerging Regional Realities.” New Political Economy 1, no. 2. ———. 1996b. “Nations, Institutions, and Technological Development.” International Journal of Technology Management 12: 651–78. Zysman, John, Eileen Doherty, and Andrew Schwartz. 1996. “Tales from the ‘Global’ Economy: Cross-National Production Networks and the Re-Organization of the European Economy.” Structural Change and Economic Dynamics 8, no. 1: 45–85. http://ideas.repec.org/a/eee/streco/v8y1997i1p45–85.html

This page intentionally left blank

The Authors The researchers who conducted the research for Global Taiwan are: Suzanne Berger, Raphael Dorman and Helen Starbuck Professor of Political Science, MIT. Marcos Ancelovici, Graduate Researcher at Industrial Performance Center and a Doctoral Candidate, Department of Political Science, MIT. Akintunde Ibitayo Akinwande, Associate Professor in the Electrical Engineering and Computer Science Department at MIT. Dan Breznitz, Graduate Researcher at Industrial Performance Center and a Doctoral Candidate, Department of Political Science, MIT. Edward Cunningham, Graduate Researcher at Industrial Performance Center and a Doctoral Candidate, Department of Political Science, MIT. Douglas B. Fuller, Graduate Researcher at Industrial Performance Center and a Doctoral Candidate, Department of Political Science, MIT. Ji-Ren Lee, Professor in the School of Business, National Taiwan University. Richard K. Lester, Professor, Department of Nuclear Science and Engineering and Director, Industrial Performance Center, MIT. Teresa Lynch, Graduate Researcher at Industrial Performance Center and a Doctoral Candidate, Department of Political Science, MIT. Sara Jane McCaffrey, Graduate Researcher at Industrial Performance Center and a Doctoral Candidate, Department of Political Science, MIT. 321

322 THE AUTHORS

Edward S. Steinfeld, Associate Professor, Department of Political Science, MIT. Charles G. Sodini, Professor, Department of Electrical Engineering and Computer Science, MIT. Timothy J. Sturgeon, Senior Research Affiliate, Industrial Performance Center, MIT. Eric Thun, Research Associate, Industrial Performance Center, and Assistant Professor, Department of Politics, Princeton University.

Index

Acer active matrix liquid crystal display (AMLCD), 90 branded firms, 13–14, 54 complementary metal-oxidesemiconductor (CMOS) logic, 82 dynamic random access memory (DRAM), 83, 86, 87, 88t global competition, 9–10, 13–14 industry co-evolution, 48, 68, 69f, 70–71 industry value chain, 148, 149, 151, 152 notebook computers, 54 original design manufacturer (ODM), 48, 68, 69f, 70–71 production shifts, 172 Actebis, 50 Active matrix liquid crystal display (AMLCD). See Electronics industry innovation ADI, 90t ADMtek, 159 Adobe, 221 Aftermarket parts sector. See Automotive industry Airoha, 159, 160 Alcatel, 58, 60, 265 ALI, 146, 157, 159, 201–2 AMD, 82, 144 AMP, 152 Analog Devices, 153 Apple industry co-evolution, 51t, 55 innovation strategies, 13 production shifts, 174, 175t Application specific integrated circuit (ASIC), 204–5 Argentina, xiv Asia electronics industry co-evolution, 56

Asia (continued) financial market crisis (1997), xiv production location, 25, 26, 27, 28 production shifts, 169, 171, 172–77, 178t See also specific country Asian Wall Street Journal, 156 Association of Southeast Asian Nations (ASEAN) automotive industry, 119–20, 124, 126 Free Trade Area (AFTA), 119–20 AST, 266 ASUSTek branded firms, 54 global competition, 9–10 industry co-evolution, 49, 52–53, 54, 68, 69f, 70–71 mobile phones, 53, 159 original design manufacturer (ODM), 49, 52–53, 54, 68, 69f, 70–71 AU Optronics, 90t, 163, 164 Australia, 56, 109t Auto Body Parts Association (ABPA), 131–32 Automotive industry aftermarket parts buyer-driven chain, 101, 110–14 exports, 111, 116 firm strategies, 109–10, 115–19, 125–29 foreign solutions, 114–15 imports, 111, 114, 115f relocation strategies, 101 research & development (R&D), 116 reverse engineering technology, 101, 111, 113–14 sectoral contrast, 100–102 assembly sector domestic market fragmentation, 102–4 323

324 INDEX

Automotive industry assembly sector (continued) exports, 106, 107t firm strategies, 125–29 foreign technology dependence, 100–101, 102, 103–6, 128–29 multinational corporations (MNCs), 127, 131 producer-driven chain, 101–2 research & development (R&D), 104–6 sectoral contrast, 100–102 state role, 104–5, 285n21 Taiwanese foreign investment, 119–21 vehicle production (1989–2002), 102–3 branded firms, 111, 134 certification, 131, 132–34 exports aftermarket parts, 111, 116 assembly sector, 106, 107t original equipment manufacturer (OEM), 107–9, 126, 130 globalization integration, 134–36 imports aftermarket parts, 111, 114, 115f original equipment manufacturer (OEM), 107–9 late development, 97–102, 130 life-cycle model, 98–99, 284n7 niche markets, 101–2, 106, 109–11, 116–18, 131, 135, 136 original equipment manufacturer (OEM) domestic market fragmentation, 106 exports, 107–9, 126, 130 firm strategies, 118–19, 125–29 foreign technology dependence, 100–101, 106–9, 124 imports, 107–9 producer-driven chain, 101–2 sectoral contrast, 100–102 Taiwanese foreign investment, 121–25, 287n85 production networks, 131, 134–36 regional hierarchy approach, 99–100 research conclusions, 134–36 research & development (R&D), 104–6, 116, 132–33, 134 research introduction, 97

Automotive industry (continued) standardization, 113, 130, 132–34 state role assembly sector, 104–5, 285n21 China, 127–28 policy recommendations, 129–34 trade liberalization, 108, 119–20, 124, 126 value chains buyer-driven chain, 100, 101, 111–14, 116–18, 134, 135–36 commodity chain perspective, 99–100, 111, 113 modular value chains, 112–13, 131, 135 producer-driven chain, 99, 100, 101–2, 110–11, 135–36 Automotive Lighting, 118 Automotive Research and Testing Center (ARTC), 132 Bank of Communications, 138 BenQ, 13–14, 159, 203 Bird, 269, 273 BMW, 118 Board-level operating system (BIOS), 49–50, 53 Borderless World, The (Ohmae), xvii Bosch-Siemens, 245 Branded firms automotive industry, 111, 134 China, 13–14, 268, 269 competitive strategies, 10, 13–14 electronics industry co-evolution, 50, 53–54, 74–75 integrated circuit (IC) design industry, 201–2 value chain, 140, 144, 147, 155, 289n8 Hong Kong, 14 Brazil, 56 Broadcom, 160 Burlington, 180, 182t Burma, 19 Calculators, 49 Canada, 118, 119 Canon, 178t Capital Iron and Steel, 245 Caribbean Basin Initiative (CBI), 180, 182–83, 185

