The Dynamics of Local Learning in Global Value Chains: Experiences from East Asia (Ide-Jetro Series)

10.1057/9780230281783 - The Dynamics of Local Learning in Global Value Chains, Edited by Momoko Kawakami and Timothy J. Sturgeon

The Dynamics of Local Learning in Global Value Chains

Other titles from IDE-JETRO: Hiroko Uchimura (editor) MAKING HEALTH SERVICES MORE ACCESSIBLE IN DEVELOPING COUNTRIES Hiroshi Sato and Mayumi Murayama (editors) GLOBALISATION, EMPLOYMENT AND MOBILITY The South Asian Experience Takashi Shiraishi, Tatsufumi Yamagata and Shahid Yusuf (editors) POVERTY, REDUCTION AND BEYOND Development Strategies for Low-Income Countries Mariko Watanabe (editor) RECOVERING FINANCIAL SYSTEMS China and Asian Transition Economies Daisuke Hiratsuka (editor) EAST ASIA’S DE FACTO ECONOMIC INTEGRATION Hisayuki Mitsuo (editor) NEW DEVELOPMENTS OF THE EXCHANGE RATE REGIMES IN DEVELOPING COUNTRIES Tadayoshi Terao and Kenji Otsuka (editors) DEVELOPMENT OF ENVIRONMENTAL POLICY IN JAPAN AND ASIAN COUNTRIES Masahisa Fujita (editor) ECONOMIC INTEGRATION IN ASIA AND INDIA Masahisa Fujita (editor) REGIONAL INTEGRATION IN EAST ASIA From the Viewpoint of Spatial Economics Akifumi Kuchiki and Masatsugu Tsuji (editors) INDUSTRIAL CLUSTERS IN ASIA Analyses of Their Competition and Cooperation Mayumi Murayama (editor) GENDER AND DEVELOPMENT The Japanese Experience in Comparative Perspective Nobuhiro Okamoto and Takeo Ihara (editors) SPATIAL STRUCTURE AND REGIONAL DEVELOPMENT IN CHINA An Interregional Input-Output Approach Akifumi Kuchiki and Masatsugu Tsuji (editors) THE FLOWCHART APPROACH TO INDUSTRIAL CLUSTER POLICY Akifumi Kuchiki and Masatsugu Tsuji (editors) FROM AGGLOMERATION TO INNOVATION Upgrading Industrial Clusters in Emerging Economies Koichi Usami (editor) NON-STANDARD EMPLOYMENT UNDER GLOBALIZATION Flexible Work and Social Security in the Newly Industrializing Countries

The Dynamics of Local Learning in Global Value Chains Experiences from East Asia Edited by

Momoko Kawakami and

Timothy J. Sturgeon

© Institute of Developing Economies (IDE), Jetro 2011 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No portion of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, Saffron House, 6-10 Kirby Street, London EC1N 8TS. Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2011 by PALGRAVE MACMILLAN Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited, registered in England, company number 785998, of Houndmills, Basingstoke, Hampshire RG21 6XS. Palgrave Macmillan in the US is a division of St Martin’s Press LLC, 175 Fifth Avenue, New York, NY 10010. Palgrave Macmillan is the global academic imprint of the above companies and has companies and representatives throughout the world. Palgrave® and Macmillan® are registered trademarks in the United States, the United Kingdom, Europe and other countries. ISBN 978–0–230–23849–7 hardback This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. Logging, pulping and manufacturing processes are expected to conform to the environmental regulations of the country of origin. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data The dynamics of local learning in global value chains : experiences from East Asia / edited by Momoko Kawakami and Timothy J. Sturgeon. p. cm. A joint research project based at Institute of Developing Economies, Japan External Trade Organization (IDE-JETRO). ISBN 978–0–230–23849–7 (alk. paper) 1. Manufacturing industries—Vertical integration—East Asia. 2. Diffusion of innovations—East Asia. 3. Business networks— East Asia. 4. Business logistics—East Asia. 5. International business enterprises—East Asia. I. Kawakami, Momoko. II. Sturgeon, Timothy J. III. Ajia Keizai Kenkyujo (Japan) HD9736.E18D96 2010 338.4095—dc22 2010012001 10 9 8 7 6 5 4 3 2 1 20 19 18 17 16 15 14 13 12 11 Printed and bound in Great Britain by CPI Antony Rowe, Chippenham and Eastbourne

Contents

List of Tables

vii

List of Figures

ix

Preface

xi

Acknowledgments

xiv

List of Contributors

xvi

Value Chain Dynamics and Capability Formation by Latecomer Firms in East Asia Momoko Kawakami 1 Inter-firm Dynamics in Notebook PC Value Chains and the Rise of Taiwanese Original Design Manufacturing Firms Momoko Kawakami 2 Value Chain Creation and Reorganization: The Growth Path of China’s Mobile Phone Handset Industry Ken Imai and Jing Ming Shiu 3 Value Chain Dynamics and Local Suppliers’ Capability Building: An Analysis of the Vietnamese Motorcycle Industry Mai Fujita 4 Local Firms’ Capability Development in Captive Value Chains: Evidence from the Indonesian Motorcycle Industry Yuri Sato 5 To Be or Not to Be a Supplier to TNCs? An Entrepreneurial Approach to Linkage Formation in the Malaysian Electronics Industry Hiroshi Oikawa v

1

16

43

68

100

136

vi

Contents

6 Inter-Country Value Distribution in the East Asian Electronics and Automobile Industries: An Empirical Global Value Chain Approach Hiroshi Oikawa 7 Learning and Earning in Global Value Chains: Lessons in Supplier Competence Building in East Asia Timothy J. Sturgeon and Greg Linden

167

207

Concluding Remarks Momoko Kawakami

227

Index

229

List of Tables

I.1

Key determinants of global value chain governance: The Gereffi, Humphrey, and Sturgeon (2005) framework 1.1 Share of OEM/ODM contracting in the total production of notebook PCs by Taiwanese firms (%) 2.1 China’s mobile phone handset industry—summary figures (million units, %) 3.1 Capability Matrix for analyzing motorcycle components suppliers 3.2 Profile of respondent suppliers 3.3 Classification of learning events by functions of capabilities acquired 4.1 A prototype of the Capability Matrix 4.2 Capability Matrix applied to the modern assembly industry 4.3 Japanese-brand first-tier suppliers by ownership in the Indonesian motorcycle industry (2004–2005) 4.4 Shift in shares of component procurement values from importing to local sourcing by Japanese motorcycle manufacturers in Indonesia (%) 4.5 Measures for reducing costs 4A.1 The 25 target firms in this study: basic information and Capability Matrix (2005) 5.1 Outsourcing structure of Company X (by component and country) 5.2 Number of suppliers of Company X (by component and country) 5.3 Company profiles 5.4 Entrepreneurial orientation score by company 6.1 Performance matrix of East Asian industrialization 6.2 Sector classification and integration: Asian international input–output table 1990, 1995, and 2000 6.3 Index of International Value Distribution vii

5 22 49 72 86 91 104 105 111

112 120 127 149 150 155 157 179 187 193

viii List of Tables

6.4

6.5

7.1

World market share of nine East Asian economies and the USA (electronics and electrical exports) in percentages (%) World market share of nine East Asian economies and the USA (automobile and auto component exports) in percentages (%) Top 5 cellular handset vendors worldwide, 2005 and 2008 (millions of units)

203

204 214

List of Figures

I.1 1.1

Research framework Quantity of Taiwan’s production and exports of notebook PCs 1.2 Worldwide Notebook PC shipments by Taiwanese firms and their share of the world total 1.3 Change in the number of notebook PC manufacturers in Taiwan 1.4 Inter-firm relationships in the late 1990s 1.5 Process of chip development by Intel 1.6 Inter-firm flow of knowledge in the late 1990s 1.7 Sales of major Taiwanese notebook PC manufacturers 1.8 Inter-firm relationships after the 2000s 1.9 Organization of Quanta Computer as of 2006 1.10 Inter-firm flow of knowledge after the 2000s 1.11 Ratio of gross margin to net sales: comparisons of Intel, Dell, and Quanta 1.12 ROA of major Taiwanese notebook PC manufacturers 2.1 Conceptual Model of Value Chain of the Mobile Phone Handset Industry 2.2 The estimated shares of local brands and foreign brands 2.3 Patterns of value chains (1)—late 1990s to circa 2003 2.4 Market shares of brand manufacturers in October 2008 2.5 Patterns of value chains (2)—circa 2003 to 2004 2.6 Patterns of value chains (3)—2005 to the present 3.1 Growth of Vietnam’s motorcycle market 3.2 Sales of motorcycles by manufacturers 3.3 HVN’s Local Sourcing 3.4 Production of motorcycles by size of production 3.5 Chain participation of respondent suppliers 3.6 Levels of capabilities reached by respondent firms in 2008 4.1 Production of motorcycles in Indonesia 4.2 Labor productivity and profitability of motorcycle component producers 4.3 Composition of Indonesian motorcycle market by brand ix

6 21 21 22 24 25 27 29 31 32 33 35 36 46 50 54 56 58 62 76 77 81 82 87 89 108 108 109

x

List of Figures

4.4 4.5 4.6 5.1 5.2 5.3

5.4 5.5 5.6

Assessing the capabilities of Indonesian local firms in the Japanese captive value chains Die and mold making and human resource allocation D Group as an example of brand-specific dedicated factories Total FDI inflows into major developing ASEAN Countries (1988–2007) Origins of purchases by Japanese manufacturing affiliates in major ASEAN countries Ownership structure of local companies purchased by Japanese manufacturing affiliates in major ASEAN countries Timing of supplier selection and involvement in the flow of product design and production Years of business of incumbent suppliers with Company X Average EO scores by supplier group

115 118 122 141 143

143 151 153 158

Preface

International trade and foreign direct investment have long been central features of the world economy, but their importance has been growing rapidly, especially since the late 1980s. While the scale of this increase is easy to see in international trade statistics, a qualitative change has also been unfolding, one that is more difficult to discern. Alongside the well-documented flow of goods in the global economy are streams of services, information, and knowledge that are very poorly documented, if they are documented at all. Official statistics reveal very little about the volume and character of these invisible flows and even less about the firm-level power asymmetries and ownership patterns that structure them and, in many ways, determine opportunities for technological learning and the distribution of gains. Nevertheless, the density of information and knowledge transfer in the world economy has increased at least as quickly as the flow of goods. Information technology and plummeting prices for international communications enable this transformation, along with the accelerated flow of people and the leveling effects of mass media, which drive the scale and homogeneity of product markets. Researchers studying this structural shift in the global economy have generated a very long list of terms to describe it, including new international divisions of labor, integrative trade, inter-industry trade, fragmentation, global production sharing, multistage production, slicing up the value chain, disintegration of production, and offshore outsourcing. The enduring structures that contain and structure these new forms of trade and investment have been referred to as global commodity chains, global production networks, and, most recently, global value chains. While the cross-border flow of intermediate goods and foreign direct investment are the primary statistical evidence for these new structures, qualitative research methods have revealed that the cross-border inter-firm relationships involved come with varying degrees of power asymmetry, requirements for explicit coordination, and potential for knowledge transfer. These qualitative features of trade, though difficult to define and measure, are important objects of xi

xii Preface

study because they help to determine how profits are divided in the global economy, and increasingly, how technological learning and economic development happen in both the developed and the developing world. Because ownership and control of economic assets are hard to discern in international statistics, and the divisions of knowledge and power in the global economy are unequal, we cannot automatically assume that countries which export large quantities of goods or services have domestic companies with the full competencies required for their development and production. We cannot even assume the country has the supply base needed to provide needed inputs, or the know-how to build and operate the infrastructure required. Factories can be owned and controlled from a great distance, products can be designed and developed in one country and assembled in another, largely from imported inputs, multinational firms can build, own, and operate port facilities, and financing can be drawn from global capital markets. Even with production spread across the globe, the world is far from being “flat.” Access to technologies can be controlled through intellectual property rights protection, careful knowledge partitioning, or by embedding capabilities in machinery, software, or advanced components. To judge where developing countries truly stand in terms of economic development, we must look beyond industrial output and export statistics to the less visible, but critically important, structures and processes that underlie them. The research collected in this book does exactly that. The central question asked here is straightforward, “How can local firms in developing countries increase their capabilities by participating in global value chains (GVCs)?” The answer, however, is not so simple. The specificities of industries, regulations and standards, technologies, policies, and regions within countries all structure the ways that learning occurs in GVCs. Since learning is cumulative and relational, outcomes hinge to a large degree on starting points, the length and intensity of global engagement, and on the strategies and practices of lead firms, suppliers, and other actors in the chain, including international and government agencies. While it is clear that the capabilities of firms in developing countries have increased and that this learning has happened in a variety of ways, we suspect that there must be generalities and best practices to be uncovered. We simply know very little about the process. This book helps to fill that gap by examining how the dynamic interactions between firms from developed and developing

Preface

xiii

economies, together with other factors, have stimulated and shaped the rise of local firm capabilities in East Asia. We see how and why local firms in late-industrializing countries have sometimes seized and sometimes been excluded from the learning opportunities offered by global integration. Timothy J. Sturgeon, Cambridge, Massachusetts, February 8, 2010

Acknowledgments

This book is the product of a two-year international joint research project based at Institute of Developing Economies, Japan External Trade Organization (IDE-JETRO). We owe a debt of gratitude to the many individuals who kindly shared their insights and experiences with us in the course of our field research. Without their generosity, this study would not have been possible. Many scholars from across Japan provided us with inspiration, insight, and opportunity for reflection. We would like to thank Tomoo Marukawa, Yumiko Nakahara, Koichi Ogawa, Hirofumi Tatsumoto, and Masanori Yasumoto for enriching our project as guest speakers at our study meetings. We also benefited greatly from the rich intellectual resources within IDE-JETRO. We are grateful to Takahiro Fukunishi for his valuable input to our research. As a specialist on African manufacturing sectors, he provided us with rich insights and crucial points of comparison that enabled us to place our East Asian cases in a broader context. Our discussions throughout the study with Makoto Abe, Ke Ding, Koichiro Kimura, Yoichi Koike, and Yukihito Sato were especially useful. Special thanks go to the administrative staff of IDE-JETRO, especially to Kyoko Miura and Ritsuko Takakusagi, for their dedication, generous support, and encouragement. We note with deep sadness the tragic passing of two colleagues who participated in this research project. The late Dr Seishi Kimura was the co-organizer of the project. He provided us with our core ideas, intellectual framework, and working hypotheses. His enthusiasm for testing GVC theory with in-depth field studies in East Asia was the main impetus and driving force for our study. Since his passing, Seishi’s influence has not waned, and this has helped to sustain us. We have carried on as best we could without him. Even with its faults, this volume stands as a testament to Seishi’s innovativeness, energy, vision, and charisma. He is deeply and constantly missed. Ken Imai passed away shortly after finishing his co-authored chapter. Ken was the most dedicated of researchers. He brought a historian’s eye for detail and accuracy to his field research and a craft-person’s sense of care to his writing, and these high standards served as an inspiration to all of his colleagues. All the authors enjoyed and deeply benefited from xiv

Acknowledgments xv

working with him. Because of his generosity, Ken’s keen insights are present in all of the chapters, and this has carried us forward in spite of his very large absence. Ken’s dedication to scholarship, warm friendship, and generous spirit will always remain in our memories. This book is dedicated to the memory of Seishi Kimura and Ken Imai. Momoko Kawakami and Timothy J. Sturgeon

List of Contributors

Mai Fujita Research Fellow, Institute of Developing Economies, IDE-JETRO, Japan Ken Imai Senior Research Fellow, Institute of Developing Economies, IDE-JETRO, Japan Momoko Kawakami Deputy Director of the Technological Innovation and Economic Growth Studies Group, Interdisciplinary Studies Center, IDE-JETRO, Japan Greg Linden Research Associate, Institute for Business Innovation, UC Berkeley, USA Hiroshi Oikawa Professor, University of Nagasaki (Siebold Campus), Japan Yuri Sato Senior Research Fellow, Institute of Developing Economies, IDE-JETRO, Japan Jing Ming Shiu Doctoral Candidate, Graduate School of Economics, the University of Tokyo, Japan Timothy J. Sturgeon Senior Research Affiliate, Industrial Performance Center, Massachusetts Institute of Technology, USA

xvi

Value Chain Dynamics and Capability Formation by Latecomer Firms in East Asia Momoko Kawakami

I.1. Purpose of the study The transformation of East Asian economic development by globalization has gathered speed since the 1990s. The expansion of foreign direct investment (FDI), the increase in international trade, and the accelerated flow of financial and human resources across country borders have triggered rapid industrialization and economic integration. A large part of this dynamism is rooted in the cross-border production networks that have emerged from the relocation and reorganization strategies of firms based in developed economies (Borrus et al. 2000; Yusuf 2004). These firms began by exploiting the low production costs and favorable investment climate in East Asia, but were soon tapping into the industrial capabilities of local manufacturers as well. As more countries were drawn into these networks over time, a complex warp and weft of value-adding activities emerged that cut across country borders. These new patterns of cross-border investment, sourcing, and contracting, which we refer to as global value chains (GVCs), have opened up important opportunities for local firms in East Asia. The insertion of local firms into GVCs has boosted exports, accelerated the growth of the manufacturing sector, served as a catalyst for knowledge diffusion, and provided new opportunities for learning and capability formation in local firms (Ernst 2002; Ernst and Kim 2002). Leveraging the linkages and interactions with firms from developed economies, local manufacturers in East Asia have, in some cases, emerged as competitive suppliers in a variety of industries, not only apparel, footwear, and other simple consumer and household goods, but more technology- and capital-intensive products such as electronics, auto parts, and textiles. 1

2

Introduction

On the other hand, GVCs can create barriers to learning and drive uneven development over time, even as they trigger rapid industrial development and upgrading. There is considerable evidence that greater rents accrue to those “lead firms” in the chain that control branding and product conception, on one hand, and to the “platform leaders” that provide core technologies and advanced components, on the other. Firms that provide routine assembly tasks and other simple services earn less, pay their workers less, and are more vulnerable to business cycles because they tend to have large-scale employment and fixed capital. If one of these categories of firms tends to dominate a specific country or region, then consequences for economic performance and social welfare can be profound and persistent, shaping the business systems of entire economies over extended periods. Specifically, entrenchment in narrow, routine, low value-added activities can lock firms and national industries into poor performing segments of GVCs. Learning might be rapid at first, but over time limits can be acutely felt. This book investigates the process of capability development in East Asia. We explore how the dynamic interactions between firms from developed and developing economies, together with a variety of other factors, have stimulated, shaped, and sometimes limited the rise of local firm capabilities in East Asia. We also explore how the local firms in these late-industrializing countries, what we term “latecomer firms,” have overcome barriers and seized the learning opportunities offered by global integration in technology-intensive sectors as electronics and motor vehicles. The research presented in this volume employs the GVC perspective as a conceptual and theoretical starting point. This perspective provides powerful tools for disentangling the “complicated skein of cross-border relationships that have evolved out of firm strategies” (Kenney 2004: 1). We extend the GVC perspective, and link it to complementary theoretical traditions, and provide an enriched analytical framework to explain the development of firm capabilities within GVCs. This introductory chapter is organized as follows. The next section reviews recent developments in the GVC perspective. The middle section presents the framework used to explain the rise in the capabilities of latecomer firms. The final section summarizes the major arguments developed in the volume.

I.2. Global value chains perspective An inter-disciplinary group of scholars formalized the GVC approach just as the twenty-first century was beginning (Gereffi and Kaplinsky

Momoko Kawakami

3

2001), but many of the underlying research motives, analytical concepts, and framework tools originated in the direct predecessor of these studies: the global commodity chain (GCC) perspective (Gereffi 1999; Gereffi and Korzeniewicz 1990, 1994).1 The GCC perspective focuses on the inter-organizational networks that cluster around one commodity or product and connect various actors in the global economy (Gereffi et al. 1994: 2). The perspective highlights four dimensions of global industries: (1) the input–output structure—a set of products and services linked together in a sequence of value-adding activities; (2) the territoriality of the networks—such as spatial patterns of dispersion or concentration; (3) the governance structure of the networks—including the authority and power relationships that determine how financial, material, and human resources are allocated and flow within a production chain; and (4) institutions— local, national, and international conditions and policies that shape the globalization process (Gereffi 1994: 96–97; 1995: 113). Based on case studies, Gereffi identified two major patterns of chain governance: (1) producer-driven commodity chains, where large manufacturers control the production system; and (2) buyer-driven commodity chains, where large-retailers, trading companies, and others organize and control the network (Gereffi 1994, 1999). Based on the criticism of the industry-dependent and static view of governance provided by the GCC perspective, the GVC framework refined the analysis of governance and changed the overarching description of cross-border networks to “global value chains” to overcome the association of the word “commodity” with undifferentiated products and focus on the fragmenting process of value creation.2 GVCs were defined as the sequence of value-adding activities, from conception to end use, after-sales service, and recycling, of a particular good or service, across country borders (Sturgeon 2001).3 GVCs therefore encompass the entire productive side of goods and services, from product conception and design to production to marketing and after-sales services. GVC studies have investigated the microfoundation of economic globalization by examining the inter-linked structure of value-added activities. The focus is on how these activities are organized and coordinated along value chains for specific products, industries, and locations. Because GVC analysis insists that the whole value chain be taken into account, it directs our attention, not only to those activities that take place in a particular location, but also linked activities that take place in other locations. As more economic activities are fragmented and spread across country borders, GVC analysis

4

Introduction

has become increasingly relevant and popular among researchers and economic development agencies. Empirical studies using the GCC/GVC frameworks have revealed that the inter-firm relationships along value-adding chains are coordinated not only through market mechanisms, but, in many cases, through the actions of especially powerful “lead firms” within the network, actions that have so far tended to emanate from developed countries. GVC research has examined how knowledge is transmitted across country borders, how it can be partitioned within developed countries, and how these GVC dynamics affect local suppliers in developing economies (Gereffi and Memedovic 2003; Humphrey and Schmitz 2004a). To coordinate economic activities that take place in their chains, lead firms define products, set and enforce parameters for production, decide on suppliers, and set the requirements for the quantity, quality, and the timing of delivery to which firms in its supply chain must adhere. The power of lead firms means that they, in many ways, determine how GVCs are governed and how value-added and profits are distributed along the chain. Among a number of studies in this line of research, we consider the GVC governance framework offered by Gereffi et al. (2005) to be a key contribution and ideal starting point for developing our own framework. Their paper identifies five types of GVC governance, that is, market, modular, relational, captive, and hierarchy,4 and argues that the character of chain governance is, in part, determined by three explanatory variables—the complexity of transactions, the ability to codify transactions, and the capabilities in the supply base (Table I.1). This simple, generic (in regard to industry), and formal theoretical framework has helped us to understand the opportunities and limits to supplier upgrading observed in our research. At the same time, the sharp focus on the determinants of chain governance and the assumption that lead firm strategies are responsible for observed patterns of chain governance has shifted the main research agenda away from the strategies and growth dynamics of firms based in late-industrializing countries.5 The work of Gereffi et al. (2005) has enhanced the applicability of the GVC framework, but at the expense of an explicit focus on learning dynamics in developing economies, a focus that was central to earlier GCC studies (Gereffi et al. 1994; Gereffi 1995). Clearly, a GVC-based framework that places the growth of latecomer firms at the center of the analysis is needed.6 As a part of our effort to create such a framework, this study adapts the GVC governance framework to better explain the rise of supplier capabilities within the context of

Momoko Kawakami

5

Table I.1 Key determinants of global value chain governance: The Gereffi, Humphrey, and Sturgeon (2005) framework Governance type

Complexity of transactions

Ability to codify transactions

Capabilities in the supply-base

Degree or explicit coordination and power asymmetry

Market Modular Relational Captive Hierarchy

Low High High High High

High High Low High Low

High High High Low Low

Low  ⏐  High

Note: As for the exclusion of three other combinations of variables, see footnote to the Table 1 of Gereffi, Humphrey, and Sturgeon (2005). Source: Gereffi, Humphrey, and Sturgeon (2005), Table 1, p. 87.

industry-specific GVCs.7 By so doing, we seek to answer the following questions: “How do the interactions among lead firms and suppliers shape the capabilities of local firms in GVCs?” “How does the organizational configuration of GVCs affect the growth of latecomer firms?” And, because GVCs are dynamic, co-evolutionary systems, we also ask, “How does the emergence of competitive suppliers from developing countries alter GVC structure?” The next section presents our research framework, developed through intensive interactions among the case study authors in this volume.

I.3. Framework of the study: value chain dynamics Our research framework is designed to help explain the rise of capabilities in latecomer firms in GVCs. Drawing on the work by Foss (1996), Christensen (1996), Teece and his colleagues (Teece and Pisano 1994; Teece et al. 1997), as well as the technological capabilities literature (Lall 1992), we define “capabilities” broadly to reflect the context of developing economies. Specifically, we refer to “capabilities” as intangible, firm-specific, locally situated, and experience-based knowledge assets. We specifically include the ability to generate and manage technological change and the management capabilities required to structure and integrate the firm’s bundle of skills and resources in ways that adapt to external changes. Based on our findings in East Asia, we argue that the rise of firm capabilities in GVCs is determined by the interactions between two sets of

6 Introduction

firm strategies—learning strategies of latecomer firms from developing economies and outsourcing strategies of lead firms from developed countries. These two strategies are in turn subject to two causal factors: the technological attributes of the concerned transactions and the institutional environment and local setting surrounding the firms (Figure I.1). We find that value chain governance is not only determined by different combinations of the three variables highlighted by Gereffi et al. (2005), but also by a process where two variables in the GVC governance framework—the complexity and codifiability of the transaction—as well as the institutional environment and local settings of the concerned economies, jointly mold firm strategies and govern the rise of firm capabilities. In other words, supplier capabilities are not given, but arise from the interaction of buyer–supplier strategies as shaped by institutional and local settings and the characteristics of the other two variables in the GVC framework. This co-evolutionary process is ultimately reflected in specific patterns of chain governance. In other words, in our model, we have shifted supplier capabilities from being an independent variable, as it is in Gereffi et al. (2005), into a new position as a dependant variable. The GVC governance framework takes supplier capabilities, at any given time, as given. Our framework opens the black box of supplier capabilities. For us, the forces that combine to create supplier capabilities are our central object of study.

Learning strategies of local suppliers

Attributes of a transaction (complexity, codifiability)

Supplier capabilities Outsourcing strategies of lead firms

Figure I.1

Institutional environment and local settings

Research framework

Note: Variables in shaded boxes correspond to the three key determinants of global value chain governance framework presented by Gereffi, Humphrey, and Sturgeon (2005). Source: Prepared based on the discussions of the case study authors.

Momoko Kawakami

7

I.3.1. Firm strategies Gibbon et al. (2008) note that Gereffi et al. (2005) privilege structural variables such as the complexity and codifiability of individual transactions as determinants of GVC governance rather than the intentional, strategic actions of firms. In our framework, the strategies of firms are deemed to be of critical importance in shaping the development of supplier capabilities. Specifically, we identify two sets of strategies that shape the trajectory of capability building in latecomer firms: local supplier learning strategies and lead firm outsourcing strategies. First, strategic and purposeful investment in learning by latecomer firms is by far the most important factor in shaping the learning process of these firms. It is only when latecomer firms seek to tap into the learning opportunities embedded in GVCs, and make substantial investments in absorbing technologies and knowledge and consolidating the acquired resources, that these firms have a chance to successfully raise their capabilities. Second, lead firm outsourcing strategies are also critical. When latecomer firms operate in GVCs driven by powerful and technologically competent lead firms from developed countries, their upgrading prospects can be enabled or constrained by lead firm outsourcing strategies, which go a long way toward determining the quantity and scope of activities to be carried out by latecomer firms. The exchange of knowledge between lead firms and suppliers is structured, not only by the technical attributes of the transaction in terms of complexity and codifiability, but also by the willingness of lead firms to share information and knowledge with their suppliers. Even more fundamentally, a supplier’s initial conception of the market, of production processes, and of technological resources might be a function of what is asked of them by lead firms. In this context, a dynamic alignment of a supplier’s capability development strategies with the evolving outsourcing strategies of its customers is the most critical determinant of successful capability formation in local firms. It is exactly what Kimura (2007: 97–98) calls “dynamic external fit,” in which, he argues, a latecomer firm operating as a supplier in GVCs identifies the domain of activities that align with the sourcing strategies of lead firms and focus on the activities that create a dynamic platform for learning and upgrading. It is only when suppliers successfully align their learning strategies to the outsourcing strategies of lead firms in this way that they effectively tap into the learning opportunities afforded by GVCs. Our central finding is that supplier learning is not unilaterally constrained by the lead firm strategies. In some instances, by

8

Introduction

their own initiatives and strategies, latecomer firms can break through many of the constraints posed by the lead firm outsourcing strategies. The examples illustrated in our case studies (Chapters 1, 2, and 3) show that some latecomer suppliers have successfully leveraged their learning opportunities by engaging in multiple chains or by trading with multiple customers, especially those in the local market.

I.3.2. Attributes of transactions, institutional environment, and local settings: factors shaping firm strategies Our framework emphasizes the key role of firm strategies in driving capability formation by latecomer firms. Nevertheless, firms are faced with structural constraints in formulating their strategies. Both outsourcing strategies and learning strategies are shaped by a number of factors that surround the firms in the chains; among these, we identify two variables as the critical contributing factors: (1) the attributes of the concerned transactions, and (2) the institutional environment and local setting surrounding the firms (see Figure I.1). First, two variables that are identified by Gereffi et al. (2005) as the explanatory variables of chain governance typology—the complexity and ability to codify the transaction in question—shape the outsourcing strategies of lead firms and confine the learning strategies that latecomer suppliers can choose. If the concerned transaction is highly complex while its codifiability is low, then outsourcing firms have little choice but to engage with their suppliers through relational linkages or internalize the linkages within boundaries of the corporate hierarchy. This eliminates the feasibility of learning strategies that focus on modular linkages. Thus, we argue that complexity of transactions and the ability to codify them determine the character and scope of strategies that both outsourcers and suppliers can pursue. Second, firm strategies of both lead firms and suppliers are influenced by institutional environments and other specificities of local settings. Institutions and local norms are especially important in the context of late-industrializing countries. Inefficient social infrastructure, inefficient market exchange, corruption, and low levels of trust may result in prohibitively high transaction costs. In such an institutional environment, lead firms find it extremely difficult to outsource to local firms, even if the codifiability of transactions is high and the capabilities of the local supply base are high enough for the job concerned.8 Also, when a small number of large-scale suppliers dominate the local market, or when the number of suppliers is too small

Momoko Kawakami

9

due to inactive entrepreneurship and an unfavorable investment climate, lead firms may take a cautious approach to expanding outsourcing due to the fear of hold-up effects associated with small supplier numbers, on one hand, or the time and effort required to build up a local supply base from scratch, on the other. This is why lead firm facilities tend to be more vertically integrated abroad than at home, and why they often encourage home-based suppliers to locate abroad with them. Both strategies limit the opportunities for local suppliers. Thus, we posit that the attributes of the transaction together with institutional environment influence firm strategies for both suppliers and lead firms.

I.3.3. Strength of the framework The framework proposed in this chapter helps us to understand how the structural constraints surrounding firms and the intentional actions of firms interact with each other to shape the growth of latecomer firms, and lead to the emergence of specific patterns of GVC-engaged development. Our framework postulates an evolutionary process where the technological attributes of transactions, institutional environment, and firm strategies jointly govern the rise of latecomer firm capabilities, and eventually lead to specific economic development outcomes. Our dynamic and multidimensional framework also highlights several aspects of value chain dynamics that have been insufficiently explored. First, our framework explains why firms operating in the same chain can perform differently in terms of capability formation. Over the course of our field studies, we have observed striking variation across firms in terms of learning attainment, even though the firms face the same transactional characteristics related to complexity and codifiability. Incorporating firm strategy into the analysis reveals one underlying mechanism behind this heterogeneity: the learning strategies adopted by suppliers are the key factor. Simple mimicking of successful strategies may not necessarily work, since lead firm outsourcing strategies may limit the number of participating firms to early movers, and because the appropriate learning strategies differ depending on firms’ resources, prior experiences, and market position. The key question is whether or not a supplier can formulate and pursue strategies that exploit learning opportunities embedded in GVCs as they arise. Second, our present framework allows us to analyze the three major topics of GVC analysis as identified by Sturgeon (2009): governance, power, and institutions.9 While the latter two realms of theoretical inquiry are relatively underexplored in the existing literature, our

10

Introduction

framework incorporates “institutions” explicitly as a variable in our conceptual model. Also, using our framework, we can explore “power in chain” by investigating how power affects the interactions between the trading parties. As we will see in the following chapters, the power of lead firms to determine the allocation of resources within its supply chain is often challenged by local assemblers and powerful component suppliers seeking to gain a larger portion of value-added created by the industry. By focusing on the interaction of firm strategies, our framework leads directly to an examination of the power dynamics in the GVCs.

I.4. The chapters This volume contains seven chapters. The first five present the results of the authors’ in-depth firm-level field studies on East Asian industrial dynamism. These in-depth field studies share the research framework described above and rely to a great extent on qualitative field research methodologies. To investigate the mechanisms underlying the development of firm capabilities, the authors carried out intensive firm visits to examine changes in technologies, firm strategies, and industrial organization in selected manufacturing sectors. We selected two very different manufacturing sectors as subjects of our case studies: the electronics and motorcycle industries. In these two sectors, we can observe the emergence of GVCs, the incorporation of a wide variety of local firms into the cross-border inter-firm division of labor, and the rise of local firms as (more or less) competitive manufacturers. More importantly, these two industries offer cases that illustrate different GVC governance criteria, and hence the divergent outcomes that GVCs can create for local firms in East Asia. Our case study research was carried out in Taiwan, China, Vietnam, Indonesia, and Malaysia. This wide geographic coverage allowed us to observe GVC engagement in different institutional settings, and observe the variety of challenges that local firms can face in pursuing GVC-engaged capability development. Chapter 1 explores how Taiwanese notebook personal computer (PC) suppliers raised their capabilities in GVCs. The chapter investigates the inter-firm relationships among major actors in GVCs, the knowledge flow among them, and the strategies Taiwanese manufacturers used to exploit the learning opportunities offered by the changing strategies of lead firms and platform leaders. By so doing, the chapter reveals how changes in the technological attributes of transactions helped to define the strategies of lead firms and contract manufacturers. The chapter also

Momoko Kawakami

11

demonstrates the critical significance of latecomer firms’ learning strategies in creating and consolidating supplier capabilities that are costly and time-consuming to replicate. Chapter 2 explores one of the most dynamic manufacturing sectors in contemporary East Asia: the Chinese mobile phone handset industry. It investigates how technological changes and the trend toward modularization have affected the value chain dynamics of the industry. It also reveals how the local firms have aggressively and nimbly pursued business opportunities that have opened in the local market. The chapter points out the co-existence of various types of value chains, those led by foreign and local lead firms, and illustrates how this co-existence has created additional space for capability enhancement by local firms. Chapter 3 examines the Vietnamese motorcycle industry. It explores how the interactions between the two different sets of value chains— those coordinated by Japanese firms and those coordinated by Vietnamese assemblers in cooperation with Chinese firms—created different learning opportunities for local suppliers. It also shows how the local suppliers exploited, and sometimes mixed, these new learning opportunities. Employing an original capability matrix methodology created jointly by the authors of Chapters 3 and 4, the chapter reveals how the capability levels of local suppliers have changed over time as they leveraged linkages with different types of lead firms to build technological capabilities. Chapter 3, again, highlights the importance of local firms’ learning strategies in driving and structuring their own growth. Chapter 4 investigates capability formation in local Indonesian motorcycle component suppliers operating in the captive value chains led by powerful brand-carrying lead firms. Using the capability matrix methodology, the author documents the attainment of capabilities by local suppliers. The findings indicate that the local suppliers’ capabilities have reached a plateau in the product area that lead firms place the highest priority on, and that process capability attainment varied across firms. The research also found that supplier capability formation in captive value chains is especially hindered in the pre-production and post-production stages, inhibiting “functional” upgrading either upstream or downstream in the value chain. Finally, the chapter documents suppliers’ efforts to formulate and pursue their own strategies in regard to capability formation. Chapter 5 highlights local entrepreneurs’ responses to the presence of global firms. While the dominant role of lead firms in GVC formation has been emphasized in the GVC literature, the role of supplier firms

12

Introduction

has been explored in less depth. To help fill this gap, the chapter examines the role of entrepreneurship in the formation of linkages between transnational corporations (TNCs) and local companies in Malaysia. In the chapter, the Entrepreneurial Orientation (EO) method is applied to compare two supplier groups: suppliers that do business with TNCs and suppliers that do not. The results show clear evidence that suppliers to TNCs tend to be more engaged in technological learning, while non-TNC suppliers generally seek more market opportunities and managerial autonomy than TNC suppliers. In this tradeoff, we see again the importance of strategy. The chapter’s findings also suggest that the institutional environment helps to structure local entrepreneurs’ perceptions and responses to the opportunities provided by TNCs. Chapter 6 differs from the other chapters in that it relies on quantitative, rather than qualitative, evidence. The chapter presents an empirical investigation of the international distribution of value-added in East Asian electronics and automobile industries. Value distribution among trading partners participating in GVCs has been a central, yet underexplored, subject of GVC studies. Using Asian international input–output tables, the chapter calculates the flow of value-added through eight major East Asian economies. The results clearly show evidence of an uneven distribution of gains among these economies. The analysis also indicates that countries where industrialization processes depend on the presence of TNCs exhibit relatively poor performance when it comes to capturing value-added. The findings of the chapter suggest the critical roles played by local firms in capturing gains from economic integration. This also supports the proposition that the rise of local firms’ capabilities matter for the economic development in late-industrializing economies. Chapter 7 revisits the major research findings of this volume, and places the contributions of this volume in the context of GVC theory. The chapter identifies six major findings derived from the empirical studies on East Asian industrial sectors. Firm strategies matter; not only those of lead firms and platform leaders but those of suppliers as well. Standards matter, since they provide the context for the cases and help to determine how GVCs are configured and profits are distributed. Firm nationality matters because it structures the opportunities for supplier learning. What’s most evident from the case studies is the variety of strategies suppliers employ when confronted by the often-severe barriers to learning and earning in GVCs. This analysis constitutes the core empirical and theoretical contribution of the volume.

Momoko Kawakami

13

Notes 1. For a literature survey on the traditions of chain research see, for example, Raikes et al. (2000), Bair (2005, 2009), and Gibbon and Ponte (2005, chapter 3). 2. For the interconnections between the GCC and GVC perspectives, see Bair (2005) and Sturgeon (2009). Besides the terminology in conceptualizing “chains,” the GVC perspective breaks from the GCC perspective in its sharp focus on the classification of chain governance typology and the investigation of their determinants. 3. For an overview of the “chain” concept originating with Porter (1985) and elaborated by GCC and GVC studies, see Henderson et al. (2002). 4. Modular, relational, and captive chains are the three network-type categories that lie between the two extreme types of chain governance—markets and hierarchies. Gereffi et al. (2005) and Sturgeon (2009) explain the three typologies as follows. In modular chains, complex information regarding the transaction is codified and passed to competent suppliers. Transactions between “turnkey service” suppliers and their customers are the typical cases of this type of chains. In relational value chains, tacit information is exchanged between buyers and competent suppliers. Mutual dependence and high asset specificity created by complex interactions between the trading parties characterize chains of this type. In captive chains, less competent suppliers engage in transaction under tight monitoring and control by lead firms. 5. As a descendant school of the World Systems approach, the GCC perspective is seriously concerned with the source of power asymmetry between lead firms and suppliers, as well as possible trajectories for upgrading local suppliers in the chains (Gereffi 1994, 1999; Gereffi et al. 1994). 6. A number of studies attempt to combine research on industrial clusters and value chain studies to explore the industrial upgrading in developing economies. See Pietrobelli and Rabellotti (2006) and Schmitz (2004). 7. Humphrey and Schmitz (2004b) share with this volume the same strong interest in the impacts of different types of value chains on local enterprises. By introducing typologies of chain governance and upgrading, they investigate how the upgrading opportunities of local firms are structured by chain governance. Meanwhile, in this volume we stress the critical importance of firm strategies, and explain the variety of learning performances across the suppliers operating in the same type of value chains by the differences in learning strategies. 8. Or they may bring in the suppliers from their home country. This also discourages the emergence of modular value chains that involve local firms as suppliers. 9. Also see Chapter 7 of this volume.

References Bair, Jennifer. 2005. “Global Capitalism and Commodity Chains: Looking Back, Going Forward.” Competition & Change 9(2), June: 153–180. Bair, Jennifer. 2009. “Global Commodity Chains: Genealogy and Review.” In Frontiers of Commodity Chain Research, ed. Jennifer Bair. Stanford: Stanford University Press: 1–34.

14

Introduction

Borrus, Michael, Dieter Ernst, and Stephan Haggard, eds. 2000. International Production Networks in Asia: Rivalry or Riches? London and New York: Routledge. Christensen, Jens Frøslev. 1996. “Analysing the Technology Base of the Firm: A Multi-dimensional Resource and Competitive Perspective.” In Towards a Competence Theory of the Firm, ed. Nicolai J. Foss and Christian Knudsen. London and New York: Routledge: 111–132. Ernst, Dieter. 2002. “Global Production Networks and the Changing Geography of Innovation Systems: Implications for Developing Countries.” Economics of Innovation and New Technologies 11(6): 497–523. Ernst, Dieter and Linsu Kim. 2002. “Global Production Networks, Knowledge Diffusion, and Local Capability Formation.” Research Policy 31(8–9): 1417–1429. Foss, Nicolai J. 1996. “Introduction: The Emerging Competence Perspective.” In Towards a Competence Theory of the Firm, ed. Nicolai J. Foss and Christian Knudsen. London and New York: Routledge: 1–12. Gereffi, Gary. 1994. “The Organization of Buyer-Driven Global Commodity Chains: How U.S. Retailers Shape Overseas Production Networks.” In Gary Gereffi and Miguel Korzeniewicz, eds. Westport: Praeger: 95–122. Gereffi, Gary. 1995. “Global Production Systems and Third World Development.” In Global Change, Regional Response: The New International Context of Development, ed. Barbara Stallings. New York: Cambridge University Press: 100–142. Gereffi, Gary. 1999. “International Trade and Industrial Upgrading in the Apparel Commodity Chain.” Journal of International Economics 48(1): 37–70. Gereffi, Gary and Miguel Korzeniewicz. 1990. “Commodity Chains and Footwear Exports in the Semiperiphery.” In Semiperipheral States in the World-Economy, William G. Martin ed. New York: Greenwood Press: 45–68. Gereffi, Gary and Miguel Korzeniewicz, eds. 1994. Commodity Chains and Global Capitalism. Westport: Praeger. Gereffi, Gary and Olga Memedovic. 2003. The Global Apparel Value Chain: What Prospects for Upgrading by Developing Countries. Vienna: United Nations Industrial Development Organization. Gereffi, Gary, Miguel Korzeniewicz, and Roberto Korzeniewicz. 1994. “Introduction: Global Commodity Chains.” In Gary Gereffi and Miguel Korzeniewicz eds.: 1–14. Gereffi, Gary, John Humphrey, and Timothy Sturgeon. 2005. “The Governance of Global Value Chains.” Review of International Political Economy 12(1), February: 78–104. Gereffi, Gary and Raphael Kaplinsky, eds. 2001. “The Value of Value Chains: Spreading the Gains from Globalisation.” IDS Bulletin 32(3), special issue. Gibbon, Peter and Stefano Ponte. 2005. Trading Down: Africa, Value Chains, and the Global Economy. Philadelphia: Temple University Press. Gibbon, Peter, Jennifer Bair and Stefano Ponte. 2008. “Governing Global Value Chains: an Introduction.” Economy and Society 37(3): 315–338. Henderson, Jeffrey, Peter Dicken, Martin Hess, Neil Coe, and Henry Wai-Chung Yeung. 2002. “Global Production Networks and the Analysis of Economic Development.” Review of International Political Economy 9(3): 436–464. Humphrey, John and Hubert Schmitz. 2004a. “Governance in Global Value Chains.” In Hubert Schmitz ed.: 95–109.

Momoko Kawakami

15

Humphrey, John and Hubert Schmitz. 2004b. “Chain Governance and Upgrading: Taking Stock.” In Hubert Schmitz ed.: 349–381. Kenney, Martin. 2004. “Introduction.” In Locating Global Advantage: Industry Dynamics in the International Economy, ed. Martin Kenney with Richard Florida. Stanford: Stanford University Press: 1–20. Kimura, Seishi. 2007. The Challenges of Late Industrialization: The Global Economy and the Japanese Commercial Aircraft Industry. Basingstoke and New York: Palgrave Macmillan. Lall, Sanjaya. 1992. “Technological Capabilities and Industrialization.” World Development 20(2): 165–186. Pietrobelli, Carlo and Roberta Rabellotti, eds. 2006. Upgrading to Compete: Global Value Chains, Clusters, and SMEs in Latin America. Washington, DC: Inter-American Development Bank. Porter, Michael E. 1985. Competitive Advantage: Creating and Sustaining Superior Performance. New York: The Free Press. Raikes, Philip, Michael Friis Jensen, and Stefano Ponte. 2000. “Global Commodity Chain Analysis and the French Filière Approach: Comparison and Critique.” Economy and Society 29(3): 390–417. Schmitz, Hubert, ed. 2004. Local Enterprises in the Global Economy: Issues of Governance and Upgrading. Cheltenham and Northampton: Edward Elgar. Sturgeon, Timothy J. 2001. “How Do We Define Value Chains and Production Networks?” IDS Bulletin 32(3): 9–18. Sturgeon, Timothy J. 2009. “From Commodity Chains to Value Chains: Interdisciplinary Theory Building in an Age of Globalization.” In Frontiers of Commodity Chain Research, ed. Jennifer Bair. Stanford: Stanford University Press: 110–135. Teece, David and Gary Pisano. 1994. “The Dynamic Capabilities of Firms: An Introduction.” Industrial and Corporate Change 3(3): 537–556. Teece, David, Gary Pisano, and Amy Shuen. 1997. “Dynamic Capabilities and Strategic Management.” Strategic Management Journal 18(7): 509–533. Yusuf, Shahid. 2004. “Competitiveness Through Technological Advances Under Global Production Networking.” In Global Production Networking and Technological Change in East Asia, ed. Shahid Yusuf, M. Anjum Altaf, and Kaoru Nabeshima. Washington, DC: The World Bank: 1–34.

1 Inter-firm Dynamics in Notebook PC Value Chains and the Rise of Taiwanese Original Design Manufacturing Firms Momoko Kawakami

1.1. Introduction Since the late 1980s, outsourcing and offshoring by large multinational electronics hardware manufacturers have driven the formation of global value chains (GVCs) producing a wide range of products, especially in East Asia. The rise of Taiwanese notebook personal computer (PC) manufacturers is, perhaps, the most striking example. By 2008 Taiwanese contract manufacturers accounted for approximately 92% of worldwide shipments of notebook PCs (Information Industry Yearbook 2009 Compilation Team 2009). In addition to manufacturing, they have increasingly provided product design services in their role as original design manufacturers (ODMs) for brand name PC firms such as Dell, HP (Hewlett Packard), Apple, and Toshiba. The learning process that has made the Taiwanese notebook PC industry so successful can only be understood by tracing their interactions with the two other main actors in the industry: the branded PC firms and principal central processing unit (CPU) supplier, Intel. Taiwanese ODMs were able to successfully tap into learning opportunities proffered by GVCs as these other firms jockeyed for position to create an appropriate value. In this chapter I focus on two dynamics: (1) how shifting inter-firm relationships structured knowledge flow among firms in the notebook PC GVC and (2) how Taiwanese ODMs exploited these relationships to enhance their capabilities and consolidate their position as the world’s dominant suppliers of notebook PCs. By so doing, 16

Momoko Kawakami

17

the chapter provides an example of successful capability formation by suppliers in late-industrializing economies. The chapter is organized as follows. Section 1.2 presents the research framework of the study. Section 1.3 traces the development of the Taiwanese notebook PC industry. Sections 1.4 and 1.5 divide inter-firm dynamics of the industry into two periods: before and after the early 2000s, and explore the process of capability development by Taiwanese ODM suppliers in each of these periods. The final section summarizes the study’s findings.

1.2. Research framework To trace the growth of Taiwanese suppliers, I employ the GVC perspective (Humphrey and Schmitz 2004; Gereffi et al. 2005). The GVC approach focuses on the organization and spatial distribution of valueadded in the global economy. It offers a conceptual starting point to explore how industry structure and the character of inter-firm linkages affect the flow of inter-firm knowledge, and in turn the opportunities for learning by firms from late-industrializing economies. 1.2.1. Global value chain analysis As global economic integration proceeded and market competition intensified during the last few decades, vertical intra-industry trade across different parts of the world has expanded rapidly (Arndt and Kierzkowski 2001; Jones et al. 2005). One of the driving forces of accelerated global integration has been the search for lower costs by powerful firms from developed economies (Fröbel et al. 1980). The result is an international division of labor that increasingly includes suppliers from developing economies. The GVC perspective highlights the organizational configuration of this type of cross-border trade. It explores the way in which “lead firms”—largely from developed economies—organize the vertical sequence of value-adding activities in global production networks. They do this by selecting suppliers and setting and enforcing the parameters and specifications related to product definition, quantity, quality, and delivery to which suppliers—from both developed and developing countries—must adhere (Humphrey and Schmitz 2004: 96–98). Recent GVC studies have sought to understand this “governance” process by focusing on the specifics of how this coordination is achieved. Arrangements in notebook PC GVCs clearly fit the basic patterns found

18

Inter-firm Dynamics in Notebook PC Value Chains

in these studies: lead firms from developed countries such as Dell, HP, and Toshiba coordinate value-adding activities by setting and enforcing parameters for outsourced transactions.1 As the “chain” metaphor indicates, the GVC perspective highlights sequences of bilateral relationships between powerful lead firms and less powerful suppliers (Gibbon et al. 2008). It explores the determinants for different patterns of chain governance and the asymmetric power relationships that tend to exist between the two parties. However, a simple application of the GVC framework to the notebook PC industry would fail to capture a critical dimension of the industry: the emergence of powerful “platform leaders” and their pivotal role in driving and structuring inter-firm dynamics in the industry. Here, the term “platform leader” refers to companies that provide the core components and technologies on which other companies build products or offer services (Gawer and Cusumano 2002: 6). As Gawer and Cusumano argue, Intel and Microsoft are considered platform leaders in the PC industry (2002: 15–16). While some have attributed this role to historical accident, or a single strategic misstep by IBM, which chose to outsource the CPU and operating system for its original PC in 1981, Intel and Microsoft have in fact devoted extensive resources to developing the ecosystem based on the “Wintel” platform and preserving their leadership position within the industry. Intel, in particular, has been extremely effective in this regard (Tatsumoto et al. 2009). Today, Intel microprocessors are found in a range of products, even in PCs that do not use the Microsoft operating system, such as Apple Macintosh computers. As I will discuss later, the evolution of the notebook PC industry has been structured by the interactions of the three major actors: (1) Intel, the industry’s most powerful component vendor and platform leader, governs the distribution of value-added among firms in the chain; (2) brand-carrying firms in the United States and Japan control access to final markets, and (3) Taiwanese ODM firms control detailed design, production, and the coordination of final assembly and the purchasing of non-core components. These firms occupy different positions in the PC GVC, not only in terms of the vertical sequence of value-added, but also in terms of geographic location. This study shows how Taiwanese ODMs took advantage of the strategic interplay among these different GVC actors to exploit learning opportunities in ways that have created both opportunities and limits for their future development.

Momoko Kawakami

19

1.2.2. Knowledge flow in global value chains and learning strategies The inter-firm flow of resources and knowledge is a critical, yet little explored, topic in GVC analysis. Gereffi (1994: 97) defined governance of chains as the authority and power relationships that determine how various types of resources flow within a chain. Sturgeon (2001: 11) defined lead firms in GVCs as “firms that initiate the flow of resources and information through the value chain by developing and marketing final products.” These definitions suggest that the strategies and actions of lead firms configure the chains and shape the flow of resources. Other studies of global production networks in the electronics industry pay special attention to the role of network “flagship” firms in transferring knowledge and the formation of capabilities by local suppliers as well (Ernst and Kim 2002). For firms from late-industrializing economies, one of the central motivations for participating in GVCs lies in the potential to acquire the strategic resources, including technological knowledge and market information, that can lead to the formation of the “dual-faceted” competences needed to generate rent (profits) as well as bargaining power within GVCs (Kimura 2007: 50–52). Put more simply, firms from developing economies can capture knowledge flowing from lead firms by participating in GVCs. If they are able to consolidate and build upon the knowledge they acquire over time, their relative position within GVCs can be improved. As will become abundantly clear from the case study presented in this chapter, this process is far from certain or automatic.

1.3. Industry background: rise of Taiwanese ODM manufacturers Before proceeding to the analysis, I present a brief history of the rise of the Taiwanese notebook PC industry.2 Japanese and American firms started to develop portable PCs from the earliest days of the PC industry and offered the first “laptop PCs” in the mid-1980s. However, the true technological breakthrough in PC miniaturization came with the advent of “notebook PCs” developed and mass-produced by Toshiba and NEC in Japan in 1989. In developing the new product, both companies designed and manufactured key components and sub-systems in-house or in cooperation with closely linked parts

20

Inter-firm Dynamics in Notebook PC Value Chains

suppliers. These included semiconductors, software, displays, floppy disk drives, and hard disk drives. They used their intimate knowledge of component and sub-system technologies to miniaturize the product through the integration of key technologies that were proprietary and in-house. Because the technological hurdles associated with developing and mass-producing this sophisticated product were high, Japanese firms continued to dominate the notebook PC market until the mid-1990s. Starting in the mid-1990s, however, the entry barriers to the notebook PC industry began to decline. First, the technological know-how associated with mechanical engineering became more mature, as we will see later. Second, and even more important, strategic moves by Intel aimed at establishing platform leadership began to dissolve the product development barriers facing new entrants, including Taiwanese ODMs. These technological changes created opportunities for Taiwanese firms by triggering a massive inflow of orders from US-based computer producers eager to increase their market share, not least by undercutting the prices charged by Japanese producers. As competition intensified, American and Japanese firms outsourced more notebook PC models to Taiwanese contract manufacturers, and began accepting design input from them on a growing range of models. By the late 1990s, notebook PC production and exports from Taiwan had swelled dramatically (see Figure 1.1). Figure 1.2 shows that the worldwide share of notebook PC shipments by Taiwanese firms rose constantly during the past decade, reaching 92% in 2008. At the same time, contract manufacturing’s share of total production by Taiwanese firms, including both original equipment manufacturing (OEM) contracts, where the lead firm provides the design, and ODM contracts, where contractors carry out some or all of the design work, increased from about 80% of 2.6 million units in 1995 to 95% of 112 million units in 2008 (see Table 1.1). While the flood of new contracts decreased the relative importance of Taiwanbranded PCs in the production mix, as almost all of this work was for foreign lead firms, even Taiwan-branded notebook PC companies such as Acer were using contract manufacturers by the early 2000s. Clearly, the rise of Taiwanese contract manufacturers as the world’s dominant producers of notebook PCs was directly related to lead firm strategies to increase global outsourcing. Another critical factor behind the dramatic rise in the share of Taiwanese firms’ global notebook PC production was the successful relocation of production sites to Shanghai and adjacent areas in China after the turn of the century. The Taiwanese government had prohibited

21

16,000 14,000 Unit: thousand sets

Production 12,000 10,000

Exports

8,000 6,000 4,000 2,000 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

0

Year Figure 1.1 Quantity of Taiwan’s production and exports of notebook PCs Source: The Internet Information Search System, Department of Statistics, Ministry of Economic Affairs, R.O.C.

100

120

90

Unit: million sets

100

80 70

Share of world total

80

60 60

50 % 40

40

30 Shipment 20

20

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

0

1995

10 0

Figure 1.2 Worldwide Notebook PC shipments by Taiwanese firms and their share of the world total Source: Information Industry Yearbook, MIC, Institute for Information Industry, various years.

22 Inter-firm Dynamics in Notebook PC Value Chains Table 1.1 Share of OEM/ODM contracting in the total production of notebook PCs by Taiwanese firms (%) Year Contracting ratio

1995

1997

1999

2001

2003

2005

2007

2008

79

81

87

92

96

96

95

95

Source: Information Industry Yearbook, MIC, Institute for Information Industry, various years.

the relocation of notebook PC production to China during the 1990s. In 2001 the government decided to lift the ban, and notebook PC suppliers, facing rising labor and land costs in Taiwan, rushed to set up production capacity in China. The ratio of Chinese production rose from 5% in 2001 to 54% in 2003 and to 99% in 2008, showing an almost complete relocation of the industry in seven years. This transition involved more than simple relocation of existing factories. First, capacity was increased to accommodate a nearly ten-fold increase in output, from 14 million to 112 million units (Figure 1.2). Second, the shift to China prompted an industry shakeout. Figure 1.3 shows the change in the number of portable PC manufacturers in Taiwan,3 50 45

Unit: number of firms

40 35 30 25 20 15 10

Figure 1.3

08/09

07/08

06/07

05/06

04/05

03/04

02/03

01/02

2000/01

99/2000

98/99

97/98

96/97

95/96

94/95

93/94

92/93

0

91/92

5

Change in the number of notebook PC manufacturers in Taiwan

Source: TEEMA (Taiwan Electrical and Electronic Manufactures’ Association) member lists, various years.

Momoko Kawakami

23

declining from about 40 firms in the early 1990s to only 19 in 2008. In recent years, consolidation has proceeded even further, with the top five suppliers, Quanta, Compal, Inventec, Wistron,4 and Asustek, accounting for more than 70% of total production (Information Industry Yearbook 2007 Compilation Team 2007: 9–2). Among them, the largest supplier, Quanta, produced approximately 29% of the world’s notebook PCs in 2008. The simultaneous scale up and absolute geographic shift of companies working in a product area undergoing continual innovation and pressure for price reduction, last-minute configuration, and rapid delivery represent a managerial achievement of significant proportions. In the following sections I will discuss some of the learning processes that enabled this extraordinary transformation.

1.4. Inter-firm relationships and the flow of knowledge: the late 1990s In this and the following section, I employ a three-step approach in my analysis.5 First, I examine the inter-firm relationships among the three major actors that comprise the industry’s value chains— platform leader(s), brand-carrying firms, and Taiwanese ODM contract manufacturers—and then analyze how the relationships among them have evolved over time. Second, I highlight how the knowledge flow among these three actors was structured by their interfirm relationships. Third, I analyze the strategies used by Taiwanese firms to exploit the learning opportunities of GVCs. The discussion in these two sections is based on in-depth interviews with managers of companies acting in all three roles. I conducted 49 interviews with Taiwanese ODM firms, brand-carrying firms, and Intel during 2004– 2008. My questioning centered on the evolution of inter-firm relationships and the degree and character of information and knowledge sharing among firms.6 The evolution of the industry’s inter-firm relationships can be divided into two phases, that is, before and after the early 2000s, with the turning point around the period 2001–2002. In this section, I examine developments prior to 2001 and in the next section I summarize the changes since. 1.4.1. Dynamics of inter-firm relationships: consolidation of platform leadership by Intel In the late 1990s, a complex dynamic of cooperation and competition developed between Intel and brand-carrying (lead) firms (Figure 1.4). On one hand, the two sides cooperated closely in developing new chips

24 Inter-firm Dynamics in Notebook PC Value Chains

Cooperation on chip and system product development Brand firms Intel

B–1 Expropriation of value-added

Technological support

B–2

Competition

B–3

Competition

Outsourcing of product development and production Taiwanese ODMs

T–1

T–2 Competition

Figure 1.4

T–3 Competition

Inter-firm relationships in the late 1990s

Source: By the author.

and system-level products based on their joint interest in cultivating PC markets. For Intel, lead firms were the indispensable agents for delivering the company’s “Intel Inside” branded components to consumers. They also provided Intel with valuable market information to help their “roadmap” for future products and applications. Intel also depended on feedback from experienced PC manufacturers during the course of product development for the verification of new chips and chip-sets (Figure 1.5). Conversely, brand-carrying firms depended on Intel. Access to the details of Intel’s technology roadmap and advance technical information about newly developed products were critical for PC firms’ own product strategies. On the other hand, intensifying competition in the notebook PC industry in the late 1990s led to conflict between the two parties over the appropriation of value-added. As already mentioned, brandcarrying firms—especially Japanese lead firms—dominated the market and enjoyed high profits in the early days of the notebook PC industry. Their in-house capability to develop chip-sets (packages of semiconductors combining Intel CPUs with other system elements) formed the core of their competitive advantage, along with their ability to verify chipsets developed by third-party vendors, which sometimes contained a large number of errors. However, beginning in the mid-1990s Intel started to challenge the dominant position of Japanese lead firms. Following the successful

Momoko Kawakami

Month (t–24)

(t–14)

Development of prototypes using new chips

Launch of new PC product t

(t–6)

Development of PC system

Development of chip Intel discusses with major brand firms about the basic concepts of new products

25

RFQ EVT Development of engineering samples, feedback from users, debug

DVT

PVT

After the 2000s, Taiwanese suppliers’ role is becoming significant with the reduced commitments by brand firms

Figure 1.5 Process of chip development by Intel Note: At EVT (engineering verification test) stage, logic design layout design, mechanical design, E-BOM release, EMI review are carried out. At DVT (development verification test) stage, logic modification, layout modification, mechanical tooling are carried out. AT PVT (production verification test) stage, preparation for mass production is carried out. Source: interviews with related firms.

strategy it had employed to dominate and extract high profits from the desktop PC industry,7 Intel began to promote its own notebook PC chip-sets, launched “mobile modules” that integrated its Pentium brand CPUs, coupled chips and second-level cache memory on a single board as a turnkey solution for latecomer firms, and kept the internal structure of its products as a proprietary black box (Tatsumoto et al. 2009; Ogawa 2007). In this way Intel was able to encapsulate more value chain functions, knowledge, and capture more value by integrating proprietary technical knowledge within its chip-sets, just as MediaTek later did in the mobile phone industry (see Chapter 2 of this book).8 In addition, the rapid spread of 3D-CAD systems for developing molds and dies and the launch of thermal modules as heat treatment solutions by Japanese parts suppliers removed additional technological barriers facing latecomer producers. All these developments negated the originally scarce and valuable engineering expertise that Japanese-branded PC firms had accumulated in earlier years (Ogawa 2007). The emergence of ready-made technological solutions with well-defined external interfaces encapsulated the know-how needed to deal with a host of potential electrical engineering problems within Intel’s design infrastructure (Tatsumoto et al. 2009). Together with better codification across most PC system elements, Intel’s strategic move increased the modularity of the product, lowering entry

26

Inter-firm Dynamics in Notebook PC Value Chains

barriers for developing and mass-producing notebook PCs. With few alternatives available, this wrested value from incumbent firms rapidly created a field of low-cost competitors that could expand the consumer marketplace with a myriad of low-cost products with superficial variations that competed away profits. Lead firms were affected by this change in different ways. American firms like Dell, Gateway, and Compaq responded positively to the emergence of Intel’s new notebook PC platforms.9 Exploiting the increased modularity of the product, the firms quickly increased outsourcing to Taiwanese firms to reduce costs, while concentrating their efforts on product conception, marketing, and distribution.10 Taiwanese firms became adept at the recursive and relatively superficial work of product design. Rapid product life cycles and intense competition justified outsourcing the most detailed aspects of design and redesign to Taiwanese ODMs. Conversely, Japanese lead firms that had based high profitability on superior Research and Development (R&D) and production capabilities were negatively affected by the commoditization of the product. Toshiba, the world’s leading notebook PC company in 1997, faced not only a decline in market share but also reduced profits from its notebook PC unit. NEC, the world’s second-largest notebook PC brand at the time, and other Japanese firms suffered similarly.11 It was the strategic move by Intel to leverage its position as a platform leader in the notebook PC GVC, and the brand firms’ reaction to it, that opened up critical learning opportunities for Taiwanese suppliers. Brand firms started to focus more on product conception, brand marketing, supplier management, and negotiations with suppliers of core components (most notably Intel), while outsourcing detailed product design, mass production, logistics, and after-sales service to contract manufacturers. By successfully embracing this role, Taiwanese ODMs soon came to dominate world production of notebook PCs.

1.4.2. Inter-firm flow of knowledge: intensive technical training and knowledge transfer As we have seen, a mix of cooperation and competition between a powerful platform leader and brand-carrying lead firms over the division of value-added in the notebook PC industry provided a critical opportunity for Taiwanese contract manufacturers. As brand firms transferred valueadding activities step by step to Taiwanese firms, an intensive flow of knowledge from lead firms to Taiwanese suppliers took place. Figure 1.6

Momoko Kawakami

Information about chips Intel

27

Brand firms B–1

B–2

B–3

Information about trends in final markets Instructions and assistance on product development

T–1

Knowledge about product development and production technologies and know-how, market information

T–2

T–3

Taiwanese ODMs Figure 1.6 Inter-firm flow of knowledge in the late 1990s Source: By the author.

conceptualizes the inter-firm knowledge flow among firms in the late 1990s. As they moved toward outsourcing production to Taiwanese firms in the late 1990s, American and Japanese lead firms sent teams of engineers on to monitor and train their ODM suppliers in the realms of R&D, parts procurement, production management, and other activities (Kawakami 2009). Production engineers from one Japanese firm trained line workers at its Taiwanese supplier by asking them to assemble and disassemble products again and again, taught them how to avoid making small scratches on exterior surfaces, and reminded them of how demanding consumers can be. An even more intensive transfer of technology and know-how took place when lead firms started to ask for product development services. Engineers from lead firms were stationed at or made frequent visits to Taiwanese suppliers to teach design engineers how to solve various technical issues and analyze data. Even after this initial transfer of knowledge, engineers from buyer and suppliers communicated with each other frequently to solve various problems. These interactions helped to expand and consolidate ODMs’ R&D capabilities. When build-to-order and configuration-to-order systems were put in place at the end of the 1990s, Taiwanese manufacturers again received intensive training from their customers. Implementing these systems was difficult for Taiwanese firms, especially as volumes soared and product variety increased dramatically over the years. They required

28

Inter-firm Dynamics in Notebook PC Value Chains

the formation of new management capabilities to handle a multitude of products and components and to configure, finish, package, and ship highly differentiated products. Intensive assistance from lead firms allowed “no-touch shipments,” in which products are shipped directly from factories to retailers and individual end users without going through the customers’ intermediate warehousing or distribution systems. HP was a fast mover in this regard, helping its suppliers to introduce the system. In the case of Quanta Computer, the first customer to launch the “direct shipment” was Apple Computer, Inc. In 1998 Quanta, in cooperation with Apple, brought in consultants and logistics vendors to help form a project team to introduce the direct shipment system (Kawakami 2009). As their role in the industry grew, Intel started providing Taiwanese ODMs with reference designs that included detailed instructions on how to develop PCs based on their chip-sets, and helped them to solve various technological issues that tend to arise during product development. The growth of Taiwanese contractors was a desirable development for the company because it lowered the cost of PC products and thereby expanded Intel’s market.

1.4.3. Learning strategies of Taiwanese firms: leveraging the diversified customer base In winning new orders and absorbing the associated flow of knowledge, Taiwanese ODMs were by no means passive actors. By consolidating the flow of technologies and expertise transferred from various notebook PC firms and Intel, Taiwanese manufacturers acquired multifaceted product development, mass production, and logistics capabilities that underpinned their rise as full service “turn-key suppliers” (Sturgeon 2002). The intense competition among contract manufacturers to win larger and more profitable orders drove the pace of this learning process. As the market expanded, the fastest-growing Taiwanese contractors were soon serving a large and diverse customer base. As the learning process unfolded, the diversity of this customer base came to assume strategic importance. Serving different types of customers from different markets (the United States, Europe, Japan, etc.) and product segments (low-end and gradually higher-end product categories) exposed ODMs to a wide variety of technologies and market information, demanded that they accommodate different types of requirements, and accelerated the pace of their learning.12

Momoko Kawakami

29

The significance of diversified customer base was especially high for two ODM suppliers, Quanta Computer and Compal Electronics, firms that have outpaced other contractors in Taiwan in the realm of ODM contract manufacturing (Figure 1.7).13 The two companies aggressively pursued customer diversification during the 1990s, whereas some other companies (e.g., Inventec and Arima) chose to become quasi-captive suppliers to powerful buyers. By interacting with multiple customers, Quanta and Compal gained access to additional market information and acquired a wider range of technological capabilities. Nevertheless, at this stage ODM supplier capabilities remained “contestable” and not “unique” to them (Langlois and Robertson 1995: 7). The knowledge transferred to and consolidated by Taiwanese suppliers during this period remained largely “substitutable” and “imitable” (Barney 1991) for their customers, and did not result in capabilities that would generate high levels of rent or bargaining power. As a result, horizontal competition between contractors was extremely intense.

900 800 700

bill. NTD

600 Compal

Quanta

500 400 300 200

Inventec 100

Figure 1.7 Sales of major Taiwanese notebook PC manufacturers Note: Arima Computer was merged with Flextronics International in 2008. Source: Annual company reports.

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

0

Arima

30

Inter-firm Dynamics in Notebook PC Value Chains

1.5. Inter-firm relationships and the flow of knowledge: after the early 2000s 1.5.1. Change in the dynamics of inter-firm relationships: shift to overlapped long-term partnerships In the early 2000s, an industry shakeout took place on the ODM side, and a small number of large Taiwanese suppliers came to dominate the notebook PC contract manufacturing business. This led to several new features in the inter-firm relationships described below. The main factors behind these changes were the following. First, the policy change by the Taiwanese government in 2001 to approve investment by Taiwanese notebook PC manufacturers in China spurred the shakeout in the industry because smaller firms did not have the resources to scale up production in China. The suppliers who had enhanced their capabilities in the late 1990s by trading with multiple customers surged ahead of competitors in this investment rush. More specifically, Quanta and Compal stretched their lead over their competitors, leveraging their rapid pace of learning further (see Figure 1.7). Second, the launch of the low-power Centrino platform by Intel in 2003 further increased the modularity of the industry by integrating even more functions within the chip-set, including wireless internet connection, power management, and a host of other functions.14 The further encapsulation of value-added by the platform leader intensified the price competition among branded firms,15 and raised the importance of scale economies. It thus spurred the concentration of contract manufacturing orders in fewer suppliers, leading to further consolidation in the supply base. Third, by lowering prices and improving notebook PC functionality (e.g., enhanced internet connectivity and longer battery life) the launch of the Centrino platform boosted demand for mobile PC products and triggered a new round of notebook PC competition. To embrace the growing market, lead firms further expanded outsourcing, but since the number of ODM firms had decreased and scale economies had become ever more important this new business was funneled to the same small set of huge ODM suppliers. As a result, branded notebook PC firms came to rely on the same set of contractors. In 2006, nine of the world’s top ten notebook PC brands were trading with Quanta Computer, the largest supplier of notebook PCs in the world. The overlapping of trading partners for both outsourcers and contractors triggered several important changes in the industry, as summarized in Figure 1.8. First, brand firms began to compete with each other

Momoko Kawakami

31

Cooperation on chip and system product concept creation Brand firms (reduced) Intel

B–1

B–2

B–3

Competition Cooperation on chip design

“Enclosure” of resources

Selection of customers

Taiwanese ODMs (consolidated) T–1

T–2 Competition

Figure 1.8 Inter-firm relationships after the 2000s Source: By the author.

to secure the attention of ODM suppliers. In other words, lead firms with the most bargaining power sought to strategically “capture” the resources possessed by their suppliers. They began to require ODM suppliers to create dedicated business units with distinct product development and mass-production capabilities. Furthermore, the largest lead firms asked their suppliers to allocate the “best” engineers and other staff to their projects, and to open financial accounts devoted to each customer, so that the returns from the customer’s investments in the relationship could be monitored and the benefits from production scale economies captured bilaterally. Taiwanese contractors responded to these requests, but only to a degree. The largest customers received their own divisions while smaller customers were sometimes grouped together (see the example of Quanta Computer in Figure 1.9). While it appears that this would be a burden to the ODM contractors, it is unclear how thoroughly operations are segmented among customers over the long term. Production equipment for core processes like circuit board assembly and material handling is highly generic, and can be moved between divisions. Warehouse space can also be re-assigned, as can personnel. Second, as a response to the strategies by lead firms to capture supplier capabilities, the largest ODM firms started, after the early 2000s, to discriminate against smaller customers by offering higher pricing, lower R&D and production quality, and slower delivery responsiveness.

32 Inter-firm Dynamics in Notebook PC Value Chains

NB unit 1 (Dell) NB Group1

NB unit 2 (IBM, Fujitsu-Simens, and others)

NB unit 5 (HP and channel customers)

General manager

NB unit 3 (Sony, NEC, and others) NB Group2

NB unit 4 (Toshiba and Acer)

NB unit 6 (Apple) Figure 1.9

Organization of Quanta Computer as of 2006

Note: Company names in parenthesis are main customers of business units. NB stands for notebook PC. Source: Company report of Quanta Computer and interviews by the author.

In other words, capabilities were distributed according to the buyer’s market power. Relationships between Intel and Taiwanese firms also changed after the early 2000s. During the 1990s, lead firms collaborated closely with Intel to solve the technological problems faced by their contractors during product development. More recently, as lead firms have been retreating from the time-consuming and strategically dubious process of working closely with their suppliers to solve technological problems related to Intel’s chips, they have encouraged such problems to be hammered out directly with Intel (see Figure 1.5). Faced with deteriorating profitability, Japanese-brand firms also reduced the human resources devoted to the notebook PC business, and began relying more on Taiwanese ODMs for design. As their relationships with Intel matured, Taiwanese ODMs began to assist Intel in debugging newly developed chips. While Intel still provided lead firms with access to market forecasts and retained the practice of negotiating chip availability and pricing with brand-carrying firms, the degree of technological collaboration between engineers of Intel and lead firms decreased. In contrast, Taiwanese ODMs used their ties to Intel to accelerate their

Momoko Kawakami

33

pace of learning and consolidate their new, more important position in notebook PC GVCs. 1.5.2. Inter-firm flow of knowledge in the early 2000s: shift to a bidirectional knowledge flow Figure 1.10 shows the inter-firm flow of knowledge after the turn of the twenty-first century. As the capabilities of Taiwanese manufacturers in terms of R&D, production, and logistics matured, lead firms gradually retreated from providing intensive training and assistance to their suppliers. As a result, the inflow of knowledge regarding product development, mass production, and deployment of global logistics from lead firms to contract manufacturers decreased substantially. In this period, the formation of new inter-firm linkage patterns and the creation of deep pools of information within Taiwanese ODMs took place as shown in Figure 1.10. First, more stable trading relationships allowed engineers and managers from both parties to forge closer relationships in which in-depth information is exchanged bilaterally. With longer-term partnerships and the protection of strategic information assumed, lead firms started to share more details about their product strategies and market observations with the ODM engineers that work (more or less) exclusively for them. Exchange on information on chip and system product concept creation (–) Intel

Brand firms

B–1

Exchange of information (increased)

Knowledge about product development and production (reduced)

B–2

B–3

In-depth information about final markets (increased)

Taiwanese ODMs T–1

T–2 Creation of information pool

Figure 1.10 Inter-firm flow of knowledge after the 2000s Source: By the author.

Proposals, suggestions, advices (increased)

34

Inter-firm Dynamics in Notebook PC Value Chains

This allowed Taiwanese suppliers, especially those with a wide customer base, to accumulate extensive knowledge about the requirements and preferences of various customers and end users. Exploiting the information pool created by the close interactions with lead firms, Taiwanese ODM firms began to prepare more prototypes and designs tailored to specific customers—especially for their top-priority customers. Taiwanese ODM firms also started to bring forward proposals for entire product lineups, including the design of new products, selection of functions, adoption of new parts, arrangement of logistics, and other issues of strategic importance to their most important customers. Although Taiwanese suppliers had been creating product roadmaps and raising various proposals earlier, lead firms have only recently come to seriously consider their proposals, and even to rely on them. Over time the exchange of information and knowledge between the lead firms and ODM contract manufacturers in the notebook PC industry has become thicker as well as bidirectional. The two parties started to exchange more in-depth knowledge, experiences, and observations about final markets and the latest technological developments. In this way, what were highly “modular” value chains, where the contracts for inter-firm transactions were highly codified in the form of detailed specifications, have become more “relational” (Gereffi et al. 2005). More tacit and hard-to-codify information is being exchanged between trading partners that are less substitutable. Paradoxically, the rise in the modularity of the product has led to the sharing of in-depth knowledge and mutual reliance between trading partners in co-creating the value-added in the industry. Furthermore, the relatively standard design parameters of notebook (and other) PCs allow ODMs to pool design knowledge in ways that are not possible with products with less design modularity (Sturgeon 2009). Still, the high modularity of the product allows lead firms to keep strong bargaining power vis-à-vis the ODM firms by switching orders among trading partners, as we will see later.

1.5.3. Strategy of learning: consolidating the information and knowledge pool In this final section I explore the learning process underpinning the acquisition of new capabilities by Taiwanese firms starting from the early 2000s. In this period, consolidating and leveraging the intra-firm information pool within leading Taiwanese ODMs, created by serving multiple customers, became an effective device for accelerating learning.

Momoko Kawakami

35

By setting up divisions assigned to different customers, ODM firms created an organization that allowed them to learn from multiple customers simultaneously (see Figure 1.9). In spite of informational “firewalls” between divisions, Taiwanese manufacturers do in fact leverage information and technologies acquired and, in essence, transfer what they have learned from one customer to another. Although proprietary customer information is protected, lessons learned in one division can be shared company-wide. Detailed sharing of customer information and data among divisions is restricted, but inter-division interaction at the senior management level provides a mechanism for diffusing more general market and technology information. Also, generic knowledge is shared across different business units, such as those regarding product verification and production management. So, in practice, ODM firms have been able to utilize a rich information and knowledge pool about final market and technological trends across their entire business. It is

70 Intel

60

50

%

40

30 Dell 20 Quanta

2008

2007

2006

2005

2004

2003

2002

2001

2000

1999

1998

1997

1996

1995

1993

0

1994

10

Year/fiscal year Figure 1.11 Ratio of gross margin to net sales: comparisons of Intel, Dell, and Quanta Source: Intel and Dell, from Annual Report Form 10-K, various fiscal years. Quanta, from annual reports, various years.

36 Inter-firm Dynamics in Notebook PC Value Chains

these information pools that enable them to make increasingly valuable and unique proposals and suggestions to their customers.16 Following Langlois and Robertson (1995: 7) we can say that during this phase, Taiwanese firms began to possess an “intrinsic core,” valuable resources that are hard to imitate. However, it should also be noted that the profitability of ODM suppliers has declined over time (see Figures 1.11 and 1.12), as lead firms have started to put ever greater pressure on their suppliers to reduce prices under threat (and practice) of switching orders between suppliers. Apparently the formation of an “intrinsic core” by Taiwanese suppliers has not provided them with enough power to raise prices and increase profitability.17 On the contrary, lead firms have started to purchase more and more components in advance (known in the industry as component “consignment”), in addition to the CPU chip-sets and other high-value components that they have always purchased to assure supply, to prevent Taiwanese suppliers from earning profits by marking up parts prices, one of the few areas of the business where they have control over pricing. To the extent that Intel’s highly integrated platform has raised the homogeneity of products, the economic returns to the “intrinsic core” of Taiwanese 60.0 50.0

Quanta

40.0 30.0

Inventec Compal

20.0 % 10.0 0.0 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 –10.0 –20.0

Arima

–30.0

Figure 1.12

ROA of major Taiwanese notebook PC manufacturers

Note: Arima Computer was merged with Flextronics International in 2008. Source: Annual company reports.

Momoko Kawakami

37

firms are limited because they remain substitutable, albeit among a small number of highly capable players.

1.6. Conclusion This chapter has explored the process by which Taiwanese notebook PC manufacturers developed their supplier capabilities, focusing on inter-firm dynamics with the global division of labor that characterizes GVCs in the electronics industry. Specifically, I have employed a three-step approach in which I studied (1) the inter-firm relationships of competition and collaboration among the three major actors of the notebook PC value chains; (2) the knowledge flow among these three actors; and (3) the strategy on the part of Taiwanese suppliers to exploit learning opportunities that have arisen from the changing configuration of knowledge flow in the industry’s value chains. I divided the evolution of inter-firm dynamics of the notebook PC industry into two phases, approximately before and after the turn of the century, and explored the above three topics in both periods. To wrest value from brand-carrying lead firms, Intel launched a platform strategy in the late 1990s that encapsulated, within its chip-sets, scarce and valuable expertise possessed by leading branded notebook PC firms in Japan. This resulted in price declines and reduced profitability for firms that had previously enjoyed lucrative returns based on their in-house product development and production capabilities. After a period where Intel’s encapsulation strategy lowered entry barriers for following firms, a similar dynamic played out, to a lesser extent, for lead firms based in the United States. As the product price declined, notebook PC firms outsourced more and more product development and mass production to Taiwanese firms. In the early stages, lead firms trained Taiwanese suppliers, but as the latter’s capabilities in product development and mass production matured, these efforts became unnecessary. After 2001, lead firms sought to capture the best resources of their suppliers for their own use, leading to a deeper level of interaction between buying and selling firms. While the effort to block information sharing within suppliers was only partly effective, the exchange of information and knowledge between the two parties became denser and increasingly bidirectional as the 2000s progressed. Through this process, Taiwanese firms came to possess an information pool and capabilities that may seem, given their relational nature, difficult to imitate and substitute by other actors in the chain. However, because these capabilities are applied

38

Inter-firm Dynamics in Notebook PC Value Chains

to the development and production of products based on modular architecture, with a core module with a very high level of functionality, they remain largely substitutable. In conclusion, three agendas need to be addressed in future research. First, the case of the notebook PC industry poses a challenge to the concept of “chain governance” used in the existing GVC literature. As Gibbon et al. (2008) point out, GVC studies tend to begin with the inter-firm linkages at a specific node in the chain—the lead firms and their first-tier suppliers—when exploring GVC governance. In the PC industry, however, it is the platform leader, Intel, that has the most important role in controlling the distribution of value-added among value chain actors. Intel possesses the power not only to set standards but also to drive the chain dynamics by redistributing the value-added in its favor. As a result, the learning process that has made the Taiwanese notebook PC industry so competitive can be understood only by tracing the evolution of inter-firm relationships among the three major actors in this GVC. This insight extends GVC governance theory beyond the dyadic relationship between lead firms and suppliers. Thus, we should seriously contemplate how to conceptualize “chain governance” and “lead firms” to incorporate the complex and diverse value chain dynamics that are common in the contemporary IT hardware industry. Second, the economic conditions that facilitated the rise of Taiwanese firms should also be examined. The emergence of a large number of manufacturers swarming into the fabrication of notebook PCs and the resulting intense competition among these firms spurred active learning in the case of Taiwan, but not every late-industrializing economy has local manufacturers with this level of entrepreneurial vigor. We must therefore consider seriously the economic and social conditions and institutions that enable, or inhibit, firms from late-industrializing economies to successfully exploit the learning opportunities offered by GVCs. Third, the notebook computer case presented in this chapter does not represent the capstone of technological learning and GVC upgrading by Taiwanese firms. Rather, it sets the stage for an unfolding GVC dynamic that will be analyzed in ongoing research. For example, Taiwanese firms have recently become very active in a convergent set of new product areas, built around new platform technologies such as Intel’s ATOM chip-set and Google’s Android operating system, that includes very lowcost “netbook” portable computers, “smart” mobile phone handsets, and personal navigation devices (Sturgeon and Kawakami 2010). The

Momoko Kawakami

39

recent rise of Acer and Asustek as successful brand-carrying suppliers of netbook PCs deserves special remarks too. The discussion of this development remains a future research topic.

Notes 1. The demand for notebook PCs is heavily concentrated in the developed markets. In 2008 the United States, Japan, and West Europe accounted for 30%, 8%, and 33% of worldwide shipments of notebook PCs, respectively (Information Industry Yearbook 2009 Compilation Team). This overwhelming share of the final market held by the developed countries has led to the strong market power of brand-carrying companies from the United States and Japan, although the multiplicity and intense competition among brand firms have weakened their bargaining power vis-à-vis Intel. 2. A part of this section is based on the author’s interviews with executive officers of Kohjinsha, Co., Ltd, former senior engineers of NEC and Panasonic Communications Co., Ltd. Existing literature on the Taiwanese PC industry includes Dedrick and Kraemer (1998), Ernst (2000), Kishimoto (2002, 2004), and Sturgeon and Lee (2005). 3. TEEMA member lists are compiled based on the self-reported data. Partly due to this, the figure suffers from inexactness. 4. Wistron was formerly the design, manufacturing, and services division of Acer. Acer is one of the few Taiwanese firms that have continuously pursued an original brand manufacturing strategy. In the early 2000s, the company split its brand business from its original design manufacturing activities and established two independent companies. This reorganization proved successful and Acer has risen to be one of the world’s largest sellers of notebook PCs. Asustek Computer Inc. followed the same strategy and split its design and manufacturing division to establish an independent company in 2008. 5. A part of this section and the next one are based on Kawakami (2009). 6. The Taiwanese ODM companies interviewed were Quanta Computer, Compal Electronics, FIC, Arima Computer, Wistron, and Uniwill Computer (acquired by Elitegroup Computer Systems in 2006). The interviewed brandcarrying companies included NEC, IBM, Toshiba, Fujitsu, and Hitachi. I made multiple visits to most of these firms. In addition, interviews were also conducted with former executive managers of the Taiwan procurement office of Dell and HP. 7. In the mid-1990s Intel started to define and promote various types of technological standards and interfaces for desktop PCs, and included these standards in their own chips. The company promoted them as technological “platforms” upon which PC makers could easily develop products. For a detailed examination of Intel’s platform strategy, see Gawer and Cusumano (2002) and Tatsumoto et al. (2009). 8. Intel kept the internal structure of its products proprietary while making their external interface specifications widely available (Gawer and Cusumano 2002: 13). By so doing, Intel succeeded in lowering the price of PC systems over time while it maintained the profitability of its chip products (Tatsumoto 2007).