INDEX 325

Celestica electronics manufacturing services (EMS), 9, 54–55, 56–59, 68, 69f, 70–71 industry co-evolution, 9, 54–55, 56–59, 68, 69f, 70–71 production shifts, 174–75 Central Bank of China, 94 Central processing unit (CPU), 48–49, 53, 82 Certification (automotive industry), 131, 132–34 Certified Automotive Parts Association (CAPA), 132–33 Chain Store Project, 133 Challenge 2008 Plan, 158 Chang, Alice H., 222 Chang, Morris, 80, 83, 156, 200 Chartered of Singapore, 83, 84t Chevrolet, 118 Chi Mei Optoelectronics, 90, 164 China Development and Industrial Bank (CDIB), 138 China External Trade Development Council, 133 China-Mexico production shifts electronics industry, 167, 176–77, 178t, 179f, 190–93, 291n17 regionalization theories, 169–70 textile/apparel industry, 166, 167, 183–86, 187, 188–89, 190–93 China Mobile, 265 China Motor Corporation (CMC), 103, 106, 107t, 108, 117, 120, 126, 127, 285n17 China-Taiwan economic relationship automotive industry aftermarket parts, 110, 116 assembly sector, 119–21, 126–29, 131, 287n78 original equipment manufacturer (OEM), 108, 109t, 119, 122– 25, 126–29, 133, 135, 289n116 China investment global, 232, 233f, 234–35, 236f, 237f, 253, 295n28 global competition, 3–4, 11–13, 17–22, 25, 26, 27, 28 industrial policy, 257–60, 278–79, 295n41 ownership control, 240–41, 278

China-Taiwan economic relationship China investment (continued) Taiwanese, 229, 230, 232, 233f, 234–35, 238f, 239f, 240f, 278–79 Taiwanese industry, 235, 238f Taiwanese province, 235, 239f, 240f Chinese domestic industry communication, 246, 294n19 corporate structure, 244–50 customer base, 246–47, 294n18 export inhibitors, 247, 248t firm size, 245 innovation, 244–50 price wars, 249–50 product transportation, 246, 294n19 supply network, 246, 249, 254, 264, 294n17 Chinese industrial policy China investment, 257–60, 278–79, 295n41 comparative advantage, 228–29, 257–60, 262–64 creative-destruction development, 259–60, 263–64 Heckscher-Ohlin-Samuelson approach, 257–58 Japanese model, 261–62, 267–68, 269–70, 272–73, 275 national champion conglomerates, 261–64, 277 South Korean model, 261–62, 263–64, 272–73, 275 Chinese multinational corporations (MNCs), 264–76 branded firms, 268, 269 industrial catch-up, 271–76, 277 industry versus national industry, 273–74 information technology (IT) industry, 266–67 innovation, 269–71, 275–76 motorcycle industry, 267–71 niche markets, 266 telecommunications industry, 265–66 Chinese state capacity, 250, 254–57 asset management companies (AMCs), 256–57 financial system, 255–57 intellectual property rights, 256

326 INDEX

China-Taiwan economic relationship Chinese state capacity (continued) nonperforming loans (NPLs), 256–57 trade regulation, 255, 258, 295n39 Chinese state reform-style bankruptcies, 254, 295n33 decentralization, 250, 251, 253–55, 261–62, 295n31 enterprise financing, 251, 252t, 253, 295n27 entrepreneurship, 253–54 experimentation, 250–51 informality, 250, 251, 253, 254–55, 261–62 mergers, 253–54, 295n32 private enterprise, 251, 294n25 socialism, 250–51 Chinese state role automotive industry, 127–28 production-cost strategies, 12–13 production location, 17–18 economic development China, 230, 231f Taiwan, xi-xii, 3–4, 52, 230 Taiwan developmental policy, 217–18, 220, 223 economic integration, 230–42, 270–71, 276–79 electronics industry value chain, 137, 140, 145, 149, 154–57, 161, 164, 290n12 global competition, 8, 9, 10 branded firms, 13–14 innovation strategies, 15–17 production-cost strategies, 11–13 production location, xii, 3–4, 17–22, 25, 26, 27, 28, 281n7, 282n20 research & development (R&D), 15–17 technology-gap preservation, 16–17 global economy, 228–30 information technology (IT) industry, 3–4 innovation Chinese domestic industry, 244–50 Chinese multinational corporations (MNCs), 269–71, 275–76 disruptive, 244, 260 global competition, 15–17

China-Taiwan economic relationship innovation (continued) new products, 249–50, 294n21 price wars, 249–50 production architectures, 244, 275–76 supply network, 246, 249, 254, 269–71, 294n17, 295n34 sustaining, 244, 260, 269–71 intellectual property rights, 20 political relationship, 229, 241, 277–78 production architectures, 229–30, 242–50 codification, 229, 242–43, 269–71, 272–73 commodification, 229, 243, 270 connectivity rules, 242–44, 246, 293n6 diversification, 270–71 innovation, 244, 275–76 modularization, 229, 242–44, 246–47, 249, 260, 269–71, 273, 274–76, 293n4 specialization, 228–29, 257, 273–76 standardization, 269–71, 276 vertically integrated organization, 242–43, 260, 263–64, 267, 270–71, 275–76 production location global competition, xii, 3–4, 17–22, 25, 26, 27, 28, 281n7, 282n20 original design manufacturer (ODM), 52, 61, 62f, 65–66, 71–72, 73–74 state role, 17–18 research conclusions, 276–79 research introduction, 228–30 SARS, 185 China Telecom, 265 Chinese Petroleum, 152 Chips & Technologies, 146 Chrysler, 119 Chung Hua, 48 Chunghwa Picture Tubes, 90 Chungshan Institute, 158–60 Cisco Systems, 50, 58, 265, 273 Cluster companies, xvi, 48, 79, 175 See also Silicon Valley (California) Codec chips, 146 Commodity chains automotive industry, 99–100, 111, 113

INDEX 327

Commodity chains (continued) cross-straits integration, 229, 243, 246, 249, 254, 264, 269–71, 294n17, 295n34 electronics industry, 38–39, 41–43, 63, 64f production shifts, 169–70 Commodore, 148 Compal, 49, 53, 68, 69f, 70–71 Compaq, 148 See also Hewlett-Packard/Compaq Compeq, 148 Competitive strategies. See Global competition Complementary metal-oxidesemiconductor (CMOS) logic. See Electronics industry innovation Computer-aided design (CAD), 80, 111 Computer-aided manufacturing (CAM), 111 Computer monitors. See Monitors Computers, desktop. See Desktop computers Computers, notebook. See Notebook computers Cone Mills, 180, 182t Contract electronics manufacturer (CEM). See Electronics manufacturing services (EMS) Council for Economic Planning and Development (CEPD), 11–12, 138 Crolles initiative, 82, 144 Crystal, 81 Cyberlink, 212–13, 221, 222 Czech Republic, 24–25 D-Link, 203 DaimlerChrysler, 103, 119, 127 Dan River, 180, 182t David Sarnoff Laboratories, 90 DBM Reflex, 118 Defense Advanced Research Projects Agency (DARPA), 80 Dell China investment, 18 cross-straits integration, 273, 275–76 industry co-evolution, 51t, 52, 61, 63 industry value chain, 139 innovation strategies, 13 production fragmentation, xvi production shifts, 174, 175t

Delphi, xv Delta, 49 Delta Woodside, 180 Desktop computers electronics manufacturing services (EMS), 55, 59, 60–68 original design manufacturer (ODM) industry co-evolution, 46–54, 60–68 industry value chain, 140, 141, 148, 149, 151t, 155 Developmental network state. See State developmental policy Developmental state theory. See State developmental policy Diversification competitive strategies, 16 cross-straits integration, 270–71 electronics manufacturing services (EMS), 58, 70 original design manufacturer (ODM), 16, 52–53 DVD products, 216, 221–22 Dynamic random access memory (DRAM). See Electronics industry innovation Eastern Europe electronics industry co-evolution, 56, 60, 66, 72 production location, 24–25, 27 production shifts, 167, 175, 290n4 See also specific country eBay Motors, 134 Economic development chronology, 4, 6 developed-country status, 137, 289n1 G7 countries, 194, 291n1 late development, 97–102, 130 lessons of Taiwan, 3 neoclassical market-based theories, 4–5 newly industrialized economies (NIEs), 194, 200 per capita GDP, 4 per capita income, 4 postwar investment, 5–6 private investment, 4–5 revisionist theories, 4–5 small/medium-sized enterprises (SMEs), 94, 95, 194, 284n17 state role, 4–8 trade liberalization, 7–8