40

Inter-firm Dynamics in Notebook PC Value Chains

9. Fang (2002) provides a detailed description of the evolution of the subcontracting trade between Dell and Taiwanese ODM suppliers. 10. In 2000 the share of outsourcing to Taiwanese suppliers in the total shipments of notebook PCs for Dell and Compaq reached 61% and 76% respectively, while the figure for Toshiba remained 16% (Chien 2001). 11. Nevertheless, Japanese firms were more reluctant to turn to outsourcing compared with their American competitors during this period. Smaller branded firms were more active in outsourcing, but Toshiba and Sony continued to retain a high proportion of in-house fabrication until the early 2000s. 12. Sturgeon and Lee (2005) point out the strong learning effects associated with a shared supply base where a number of lead firms that compete head-tohead in final product markets trade with groups of suppliers. 13. I did not include Wistron in Figures 1.7 and 1.12, as the company’s key products are diversified compared with other ODM firms shown in these figures. 14. Only the mobile PCs that contain Pentium M, the coupled chip-sets and wireless LAN chips—all of which are supplied by Intel—are allowed to use the “Centrino” brand. 15. The average selling price of notebook PCs dropped from US$1827 in 2001 to US$1065 in 2006 (Topology Research Institute 2005: 47; 2008: 12). In response to the intensified price competition, Japanese firms started to rely more heavily on subcontracting. For example, Toshiba changed its in-house production policy and started to outsource all of their low-cost models from Taiwanese firms in 2003. 16. At the same time, customers also benefit from access to the information pools of their suppliers who sometimes know even better than the customers themselves about the overall development of the world notebook PC market, and they enjoy the improved product development capabilities of their suppliers. 17. Based on a teardown report of a specific HP model (HP nc6230), Dedrick et al. (2010) calculate the value captured by the firms participating in the value chain; according to their calculation of gross profit along the value chain, HP captures 28%, and Microsoft and Intel jointly capture 15% of the wholesale price of the product, whereas the Taiwanese ODMs are only 2%.

References Arndt, Sven W. and Henryk Kierzkowski, eds. 2001. Fragmentation: New Production Patterns in the World Economy. New York: Oxford University Press. Barney, Jay. 1991. “Firm Resources and Sustained Competitive Advantage.” Journal of Management 17(1): 99–120. Chien, Meng-Chu. 2001. “Fenxi Guoji Dachang Jingqi de NB Daigong Celüe” in Chinese (“Analysis of Recent Notebook PC Outsourcing Strategies of Global Large-scale Firms”) Chanye Baogao (Industry Report), KGI Securities Co., Ltd. Dedrick, Jason and Kenneth L. Kraemer. 1998. Asia’s Computer Challenge: Threat or Opportunity for the United States and the World? New York and Oxford: Oxford University Press.

Momoko Kawakami

41

Dedrick, Jason, Kenneth L. Kraemer, and Greg Linden. 2010. “Who Profits from Innovation in Global Value Chains? A Study of the iPod and notebook PCs.” Industrial and Corporate Change 19(1): 81–116. Ernst, Dieter. 2000. “What Permits David to Grow in the Shadow of Goliath? The Taiwanese Model in the Computer Industry.” In International Production Networks in Asia: Rivalry or Riches?, ed. Michael Borrus, Dieter Ernst, and Stephan Haggard. London and New York: Routledge: 110–140. Ernst, Dieter and Linsu Kim. 2002. “Global Production Networks, Knowledge Diffusion, and Local Capability Formation.” Research Policy 31(8–9): 1417–1429. Fang, Kuo-chien. 2002. Hai-kuo Tiankong: Wo zai DELL de Suiyue, Tien-hsia Yuan-chien Chupan: Taipei, in Chinese (My Days at Dell). Commonwealth Publishing Co., Ltd. Fröbel, Folker, Jürgen Heinrichs, and Otto Kreye. 1980. The New International Division of Labor. Cambridge: Cambridge University Press. Gawer, Annabelle and Michael A. Cusumano. 2002. Platform Leadership: How Intel, Microsoft and Cisco Drive Industry Innovation. Boston: Harvard Business School Press. Gereffi, Gary. 1994. “The Organization of Buyer-Driven Global Commodity Chains: How U.S. Retailers Shape Overseas Production Networks.” In Commodity Chains and Global Capitalism, ed. Miguel Korzeniewicz. Westport: Praeger: 95–122. Gereffi, Gary, John Humphrey, and Timothy Sturgeon. 2005. “The Governance of Global Value Chains.” Review of International Political Economy 12(1): 78–104. Gibbon, Peter, Jennifer Bair, and Stefano Ponte. 2008. “Governing Global Value Chains: An Introduction.” Economy and Society 37(3): 315–338. Humphrey, John and Hubert Schmitz. 2004. “Governance in Global Value Chains.” In Local Enterprises in the Global Economy: Issues of Governance and Upgrading, ed. Hubert Schmitz. Cheltenham: Edward Elgar: 95–109. Information Industry Yearbook Compilation Team. Zixun Gongye Nianjian (Information Industry Yearbook), various years. Taipei: Market Intelligence & Consulting Institute (MIC), Institute for Information Industry. Jones, Ronald, Henryk Kierzkowski, and Chen Lurong. 2005. “What Does Evidence Tell Us about Fragmentation and Outsourcing?.” International Review of Economics and Finance 14(3): 305–316. Kawakami, Momoko. 2009. “Learning from Customers: Growth of Taiwanese Notebook PC Manufacturers as ODM Suppliers.” China Information 23(1): 103–128. Kimura, Seishi. 2007. The Challenges of Late Industrialization: The Global Economy and the Japanese Commercial Aircraft Industry. Basingstoke and New York: Palgrave Macmillan. Kishimoto, Chikashi. 2002. “The Taiwanese Personal Computer Cluster: Trajectory of Its Production and Knowledge Systems.” D.Phil. dissertation, Brighton: Institute of Development Studies. Kishimoto, Chikashi. 2004. “Clustering and Upgrading in Global Value Chains: The Taiwanese Personal Computer Industry.” In Local Enterprises in the Global Economy: Issues of Governance and Upgrading, ed. H. Schmitz. Cheltenham: Edward Elgar: 233–264.

42

Inter-firm Dynamics in Notebook PC Value Chains

Langlois, Richard N. and Paul L. Robertson. 1995. Firms, Markets and Economic Change: A Dynamic Theory of Business Institutions. London and New York: Routledge. Ogawa, Koichi. 2007. “Wagakuni erekutoronikusu sangyo ni miru purattofomu no keisei mekanizumu: akitekucha-besu no purattofomu keisei niyoru erekutoronikusu sangyo no saiko ni mukete” in Japanese (“Mechanism of platform formation in the Japanese electronics industry: towards a resurgence of the electronics industry based on the formation of an architecture-based platform”). MMRC Discussion Paper, no. 146. Tokyo: Manufacturing Management Research Center. Quanta Computer Inc., “Quanta Computer Annual Report 2008.” April 30, 2009. http://www.quantatw.com, accessed on February 18, 2010. Sturgeon, Timothy J. 2001. “How Do We Define Value Chains and Production Networks?” IDS Bulletin 32(3): 9–18. Sturgeon, Timothy J. 2002. “Modular Production Networks: A New American Model of Industrial Organization.” Industrial and Corporate Change 11(3): 451–496. Sturgeon, Timothy J. 2009. “From Commodity Chains to Value Chains: Interdisciplinary Theory Building in an Age of Globalization.” In Bair Jennifer (ed.) Frontiers of commodity chain research. Palo Alto, CA: Stanford University Press. Sturgeon, Timothy J. and Momoko Kawakami. 2010. “Global Value Chains in the Electronics Industry: Was the Crisis a Window of Opportunity for Developing Countries?” Policy Research Working Paper 5417, The World Bank. Sturgeon, Timothy J. and Ji-Ren Lee. 2005. “Industry Co-Evolution: A Comparison of Taiwan and North American Electronics Contract Manufacturers.” In Global Taiwan: Building Competitive Strengths in a New International Economy, ed. Suzanne Berger and Richard K. Lester. Armonk and London: M.E. Sharpe: 33–75. Taiwan Electrical and Electronic Manufacturers’ Association (TEEMA). Huiyuan Minglu (Member list), various years. Taipei: Taiwan Electrical and Electronic Manufacturers’ Association. Tatsumoto, Hirofumi. 2007. “PC no basu akitekucha no hensen to kyoso yui: naze Intel ha purattofomu ridashippu wo kakutoku dekitaka” in Japanese (“The evolution of PC bus architecture and competitive advantages: why could Intel acquire platform leadership?”). MMRC Discussion Paper, no. 171. Tokyo: Manufacturing Management Research Center, The University of Tokyo. Tatsumoto, Hirofumi, Koichi Ogawa, and Takahiro Fujimoto. 2009. “The Effect of Technological Platforms on the International Division of Labor: a Case Study of Intel’s Platform Business in the PC Industry.” In Platforms, Markets and Innovation. ed. Annabelle Gawer. Cheltenham, UK and Northampton, US: Edward Elgar: 345–369. Topology Research Institute. 2005. “Xingdong Diannao Ruqin Ershiyi Shiji Jiating Zhongxin” in Chinese (“Mobile PCs Becoming Central to the Families in the 21st Century”). Topology Technology, Inc. Topology Research Institute. 2008. “NB Liangjihua Fazhan Kuayue PC yu CE de Fanli” in Chinese (“The Polarizing Growth of Notebook PCs: Crossing the Border between PCs and CE”). Topology Technology, Inc.

2 Value Chain Creation and Reorganization: The Growth Path of China’s Mobile Phone Handset Industry Ken Imai and Jing Ming Shiu

2.1. Introduction The rise of the East Asian electronics industry has attracted a great deal of research interest (Borrus et al. 2000; Ernst and Guerrieri 1998; Ernst and Kim 2002; Sturgeon 2002; Sturgeon and Lester 2004; Sturgeon and Lee 2005; Yeung et al. 2006; Vind and Fold 2007). These studies have highlighted the key roles that lead firms from developed countries—global buyers and “flagship” transnational corporations (TNCs)—have played in fostering the export-led growth in developing East Asia. In their efforts to reduce costs and tap new markets, lead firms place orders, establish affiliates, and contract portions of production to East Asian manufacturers. In some cases, lead firms have also outsourced the product design tasks to local firms in East Asia. In the process, extensive cross-border production networks, or global value chains (GVCs) have come into being, with local firms in developing countries generally fulfilling limited, specialized, but gradually expanding roles. One of the central concerns of the existing literature is how these local firms can gain more autonomy and become involved in higher value-added activities in GVCs. This chapter examines the creation and reorganization of China’s domestic mobile phone handset value chain in the context of broader industry GVCs. With production of US$413.1 billion in 2008, China’s electronics industry has overtaken the United States as the largest in the world.1 However, as is the case with other newly industrializing economies in East Asia, the industry’s growth in China has been driven, 43

44

Value Chain Creation and Reorganization in China

to a significant degree, by lead firms’ decisions to locate production in East Asia, either through TNC affiliate production, joint venture production, component sourcing, or contract manufacturing. One feature that distinguishes China’s experience from that of other East Asian countries is the huge and highly competitive China’s domestic market. This and China’s increasingly open economic policies have attracted investment from all over the world. As a result, nearly all major mobile phone TNCs have entered or sought to enter China’s domestic market. Of course, many local firms have emerged, survived, and even thrived, not only as suppliers and contract manufacturers but as competitors to TNCs. In this chapter we describe the emergence of local firms in China’s mobile phone handset industry and how their development has co-evolved with the larger GVCs that make up the worldwide industry. China’s domestic mobile phone handset market began to take off in the late 1990s, a period of dramatic change in the global industry. In the 1990s and beyond, advanced economy markets were reaching saturation, cost competition was driving handset makers to outsource manufacturing, and semiconductor, or “chipset,” vendors were introducing a series of significant technological changes with profound effects on the industry’s division of labor. In order to analyze the growth path of the industry, we adopt the GVC conceptual framework developed by Gereffi et al. (2005) and Humphrey and Schmitz (2004). However, instead of focusing on how local firms are inserted into GVCs organized by lead firms from developed countries, as GVC governance and upgrading theory does, our analysis concentrates on how domestic value chains have been created by local firms, how their development has been affected by technological changes induced by foreign technology vendors, and how local firms have tried to leverage their ability to detect and act on business opportunities from their position deep within a huge and diversified local market. The remaining sections of the chapter are organized as follows. In Section 2.2 we present our analytical framework. After a brief portrait of China’s mobile phone handset industry in Section 2.3, we examine the growth dynamics of the industry in Section 2.4. In this section, we highlight the role of three types of firms—brand manufacturers, independent design houses, and core chipset vendors—and describe how the prospects of these firms have changed dramatically over very short time periods. In Section 2.5, we conceptualize these dynamic changes, focusing on local firms’ entrepreneurial responses to technological change, especially the encapsulation of product functionality within larger platform chipset modules provided by semiconductor firms. Section 2.6

Ken Imai and Jing Ming Shiu 45

concludes the chapter with a summary of the theoretical implications of the case study.

2.2. Framework An industry can be thought of as consisting of a series of activities, starting with product development and ending with sales of finished products. These activities produce economic value and are referred to as “value chains.” The manner in which the value is distributed among participants is largely determined by the structure of the chain. The theory of GVC governance developed by Gereffi et al. (2005) spotlights cross-border transactions between the global “lead firms” that organize value chains and their suppliers. Based on comprehensive stocktaking of case studies across industries and countries, the theory identifies three variables that determine how GVCs are organized: the complexity of transactions; the ability to codify transactions; and the capabilities in the supply base. In correspondence to variance in the three explanatory variables, the theory argues that transactions will be shaped into one of five typical patterns—market, modular, relational, captive, or hierarchy—which range from low to high levels of explicit coordination and power asymmetry.2 While this theory was originally devised to analyze the relationship between buyers from advanced industrialized economies and suppliers from latecomer economies, the GVC framework is applicable to a broad range of intra-industrial divisions of labor regardless of the geographical location or the nationality of firms in question, as emphasized by Sturgeon (2009). In this chapter, we apply the GVC framework to the case of China’s domestic mobile phone handset industry as a way of understanding how new value chains are created by local firms and how these value chains respond to environmental changes. In doing so, we must realign the GVC framework to explore the value chain dynamics in a large domestic market: the mobile phone handset industry in China. Our conceptual model for the mobile phone handset value chain is summarized in Figure 2.1. Instead of focusing on the characteristics of transactions between two actors within a chain, as conventional GVC analysis does (e.g., marketing versus manufacturing or system integration versus component manufacturing), our analysis highlights three pivotal activities in the mobile phone handset value chain: product marketing; product development; and the development of a core set of integrated and interoperable semiconductor chips referred to in the industry as a “baseband chipset.”3 Baseband chipsets are the key

46 Value Chain Creation and Reorganization in China

Market

Product marketing Product manufacturing Product development

Core chipset development : Activity in the chain Figure 2.1 Industry

Conceptual Model of Value Chain of the Mobile Phone Handset

functional components in mobile phone handsets, responsible, most basically, for voice signal processing. Starting around 2000, baseband chipsets started to become more highly integrated, or “bundled,” with other important functions, such as key software called the “protocol stack,” which defines the procedures for network connection and signal processing based on prevailing mobile telecommunication standards. These functions, taken together, constitute what is called the handset’s “technology platform.” As we will see, the technology platform plays a critical role in shaping value chains in the industry. In our value chain model, we put aside product manufacturing as a subsidiary activity, chiefly in order to avoid excessive complication, though we need to make reference to it occasionally in the analysis. Our analysis concentrates on the emergence of local firms in each of the three activities, which may or may not take place in sequence—(1) product marketing, (2) product development, and (3) core chipset platform development. Because the market in question is domestic, local firms can potentially exploit their knowledge of local consumers’ preferences and local clients’ needs as an advantage over TNCs. Therefore, we address not only technological capabilities but also the capability to sense and capture market opportunities.4 Among the five types of transactions identified by Gereffi et al. (2005), modular-type transactions are prominent in the case of China’s mobile

Ken Imai and Jing Ming Shiu 47

phone handset industry. Modularization can both decrease the requirements for interaction between product development and production (which usually results in shorter lead times and fewer mistakes in the process) and facilitate capability building by allowing participants to concentrate on activities where their comparative advantages lie (Baldwin and Clark 2000). In the context of this study, modularization is most notably present in the evolution of the technology platform, which encapsulates the increasingly comprehensive and complex core functions of the handset system. Because local Chinese firms were able to purchase complete technology platforms from foreign chipset vendors, the trend toward modularization opened up opportunities for local firms with lower technological capabilities to enter the industry. However, in practice, the path of modularization is not always straightforward or unidirectional. The common view is that modularization proceeds as the technologies involved become widely understood and deployed, more standardized and more codifiable. This is indeed what happened with the “second-generation” (2G) mobile phone standards and the semiconductor technologies that vendors used to develop 2G baseband chipsets. However, when technology platforms with new technologies and functions were adopted in the initial phase of thirdgeneration (3G) handset development, this link in the value chain was essentially decodified, driving close collaboration between technology platform vendors and firms that develop handsets.5 In this case, the dynamics of technological change rendered the modularity of transactions incomplete, driving them in the direction of relational-type transactions. We call this an incompletely modular transaction. To put it differently, if a transaction is modular in a stricter sense (i.e., with little explicit coordination between firms), what we call a highly modular transaction, it may limit the opportunities for parties in the relationship to engage in innovation that spans more than two activities in the chain (Langlois and Robertson 1995). In this context, we can assume that firms that put emphasis on developing innovative products will be inclined to pursue incompletely modular transactions, while technologically backward firms will be more likely to opt for highly modular transactions. As we will see later, in China’s mobile phone handset industry the two cases essentially correspond to value chains of TNCs and local firms, respectively. In the former case, because TNCs control market channels, engage deeply in the standard setting process, and have broad technological capabilities, they can exercise dominant power in their supply chains, choosing close, relational partnerships with other technologically competent firms to

48

Value Chain Creation and Reorganization in China

create proprietary, customized technology solutions. In the latter case, weak lead firms tend to rely on modular linkages with technology platform vendors that provide highly integrated generic modules. Such platform vendors tend to exert strong influence in reshaping the value chain, and extract a large share of the value created. This case study illustrates a dynamic relationship between modularization and the growth of local firms engaged in the three activities listed in Figure 2.1. While modularization affords opportunities for local firms to create new products, the growth of local firms in turn generates demand for further modularization. The trend toward modularization, however, may potentially compress the space for value capture by local firms, which in turn creates strong pressure for value chain reorganization. We will explore this argument more closely after discussing the industry’s development in Section 2.4. Prior to this, however, we present a brief profile of China’s mobile phone handset industry, highlighting the complex nature of the domestic market, where increasingly harsh competition involving major foreign brands and various types of local Chinese manufacturers has been unfolding.

2.3. A profile of the mobile phone handset industry in China Worldwide mobile telecommunication service began to grow rapidly with the introduction of 2G technologies in the early 1990s.6 In China, the number of mobile phone subscribers grew explosively during the second half of the decade. Although the rate slowed somewhat in the mid-2000s, the momentum of the market is still strong, with an annual increase of around 70 million subscribers. More than a decade of rapid growth has made China by far the largest market for mobile phone handsets in the world, with more than 500 million users. Not only is China’s market large and growing, but China has emerged as the world’s largest exporter of mobile phone handsets. In sum, driven by the surge in demand in both the global and domestic markets, China’s mobile phone handset industry has exhibited spectacular growth since the late 1990s (Table 2.1). The country’s share of world production shot up from a negligible percentage in the mid1990s to nearly 50% in 2007. Exports have grown more rapidly than domestic consumption in recent years. It is estimated that TNCs were responsible for approximately 90% of China’s mobile phone exports in 2007. Thus, the export growth of China’s mobile phone handset industry has been overwhelmingly dominated by TNCs, just as has been the case in other developing countries. However, as we will discuss later,

Ken Imai and Jing Ming Shiu 49 Table 2.1 China’s mobile phone handset industry—summary figures (million units, %)

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007

Production (A)

Exports (B)

Imports (C)

Domestic sales (A–B+C)

Export ratio (B/A) (%)

Share in world production (%)

4.0 22.6 52.5 80.3 121.5 182.3 237.5 303.5 480.1 548.6

2.2 5.7 22.8 39.7 63.3 95.3 146.0 228.3 385.5 483.7

1.6 3.0 6.0 7.5 17.2 22.1 12.7 12.8 28.9 16.9

3.4 19.9 36.6 54.8 85.5 109.0 100.1 88.0 123.6 81.8

55.1 25.2 42.6 45.6 48.1 52.3 62.6 75.2 80.3 88.2

2.3 8.0 12.2 21.8 31.2 35.6 34.6 37.2 47.1 47.8

domestic Chinese handset manufacturers have become increasingly outward-looking in recent years. In contrast to the export market, where volumes are high and dominated by a few firms, the situation in the Chinese domestic market is highly complex. Figure 2.2 presents the shares of local brands and foreign brands in China’s domestic market according to official data released by the Ministry of Information Industry (MII). Starting from just around 5% in 1999, the local brands’ share of the Chinese market increased very sharply until 2003, when the official media triumphantly announced that Chinese handset manufacturers had captured more than 50% of the domestic market. Then, quite suddenly, came a reversal. In 2004, local handset manufacturers’ share started to slide. The figure shows that the trend continued up to the third quarter of 2007, when local producers held only 33% of the market. However, the data presented in Figure 2.2 are incomplete because it counts only shipments by “authorized” handset manufacturers. If shipments by unauthorized and newly emerging manufacturers were included, the margin of local firms’ decline would be slim (see the broken line in Figure 2.2). Estimates suggest that unauthorized and newly emerging manufacturers accounted for more than 40% of shipments by local manufacturers in 2007. This means that the actual shares, both of foreign manufacturers and authorized local manufacturers, are substantially lower than represented in official statistics.7 To sum up, during the recent decade, China’s mobile phone handset market has been characterized by rapid growth and dramatic shifts in

50 Value Chain Creation and Reorganization in China

100% 90% 80% 70% 60%

Estimated share of local manufacturers including unauthorized manufacturers

50% 40% 30% 20% 10% 0% 1999 2000 2001 2002 2003 2004 2005 2006 2007 Q3 Foreign brands Figure 2.2

Local brands (authorized)

The estimated shares of local brands and foreign brands

Note: Because of the lack of coverage of unauthorized manufacturers’ shipments, the share of local brands is biased downward. The broken line represents a very rough estimate of the share of local brands including unauthorized manufacturers. See text for details. Source: Compiled by the authors based on estimates by MII, CCID, Pday Research Center, and other sources.

market share. After local brand manufacturers broke the initial market dominance of foreign manufacturers, they soon slid into decline, and this was followed by the emergence of a new set of local manufacturers, including unauthorized ones. In our examination of the value chain dynamics of the industry in the next section, we argue that market share volatility has been driven primarily by the interaction of two factors—the modularization of baseband chipset technologies and the entrepreneurial, strategic reactions of local and foreign firms to these changes in technology. We begin with a description of the global disintegration of the mobile phone handset industry, a set of changes that provide the context for the GVC dynamics we observe in China.

2.4. The GVC dynamics of China’s mobile phone handset industry8 2.4.1. Disintegration of the mobile phone handset industry A significant turning point in the world’s mobile telecommunication industry was the introduction of a global system for mobile communications (GSM) in 1991, the first 2G wireless

Ken Imai and Jing Ming Shiu 51

telecommunication standard. It originated in European countries and was later adopted by more than 200 countries, becoming a de facto global standard. Digital mobile telecommunication standards, including GSM, relate to a highly complicated set of technologies. Until the late 1990s, due to the low codifiability of the relevant technologies (e.g., analog signal processing), the GSM handset industry developed in a highly integrated manner, and the great majority of mobile phone handsets were provided by global telecommunication network providers close to the standard-setting process, such as Nokia, Motorola, and Ericsson (Funk 2002). These firms also developed baseband chipset and protocol stack in-house in order to achieve optimal communication performance, low battery consumption, and smaller handsets. The disintegration of the mobile phone handset industry started in the mid-1990s when US and European specialized semiconductor vendors such as Texas Instruments (TI), Analog Devices (ADI), Lucent, and Philips began to offer generic baseband chipsets. To facilitate this, these chipset vendors bundled baseband chipset with protocol stack, creating highly functional modules referred to in the industry as “technology platforms.” Hence we refer to such firms as “technology platform vendors.” While some major handset manufacturers such as Nokia preferred to customize technology platforms by collaborating closely with technology platform vendors to realize cutting-edge product performance, the availability of platforms embodying the most pivotal functions of the handset system enabled firms that lacked comprehensive knowledge of mobile telecommunication technologies to enter into the handset market. For example, in the late 1990s Taiwanese and Korean original design manufacturing (ODM) contract manufacturers that design and manufacture products for sale under the brand names of other firms were able to provide complete handset to their customers, typically network operators seeking to provide entry-level products for new subscribers. Similarly, some firms, called independent design houses (IDHs), specialized only in the design and development of handset products, designing phones that could be assembled by contract manufacturers and sold under the brand names of other firms. These new value chain actors played important roles in the earliest phase of the growth of China’s mobile phone handset industry.

2.4.2. Rise and decline of local Chinese handset manufacturers In 1994, two Chinese telecommunication network operators, China Telecom and China Unicom, adopted GSM as their 2G standard for

52

Value Chain Creation and Reorganization in China

mobile telecommunication. This decision effectively opened up international competition in the domestic market. By the mid-1990s GSM had become the most widely used 2G standard in the world and any phone using the standard could be sold in the Chinese market.9 Foreign manufacturers needed to make only small adjustments to their existing products in order to make them compatible with China’s telecommunication networks. Later, as the technologies of GSM matured, local firms benefited from the rich technological resources available in GSM technology platforms. Until the late 1990s, China’s mobile phone handset market was virtually monopolized by two major TNCs—Motorola and Nokia. Because of policy restrictions such as strict import control, TNCs established joint ventures, typically with state-owned telecommunication equipment manufacturers, to manufacture handsets in China to sell in the local market. Technological spillover from TNCs to their Chinese partners, for example in the area of product development, was very limited because local joint ventures were only responsible for assembling mostly imported components. In the earliest phase of the industry’s growth, policy interventions played an important role in supporting the entry of local firms. As market opportunities opened up in the 1990s, Chinese government policymakers came to recognize the importance of the mobile telecommunication industries, including handset development and production. In early 1990, the government issued a decree aimed at curbing further expansion of foreign brands and promoting the growth of an indigenous mobile phone handset industry. The decree required all manufacturers to acquire a license from MII for producing and marketing mobile phone handsets and to acquire permission from MII for adding new production lines. Although this industrial policy gave local companies a chance to enter the lucrative market for mobile phone handsets, most local firms came from the fixed line telecommunication equipment industry or the consumer electronics industry and had very limited or virtually no technological capabilities in the area of mobile phone handsets. A few firms succeeded in developing their own handsets, but their limited capabilities did not allow them to keep pace with changes in the market. At this juncture, local firms opted for the strategy of outsourcing most or all of their development and production, especially to Korean IDHs and Taiwanese ODM contract manufacturers, while focusing on sales and product marketing, where they had advantages vis-à-vis TNCs.

Ken Imai and Jing Ming Shiu 53

Despite these new sources of design, the two main telecommunication network operators, China Telecom and China Unicom, tended to rely on other brand name producers for their handsets.10 Typically, TNC manufacturers sell products to large first-tier distributors with networks that cover the entire country or several provinces. The first-tier distributors then distribute the products to second-tier dealers who cover much smaller areas, who in turn sell the products to retailers. There are often even third-tier dealers between second-tier dealers and retailers. In the early 1990s, major nationwide distributors, earning hefty profits from selling foreign brand products, had little reason to be interested in dealing seriously with new local entrants whose brands were barely recognized by consumers at the time. In response, local manufacturers turned to markets neglected by the TNCs and their nationwide dealers: rural areas, townships, small cities, and hinterlands. Because the adoption of mobile phones had only recently begun in these areas, consumers’ recognition of foreign brands was far less established than in mainstream urban markets. Focusing on these potentially large but as yet ignored segments of the domestic market, local manufacturers approached second- or third-tier dealers and offered them better incentives than those offered by foreign brands. In addition, in an attempt to make up for their technological disadvantages, local manufacturers placed great emphasis on exterior designs and user interfaces that suited the preferences of Chinese consumers. By fully exploiting the strategy of focusing on marketing and outsourcing development and production, the local manufacturers best able to pursue agile marketing strategies, such as Bird and TCL, scored outstanding successes and within a few years gained market share comparable to that of the major TNCs. Figure 2.3 presents conceptualized models comparing the new value chains created by local manufacturers in the earliest phase of the industry’s growth with the pre-existing value chains of major TNCs. As we mentioned earlier, TNCs’ transactions with technology platform vendors (presented on the left side of the figure) tend to be incompletely modular, owing to their emphasis on the development of cutting-edge products. In order to achieve proprietary product designs, TNCs also developed the majority of their products themselves, outsourcing only a portion of development and manufacturing for low-end products to Taiwanese ODM manufacturers and other contract manufacturers. The pattern of transactions within the local firms’ value chain (on the right side of the figure) is very different. First, as already mentioned, marketing and product development were vertically disintegrated. Local branded handset firms were only responsible only for

54 Value Chain Creation and Reorganization in China

Market

Product marketing

Product marketing

F

L

Highly modular F (Korean & Handset development Taiwanese)

Handset development

Incompletely modular

Modular

: Firms’ boundaries Core chipset development

F

F : Foreign firms L : Local firms

Figure 2.3

Patterns of value chains (1)—late 1990s to circa 2003

product marketing. Transactions between local brands and Korean IDHs or Taiwanese ODM manufacturers were highly modularized. In other words, ODM contract manufacturers typically provided local branded lead firms with pre-designed product prototypes, fully designed and assembled printed circuit boards (PCBs), or even finished products. Second, transactions between IDHs and ODM manufacturers on one hand and technology platform vendors on the other were more modular than in the case of TNCs. This is because IDHs and ODM manufacturers developed handsets based on generic technology platforms. Coordination was usually limited to problem-solving when adopting new technology platforms during the initial phase of mobile phone handset development. While high levels of transaction modularity meant that products sold by local brands were generally more generic in terms of technology and features compared with TNCs’ products, there was an advantage in terms of responsiveness. Moreover, the demand for low-cost mobile phones in China was more than enough for them to sustain growth until the boom of local brands reached its peak in 2003. Handsets sold by local manufacturers were, although inferior to TNCs’ products in quality and functionality, more appealing to many low-income

Ken Imai and Jing Ming Shiu 55

consumers who were sensitive not only to prices but also to the high levels of variation in exterior designs and user interfaces. Subsequent developments, however, demonstrated the fragility of success built predominantly on a marketing-centric strategy for highly modularized products. Major TNCs soon learned their lesson from their lack of emphasis on sales to underserved markets, and began to extend their marketing efforts beyond the mainstream market and place more emphasis on lower-end products. At the same time, they intensified efforts to fend off further penetration by local brands in large urban markets through aggressive diversification of their product lineup. The introduction of brand new technologies such as color LCD screens and cameras also favored TNCs because local brand manufacturers lagged behind in assimilating the new technologies and features into their products. As a result, the share of local brand manufacturers started to decline in 2004 (as shown in Figure 2.2). In 2005, almost all the leading local brand manufacturers suffered significant financial losses. However, the resurgence of TNCs was not the sole reason for the poor performance of major local handset firms. Because the size of the market was growing rapidly, new entrants proliferated, which led to further decline in the average selling prices and in the profitability of local firms. In addition, most new entrants circumvented the entry regulations by borrowing licenses from inactive license-endowed manufacturers. When license restrictions were substantially relaxed in 2005 and finally abolished in 2007, more than 100 handset firms entered the market.11 These newcomers included many unauthorized firms that ignored the compulsory product certification required by law. Figure 2.4 shows the estimated share of brand name lead handset firms in the Chinese market in October 2008. According to the figure, three global brands—Nokia, Samsung, and Motorola—captured more than 50% of the market, leaving the remainder to Sony Ericsson, LG, and local brand manufacturers. As is the case in Figure 2.2, it can be assumed that this estimate fails to count many newcomers, especially unauthorized manufacturers. This means that, in reality, China’s domestic market is likely to be much more fragmented than the estimate suggests. Two questions that arise from this analysis are: “What are the factors behind this ever-continuing fragmentation of China’s mobile phone handset market?” and “How are the value chains of minuscule local manufacturers structured?” To answer these questions we focus on two events that took place since the local brands appeared on the scene—the

56 Value Chain Creation and Reorganization in China

Figure 2.4

Nokia 38.9%

Samsung 19.7%

Motorola 7.5%

TIANYU 6.3%

Lenovo 4.4%

Sony Ericsson 3.5%

Gionee 2.2%

LG 2.0%

Changhong 1.7%

Haier 1.6%

Others 12.2%

Market shares of brand manufacturers in October 2008

Source: Estimated by CCID Consulting. http://www.ccidconsulting.com/insights/content. asp?Content_id= 20372

evolution of local IDHs and the arrival of new, highly functional technology platforms created by Taiwanese and local chipset vendors. 2.4.3. The rise of independent design houses in China In the context of the mobile handset industry, an independent design house is a firm that specializes in the development of handsets on a contract basis. The most capable IDHs are able to undertake the entire process of development, from product definition to certification acquisition. More recently, many IDHs have taken on procurement of components, management of contract manufacturing, and shipment of completed handsets (or semi-assembled kits) to customers. On the other hand, there are numerous minuscule IDHs that undertake only one or two design processes (e.g., exterior design or mechanical design) for handset manufacturers or that do so for other IDHs as subcontractors.

Ken Imai and Jing Ming Shiu 57

Mobile phone IDHs first emerged in Korea in the late 1990s. As we mentioned previously, Korean IDHs were, along with Taiwanese ODM manufacturers, the major providers of handset product design to local manufacturers during the earliest years of China’s handset industry. However, as competition in the domestic handset market intensified, local handset manufacturers became increasingly dissatisfied with outsourcing product development to Korean IDHs because of high costs and their inability to make quick adaptations to ever-changing market conditions. Chinese local IDHs appeared soon after the initial growth of the handset industry in China. Most of them were established either by Chinese engineers who had acquired technological knowledge about handset development through job experience in TNCs, the most competent Chinese telecom manufacturers such as ZTE, or electronic device distributors who had market knowledge and connections with engineers. The growth of local IDHs accelerated after 2002. By 2004 or 2005, local IDHs had virtually replaced their Korean and Taiwanese rivals as the mainstream players in the design outsourcing business for local manufacturers. Estimates suggest that about 40–50% of the handsets shipped in China in 2006 by local manufacturers were designed by IDHs. In that year, six major IDHs accounted for around 80% of shipments. The largest IDHs are responsible for more than 10 million units per year, which is comparable to the scale of shipments by the largest local firms. Thus, the presence of IDHs in China’s handset market is large in comparison with the global market, where in-house development by handset manufacturers remains the dominant mode. There are several interrelated reasons why local IDHs could overtake their Korean and Taiwanese predecessors within a very short period. First, the technologies for platforms, peripheral components, and other modules that enable IDHs to include additional functions had become increasingly mature. Most importantly, perhaps, technology platform vendors started to provide customers with reference designs for the main circuits. As a result, in the early 2000s, product development of 2G mobile phone handsets could be realized by assimilating and combining existing technologies embodied in technology platforms and readily available peripheral components and software according to reference design provided by technology platform vendors. Local IDHs—led by a small number of core engineers who had acquired knowledge through their job experience in TNCs, telecommunication system vendors, and public research institutes—became increasingly

58 Value Chain Creation and Reorganization in China

capable of executing the task at a much lower cost than Korean and Taiwanese firms, especially since they could utilize the growing pool of young Chinese engineers for knowledge-intensive tasks such as software development.12 Figure 2.5 is an illustration of the value chain patterns that emerged in China around 2004. Korean IDHs and Taiwanese ODM manufacturers were displaced by local IDHs (see the right-hand side of the figure). Although the structure of the value chains appears similar, there was one crucial difference between local IDHs and their foreign competitors. Because of their proximity to the market, local IDHs had better knowledge of Chinese consumers’ preferences and the needs of their clients (local branded handset firms). They could be more flexible and responsive, instead of developing products with long lead time based on predetermined assumptions about market conditions. This was a major advantage for local handset firms who, in an increasingly competitive environment, desperately needed product solutions that could be brought to market quickly.

Market

Handset development

Incompletely modular

Handset L development

Modular

Core chipset (platform) F development

L

Highly modular

L

F Handset development

Product marketing

Product marketing

Product marketing

Modular

: Firms’ boundaries F : Foreign firms L : Local firms

Figure 2.5

Patterns of value chains (2)—circa 2003 to 2004

Ken Imai and Jing Ming Shiu 59

At the same time, some of the largest local manufacturers began to develop in-house design capabilities in an effort to alleviate their disadvantage vis-à-vis TNCs in product development (represented in the center of Figure 2.5). The process in which large local handset firms develop products is largely the same as IDHs. However, there are some important differences that favor IDHs in terms of efficiency and cost in product development. First, because they are independent private firms specializing in product development, IDHs can generally offer better incentives to engineers, such as stock options, than brand manufacturers can. This explains why there is a constant flow of engineers from TNCs or competent local telecommunication system vendors to establish or join IDHs. The difference in incentive structure, along with management agility due to the small firm size and flat organizational structure, makes IDHs more responsive to the market than average brand manufacturers. Another distinctive feature of the IDH business model is their effective reuse of basic models, which are conventionally called “product platforms” or solutions. For example, in 2008 a major IDH developed 200 products based on 37 product platforms.13 This product platform strategy, which in some respects is similar to the strategies of major TNCs such as Nokia, can greatly decrease the cost of developing new models. Because of these advantages, most of the major local handset firms combined in-house development with outsourcing to IDHs in order to construct an optimal product lineup. In spite of dramatic decline in the relative fortunes of local handset firms after 2003, IDHs continued to thrive and enjoyed relatively high profit margins until 2005 because of overall market growth and the expansion of their customer base from new entrants, including unauthorized manufacturers.14 As local handset firms’ sales grew along with the capabilities in the indigenous industrial base grew,15 competition in China’s handset market further intensified. Local manufacturers and IDHs felt mounting pressure to launch new products that appealed to the increasingly diverse and discerning preferences of Chinese consumers. This set the scene for the next stage of development—the emergence of new technology platforms aimed at filling the needs of local handset firms and IDHs in China.

2.4.4. Evolution of technology platforms and its consequences As discussed earlier, the development of 2G mobile handsets was based on technology platforms that had traditionally been provided by US or

60

Value Chain Creation and Reorganization in China

European semiconductor vendors. The lack of capabilities to assimilate these platforms and combine them with other key functions blocked local handset development in China, but when low cost, more highly integrated 2G technology platforms became available, Chinese IDHs with relatively meager capabilities could develop handsets and earn substantial profits, at least at first. However, as multi-functional 2.5G and 3G phones became the prime driver in the Chinese handset market, beginning around 2003, the new, more complex technologies required began to make it difficult for local handset manufacturers and IDHs in China to continue with handset development. Dominant platform vendors like TI were less interested in adjusting their platforms to meet the lower requirements of local Chinese firms because their principal focus was on serving the needs of global giants such as Nokia, which carry out most of their product development outside China and focus on high-end markets and leading-edge products.16 These circumstances created a market opening for MediaTek, a Taiwanese “fabless” semiconductor design firm that, due to its success in designing DVD-controller chipsets, had became the largest such firm in Asia.17 In the early 2000s, in an effort to diversify its business, MediaTek focused on the growing popularity of multimedia functions in the mainland’s handset market and began to provide highly integrated technology platforms to local handset firms and IDHs. By exploiting its expertise in multimedia data processing, the company integrated multimedia functions into their technology platforms. In this way, they could not only lower cost of PCBs, but also reduce the complexity of circuit design. At the same time, they packaged their technology platform with software for multimedia functions such as music and video players so their clients could quickly develop multi-functional handsets in a highly modular fashion. MediaTek’s platforms were adopted first among Chinese IDHs in 2004 because, in contrast to local brand manufacturers, they were daring and flexible enough to try a novel solution. Then, with the growing popularity of multimedia phones in the low- and middle-end markets, MediaTek chipsets were adopted by major manufacturers at remarkable speed, replacing TI as the dominant platform provider in China by 2005 and capturing nearly 40% of the domestic market for handset technology platforms. Our field research suggests that the great majority of unauthorized handsets designed by IDHs utilize MediaTek chipsets as well. The spectacular success of MediaTek opened the way for a highly competent local follower, Spreadtrum Communications, established in 2001

Ken Imai and Jing Ming Shiu 61

by Chinese returnee engineers from Silicon Valley with funding by local and overseas venture capital. In early 2004 Spreadtrum became the first Chinese firm to develop GSM baseband chipsets and launched its first multimedia-focused technology platform for low-end handsets the next year. By launching turnkey-type platforms similar to MediaTek’s at lower prices, Spreadtrum grew very rapidly. It began with sales to relatively minor IDHs and manufacturers (including unauthorized brands, as has been the case with MediaTek), but by 2006 larger manufacturers and IDHs began to adopt Spreadtrum’s platform as well. In 2007 Spreadtrum accounted for 10% of the baseband chipset market in China, becoming the second largest technology platform provider next to MediaTek. In the same period, other local fabless semiconductor firms started to launch products for handsets, such as multimedia processors. Local manufacturers and IDHs tend to welcome locally designed technology platforms because of their potential for a closer partnership, as long as their functionality, reliability, and future roadmap are secure. This suggests there is substantial market potential to be exploited by local fabless semiconductor firms in China.18 An important factor behind the rise of local fabless semiconductor firms such as Spreadtrum is the recent trend toward deepening modularization in semiconductor design and mobile communications technologies. Baseband chipsets and application processors are based on two core technologies: the digital signal processor (DSP) and the central processing unit (CPU). Spreadtrum sources these two core technologies, or so-called core IPs (functional modules of semiconductor circuits), from CEVA and ARM respectively, which are Israeli and British mainstream core IP providers.19 Vendors of computer-based semiconductor design tools (known as EDA, or so-called electronic design automation) and semiconductor foundries also provide fabless semiconductor firms with intellectual property and know-how for chipset design. While semiconductor design requires more advanced skills than handset development, the modularization of core technologies and design tools helps to explain the rapid rise of local fabless semiconductor firms. The most recent value chains patterns in China’s local mobile phone handset industry are presented in Figure 2.6. The notable change is the diversification of technology platform vendors as a result of the entry of MediaTek and Spreadtrum. Local firms in all three activities discussed in the paper—handset lead firms, IDHs, and fabless platform vendors— have now emerged. Transactions between turnkey technology platform vendors (MediaTek and Spreadtrum) and their clients are characterized by high levels of

62 Value Chain Creation and Reorganization in China Market

Product marketing

Product marketing

F

Product marketing

L

Handset development

Highly modular

Handset development

Handset development

Highly modular

Highly modular Core chipset (platform) development

F

Core chipset (platform) development

F

(Mediatek) : Firms’ boundaries

Figure 2.6

L

Modular

Modular

Incompletely modular

L

F: Foreign firms

Core chipset (platform) development

L

(Spreadtrum) L: Local firms

Patterns of value chains (3)—2005 to the present

modularity. As technology platform vendors encapsulate more and more functions beyond simple signal trafficking, the capabilities required for local manufacturers and IDHs to develop multi-functional phones decreases substantially. Initially, this was good news for local handset firms and IDHs desperate to launch a greater variety of products within a shorter period of time. However, as the adoption of turnkey technology platforms became widespread and the technological barriers to handset development were lowered over time, the most competent local firms found the economic value of their hard won capabilities seriously diminished. As turnkey technology platforms were widely adopted, the financial performance of these firms deteriorated. This trend has compelled technologically competent local manufacturers and IDHs to adjust their value chain strategies. As the value of their design capabilities declines, major IDHs have been shifting their emphasis to manufacturing management—taking responsibility for purchasing components, managing contract manufacturers, and selling finished products or PCBs to brand manufacturers. At the same time, many of them have forged strategic alliances with MediaTek or

Ken Imai and Jing Ming Shiu 63

Spreadtrum so that they can launch handsets utilizing new platforms earlier than their rivals.20 In addition, both local handset firms and IDHs have been intensifying their efforts to extend their customer bases to overseas markets, mainly in low-income countries. By pursuing these measures, they are trying to alleviate the drawbacks of increasing modularization in mobile phone handset design.

2.5. Value chain dynamics Based on the case study in the preceding section, we will now discuss the value chain dynamics behind the evolution of China’s mobile phone handset industry. The case study illustrates that two key factors have been driving the creation and reorganization of the value chains populated by local Chinese firms: the trend toward modularization in core technologies and the entrepreneurial responses of local firms to opportunities in the domestic market. A crucial point is that the fate of local firms has become increasingly dependent on patterns of change at the level of components and technology platforms. To sum up, many of the complex technologies and functions of mobile telecommunication became encapsulated into technology platforms in the 1990s. The increasing modularity in GSM technology platforms made it possible for local firms in China and other developing countries that lacked comprehensive knowledge of mobile telecommunication technologies to enter the handset development business. After the rise of Korean IDHs and Taiwanese ODM manufacturers, local Chinese handset design firms entered the market, assisted by institutional factors such as government policy and the availability of human resources in electronic engineering. Chinese firms, especially IDHs, were quick to learn how to assimilate emerging modularized technology platforms and develop products tailored to the needs of Chinese consumers that had previously gone unnoticed by TNCs. Today, China’s domestic market is characterized not only by its huge scale but also by the diversification of consumer demands, due largely to the wide income gap among different social strata and regions. Taking advantage of their proximity to the market, local firms have been able to detect opportunities in spite of their technological disadvantages. TNCs’ resurgence after 2003 and the constant entry of new local manufacturers greatly intensified competition in the Chinese handset market. In order to survive in an increasingly difficult competitive environment, local manufacturers and IDHs desperately needed to launch phones with new features within shorter time periods and at lower

64

Value Chain Creation and Reorganization in China

costs. This induced Taiwanese and local semiconductor vendors to develop turnkey technology platforms combined with additional functions such as multimedia. As the value of the design capabilities of IDHs has been eclipsed by the emergence of turnkey technology platforms, they have tried to adjust their position in value chains to take on roles similar to ODM manufacturers in Taiwan: to provide complete handsets for sale under the brand names of their customers. Thus, our case study illustrates how the trend toward modularization, and local firms’ entrepreneurial reactions to it, has not only driven growth, but set the character of the indigenous industrial base. In the early stages of the industry in China, the trend toward modularization took place only on a global scale and was completely exogenous to China’s industry. However, the rise of local value chains in China eventually called for further modularization to satisfy the needs of local firms. Although the customers of turnkey technology platforms are limited almost exclusively to Chinese firms at the time of this writing, exports of Chinese handsets using these platforms are reported to be increasing rapidly. This suggests that a country like China, with a huge domestic market and a growing indigenous industrial base, may have the potential to affect the configuration of global industries in the long run.

2.6. Conclusion In this chapter, we have examined the evolution of China’s mobile phone handset industry, adopting some of the key concepts of GVC theory. Beginning in the late 1990s, new value chains were created by local firms. While the trend toward modularization favored local firms that were technologically disadvantaged, the eventual drawbacks of high modularity made local firms’ value chains vulnerable to easy and rapid disruption. In spite of the rapid growth of the industry, the scope of innovation by local firms remains extremely limited. The precariousness of local firms and their limited innovative capacity, however, by no means denies the significance of China’s mobile phone handset industry. Beginning with marketing, which is the shallowest part of the value chain in terms of technological competence, local firms’ capabilities now extend to activities ranging from product design to technology platform development. The consequences of value chain evolution during the decade is a large and dynamic indigenous industrial base that is responsive to the domestic market, and perhaps to international markets with similar characteristics.21

Ken Imai and Jing Ming Shiu 65

To conclude, let us return to the issue presented in the introductory part of this chapter. Previous studies of the rise of the electronics industry in East Asia emphasized the importance of local firms in newly industrializing economies gaining access to GVCs organized by TNCs in order to upgrade their technological competence. Our study reveals that, in a large country with China’s somewhat unusual characteristics— a growing, fragmented market, a large supply of engineers, and an abundance of entrepreneurs ready to start own-brand firms—gaining access to TNCs’ networks is not the sole path to industrial upgrading. We explored how local firms can build on the increasingly modularized technologies provided by technology platform vendors in order to jump-start a significant indigenous industrial base. In the case of the mobile phone handset industry, these new value chains are characterized by modular transactions between local brand manufacturers, IDHs, and technology platform vendors. In spite of their trailing-edge approach to innovation in comparison with TNCs’ value chains, local firms appear to be capable of responding quickly to the diversified and ever-changing demands in China’s huge domestic market, and perhaps in other developing countries as well.

Notes 1. Source: The Yearbook of World Electronics Data 2008. 2. For a detailed review of the theory of GVC, see the Introduction and Chapter 7 of this volume. 3. In this respect, we follow the spirit of Vind and Fold (2007), which applies the GVC framework (“global production network” in their terminology) to multilayer structures of intra-industrial divisions of labor in Singapore’s electronics industry. 4. Of course, local firms’ advantages in marketing are not always guaranteed or permanent, as TNCs can potentially learn from their behaviors and performance. As we will see later, this is what has actually happened in China’s handset product market since 2003. 5. Yasumoto and Shiu (2007) elaborate on this point in a comparative case study of the mobile phone handset industry in China, Japan, and Taiwan. 6. 2G is the first generation of digital mobile telecommunication. The introduction of digital mobile telecommunication increased the capacity and efficiency of signal traffic in a revolutionarily manner. 7. We will return to this issue again in the next section. 8. This section is based mainly on (1) our interviews with local and foreign handset manufacturers, local IDHs, local and foreign semiconductor vendors, and local distributors conducted during 2004–2007, (2) corporate disclosure materials, and (3) secondary sources such as business reports and periodicals. 9. China also adopted CDMA 1x in 2001, a 2G standard used mainly in a few countries such as the United States and Korea. The share of CDMA 1x service,

66

10.

11.

12.

13. 14.

15.

16.

17.

18. 19. 20.

21.

Value Chain Creation and Reorganization in China however, remains at around 10%. Thus, our analysis in this chapter focuses essentially on the GSM handset market. As telecom operators promote value-added services based on customized handsets, the share of handsets sold through operators’ own channels has been rising recently, though it still remains at about 20% (including CDMA 1x phones). This new trend may potentially change the nature of competition in China’s mobile phone handset market, a point on which we cannot afford further elaboration in this chapter. The Japanese market, which is half as large as the Chinese market in terms of handset shipments, is commonly thought to be overcrowded with too many manufacturers, including foreign manufacturers whose market shares are almost negligible. Motorola (China) and ZTE, two companies that emphasize in-house human resource development, are known as the most important source of founders and senior engineers at local IDHs. SIM Technology, Annual Report 2008. The almost simultaneous overseas stock-listing of three IDHs—Techfaith, Longcheer, and SIM Technology—epitomized the success of Chinese IDHs at the time. Apparently, this does not mean that the actual amount of value captured by local firms increased (though some exceptionally successful firms achieved that) because product prices had been driven down by the heated competition in the product market. A number of Chinese customers of TI whom we interviewed pointed out that the TI technology platform had higher technological complexity and that the company’s insufficient adaptation to the Chinese market was a concern (from interviews with IDHs and handset manufacturers in August 2006). “Fabless” design firms use semiconductor “foundries” such as TSMC and UMC in Taiwan, or SMIC in China, to manufacture chips. Although this model is popular in the United States, most successful semiconductor firms in South Korea and Japan are vertically integrated, performing both design and manufacturing in-house. Interviews with local manufacturers and IDHs (August 2006). CEVA and ARM are the largest providers of embedded DSPs and CPUs, respectively. From the viewpoint of the turnkey technology platform vendors, the alliance enables them to utilize the knowledge of IDHs in mobile phone handset design and the final product market in the process of development of new platforms. One example is the recent boom of phones with digital TV function that adopt China’s original standard called CMMB. Processors for CMMB are supplied by several local fabless semiconductor firms.

References Baldwin, Carliss Y. and Kim B. Clark. 2000. Design Rules, Volume 1: The Power of Modularity. Cambridge, MA: The MIT Press. Borrus, M., D. Ernst and S. Haggard, eds. 2000. International Production Networks in Asia. London and New York: Routledge.

Ken Imai and Jing Ming Shiu 67 Ernst, Dieter and Paolo Guerrieri. 1998. “International Production Networks and Changing Trade Patterns in East Asia: The Case of the Electronics Industry.” Oxford Development Studies 26(2): 191–212. Ernst, Dieter and Linsu Kim. 2002. “Global Production Networks, Knowledge Diffusion, and Local Capability Formation.” Research Policy 31(8–9): 1417–1429. Funk, Jeffrey L. 2002. Global Competition Between and Within Standards: The Case of Mobile Phones. London: Palgrave. Gereffi, Gary, John Humphrey, and Timothy J. Sturgeon. 2005. “The Governance of Global Value Chains.” Review of International Political Economy 12(1): 78–104. Humphrey, John and Hubert Schmitz. 2004. “Governance in Global Value Chains.” In Local Enterprises in the Global Economy: Issues of Governance and Upgrading, ed. Hubert Schmitz. Cheltenham: Edward Elgar: 95–109. Langlois, Richard N. and Paul L. Robertson. 1995. Firms, Market, and Economic Change: A Dynamic Theory of Business Institutions. London: Routledge. Sturgeon, Timothy J. 2002. “Modular Production Networks. A New American Model of Industrial Organization.” Industrial and Corporate Change 11(3): 451–496. Sturgeon, Timothy J. and Richard K. Lester. 2004. “The New Global Supply-base: New Challenges for Local Suppliers in East Asia.” Chapter two in Global Production Networking and Technological Change in East Asia, ed. Shahid Yusuf, Anjum Altaf, and Kaoru Nabeshima. Washington, DC: The World Bank: 35–87. Sturgeon, Timothy J. and Ji-Ren Lee. 2005. “Industry Co-Evolution: Electronics Contract Manufacturing in North American and Taiwan.” Chapter three in Global Taiwan: Building Competitive Strengths in a New International Economy, ed. Suzanne Berger and Richard Lester.Armonk, NY: M.E. Sharpe: 33–75. Sturgeon, Timothy J. 2009. “From Commodity Chains to Value Chains: Interdisciplinary Theory Building in an Age of Globalization.” In Frontiers of Commodity Chain Research, ed. Jennifer Bair. Stanford: Stanford University Press: 110–135. Vind, Ingeborg and Niels Fold. 2007. “Multi-level Modularity vs. Hierarchy: Global Production Networks in Singapore’s Electronics Industry.” Danish Journal of Geography 107(1): 69–83. Yasumoto, Masanori and Jing Ming Shiu. 2007. “An Investigation into Collaborative Novel Technology Adoption in Vertical Disintegration: Interfirm Development Process for System Integration in the Japanese, Taiwanese, and Chinese Mobile Phone Handset Industries.” Annals of Business Administrative Science, 6: 35–70. Yeung, Henry Wai-chung, Liu, Weidong and Peter Dicken. 2006. “Transnational Corporations and Network Effects of a Local Manufacturing Cluster in Mobile Telecommunications Equipment in China.” World Development, 34(3): 520–540.

3 Value Chain Dynamics and Local Suppliers’ Capability Building: An Analysis of the Vietnamese Motorcycle Industry Mai Fujita

3.1. Introduction The Vietnamese motorcycle industry has experienced remarkable growth since the mid-1990s. Being a latecomer to the motorcycle industry in the Asian region, motorcycle production in Vietnam started in the mid-1990s with a very small market, heavy protection, and a nascent local supply base. The few Japanese firms with local production manufactured high-quality, expensive models that relied on imported components and were beyond the reach of ordinary Vietnamese customers. However, in less than ten years the country emerged as the world’s fourth largest market for and producer of motorcycles after China, India, and Indonesia.1 This remarkable transformation was triggered by the arrival of Chinese models that were assembled by local Vietnamese firms, often with the assistance of Chinese firms. Fierce competition between the Japanese industry leaders and local assemblers of Chinese motorcycles led to lower prices, increased local content, and rapid expansion of the market (The Motorbike Joint Working Group 2007; Fujita 2008). This chapter explores whether or not the growth of the Vietnamese motorcycle industry has been accompanied by the formation of a viable local component supply base. While it had long been argued that the Vietnamese supply base was seriously underdeveloped, recent research has shown that Japanese motorcycle manufacturers have started to source components from an increasing number of local suppliers, both at the first tier (Ohno 2005: 48) and at the second and third tiers (Nguyen Duc Tiep 2006). 68

Mai Fujita 69

Changes in the sourcing strategies of Japanese producers, however, tell only part of the story. After 2002 the development of the local supply base was also stimulated by the emergence of local Vietnamese motorcycle assemblers (Ha Huy Thanh et al. 2003). The sourcing strategies of Japanese motorcycle manufacturers and Vietnamese assemblers changed over time as they competed with each other and adapted to the changing market and policy environment in Vietnam. The key question is if and how local Vietnamese supplier firms developed firm-level capabilities through their supply relationships with Japanese and/or local assemblers in the context of rapid market growth and intense competition. This chapter examines this question by integrating the global value chain (GVC) approach, which seeks to characterize and explain the nature and dynamics of inter-firm relationships, and the technological capability (TC) approach, which is designed to characterize and explain the development of firm-level capabilities.2 The remainder of the chapter is organized as follows. Section 3.2 presents the analytical framework. Section 3.3 presents the global, regional, and local context of the industry. Sections 3.4 and 3.5 present the empirical analysis of local suppliers’ capability formation in the Vietnamese motorcycle industry. As the first step, Section 3.4 discusses how the changes in Japanese and Chinese motorcycle value chains since the late 1990s opened up new opportunities for local motorcycle component suppliers in Vietnam. As the second step, Section 3.5 analyzes how local suppliers exploited these opportunities to build their own capabilities. The concluding section provides a summary of findings and discusses the implications of the case study for the GVC and TC approaches.

3.2. Analytical framework The analytical framework used in this chapter combines two distinct yet closely related approaches: the GVC and TC approaches. The first element of the framework draws on the GVC approach. Using the notion of “chains” to refer to the sequence of value-adding activities required to bring products or services to market, this approach highlights the character of inter-firm linkages, especially those that extend across borders.3 Gereffi et al. (2005) present a simple framework, independent of industries and geographical, historical, or social contexts, to analyze the patterns by which inter-firm relationships are coordinated. They have extracted three fundamental variables that shape patterns of inter-firm coordination: (1) complexity of information exchanged in transactions,

70

Value Chain Dynamics and Local Capability Building in Vietnam

(2) codifiability of the information exchanged, and (3) suppliers’ capability levels relative to the requirements of transactions. Gereffi et al. (2005) argue that different combinations of these variables result in five basic types of inter-firm coordination, with markets and hierarchy (vertical integration) at the two opposite ends of the spectrum, and three intermediate forms of coordination (modular, relational, and captive) falling between, each with ascending requirements for explicit coordination. Of the three intermediate forms, captive and relational chains play important roles in the case study presented in this chapter. In captive chains, smaller suppliers are transactionally dependent on larger customers. Captive chains emerge when product specifications are complex but can be codified in the form of detailed instructions, and low supplier competence calls for a great deal of intervention and control on the part of the customer. Relational chains, by contrast, involve complex interactions between customers and highly competent suppliers, which create mutual dependence between the two parties. This form of coordination emerges when complex product specifications that cannot be codified force firms to exchange tacit knowledge, and when the supplier is able to provide complementary competencies to the customer (Gereffi et al. 2005). The second element of the framework is based on the TC approach (Lall 1992; Bell and Pavitt 1997). Following Bell and Pavitt (1997: 89), this chapter defines “firm-level capabilities” as “resources needed to generate and manage technical change, including skills, knowledge and experience, and institutional structure and linkages.” A key insight of the TC approach is that accumulating firm-level capabilities is a complex and specialized process that requires the firms’ deliberate efforts such as explicit intra-firm training and managed experience accumulation (Bell and Pavitt 1997). This chapter analyzes capabilities in two dimensions: their functions and levels. The analytical approach pioneered by Lall (1992) and adapted by Ariffin (2000), Figueiredo (2008), and others has been tailored to the attributes of the motorcycle industry. One of the original contributions of this framework is to integrate the GVC perspective with the classification of functional categories. By mapping the functions to several key functions in the motorcycle value chain, this framework seeks to highlight how the division of labor between the lead firm and the supplier affects the functional scope of capabilities acquired by the suppliers. Specifically, the key functions in the motorcycle value chains are classified into (product) planning, production, and marketing. Production capability is further divided into equipment-related capability, which directly deals with the operation, maintenance, design, and

Mai Fujita 71

manufacture of machinery and equipment (i.e., the “hard” aspects of production), and production management capability, which is concerned with effective management of the various elements of production including equipment, materials and components, human resources, and information (i.e., the “soft” aspects of production).4 This results in four functional categories of capabilities: 1) Planning capability: The capability to conduct market research, develop product concept, develop and design a new product according to the needs of the market. 2) Equipment-related capability: The capability to operate, maintain, design, and manufacture equipment, dies and molds, jigs and tools. 3) Production management capability: The capability to organize and manage the various elements of production. 4) Marketing capability: The capability to market products so as to strengthen the relationship with customers, develop own brand, and explore markets. The levels of capabilities are designed to reflect the process of capability building as experienced by developing country firms. Typically, these firms first import mature, standardized technology and then adopt more advanced technologies over time. Simple as this might seem, it generally takes a great deal of time and tedious effort for developing country firms to master imported technology and maintain stable operations on their own (Kim 1997, 2004). With time, firms can gradually start to adapt technology to their needs in the domestic market, improve on it, and finally create new technology of their own (Lall 1992; Bell and Pavitt 1997). Accordingly, the framework sets the following four levels of capabilities: 1) Operational level: Operating the existing technology. 2) Assimilative level: Mastering the existing technology and maintaining the operation over time. 3) Adaptive level: Making original improvements to the existing technology prevailing among local producers in the country concerned. 4) Innovative level: Creating something new with significant elements of originality and novelty compared to the existing technology prevailing in the world. The above classification of functions and levels of capabilities generates a two-dimensional Capability Matrix (Sato and Fujita 2009). Table 3.1 presents the matrix adapted to the capability building steps typical of

72

Table 3.1 Capability Matrix for analyzing motorcycle components suppliers Pre-production

Production

Post-production

Planning

Equipment-related

Production management

Marketing

Operational

Replication of an existing/given product in the domestic market by recreating the design drawings.

Basic operation of machinery and equipment, dies, molds, jigs, and tools to process components to the level required in the domestic market.

Routine production management required in the domestic market.

Routine marketing methods/activities.

Assimilative

Replication of existing international-standard products by recreating the design drawings.

Processing components and manufacturing dies, molds, jigs and tools to the level required by foreign customers; maintenance and repair of machinery and equipment, dies, molds, jigs, and tools.

Maintaining stable production management fulfilling the levels required by foreign customers.

Adopting and stably managing improved marketing methods/ activities.

Adaptive

Making original improvements to the existing products prevailing among local producers.

Making original improvements to the existing machinery and equipment prevailing among local producers.

Making original improvements in production management practices prevailing among local producers.

Making original improvements to marketing methods prevailing among local producers; developing brands recognized in the domestic market.

Innovative

Planning and designing of new products with significant elements of originality and novelty compared to the existing products prevailing in the world.

Designing and developing new machinery and equipment with significant elements of originality and novelty compared to the existing machinery and equipment prevailing in the world.

Establishing production management system so as to achieve the world’s topmost level in production management.

New marketing methods with significant elements of originality and novelty to explore new markets abroad; establishing internationally recognized brands.

Source: Sato and Fujita (2009).

73

74

Value Chain Dynamics and Local Capability Building in Vietnam

motorcycle component suppliers. In this chapter, this matrix will be used to analyze how capability levels of local suppliers for different functional value chain categories have changed over time.

3.3. Industry settings This section presents the global, regional, and local Vietnamese context of the low-displacement motorcycle industry. 3.3.1. Global and regional context Low-displacement motorcycles often serve as an important means of daily transportation in developing countries, particularly in Asia. The presence of substantial market potential for motorcycles, combined with the opportunity to stimulate the development of component industries, has led governments in developing countries to implement import-substitution policies and local content rules. At the global level, production is highly concentrated in the hands of four large Japanese motorcycle manufacturers: Honda, Yamaha, Suzuki, and Kawasaki. After firmly establishing their positions in Japan, these four companies expanded overseas operations via exports and later, because of growing barriers to trade, foreign direct investment (FDI). By the mid-1980s these four firms had captured roughly half of the global market.5 This highly concentrated market structure is closely related to the nature of motorcycle technology. Since Honda’s launch of the highly successful “Super Cub” in 1958, there have been few if any radical product innovations. Fifty years on, the basic technology remains the same and the integrated nature of product design is unchanged. Thus, the focus of innovation has been incremental process improvements (Abernathy and Utterback 1978: 44) and the continuous improvement of “quality, costs, and delivery” (QCD) has become critical in achieving a competitive advantage. The leadership of Japanese motorcycle manufacturers has been sustained primarily by the superior QCD performance of their internal operations and in closely aligned component suppliers. In Southeast Asia the market leadership of Japanese motorcycle manufacturers has been firmly in place since the 1960s. Local content rules in each country led them to undertake significant FDI for production within each country. While most core components sourced in the host country are from subsidiaries of Japanese suppliers which have also engaged in FDI, Japanese motorcycle manufacturers have also nurtured local suppliers. Since basic designs for the models produced in Southeast Asia are developed in Japan and Thailand,6 the majority of suppliers

Mai Fujita 75

in the Southeast Asian region, with the exception of a limited number of Japanese suppliers in Thailand that participate in co-design of components, are assigned the task of manufacturing components in accordance with the detailed drawings and QCD requirements provided by the motorcycle manufacturers, who closely monitor and control the suppliers’ performance. The supplier, in turn, dedicates most of its output to the motorcycle manufacturer. In the terminology of the GVC approach, linkages of this sort are typical examples of “captive chains,” whereas linkages between the Japanese motorcycle manufacturers and a limited number of core component suppliers that have proprietary technology and participate in co-design of components in Japan are characterized as “relational chains.” Since the late 1990s Japanese motorcycle manufacturers have been challenged by the rise of indigenous motorcycle manufacturers from the emerging economies. The rise of Chinese firms has been particularly threatening because these firms dominate their huge domestic market and base their competitiveness on a set of capabilities different from that of Japanese firms. Chinese indigenous motorcycle manufacturers produce low-priced imitations of Japanese base models, sometimes with minor improvements. In a market where intellectual property rights are weakly protected, several very popular base models including the Honda Cub mentioned earlier, which had been introduced to Chinese state-owned motorcycle manufacturers in the 1980s under formalized technological transfer agreement and came to be copied by numerous private motorcycle manufacturers in the 1990s, have become industrywide de facto standard designs (Ohara 2001: 32).7 Competition among these largely undifferentiated, homogenous products is based primarily on price. The emergence of Japanese base models as a de facto standard in the industry enabled firms to operate with weak forms of coordination. De facto standardization of the motorcycle and its component parts allows motorcycle manufacturers and suppliers to engage in arm’slength transactions with little coordination, even though these designs remain integral and proprietary. This apparent puzzle—the emergence of arm’s-length transactions where the product design is integral and proprietary—is explained by the standardization of components used to produce de facto standard models. It is also enabled by the low product quality requirements in the Chinese market, which significantly lessens the need for coordination between assemblers and suppliers.8 Following the GVC approach, the low complexity of information exchanged in transactions involving standardized products enables market-based value chains to prevail in China.

76 Value Chain Dynamics and Local Capability Building in Vietnam

The challenge to the Japanese industry leaders posed by the Chinese firms began in China in the 1990s and then spread overseas. The first major external market where the Chinese motorcycle manufacturers established a strong presence was Vietnam. 3.3.2. The development of the motorcycle industry in Vietnam The history of the Vietnamese motorcycle industry has passed through three stages.9 The first, from the middle to the end of the 1990s, was the start-up stage. Vietnam promoted the domestic production of motorcycles by erecting barriers to imports and incentives for foreign motorcycle manufacturers to invest in the country. By the late 1990s one Taiwanese and three Japanese motorcycle manufacturers (Sanyang, Suzuki, Honda, and Yamaha) had invested in Vietnam. However, sales failed to expand because the prices of their motorcycles were very high when compared with the average income level of the population (Figure 3.1). This situation was completely transformed in the second stage (2000– 2004). During the years 2000 and 2001, massive numbers of low-priced motorcycles were imported from China, a time referred to by incumbent producers as the “China shock.” Because the Vietnamese government had prohibited the import of assembled vehicles since 1998, Chinese

20

2000 15 1500 10 1000 5

Sales (mopeds) Figure 3.1

Sales (scooters)

2007

2006

2005

2004

Registered motorcycles/ population

Growth of Vietnam’s motorcycle market

Source: Honda Motor Co., Ltd., various years.

2003

2002

2001

2000

1999

1998

1997

1996

1995

1994

0

1993

500

1992

Sales (thousand units)

2500

0

Registered motorcycles/ population (%)

25

3000

Mai Fujita

77

imports arrived in the form of “knockdown” component kits. Soon, more than 50 local firms were engaged in the assembly of Chinese motorcycle component kits (hereafter referred to as “local assemblers”) that were, in essence, imitations or slightly modified versions of the older Japanese base models already mentioned. With prices as low as one-third to one-fourth of Japanese-brand models, “Chinese motorcycles”10 quickly penetrated medium- and low-income consumer markets in urban and rural areas that had remained unexploited by Japanese firms. This led to a rapid expansion of the market. By 2001 local assemblers of Chinese motorcycles held an 80% share in the expanded market in Vietnam (Figure 3.2). The China shock provoked a series of reactions from incumbent producers and policymakers. The Japanese motorcycle manufacturers, 2500

Sales (thousand units)

2000

1500

1000

500

0

1998

1999

2000

2001

Local assemblers & others

2002

2003

2004

2005

Imported scooters

VMEP (a wholly-invested subsidiary of Taiwan’s Sanyang Motor) Vietnam Suzuki

Yamaha Vietnam

Honda (imported)

Honda Vietnam

Figure 3.2 Sales of motorcycles by manufacturers Source: The Motorbike Joint Working Group (2007).

78

Value Chain Dynamics and Local Capability Building in Vietnam

perceiving Vietnam as a symbol of an expanded Chinese threat that had already become apparent in the Chinese market, initiated companywide efforts to regain market shares. In 2002 Honda Vietnam (HVN) launched a new model, the “Wave Alpha,” priced at approximately one-third of its previous models. In the policy arena, the Vietnamese government reacted by engaging key policy changes to restore order and to promote sound development of the industry, such as stepping up the enforcement of local content rules and import tariffs, which had been circumvented by local assemblers, and introducing product quality and environmental standards. In a desperate attempt to prevent an uncontrolled proliferation of motorcycles on Vietnam’s streets, the government also resorted to more direct interventions including quantitative restrictions on the number of motorcycle components imported and a ban on the registration of motorcycles in certain localities. The fact that many of these interventions were implemented in an arbitrary manner, often going against the commitments previously announced by the government, severely affected business confidence.11 Moreover, foreign motorcycle manufacturers suffered, in particular, because they were prevented from investing in additional production capacity beyond the original plans granted by the Vietnamese authorities upon receiving FDI license.12 As a result, although the newly launched “Wave Alpha” rapidly gained popularity in the Vietnamese market, the sales of Japanese motorcycle manufacturers did not expand rapidly during this period (Figure 3.2). The third stage (2005–2008) was a period of FDI-led development. As Vietnam stepped up efforts to enter the World Trade Organization (WTO) and implemented its tariff reduction commitments under the ASEAN Free Trade Agreement, it dismantled its prohibition of motorcycle imports,13 abolished local content rules, and reduced tariffs on components. At the same time, the government abolished the restrictive rules on motorcycle registration and investment in motorcycle production that had been introduced during the previous period. This move, combined with the country’s economic boom, significantly enhanced domestic motorcycle sales, which climbed to 2.8 million units in 2007, far exceeding the figures during the China shock. Even though import restrictions were relaxed under WTO rules and AFTA commitments, Japanese firms chose to satisfy the growing market in Vietnam via FDI for local production, following their conventional approach to localize production in countries with large demand for their products. However, local Vietnamese assemblers, despite the challenge posed by the radical price reductions introduced by Japanese motorcycle manufacturers and changes in the

Mai Fujita 79

policy environment, still held roughly one-third of the market as of 2006.

3.4. Value chain dynamics in the Vietnamese motorcycle industry This section examines the inter-firm dynamics in the industry. The analysis focuses on two sets of value chains: (1) those coordinated by Japanese motorcycle manufacturers, which will hereafter be called “Japanese chains” (and which are represented most significantly by the captive value chains driven and coordinated by the dominant Japanese firm, HVN), and (2) the market-based value chains coordinated by local assemblers often in cooperation with Chinese firms, which will be referred to as “Vietnamese–Chinese chains.” 3.4.1. Japanese chains: emergence and consolidation of HVN’s captive chains As discussed in Section 3.3.1, the “captive” nature of the Japanese chains was fundamentally shaped by the integrated nature of product design and the global and regional configuration that compartmentalized relational assembler–supplier interactions in countries outside of Vietnam. Since the operations of the Vietnamese subsidiaries of Japanese firms focused on production and distribution, not product development or design, suppliers in Vietnam were assigned a narrow range of components manufacturing activities and operated under the strict control of powerful industry leaders. It took time for HVN to establish captive chains in Vietnam. During the industry’s start-up stage in the 1990s, Japanese firms engaged in very low levels of local sourcing as a result of their small production scale and Vietnam’s limited industrial capabilities. By the end of the 1990s, HVN’s local content ratio had risen to slightly over 50%, but this was largely achieved through in-house component production and the arrival of a few Japanese suppliers. However, at the time when it launched its Vietnamese operations, HVN selected a set of local firms as suppliers, none with prior experience in supplying Japanese companies. Since HVN was not under severe competitive pressure at the time, suppliers were given ample time to study HVN’s requirements and endeavor to meet them. In the second stage (2000–2004), new opportunities opened up for local firms to be selected as suppliers to HVN. A major impetus was Honda’s company-wide drive to fight the competition from “Chinese motorcycles” by launching the Wave Alpha in 2002, and to abide with

80

Value Chain Dynamics and Local Capability Building in Vietnam

the local content rules that had been implemented by the Vietnamese government around 2000. With the development of the Wave Alpha, incumbent suppliers were notified that HVN would buy components from suppliers of any nationality as long as the components reached the required QCD levels. Apart from sourcing some components from suppliers in China, Honda launched an extensive search for suppliers in Vietnam by dispatching experts from Japan and started to source components from a greater number of suppliers in Vietnam. Many non-Japanese firms, especially Taiwanese, Korean, and Vietnamese companies, became HVN’s suppliers at this stage (Figure 3.3). As discussed in Section 3.3.2, although the Wave Alpha rapidly gained popularity, HVN was held back from making investments in production capacity expansion at this stage. For this reason, the volume of HVN’s orders stagnated and the suppliers’ dependence on HVN remained low. The development of the Wave Alpha, combined with the local content rules, also brought about an increase in the number of second-tier suppliers. As first-tier suppliers came under pressure to reduce costs, they attempted to replace imported components with locally sourced ones, and components sourced from Japanese suppliers with ones sourced from Taiwanese or Vietnamese suppliers. The author interviewed six suppliers in Vietnam from Japan, Taiwan, and Korea in 2004 and 2005. They used a total of 162 second-tier suppliers, at least 106 of which were Vietnamese firms.14 During the third stage, restructuring and consolidation of Japanese value chains took place alongside the remarkable expansion of HVN’s production scale in Vietnam, from roughly 400,000 units per year in 2002–2003 to over 1 million units in 2007 (Honda Motor Co., Ltd. 2008). The number of Japanese suppliers increased rapidly with a new wave of FDI, while the persistent need for cost reduction also pushed HVN to look for more local and other non-Japanese suppliers. As competition among suppliers intensified, less competitive suppliers faced decreasing orders from HVN.15 Scale expansion and intensified competition meant that suppliers faced greater pressure to increase volumes, while meeting tighter delivery requirements, lowering prices, and maintaining (or even improving) quality levels. Because orders increased, first-tier suppliers’ dependence on HVN increased dramatically. Scale expansion also compelled first-tier suppliers to outsource a part of their production to meet their production plans without excessive investment in their own factories. Overall, while more opportunities opened up for local second-tier suppliers at this stage, requirements on first-tier suppliers became increasingly demanding. Competition intensified at all levels.

81

(1) Local content ratio and number of suppliers 70

100

60

80 70

50

60

40

50 30

40 30

20

20 10

10 0

1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Number of suppliers in Vietnam

70

Local content ratio (%)

Number of suppliers

90

0

Local content ratio

(2) Number of suppliers in Vietnam by nationality

Number of suppliers

60 50 40 30 20 10 0 2001 Vietnamese

2004 Taiwanese/Korean

2007 Japanese

Figure 3.3 HVN’s local sourcing Source: Prepared by the author based on the author’s interviews and presentation material on “Requirements of Japanese Companies in Parts and Components Procurement from Local Companies” at the seminar on Vietnam-Japan SI Business Promotion held in Hanoi on January 22, 2007.

82 Value Chain Dynamics and Local Capability Building in Vietnam

3.4.2. The Vietnamese–Chinese chains: evolving market-based relationships In 2000, 51 local assemblers assembled 1.37 million motorcycles in Vietnam.16 The majority operated on a small scale, producing fewer than 40,000 units per year (Figure 3.4). Although most of these assemblers began by assembling component kits imported from China, they increased local component sourcing because: (1) Chinese kits often had missing or defective components,17 and (2) the Vietnamese government introduced local content rules, even though these were weakly enforced until 2002. Furthermore, the proliferation of Chinese motorcycles created a huge demand for replacement components. In all, the demand for standardized motorcycle components without high precision or quality 3000

Production (thousand units)

2500

2000

1500

1000

500

0

2001

2002

2003

2004

2005

2006

Foreign motorcycle manufacturers

Local assemblers producing less than 10,000 units/per year (27 firms in 2005)

Local assemblers producing 20,000–40,000 units/per year (10 firms in 2005)

Local assemblers producing over 40,000 units/per year (6 firms in 2005)

Figure 3.4

Production of motorcycles by size of production

Source: The Motorbike Joint Working Group (2007).

Local assemblers producing 10,000–20,000 units/per year (14 firms in 2005)

Mai Fujita 83

requirements created an impetus for numerous Vietnamese and foreign firms to produce motorcycle components. The bulk of the motorcycle components used in the emerging Vietnamese–Chinese value chains followed designs of a few older Japanese base models.18 As in the Chinese motorcycle industry, the de facto standardization of these base models and low quality requirements enabled many assemblers and suppliers to engage in arm’s-length market transactions. Switching of suppliers frequently took place, mainly on the basis of price. After 2002 local assemblers were confronted with a number of challenges. On the market front, HVN fought back and rapidly recovered its market share. On the policy front, the government stepped up enforcement of local content rules and introduced new regulations requiring motorcycle assemblers and manufacturers to invest in in-house production of key components. While many local assemblers either abandoned or significantly reduced production, some invested in in-house production of components and promoted local sourcing with the aim of becoming fully fledged motorcycle producers, rather than assemblers of Chinese kits. From 2005 onwards, local content rules were abandoned and replaced with import tariffs on components, which alleviated the pressure for local sourcing to some extent. Competition with foreign assemblers, however, intensified further. These market and policy changes triggered the consolidation of local assemblers and their suppliers. By 2006 several large-scale firms had captured the bulk of the market share for local assemblers (Figure 3.4). In fact, four local assemblers produced more than 100,000 units per year in 2005 (Institute for Industry Policy and Strategy 2007: 36). The author’s fieldwork in 2004 and 2007–2008 found that firms with the most impressive growth records focused on the low-end segment of the market, selling models priced at roughly 40% of HVN’s Wave Alpha, incorporating minor improvements that affected only the external appearance of products (e.g., plastic side covers, frames, and engine covers).19 On the suppliers’ side, too, consolidation was taking place. A small number of very large suppliers emerged, producing components in large quantities and selling them to many local assemblers. The majority of 18 Vietnamese, Taiwanese, and Chinese suppliers to local assemblers interviewed by the author during three rounds of fieldwork in 2007–2008 had lost orders from local assemblers by 2007. However, five (one Vietnamese-owned and four Chinese-owned) had started to supply components to local assemblers on a relatively large scale. In particular, those supplying plastic covers, frames, and silencers conducted their own

84

Value Chain Dynamics and Local Capability Building in Vietnam

market research and made external modifications to the standardized component design. One Chinese supplier sold 860,000 plastic covers and frames per year to 43 local assemblers.20 These figures are equivalent to 74% of the total number of motorcycles produced by local assemblers and 84% of operating local assemblers as of 2006. The consolidation of the Vietnamese–Chinese chains during this period was the result of assemblers’ and suppliers’ strategies. Exploiting the de facto standardization of the Japanese base models to sell components to many assemblers was an attractive strategy for suppliers because the small and (initially) fragmented market structure meant that pooling orders from many assemblers was the only way to achieve sufficient economies of scale. For local assemblers who lacked component design capabilities, it made economic sense to rely on large and highly efficient suppliers. The largest assemblers wanted to use large suppliers to help them achieve the mass production of low-priced motorcycles with regular modifications to meet rapidly changing market demand. In a pattern similar to that found by Imai and Shiu in the case of local mobile phone handset producers in China, local Vietnamese assemblers succeeded by targeting low-income consumers in the rural market, which the foreign motorcycle manufacturers had not been able to penetrate. For suppliers, however, the design and production capabilities needed were not easy to acquire, as will be discussed in the following section.

3.5. The buildup of firm-level capabilities in local suppliers This section turns the focus to the local Vietnamese component suppliers. It will examine how they exploited the opportunities created by the value chain dynamics discussed in the previous section for accumulating their firm-level capabilities. 3.5.1. Research methodology This section presents an in-depth case study of 17 local motorcycle component suppliers. Since this research explores how local suppliers accumulated their capabilities as they participated in different value chains, the cases were selected to include suppliers that participated in Japanese and/or Vietnamese–Chinese chains at different points in time.21 In an effort to ensure the quality and accuracy of retrospective data (i.e., late 1990s–2008), priority was placed on those suppliers that had been previously interviewed or surveyed by the author in the period 2003–2005.22 To enable comparison, the cases were limited to suppliers

Mai Fujita 85

specialized in production of a small set of components and production processes: 1) suppliers of metal or plastic components, 2) firms that specialized in specific production processes such as plating, and 3) suppliers of dies and molds. Profiles of all 17 firms are presented in Table 3.2. Data on the 17 suppliers were collected via interviews carried out by the author in September and/or November 2008. The interviews consisted of two parts. The first focused on the overall business performance, product and customer structure, and relationships with main customers since the late 1990s. The second focused on “tracking” the capability-building trajectories of the firms using the following two-stage procedure. 1) Interviews began by identifying the status of the firms’ capabilities at the “point of departure,” that is, just before they started to produce motorcycle components in either Japanese or Vietnamese–Chinese chains. Questions were asked about how each of the four value chain functions in the Capability Matrix (Table 3.1) was conducted at this stage. 2) Having identified the point of departure, questions were then asked about how the ways in which the four functions were conducted changed over time. An important aim of this line of questioning was to identify major “learning events,” which are defined as events that involved either progress from a certain level to a higher level of capability or substantial progress within a given level of capability. For example, an event may have been associated with the launch of a new product or technology, or company-wide initiatives to substantially increase QCD levels in response to customer demands. Some events may have been short term, while others may have extended over years. In many cases, a particular event involved changes in the levels of capabilities in more than one function. Although the goal was to identify three key learning events per firm, the research revealed two to four events per firm, depending on its growth path. 3.5.2. Respondent firms’ value chain participation Figure 3.5 shows the patterns of value chain participation by respondent firms during 1996–2008, as well as the relative importance of the firms’ major products/customers as a percentage of total sales in 2008. As of the mid-1990s, only 3 of the 17 firms (A, B, and E) participated in motorcycle value chains. As opportunities for local firms in motorcycle value chains increased in the period 1996–2008, some suppliers (K, L, M, and N) participated in both Japanese and Vietnamese–Chinese

86

Table 3.2 Profile of respondent suppliers Motorcycle value chains A B C D E F G

J (1st-tier) J (1st-tier) J (1st-tier) J (1st-tier) J (1st-tier) J (1st tier) J (2nd tier)

H I J K L M N O P Q

Establishment (start of business)

Ownership

Number of employees

Revenue (2007, bil. Dongs)

1974 1972 1974 1994 1980 2004 1969

State State State Private State Private State

1,350 1,000 1,000 500 550 81 1,100

374 248 297 88 156 n.a. 350 (2008)

J (2nd tier) J (2nd tier) J (2nd tier) V-C→ J (1st tier) V-C→ J (2nd tier)

(1988) 2002 1994 1980 1960 2001

Private Private State State Private

150 182 850 600 150

24 48 254 150 30

V-C→ J (2nd tier) V-C→ J (2nd tier) V-C V-C V-C

(1986) 2001 (1981) 2000 (1987) 1995 (1959) 1997 (1996) 1999

Private Private Private Private Private

200 400 450 150 170

130 101 32 >10 40

Types of components

Metal components Metal components Metal engine components Wire harnesses Plastic components Molds Crankshafts (forging process only) Plating Plastic components Plating Metal components Clutches, Aluminum diecast components Aluminum diecast components Silencers, wheels Metal engine components Metal components Metal components (silencers)

Note: Type of motorcycle value chains: “J” denotes Japanese chains, and “V-C” denotes Vietnamese–Chinese chains. Source: Prepared by the author.