328 INDEX

Economic development (continued) See also China-Taiwan economic relationship; Global competition; Global economy; specific industry; State developmental policy Economist, The, 18 Economist Intelligence Unit (EIU), 19 Educational system, 96 Elan MicroElectronics, 201t Elcoteq Americas, 187 Electrolux, 245, 273, 294n15 Electronic Business Magazine, 57–58 Electronics contract manufacturer (ECM). See Electronics manufacturing services (EMS) Electronics industry economic development, 4 global competition, 8–10 original design manufacturer (ODM), 8–9 original equipment manufacturer (OEM), 8–9 production location, 18, 19, 20, 22, 23–24 See also Integrated circuit (IC) design industry; Production shifts; specific company/product; State developmental policy Economic development analysis level macro-level, 34–35, 36, 44, 69–70 meso-level, 36, 44 micro-level, 34–35, 36, 44 branded firms, 50, 53–54, 74–75 co-evolutionary framework development, 34–45 co-evolutionary process, 35–36, 40 collaboration value, 36–37, 42–43, 44–45, 283n8 competencies firm, 35, 36–38, 40, 42–43, 45 geographic scope, 46, 60–63 product/customer scope, 46, 60–63 supplier, 35–36, 37, 38, 40–43, 45 value-chain scope, 46, 60–63 competitive strategy contract manufacturer contrast, 63–72 firm-supplier relationships, 35, 45f implications, 72–75

Economic development (continued) contract manufacturer contrast competition, 63–72 convergence, 63–72 design services group, 64–65 differences, 60–61, 62f production location, 61, 62f, 65–66, 67t, 71–72, 73–74 relationship variations, 63, 64f revenues, 68–72 similarities, 61–63 defined, 35, 44–45 electronics manufacturing services (EMS) (United States) acquisitions, 59–60, 68–70 desktop computers, 55, 59, 60–68 development, 55–58 diversification, 58, 70 global economy, 58–60 information technology (IT) industry, 58–60 origins, 54–55 overexpansion (1990s), 59–60 overview, 46 production location, 56–58, 59–60, 61, 62f, 66, 67t, 70, 71–72, 73–74 product mix (2001), 58–59 revenues, 56f, 57t, 68–72 firm-supplier relationships competitive, 35, 45f exploitative, 35, 36–37, 41, 44–45 symbiotic, 35, 41, 44–45 geographic fragmentation, 33–34, 40, 53, 60–63, 73 global economic integration market level, 33–34 operations level, 33–34 horizontally-integrated companies, 40, 49 innovation strategy, 72–75 intellectual property rights, 43 manufacturing facility investment, 39–43 modular value chains, 39–41, 44–45, 49–50, 58, 60–63, 67–68, 73, 74 organizational fragmentation, 33–34, 53, 73 original design manufacturer (ODM) (Taiwan) board-level operating system (BIOS), 49–50, 53

INDEX 329

Economic development original design manufacturer (ODM) (Taiwan) (continued) calculators, 49 central processing unit (CPU), 48–49, 53, 82 desktop computers, 46–54, 60–68 development, 46–51, 282n3 diversification, 52–53 global economy, 51–54 information technology (IT) industry, 50–52, 283n6 integrated circuit (IC) industry, 47–48 mobile phones, 50, 53 monitors, 47f, 48, 50, 66f motherboards, 46–47, 49, 50, 52–53, 63–64, 65–66 notebook computers, 46–47, 48, 49, 52–53, 54, 63–64, 65–66, 67 operating system, 48–49, 53 origins, 46 overview, 46 peripherals, 46–48, 49 production location, 52, 61, 62f, 65–66, 71–72, 73–74 revenues, 68–72 production location China, 52, 60, 65–66, 70, 71–72 contract manufacturer contrast, 61, 62f, 65–66, 67t, 71–72, 73–74 electronics manufacturing services (EMS) (United States), 56–58, 59–60, 61, 62f, 66, 67t, 70, 71–72, 73–74 original design manufacturer (ODM) (Taiwan), 52, 61, 62f, 65–66, 71–72, 73–74 research implications, 72–75 research introduction, 33–34 resources generic assets, 36, 39–43, 44, 63, 64f, 73 specific assets, 36–37, 38–40, 41, 42f, 44, 63, 64f, 73 shared supplier networks, 34–36 state role, 51 strategic outsourcing, 34–45, 282n1 limitations, 43 multiple sourcing, 40–43, 44 parallel sourcing, 40

Economic development (continued) supplier characteristics captive suppliers, 38–39, 41–43, 63, 64f commodity suppliers, 38–39, 41–43, 63, 64f turnkey suppliers, 41–45, 63, 64f surface mount technology (SMT), 55 transaction cost theory, 38–39, 74 transaction-value proposition, 36–37 value-chain co-evolution, 35–45 supplier perspective, 37–44 vertically integrated organization, 37–40, 49, 67–68 See also specific company Electronics industry developmental policy. See State developmental policy Electronics industry innovation absorption defined, 77 active matrix liquid crystal display (AMLCD), 76 classic technological followers, 78, 89–91 flying geese model, 77–78 foreign technology transfer, 89–90, 91, 92 granularity, 92 human capital, 90 industry value chain, 140, 158, 163–65 patient capital, 92 product characteristics-economy connection, 78, 92 research & development (R&D), 90–91 state role, 90 technological absorption, 77 volume production, 95–96 cluster companies, 79 complementary metal-oxidesemiconductor (CMOS) logic, 76 flying geese model, 77–78 foreign technology transfer, 79 global recognition, 78, 81–82 granularity, 83, 91–92 human capital, 79–80 industry value chain, 153 patient capital, 91–92 product characteristics-economy connection, 78, 79–80, 83, 91–92

330 INDEX

Electronics industry innovation complementary metal-oxidesemiconductor (CMOS) logic (continued) pure-play foundry model, 80–83, 91–92 research & development (R&D), 79 state role, 79–80 successful innovation, 78, 79–83 volume production, 92, 95–96 computer aided design (CAD), 80 dynamic random access memory (DRAM), 76 debt-to-equity ratios, 85–86, 284n6 flying geese model, 78 foreign technology transfer, 83–84, 86–88, 95 granularity, 91–92, 95 human capital, 83–84 industry value chain, 163, 164 patient capital, 85–87, 91–92, 94–95 precarious technological position, 78, 83–89 product characteristics-economy connection, 78, 91–92 research & development (R&D), 84, 85, 86–87, 95 state role, 83–84 volume production, 85–86, 92, 95–96 equity/debt ratio, 85–86, 93–95, 200, 284n6 Europe, 83 fabless design houses, 81, 83 financial market crises, 78, 200 flying geese model, 77–78 foreign technology transfer, 77, 95 AMLCD technology, 89–90, 91, 92 CMOS logic technology, 79 DRAM technology, 83–84, 86–88, 95 granularity, 76–77, 91–93, 95 AMLCD technology, 92 CMOS logic technology, 83, 91–92 complementary metal-oxidesemiconductor (CMOS) logic, 83 defined, 77 DRAM technology, 91–92, 95 human capital, 78, 79–80, 83–84, 90