87

1996

A B C

D

E F G H I J

K

L

M

N

O

P

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

Share in total 2008 revenue (2007)

Replacement components for bicycles HVN (1st & 2nd tier) Stainless steel products for export HVN, Yamaha Vietnam (1st & 2nd tier) Agricultural machinery & components for domestic market HVN, Yamaha Vietnam (1st & 2nd tier) Wire harnesses for export HVN (1st tier) Wire harnesses for electronic & auto makers in Vietnam HVN, Yamaha Vietnam (1st & 2nd tier) components for electronic products, construction products Household products HVN (2nd & 1st tier) Diesel engines, machinery components on OEM basis (domestic market exports) HVN (2nd tier) Yamaha Vietnam (ceased after 2004), HVN, Vietnam Suzuki (2nd tier) Replacement components Machinery components for mainly Japanese customers Consumer electronics (ODM) Plastic containers and household products HVN &Yamaha Vietnam (2nd tier) Metal furniture (domestic market exports) HVN & Yamaha Vietnam (2nd tier) Bearings Local assemblers HVN (1st & 2nd tier) Local assemblers HVN (2nd tier) Replacement components, etc. Other foreign (VMEP) Other foreign (VMEP) HVN (2nd tier) Local assemblers Replacement components Bicycle components Replacement components Local assemblers HVN (2nd tier) Replacement components Local assemblers Local assemblers Bicycle Other foreign (VMEP) Compo Machinery & components nents (domestic market & exports) Local assemblers

Q

Other foreign (VMEP, Lifan)

Note: (1) Not all chains are included (i.e., the total does not necessarily equal 100%). (2) Shaded cells denote the types of chains as follows. Japanese motorcycle manufacturers

Local assemblers

Other foreign motorcycle assemblers

Other Industries

Figure 3.5 Chain participation of respondent suppliers Source: Author’s fieldwork.

0% 90% 20% 80% 30% 70% 20% 60% 20% 50% 50% 100% 30% 33% 45% 45% 90% 10% 90% 10% 20% 0% 80% 0% 60% 40% 20% 50% 30% 10% 20% 20% 30% 95% 5% 0% 30% 70% 10% 85%

88

Value Chain Dynamics and Local Capability Building in Vietnam

chains (either simultaneously or shifting from the latter to the former). All the suppliers, except for F, did not specialize in motorcycle components; they also produced products such as agricultural machinery or household products. A number of observations can be made about the suppliers’ participation in Japanese and Vietnamese–Chinese value chains. Regarding the first-tier suppliers to HVN, there were two main periods when chain participation began. For suppliers A, B, and E it was the mid-1990s, when HVN started its operations in Vietnam. For suppliers C, D, G, and K, it was the period 2000–2004, within a few years of HVN’s launch of the Wave Alpha in 2002. For all of these suppliers, once a relationship was established with HVN, it lasted for a long time. As HVN’s production expanded rapidly after 2005, most of the first-tier suppliers became highly dependent on them. A, B, and K depended on HVN for as much as 80% of their total sales in 2008. The story was slightly different for HVN’s second-tier suppliers. The majority (F, G, H, I, J, L, and M) joined HVN’s chains between 2000 and 2004. Whereas all the first-tier suppliers except D and F were large stateowned enterprises with the machinery and human resources to produce metal parts, the majority of second-tier suppliers were smaller private firms engaged mainly in the production of replacement components, bicycle components, or other small metal or plastic products. Unlike the first-tier suppliers, the second-tier suppliers’ dependence on HVN remained low. Suppliers K–Q joined Vietnamese–Chinese chains between 1996 and 2002, when local assemblers emerged and expanded their market shares. These firms were smaller private firms in related industries. Although they expanded the production of motorcycle components in the period 2000–2002, by 2008 most of these suppliers had either stopped supplying components to local assemblers or significantly reduced the volume of transactions. As they faced diminishing orders from the local assemblers, some suppliers in the Vietnamese– Chinese chains entered the Japanese chains as either first-tier (K) or second-tier (L, M, and N) suppliers. Others either shifted back to traditional markets or explored new markets in other industries. The only exception was supplier M. Like the Chinese supplier discussed in Section 3.4.3, supplier M was able to sustain a relationship with local assemblers because it took its own initiatives to conduct market research and make modifications to the standardized component design. These modifications were neither requested nor specified by the local assemblers.

Mai Fujita

89

Overall, the value chain dynamics at first created new opportunities for an increasing number of local firms, including small firms without abundant capital or technological and human resources, but these opportunities soon diminished for all but a handful of suppliers. 3.5.3. Local suppliers’ capability building 3.5.3.1. Status of firm-level capabilities in 2008 A “snapshot” of respondent suppliers’ capabilities as of 2008 is provided in Figure 3.6. For virtually all the suppliers interviewed, the point of departure was the pre-operational or operational level, since they either had not started business or were producing only simple items such as household products or bicycle components for the domestic market. Given this basic starting point, the attainment of capabilities shown in Figure 3.6 is a remarkable achievement. While none of the respondent firms reached the innovative level in any of the four functions, many reached the assimilative or even adaptive level for one or more functions. At the same time, the figure reveals diversity of functions in Motorcycle value Firm chain participation A

J (1st tier)

B

J (1st tier)

C

J (1st tier)

D

J (1st tier)

E

J (1st tier)

F

J (1st tier)

G

J (2nd tier)

H

J (2nd tier)

I

J (2nd tier)

J

J (2nd tier)

K

V-C → J (1st tier)

L

V-C → J (2nd tier)

M

V-C → J (2nd tier)

N

V-C → J (2nd tier)

O

V-C

P

V-C

Q

V-C

Planning Op

As

Ad

Production management

Equipment In

Op

As

Ad

In

Op

As

Ad

Marketing In

Op

As

Ad

In

n.a. n.a.

Figure 3.6 Levels of capabilities reached by respondent firms in 2008 Notes: (1) Levels of capabilities are as follows: Op: Operational; As: Assimilative; Ad: Adaptive, In: Innovative. (2) Dark-shaded cells denote full mastery of the respective level of capabilities, while lightshaded cells denote incomplete mastery of the respective level of capabilities. Source: Author’s fieldwork.

90

Value Chain Dynamics and Local Capability Building in Vietnam

which capabilities were acquired, as well as in the levels reached, even among the suppliers that participated in the same value chains. In order to look more closely at the learning process that enabled respondent firms to reach the capability levels they did, I analyze the “learning events” they experienced between 1996 and 2008. For 17 respondent firms, 50 learning events were identified. Table 3.3 presents the features of the learning events classified according to the value chains in which the events took place. 3.5.3.2. Learning in Japanese chains Table 3.3 shows that learning in Japanese chains was concentrated in two functions: equipment and production management. This is closely related to the nature of HVN’s requirements, which are most often transferred to suppliers in the form of detailed drawings, product specifications, and target QCD levels. In response to the stringent QCD requirements imposed on by the lead firms, suppliers improved their QCD levels by improving equipment-related and/or production management capabilities. Given the lack of opportunities for product development and design, which was undertaken in-house by R&D headquarters in Thailand and Japan, none of the learning events in the Japanese chains involved the product planning function. For first-tier suppliers in the Japanese-led chains, capability building was a cumulative process. During the 1990s, when production scale was small and HVN faced little competitive pressure, learning in the Japanese chains was slow. In the 2000s, suppliers were required to achieve continuous improvements in capability levels as HVN introduced progressively higher QCD requirements. Most suppliers experienced particularly intensive learning during the 2000–2002 period, corresponding with the launching of the Wave Alpha, and during 2005–2008, when HVN rapidly expanded its production volume. During these periods HVN’s supplier requirements increased significantly. While HVN’s more stringent requirements acted as a broad driver of supplier learning, QCD improvements were achieved through a variety of approaches. Supplier D’s learning, for example, consistently focused on production management in its factory and management of suppliers of sub-components, rather than in-house equipment-related activities. Instead of investing in in-house manufacturing capacity, the firm focused on improving the quality and delivery performance of its suppliers, which provided the numerous sub-components required for its products. Supplier D sought to improve its suppliers’ quality by providing training on quality control schemes, inspecting

Table 3.3

Classification of learning events by functions of capabilities acquired # of events

Classification by functions Planning # of events

Maximum level reached

Equipment-related # of events

Maximum level reached

Production management # of events

Japanese chains (1st tier)

15

0

—

15

Adaptive

15

Japanese chains (2nd tier)

7

0

—

7

Adaptive

Vietnamese– Chinese chains

5

4

Adaptive

3

Other motorcycle chains

3

0

—

3

Other chains

20

12

Total

50

Adaptive

18

Maximum level reached

Marketing # of events

Maximum level reached

Adaptive

0

—

7

Assimilative

1

Assimilative

Operational

3

Operational

1

Assimilative

Assimilative

3

Assimilative

0

—

10

Assimilative

5

Adaptive

Adaptive

Source: Author’s fieldwork.

91

92

Value Chain Dynamics and Local Capability Building in Vietnam

sub-components, and regularly monitoring the suppliers’ factories. On the other hand, supplier A made substantial progress in in-house equipment-related improvements. Starting by supplying a few simple components (e.g., chain cases) in the late 1990s, supplier A gradually learned to design production processes, manufacture dies and molds, and process increasingly complex and high-precision components by investing in computer numerical controlled (CNC) machine tools. By placing process design engineers in each factory, the company also consistently improved production processes, achieving incremental cost reduction and quality improvements. Serious attempts at improving production management, however, only started around 2006 and were still ongoing as of the time of the interview. The company’s manager admitted that the company was still weak in this area.23 Levels of attainment also differed across suppliers. While some suppliers only partially mastered the assimilative level of equipment-related or production management capabilities, a few managed to go beyond the assimilative level by making their own original improvements to widely adopted production management practices or imported machinery, which is equivalent to the adaptive level. For instance, supplier E obtained ISO9001 certification without assistance from external consultants, as is often the case with local suppliers. The general director organized a company-wide initiative by which the top managers and the managers of individual departments constructed a comprehensive quality management system. By mobilizing the managers and staff of various departments, and engaging in repeated discussions, the initiative improved the coordination between different departments and raised the quality consciousness of the managers and workers. Second-tier suppliers’ capability building also focused on productionrelated activities, but the levels attained diverged even more widely than for the first-tier suppliers. Most second-tier suppliers were assigned simple processing activities using dies and molds provided by the firsttier suppliers, which limited their progress in improving equipmentrelated capability. For instance, suppliers J, L, M, and N were unable to completely achieve the assimilative level for equipment-related and/or production management capabilities. Nevertheless, some second-tier suppliers were engaged in the processing of core components, such as engine components, that were subject to stringent requirements. This is why supplier K was closely assisted not only by its direct customers, Japanese and Taiwanese first-tier suppliers, but also by Yamaha

Mai Fujita 93

Vietnam. Supplier K substantially decreased defect ratios by introducing a new quality control system a few years after it began to supply Japanese motorcycle manufacturers. In order to fulfill the stricter environmental standards imposed by its customers and to diversify its products, supplier K also mastered new plating technologies such as trivalent chromium plating and plating on plastics. 3.5.3.3. Learning in Vietnamese–Chinese chains Unlike HVN and Yamaha Vietnam, local assemblers did not provide design drawings or detailed specifications to suppliers. For most components, suppliers were required to replicate samples provided by the assemblers and re-create design drawings on their own. Since participating in Vietnamese–Chinese chains required suppliers to possess an operational capability level for all the value chain functions, learning in the Vietnamese–Chinese chains tended to be wider in functional scope than it was in the Japanese chains. Unlike suppliers in the Japanese chains, the majority of suppliers in the Vietnamese–Chinese chains did not experience cumulative learning toward progressively higher-level capabilities. Many of the 51 local assemblers that operated in the early 2000s eventually reduced or phased out their motorcycle assembly operations. Those that survived did not ask suppliers for further capability level improvements. Instead, they came to depend on the handful of large suppliers that could provide cost competitiveness and conduct component design work. Most of the respondent suppliers had stopped supplying components to local assemblers by 2008. Supplier M, however, was an exception. In 2002 this supplier established an R&D department and invested in design equipment, software, and measuring and testing devices. The engineers in the R&D department cooperated closely with the marketing department, which undertook in-depth market research via frequent visits to local assemblers and motorcycle dealers. Through these activities, rarely observed among other suppliers in Vietnamese–Chinese chains, supplier M was able to reach the adaptive level of planning and marketing capabilities. Supplier M continued to maintain relationships with local assemblers, because it took its own initiative to conduct market research and make modifications to the existing product designs on their own. 3.5.3.4. Learning in other chains With the exception of supplier G, virtually all of the 17 respondent suppliers were engaged in activities outside the motorcycle industry.

94

Value Chain Dynamics and Local Capability Building in Vietnam

Apart from the few first-tier suppliers that became increasingly dependent on HVN, suppliers’ dependence on specific customers was low. Even some of HVN’s first-tier suppliers actively attempted to diversify their customer base by leveraging the production capabilities they had acquired by serving HVN. Since the mid-1990s supplier E consistently sought to enhance its capabilities to produce higher precision products and design and manufacture molds with the aim of supplying plastic products to foreign buyers in various sectors. While HVN had been a major customer since 1997, supplier E also won orders for a different customer, for telecommunication equipment components, and participated in joint product design with this customer. This enabled the company to acquire planning capabilities that it could not through its relationship with HVN. Faced with unstable and declining orders from local assemblers after 2003, some suppliers in Vietnamese–Chinese chains shifted to other customers and markets, such as Taiwanese and Chinese motorcycle assemblers based in Vietnam and selling in the domestic market for replacement components. Supplier O reached the adaptive level for planning capability, but this was achieved in the realm of improving the designs of replacement components for sale in the domestic market.

3.6. Conclusions This chapter has shown that the dramatic growth of the Vietnamese motorcycle industry since the late 1990s has been accompanied significant, if uneven improvement in the local supply base. The analysis was conducted in two steps. The first focused on value chain dynamics in the Vietnamese motorcycle industry. The analysis showed that competition and adaptation within two value chains with contrasting GVC governance patterns structured the opportunities for local suppliers’ growth and capability formation. The second step looked at how the suppliers exploited those new opportunities. While the opportunities and limits provided by the two value chains were very different, so too were local firms’ capacity to exploit them. While most suppliers in captive value chains coordinated by Japanese motorcycle manufacturers improved equipment-related and production management capabilities, the areas of focus and the levels of attainment diverged across suppliers. Suppliers in market-based chains of local assemblers engaged in a wider range of activities but their capability levels, with the exception of one supplier, remained low. Some suppliers also leveraged participation

Mai Fujita 95

in other value chains to help them acquire new capabilities, including capabilities not nurtured in the Japanese or Vietnamese–Chinese chains. Two implications can be drawn from this study for GVC and TC theory. First, the study contributes to a better understanding of the factors shaping the patterns of inter-firm coordination. Previous studies have emphasized the importance of industry standards and changes in product design architecture (e.g., modularization) in driving shifts from highly integrated value chains to more loosely coordinated modular and market value chains (Sturgeon 2002; Gereffi et al. 2005). By contrast, the case of Vietnam’s motorcycle industry shows the possibility of such a shift occurring in the absence of institutionalized standards enforced by lead firms or other global actors (Nadvi and Waltring 2004; Ponte and Gibbon 2005), or of changes in product design architecture. The Japanese base motorcycle models became de facto standards in the Vietnamese market through an endogenous and bottom-up process that emerged out of the actions taken by numerous Chinese and Vietnamese firms under weak protection of intellectual property rights and low product quality requirements. The case raises the possibility that inter-firm coordination can be structured by the emergence of local/regional standards that arise under the market and institutional conditions specific to developing countries. Second, while the research presented here supports previous GVC research that has found that the mode of chain “governance” is a major determinant of local firms’ upgrading (Humphrey and Schmitz 2004), my research shows that firm-level capability-building trajectories significantly reflect the firm’s learning approaches as well. In short, supplier strategies matter. The diversity of capability-building achievement, even among suppliers in the same value chain, suggests that supplier upgrading is shaped not only by value chain governance but also by suppliers’ strategy and determination to succeed. While this chapter has emphasized differences in supplier learning approaches, capability achievement was also influenced by the suppliers’ initial stock of capabilities, their overall business and/or learning strategies, and their capacity to mobilize financial and human resources. By taking a firm-level approach to the analysis of capability building, this chapter has highlighted the strategic, intentional moves some suppliers made to leverage their participation in multiple chains. While captive chains were found to promote acquisition of production capabilities, as argued by Humphrey and Schmitz (2001), many of HVN’s suppliers were not as subordinate to the customer’s control as previous

96

Value Chain Dynamics and Local Capability Building in Vietnam

research has suggested. Some suppliers tried to take advantage of the production capabilities they had acquired in the HVN’s chains to diversify their products and customers. Analyzing local firms’ upgrading or capability building, therefore, requires a holistic view of the markets suppliers serve, or might potentially serve, including those outside the main value chain, industry, or country under investigation. This can be achieved by combining GVC and TC approaches.

Notes This research was partially supported by Grant-in-Aid for Scientific Research (C) on “Assembler-Supplier Relationship and the Growth of Local Component Suppliers in the Vietnamese Motorcycle Industry” (Project No. 20510243) of the Japan Society for Scientific Research. 1. Based on the motorcycle production and sales figures in the respective countries in 2006 (Honda Motor Co., Ltd. 2008). 2. The need to combine the GVC and TC approaches has been argued by Morrison, Pietrobelli, and Rabelotti (2008), though they did not discuss how this can be put into practice. 3. While the conventional version of GVC approach focused on the value chains that extend across international borders, the approach has been applied to chains serving the domestic markets of developing countries (Bazan and Navas-Alemán 2004). With regard to the GVC framework adopted in this chapter (Gereffi et al. 2005), Sturgeon (2008) argues that it operates equally well at regional, national, and local levels. 4. For the background behind this distinction between “hard” and “soft” aspects of production, see Suehiro (2008). 5. This figure includes production in Japan as well as exports of vehicles and knockdown components from Japan that are assembled overseas (Honda Motor Co., Ltd. 1986: 4). If production by firms outside Japan that received technology transfer from the four companies is included, the four Japanese firms exercised control over 70% of the world’s production of motorcycles in 1996 (Otahara 2000). 6. Honda has a regional R&D headquarters in Thailand, where basic designs developed in Japan are modified to meet the market requirements of each country in Southeast Asia. 7. To be precise, most of the chinese motorcycle manufacturers do not simply copy the standard models; they instead use these models to experiment varieties of minor improvements for revamping product performance, making functional or cosmetic changes, or reducing costs (Ohara 2001: 10–11). 8. Of course, the reality is more complex than this. The level of inter-firm coordination differed across firms and has changed over time. Ohara (2006) notes the shift toward closer assembler–supplier cooperation for product development in the 2000s. 9. For further details of the development of the industry, see Fujita (2006, 2008).

Mai Fujita 97 10. This chapter refers to the motorcycles assembled by local Vietnamese firms using Chinese component kits as “Chinese motorcycles,” as they are called in Vietnam. 11. As a typical example, in September 2002 the Vietnamese government suddenly announced that it would limit the import of motorcycle components to 1.5 million units for the year and allocated the quota to individual motorcycle manufacturers. As HVN and Yamaha Vietnam used up the quota initially allocated to them, they had to suspend production of motorcycles until additional quota was granted (“Deputy PM Dung feeds hungry motorbike makers more imports,” Viet Nam News, November 4, 2002). 12. The Vietnamese government required foreign motorcycle manufacturers to operate according to the feasibility studies they submitted to the authorities when they applied for investment licenses. This prevented the foreign motorcycle manufacturers from investing in the rapid expansion of production capacity, which had not been envisaged in late 1990s. 13. High tariffs on imported motorcycles remain as of 2008. 14. It is possible that some suppliers were counted more than once. 15. The author’s fieldwork in 2007. 16. Based on data from the Ministry of Industry. 17. Based on the author’s interview in 2005 with a state-owned enterprise which had been engaged in motorcycle assembly in the early 2000s. 18. HVN’s “Dream” and “Wave,” both of which are derivative models based on the old Super Cub, are the two most widely imitated base models in Vietnam. 19. Four of the five assemblers surveyed were included in the top ten local assemblers in terms of the value of turnover in 2006. 20. The author’s interview in November 2007. 21. Since there are no readily available databases of Vietnamese motorcycle component firms classified according to the types of value chains, the cases were selected from the database prepared by the author on the basis of published and unpublished data sources including the lists of firms complied by Japanese government-related and international organizations, newspapers and magazines, company directories, and interviews with Japanese motorcycle manufacturers. 22. While the previous interviews and surveys did not explicitly address the issues of capability building, information on the suppliers’ activities, products, customers, machinery and equipment, and technology included in their results served as a basis to develop questions for the interviews in 2008 and to conduct consistency checks. 23. Interview with the deputy director of the Administrative Department of Supplier A (September 2008).

References Abernathy, William J. and James M. Utterback. 1978. “Patterns of Industrial Innovation.” Technology Review 80(7) June/July: 40–47. Ariffin, Norlela. 2000. The Internationalisation of Innovative Capabilities: The Malaysian Electronics Industry. Ph.D. Thesis. Brighton: SPRU, University of Sussex.

98

Value Chain Dynamics and Local Capability Building in Vietnam

Bazan, Luiza and Lizbeth Navas-Alemán. 2004. “The Underground Revolution in the Sinos Valley: A Comparison of Upgrading in Global and National Value Chains.” In Local Enterprises in the Global Economy: Issues of Governance and Upgrading, ed. H. Schmitz. Cheltenham and Northampton: Edward Elgar: 110–139. Bell, Martin and Keith Pavitt. 1997. “Technological Accumulation and Industrial Growth: Contrasts Between Developed and Developing Countries.” In Technology, Globalisation and Economic Performance, ed. D. Archibugi and J. Mitchie. Cambridge: Cambridge University Press: 83–137. Figueiredo, Paulo N. 2008. “Government Policies and Sources of Latecomer Firms’ Capability Building: A Learning Story from Brazil.” World Development 36(1): 59–88. Fujita, Mai. 2006. “Betonamu no Nirinsha Sangyo: Shinko Shijo ni Okeru Jiba Kigyo no Sannyu to Sangyo Hatten” in Japanese (“Vietnam’s Motorcycle Industry: The Entry of Local Enterprises into a Newly Emerging Market and Industrial Development”). In Ajia no Nirinsha Sangyo: Jiba Kigyo no Bokko to Sangyo Hatten Dainamizumu, in Japanese (Asia’s Motorcycle Industry: The Rise of Local Companies and the Dynamism of Industrial Development) ed. Y. Sato and M. Ohara. Chiba: Institute of Developing Economies: 323–365. Fujita, Mai. 2008. “Value Chain Dynamics and Growth of Local Firms: The Case of Motorcycle Industry in Vietnam.” Discussion Paper No. 161. Chiba: Institute of Developing Economies. Gereffi, Gary, John Humphrey, and Timothy Sturgeon. 2005. “The Governance of Global Value Chains.” Review of International Political Economy 12(1): 78–104. Ha Huy Thanh, Bui Tat Thang, Do Huu Hao, Do Thu Trang, Tran Thu Hien, and Phan Mai Huong. 2003. “Study on Industrial Policies in Vietnam.” In Chiiki Keizai Togo to Betonamu, in Japanese (Regional Economic Integration and Vietnam) ed. A. Ishida. Chiba: Institute of Developing Economies: 305–405. Honda Motor Co., Ltd. Various years. Sekai Nirinsha Gaikyo (World Motorcycle Facts & Figures). Humphrey, John and Hubert Schmitz. 2001. “Governance in Global Value Chains.” IDS Bulletin 32(3): 19–29. Humphrey, John and Hubert Schmitz. 2004. “Chain Governance and Upgrading: Taking Stock.” In Local Enterprises in the Global Economy: Issues of Governance and Upgrading, ed. H. Schmitz. Cheltenham and Northampton: Edward Elgar: 349–381. Institute for Industry Policy and Strategy. 2007. “Masterplan for the Development of Vietnam’s Motorcycle Industry in the Period of 2006–2010, With a Vision to 2020.” Hanoi: Ministry of Industry. Kim, Linsu. 1997. Imitation to Innovation: The Dynamics of Korea’s Technological Learning. Boston, MA: Harvard Business School Press. Kim, Linsu. 2004. “The Multifaceted Evolution of Korean Technological Capabilities and its Implications for Contemporary Policy.” Oxford Development Studies 32(3): 341–363. Lall, Sanjaya. 1992. “Technological Capabilities and Industrialization.” World Development 20(2): 165–186. Morrison, Andrea, Carlo Pietrobelli, and Roberta Rabellotti. 2008. “Global Value Chains and Technological Capabilities: A Framework to Study Learning and Innovation in Developing Countries.” Oxford Development Studies 36(1): 39–58.

Mai Fujita 99 Nadvi, Khalid and Frank Waltring. 2004. “Making Sense of Global Standards.” In Local Enterprises in the Global Economy: Issues of Governance and Upgrading, ed. H. Schmitz. Cheltenham and Northampton: Edward Elgar: 53–94. Nguyen Duc Tiep. 2006. “Chinese Motorcycle Penetration into Vietnam and the Existing Motorcycle Makers: A Study of Honda Company.” Economics Bulletin 13(4): 1–9. Ohara, Moriki. 2001. “Chugoku Otobai Sangyo no Sapuraiya Sistemu; Risuku Kanri to Noryoku Kojo Sokushin Mekanizumu kara mita Nicchu Hikaku” in Japanese (“The Supplier System of the Chinese Motorcycle Industry: A Comparative Study with the Japanese System in View of the Mechanisms of Risk Management and Capability Upgrading”). Ajia Keizai XLII-4: 2–38. Ohara, Moriki. 2006. Interfirm Relations under Late Industrialization in China: The Supplier System in the Motorcycle Industry. Chiba: Institute of Developing Economies. Ohno, Kenichi. 2005. “Renovating Industrial Policy.” In Improving Industrial Policy Formulation, ed. K. Ohno and Nguyen Van Thuong. Hanoi: The Publishing House of Political Theory: 29–59. Otahara, Jun. 2000. “Nihon Nirin Sangyo ni Okeru Kozo Henka to Kyoso: 1945–1965” in Japanese (“Structural Changes and Competition in Japanese Motorcycle Industry: 1945–1965”). Keiei Shigaku 34(4): 1–28. Ponte, Stefano and Peter Gibbon. 2005. “Quality Standards, Conventions, and the Governance of Global Value Chains.” Economy and Society 34(1): 1–31. Sato, Yuri and Mai Fujita. 2009. “Capability Matrix: A Framework for Analyzing Capabilities in Value Chains,” Discussion Paper No. 219, Chiba: Institute of Developing Economies. Sturgeon, Timothy. 2002. “Modular Production Networks: A New American Model of Industrial Organization.” Industrial and Corporate Change 11(3): 451–496. Sturgeon, Timothy. 2008. “From Commodity Chains to Value Chains: Interdisciplinary Theory Building in an Age of Globalization.” In Frontiers of Commodity Chain Research, ed. J. Bair. Stanford, California: Stanford University Press: 110–135. Suehiro, Akira. 2008. Catch-Up Industrialization: The Trajectory and Prospects of East Asian Economies. Singapore: NUS Press. The Motorbike Joint Working Group. 2007. For Sound Development of the Motorbike Industry in Vietnam. Hanoi: The Publishing House of Social Labour.

4 Local Firms’ Capability Development in Captive Value Chains: Evidence from the Indonesian Motorcycle Industry Yuri Sato

4.1. Introduction This volume explores global value chain (GVC) dynamics from the perspective of local firms in developing countries. The focus is on supplier strategies, capability acquisition, and interactions with lead firms from developed countries. This chapter explores these dynamics through the case of the motorcycle industry in Indonesia. Rapid capability acquisition by local firms can be clearly observed in GVCs where linkages with lead firms contain dense interactions, exchange of complex specifications, and have relatively low power asymmetry, such as relational and modular value chain linkages prevalent in the notebook computer GVCs discussed in Chapter 1. But modular and relational linkages are just two of the five linkage, or GVC governance types, developed by Gereffi et al. (2005). The five types of value chain governance—market, modular, relational, captive, and hierarchy—range from a low to a high degree of power asymmetry between lead firms and suppliers and come with increasing requirements for explicit coordination. This chapter discusses local firms’ strategies, capability acquisition, and relationships to lead firms in captive value chains. Captive value chains are common in GVCs where suppliers are highly dependent on lead firms, supplier capabilities are relatively low, and lead firm intervention is high. Because suppliers in captive value chains are faced with high switching costs, they are commonly locked into their relationship with their main customer and frequently confined to a narrow range of functions (Gereffi et al. 2005: 84–87). 100

Yuri Sato 101

While captive value chains facilitate rapid supplier upgrading in the realm of production, they can block suppliers from further upgrading beyond production and leave suppliers dependent on powerful lead firms (Humphrey and Schmitz 2004). This chapter examines the details of local suppliers capability development in captive value chains. It analyzes how strategies of lead firms and local suppliers combine to stimulate, limit, and structure capability development among suppliers. The case of the Indonesian motorcycle industry provides a clear example of captive GVCs, where lead firms exert disproportionate power and engage in explicit coordination of the chain. However, the chapter demonstrates that even in captive value chains, local firms can acquire “focal” capabilities and, in some cases, execute strategies that drive the course of their own capability development. To gauge capabilities of local firms, the chapter uses a Capability Matrix, a framework that combines GVC and technological capability (TC) approaches. The chapter is organized as follows. Section 4.2 reviews the GVC and TC literatures and explains how the Capability Matrix draws from these two approaches. Section 4.3 outlines the main characteristics of the motorcycle industry in Indonesia. Section 4.4 examines the sourcing strategies of Japanese lead firms in Indonesia. Section 4.5 analyzes the capability development and strategies of local suppliers. The last section summarizes the research findings.

4.2. Analytical framework Most researchers using the GVC approach have focused on how internationally dispersed value-adding activities are coordinated by global buyers and lead firms from advanced economies. Lead firms set and enforce parameters on product design, specifications, quality, quantity, and production scheduling, under which other firms in the chain operate as suppliers (Humphrey and Schmitz 2001, 2004). The degree of explicit coordination or “governance” in value chains, largely determined by lead firm strategies and practices, is of central importance in GVC studies. As mentioned earlier, Gereffi et al. (2005) have formulated five modes of chain governance—market, modular, relational, captive, and hierarchy—based on the characteristics of inter-firm linkages and supplier capabilities, in which power asymmetry between lead firms and local suppliers ascends from low to high. Humphrey and Schmitz (Humphrey and Schmitz 2000, 2001; Schmitz 2004) have identified four types of “upgrading” of local suppliers in value chains—process,

102 Capability Development in Captive Value Chains in Indonesia

product, functional, and inter-sectoral—and have examined how chain governance affects the upgrading prospects for local suppliers along these four types.1 The largely accepted view of how the captive GVC governance affects the upgrading of local suppliers is that it provides local suppliers with a fast track for process and product upgrading, but tends to block their functional upgrading.2 Lead firms assist local suppliers in upgrading production activities in order to diminish the risk of supplier failure, but do not assist in upgrading non-production functions (e.g., product development, design, branding, and marketing) because the acquisition of capabilities by suppliers in these functions will clash with the core competence of the lead firms. Thus, functional upgrading is expected to occur only when value chain modularity through codification and standardization is possible and/or suppliers somehow acquire enough competence on their own to reduce the risk of failure. In these instances power asymmetry can diminish, and the chain governance can evolve toward relational, modular or market types (Humphrey and Schmitz 2004). Since the mechanisms for upgrading without lead firm assistance are not specified, this research implies that functional upgrading hardly ever occurs in captive value chains. This perception of captive value chains in the GVC literature is the starting point for the study presented in this chapter. To test the assumption that captive value chains limit functional upgrading by local suppliers in developing countries, this chapter examines the endogenous dynamics in which local suppliers learn, innovate, and develop capabilities. I argue that supplier strategies and practices need to be examined more closely and linked to a framework that can objectively assess capabilities. A body of literature that has dealt centrally with the dynamics of capability formation by local firms in developing countries is the TC approach.3 As an analytical tool, Lall (1992) first developed the TC matrix that aligns generic functions in columns and the degree of complexity or difficulty in rows. Recent TC studies (e.g., Figueiredo 2002, 2008; Ariffin and Figueiredo 2004) expanded Lall’s matrix, used it to measure the rate of TC development, identified the determinants of TC development, and revealed divergent paths to capability development. However, the TC literature does not consider the importance of interrelationships between local firms and lead firms in GVCs. Thus, as Morrison et al. (2008) have proposed, it makes sense to integrate the local firm perspective on TC development with the GVC perspective.

Yuri Sato 103

In an attempt to achieve this kind of theoretical integration, Sato and Fujita (2009) created a Capability Matrix specifically designed to characterize the firm-level capabilities of developing country suppliers operating in value chains. The Capability Matrix is a simple, practical, and industry-neutral conceptual tool (Table 4.1). It assesses depth of capabilities,4 arranged in rows, according to the value chain functions, pre-production (planning), production, and post-production (marketing), arranged in columns. Drawing on the TC literature, capabilities are defined as resources needed to generate and manage technological change, such as skills, knowledge, experience, organizational systems, and linkages (Bell and Pavitt 1995; Figueiredo 2003). The main benefits of the Capability Matrix are as follows. First, unlike the TC approach, the Capability Matrix explicitly incorporates the value chain perspective. The columns show, in a simplified form, the spectrum of functions that might be allocated to local suppliers in value chains. Second, the Capability Matrix assesses the depth of capabilities for each value chain function and maps capabilities in a two-dimensional surface that consists of width and depth. The assumption that underlies this matrix differs from the thinking that provides the basis for the GVC upgrading literature, which conceptualizes functional upgrading as the process of “climbing up the ladder” from lower value-adding functions, such as production, to higher ones, such as pre-production (planning) and post-production (marketing). This perspective regards functional upgrading as a superior dimension to product and process upgrading within the production function, but does not discuss the degree of upgrading within non-production functions. By contrast, the Capability Matrix assigns capability levels across any functional area of the value chain. It does not regard non-production capabilities as superior to production-related capabilities.5 This same conceptual approach, although used here in the simplified value chain shown in Table 4.1, can be applied to more detailed value chains as well. Third, the Capability Matrix can visually portray an individual firm’s capabilities by coloring cells in the matrix where the firm in question fulfills relevant criteria. Table 4.2 shows an example of the Capability Matrix in which criteria are designed to apply to the modern assembly industry. This framework is shared by the Vietnamese motorcycle industry case study by Fujita in Chapter 3. This applied Capability Matrix adopts a twofold distinction within the production function by dividing it into the “hard” and “soft” aspects

104 Table 4.1 A prototype of the Capability Matrix Width of functions

Depth of capabilities Operational

Operate the existing technology

Assimilative

Master the existing technology and maintain the operation over time

Adaptive

Make original improvements to the existing technology prevailing among local producers in the country concerned

Innovative

Create something new with significant elements of originality and novelty compared to the existing technology prevailing in the world

Planning Conduct market research; develop product concept; develop and design a new product according to the needs of the market

Source: Arranged from Sato and Fujita (2009).

Production

Marketing

Conduct production activities so as to achieve targeted performance; operate, maintain, design, and manufacture equipment

Market products so as to strengthen the relationship with customers; develop own brand; explore markets

105 Table 4.2 Capability Matrix applied to the modern assembly industry Functional width

Planning

Depth (Level)

Production Equipment-related

Marketing

Production management

Operational

Replication of an existing/given product in the domestic market by recreating the design drawings

Basic operation of machinery and equipment, dies, molds, jigs, and tools to process components to the level required in the domestic market

Routine production management required in the domestic market

Routine marketing methods/ activities

Assimilative

Replication of existing internationalstandard products by recreating the design drawings

Processing components and manufacturing dies, molds, jigs, and tools to the level required by foreign customers; maintenance and repair of machinery and equipment, dies, molds, jigs, and tools

Maintaining stable production management fulfilling the levels required by foreign customers

Adopting and stably managing improved marketing methods/ activities

Adaptive

Making original improvements to the existing products prevailing among the local producers

Making original improvements to the existing machinery and equipment prevailing among the local producers

Making original improvements to production management practices prevailing among the local producers

Making original improvements to marketing methods prevailing among the local producers; developing brands recognized in the domestic market

Innovative

Planning and designing new products with significant elements of originality and novelty compared to the existing products prevailing in the world

Designing and developing new machinery and equipment with significant elements of originality and novelty compared to the existing machinery and equipment prevailing in the world

Establishing a production management system so as to achieve the world’s topmost level in production management

New marketing methods with significant elements of originality and novelty to explore new markets abroad; establishing internationally recognized brands

Source: Arranged from Sato and Fujita (2009).

106 Capability Development in Captive Value Chains in Indonesia

of production, namely equipment-related and production management capabilities. These concepts are drawn from the literature on industrial and organizational development (Hayashi 1986; Cole 1989; Appelbaum and Batt 1994; Suehiro 2000; Fujimoto 2001). In the context of developing countries, where local firms most often face shortages of capital for investment in modern technology, it makes sense to distinguish the “soft” elements of production that require smaller amounts of physical investment from the “hard” elements of production that need larger investment. Improving production management can be an important channel for boosting industrial performance without the need for substantial investments in fixed capital. This chapter uses the Capability Matrix shown in Table 4.2 to assess the capabilities of local suppliers operating in captive value chains led by Japanese lead firms in the Indonesian motorcycle industry. Based on the author’s field studies, the chapter examines the proposition that captive value chains can enhance supplier capabilities in production while blocking capability formation in non-production functions. The study also identifies local suppliers’ strategies and how they affect their capability development.

4.3. Industrial setting 4.3.1. The motorcycle industry as a venue for local firm participation The motorcycle industry has several distinct features that affect the growth prospects for local firms (Ohara and Sato 2008). First, the industry has lower entry barriers in terms of capital and technological requirements than the automobile and electronics industries because motorcycles are relatively simple. Second, in developing countries, the demand for motorcycles generally expands ahead of demand for automobiles because prices are lower. Third, motorcycle technology is quite mature, which is not the case in the automobile and electronics industries. Since the launch of the “Super Cub” (C100) by Honda Motor in 1958, the core technology of the C100, which has become a de facto standard model in many developing countries, has not changed significantly. Fourth, the world’s motorcycle industry is overwhelmingly dominated by just a handful of large Japanese multinationals. This is in contrast to the automobile industry where the leading manufacturers in Japan, the United States, and Europe compete fiercely in global markets. Otahara (2006) found that Japanese motorcycle manufacturers

Yuri Sato 107

have a firmer technological grip on Japanese suppliers than is the case in the automobile industry. As a result there is even less room for suppliers to co-develop products with Japanese lead firms. This suggests that, in the motorcycle industry, the degree of power asymmetry between lead firms and suppliers is extreme. As such it is an ideal case for examining the dynamics of supplier upgrading in captive value chains. Once local governments institute policies to stimulate local production, the main attributes of the motorcycle industry just mentioned— low capital and technological requirements, early demand growth due to affordability, mature technology, and the existence of a de facto standard model—promote rapid entry of developing country firms in the realms of final assembly and component manufacturing. However, eventually local firms have to confront or become captured by the overwhelming power of a few Japanese manufacturers.

4.3.2. Dynamics of the Indonesian motorcycle industry Indonesia is the world’s third largest producer of motorcycles after China and India. As Figure 4.1 shows, the annual production volume began to expand in the early 1990s. Then, following a sharp drop after the Asian financial crisis, the industry expanded even more rapidly, reaching an output of over 5 million units per year by 2005. The average annual growth rate rose from 26% before the crisis (1990–1997 annual average) to 43% after the crisis (2000–2005 annual average). The rapid growth of production corresponded to the expansion of the domestic market, as exports and imports of finished vehicles were limited. The accelerated market expansion during the 2000–2005 period was partly attributable to credit relaxation such as a decline in interest rates and the down payments required for financed purchases. However, a more fundamental factor was the decline in the price of motorcycles, triggered by an influx of Chinese-branded low-priced motorcycles in 2000. Even though the boom turned out to be short-lived, its impact created a new demand for low-priced motorcycles. Targeting this new demand, the Japanese lead firms began to launch low-priced models one after another beginning in 2003. As a result, the market share for lowpriced models jumped from 23% in 2003 to 43% in 2004 in the context of a total expansion of the market.6 One consequence of the growing market share for inexpensive motorcycles was a decline in profits for component producers. Figure 4.2

108

6

5

Million unit

4

3

2 1

Figure 4.1

2005

2003

2001

1999

1997

1995

1993

1991

1989

1987

1985

1983

1981

1979

1977

1975

1973

1971

0

Production of motorcycles in Indonesia

350

40

300

35

250

30 25

200

20 150

15

100

10

50

5

0

1997

2000

2003

2004

2005

Rp. million, %

Rp. million

Source: Indonesian association of motorcycle Industry.

0

Sales per worker (left) Profits per worker (right, Rp.million) Profitability over sales (right, %) Figure 4.2 Labor productivity and profitability of motorcycle component producers Source: Calculated from the author’s study.

Yuri Sato

109

shows that labor productivity (sales per worker) at component producers increased in line with increasing volumes 1997–2005, but that profitability (profits per worker as well as profitability over sales) began to decline after 2003. Profitability declined because prices dropped as sales of low-priced models increased. To sum up, the Indonesian motorcycle industry experienced unprecedented growth during 2000–2005, with the expansion passing through two different phases: growth in terms of production and profitability (2000–2002), followed by a phase of even faster growth accompanied by declining profitability (2003–2005).

4.3.3. Dominance of Japanese-brand motorcycles While in Taiwan, China, and India, local brands hold a significant share of domestic markets, in Indonesia, as in Thailand, Japanese-branded motorcycles hold a dominant market share. In Indonesia, Japanesebrand motorcycles—Honda, Yamaha, Suzuki, and Kawasaki—account for more than 90% of the market (Figure 4.3). Other foreign brands, such as Taiwan’s Kymco, Italy’s Piaggio, India’s Bajaj and TVS, and

100% 90%

Others

80%

Piagio

70% 60%

Kymco

50%

Kawasaki

40% Suzuki

30% 20%

Yamaha

10%

Honda

0%

1995

2000

2001

2002

2003

2004

Figure 4.3 Composition of Indonesian motorcycle market by brand Note: Others includes several Chinese brands (see footnote 7) and one local brand, Kanzen. Source: Same as Figure 4.1.