Electronics industry innovation (continued) innovation defined, 77 integrated device manufacturer (IDM), 80–81, 83, 92, 200, 283n2 Japan, 82, 83, 85–86, 87, 89, 91, 92, 93, 284n14 manufacturing-based outputs, 96 modularity defined, 76–77 patient capital, 91–96, 200 AMLCD technology, 92 CMOS logic technology, 91–92 defined, 76 DRAM technology, 85–87, 91–92, 94–95 product characteristics-economy connection, 91–96 AMLCD technology, 78, 92 CMOS logic technology, 78, 79–80, 83, 91–92 defined, 76, 77 DRAM technology, 78, 91–92 product cycle theories, 78 research conclusions, 92–96 research & development (R&D), 93, 95 AMLCD technology, 90–91 CMOS logic technology, 79 DRAM technology, 84, 85, 86–87, 95 research introduction, 76–78 small/medium-sized enterprises (SMEs), 94, 95, 284n17 South Korea, 83, 85–86, 89, 91, 92, 93, 94, 284n14 state role, 77, 78, 93–95 AMLCD technology, 90 CMOS logic technology, 79–80 DRAM technology, 83–84 strategic heuristics, 96 Taiwan educational system, 96 trade liberalization, 94 United States, 78, 79, 81, 82, 83, 90 vertically integrated organization, 80 volume production, 95–96 AMLCD technology, 95–96 CMOS logic technology, 92, 95–96 defined, 76 DRAM technology, 85–86, 92, 95–96 See also State developmental policy

INDEX 331

Electronics industry value chain active matrix liquid crystal display (AMLCD), 140, 158, 163–65 complementary metal-oxidesemiconductor (CMOS) logic, 153 contract electronics manufacturer (CEM), 149, 150 dynamic random access memory (DRAM), 163, 164 industry development characteristics, 138–39 gross domestic product (GDP), 137, 289n3 information technology (IT) industry, 137 integrated circuit (IC) industry, 137, 289n2 Taiwan economy, 137, 289n1 technoglobalist approach, 138, 140 technonationalism, 138 industry opportunities China, 154–57, 290n12 intrafirm capabilities, 150–54 research & development (R&D), 151–54 state role, 151–52, 154, 156, 157 integrated circuit (IC) industry branded firms, 144, 289n8 Codec chips, 146 dependent relationships, 140, 141, 144 Ethernet chips, 146 fabless design houses, 144, 145–46, 153–54, 155, 157, 160 human capital, 142, 145, 146, 157 IC design, 140, 141, 145–47 IC fabrication, 140–41, 142–45 integrated device manufacturer (IDM), 141, 142–45, 153, 157 interdependent relationships, 139, 140–47 pure-play foundry model, 139, 140, 142 research & development (R&D), 142–43, 144–45, 146–47 state role, 140–41, 146–47 supplier model, 139 multinational corporations (MNCs), 140, 146, 160–61, 162t, 290n10 original design manufacturer (ODM) branded firms, 140, 147, 155

Electronics industry value chain original design manufacturer (ODM) (continued) dependent relationships, 139, 140, 141, 149 desktop computers, 140, 141, 148, 149, 151t, 155 exports, 141 monitors, 151t, 155 motherboards, 151t, 155 notebook computers, 140, 141, 144, 148–49, 151t, 154, 155 state role, 140–41, 147–49, 289n6 supplier model, 139 systems manufacturing, 140–41, 147–50, 151t upgrading, 148, 290n11 value (smile) curve, 149, 150f, 155 original equipment manufacturer (OEM), 144, 147–49, 290n9 own brand manufacturer (OBM), 147–48, 149, 159–60 patents, 138, 146, 152–54, 289n5 research conclusions, 164–65 research & development (R&D) industry opportunities, 151–54 integrated circuit (IC) industry, 142–43, 144–45, 146–47 ITRI model, 138–39, 140, 141, 157–58 state role, 138–39, 140–41, 157–58, 160–61, 162t, 164–65 research introduction, 137–40 state role Challenge 2008 Plan, 158 industry opportunities, 151–52, 154, 156, 157 integrated circuit (IC) industry, 140–41, 146–47 limitations, 140, 161, 163–64 original design manufacturer (ODM), 140–41, 147–49, 289n6 research & development (R&D), 138–39, 140–41, 157–58, 160–61, 162t, 164–65 wireless technologies, 139, 158–60 wireless technologies, 139, 154 radio frequency (RF) circuit design, 158–60 wireless local area network (WLAN), 158–60, 161

332 INDEX

Electronics manufacturing services (EMS) contract electronics manufacturer (CEM), 149, 150 electronics contract manufacturer (ECM), 172, 173–74, 290n2 global competition, 9, 17 production shifts, 172, 173–74, 290n2 See also Electronics industry coevolution Elite Semiconductor, 201t Elpida, 87, 88t Embedded autonomy industrial bureaucracy. See State developmental policy Emerson Electric, 178t Ericsson, 58, 269 ESS, 146 Ethernet chips, 146 Etron, 203 Europe automotive industry, 107, 109t, 122 electronics industry co-evolution, 50, 53, 54, 56, 67–68 innovation, 83 value chain, 141, 149, 158–60, 162t, 165 production location, 24–28 production shifts, 167, 169, 171, 173, 175 Taiwan developmental policy, 206–7, 222 See also Eastern Europe; specific country Executive Yuan Development Fund, 138 Exports. See Automotive industry; Production location; Production shifts; Trade liberalization; Trade regulation Exxon/Mobil, 245 Fabless design houses. See Integrated circuit (IC) design industry Faraday, 157, 201t Ferrari, 118 Financial market crises, xiv, 200 Finland, 15 First International Computer, 9–10 Flexible developmental state (FDS). See State developmental policy

Flextronics electronics manufacturing services (EMS), xv, 9, 54–55, 56–59, 65, 66, 68, 69f, 70–71, 72, 283n7 global competition, 9 industry co-evolution, xv, 9, 54–55, 56–59, 65, 66, 68, 69f, 70–71, 72, 283n7 original design manufacturer (ODM), xv, 9 production shifts, 172, 174–75, 178t, 189, 190 Flying geese model, 77–78 Fodus, 159 Ford, 103, 106, 118, 128 Ford Lio Ho, 103–4, 107t Foreign investment global economic development, 5–6 Mexican economy, 171, 172t, 172–73, 175, 181, 182t, 183t postwar Taiwan, 5–6 Taiwan automotive industry, 119–25, 287n8 global competition, 3–4, 11–13, 17–22, 25, 26, 27, 28 Mexican textile/apparel industry, 171, 172t See also China-Taiwan economic relationship; Production location; Production shifts; Trade liberalization Formosa Plastics Group, 84 Fragmentation domestic market (automotive industry), 102–4, 106 geographic (electronics industry), 33–34, 40, 53, 60–63, 73 organizational (electronics industry), 33–34, 53, 73 production, xvi, 17 France, 60, 153 Free trade. See North American Free Trade Agreement (NAFTA); Production location; Production shifts; Trade liberalization; Trade regulation Fujitsu, 50, 90t Fujitsu/Siemens, 50 Full Soft, 218 Fuso, 119

INDEX 333

Galanz Enterprise Group, 249–50 Galey and Lord, 180, 182t Gamania, 218 Game sector, 218–20 Gap, 180 GE Medical, 58 Genera Corporation, 117–18, 119 General Electric, 245, 266, 294n15 General Motors, 35, 123, 128, 135, 235 Geographic fragmentation. See Fragmentation Germany automotive industry, 119 craft production, 76, 95–96 globalization research, xiii research & development (R&D), 160–61 Gigabyte, 9–10, 54, 202 Global competition, 8–10 brand strategies, 10, 13–14 diversification strategy, 16 electronics industry strategies contract manufacturer contrast, 63–72 firm-supplier relationships, 35, 45f implications, 72–75 innovation state role, 28–32 strategies, 10, 14–17 production-cost strategies, 10, 11–13 production location China, xii, 3–4, 17–22, 25, 26, 27, 28, 281n7, 282n20 global, 24–28 Taiwan, 21–24 product/services differentiation, 13–17 state role innovation, 28–32 intellectual property rights, 31–32 research & development (R&D), 29–31 trade liberalization, 8 Global economy automotive industry, 134–36 cross-straits integration, 228–30 electronics industry co-evolution electronics manufacturing services (EMS), 58–60 market-level integration, 33–34 operations-level integration, 33–34

Global economy electronics industry co-evolution (continued) original design manufacturer (ODM), 51–54 information technology (IT) industry, xi-xii, xiii, xiv production shifts globalization theories, 166–67, 168–71, 172t, 175 global producer future, 192–93 Taiwan miracle, 3–4 trade liberalization, xiii-xiv transportation technology, xiii See also China-Taiwan economic relationship; Economic development; Global competition; State developmental policy Globalization Study (IPC). See Industrial Performance Center (IPC) (MIT) Global Validators, 132 Gold Lion, 14 Gordon Auto Body Parts Company, 112, 115–16 Grace Group, 156, 235 Granularity. See Electronics industry innovation Guangzhou Honda, 128 Guide Corporation, 118 Guilford Mills, 180, 182t Haier, 245, 266, 272, 273, 294n23 Haier America, 266 Hannstar Display, 90t Hejian, 156, 157 Hewlett-Packard industry value chain, 139 production shifts, 172, 174, 175, 178t state developmental policy, 217 Hewlett-Packard/Compaq industry co-evolution, 51t, 52, 55–56, 61, 63 innovation strategies, 13 Hino, 103 Hitachi, 50, 87, 175 Holtek Semiconductor, 82, 201t Honda, 103, 126, 128 cross-straits integration, 267, 268, 269, 270 Honda (USA), 103 Honda Wave Alpha, 268

334 INDEX

Honea, 267 Hong Ho Precision Textile, 166–67 Hong Kong branded firms, 14 China investment, 232 China relations, 11, 266–67 globalization research, xiii production location, 21–22 production shifts, 166–67, 178t, 185, 189 SARS, 185 Hon Hai (Foxconn) global competition, 9–10 industry co-evolution, 49, 53, 68, 69f, 70–71, 72 industry value chain, 152–53, 161 mobile phones, 53 original design manufacturer (ODM), xv, 49, 53, 68, 69f, 70–71, 72 Hood, 134 Horizontally-integrated companies, 40, 49 Hsinchu Science-Based Industrial Park (HSBIP), 79, 152, 216, 220, 224, 226 Huawei, 265, 272, 273 Human capital (electronics industry) industry innovation, 78, 79–80, 83–84, 90 industry value chain, 142, 145, 146, 157 Hungary, 25 Hyundai Motors, 89, 95, 103, 118, 128 IBM active matrix liquid crystal display (AMLCD), 90t cross-straits integration, 245, 266, 269, 271, 273, 274, 275 dynamic random access memory (DRAM), 87, 88t industry co-evolution, 48–49, 51t, 52, 55–56, 60, 63 industry value chain, 145, 148 innovation strategies, 13 model AT, 48–49, 52 production shifts, 172, 173–74 ICreate, 203 Imports. See Automotive industry; Production location; Production shifts; Trade liberalization; Trade regulation

India, 8, 10, 212–13, 214t Indonesia, 19, 111, 267–68 Industrial Performance Center (IPC) (MIT) globalization research collaboration, xi-xiii, xv convergence theory, xvii-xviii macro-level analysis, xiii, xvii-xviii national models theory, xvii-xviii Globalization Study collaboration, xi, xix-xxi countries, xix, xxt industrial performance analysis, xiixix industries, xiit, xiiit, xix interviews (1999–2003), xii, xiiit, xix, xxt, 139, 196, 292n5 micro-level analysis, xiii, xviii-xix open pathways approach, xviii-xix research methodology, xii, xiiit, xvii-xix, xxt research objectives, xi research overview, xix-xxi theoretical foundations, xvii-xviii Industrial Technology Research Institute (ITRI) active matrix liquid crystal display (AMLCD), 90t, 140 automotive industry, 105 complementary metal-oxide semiconductor (CMOS) logic, 79, 82 Computer and Communications Laboratory (CCL) electronics industry value chain, 146, 148–49, 158–59 integrated circuit (IC) design industry, 196, 205, 211–12 research & development (R&D), 30 dynamic random access memory (DRAM), 84, 88t Electronic Research Service Organization (ERSO), 30, 79, 84, 90, 146–47 ASTRO, 151–52 integrated circuit (IC) design industry, 196, 199–200, 204–6 Opto-electronics Science Laboratory (OESL), 146 electronics industry value chain, 138–39, 140, 141, 146–47, 148–49, 151–52, 157–59, 161

INDEX 335

Industrial Technology Research Institute (ITRI) (continued) innovation, 29–31 integrated circuit (IC) design industry, 196, 199–200, 204–6, 208–9, 210, 211–12, 225, 226 Mechanical Industrial Research Laboratory (MIRL), 105, 132–33, 285n29 patents, 15, 152 research & development (R&D), 29–31 Submicron Project, 141, 151 Very Large Scale Integration Project, 151 Infineon, 86, 87–88, 144 Information technology (IT) industry China, 3–4 electronics industry value chain, 137 electronics manufacturing services (EMS), 58–60 global economy, xi-xii, xiii, xiv innovation strategies, 13 original design manufacturer (ODM), 50–52, 283n6 production location, 3–4 See also Electronics industry; Integrated circuit (IC) design industry; Production shifts; specific company/product; State developmental policy Innovation China, 15–17 competitive strategies, 10, 14–17 electronics industry co-evolution, 72–75 patents, 15 state role, 28–32 electronics industry, 51, 77, 78, 79–80, 83–84, 90, 93–95 intellectual property rights, 31–32 multipolar structure, 31 research & development (R&D), 29–31 See also China-Taiwan economic relationship; Electronics industry innovation; Integrated circuit (IC) design industry; State developmental policy Institute for Information Industry (III), 196, 212, 213, 215–20, 222, 223–24, 225, 226–27, 292n4

Integrated circuit (IC) design industry cluster companies, 48 fabless design houses, 48, 81, 83, 282n4 industry innovation, 81, 83 industry value chain, 144, 145–46, 153–54, 155, 157, 160 state developmental policy, 201–12, 226 original design manufacturer (ODM) foundry services, 47, 80 packaging services, 47–48, 80 testing services, 47–48, 80 pure-play foundry model electronics industry innovation, 80–83, 91–92 electronics industry value chain, 139, 140, 142 state developmental policy, 200, 202, 203–4, 206, 225–26 See also Electronics industry value chain; State developmental policy Integrated device manufacturer (IDM) industry innovation, 80–81, 83, 92, 200, 283n2 industry value chain, 141, 142–45, 153, 157 Intel central processing unit (CPU), 48–49, 53, 147, 202, 203 cross-straits integration, 18, 269, 271, 275–76 industry co-evolution, 48–49, 50, 53, 61, 63 industry value chain, 145 Pentium processors, 202, 269 production shifts, 175 Rambus technology, 202 state developmental policy, 202, 203 Intellectual property rights, 20, 31–32, 43, 156 International Telephone and Telegraph, 148 Interserv, 218 Ireland, 212–13, 214t, 227 Israel patents, 15 software industry, 203, 205, 212–13, 214t Italy, 76, 95–96 Iventec, 49

336 INDEX

Jabil Circuit electronics manufacturing services (EMS), 9, 54–55, 56–59, 68, 69f, 70–71 global competition, 9 industry co-evolution, 9, 54–55, 56–59, 68, 69f, 70–71 production shifts, 174–75, 190 Japan automotive industry, 99 aftermarket parts, 111, 112, 113–14, 118, 286n48 assembly sector, 101, 104, 120, 125, 126, 127, 128 original equipment manufacturer (OEM), 100, 101, 106–9, 122, 124, 126 technological dependence, 101, 104, 106–9, 112, 113–14, 118, 128, 129, 285n20, 286n48, 288n101 cross-straits integration, 261–62, 267–68, 269–70, 272–73, 275 electronics industry co-evolution, 50, 67–68 innovation, 82, 83, 85–86, 87, 89, 91, 92, 93, 284n14 value chain, 137, 141, 144, 147, 153, 159, 160–61, 162t, 165 globalization research, xiii motorcycle industry, 267–68 patents, 15 production location, 24 strategic outsourcing, 67–68 Taiwan developmental policy, 200, 206–7, 221, 222 Jau Huang, 222 Jeep, 119 Johnson Controls, 135 Johnson & Johnson, 58 Kellwood, xv Keystone, 134 Kodak, 175 Konka, 249, 294n22 Kuozui Motors, 103, 106, 107 Leading Product Development Program (LPDP), 209–10 Lenovo, 245, 266–67, 269, 271, 273, 275–76, 277