110 Capability Development in Captive Value Chains in Indonesia

the only Indonesian brand Kanzen, hold only small percentages of the market. An exception to this pattern is the sudden appearance of Chinese-brand motorcycles. In Indonesia, Chinese brands reached 18% market share in 2000, but, in contrast to the Vietnam case presented in Chapter 3, sales declined to below 10% after 2002, mainly due to consumers’ dissatisfaction with low quality and lack of after-sales services. Most of the Chinese-brand motorcycle assemblers relied entirely on imported parts, and only had final assembly lines in Indonesia.7 These firms faded away quickly after 2002, well before establishing a local supply base. The dominant market share held by Japanese-brand motorcycles has been supported by strong local production and distribution networks. Operating since the 1970s, Japanese motorcycle manufacturers have established deeply rooted value chains in Indonesia, consisting of several layers of component suppliers, as well as dealers, distributors, and after-sales service outlets. The relatively long history of value chain development by the Japanese manufacturers can explain why newcomers such as Chinese assemblers and the Indonesian assembler Kanzen faced difficulties expanding in Indonesia, unlike in the newly emerging Vietnamese motorcycle industry. In Vietnam, where Japanese manufacturers set up local production only in the late 1990s, Vietnamese assemblers using Chinese components have grown side by side with Japanese manufacturers. Table 4.3 shows ownership of first-tier suppliers for the three largest Japanese-brand motorcycle manufacturers in Indonesia. Japaneseaffiliated suppliers account for 47% of the total number and 71% for total procurement value, while local suppliers account for 44% and 27% respectively. The affiliated suppliers, in which Japanese-branded firms have invested directly, produce relatively high value-added items such as cast and machined engine components, gears, drive shafts, braking systems, and electrical components; while local suppliers engage in the production of relatively low value-added items such as pressed and welded metal parts. Nevertheless, the fact that Honda—the dominant firm with a market share of around 50%—has actively utilized local suppliers has had a significant impact on the development of the local supply base. This study focuses on local suppliers operating in value chains led by the three major Japanese-brand motorcycle manufacturers in Indonesia: Honda, Yamaha, and Suzuki.8

Table 4.3

Japanese-brand first-tier suppliers by ownership in the Indonesian motorcycle industry (2004–2005)

Ownership classification

Japanese-affiliated of which, directly invested2 Non-Japanese foreign Local Other, unknown3 Total

Number of firms1 (%)

Value of procurements1 (%)

Honda No. of firms (unit)

(%)

Yamaha Value No. of firms (%) (unit) (%)

Suzuki Value No. of firms (%) (unit) (%)

Value (%)

47 — 3 44 5

71 — 2 27 1

43 11 3 77 9

33 8 2 58 7

67 55 3 29 1

70 3 10 30 10

58 3 8 25 8

NA 37 NA NA NA

60 2

50 2

75 NA

60

50

25

—

—

—

100

100

132

100

100

120

100

100

120

100

100

Note: 1 The simple average of the three Japanese-affiliated manufacturers. 2 Component producers with direct investment by the principals of each brand firm in Japan or by their subsidiaries in Indonesia, Thailand and Singapore. For Yamaha, only ones directly invested by the subsidiary in Indonesia. 3 Other refers to companies other than those in component production, such as trading. Source: Interviews at the Japanese-affiliated motorcycle manufacturers, and various directories.

111

112 Capability Development in Captive Value Chains in Indonesia

4.4. Lead firms’ strategy During the period of rapid market expansion after 2000, Japanese motorcycle manufacturers changed their procurement strategy from imports to local sourcing. As shown in Table 4.4, the share of imports in the total value of components procured by the three major lead firms fell dramatically from 57% prior to 2001 to just 9% in 2003–2005. As local procurement increased, external sourcing soared to 78%. This dramatic shift to local sourcing in Indonesia was driven by the strategies of Japanese lead firms who were trying to adjust to exchange rate fluctuations and decline in final product prices. After the 1997 Asian financial crisis, the Indonesian rupiah depreciated to around one-fourth of its previous value. As a result the cost of imported components effectively quadrupled. Then the Chinese-brand motorcycle boom came. Even after the boom passed, Japanese lead firms were confronted with an increased demand for low-priced models. Since they could not raise product prices, they sought to reduce costs by switching from imported to locally produced components. Japanese motorcycle manufacturers in Indonesia had preferred in-house manufacturing to local external sourcing, as it was easier to control component quality, costs, and delivery (QCD). In order to further reduce costs, however, they were compelled to shift to external sourcing from domestic suppliers. When they switched to external sourcing, either from Japaneseaffiliated or local suppliers, Japanese lead firms redoubled their emphasis on QCD. They insisted on the maintenance of consistent performance in quality and delivery despite increasing the size of orders. They also started to place greater emphasis on cost reduction, particularly after they had launched low-priced models. Table 4.4 Shift in shares of component procurement values from importing to local sourcing by Japanese motorcycle manufacturers in Indonesia (%) Source of component procurement Imports from Japan from Asia excluding Japan Indonesian domestic In-house manufacturing External sourcing Total

Prior to 2001

2003–2005

57 51 6 44 10 34

9 4 4 91 13 78

100

100

Note: The simple average of the three Japanese manufacturers. Source: Interviews at the Japanese motorcycle manufacturers.

Yuri Sato 113

4.5. Analysis of local suppliers’ capabilities This section investigates capability formation among local suppliers in captive value chains led by Japanese lead firms in the Indonesian motorcycle industry. The first sub-section introduces, in general terms, the target firms that were used in this study. The second sub-section uses the Capability Matrix to map the distribution of capabilities in the target firms and to characterize their capability formation. The third and fourth sub-sections focus on the capability categories pertinent to the production function, namely the equipment-related and production management capabilities, because they are the two capabilities that were most commonly pursued by suppliers in Japanese-led value chains. The fifth sub-section discusses the strategies local firms put in place to enhance their capabilities. 4.5.1. The target firms in this study In 2005, there were approximately 220 first-tier motorcycle component suppliers in Indonesia, of which about 120 were locally owned. Of the local suppliers, about 80 firms participated in value chains led by Japanese lead firms.9 This study selected 25 local firms that directly or indirectly served the three Japanese motorcycle manufacturers: Honda, Yamaha, and Suzuki. To create greater variation, non-first-tier suppliers were included. The firms selected for in-depth study consist of 15 first-tier suppliers out of the above-stated 80 firms,10 5 second-tier suppliers, 2 producers of genuine spare parts, and 3 producers of dies and molds. All 25 suppliers selected were wholly locally owned.11 All but three of the firms were mainly engaged in the motorcycle industry, which accounted for 50–100% of their total sales (see Appendix). They either exclusively or mainly produced for the manufacturers of Japanese-brand motorcycles. During 2000–2005, when the motorcycle market expanded rapidly and the lead firms’ strategies shifted from imports to local sourcing, local production of components grew at annual rates as high as 60%. With a limited number of new entries, existing local suppliers expanded the scale of their production dramatically. Major suppliers raised their output to an equivalent of about 1 million motorcycle units annually, a volume without precedent in any sector of the Indonesian machinery industry. Reflecting this dramatic growth, 76% of the firms targeted in this study can be categorized as “large firms” with more than 100 workers.12 Approximately 70% of the target firms produced components for only one or two brands. On average, their largest customer accounted for

114 Capability Development in Captive Value Chains in Indonesia

more than 60% of their business, a share which rose over time. Nevertheless, the number of customers the target firms traded with also increased. This reflects both increasing concentration in the supply base and overt supplier strategies for diversification. Some 65% of the target firms had long-term relationships of more than 10 years’ standing with their largest customer. Almost 90% of the target firms delivered products one to three times a day to their largest customer located within 50 km. In sum, the majority of the target firms were highly dependent on their largest customer, to whom they were making a high frequency of deliveries. At the same time, these firms were diversifying their secondary customer base. So, while the target firms were engaged in captive relationships with their main customer, they still had some autonomy.

4.5.2. Mapping firm capabilities As discussed earlier, this study applies a Capability Matrix framework to the motorcycle industry, using criteria for each cell as shown in Table 4.2. We assess whether a target firm fulfilled the corresponding functions and what level of capabilities it acquired in each function. The cells are shaded if the firm met the criteria defined for each cell. We then calculate the share of target firms that have shaded cells. Figure 4.4 shows the aggregate results of 25 firms in 2005. The darker the shading of the cell is, the higher the percentage of target firms possessing capabilities corresponding to the cell. The main findings from Figure 4.4 are as follows. First, looking at the overall results, the two columns in the production function have darker shading than those corresponding to the pre-production and postproduction functions. This result is, in a broad sense, in line with the argument advanced in the GVC literature that upgrading of suppliers in captive value chains tends to be confined to the production function. This study, however, attaches more importance to the “depth” of production capabilities represented vertically in the Capability Matrix, and regards these assets as evidence of significant learning in captive value chains. In addition, it must be noted that some target firms, though not many, acquired capabilities in non-production functions. Second, comparing the two production function categories, the production management column is darker than the other column, indicating that more target firms reached a deeper level in production management than in equipment-related capabilities. This suggests that the capabilities most commonly acquired by local suppliers in Japanese captive value chains are in production management, while

Yuri Sato

115

(%) Functional width Depth (Level)

Preproduction

Postproduction

Production

Planning

Equipmentrelated

Production management

Marketing

Operational

12

96

96

88

Assimilative

0

62

84

0

Adaptive

8

4

24

12

Innovative

0

0

0

0

Figure 4.4 Assessing the capabilities of Indonesian local firms in the Japanese captive value chains Source: Calculated from the author’s study.

attainment in equipment-related capability acquisition tends to vary. Those local suppliers that had been learning the basics of Japanese product management systems since the 1980s, such as the 5Ss, Kaizen, and QC circles,13 were able to quickly enhance their capabilities during the period of rapid market expansion in the 2000s. Otherwise, the defect rates would have increased along with production volume increases. The dark shading at the assimilative level of production management means that most of the target firms had acquired the production management capabilities needed to fulfill the requirements of the Japanese lead firms. Production management constitutes a core knowledge set which Japanese manufacturing firms and many of their suppliers accumulated over the course of their development. As Ohara and Sato (2008) point out, local suppliers’ production management capabilities, inculcated and passed on by Japanese lead firms in GVCs, have served as a vital source of industrial and corporate competitiveness throughout East Asia. The different supplier achievement levels in equipment-related and production management capabilities are a noteworthy finding. Why does capability acquisition differ between the hard and soft aspects

116 Capability Development in Captive Value Chains in Indonesia

of production? My field observations indicate that equipment-related capabilities have a dual nature. On the one hand, they are closely interrelated with production management capabilities, but, on the other, they leave more room for supplier discretion. This is because Japanese lead firms generally do not exert direct control over their suppliers’ investment in equipment, whereas they are highly concerned about and exert control over suppliers’ production management practices. Thus, equipment-related capabilities tend to reflect the extent that suppliers attach importance to equipment-related improvement. Suppliers may choose to purchase higher-performing machinery or may seek costsaving through incremental improvement to existing equipment. Some may not care much about equipment-related improvement. The third finding from Figure 4.4 is that in all four functional categories some of the target firms reached the adaptive level. This indicates that even in captive value chains, suppliers do not only follow orders from lead firms, but can also generate something original based on their own initiative. It also suggests that suppliers are pursuing their own independent strategies. The fourth finding is that capability deepening is not always a continuous process. While equipment-related and production management capabilities exhibit a trajectory of continuous across-the-board deepening, planning and marketing capabilities show a discontinuity. Some target firms had acquired capabilities at the adaptive level without going through the assimilative level. This result suggests that the exercise of originality is not necessarily a direct extension of the knowledge acquired from the Japanese lead firms, as will be shown later.

4.5.3. Equipment-related capabilities in captive value chains Regarding equipment-related capabilities, 96% of the target firms had reached the operational level, and 62%—a far lower figure in comparison to production management capabilities, but still more than 50%—had reached the assimilative level: they were able to maintain and repair machinery, equipment, dies, molds, jigs, and tools to the level set by the Japanese lead firms. However, except for Firm C, the supplier that designed improved assembly machines on its own, the target firms had not reached the adaptive level. The fact that only one out of 25 target suppliers, or 4%, had reached the adaptive level in equipmentrelated capabilities reveals a lower level of achievement in Indonesia in comparison to suppliers in Vietnam. Chapter 3 shows that 21% of the target suppliers attained adaptive-level equipment-related capabilities.

Yuri Sato 117

Of the 62% of the target firms that attained the assimilative level in equipment-related capabilities, my field research revealed two contrasting—but not strictly exclusive—approaches for ensuring quality during the expansion of production. The first approach was to reduce human error by making massive investments in high-performance machinery that replaced workers. A typical example is the introduction of welding robots, which some large-scale, well-financed suppliers purchased in significant numbers. Management believed that the robots could be operated by a smaller number of operators, less skilled than their current workforce, and could produce fewer defects. Firm A, one of the largest first-tier suppliers in Japanese value chains, reduced its defect rate from 1% in 2003 to 0.5% in 2005 by introducing more than 100 welding robots. The second approach was to improve proficiency in the use of dies, molds, jigs, and tools (DMJT) while utilizing existing machinery and training workers. Japanese lead firms have emphasized improving capabilities related to DMJT because they regard DMJT as a key determinant of product quality. Suppliers were encouraged to design and manufacture DMJT in-house and were sometimes explicitly taught more effective ways of using, maintaining, and repairing DMJT equipment, some of which was provided by lead firms. In fact, almost all of the target firms made jigs in-house, 95% made dies and molds in-house, and by 2005 45% were no longer being provided with DMJT by lead firms (Figure 4.5). However, even though the in-house ratios of DMJT manufacturing were high, we need to carefully assess whether target firms really were able to manage DMJT over time. One of the indicators for assessing this is the percentage of defects. Firm E experienced increased defect rates when running at full capacity due to insufficient maintenance of molds. In this case the firm failed to attain the assimilative level. Another possible indicator of assimilative-level equipment-related capabilities is the ratio of personnel in the workshop in charge of designing, manufacturing, maintaining, and repairing DMJT. Firms typically allocate their best workers to these tasks. Figure 4.5 shows diverging results: 44% of the target firms had low ratios of DMJT personnel (less than 4%), 28% had relatively high ratios (8–10%), while the mid-level ranges (4–6% and 6–8%) accounted for smaller shares. This indicates that, although most firms had set up workshops and fabricated DMJT in-house, commitment to these activities varied across firms depending on the importance attached to DMJT. Of the target firms, 20% were assessed as failing to reach the assimilative level, as their ratios

118 Capability Development in Captive Value Chains in Indonesia

(a) Die and mold making by local firms

All provided

(b) Human resources allocated in the workshop

~4%

0% Figure 4.5

Provided + made in-house

All made inhouse

4–6%

Made in-house + procured

6–8%

8–10%

10% 20% 30% 40% 50% 60% 70% 80% 90% 100%

Die and mold making and human resource allocation

Source: Calculated from the author’s study.

of workshop personnel were in the lowest range, even though they fabricated DMJT in-house.14 Equipment-related capabilities thus have two aspects. While DMJT management is what Japanese lead firms stress, the choice of machinery and personnel allocation in workshops are matters that lie beyond their control. That is why the equipment-related capabilities show a varied performance in the Capability Matrix. 4.5.4. Production management capabilities in captive value chains A high percentage of the target firms reached the operational and assimilative levels in production management, and about a quarter attained the adaptive level. This means that most of the target firms had acquired capabilities in production management high enough to fulfill the requirements of Japanese lead firms, and that some were even able to make original improvements of their own. The deepening of production management capabilities can be regarded as a central achievement for Indonesian suppliers in Japanese captive value chains. This type of capability formation can take place in captive chains because Japanese lead firms emphasize improvement in production management. Since the 1980s, suppliers have developed production management capabilities because they have been exposed to ceaseless demands from Japanese lead firms to enhance quality and delivery. Besides, unlike equipment-related capabilities, the acquisition of capabilities in production management does not require large financial investment, since knowledge, being intangible, can be absorbed regardless of the size and financial status of suppliers.

Yuri Sato 119

In addition to insisting on high standards in quality and delivery, Japanese lead firms have pressed suppliers to cut costs, especially after 2003. Table 4.5 shows the measures that the target firms have taken to cope with this cost-cutting pressure. The measures are listed in the order of their frequency of application. The table clearly shows that the measures taken most frequently fall into the category of production management. This indicates that knowledge about production management has penetrated into the Japanese supply base in Indonesia. In contrast, the measures that fall into the equipment-related category are fewer in number. Some improvement measures in the equipment-related category are closely related to production management. For instance, in order to raise efficiency, some target firms reduced the number of processes they performed, an effort that required redesigning of dies and molds. Other firms redesigned work standards and production systems in introducing high-performance machinery. Other target firms improved jigs, and concurrently developed systems for their maintenance, such as the daily calibration of process jigs and tri-monthly calibration of inspection jigs. The production management capabilities needed to take measures illustrated in Table 4.5 fall into the assimilative level in the Capability Matrix. Having attained this level, some firms deepened further to the adaptive level, achieving something original beyond the requirements of lead firms. One example is the case of Firm H, which utilized inspection jigs in-house. The firm had its own internal rules governing the use of the jigs. For example, workers were required to use an inspection jig after every ten pressing shots. In order to fully enforce its internal rules, the firm introduced a system where workers who neglected the rules and produced defective units had to pay compensation to the firm for the lot concerned. This type of compensation system, brought to bear on individual workers to enforce quality control, and which can be often observed in China, is rarely seen in Indonesia. Firm H’s owner, however, balanced this punitive system with the provision of labor welfare facilities, such as an in-house credit union and company accommodation, superior to those of the other target firms. Compensation systems of this kind can be regarded as a demonstration of a firm’s capabilities to create its own production management systems, or what may be called adaptive-level capabilities. Adaptive-level production management capabilities can also be seen in improved QC techniques developed by suppliers themselves. Firm K expanded intra-divisional quality and cost control (QCC) circles into inter-divisional QCC circles and held a company-wide contest for

120

Table 4.5 Measures for reducing costs Measures

The number of answers

Planning

Production Equipmentrelated

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Intensification of Kaizen Reduction in the number of working processes Introduction of high-performance machinery Reduction in the processing/exchanging time of molds Improvement of process/inspection jigs and systems of using jigs Reduction of inventory (Kanban) Saving on labor Investment in energy-saving/waste-reducing equipment Revision of layout Differentiation of products Revision of quality in excess of need Revision of material composition designs Downsizing of lots Utilization of subcontractors Procurement of lower-priced materials Conversion to lower-priced fuels (coals)

Note: Multiple answers were allowed. Source: Calculated from the author’s study.

6 5 4 3 3 3 2 2 1 1 1 1 1 1 1 1

       

Production management        

  

Yuri Sato 121

improvement suggestions. Firm O introduced a system of daily mutual evaluation on the 5Ss and safety, and a system of monthly reviews where divisions evaluate each other and review patterns collected in daily evaluations. Firm C developed a policy of “radical manufacturing” including rationalization of production processes and waste reduction. The firm also introduced a proper Kanban system by requesting that a subsidiary of Toyota Motor (the originator of the Kanban system) in Indonesia provide training, even though it had no trading relationship with the company. 4.5.5. Local suppliers’ strategies for survival Clearly, local suppliers operating in Japanese captive value chains can, to some degree, formulate and pursue their own strategies outside the control of lead firms. However, for the most part, the objective of these strategies is to maximize profits, not to enhance capabilities per se. Consequently, these strategies do not always lead to capability enhancement. This section discusses three strategies adopted by local suppliers and examines whether or not they have furthered capability development. In cases where they have done so, I explore whether those capabilities are utilized within or outside Japanese value chains. 4.5.5.1. Integration of production processes Faced with declining profitability, some target firms sought to create new sources of profits through integration of production processes. Firm E’s profitability in its die-casting business fell to 4–5%. In response the firm stopped outsourcing its machining work, brought the process in-house, and integrated it with die-casting, raising its profitability to 10%. The firm then raised its profit rate to 18% by moving the painting process and the sub-assembly of hubs and other parts in-house. Firm J, a forging and heat treatment supplier, integrated a machining process of the forged motorcycle components vertically. It also pursued horizontal diversification by offering machining services to producers of automobile spare parts. Within the conceptual framework of the Capability Matrix, integration of production processes can be regarded as an extension of the width of the production function, so that capabilities were widened within the production function. This is called functional upgrading in the GVC literature. However, it needs to be distinguished from efforts to deepen capability levels. In other words, process integration to raise profitability per se does not necessarily lead to a deepening of capabilities. In the above cases, the extension of the width of the production

122 Capability Development in Captive Value Chains in Indonesia

function occurred within the existing Japanese value chains and was aimed at raising the firm’s status in those chains, with the exception of Firm J’s move into machining automobile spare parts. 4.5.5.2. Brand-specific dedicated factories One of the strategies pursued by suppliers to cope with production expansion was to establish factories dedicated to specific customers, as observed in Firms D and M. Figure 4.6 illustrates the case of Firm D. If a firm seeks to secure orders for increased volume and additional items from its largest customer, while at the same time maintaining business with minor customers, then dedicating a factory to the main customer is one way forward. According to Firms D and M, customer-specific factories were established in order to make dealing with visits and audits from lead firm personnel easier, as well as to more effectively meet customer-specific requirements. The strategy can thus be regarded as an attempt to strengthen bilateral relationships with customers. Setting up dedicated factories also helped to avoid mixing-up similar items produced for different customers, and made possible more efficient management of a large number of items and frequent deliveries to each lead firm. By choosing to build dedicated factories, Firms D and M prioritized effective marketing and production management, albeit with some loss of efficiency from lower economies of scale and divisions of labor. Some

Automobiles Yamaha

Honda

Da

Kawasaki

D

Plating

Kymco

Kanzen

Db

Molds D Group

Figure 4.6

D Group as an example of brand-specific dedicated factories

Note: 1. Firms D, Da, Db are engaged in pressing, welding and machining of frames, handlebars, frontsteps, exhaust pipes, etc. 2. Kymco is a Taiwanese brand, and Kanzen is an Indonesian brand. Source: Interviews at Firm D.

Yuri Sato 123

firms have created dedicated factories for Japanese lead firms to conceal their business with non-Japanese customers. In such cases, the strategic intention is to strengthen their position within Japanese value chains while maintaining or expanding their position in other chains. In terms of the Capability Matrix this places firms in adaptive-level capability category because changes were implemented as a part of supplier strategy, not in response to requests from lead firms.

4.5.5.3. Improvement of product design Confronted with pressure to lower costs and maintain profits, some suppliers in the study began to make their own modifications to existing designs, technologies, and products. This behavior, though rare among the target firms, falls into the planning function at the adaptive level, and is an unexpected occurrence in captive value chains. The fact that it was rare supports GVC theory, but also shows that non-production-related upgrading is possible in captive value chains. Firm C, a producer of motorcycle drive chains, judged that the existing products were of “excess quality,” and revised the specifications downward, within an acceptable range, in order to realize “appropriate quality” with lower costs. For example, the firm replaced certain materials made in Japan with lower-cost materials with the same part number made in Taiwan and India. At the same time, the firm developed a differentiated set of higher-priced drive chains. Both these efforts came at the suppliers’ own initiative, outside the formal requirements provided by lead firms. Furthermore, Firm C decided to produce a new product, industrial chains, having experienced a decline in the profitability in its motorcycle drive chains from 10% in 2000 to 6% in 2005. Applying accumulated knowledge of making motorcycle chains based on drawings provided by a Japanese chain manufacturer under a technical agreement, the firm succeeded in developing a chain of its own design for industrial applications. Similarly, Firm F, a producer of rubber components, changed the composition of materials in order to reduce costs without affecting product performance, an initiative based on its own R&D. The firm also began production of a new component—extruded fuel hoses for engines—for the purpose of offering higher value-added products to Japanese lead firms. In order to do this, it explored a new technology—“extruded injection” instead of the “molded injection” that it was using at the time—and signed a technical assistance agreement with a Japanese rubber manufacturer.

124 Capability Development in Captive Value Chains in Indonesia

These cases suggest that, while unusual, it is possible for local suppliers in Japanese captive value chains to achieve adaptive capability levels in the planning function. What conditions, therefore, might promote the attainment of planning capabilities at the adaptive level? One condition may lie in the attributes of products. Chains and rubber components have generalizable attributes that, to a certain degree, are independent of lead firms’ brand-specific control. This leaves room for local firms to make changes in design, to explore relationships with specialized technological sources, and to seek new markets without alienating the dominant customer. This is also related to what Sturgeon (2002) has referred to as a generic “base” product or process, and one that can be turned toward multiple applications and markets. Such generic products and processes can form the basis of modular value chain linkages and can begin to ease suppliers in captive value chains away from reliance on their dominant customer. As stated earlier, Japanese motorcycle manufacturers maintain a firm technological grip on their suppliers, even in Japan. This creates little room for joint product development with suppliers. Thus, suppliers have limited opportunities for deepening planning capabilities in Japanese motorcycle value chains. One consequence of this, as shown above, is that the successes in planning take place through firms’ departure from the Japanese motorcycle value chains. However, this study also shows that adaptive-level planning capabilities can be achieved even within the captive value chains, provided that products have certain independence with regard to their technological attributes.

4.6. Conclusion This chapter has examined local suppliers’ capability development in captive value chains led by Japanese manufacturers in the Indonesian motorcycle industry, focusing on the strategies of both lead firms and suppliers and the way these strategies affect the development of supplier capabilities. The study confirmed, in a broad sense, the argument put forward in the GVC literature that suppliers in captive value chains are able to develop capabilities in production activities, but are hindered from diversifying into non-production functions. This study, however, delved deeply into this argument and offered specific evidence for what suppliers learned, and how they leveraged their experience and/or pursued their strategies to become more innovative, profitable, and autonomous

Yuri Sato 125

without alienating their dominant customer(s). The following are the main findings. First, the capabilities most commonly attained by suppliers in Japanese captive value chains are those related to production management. This is because Japanese lead firms have emphasized the acquisition of these capabilities over a long period of time in Indonesia; and also because production management is an intangible asset that does not require large-scale investment and, thus, can be absorbed by suppliers regardless of their size and financial resources. Most of the target firms have achieved the assimilative level by meeting the standards required by Japanese lead firms. The honing of production management was a central focus during the development of Japan’s manufacturing industry, and these capabilities continue to be a fundamental element of the industrial and corporate competitiveness of Japanese manufacturing firms. This focus has clearly been passed on to the Japanese supply base in East Asia. Thus, the growing depth of local suppliers’ production management capabilities has become a key feature of Japanese-led captive GVCs, critical to any understanding of East Asian industrial development. Second, equipment-related capabilities showed a lower degree of attainment compared with production management capabilities. While the Japanese lead firms monitor production management, they exert limited control over suppliers’ equipment. While equipment-related activities are closely linked to production management, investments in machinery and human resources are left to suppliers’ discretion. While greater discretion does not necessarily lead to lower attainment, in the case of Indonesian suppliers this has resulted in a varied and ultimately limited attainment in equipment-related capabilities. Third, counter to the broad assumption made in the GVC literature, this study has found evidence suggesting that even in captive value chains suppliers have been able to acquire capabilities in nonproduction functions. Moreover, a moderate number of suppliers were able to generate something original and worthwhile on their own initiative, as evidenced by the ratio of firms that had reached the adaptive level in all the functions in the Capability Matrix. This is why I argue that the strategies of the suppliers matter. The planning function showed the least capability formation by suppliers, but even so there were a few suppliers that acquired the adaptive level of capabilities in product design and development. The attainment of non-production capabilities and the demonstration of originality were in some cases directed toward customers outside Japanese-led captive value chains, but

126 Capability Development in Captive Value Chains in Indonesia

in other cases they occurred within them and were based on knowledge gained over time from Japanese lead firms. How, then, did the strategies of the lead firms affect the formation of supplier capabilities? The shift of lead firms’ procurement strategy from importing to local sourcing had a great impact on the growth of local suppliers. The ceaseless pressure of Japanese lead firms to maintain high standards of quality and delivery in the context of rapidly increased production, and to reduce component costs for low-priced models, with special focus on DMJT and production management, directly affected the deepening of suppliers’ production management capabilities. Moreover, their cost-cutting pressure had wider effects in that it forced suppliers to develop and implement a variety of survival strategies. Supplier strategies, broadly speaking, operated in two ways. One was the application of knowledge learned in the Japanese captive GVCs. In activities where Japanese lead firms exerted direct control, most suppliers acquired capabilities to assimilate existing technology, and if suppliers employed strategies to leverage what they had learned, they attained the capabilities to make original improvements. This strategy worked most typically to deepen production management capabilities to the adaptive level, and contributed to strengthening the firm’s position within Japanese value chains. The second set of strategies operated in areas outside of lead firms’ control, to seek out and make use of any room for maneuver left open. This was especially evident in cases where suppliers created something original in the realm of non-production functions, such as the strategies of creating dedicated factories, making changes in product design, seeking out new technology sources, and introducing new products for markets outside the motorcycle industry. While successes in developing non-motorcycle products typically meant departing from Japanese motorcycle GVCs, other successes in exhibiting adaptivelevel non-production capabilities took place inside the captive value chains. Lastly, this study used a Capability Matrix for empirical studies. Specifically, the Capability Matrix was used to map the achievements of supplier capabilities. This helped to show the capability achievements not only in production but also in non-production functions, and to reveal the difference in achievement between the hard and soft aspects of production capabilities. It has also been useful in identifying the areas in which the local firms exhibited originality, either based on the knowledge acquired in Japanese-led captive value chains or by way of their own independent survival strategies.

Appendix Table 4A.1

The 25 target firms in this study: basic information and Capability Matrix (2005)

No. Code Year No. of Ownership2 manufacturing workers began1 (persons)

Status in chain3 OEM 1-tier

2-tier

1

A

1988

1050

PC





2

B

1994

900

PC





3

C

1985

835

P∗



4

D

1983

800

PC



5

E

1990

693

P∗





Main products

After-sales service market

% of sales Notes in motorcycles

80

High-performance machinery



Frames, wheelrims, mufflers

75

High-performance machinery



Chains, control cables

60

Design machinery Differentiation New product development Market exploration

Frames, wheels

75

High-performance machinery Brand-specific dedicated factories

Enginecasting parts, hubs

73

Process integration Increasing defects

127

Frames, starters, wheels

Capability matrix

128

Table 4A.1

(Continued)

No. Code Year No. of Ownership2 manufacturing workers began1 (persons)

Status in chain3 OEM 1-tier

6

F

1988

618

P∗



7

G

1985

600

PP



8

H

1984

600

PC

9

I

1995

578

10

J

1992

567

2-tier

Main products

After-sales service market 

% of sales Notes in motorcycles

Engine rubber parts

60

Design material composition Low ratio of technical personnel5 New technology introduction



Stays, brackets, fuel tanks

90

Low ratio of technical personnel5





Uppertanks, wheels

60

Compensation system

PP





Wireharnesses

90

Low ratio of technical personnel5

P∗





Forging parts

44

Process integration In-house forging dies

Capability matrix





Crank-case covers, hubs

68



Cylinder parts, starters

75



Frames

90



Plating service

80

Pivot oil pumps

92

Press parts, dies

10

11

K

1983

500

S

12

L

1978

440

PC

13

M

1981

427

PC

14

N

2001

400

PC

15

O

1981

400

PC





16

P

1998

181

PP









Energy-saving machinery Low ratio of technical personnel5 Extensive QC circles/contests

Brand-specific dedicated factories

5S mutual evaluation system

129

(Continued)

130

Table 4A.1

No. Code Year No. of Ownership2 manufacturing workers began1 (persons)

Status in chain3 OEM 1-tier

2-tier

Main products

After-sales service market 4

% of sales Notes in motorcycles

17

Q

1988

170

PC

18

R

1994

160

PC

19

S

1981

115

PC

20

T

1997

60

PC



Sprockets, stays, machining

53

21

U

2003

38

PP



Brakets

60

22

V

1998

30

PC



Muffler parts, dies

60





4



Brake shoes/ pads

60

Unstable production due to price competition

Forging parts

36

In-house forging dies Low ratio of technical personnel5

Brake shoes/ pads

100

Capability matrix

23

W

1994

90

PP





Dies and molds

70

24

X

2001

90

PP





Plastic/metal dies, jigs

80

25

Y

1996

45

PP





Forging dies

60

Designing dies from products

Designing dies from drawings of products with defects

Note: 1 The year that manufacturing began in the motorcycle industry, and not necessarily coinciding with the year of establishment. 2 PC: Private and Chinese Indonesian, PP: Private and Pribumi, P∗ : Private and corporate ownership, S: State-owned enterprise. 3 : Major status, : Subordinate status. 4 Genuine spare parts for Japanese-brand motorcycles. 5 The ratio of personnel allocated to the workshop is in the lowest range of below 4%. Source: Based on the author’s study.

131

132 Capability Development in Captive Value Chains in Indonesia

Notes 1. “Process upgrading” is defined as “transforming inputs into outputs more efficiently by re-organizing the production system or introducing superior technology,” “product upgrading” as “moving into more sophisticated product lines . . . in terms of increased unit values,” “functional upgrading” as “acquiring new functions in the chain (or abandoning existing functions) to increase the overall skill content of activities,” and “inter-sectoral upgrading” as “using the knowledge acquired in particular chain functions to move into different sectors” (Schmitz ed. 2004). 2. The names of governance types differ according to the literature. But as Humphrey and Schmitz (2004: 377) noted, the “captive network” of Sturgeon (2002), “controlled network” of McCormick and Schmitz (2002), and “quasi-hierarchy” of Schmitz (2004) are very similar concepts. This chapter calls these similar types uniformly “captive,” drawing on Gereffi et al. (2005). 3. See, for instance, Fransman and King (1984); Lall (1987, 1992); Bell and Pavitt (1995); Ernst et al. (1998). 4. Unlike the matrix of Lall (1992), the depth of capabilities in the Capability Matrix is not set by the degree of complexity or difficulty of technology. This is because there is no universal tool that can assess the levels of complexity or difficulty in the absolute sense for all industries and products with varied attributes. Instead, the depth of capabilities is set on the basis of capability development from the learning of the existing technology to generation of something original. So, even if the existing technology is highly complex, the Capability Matrix assesses it at the operational/assimilative levels. Contrariwise, even if original elements are not very technologically complex, the Capability Matrix can assess them at the adaptive/innovative levels. The reason for our concern with elements of originality is that we consider responding to challenges in an original way, rather than just doing as required, to be a key step forward on the path toward innovation. 5. For further discussion on conceptual differences between upgrading in the GVC literature and capabilities in the Capability Matrix, see Sato and Fujita (2009). 6. Figures are based on the author’s hearing study at PT Astra Honda Motor and PT Yamaha Indonesia Motor Manufacturing. For details, see Sato (2005, 2006). 7. Jialing started local assembly before 2000. In 2000–2001 Jianshe, Qianjiang (Sanex), Qingqi, and Zongshen started assembly in joint ventures, and Jincheng, Lifan, Loncin, Chunlan, and Dayang in the form of technical assistance to local firms. 8. These are PT Astra Honda Motor, PT Yamaha Indonesia Motor Manufacturing, and PT Indomobil Suzuki International. Japanese ownership is 50%, 100%, and 49% respectively. Honda and Suzuki brands are manufactured by joint ventures, where the strategy of local partners is also reflected in management. However, in this study such joint ventures are referred to as “Japanese lead firms,” since they are firmly linked with their Japanese parent firms via global value chains.

Yuri Sato 133 9. These estimates are based on: (1) the author’s interviews at the three Japanese lead firms and one local motorcycle manufacturer; (2) directories such as GIAMM (various years); CISI Raya Utama (2004); JETRO Jakarta Center (various years); VISDATIN (2002); and (3) previous studies such as Thee (1997) and Sato (1998). 10. The firms were randomly selected, and were adopted in order from those accepting the study. The study was conducted in 2004 and 2005, by visiting the target firms, interviewing president directors or factory directors, filling questionnaires by the author, and observing production lines at first hand. 11. Of the 25 suppliers, 13 are owned by Chinese Indonesian, 7 by Pribumi (native Malay) Indonesian, 4 by diversely-owned private corporations, and one is state-owned. See also Appendix. 12. The Central Board of Statistics (BPS) in Indonesia defines the size of firms in the manufacturing sector by the number of workers: large firms have more than 100, medium-size firms 20–99, small firms 5–19, and micro firms 1–4 workers. 13. The 5Ss is a term that refers to activities for improving Seiri (arrangement), Seiton (order), Seis¯ o (cleaning), Seiketsu (cleanliness), and Shitsuke (discipline). Kaizen, meaning “improvement,” refers to the improvement of production performance through bottom-up proposals and circle activities. Quality control (QC) circles are a form of Kaizen. These Japanese terms are often used in Indonesia. 14. This study counts as half a point, or 0.5, in the cell for equipment-related capabilities at the assimilative level, if the firm can manufacture DMJT in-house but the ratio of personnel allocated to the workshop is in the lowest range of below 4%, the range where a firm is assessed as being unable to effectively maintain and repair DMJT over time.

References Applebaum, Eileen and Rosemary Batt. 1994. The New American Workplace: Transforming Work Systems in the United States. Ithaca, NY: ILR Press. Ariffin, Norlela and Paulo N. Figueiredo. 2004. “Internationalization of Innovative Capabilities: Counter-Evidence from the Electronics Industry in Malaysia and Brazil.” Oxford Development Studies 32(4): 559–583. Bell, Martin and Keith Pavitt. 1995. “The Development of Technological Capabilities.” In Trade, Technology and International Competitiveness, ed. Irfan ul Haque. EDI Development Studies, Washington DC: The World Bank: 69–101. Cole, Robert E. 1989. Strategies for Learning: Small Group Activities in American, Japanese, and Swedish Industry. Berkeley, CA: University of California Press. CISI Raya Utama. 2004. “Motorcycle Market in Indonesia.” An unpublished study submitted to IDE, Jakarta: CISI Raya Utama. Ernst, Dieter, Tom Ganiatsos, and Lynn Mytelka (eds) 1998. Technological Capabilities and Export Performance: Lessons from East Asia. Cambridge: Cambridge University Press. Figueiredo, Paulo N. 2002. “Does Technological Learning Pay Off? Inter-Firm Differences in Technological Capability-Accumulation Paths and Operational Performance Improvement.” Research Policy 31(1): 73–94.

134 Capability Development in Captive Value Chains in Indonesia Figueiredo, Paulo N. 2003. “Learning, Capability Accumulation and Firms Differences: Evidence from Latecomer Steel.” Industrial and Corporate Change 12(3): 607–643. Figueiredo, Paulo N. 2008. “Industrial Policy Changes and Firm-Level Technological Capability Development: Evidence from Northern Brazil.” World Development 36(1): 55–88. Fransman, Martin and Kenneth King (eds) 1984. Technological Capability in the Third World. London: MacMillan. Fujimoto, Takahiro. 2001. Seisan Manejimento Ny¯umon (Introduction to Production Management) I, II. Tokyo: Nihon Keizai Shimbun, Inc. Gereffi, Gary, John Humphrey, and Timothy Sturgeon. 2005. “The Governance of Global Value Chains.” Review of International Political Economy 12(1): 78–104. GIAMM (Gabungan Industri Alat-Alat Mobil & Motor/Indonesian Automotive Parts and Components Industries Association). Various years. List GIAMM Members. Jakarta: GIAMM. Hayashi, Takeshi (ed.) 1986. (1990 in English) The Japanese Experience in Technology: From Transfer to Self-Reliance. Chapter 3 Theoretical Summary: A Preliminary Examination and an Interim Conclusion. Tokyo: United Nations University Press (http://d-arch.ide.go.jp/je_archive/english/society/book_jes2_ d03.html). Humphrey, John and Hubert Schmitz. 2000. “Governance and Upgrading: Linking Industrial Cluster and Global Value Chain Research.” IDS Working Paper 120. Brighton: Institute of Development Studies, University of Sussex. Humphrey, John and Hubert Schmitz. 2001. “Governance in Global Value Chains.” IDS Bulletin 32(3): 19–29. Humphrey, John and Hubert Schmitz. 2004. “Chain Governance and Upgrading: Taking Stock.” In Local Enterprises in the Global Economy: Issues of Governance and Upgrading, ed. H. Schmitz. Cheltenham/Northampton: Edward Elgar: 349–381. JETRO Jakarta Center. various years. Directory of Japanese Companies and Representative Offices in Indonesia. Jakarta: JETRO Jakarta Center. Lall, Sanjaya. 1987. Learning to Industrialise: The Acquisition of Technological Capability by India. London: Macmillan. Lall, Sanjaya. 1992. “Technological Capabilities and Industrialisation.” World Development 20(2): 165–186. McCormick, D. and Hubert Schmitz. 2002. Manual for Value Chain Research on Homeworkers in the Garment Industry. Brighton: Institute of Development Studies. Morrison, Andrea, Carlo Pietrobelli, and Roberta Rabellotti. 2008. “Global Value Chains and Technological Capabilities: A Framework to Study Learning and Innovation in Developing Countries.” Oxford Development Studies 36(1): 39–58. Ohara, Moriki and Yuri Sato. 2008. “Asian Industrial Development from the Perspective of the Motorcycle Industry.” IDE Discussion Paper No.182. Chiba: Institute of Developing Economies (IDE). Otahara, Jun. 2006. “Nihon no Nirinsha Buhin Sapuraiya: Bungy¯ o k¯ oz¯ o to Torihiki Kankei” (“Japanese Motorcycle Component Suppliers: The Structure of the Division of Labor and Business Relationships”). In Ajia no Nirinsha Sangy¯o: Jiba Kigy¯o no Bokk¯o to Sangy¯o Hatten Dainamizumu (Asia’s Motorcycle Industry:

Yuri Sato 135 The Rise of Local Firms and the Dynamism of Industrial Development), ed. Yuri Sato and Moriki Ohara. Chiba: IDE: 95–130. Sato, Yuri. 1998. “The Machinery Component Industry in Indonesia: Emerging Subcontracting Networks.” In Changing Industrial Structures and Business Strategies in Indonesia, ed. Y. Sato. Tokyo: IDE: 107–148. Sato, Yuri. 2005. “Indonesia no Nirinsha Sangyo” (Indonesia’s Motorcycle Industry). In Ajia no Nirinsha Sangy¯o: Kiso J¯oh¯o to Kigy¯o Ichiran (Asia’s Motorcycle Industry: Basic Information and Company Directory), ed. Y. Sato and M. Ohara. Chiba: IDE: 367–428. Sato, Yuri. 2006. “Indoneshia no Nirinsha Sangy¯ o: Jiba Kigy¯ o no N¯ oryoku Keisei to Sangy¯ o Kiban no Kakudai” (“Indonesia’s Motorcycle Industry: Capability Development of Local Firms and the Expansion of the Industrial Base”). In Ajia no Nirinsha Sangy¯o: Jiba Kigy¯o no Bokk¯o to Sangy¯o Hatten Dainamizumu (Asia’s Motorcycle Industry: The Rise of Local Firms and the Dynamism of Industrial Development), ed. Yuri Sato and Moriki Ohara. Chiba: IDE: 281–322. Sato, Yuri and Mai Fujita. 2009. “Capability Matrix: A Framework for Analyzing Capabilities in Value Chains.” IDE Discussion Paper No. 219. Chiba: IDE. Schmitz, Hubert, ed. 2004. Local Enterprises in the Global Economy: Issues of Governance and Upgrading. Cheltenham/Northampton: Edward Elgar. Sturgeon, Timothy. 2002. “Modular Production Networks: A New American Model of Industrial Organization.” Industrial and Corporate Change 11(3): 451–495. Suehiro, Akira. 2000. (2008 in English). Catch-Up Industrialization: The Trajectory and Prospects of East Asian Economies. Nagoya: Nagoya University Press (Hawaii: University of Hawaii Press in English). Thee, Kian Wie. 1997. “The Development of the Motorcycle Industry in Indonesia.” In Waves of Change in Indonesia’s Manufacturing Industry, ed. Mari E. Pangestu and Yuri Sato. Tokyo: IDE: 95–135. VISDATIN. 2002. Study of Motorcycle Industry in Indonesia. Jakarta: VISDATIN.