LG Electronics, 83 Lifescan, 58 Liz Claiborne, 180 Lucent Technology, 58, 175, 178t Macauto Industrial Company, 118–19 Machine that Changed the World, The (MIT), xv Macronix, 82 Made in America (MIT), xiii, xv Made in Hong Kong (MIT), xiii Magna, xv Malaysia, 65, 83 Malcolm Baldrige Award, 55 Manufacturers’ Qualification and Validation Program (MQVP), 132 Market Intelligence Center (MIC), 47, 52, 67, 148 Mazda, 103–4 MediaTek industry value chain, 141, 146, 159 state developmental policy, 201t, 202, 206, 212 Mercedes-Benz, 118, 119, 127 Mexico electronics industry co-evolution, 66, 72 financial market crisis (1994), xiv production-cost strategies, 12 production location, 26–27, 281n19 See also Production shifts Microsoft cross-straits integration, 265–66, 271, 274–76 industry co-evolution, 48–49, 53, 61, 63 industry value chain, 147, 159 operating system, 48–49, 53, 147 production shifts, 189 state developmental policy, 217, 218 Microtek, 221 Ministry of Economic Affairs (MoEA) CMOS logic technology, 79 Department of Industrial Technology, 209–10, 292n8 Industrial Development Bureau (IDB), 104–5, 158, 209, 223 research & development (R&D), 138, 154 Ministry of Finance, 105 Mitac, 9–10, 48, 141, 148

INDEX 337

Mitsubishi, 87, 88t, 103, 117, 126, 127 Mobile phones, 50, 53, 249, 269 Mobiletron, 116–17 Modularity automotive industry, 112–13, 131, 135 cross-straits integration, 229, 242–44, 246–47, 249, 260, 269–71, 273, 274–76, 293n4 defined, 76–77 electronics industry, 39–41, 44–45, 49–50, 58, 60–63, 67–68, 73, 74, 76–77 Monitors original design manufacturer (ODM) industry co-evolution, 47f, 48, 50, 66f industry value chain, 151t, 155 production shifts, 174 Mosel Vitelic, 86, 87 MOSIS, 80 Motherboards industry co-evolution, 46–47, 49, 50, 52–53, 63–64, 65–66 industry value chain, 151t, 155 Motorola cross-straits integration, 235, 269, 273 industry co-evolution, 50 industry value chain, 144, 159 production shifts, 172, 178t Mu Chip, 159 Multinational corporations (MNCs) automotive industry, 127, 131 China, 264–76 economic development, 5, 194 electronics industry, 140, 146, 160–61, 162t, 290n10 production shifts, 167, 170, 189–90 research & development (R&D), 15, 140, 160–61, 162t

NEC (continued) industry co-evolution, 50 Neodevelopmental state theory. See State developmental policy Netherlands, 24–25 Nicaragua, 26 Niche markets automotive industry, 101–2, 106, 109–11, 116–18, 131, 135, 136 cross-straits integration, 266 production shifts, 191 research & development (R&D), 30 software industry, 217 Nihon Yamaha, 267 Nippon Steel, 144, 245 Nissan, 103, 106, 118, 128 Nokia, 50, 235, 269 Nortel, 58, 265 North American Free Trade Agreement (NAFTA) production location, 25–28 production shifts global producer future, 191–93 Mexican electronics industry, 172–76 Mexican textile/apparel industry, 178–80, 183, 185, 186, 188–89 regional trading blocks, 167, 168– 71 Notebook computers original design manufacturer (ODM) industry co-evolution, 46–47, 48, 49, 52–53, 54, 63–64, 65–66, 67 industry value chain, 140, 141, 144, 148–49, 151t, 154, 155 production shifts, 173–74 Novatek Microelectronics, 201t, 202

Nanya, 86, 87–88 National Semiconductor, 146 National Software Directorate (NSD), 227 National Taiwan University, 222 NatStell, 175 NEC complementary metal-oxidesemiconductor (CMOS) logic, 82 cross-straits integration, 265–66 dynamic random access memory (DRAM), 87

Ohmae, Kenichi, xvii Oki, 87, 88t Operating system, 48–50, 53, 147 Oracle, 217 Organizational fragmentation. See Fragmentation Original design manufacturer (ODM) branded firms, 13–14 diversification strategies, 16, 52–53 electronics industry, 8–9 global competition, 8–9, 13–14, 16 global economy, xv, xvi

338 INDEX

Original design manufacturer (ODM) (continued) integrated circuit (IC) design industry, 211–12, 293n11 production location, 19 production shifts, 177 state developmental policy, 211–12, 293n11 See also Electronics industry coevolution; Electronics industry value chain; specific manufacturer Original equipment manufacturer (OEM) branded firms, 13–14 diversification strategies, 16 electronics industry, 8–9 co-evolution, 46, 47, 49, 282n2 value chain, 144, 147–49, 290n9 global competition, 8–9, 13–14, 16 global economy, xv integrated circuit (IC) design industry, 203–7, 211, 225–26, 293n11 production location, 19 production shifts, 172–76 software industry, 221–22 state developmental policy, 203–7, 211, 225–26, 293n11 See also Automotive industry; specific manufacturer Osborne, 55 Own brand manufacturer (OBM) electronics industry value chain, 147–48, 149, 159–60 integrated circuit (IC) design industry, 203–7 state developmental policy, 203–7 Patents, 15, 138, 146, 152–54 high-technology utility patents, 138, 289n5 Patient capital. See Electronics industry innovation Philippines, 12, 19, 20, 120 Philips, 50, 144, 152, 160, 235 Pioneer, 178t Poland, 25 Powerchip, 86, 87, 88 Prime View International, 90t Prince Motors, 107t Product cycle theories, 78, 98–99, 284n7

Production architectures. See ChinaTaiwan economic relationship Production costs, 10, 11–13 Production fragmentation. See Fragmentation Production location China electronics industry co-evolution, 52, 60, 65–66, 70, 71–72 global competition, xii, 3–4, 17–22, 25, 26, 27, 28, 281n7, 282n20 electronics industry co-evolution China, 52, 60, 65–66, 70, 71–72 contract manufacturer contrast, 61, 62f, 65–66, 67t, 71–72, 73–74 electronics manufacturing services (EMS), 56–58, 59–60, 61, 62f66, 67t, 70, 71–72, 73–74 original design manufacturer (ODM), 52, 61, 62f, 65–66, 71–72, 73–74 global, 24–28 North American Free Trade Agreement (NAFTA), 25–28 Taiwan, 21–24 Production shifts Asia, 169, 171, 172–77, 178t Caribbean Basin Initiative (CBI), 180, 182–83, 185 China electronics industry, 167, 176–77, 178t, 179f, 190–93, 291n17 regionalization theories, 169–70 textile/apparel industry, 166, 167, 183–86, 187, 188–89, 190–93 globalization global producer future, 192–93 theories, 166–67, 168–71, 172t, 175 introduction, 166–68 Mexican economy decline, 176–77, 178t, 182f foreign direct investment (FDI), 171, 172t, 172–73, 175, 181, 182t, 183t labor costs, 166, 167, 171, 186–87 peso devaluation, 187–88 product transportation, 170, 183–84, 291n14 trade regulation, 166, 167, 170, 188–89, 291n14

INDEX 339

Production shifts (continued) Mexican electronics industry clusters, 175 electronics contract manufacturer (ECM), 172, 173–74, 290n2 future of, 190–93 NAFTA impact, 172–76 niche markets, 191 original design manufacturer (ODM), 177 original equipment manufacturer (OEM), 172–76 transformation of, 171–77, 178t, 179f turnaround time, 170, 173–74, 190 vertically integrated organization, 174 Mexican research & development (R&D), 191, 291n17 Mexican textile/apparel industry, 177–90, 191f commodity chains, 169–70 foreign direct investment (FDI), 171, 172t future of, 190–93 lean retailing, 169–70, 171–72 NAFTA impact, 178–80, 183, 185, 186, 188–89 niche markets, 191 turnaround time, 169–70, 171, 190 vertically integrated organization, 180 multinational corporations (MNCs), 167, 170, 189–90 North American Free Trade Agreement (NAFTA) electronics industry, 172–76 global producer future, 191–93 regional trading blocks, 167, 168–71 textile/apparel industry, 178–80, 183, 185, 186, 188–89 regionalization implications for, 186–90 theories, 166–67, 168–71, 172t, 177–78, 187, 190–91 trading blocks, 166–67, 169–71, 175 Taiwan foreign direct investment (FDI), 171, 172t

Production shifts (continued) global producer future, 192–93 original design manufacturer (ODM), 177 textile/apparel industry, 166 United States electronics industry, 172–77, 179f, 190–93 global producer future, 192–93 market access, 166, 167, 171, 173, 175–76 market proximity, 166, 167, 169, 171, 175–76 original equipment manufacturer (OEM), 172–76 recession impact, 166, 176–77, 186 regional trading blocks, 166–67, 169–71 textile/apparel industry, 178–90, 190–93, 291n7 Program for Sectoral Promotion (PROSEC), 188–89 Promos, 86, 87, 88t Public policy. See State developmental policy; State role Pure-play foundry model. See Integrated circuit (IC) design industry Quanta active matrix liquid crystal display (AMLCD), 90 calculators, 49 global competition, 9–10 industry co-evolution, 49, 53, 68, 69f, 70–71 industry value chain, 139 mobile phones, 53 notebook computers, 49, 54 original design manufacturer (ODM), xv, xvi, 49, 53, 68, 69f, 70–71 production fragmentation, xvi production location, 24–25 state developmental policy, 203 Quasel, 86, 87, 89, 283n4 Radio frequency (RF) circuit design, 158–60 Ralink, 159 Ralph Lauren, 180 RCA, 79, 200

340 INDEX

RealTek, 146, 157, 159, 201t Regionalization. See Production shifts Renault, 103 Research & development (R&D) automotive industry, 104–6, 116, 132–33, 134 China, 15–17 electronics industry innovation, 79, 84, 85, 86–87, 90–91, 93, 95 electronics industry value chain industry opportunities, 151–54 integrated circuit (IC) industry, 142–43, 144–45, 146–47 state role, 138–39, 140–41, 157–58, 160–61, 162t, 164–65 global competition, 15–17, 29–31 innovation, 14–17, 29–31 integrated circuit (IC) industry electronics industry value chain, 142–43, 144–45, 146–47 IC design, 195, 196, 197, 199–200, 204–6, 207–12, 226 multinational corporations (MNCs), 15, 140, 160–61, 162t production location, 21–22, 23, 24–25, 27–28 production shifts, 191, 291n17 software industry, 195, 196, 197, 215–16, 223–24, 226–27 state developmental policy, 195, 196, 197, 199–200, 204–6, 207–12, 226 state role, 29–31, 138–39, 140–41, 157–58, 160–61, 162t, 164–65 Taiwan investment, 14–15 Romania, 25, 60 Royal Philips Electronics, 178t, 189 Russia, xiv SAIC, 123 Samsung Electronics, 86, 89, 95, 165 Sanmina-SCI electronics manufacturing services (EMS), 9, 54–55, 56–59, 60, 68, 69f, 70–71 global competition, 9 industry co-evolution, 9, 54–55, 56–59, 60, 66, 68, 69f, 70–71 production shifts, 172, 174–75 Sanyang, 103, 126 Sanyo, 178t

Science and Technology Advisory Group (STAG), 79, 138 SCI Systems. See Sanmina-SCI Shanghai Volkswagen, 121, 128–29 Sharp, 90t Shih, Stan, 149, 150f, 155 Shinwave, 215 Siemens, 50, 175, 245 Silicon Integrated System (SiS), 141, 146, 204 Silicon Valley (California), xvi, 24–25, 54, 56, 58, 72, 75, 157, 159 Singapore, 56, 65, 83, 144 Sinopec, 245 Small Business Innovation Research (SBIR), 205, 209–10 Small/medium-sized enterprises (SMEs), 94, 95, 194, 284n17 Smart, 119 SMIC, 156 Softstar, 218 Software industry. See State developmental policy Soft-World, 218 Solectron electronics manufacturing services (EMS), xv, 9, 50, 54–55, 56, 57t, 58–59, 60, 66, 67t, 69f, 70–71 global competition, 9 industry co-evolution, xv, 9, 50, 54–55, 56, 57t, 58–59, 60, 66, 67t, 69f, 70–71 production shifts, 172, 174–75 Sony China investment, 18 industry co-evolution, 50, 51t, 52–53, 67 innovation strategies, 13 South Korea automotive industry, 99, 108, 109t, 118, 119 cross-straits integration, 261–62, 263–64, 272–73, 275 electronics industry innovation, 83, 85–86, 89, 91, 92, 93, 94, 284n14 value chain, 143, 147, 149 integrated device manufacturer (IDM), 143 patents, 15 Taiwan developmental policy, 200, 206–7, 219

INDEX 341

Specialization, 228–29, 257, 273–76 Standardization automotive industry, 113, 130, 132–34 cross-straits integration, 269–71, 276 State developmental policy China, 217–18, 220, 223 developmental network state, 198 developmental state theory, 198 economic development, 4–8, 194–95 embedded autonomy industrial bureaucracy, 198 Europe, 206–7, 222 flexible developmental state (FDS), 198–99 India, 212–13, 214t information technology (IT) growth, 196, 197 labor division, 195, 199 theoretical review, 198–201 innovation objective integrated circuit (IC) design industry, 194–95, 196–97, 200, 201–7, 208–12, 225–26, 291n2 software industry, 194–95, 196–97, 291n2 integrated circuit (IC) design industry application specific integrated circuit (ASIC), 204–5 branded firms, 201–2 business models, 194, 196–97, 200, 202, 203–7, 211 Computer and Communications Laboratory (CCL), 196, 205, 211–12 Department of Industrial Technology, 209–10, 292n8 development of, 201, 203–7 Electronic Research Service Organization (ERSO), 196, 199–200, 204–6 fabless design houses, 201–12, 226 financial analysis, 194, 196–97, 207–10, 226 foreign technology dependency, 201–2 growth of, 201 Industrial Development Bureau (IDB), 209

State developmental policy integrated circuit (IC) design industry (continued) Industrial Technology Research Institute (ITRI), 196, 199–200, 204–6, 208–9, 210, 211–12, 225, 226 innovation capabilities, 194–95, 196–97, 200, 201–7, 208–12, 225–26, 291n2 institutional role, 194, 195, 196–97, 207–12 Leading Product Development Program (LPDP), 209–10 original design manufacturer (ODM), 211–12, 293n11 original equipment manufacturer (OEM), 203–7, 211, 225–26, 293n11 own brand manufacturer (OBM), 203–7 policy limitations, 197, 200, 225–26 pure-play foundry model, 200, 202, 203–4, 206, 225–26 research & development (R&D), 195, 196, 197, 199–200, 204–6, 207–12, 226 Small Business Innovation Research (SBIR), 205, 209–10 stock options, 207–8 structure of, 203–12 System on Chip Project, 208–9 Taiwan stock exchange, 210, 293n10 tax incentive program, 209 Technology Development Program (TDP), 209–10 venture capitalists (VCs), 210, 293n10 Ireland, 212–13, 214t, 227 Israel, 203, 205, 212–13, 214t Japan, 200, 206–7, 221, 222 limitations of integrated circuit (IC) design industry, 197, 200, 225–26 software industry, 197, 217–20, 223–24 neodevelopmental state theory, 198–99, 224–25, 292n6 public-research-institution-based policy, 197, 200