5 To Be or Not to Be a Supplier to TNCs? An Entrepreneurial Approach to Linkage Formation in the Malaysian Electronics Industry Hiroshi Oikawa

5.1. Introduction East Asia has emerged as a major global manufacturing hub for a number of industries, not only for labor-intensive products such as consumer goods and apparel, but also for technology-intensive products such as electronics and automobiles. In technology-intensive industries, in particular, affiliates of transnational corporations (TNCs) have played a vital role in boosting the rapid export-oriented industrialization of this region. The contributions of TNCs appear in various forms: the creation of direct and indirect employment, the generation of foreign exchange, and contributions to the development of local industrial clusters and related activities. Moreover, TNCs provide local manufacturers with new opportunities to access global markets and upgrade their technological and managerial capabilities. Governments in developing countries have been motivated to formulate policies to attract TNCs’ affiliates in hope of enjoying these benefits. Such developmental strategies are particularly pertinent to the economies of Southeast Asia, which have arguably been among the most successful in leveraging economic development brought by TNCs, so much so that Hobday (2001) described the pattern of ASEAN’s economic development as “TNC-led industrialization.” The present study focuses on Malaysia. The country has adopted a TNC-focused development strategy, which has also been pursued by other major ASEAN economies such as Singapore, Indonesia, Thailand, and the Philippines. In the mid-1990s, the country was one of the top ten recipients of foreign direct investment (FDI) among developing countries (UNCTAD 1999). Indeed, TNCs have made a major 136

Hiroshi Oikawa

137

contribution to Malaysia’s core manufacturing industries such as electronics and autos in terms of capital formation, output, export, employment, and value-added (Sieh 2006). Malaysia’s relative success in attracting TNCs can be explained by a variety of factors, including a relatively cheap and highly trainable labor force, political stability, efficient government policies, and the existence of local supporting industries (Rasiah 1995; Best and Rasiah 2003; Giroud 2003; Jomo et al. 2003). Among these factors, the availability of a dynamic local supply base gives the country a crucial and fundamental competitive edge (Rasiah 1995, 2002), by helping it to attract and retain TNCs’ affiliates. The study presented in this chapter examines this process, focusing on the linkage formation between TNCs’ affiliates and local suppliers in Penang, Malaysia. The case of Penang is examined because of the large number of electronics TNC affiliated that have located in this state, more than in other successful industrial districts like Johor and Kelang Valley in Malaysia (Best 1999; Rasiah 2005). The electronics and electrical industries are the main foci of this chapter since these industries are the largest recipients of FDI and the principal drivers behind industrial development in the Malaysian manufacturing sector (Nazari 1995; Best and Rasiah 2003). The degree to which TNCs commit to the host economy is one of the central concerns for development economists and policymakers in developing countries. Are investments long lasting or “footloose”? Do affiliates create robust linkages to the local supply base? Is any research, technical development, or design work conducted? This study investigates determinants, degree, and character of TNCs’ embeddedness in the Malaysian economy through linkage creation. It finds that the overall degree and character of linkage formation are determined by three sets of factors: the outsourcing strategies of TNCs’ local affiliates; local entrepreneurial response to the opportunities and constraints TNCs create; and host government policies. Among these, the main focus in what follows is on the first and second factors, particularly the response of local firms to the opportunities provided by foreign affiliates. While the importance of government policy in supporting and shaping linkage formation cannot be denied, it is not a central concern here since a number of studies have already examined this issue in depth (Altenburg 2000; UNCTAD 2001; Best and Rasiah 2003; Rasiah 2005). By contrast, little attention has been paid to the role of entrepreneurship in linkage formation. Hence, the present study focuses on this aspect. Like other chapters in this volume, I ask the question, “Do supplier strategies matter?”

138 To Be or Not to Be a Supplier to TNCs in Malaysia?

It is a well-established, but not well-studied, fact that local entrepreneurship is a prerequisite for successful regional industrial development (Kilby 1971; Leff 1979). Entrepreneurship encompasses the motivation, goals, strategies, limitations, and other factors relevant to the potential of local firms to learn via the formation and development of linkages with foreign affiliates. The research I present in this chapter began with the premise that successful entrepreneurship is not the result of a single common trait; that a variety of characteristics will be present among successful entrepreneurs. It is not easy to become a supplier to a global company; so, arguably, a person who has successfully achieved this must possess some positive entrepreneurial capabilities. Using the entrepreneurial orientation (EO) method developed by Covin and Slevin (1989, 1991), my research compares TNC suppliers to non-TNC suppliers with the goal of gaining insights that will be useful in shaping entrepreneurial-based development policy. The issue of linkage creation is central to global value chain (GVC) analysis (Gereffi and Korzeniewicz 1994; Henderson 1998; Gereffi 1999; Gereffi and Kaplinsky 2001; Schmitz 2004). GVCs refer to the full range of activities from upstream processes to the final stage of production to sales and end use, activities which increasingly span a number of national boundaries. One of the central questions addressed by GVC analysis is whether insertion into GVCs is beneficial or harmful to the firms in developing countries. To answer this question, GVC research needs to provide a systematic explanation of the process of global value chain formation, particularly in its initial stages, when interactions between TNCs and local firms are formed and entrepreneurs inculcate the business culture of their companies. This chapter is organized as follows. Section 5.2 presents a detailed discussion of how linkages have been examined in GVC studies. Section 5.3 provides a brief sketch of the history and performance of Malaysian FDI and TNC policies. Section 5.4 describes the methodology used for my fieldwork in Penang, Malaysia. Section 5.5, based on the results of this fieldwork, discusses TNCs’ local purchasing practices and the reaction of local entrepreneurs. The results are summarized and implications are also discussed in the final section.

5.2. Study of linkages in GVC analysis In the literature on GVCs, dominant TNCs, or lead firms, are often expected to make a substantial contribution to upgrading and restructuring their supply chains. Because lead firms want local suppliers

Hiroshi Oikawa

139

with capabilities that will enable them to function efficiently, insertion into GVCs may mean substantial impetus and support for upgrading. On the other hand, some lead firms may bring their own in-house resources and/or suppliers with them, asking very little if anything from local suppliers, or simply move on if capabilities are inadequate. Nevertheless, for local producers, developing supply relationships with local TNC affiliates can provide opportunities for “upgrading within production and subsequently into design, marketing and branding as a combination of learning by exporting and organizational succession” (Humphrey and Schmitz 2002), although the process is typically difficult and fraught with limitations (Sturgeon and Lester 2004). This study can contribute to GVC analysis in three ways. First, the present research examines the process of linkage creation and development between TNC affiliates and local firms. The study finds that the process of initial linkage creation takes precedence over all others aspects related to the issues of GVCs including the three key variables of GVC governance theory proposed by Gereffi et al. (2005), i.e. complexity, codifiability, and supplier competence. Clearly, one of the most formidable challenges for firms in developing countries is finding a way to approach TNCs. It is therefore imperative that GVC analysis provides an in-depth examination of this process. The research presented in this chapter provides a detailed analysis of GVC linkage formation in the Penang, Malaysia region. Second, care must be taken when examining the flow of value across national boundaries. Stated simply, the more local firms participate in GVCs, the more the host country receives value-added from TNCs’ global operations. This view may remind us of the old debate about economic enclaves, where special economic zones and free trade industrial estates seek to capture value added of any kind and at all costs. In contrast, current GVC analysis pays greater attention to the quality of the linkage, that is, the question of how different GVC governance modes affect value capture, power, autonomy, and control in the context of a single bilateral buyer–suppler relationship (Gereffi et al. 2005). While the importance of this micro view of chain dynamics cannot be denied, from the perspective of development economics, equal importance should be given to macro and dynamic perspectives that lead an examination of the process in which local firms become embedded in GVCs. Linkage studies can contribute to this issue by examining incipient interactions between TNCs and local firms in developing countries.

140 To Be or Not to Be a Supplier to TNCs in Malaysia?

Finally, in GVC studies, more attention should be paid to the indigenous factors that may or may not facilitate economic development, particularly the purposeful efforts embodied in the strategies of local firms in developing countries. Again, three variables have been emphasized in GVC governance theory—complexity, codifiability, and supply base capabilities (Gereffi et al. 2005; Sturgeon 2007). The first two variables are principally determined by technical attributes of the products and processes within the chain. But what determines supply base capabilities? In the GVC literature, local firms in developing countries are largely presented as passive and submissive to the supremacy of TNCs. However, in reality, local firms can, and sometimes do, aggressively pursue technological learning and other competence building efforts. This too can influence the magnitude and direction of supply base development. Clearly, aggressive technological and managerial efforts by local suppliers can work to alter inter-firm dynamics in GVCs, especially over time. However, the research presented here suggests that the initial character of linkage formation matters. Therefore, close attention is paid here to how local entrepreneurs’ motivations, strategies, plans, and goals relate to GVC governance at the time when linkages are created. I believe that investigating the role of entrepreneurship and linkage creation yields important insights into supply base upgrading in GVCs, and should be formalized within GVC analysis.

5.3. FDI-led development of the Malaysian electronics industry Foreign direct investment has played a prominent role in Malaysia’s economic development. The initial attraction of Malaysia as an investment destination was the availability of a cheap but highly trainable, English-speaking workforce, combined with access to the protected local market. Since 1990, Malaysia has ranked as the 15th largest recipient of FDI inflows in the world and fourth among non-OECD (Organisation for Economic Co-operation and Development) countries after China, Brazil, and Singapore. Figure 5.1 shows the level of inward FDI flowing to Malaysia in comparison with other major ASEAN economies. Malaysia was among the largest FDI recipients in the region, although the relative level of inflows has declined notably since the Asian financial crisis of 1997–1998, after which China, and later Vietnam, attracted the largest share of investment.1 In addition, Malaysia has the second highest ratio of FDI to gross fixed capital formation (19.4% during the early 1990s) among the

Hiroshi Oikawa

141

7000 Malaysia

Vietnam Thailand

US$ million

5000 3000 1000

–1000 –3000

1990

1995

2000

2005

Philippines

Indonesia

–5000

Figure 5.1 Total FDI inflows into major developing ASEAN Countries (1988– 2007) Source: Asian Development Bank. Key Indicators (various issues).

ASEAN member countries, after Singapore (30.3%) (UNCTAD 2001). Thus, FDI has played a key role in Malaysia’s industrialization. Over the same period, FDI was concentrated in the manufacturing sector, especially in the electrical and electronics industries, which accounted for more than half of the gross national capital invested (Sieh 2006). Penang was the first popular destination for TNC investment in Malaysia, and it has maintained a strong TNC presence in electronics manufacturing that began in the early 1970s. In 1971, Clarion and National Semiconductor set up assembly plants in a free trade zone in Penang Island. By 1974, Intel, Motorola, Hewlett Packard, AMD, and Hitachi had followed. TNCs came from the United States, Japan, Europe, and other newly industrializing Asian economies. Consumer electronics assembly became an important sector in the late 1970s, and even more so in the late 1980s after Sony and Toshiba started operations there. Although the local procurement ratio of TNC affiliates remains quite low, Penang is often mentioned as a successful industrial cluster development where TNC affiliates enjoy strong linkages with local firms and good coordination with support institutions (Rasiah 1995, 2005; Eiteman 1997; Yazicioglu 2003). However, it has taken a long time to reach this stage of development. It has been widely observed that linkages between local firms and TNCs in free trade zones (FTZs) are

142 To Be or Not to Be a Supplier to TNCs in Malaysia?

generally weak, and Penang is no exception. Warr (1989) compared the performance of four export processing zones (EPZs)—in Indonesia, Korea, Malaysia, and the Philippines—during the 1970s and early 1980s. In most industries, Malaysia recorded the highest local content ratios, although the ratio of raw material purchasing was still less than 10%. Driffield and Noor (1999) summarized the results of local procurement performance reported in previous literature, and reported mixed results in which TNCs’ local procurement ratio in Malaysia in the late 1980s varied widely from TNC to TNC, from 9% to 77%. During the 1990s, the performance of the Malaysian electronics sector was not outstanding in terms of retaining domestic value-added, although the country outperformed other ASEAN economies except for Indonesia (Oikawa 2011). While the variability of these results is too great to provide any decisive conclusions, the fact that even the smallest local procurement ratio exceeds Warr’s (1989) estimation strongly suggests that TNCs in Malaysia gradually increased their local procurement during the 1970s and 1980s. In 2007, an in-depth survey of Japanese manufacturing companies’ purchasing in Southeast Asia was conducted by Japan’s External Trade Organization (JETRO) (JETRO 2008). Using average data gained from a total of 612 TNC affiliates, the results reveal the origin of procurements and ownership structure of suppliers to Japanese TNC affiliates, summarized in Figures 5.2 and 5.3. These data reveal clear differences in the local purchasing practices of Japanese TNC affiliates in different ASEAN economies. In less industrialized economies such as Vietnam and the Philippines, Japanese TNC affiliates tended to depend less on local procurement (Indonesia is an exception, which may require more detailed examination). Among the ASEAN countries, a relatively high rate of local procurement was reported in Thailand and Malaysia. Figure 5.2 shows that local procurement in Thailand may be comparable to, or greater than, that in Malaysia; however, substantial procurement ratio from locally owned suppliers was larger in Malaysia than Thailand. The figure indicates that local TNC affiliates in Thailand purchased mainly (53.2%) from Japanese-owned local suppliers, while those in Malaysia purchased mainly (61.7%) from locally owned and marginally (34.6%) from Japanese-owned suppliers. Considering the direct and indirect effects of inter-firm linkages, it may be that locallyowned suppliers in Thailand are less involved in Japanese GVCs than those in Malaysia. Thus, it is reasonable to conclude that, among the ASEAN economies, industrial linkages between locally-owned suppliers and TNCs are relatively strong in Malaysia.

Hiroshi Oikawa

0% 10%

20%

30%

Thailand

40%

50%

53.9

Malaysia

Indonesia

Vietnam

8.3

52.3

23.0

Local

9.9

38.8

12.1

26.5

9.5

18.0

34.0 11.9

27.3

Philippines

90% 100%

13.7

32.1

18.4

39.5

80%

31.5

13.0

29.6

70%

5.2

41.3

Singapore

60%

9.4

41.2

ASEAN

143

Japan

Others

Figure 5.2 Origins of purchases by Japanese manufacturing affiliates in major ASEAN countries Source: JETRO 2008.

0%

10%

Thailand

20%

30%

40%

50%

60%

41.5

Malaysia

70%

43.3

Indonesia

44.4

Locally owned

3.7 12.3 4.7

58.4 50.5

Vietnam

5.3

44.9

36.0

32.4

Japanese-owned local

100%

34.6

50.4

Philippines

90%

53.2 61.7

Singapore

80%

5.7 17.2

Other foreign-owned local

Figure 5.3 Ownership structure of local companies purchased by Japanese manufacturing affiliates in major ASEAN countries Source: JETRO 2008.

The conclusions regarding local content ratio drawn from the previous empirical studies are also supported by several qualitative investigations that suggest that local content and linkage creation in Penang has gradually been improving. Rasiah (1995) described the rapid emergence of local firms in Penang supplying machine-tools and components to overseas markets as well as to locally operating TNC affiliates. Penang is

144 To Be or Not to Be a Supplier to TNCs in Malaysia?

often pointed to as a successful industrial area whose success has come from vigorous local suppliers that support the operations of local affiliates of leading global TNCs like Intel, Bosch, SONY, and Hewlett Packard (UNCTAD 2001; Jomo et al. 2003). In summary, previous empirical studies suggest that inter-firm linkages in the Malaysian electric and electronics sectors have slowly but steadily strengthened, particularly in Penang area. Consequently, the Malaysian economy has performed relatively well in comparison with some other ASEAN economies with weaker supply bases. As is mentioned above, the emergence of competent local suppliers based on vigorous entrepreneurship is among the factors underlying the emergence of a successful industrial agglomeration in Penang. The following sections examine the entrepreneurial factors that facilitate linkage creation between TNCs and local firms.

5.4. In what way does entrepreneurship matter? Use of entrepreneurial orientation The presence of vigorous and technologically competent local suppliers is a prerequisite for successful linkage formation with TNCs. Only aggressive entrepreneurial firms will try to find and grasp the profit and learning opportunities offered by the TNCs that use local suppliers. On the other hand, local firms with weak or scant entrepreneurial motivation may be inert or inattentive and, therefore, not respond swiftly or appropriately to the potential profit opportunities presented by TNC affiliates. Hence, it follows that strong inter-firm linkages will be in some way related to the availability of local entrepreneurs keen to start business with TNCs. This study contends that not all entrepreneurs are identical. Local firms that become successful suppliers of TNCs will exhibit characteristics that differ from those of other local firms. I also hypothesize that there are differences between the entrepreneurial characteristics of firms that supply TNCs and those that do not. The objective of this study is to test this hypothesis by identifying and examining these differences using the entrepreneurial orientation (EO) criteria (Miller and Friesen 1982; Covin and Slevin 1989; Lumpkin and Dess 1996). 5.4.1. Characterizing entrepreneurship; the entrepreneurial orientation method The EO measure was first proposed by Covin and Slevin (1989), proving a useful way to characterize entrepreneurs. Because entrepreneurship is a multifaceted and fuzzy concept that cannot be represented by a

Hiroshi Oikawa

145

single measurement, these authors divide entrepreneurship into three broad aspects: risk-taking, innovation, and proactivity. While I follow Colvin and Slevin’s approach to a degree, I found that their typology was not immediately applicable to developing countries. I also found that a broader range of EO types are needed to account for the whole spectrum of entrepreneurial tasks, as described by various scholars such as Knight (1921), Schumpeter (1934), McClelland (1961, 1987), Leibenstein (1968), Kirzner (1973, 1997), Baumol (1990), and others. For example, in the context of developing countries, where local firms are generally trying to catch up to the established global frontier, Covin and Slevin’s term “innovation orientation” should be replaced by “technology-learning orientation.” This is because most rapid productivity growth is achieved through a learning process rather than an innovation process. Therefore, I have reordered and extended the three original EO measures (i.e. risk-taking, innovation, and proactivity) into four: (a) market orientation, (b) technology-learning orientation, (c) risk and uncertainty tolerance, and (d) autonomy orientation. (a) Market orientation Discovering, accessing, and capturing market opportunities are fundamental entrepreneurial functions (Kirzner 1973). Market-oriented entrepreneurs are adept at recognizing and responding to changes in consumer preferences and industrial structure when a number of firms are in keen competition. Entrepreneurs that are alert to market changes have an ability to anticipate, address, and capitalize on opportunities. Market-oriented entrepreneurs tend to seek competitive advantage through sales promotion, low-cost production, cultivation of unknown markets, and better customer service such as quick delivery and substantive after-sales support. A fundamental element of market orientation is entrepreneurs’ ability to adapt to changing and uncertain environments. These efforts may not always require innovation or technological learning, as a highly alert entrepreneur can earn profits in unexplored, unformed, or uncultivated markets with little or no technological improvement or truly novel products or processes. Therefore, it is possible for a successful entrepreneur to have high market orientation but low technology-learning orientation (see below). (b) Technology-learning orientation According to Schumpeter (1934), the title of “entrepreneur” is closely associated with the notion of “innovation.” Typically, entrepreneurs are motivated to actively shape their environment, sometimes initiating a process of “creative destruction,” where existing technologies, markets,

146 To Be or Not to Be a Supplier to TNCs in Malaysia?

and firms are swept aside to make way for the new. However, in the context of developing countries, such radical innovation of this sort is rare for local firms. Instead, the nature of technological advancement tends to be adaptive and assimilative (Lall 2002). Thus, in the EO construct adopted here, the word innovation has been replaced by the concept of “technology learning.” A technology-learning orientation signifies an entrepreneur’s inclination to use technological improvement as a means of improving business opportunities. Bear in mind that not all entrepreneurs reveal a strong technological orientation: some are conservative, conventional, uncreative, and are willing to remain followers in the realm of technology. Entrepreneurs with a lesser technology-learning orientation may have more or less market alertness and ability to adapt to changing markets. Thus, as mentioned above, technology-learning and market orientations can occur together or even separately. (c) Risk and uncertainty tolerance Entrepreneurial behavior always involves elements of risk, uncertainty (Schumpeter 1934), and speculation (Kirzner 1997). Entrepreneurs take these risks as long as they expect the overall return from the venture to be positive. Entrepreneurs tend to possess a high tolerance for uncertainty, particularly during the early stages of market entry. It is hypothesized that TNC suppliers take higher risks than non-TNC suppliers because the former tend to be exposed to greater technological uncertainty, competition from overseas suppliers, and financial risks from large-scale capital investment. Risk and uncertainty is less for developing country suppliers that concentrate on more stable domestic markets with lesser demand from customer companies. (d) Autonomy orientation Entrepreneurs are often driven by a desire to establish their own domain of control and power, or “realm” (Schumpeter 1934). Entrepreneurs with high autonomy orientation tend to express their own individuality in the workplace, dislike taking orders from superiors, and refuse to be a mere cog in an organizational machine. Having an autonomy orientation means that the entrepreneur places high importance on his or her own decisions and is averse to being in an ancillary position. Thus, it is expected that entrepreneurs with a strong autonomy orientation will possess a strong motivation to realize their own ideas and visions. However, due to the extreme asymmetry in power between local firms and global TNCs, TNC suppliers tend to play a subservient role

Hiroshi Oikawa

147

in GVCs. Thus, given the power asymmetry typical of captive GVCs, an interesting question is whether entrepreneurs possessing higher autonomy orientations might choose not to become TNC suppliers.

5.5. Research method One goal of this study is to examine which EO combinations are present in TNC suppliers. Most data were collected during semi-structured interviews conducted by the author in Penang, Malaysia, over a two-week period in October 2008. Interviewees included managers at one large Japanese TNC, five suppliers subcontracted to this TNC, as well as four non-TNC suppliers. Due to time constraints, the sample size was small and rigorous statistical analysis could not be conducted. Hence, a case study approach is employed. The TNC concerned (called Company X, hereafter, for confidentiality reasons) is a 100% Japanese-owned home electronics manufacturer. Company X came to Bayan Lepas Industrial Estate, an FTZ on Penang Island, in 1984. The main products manufactured at this location are OEM car-audio systems and personal computer batteries, sold mainly to overseas customers. Domestic sales are negligible. The company is a relatively large-scale firm in the zone, employing nearly 3000 workers. Interviews were conducted with general managers, purchasing officers, and engineers, including both Japanese and local staff. The interviews took more than two days, in which particular attention was paid to the management of purchasing practices. The purchasing manager provided a full list of approved suppliers along with his evaluation of the capability of each supplier. Interviews were then conducted at two different groups of suppliers, Company X’s suppliers and local suppliers that were not linked to any TNCs (non-TNC suppliers). The two groups were approached in different ways. Contacts with TNC suppliers were facilitated by introductions from Company X. Non-TNC suppliers were selected from the Malaysian Trade Directory 2007 and the “investPenang” (a government agency) website. Contacts with them were made via fax or email. In most cases, interviews were conducted with a company’s founder or managing director. In general, supplier interviews lasted between one and a half and two hours. The goal of the interviews was to understand the entrepreneur’s career path, personal profile, motivations, technological efforts, and managerial perceptions. Interview questions were related to the four EO elements presented above. Traditional EO assessment is based on self-reports, primarily by

148 To Be or Not to Be a Supplier to TNCs in Malaysia?

owners or managing directors. However, the approach adopted in this study was slightly different in the sense that entrepreneurs’ anecdotes were recorded and later analyzed by the author. Based on this analysis, the strength of entrepreneurial orientation was rated by the author for each of the four EO elements from 1 (weak) to 5 (strong). To ensure strict comparison, the average rating with four EO elements was standardized to 3.5 for individual entrepreneurs.

5.6. Results In this section, the empirical results of the research are described. The analysis is exploratory in nature, and intended to set the stage for future research examining the purchasing strategies of TNCs and the EO differences between TNC and non-TNC suppliers. The data gathered serve to test the four-element EO model defined above and identify relevant theoretical issues. At the end of each subsection, the findings are summarized. 5.6.1. TNCs’ purchasing structure Purchasing by TNCs’ local affiliates creates outsourcing demand and the potential for linkages with local suppliers. A TNC’s intention to purchase locally, for whatever reason, constitutes a business opportunity for local suppliers as long as the work is not allotted to suppliers on a predetermined basis; for example, to a supplier from Japan or another country where the lead firm operates. Thus, this study examines Company X’s outsourcing strategies in order to identify the nature of these opportunities. Although it is not appropriate to draw general conclusions from a single case study, the case study method can nevertheless provide insights onto the interaction between a phenomenon and its context (Yin 1994). The data gathered through fieldwork are presented in Tables 5.1 and 5.2 and Figures 5.4 and 5.5. From these data, four notable features stand out as follows. 5.6.1.1. Increases in local purchasing Table 5.1, which presents Company X’s sourcing structure by component and source country, shows an increase in the local purchasing ratio over a short, three-year period. The local purchasing ratio increased steadily from 18.5% in 2005 to 36.4% in 2007. However, care must be taken in interpreting these results because imported materials may be included in local procurement if they were obtained through local sales representatives, particularly in the case of electrical parts.

Hiroshi Oikawa

149

Table 5.1 Outsourcing structure of Company X (by component and country)

2007

Local Japan Korea/ Singapore Others Total China

Electrical parts

21.2

0.0

1.3

77.5

0.0

100

Metal parts

94.6

4.6

0.0

0.8

0.0

100

6.8

Rubber, plastic

99.8

0.0

0.0

0.2

0.0

100

15.1

Electromechanical

15.9

0.1

0.1

83.4

0.6

100

54.0

0.0

35.4

0.5

64.2

0.0

100

4.1

100.0

0.0

0.0

0.0

0.0

100

0.0

Packaging & non-core

94.8

0.0

0.0

5.2

0.0

100

2.9

Share to total cost

36.4

1.8

0.0

61.2

0.3

0

100.0

Electrical parts

41.0

0.0

0.8

58.2

0.0

100

33.5

Metal parts

87.7

10.9

0.0

1.4

0.0

100

4.7

Rubber, plastic

99.6

0.0

0.0

0.4

0.0

100

9.7

Electromechanical

10.7

0.0

0.1

88.5

0.7

100

46.1

0.0

42.4

0.0

57.6

0.0

100

4.3

100.0

0.0

0.0

0.0

0.0

100

0.0

Packaging & non-core

92.4

0.0

0.0

7.6

0.0

100

1.7

Share to total cost

34.0

2.3

0.3

63.0

0.3

0

100.0

9.7

0.0

1.0

89.3

0.0

100

30.7

Metal parts

77.2

20.4

0.0

2.4

0.0

100

3.1

Rubber, plastic

99.9

0.0

0.0

0.1

0.0

100

9.5

Mechanical parts

2006

Semi-assembled

Mechanical parts Semi-assembled

Electrical parts

2005

(Percentage) Share of item to total costs 17.1

Electromechanical

5.1

0.0

0.1

94.0

0.9

100

52.4

Mechanical parts

0.0

63.2

0.0

36.8

0.0

100

3.2

100.0

0.0

0.0

0.0

0.0

100

0.0

Packaging & non-core

90.3

0.0

0.0

9.7

0.0

100

1.1

Share to total cost

18.5

2.7

0.3

78.1

0.5

0

100.0

Semi-assembled

Source: Interviews by author.

Whatever the exact values are, the increase in local purchasing can be partly explained by Company X’s expanded efforts to find competent local suppliers. As shown in Table 5.2, local suppliers accounted for nearly half of Company X’s suppliers in 2007. All Company X’s local suppliers were Malaysian-Chinese owned. Cost advantage is an important factor for any company, including Company X, which always sought lower cost sources of supply. Local suppliers are in an advantageous position because (i) in general, they can offer lower prices than importers or foreign-owned local suppliers

150 To Be or Not to Be a Supplier to TNCs in Malaysia? Table 5.2 Number of suppliers of Company X (by component and country) Local

Japan

Korea/ Singapore China

Electrical parts Metal parts Rubber, plastic Electromechanical Mechanical parts Semi-assembled Packaging & non-core

11 6 7 22 0 1 19

0 1 0 1 1 0 0

1 0 0 3 1 0 0

Total

66

3

5

Others

Total

27 3 3 24 1 0 2

0 0 0 3 0 0 0

39 10 10 53 3 1 21

60

3

137

Source: Interviews by author.

do because of lower operating costs, (ii) local purchasing can reduce transportation costs and import duties, (iii) sourcing locally reduces currency risks, and (iv) lead time can be reduced with easy and regular communication with suppliers. Government policies such as local content requirements may have stimulated this process, but this was not a necessary condition. It is important to note that Company X’s local affiliate cannot expand local purchasing at its sole discretion. The company’s main products are “original equipment” items, meaning items made to customer specifications. Thus, the production process, quality control, and supply chain management are strictly monitored and frequently inspected by the parent and customer companies outside Malaysia. Suppliers are carefully scrutinized and need to be approved, not only by the local affiliate, but also by the quality inspection department located in Japan. As this is a time-consuming and costly process, the parent company is often reluctant to add new suppliers. In this regard, being far from headquarters can be disadvantageous for local suppliers who seek approval as qualified suppliers. This is a key insight from the GVC literature. Coordination and control of the GVC often resides at the headquarters of lead firms in developed economies. These decisions are nearly always taken without consultation with local governments and firms in host countries, and even local affiliates often have limited input. When we consider the prospects for TNC suppliers, these conditions place severe limits of what can be achieved in developing countries though entrepreneurial initiative. However, in the case of Company X, ongoing global price competition and the need for a faster response to changing market demands did drive the company to streamline its organizational structure to become

Hiroshi Oikawa

151

leaner, more flexible, and less centralized. Beginning in 2005, the local affiliate in Penang was given more autonomy in various areas, including purchasing management. While the procurement of core components such as semiconductors and batteries remained under the strict control of headquarters in Japan, the local affiliate was given more leeway to select suppliers of non-core items. Without such devolution of authority, the increase in local purchasing would have surely been less than what observed in my fieldwork. 5.6.1.2. Limitation to marginal production processes Despite an increase in local purchasing, a structural obstacle to further increasing local procurement remains. As illustrated in Figure 5.4, the involvement of local suppliers in Penang is strictly limited to marginal processes. The figure shows the four stages at which local suppliers have the opportunity to become involved in the value chain. These are: (A) R&D, (B) design modification, (C) bidding, and (D) supplier change after starting mass production. Obviously, core and strategically important suppliers are invited to provide input at the early stages of product development, such as (A) and (B). They are guaranteed to receive orders in exchange for their contribution to product development and provision of specialized knowledge. With only a few

Basic research Market research

(A) Product planning

(B) Detailed design

Basic design

Early supplier involvement

Technical testing Prototype manufacturing

(C) Arrangement of materials & parts Mold and parts manufacturing

Prototype mass production, testing

Re-equipment facilities (D) Production, inspection

Engineering change

Shipment, sales

Figure 5.4 Timing of supplier selection and involvement in the flow of product design and production Source: By the author.

152 To Be or Not to Be a Supplier to TNCs in Malaysia?

exceptions, Company X maintained the functions of (A) and (B) at home in Japan, in proximity to its main customers. As a result, only the relatively marginal entry points (C) and (D) are open to Malaysian suppliers. The detailed technical specifications of each component are largely predetermined in Japan. Since suppliers in Malaysia are provided with these specifications, Company X’s value chain offers only limited opportunities for local firms to upgrade their technologies and to fully participate in the larger value-added segments along the value chain. Again, the GVC approach emphasizes global divisions of labor of this sort and how they can place limits on opportunities for economic development and technological learning in developing countries. 5.6.1.3. Concentration of locally purchased items A third finding refers to types of products that are purchased locally. Tables 5.1 and 5.2 indicate the specific product categories where local sourcing increased: rubber, plastic, and metal parts; packaging; and other technologically simple and non-core products such as stickers, plastic bags, and silica-gel. These products were predominantly purchased from local suppliers, although ‘local’ may include foreign-owned local suppliers and local sales representatives of trading companies. On the other hand, electronic, electromechanical, and precision mechanical parts (e.g., complex electromechanical subassemblies, batteries, and advanced electronic components like semiconductors) are largely imported, mainly via Singapore. The role of Singapore again suggests a regional division of labor where more advanced functions in the GVC led by Company X are accomplished outside of Malaysia. The reasons for this pattern of specialization are partly technical. Locally purchased items are generally bulky and/or require relatively simple technology to manufacture, such as corrugated boxes, plastic items, and simple metal parts. For such products, local firms enjoy a proximity advantage. On the other hand, imported items are technologically sophisticated and too complicated for locally owned suppliers to produce on their own. Thus, these items continue to be imported. Moreover, as these items are generally high-priced but non-bulky, the per-unit logistic costs of importing them are relatively low. For such items, regional and global purchasing is a viable strategy and there is a higher barrier of entry into GVC for local suppliers. 5.6.1.4. Emphasis on the long-term commitment Figure 5.5 shows the nationality and the number of firms (including both local and overseas firms) that have been supplying Company X for

Hiroshi Oikawa

153

45 40

Others Singapore

Number of suppliers

35 30 25

Japan

20 15 10 5 0

Malaysia 0–4

5–9

10–19

20+

Years Figure 5.5 Years of business of incumbent suppliers with Company X Source: By the author.

a considerable time. The Company listed over 130 suppliers in 2007. The largest group of 40 suppliers has done business with the Company for 20 years or more, a smaller second group of 38 suppliers between 10 and 19 years, a third group of 26 between 5 and 9 years, and a twenty fewer than 5 years. These figures clearly demonstrate that Company X, like many large Japanese companies, seeks long-term relationships with its suppliers. In fact, according to a purchasing manager interviewed for the study, supplier replacement is not frequent, with only a few cases annually. As a result, the number of new suppliers is gradually decreasing. This implies that, once established, buyer–supplier relationships tend to last for a long time and, on the other hand, new opportunities for local suppliers to participate in GVCs diminish. This is one reason why the process of initial linkage formation is so important. However limiting the lead firm’s outsourcing strategies are, Figure 5.5 suggests that, in recent years, opportunities for Malaysian firms to become suppliers has been expanding. While Japanese suppliers dominated the oldest group (i.e., 73%), the share was only 40% in the youngest group, in which the share of Malaysian suppliers was 35%. The growing presence of local suppliers was consistent with the Company’s efforts to increasing local purchasing. To summarize, the purchasing practices of Company X present both favorable and unfavorable conditions for local suppliers wishing to

154 To Be or Not to Be a Supplier to TNCs in Malaysia?

participate in the company’s supply chain. On the one hand, the proportion of local purchasing is increasing, which suggests that Company X is open to being approached by new local firms. But clearly, opportunities are limited to the simple processes and inputs that are open for new bidding and/or supplier replacement. In addition, because the company prefers stable, long-term relations with a limited number of suppliers, the number of opportunities for new suppliers to make inroads remains limited. Obviously, unstable supplier relations are usually not compatible with rising product standards in either quality or reliability. As a result, increasing barriers to entry can prevent potential suppliers from becoming approved suppliers. However, the implication is that once selected, local suppliers have a chance to forge a long-term relationship. Balancing this, with limits, are new opportunities from Company X’s drive to increase local sourcing. 5.6.2. Entrepreneurial characteristics of TNC and non-TNC suppliers This section examines the entrepreneurial characteristics of local suppliers in Penang. The objective is to identify entrepreneurial factors that distinguish TNC from non-TNC suppliers. In total, ten locally owned suppliers were interviewed, five TNC suppliers and four non-TNC suppliers. One local supplier presented a mixed case, acting at times as a TNC supplier and at others as a non-TNC supplier; therefore, it was excluded from the sample. All nine entrepreneurs interviewed were Chinese Malay. The non-TNC suppliers had no previous or current relations with TNCs. They were randomly selected, in order to produce a non-biased sample. 5.6.2.1. Examination of company profiles One of the objectives of the interviews with local entrepreneurs was to gain an understanding of the company history, and the career path, motivation, and family background of each founding entrepreneur. The profiles of the entrepreneurs and their businesses are summarized in Table 5.3. Based on the interviews, the following themes emerged. 5.6.2.1.1. Similarities in period of establishment. With the exception of Supplier J, whose company was passed down to the current entrepreneur from his father, all suppliers were established in the period 1985–1995. As is explained in Section 5.2, Penang experienced a rapid increase in

1985

3

27

93

Junior high school

New

Technician

Encouragement3

Family

Self study

Supplier B

Auto injection parts

1990

1

26

57

Junior high school

Spin-off

Worker

Invitation4

Self financed

Machine Non dependence6

Supplier C

Packaging

1991

5

23

80

Diploma

New

Technician

Discovered business chance

Self financed

Internal7 & self study

Non

Supplier D

Printing

1984

8

32

120

High school

New

Sales

Discovered business chance

Family

Machine & self study

Non

Supplier E

Wire cutting

1994

2

32

80

High school

Spin-off

Technician

Sought higher status

Family

Self study

Non

Supplier F

Metal piping

1992

NA

34

150

High school

Diversification1 Sales

Discovered business chance

Self financed

Suppliers

Non

Supplier G

Screws

1994

6

39

430

Bachelor’s

Spin-off

Sales

Discovered business chance

Partners

Machine dependence

Non

Supplier H

Electronics

1990

6

41

26

Master’s

New

Technician

Dream pursuit

Self financed

Self study

Procurement

Supplier J

Boilers

1975

20

24

80

Bachelor’s

Succession2

Engineer

Obligation5

Family

Self study

Non

Type of establishment

Educational attainment

Provision of support by government

Number of current employees

Engineering

Main sources for tech. upgrading

Age of the entrepreneur

Supplier A

Major sources of finance

At time of establishment Motivation for independent start-up

Industry

Previous career

Company

Number of employees

Company profiles

Year of establishment Non-TNC

TNC suppliers

Table 5.3

Non

155

Note: 1 “Diversification” refers to a strategy by the parent company to establish a new organizational division for the diversification of production. 2 “Succession” literarily means taking over the company from a parent or relatives. 3 “Encouragement,” which often appears as a major motivation for independence, means a decision based on advice from others within the same company. 4 “Invitation” is a similar motivation to “encouragement” but is made by people outside the company, such as a purchasing manager of the buyer firm. 5 “Obligation” refers to a case in which the entrepreneur is expected to succeed or start up a company for reasons involving the family or other obligation. 6 “Machine dependence” is one method of technological upgrading, relying on embedded technologies embodied in newly purchased machines. 7 “Internal” implies a technological learning method mainly pursued by independent efforts, including reference books, professional journals, internet searches, in-house labs, and R&D. Source: Interviews by author.

156 To Be or Not to Be a Supplier to TNCs in Malaysia?

FDI inflow in the late 1980s. The surge in TNC purchasing undoubtedly created opportunities for local entrepreneurs to start businesses. In particular, it appears that the years around 1990 were a critical period where conditions were favorable for the formation of TNC supply relationships. According to the interviewees, it was difficult for local suppliers before this period because TNCs were able to meet their cost objectives with imported inputs. After the mid-1990s it became difficult to start up a company as a supplier because TNCs began awarding larger contracts to a smaller number of local suppliers. 5.6.2.1.2. The importance of working experience. Founders’ higher educational attainment did not seem to be a requirement for becoming a TNC supplier. Rather, lower educational levels, for example, high school or junior high school, were common among the founders of TNC suppliers. Founders of non-TNC suppliers tended to have more education. In the interviews, most of the entrepreneurs expressed the opinion that formal education was of little practical use. Instead, they emphasized the importance of their experiences working in large companies as well as formal and on-the-job training. For instance, Supplier A produces highly complicated automated machines for use by the affiliates of US-based semiconductor firms in Penang. The founder acquired his basic knowledge of the machines while working for one of these companies in the mid-1980s. Technological improvements to the equipment were initiated and carried out by the founder himself. The research findings do not necessarily reject the importance of formal education in enhancing general human capital; however, for entrepreneurs serving TNCs, acquiring practical knowledge and technological know-how through direct work experience was critically important. 5.6.2.1.3. Self-finance. There was a general tendency for self or family financing at the start-up phase at both TNC and non-TNC suppliers. Since most of the entrepreneurs came from poor families, they had limited funds to cover start-up costs, and most started on a small scale. In my interviews, the “Bumiputra policy,” where Chinese Malay entrepreneurs often receive less government support in the areas of financial assistance and public contracts, was often blamed for this. However, as some entrepreneurs admitted, the difficult initial financial conditions were useful because they encouraged strict financial discipline that proved helpful in later stages of the company’s development.

Hiroshi Oikawa

157

5.6.2.1.4. Attitudes toward technology learning. Technological upgrading is an essential element for developing a competitive edge; therefore, TNC and non-TNC suppliers equally expressed a keen interest in improving their technological know-how. However, the means of learning new technologies differed considerably between the two groups. TNC suppliers tended to put more importance on internal resources, such as in-house R&D facilities and training programs. On the other hand, non-TNC suppliers tended to rely on external sources, such as information from machinery and material suppliers and technical consultants. 5.6.2.2. EO analysis The entrepreneurial history, strategies, decision-making practices, and practical behaviors collected in each entrepreneur’s interview were analyzed using the EO framework. The results of this evaluation are presented in Table 5.4 and Figure 5.6. While the assessments were based

Industry

Technology learning

Risk-taking

Autonomy

Average

Supplier A

Engineering

2

5

4

3

3.5

Supplier B

Auto injection parts

3

4

4

3

3.5

Supplier C

Packaging

4

5

3

2

3.5

Supplier D

Printing

5

2

4

3

3.5

Supplier E

Wire cutting

3.5

Average

Non-TNC

Elements of entrepreneurial orientation (EO)

Company

Market TNC suppliers

Table 5.4 Entrepreneurial orientation score by company

2

5

5

2

3.2

4.2

4

2.6

Supplier F

Metal piping

5

1

3

5

3.5

Supplier G

Screws

5

2

4

3

3.5

Supplier H

Electronics

3

4

2

5

3.5

Supplier J

Boilers

4

4

3

3

3.5

4.25

2.75

3

4

Average Source: Assessed by author.