342 INDEX

State developmental policy (continued) research conclusions, 197, 224–27 research introduction, 194–97 research methodology interviews, 196, 292n5 macro-level analysis, 195–96 micro-level analysis, 195–96, 199, 224–26, 291n3 software industry business models, 194, 196–97, 216–22, 227 development of, 216–22 DVD products, 216, 221–22 exports, 212–13, 214t financial analysis, 194, 196–97, 223–24, 226 game sector, 218–20 growth of, 212–13, 216–17 innovation capabilities, 194–95, 196–97, 291n2 Institute for Information Industry (III), 196, 212, 213, 215–20, 222, 223–24, 225, 226–27, 292n4 institutional role, 194, 195, 196–97, 223–24 niche markets, 217 original equipment manufacturer (OEM), 221–22 policy limitations, 197, 217–20, 223–24 research & development (R&D), 195, 196, 197, 215–16, 223–24, 226–27 structure of, 216–24 Taiwan stock exchange, 218 tax incentive program, 223 venture capitalists (VCs), 223–24, 226 Video Graphics Array (VGA), 222 South Korea, 200, 206–7, 219 United States, 202, 203, 204f, 205, 217–18, 222 State role automotive industry, 104–5, 129–34, 285n21 economic development, 4–8, 194–95 innovation, 28–32 electronics industry, 51, 77, 78, 79–80, 83–84, 90, 93–95 intellectual property rights, 31–32

State role innovation (continued) multipolar structure, 31 research & development (R&D), 29–31 production-cost strategies, 11–12 research & development (R&D), 29–31, 138–39, 140–41, 157–58, 160–61, 162t, 164–65 See also Electronics industry value chain; specific institution; State developmental role ST Microelectronics, 144 Stock exchange (Taiwan), 210, 218, 293n10 Stock options (electronics industry), 207–8 Submicron Project, 141, 151 Summit, 216, 218 Sunkist, 134 Sunplus Technology, 157, 201t, 204–5 Surface mount technology (SMT), 55 Suzaki, 267 Suzuki, 267, 268 Swaziland, 12 Switzerland, 24 System on Chip Project, 208–9 Taiwan Auto Body Parts Association (TABPA), 130, 131 Taiwan Isuzu, 107t Taiwan Mask, 79 Taiwan Semiconductor Manufacturing Corporation (TSMC) complementary metal-oxidesemiconductor (CMOS) logic, 79, 80–82, 84t dynamic random access memory (DRAM), 87 economic development, 6 global competition, 9–10 industry co-evolution, 47 industry value chain, 141, 142, 144–45, 151, 152, 153, 156, 157, 160 integrated circuit (IC) foundry services, 47 state developmental policy, 200, 202, 208 Taizhou Yamaha, 267 Tatung, 48 Technoglobalist approach, 138, 140

INDEX 343

Technology Development Program (TDP), 209–10 Technonationalism, 138 Texas Instruments (TI), 83, 87, 88t, 151 Textile/apparel industry. See Production shifts Textron, 118 Thailand, 65, 108, 109t Toppoly, 90t Toshiba active matrix liquid crystal display (AMLCD), 90t dynamic random access memory (DRAM), 87, 88t industry co-evolution, 50, 67–68 Toyota, 35, 103, 106, 134 Trade liberalization Association of Southeast Asian Nations (ASEAN), 119–20 automotive industry, 108, 119–20, 124, 126 economic development, 7–8 electronics industry, 94 global competition, 8 global economy, xiii-xiv See also North American Free Trade Agreement (NAFTA); Production location; Production shifts Trade regulation China, 255, 258, 295n39 Mexico, 166, 167, 170, 188–89, 291n14 See also North American Free Trade Agreement (NAFTA) Transaction cost theory, 38–39, 74 Transportation China, 246, 294n19 global economy, xiii production shifts, 170, 183–84, 291n14 Trend Micro, 24–25, 221, 222 TYC, 117, 118 Ulead, 212–13, 221 United Kingdom, 24 United Microelectronics Corporation (UMC) active matrix liquid crystal display (AMLCD), 90 complementary metal-oxidesemiconductor (CMOS) logic, 79, 82, 84t

United Microelectronics Corporation (UMC) (continued) economic development, 6 global competition, 9–10 industry co-evolution, 47 industry value chain, 141, 142, 144–45, 151, 152, 153, 156, 157, 160 integrated circuit (IC) foundry services, 47 state developmental policy, 200, 202 United States automotive industry aftermarket parts, 100, 110, 111, 112–13, 114–16, 117–18, 125–26, 128, 131, 132, 134, 135 original equipment manufacturer (OEM), 107, 109t, 118, 119, 122, 123, 125–26, 130 complementary metal-oxidesemiconductor (CMOS) logic, 79 electronics industry innovation, 78, 79, 81, 82, 83, 90 electronics industry value chain, 137, 141, 144, 147–48, 149, 150, 155, 157, 165 human capital, 142, 145, 146, 157 integrated circuit (IC) design, 145–46 patents, 138, 152–54 research & development (R&D), 160, 162t wireless technologies, 158–60 global competition, 8 globalization research, xiii nonmilitary aid, 5 patents, 15, 152–54 production location, 24–28 Silicon Valley (California), xvi, 24–25, 54, 56, 58, 72, 75, 157, 159 Taiwan developmental policy, 202, 203, 204f, 205, 217–18, 222 See also Electronics industry coevolution; Production shifts; specific company Universal Studios, 117 Valeo, 118 Value chains. See Automotive industry; Electronics industry value chain Vanguard dynamic random access memory (DRAM), 84, 86, 87, 88t, 94, 284n8

344 INDEX

Vanguard (continued) industry value chain, 141, 151, 153 state developmental policy, 200 Venture capitalists (VCs) integrated circuit (IC) design industry, 210, 293n10 software industry, 223–24, 226 Vertically integrated organization best industrial practices, xv-xvi cross-straits integration, 242–43, 260, 263–64, 267, 270–71, 275–76 electronics industry, 37–40, 49, 67–68, 80 Mexico, 174, 180 Very Large Scale Integration Project, 151 VIA industry value chain, 146, 157, 159 state developmental policy, 201–2, 203 Video Graphics Array (VGA), 222 Vietnam motorcycle industry, 267–68 production-cost strategies, 12 production location, 19, 21 trade liberalization, 108 Viewsonic, 50 Visteon, xv, 118 VLSI Technology, 81 Volkswagen, 118, 123, 128–29 Volume production. See Electronics industry innovation

V-Tech, 166–67, 178t, 189 Walsin Lihwa, 84, 90, 283n5 Whirlpool, 245, 266, 294n15 Winbond complementary metal-oxidesemiconductor (CMOS) logic, 79, 82, 283n1 dynamic random access memory (DRAM), 86, 87, 88t state developmental policy, 200 Wintel, 147 Wireless local area network (WLAN), 158–60, 161 Wireless technologies, 139, 154, 158–60, 161 Wistron, 54, 139 World Bank (2001 survey), 245–47, 251, 294n12 World Trade Organization (WTO), 18, 100, 105, 108, 120, 184, 189, 258, 262 Yamaha, 267, 268, 269 Yameha, 267 Yulon Motor, 103, 106, 107t, 108, 120, 126 Yulon-Nissan, 126 Zoran, 203 ZTE, 265–66