158 To Be or Not to Be a Supplier to TNCs in Malaysia?

Market

Non-TNC suppliers

Autonomy

4.5 4 3.5 3 2.5 2 1.5 1 0.5 0

Tech-learn

TNC suppliers

Risk-taking Figure 5.6

Average EO scores by supplier group

Source: Same as Table 5.4.

on my interpretations of the entrepreneurs’ words and stories, I tried to evaluate them as objectively as possible. The following example illustrates how scores for each of the EO elements were assigned from interview transcripts. Example: The founder of Supplier C explained his start-up motivation as follows: I was working for polystyrene packaging company in Singapore after I graduated from a vocational school in Kuala Lumpur. I worked very hard because I liked this job. I went without sleep in order to make my work perfect. Probably due to my diligence, my performance was outstanding and I was promoted to a line manager during my twoyear career at that company. I had no intention of starting my own company; but my boss encouraged me to start operations in Penang unexpectedly, because there was growing demand for our products there and the location was too far from Singapore. The benefit to my boss was payment of dividends if my company was successful. Risk was minimal, because for the first five years my orders were facilitated by that company. I decided to take on this challenge because I was

Hiroshi Oikawa

159

confident that by doing quality work I would surely be able to satisfy our customers’ needs and expectations. (a) Market orientation: 2 or 3 (for start-up, he did not need to be seriously concerned with marketing issues) (b) Technology-learning orientation: 5 (he attached great importance to technology learning) (c) Risk and uncertainty tolerance: 2 or 3 (he took some risk but it was mitigated by receiving orders from his boss in the context of an expanding market) (d) Autonomy orientation: 1 or 2 (he was not motivated to start his own company until he was encouraged by his boss and was dependent on his prior employer for orders) Of course, a single anecdote does not provide enough information for an assessment of all the entrepreneurial characteristics. For a more comprehensive evaluation, numerous interview anecdotes were taken into consideration. An individual’s EO was measured in the four dimensions (a)–(d) discussed in Section 5.3. Mean scores for TNC and non-TNC groups were compared. For appropriate comparison, the average EO score for individual entrepreneurs was standardized to 3.5. The results are presented in Table 5.4 and Figure 5.6. In the following section, each EO element is examined for each supplier group, and the implications for supplier development are discussed. 5.6.2.2.1. Market orientation. Market orientation measures the extent to which an entrepreneur attaches importance to finding market opportunities. Entrepreneurs at non-TNC suppliers exhibited higher scores for this orientation than TNC suppliers, although the difference was not large. The mean score for the non-TNC supplier group was 4, and 3.4 for the TNC group. I found that market orientation was closely related to start-up motivation. For most non-TNC suppliers, the decision to start a business was motivated by the desire to capture opportunity and profit. Such suppliers tended to give greater importance to discovering new market opportunities than investing in technological upgrading. The machines used by non-TNC suppliers were often second hand, implying that their competitiveness did not rely on having the newest, most advanced technology. A typical example is presented in the case of Supplier F. The founder had no engineering experience in aluminum manufacturing; however,

160 To Be or Not to Be a Supplier to TNCs in Malaysia?

he started this business because “there used to be no aluminum suppliers in this area. Domestic industries, as well as local TNCs, depended on the import of high-priced aluminum materials.” This comment indicates a strong market orientation. To make up for his lack of technical know-how, he hired technicians, including expatriates who had retired from local TNCs. Interestingly, as the entrepreneur himself admitted, technological upgrading was not seriously pursued because the market did not require it. Entrepreneurial alertness in detecting new opportunities as the market environment changed was the key to the company’s success. Expansion of personal business networks was a constant goal of this entrepreneur. Supplier G had a similar story. In contrast, as explained in the following section, TNC suppliers tended to place more importance on technological improvement than marketing and market-seeking activities. 5.6.2.2.2. Technology-learning orientation. Technology-learning orientation characterizes the extent to which an entrepreneur emphasizes technological learning as a means to achieve business success. There is often, but not always, a trade-off between a technology-learning and market orientation. Techno-entrepreneurs perceive their competitive advantage as lying principally in their technological superiority, either in process or product technology. They believe that excellent products create demand, and therefore, technology learning should trump marketing efforts. In general, TNC suppliers showed a stronger technology-learning orientation than non-TNC suppliers. This could be because TNC affiliates often require higher-tech components, and only those local firms with high standards of technological and managerial skills were accepted as qualified suppliers. Interestingly, with the sole exception of Supplier D, TNC supplier entrepreneurs told similar stories about how they became TNC suppliers. When working in other companies prior to starting their own, they enthusiastically devoted themselves to the technical aspects of the operation. They showed excellent aptitude as engineers. Thanks to their skills, they were rapidly promoted within the companies and were later encouraged by other people to start their own businesses. Sometimes managers of local TNCs encouraged and gave partial support for the start-up (e.g., Supplier E). In this case, the entrepreneurs did not need to worry about finding customers for their products and, therefore, a strong market orientation was not necessarily required initially. Non-TNC suppliers were, on average, less oriented toward technology learning. As a matter of course, they understood the importance of

Hiroshi Oikawa

161

technology as an element of their company’s competitive advantage. However, as illustrated by the case of Supplier F, they placed less emphasis on technology learning. Most of these entrepreneurs did not conduct internal R&D on the grounds that it was not profitable to do so. In fact, surprisingly, one shop-floor manager for a non-TNC supplier did not even know the term “R&D.” When needed, they called on outside technical support to deal with equipment problems, and viewed this as a good strategy for both managerial and financial reasons. 5.6.2.2.3. Risk-taking and uncertainty tolerance. Risk-taking and tolerance of uncertainty are characterized by the extent to which entrepreneurs take on technological, managerial, and financial risk. It includes the opportunity costs taken on by an entrepreneur who sacrifices his or her salary from previous employment. There was a little difference in the scores in this measure between TNC and non-TNC suppliers, with TNC suppliers being slightly greater risk takers. Although the score differences were negligible, some important qualitative differences were detected. There are two reasons for the slightly higher risk-taking scores for TNC suppliers. First, TNC suppliers were willing to tackle new technological challenges even though they were uncertain of the outcome. They took a more proactive stance to trying new and unfamiliar technologies. Second, because business contracts with TNCs often call for high volumes, substantial investment was required to expand production capacity. Suppliers were often asked to invest in specific assets such as narrow purpose machinery that was not useful for manufacturing other products. All of these factors exposed TNC suppliers to higher investment risk. In contrast, non-TNC suppliers tended to follow the opposite strategy. In fact, they sometimes chose not to become a TNC supplier because TNCs often request new investments and require substantial technological upgrading. The high risk associated with supplying TNCs was not attractive because technological efforts were perceived as difficult to manage and not always profitable. Instead of taking on such risks, nonTNC suppliers tended to seek stable domestic markets which could be served with the technology they had on hand. 5.6.2.2.4. Autonomy orientation. Autonomy orientation characterizes an entrepreneur’s desire to run his or her company independently. The research results show a substantial difference in this orientation between TNC and non-TNC suppliers. As described earlier, some TNC suppliers had not initially intended to start their own companies. TNC customers

162 To Be or Not to Be a Supplier to TNCs in Malaysia?

provided the initial contract and most subsequent orders. Because most TNC suppliers are technology-learning-oriented, they are motivated by the potential for technological learning that comes from interacting with the technologies found rather on the TNC’s shop floor. Because they began with “captive” style GVC linkages with a technologically dominant customer, business independence was clearly not a central motivation for these entrepreneurs. This may be a reason for the lower mean average score of TNC suppliers in regard to autonomy orientation. Non-TNC suppliers, on the other hand, tended to put more importance on independence. This is partly because business autonomy is necessary to respond quickly to market opportunities. For these entrepreneurs, running their own companies was often a life-long dream, and the domestic market was the first accessible target for them. A typical case was the entrepreneur, a founder of Supplier H, who obtained a Master’s degree from a UK university in electronics, and worked as a senior technical manager for the Penang affiliate of a US semiconductor firm. Because his life-long dream was to become the “Asian Edison,” he resigned his position and started his own small company. The company manufactured several types of electronics products under its own brand name. The main market was Malaysia where the Malaysian government eventually became its largest customer.

5.7. Conclusions and discussion It is commonly argued that FDI represents a relatively fast track to economic development for countries with few own technological resources (Lall 2002). In an effort to catch up technologically, and to accelerate capital formation, many developing countries have passed legislation permitting foreign firms to establish local affiliates and even provide incentives to attract international investment. In 1970, the export processing zones (EPZ) were introduced on Penang Island for this reason. The simplistic view is that open export processing zones, where local and foreign firms are encouraged to co-locate, facilitate local sourcing. This approach has been successful in some countries, and Malaysia is often referred to as an example (Lall 1995; Rasiah 1995). Still, the origins, character, and quality of these linkages have not been explored in depth, and systematic comparisons of TNC suppliers and non-TNC suppliers have not been made. This study examines the GVC linkages between TNC’s affiliate and locally owned suppliers, and the entrepreneurial orientation of a small number of TNC and non-TNC suppliers. The study shows that both

Hiroshi Oikawa

163

TNCs’ strategies and local suppliers’ entrepreneurial characteristics help to structure linkage formation. The findings confirm that TNCs’ strategies, such as expanding local purchasing, significantly affect the demand side of linkage formation, while the characteristics of local entrepreneurs, such as having strong technology-learning orientations or risk aversion, affect the supply side of linkage formation. The EO approach adopted for this study was useful in distinguishing the entrepreneurial characteristics of TNC and non-TNC suppliers. The TNC suppliers in the sample were generally technology-oriented and considered technology to be central to their competitive advantage, while non-TNC suppliers tended to put more importance on seeking out new market opportunities. The TNC suppliers in the sample tended to take greater risks than non-TNC suppliers in regard to asset-specific and/or large-scale investments, and both TNC and non-TNC suppliers were exposed to market risk. TNC suppliers seek less autonomy than non-TNC suppliers, and are indifferent to stay in subservient, “captive” relationships with TNCs. The findings presented here have significant implications for GVC analysis. Traditionally, GVC analysis has tended to emphasize the role of “lead firms” in determining the positioning of supplier firms. In contrast, this study examines the interaction between local entrepreneurs and TNCs in creating and developing industrial linkages. While the research bears this out, forming a TNC supplier is not the only path to economic development. For entrepreneurs with a relatively low autonomy orientation and a high technological learning orientation, developing a supply relationship with a local TNC affiliate can be a powerful path toward growth and upgrading. But for entrepreneurs with the opposite orientation, pursuing opportunities in the local market may be a more profitable path. On the other hand, if the goal is to accelerate economic development through technological learning, TNC suppliers clearly have the edge over non-TNC suppliers, and the EO of TNC suppliers in regard to technology orientation reflects this. There is no way, given the research methodology, to understand causation for the observed patterns of EO with certainty. Entrepreneurs whose company formed a captive GVC relationship with Company X may have a low autonomy orientation because it would have served no purpose to attempt to do otherwise, and so on. The small sample size is another drawback of this study, with only one TNC, five of its suppliers, and four non-TNCs as a comparison. Additional research with larger samples and more robust methods will be needed to probe more deeply

164 To Be or Not to Be a Supplier to TNCs in Malaysia?

into the causation of a founder’s EO, develop the arguments further, and render the conclusions more generalizable.

Notes 1. In Figure 5.1, Malaysia has suffered from a severe, steady decline in net FDI inflows this decade. There are several reasons for this: (i) after the financial crisis, outward investment from Malaysia has been steadily increasing, (ii) neighboring countries, such as Vietnam and Thailand, which have a larger population and younger workforce are more attractive to potential foreign investors, (iii) foreign investment is discouraged by Malaysia’s highly regulated financial system, as illustrated by the tying of the Malaysian Ringgit to the US dollar (although this policy was abandoned in 2005), and (iv) investors are concerned with potential political instability following the reign of Premier Mahathir. These are fundamental obstacles to the Malaysian economy attracting more FDI in the long term. 2. Our research does not imply that entrepreneurship of non-TNC suppliers is less productive, less useful, or less important. On the contrary, the contribution of non-TNC suppliers to economic development is equally important, particularly in domestic service-oriented industries. However, as indicated in the previous section, it is important to nurture more TNC suppliers to retain more value-added within the economy, particularly for countries such as Malaysia, which are heavily dependent on TNCs. 3. As can be seen in Table 5.3, the sourcing structure fluctuated significantly over a short, three-year period, particularly with respect to local and Singaporean suppliers. This is partly explained by production changeovers that necessitated adjustments of the outsourcing structure.

References Altenburg, Tilman. 2000. TNC-SME Linkages for Development: Issues-ExperiencesBest Practices. Geneva: UNCTAD, United Nations. Baumol, William J. 1990. “Entrepreneurship: Productive, Unproductive, and Destructive.” Journal of Political Economy 98(5): 893–921. Best, Michael. 1999. “Cluster Dynamics in Theory and Practice: Singapore/Johor and Penang Electronics.” Judge Institute of Management Studies, Research Paper in Management Studies 9. Best, Michael and Rajah Rasiah. 2003. Malaysian Electronics: At the Crossroad. Small and Medium Enterprise Branch. Vienna: UNIDO. Covin Jeffrey, G. and Dennis P. Slevin. 1989. “Strategic Management of Small Firms in Hostile and Benign Environments.” Strategic Management Journal 10(1): 75–87. Covin Jeffrey, G. and Dennis P. Slevin. 1991. “A Conceptual Model of Entrepreneurship as Firm Behavior.” Entrepreneurship: Theory and Practice 16(1): 7–24. Driffield, Nigel and Abd HalimMohod Noor. 1999. “Foreign Direct Investment and Local Input Linkages in Malaysia.” Transnational Corporation 8(3): 1–24.

Hiroshi Oikawa

165

Eiteman, David K. 1997. “Multinational Firms and the Development of Penang, Malaysia.” International Trade Journal 11(2): 169–185. Gereffi, Gary. 1999. “International Trade and Industrial Upgrading in the Apparel Commodity Chain.” Journal of International Economics 48(1): 37–70. Gereffi, Gary and Raphael Kaplinsky. eds 2001. The Value of Value Chains: Spreading the Gains from Globalisation. IDS Bulletin 32(3). Gereffi, Gary and Miguel Korzeniewicz, eds 1994. Commodity Chains and Global Capitalism. Westport, CT: Praeger. Gereffi, Gary, Humphrey, John and Timothy Sturgeon. 2005. “The Governance of Global Value Chains.” Review of International Political Economy 12(1): 78–104. Giroud, Axele. 2003. Transnational Corporations, Technology and Economic Development: Backward Linkages and Knowledge Transfer in South-East Asia. Cheltenham, Edward Elgar. Henderson, Jeffrey. 1998. “Danger and Opportunity in the Asia-Pacific.” In Economic Dynamism in the Asia-Pacific, ed. G. Thompson. London: Routledge: 356–384. Hobday, Mike. 2001. “The Electronics Industries of the Asia-Pacific: Exploiting International Production Networks for Economic Development.” Asia-Pacific Economic Literature 15(1): 13–29. Humphrey, John and Hubert. Schmitz. 2002. “How Does Insertion in Global Value Chains Affect Upgrading Industrial Clusters?” Regional Studies 36(9): 1017–1027. JETRO. 2008. Business Conditions of Japanese Companies in ASEAN and India. (in Japanese). JETRO. Jomo, K. Sundaram, Rasiah, Rajah, Alavi, Rokiah, and Gopal, Jaya. 2003. “Industrial Policy and the Emergence of Internationally Competitive Manufacturing Firms in Malaysia.” In Manufacturing Competitiveness in Asia: How Internationally Competitive National Firms and Industries Developed in East Asia, ed. K. S. Jomo. London and New York: RoutledgeCurzon: 106–172. Kilby, Peter. ed. 1971. Entrepreneurship and Economic Development. New York and London: Free Press. Kirzner, Israel M. 1973. Competition and Entrepreneurship. Chicago: University of Chicago Press. Kirzner, Israel M. 1997. “Entrepreneurial Discovery and the Competitive Market Process: An Austrian Approach.” Journal of Economic Literature 35(1): 60–85. Knight, Frank H. 1921. Risk, Uncertainty and Profit. Boston: Houghton Mifflin. Lall, Sanjaya. 1995. “Malaysia: Industrial Success and the Role of the Government.” Journal of International Development 7(5): 759–773. Lall, Sanjaya. 2002. “Linking FDI and Technology Development for Capacity Building and Strategic Competitiveness.” Transnational Corporations 11(3): 39–88. Leff, Nathaniel H. 1979. “Entrepreneurship and Economic Development.” Journal of Economic Literature 17(1): 46–64. Leibenstein, Harvey. 1968. “Entrepreneurship and Development.” American Economic Review 58(2): 72–83. Lumpkin, G. Tom and Gregory G. Dess. 1996. “Clarifying the Entrepreneurial Orientation Construct and Linking it to Performance.” Academy of Management Review 21(1): 135–172. McClelland, David C. 1961. The Achieving Society. New York: Free Press.

166 To Be or Not to Be a Supplier to TNCs in Malaysia? McClelland, David C. 1987. “Characteristics of Successful Entrepreneurs.” Journal of Creative Behavior 21(3): 219–233. Miller, Danny and Peter H. Friesen. 1982. “Innovation in Conservative and Entrepreneurial Firms: Two Models of Strategic Momentum.” Strategic Management Journal 3(1): 1–25. Nazari Ismail, Mohd. 1995. Transnational Corporations and Economic Development: The Case of the Malaysian Electronics Industry. Kuala Lumpur: University of Malaya Press. Oikawa, Hiroshi. 2011. “Inter-Country Value Distribution in East Asian Electronics and Automobile Industries: An Empirical Global Value Chain Approach.” Chapter 6 in this volume. Rasiah, Rajah. 1995. Foreign Capital and Industrialization in Malaysia. London: Macmillan. Rasiah, Rajah. 2002 “Systemic Coordination and the Development of Human Capital: Knowledge Flows in Malaysia’s TNC-driven Electronics Clusters.” Transnational Corporation 11(3): 89–129. Rasiah, Rajah. 2005. “Fostering Clusters in the Malaysian Electronics Industry.” LAEBA 2005 Second Annual Meeting, Buenos Aires, Argentina. 28–29 November 2005. Schmitz, Hubert. ed. 2004. Local Enterprise in Global Economy: Issues of Governance and Upgrading. Cheltenham, Edward Elgar. Schumpeter, Joseph A. 1934. The Theory of Economic Development. Cambridge: Harvard University Press. Sieh, Lee Mei Ling. 2006. “FDI Inflows and Economic Development: The Postcrisis Experience of Malaysia.” In Multinationals and Economic Growth in East Asia: Foreign Direct Investment, Corporate Strategies and National Economic Development, ed. S. Urata, C. S. Yue and F. Kimura. London and New York: Routledge: 287–318. Sturgeon, Timothy. 2007. “From Commodity Chains to Value Chains: Interdisciplinary Theory Building in an Age of Globalization.” In Frontiers of Commodity Chain Research, ed. J. Bair. Stanford, California: Stanford University Press. Sturgeon, Timothy J. and Richard 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, ed. S. Yusuf, A. Altaf and K. Nabeshima. Oxford: Oxford University Press: 35–87. UNCTAD. 1999. World Investment Report 1999: Challenge for Development. UNCTAD. 2001. World Investment Report 2001: Promoting Linkages. Warr, Peter G. 1989. “Export Processing Zones: The Economics of Enclave Manufacturing.” World Bank Research Observer 4(1): 65–88. Yazicioglu, Yaz. 2003. “Big Bang in Penang.” Electronics Business November, 70–84. Yin, Robert. 1994. Case Study Research: Design and Methods. 2nd ed. Thousand Oaks, CA, London: Sage.

6 Inter-Country Value Distribution in the East Asian Electronics and Automobile Industries: An Empirical Global Value Chain Approach Hiroshi Oikawa

6.1. Introduction Over the past several decades regional production networks have steadily expanded and deepened in East Asia, growing alongside the trans-continental business networks that have been the focus of much of the global value chain (GVC) literature (e.g., Gereffi 1999; Schmitz 2004). Local firms in East Asia have been drawn into regional production systems in addition to, and in some cases because of, the growth of more extensive GVCs. In this view, regional production systems are nested within, and fully compatible with, the rise of trans-continental GVCs (Sturgeon et al. 2008). Indeed, Petri (2006) clearly shows the interdependence of the region’s economies growing to historic heights. Other studies suggest that East Asia is more deeply integrated through trade and investment networks than other developing areas (Ng and Yeats 1999; Kimura and Ando 2003; Lall et al. 2004). There are several notable features in the structure and processes driving East Asian economic integration. First, transnational corporations (TNCs) and global buyers have played a central role in organizing and coordinating GVCs and have thus been important forces in the formation of the region’s networks (Feenstra and Hamilton 2006; Urata et al. 2006). Second, in comparison with other developing regions, East Asia has achieved outstanding economic results. Firms and industries 167

168 Inter-Country Value Distribution in East Asia

in the region have reaped enormous benefits from regional and global economic integration (Borrus et al. 2000). Obviously, these two features are related. The region’s production networks provide important opportunities for learning, capability building, marketing, and finance for local firms and individual workers. Dynamic relationships with demanding TNCs and global buyers have been essential for local firms and workers seeking to acquire, refine, and upgrade their technological and managerial capabilities (Ernst and Kim 2002; Kawakami, Fujita, Sato, and Oikawa, this volume). The region’s trade and investment networks are among the key factors explaining the success of East Asia’s industrialization (Whittaker et al. 2010). However, several important questions remain unanswered in regard to the impact of regional integration on East Asia’s economic development. These are principally related to distribution and causes of gains from cross-border networks. For instance, do the region’s production networks provide the same benefits for all participants at country, firm, and individual levels, or do some economies, firms, and individuals gain in spite of, or even at the expense of others? Are all East Asian economies embedded into the networks in the same ways? What roles have US- and Japan-based TNCs played in East Asia’s production networks? These questions are underexplored in the existing empirical literature, which has largely examined regional trade and investment flows without unveiling the critical role of TNCs in driving foreign direct investment (FDI) (Sohn 2002; Kawai 2005; Kimura and Ando 2005). These studies provide evidence that the East Asian economy has been deeply integrated, but shed little light on the sources, features, and economic consequences of this integration. Global value chain (GVC) analysis provides a useful framework for examining these questions. It focuses on the drivers, features, and distribution of gains from international production networks (Gereffi and Korzeniewicz 1994; Henderson 1998; Gereffi 1999; Kaplinsky 2000; Gereffi and Kaplinsky 2001), and on the institutional characteristics of firms, industries, and countries (Borrus 2000; Sturgeon 2007). GVC analysis also focuses on the nature of cross-border transactions, arguing that the role of a nation’s industry within GVCs helps to structure economic development processes and learning opportunities, in which the powerful role of large TNCs and global buyers has been emphasized in creating and organizing the international division of labor. Regional production networks in the electronics and automobile industries are my principal focus for several reasons. First, these

Hiroshi Oikawa

169

industries have been particularly important for export-oriented industrialization in East Asia. They have generated an immense amount of employment and wage growth in the region, and provided opportunities for outsourcing and for the technological upgrading of local suppliers (Sturgeon and Lester 2004). Second, these industries have extensive and dense production networks within the region. Global trade in automotive, and especially electronics intermediates (a useful proxy for GVCs in goods producing industries), dominate trade in all other manufactured intermediates by a wide margin, and the share of electronics in total intermediate goods trade has been growing rapidly since 1990. A rising share of this trade involves countries in East Asia (Sturgeon and Memedovic, forthcoming). Third, in large measure, the development of these industries has been fueled by a massive inflow of FDI. Because of differing FDI policies set by various East Asia’s governments and the different technical requirements associated with the different industries, the character of production networks varies widely in individual countries as well as between the two industries. The automotive and electronics industries, therefore, are ideal laboratories for examining the character and implications of regional production networks in East Asia. A core concept used in this study is value-added. The overall welfare of a country can be measured plainly with this variable. Specifically, value is added along various steps in the R&D, production, and distribution chain and then distributed in the form of wages, profits, taxes, and other rewards. In this study, I use newly available regional input– output tables to estimate how much value-added is created and retained at home and how much is moved abroad. East Asian international input–output tables for 1990, 1995, and 2000 are used to examine interindustry flows of value added, and the empirical results are carefully vetted with my fieldwork in several Southeast Asian countries as well as in Japan. The study is structured as follows. Section 6.2 provides a brief theoretical review, with a focus on GVC analysis. Section 6.3 examines the drivers of development in the electronics and automobile industries in East Asia, mainly by focusing on variation in FDI and trade policies of governments in the region. Section 6.4 explains the input– output method and applies it to an estimation of cross-country value distribution. Section 6.5 offers the results and their implications. A summary of the findings and a conclusion are presented in the final section.

170 Inter-Country Value Distribution in East Asia

6.2. Global value chains and uneven development Questions related to the relationship between globalization and the international distribution of income have stimulated a great deal of intellectual concern (e.g., Kaplinsky 2005). In this section, I summarize two theoretical frameworks that can be used to examine this issue. The first is mainstream international trade theory. This literature tends to emphasize the gains from trade, demonstrating that the international division of labor benefits all participants. Each country should specialize in the kind of production in which it has a comparative advantage so as to increase real national income through trade. The theory shows excellent proof of the existence of the gains from trade; however, it does not provide a clear explanation about the allocation of gains among trading partners. Second, GVC analysis can be used to analyze the question of allocation of gains from trade. The concept of GVCs highlights the full range of activities related to the production and consumption of a specific product or service: from product conception and design, to material and component production, the final stages of production, sales and marketing, and finally after-sales support, disposal, and recycling. The initiative to form GVCs is often undertaken by TNCs and/or large “global buyers” (Gereffi 1999); therefore, the focus of the analysis is centered on these “lead firms” in the chain, firms which coordinate and control the organization of their foreign affiliates, if they have them, and/or, to a greater or lesser some degree, their suppliers. By deciding who produces what, where, and how, lead firms can structure the upgrading potential of the entire chain. GVC analysis stresses that “the value-added capacities of companies at each nodal point in the chain tend to vary and in any case are constrained by the way the chain is organized and by the nature of corporate power within it” (Henderson 1998: 369). In this way, the GVC concept provides a mechanism for upgrading and an institutional context that includes asymmetric power structures within the chain, both useful concepts for explaining uneven distribution of the gains from trade. The GVC approach holds the view that uneven gains from trade and investment are closely associated with the spread and penetration of TNCs’ global activities. In other words, inter-country economic polarization is sometimes exacerbated by the expansion of the TNCs’ global operations. This critical perspective is explained as follows. A country that possesses weak technological capabilities and skills and little fixed capital is forced to offer falling real wages if it hopes to attract

Hiroshi Oikawa

171

TNCs by providing an attractive base for simple manufacturing (Fröbel et al. 1980; Kaplinsky 2005). Profit is usually greatest in the nodes of the chain (upstream and downstream) that are located in developed countries, while less profitable segments are located in less developed countries (Schmitz 2004; Sturgeon and Lester 2004; Song 2007). Moreover, dependence on TNCs often results in outflows of value-added in terms of payments for imported materials, license fees, royalties, and the repatriation of profits (Sturgeon and Linden, this volume). Thus, the GVC approach allows that the economic consequences of TNC-led production networks could result in uneven development among network participants. It is important to note that the views of the gains from trade derived from mainstream trade economics and GVC analysis are not necessarily in conflict. Instead, the difference lies in emphasis. The former highlights the aggregate gains from trade, while the latter pays greater attention to the distributive aspects of these gains. In this regard, as the electronics and automobile industries in East Asia increased their share of world production and trade, the region gained substantially in absolute terms. Then, this study aims to examine the distribution of value in the region. The goal is to identify which economy gained from trade in these industries the most and which the least.

6.3. Electronics and automobile value chains in East Asia Again, the East Asian electronics and automobile industries were chosen as the primary foci of this study. There are several reasons for this. First, the electronics and automobile industries have been major forces behind rapid economic development in the region. While these industries have been two of the fastest growing industries in world production and exports, this is particularly true in East Asia, where export and output growth in these industries has far exceeded the world average, allowing the region to steadily gain market share over the last few decades (Sturgeon and Memedovic, forthcoming). Second, with this expansion in output and exports, the electronics and automotive industries have made a remarkable contribution to employment creation, the generation of value-added, and technological advancement (Lall et al. 2004). Third, the rapid development in output and exports in these industries has long been boosted by massive inflows of FDI from developed nations. An important outcome of FDI-led industrialization has been to create closer and more complex cross-border production linkages than traditional arm’s length trade

172 Inter-Country Value Distribution in East Asia

linkages. Huge FDI inflows not only support TNCs’ local operations but also potentially provide local firms in the surrounding area with opportunities to participate in international production networks. Finally, for technical reasons, electronics and automobile production are more easily fragmented than other industries. These industries cover a broad range of products that include high-tech products, such as semiconductors, flat panel displays, and engines, as well as fairly traditional mass-produced commodities, such as tires, keyboards, batteries, steel, and simple metal and plastic parts. Thus, in comparison with industries that produce products that are more monolithic (on one hand) and customized (on the other), these industries create demand for a large number of intermediate goods, broadening the opportunities for local and international outsourcing and subcontracting. The focus on the electronics and automobile industries is therefore ideal for capturing both a wide range of value chain characteristics, depending on the nature of the input or stage of production, analyzing sectoral variation, and revealing the elaborate structure of the region’s production networks. Despite their similarities, such as dependence on TNCs and aggressive support of host governments for indigenous input sectors, there are important differences as well. First, the electronics industry is more export-oriented than the automobile industry. During the 1990s, East Asian electronics exports grew by 13.3% per annum while auto exports grew by only 4.7%. Second, production of electronics products is more easily fragmented than automobile production. These differences are rooted in the nature of products and processes and reflected in the different industrial organization of each industry. Electronics industry value chains are relatively open, operate on a much shorter-term basis, are more decentralized, and tolerate supplier switching, while automobile production networks are relatively slow, centralized, stable, and oriented toward the long term contract (Lall et al. 2004; Sturgeon et al 2008). In addition to the industry comparison, the study aims to clarify the characteristics of the region’s production networks through a comparison of each country’s performance and of government policies, which differ widely in regard to the promotion of local support industries and the encouragement of FDI in general and the attraction TNC affiliates in particular (Lall 1995). Korea and Taiwan have pursued similar industrial policies to promote indigenous firms and deepen technological upgrading with minimum dependence on FDI. In these countries, FDI is assigned a

Hiroshi Oikawa

173

secondary role in obtaining advanced technologies. These strategies are fairly consistent with the other industrial policies adopted by these economies. The domestic market was not fully opened to free trade to reserve it for national champion companies. The harmful effects of protection were partly offset by incentives and pressures on local firms to export in competitive international markets. Korean and Taiwanese firms sought product assembly joint ventures with US, Japanese, and European companies. As these joint venture firms acquired more knowledge and production know-how in the area of assembly, they gradually took on more functions and eventually become full service, independent contract manufacturers, purchasing components and producing according to the specifications of foreign brands, a sourcing arrangement known in the industry as original equipment manufacture (OEM) (Hobday 1995). As local contract manufacturers aggressively accumulated process and component purchasing capabilities, they began to offer product design services as well, in a sourcing arrangement known as original design manufacture (ODM). Clearly, ODM contracting requires more comprehensive technological capabilities, including systems design, component selection, and post-production logistics. Hobday (2001) calls this catch-up mechanism an OEM/ODM system, and Korea and Taiwan are cited as successes in this regard. In stark contrast to Korea and Taiwan, several ASEAN countries, especially Malaysia, Thailand, Vietnam, and Singapore, have largely accepted the dominance of TNC investments. The government’s primary role has been to provide infrastructure and incentives for TNC affiliates, including a relatively stable macroeconomic environment, tax holidays, the establishment of free trade zones and, commonly, high tariff protection to encourage domestic production. These factors succeeded in attracting TNC investment, particularly in the field of electronics. This industrial strategy allowed large TNCs to enter local markets with their own production and suppliers, and avoided relying heavily on the development of indigenous firms as was the case in Korea and Taiwan. Hobday (2001) calls this development strategy TNC-led industrialization. TNCs’ dominance of ASEAN’s core industrial sectors is partly explained by the relative absence of competitive local firms. ASEAN economies are still at a relatively early stage of their development. When a country does not have the indigenous technological resources to deepen its industrial capabilities, a TNC-led industrial strategy can lead

174 Inter-Country Value Distribution in East Asia

to better outcomes than policies that keep TNCs out to make more space for indigenous firms. The Malaysian electronics and Thai automobile industries have certainly followed this path. This discussion raises important questions. Have East Asian economies enjoyed equal benefits from these different industrialization paths, or have different industrial strategies resulted in different economic results? Have indigenous capabilities arisen in Korea and Taiwan, where TNCs have a relatively small presence? Or, has TNC participation successfully enhanced, or at least compensated for, the relatively weak local technological capabilities of ASEAN economies? Is there any evidence that TNCs have crowded out local entrepreneurship and taken benefits away from these local economies? The next section explains the quantitative method used to approach these questions.

6.4. Methodology This section explains the methodologies used to gauge the relative performance of countries in regional production networks. Positive performance is measured by a rising share of value-added in the regional value chain of the industry under examination. This is defined in relative terms; therefore, even a “poor” performing country may capture a greater level of value-added over time in absolute terms. Clearly, value is an important concept in GVC studies. In this study, value can be defined as the sum of value-added, including wages, profits, and natural resource rents. Although economic success can also be detected in growth of output, employment, or exports, the generation of value-added is a fundamental variable in determining a country’s overall well-being. Value is normally generated in a profit-seeking activity that is conducted by an individual corporation for which a number of intermediate inputs are used. Such intermediates are often products or services supplied by other firms. Therefore, value-added, or value in general, is created within a unit of a system of industrial linkages where the costs of inputs are deducted from the value of outputs to arrive at value added. Such systems can encompass both domestically produced and imported inputs. As a result, imported intermediate inputs show up as international transfers of value-added among different industries and countries. Obviously, these international inter-firm relationships change over time. Some firms and industries in a country are successful in

Hiroshi Oikawa

175

increasing their share of value-added while others are not. Thus, I measure economic performance in GVCs through the use of an international value distribution (IVD) index. This index indicates how much value is retained within an economy, flows out to other economies and regions, or disappears as tariffs or transportation costs. The following sections explain the formula for quantification. 6.4.1. Model To quantify the region’s value distribution, I use a simple input–output method based on the Asian International Input–Output tables for 1990, 1995, and 2000. The immediate advantage of the “value-added” definition is the applicability of input–output methods for mapping the flows of value-added among industries. Input–output analysis sees an economy as a system of inter-linkages between all inputs and outputs. An important feature of the tables is that all the countries listed are treated as non-competitive and endogenous, following Matsumura and Fujikawa (1998), Oikawa (2004), and Fujikawa et al. (2005). Suppose that the total numbers of sectors and countries are n and s, respectively. Then, the following supply–demand balance equation holds in the matrix form. X = AX + F

(6.1)

where   X = x11 , . . . , x1n ; x21 , . . . x2n ; . . . ; xs1 , . . . xsn   F = f11 , . . . , fn1 ; f12 , . . . fn2 ; . . . . . . ; f1s , . . . fns ⎡ ⎢ ⎢ ⎢ A ≡⎢ ⎢ ⎣

where

⎡ ⎢ ⎢ ⎢ A ≡⎢ ⎢ ⎣ hk

A11 A21 .. . As1

ahk 11 ahk 21 .. . ahk n1

A12 .. . .. . ···

··· .. . .. . ···

A1s .. . .. . Ass

ahk 12 .. . .. . ...

... .. . .. . ...

ahk 1n .. . .. . ahk nn

⎤ ⎥ ⎥ ⎥ ⎥ ⎥ ⎦

⎤ ⎥ ⎥ ⎥ ⎥ ⎥ ⎦

176 Inter-Country Value Distribution in East Asia

In equation (6.1) xsn and fns denotes the vectors of gross output and final demand of sth country’s nth sector. A prime ( ) denotes the transpose. Ahk is the coefficient matrix of hth country’s product used for kth country’s hk unit output. ahk matrix, represents the value-termed ij , an element of A coefficient of the hth country’s ith sector used for the unit output of  hk the jth sector in kth country aij ≥ 0 . A is a large square matrix of (ns × ns). Two points should be noted for the subsequent argument. The first is Leontief’s inverse matrix [I − A]−1 , representing how much incremental gross output is induced directly and indirectly by a unit increase in final demand. The word “indirectly” refers to the recursive increase in output due to sectoral interdependence. Hence, both direct and indirect impacts are always considered. Second, for simplicity’s sake, the final demand vector is omitted hereafter; increases in final demand are assumed to be evenly set to unity across all the relevant sectors. By pre-multiplying this with the diagonal matrix of the value-added ratio, G, the international distribution of value among s countries is obtained as follows. D = G [I − A]−1

(6.2)

where ⎡ ⎢ ⎢ ⎢ G≡⎢ ⎢ ⎣

g1 .. . .. . 0

...

...

g2 .. . ...

... .. . ...

r1k

0

0 .. . 0

r2k .. . ...

... .. . .. . ...

0 .. . .. . gs

⎤ ⎥ ⎥ ⎥ ⎥ ⎥ ⎦

where ⎡ ⎢ ⎢ ⎢ g ≡⎢ ⎢ ⎣ k

0 0 .. . rnk

⎤ ⎥ ⎥ ⎥ ⎥ ⎥ ⎦

I is the identity matrix. rji denotes the value-added ratio of ith country’s jth sector. D appears as the (ns × ns) matrix, specifically expressed in matrix form as follows.

Hiroshi Oikawa

⎡ ⎢ ⎢ ⎢ D=⎢ ⎢ ⎣

D11 D21 .. . Ds1

D12 .. . .. . ...

... Dhk .. . ...

D1s .. . .. . Dss

... .. . .. . ...

hk v1n .. . .. . hk vnn

177

⎤ ⎥ ⎥ ⎥ ⎥ ⎥ ⎦

where ⎡ ⎢ ⎢ ⎢ D ≡⎢ ⎢ ⎣ hk

hk v11 hk v21 .. . hk vn1

hk v12 .. . .. . ...

⎤ ⎥ ⎥ ⎥ ⎥ ⎥ ⎦

In the above matrices, vijhk represents the share of the value-added gained by hth country’s ith sector, which was originally generated by production activities conducted in the jth sector of kth country. In other words, the unit production of kth country’s jth sector outflows the amount of vijhk value-added to hth country’s ith sector. Imports from Hong Kong and the rest of the world (ROW) are exogenous in the Asian International Input–Output tables. Therefore, for such a technical reason, it is not possible to trace the outflows of valueadded to these areas. However, this is not a crucial shortcoming because the approach is still useful for comparing country performance in the regional context. Moreover, as discussed below, because of its importance in East Asian GVCs, the United States is included in the analysis. Nevertheless, due to this technical constraint, the leakage of import costs to these areas needs to be treated as technically equivalent to valueadded. The direct and indirect cost leakages to Hong Kong and ROW are calculated by using the following formula. M h = ah [I − A]−1

(6.3)

where h = HKG and ROW In equation (6.3), aHKG and aROW are (1 × ns) row vectors whose elements are import coefficients from Hong Kong and ROW. Note again that this is not value, but payments for imported materials from Hong Kong and ROW, respectively. The matrix M h appears as a (1 × ns) row vector.

178 Inter-Country Value Distribution in East Asia

The principal focus of this study is not to obtain the absolute scale of value-added multipliers, but to examine the relative distribution of value-added in sectors across the economies concerned. Thus, a standardized scale is required. For this, the column sum of D, M HKG and M ROW has been set to equal unity. After this manipulation, the modified elements of the D, M HKG and M ROW matrices are expressed with the hat (∧ ). Finally, the variables concerned are transformed into an aggregated index. With the notations used above, the IVD is defined as follows. Vjk→h ≡

n 

vˆ ijhk

(6.4)

i=1

The IVD index refers to the total share of value-added retained domestically (i.e., k → k) as well as of value leaked out to other economies h (i.e., k → h). The index clearly shows how much valueadded leaves the country from the industry in question and how much other countries in the region gain from this leakage. 6.4.2. Determining relative country performance in regional production networks Mainstream trade theory posits that the international division of labor will benefit all participants. Whether or not this is true is not a central concern of this chapter. Instead, this study evaluates the gains in relative terms—which sector in which country gains the most value-added and which gains the least? The implication of the word “relative” may be important in the context of East Asian development, as East Asia as a whole has enjoyed larger benefits from international trade than other developing regions in terms of export and production expansion. In the 1990s, for example, East Asia exports in electronics and automobiles far outpaced those of the world average (Lall et al. 2004). The question this study seeks to answer is how these benefits were distributed among all East Asian economies. A performance matrix is proposed for this evaluation because an assessment using only a single index would fail to grasp the complex nature of an industry’s performance. For example, an increase of a country’s domestic IVD could result from the introduction of an autarkical trade policy. However, this type of policy tends to undermine international competitiveness by detaching local industry from international technological frontiers. Conversely, export expansion can be attained with the help of TNCs; however, this tends to increase dependence on imported inputs and, as a result, allows larger outflows of value-added

Hiroshi Oikawa

179

Table 6.1 Performance matrix of East Asian industrialization

IVD Index

World market shares Increased

Decreased

Increased

Absolute winners

Substantial winners

Decreased

Questionable winners

Absolute losers

Source: By the author.

from the country. To examine cases between these two extreme examples, I use a matrix evaluation which combines global export shares with the IVD index. The four possible types of industrial performance that result are presented in Table 6.1. First, “Absolute Winners” are defined as a national industry that increases not only its share of world exports but also its percentage share in the IVD index. This suggests that the country’s industry captures more value-added from regional production networks than the industry in other countries in the region. The second category, “Substantial Winners,” refers to an industry whose share of world exports decreases but whose position on the IVD index rises. This is an interesting case in which an industry loses in the world market yet may enjoy more gains due to less value leakages to other countries. The third category, “Questionable Winners,” is the opposite case of “Substantial Winners” just mentioned above, where increasing exports are combined with a falling share of IVD. The last category, “Absolute Losers,” has a reduced share of the world’s exports and also more value-added is leaked out to other economies. 6.4.3. Sources of data and notes Two sets of data are the principal sources of information in this study. The first is the Asian International Input–Output Table in 1990, 1995, and 2000 (the latest currently available), which is published by the Institute of Developing Economies in Japan. Seventy-eight sectors are available in the 1990 and 1995 tables and 76 sectors in the 2000 tables. To simplify the analysis, I combined related sectors to reduce the sectors to 26 in 1990 and 1995, and to 25 in 2000. Table 6.2 (see at the end of this chapter) explains the sector classification, as modified for this study. This chapter focuses on two sectors, electronics and autos. The second data set used in the study is the UN Comtrade database, which is used to calculate the world market share for the electronics and automobile

180 Inter-Country Value Distribution in East Asia

industries for individual countries in East Asia. The industrial classification of trade we use is, in most cases, the two-digit level of SITC Rev.2. This study focuses on the ten major economies in the AsiaPacific region, that is, China, Indonesia, Japan, Korea, Malaysia, the Philippines, Singapore, Taiwan, Thailand, and the United States. These economies appear on the Asian International Input–Output tables as endogenous, while the imports from Hong Kong and the ROW are treated as, in I-O terminology, non-competitive and exogenous. This study explicitly includes the United States, as multiple studies show that its economic influence on East Asian production networks is very important (e.g., Borrus et al. 2000). So, in the following analysis, the term “East Asia” includes the United States, unless otherwise noted. Again, the electronics and automobile industries are the central focus of this study; therefore, the following analysis concentrates on the country-level performance of these two industries. Tables 6.3, 6.4 and 6.5 (see at the end of this chapter) set out the main results. Table 6.3 presents the IVD index by industry and country. The tables show the share of the industries’ value-added that is retained by each country as well as the value lost to other economies. For instance, in 2000, China retained 85.16% of the total value-added generated by its local automobile industry, while the rest shifted to other economies or disappeared as freight and insurance costs or tariffs. In this year, Japan was the largest beneficiary. It captured 4.65% of the value-added that the Chinese automobile sector created. Note that the use of the word “local” or “domestic” industry does not denote nationality but territory, so it does not necessarily mean locally owned. It therefore often includes foreign-owned companies operating within the territory of the country concerned. Table 6.4 and 6.5 show the global export market share of East Asia’s electronics and automobile industries, respectively, from 1990 to 2000.

6.5. The results 6.5.1. Overall Results The results can be summarized using the four categories proposed in the previous section. (a) Absolute Winners (gaining larger world export market share and IVD) Electronics industries: Indonesia Automobile industries: Indonesia, Malaysia, Taiwan, Thailand

Hiroshi Oikawa

181

(b) Substantial Winners (losing world market share but gaining higher IVD) None (c) Questionable Winners (gaining world export market share with decrease in IVD) Electronics industries: China, Korea, Malaysia, Taiwan, Philippines, Thailand (d) Absolute Losers (losing world export market share and decrease in IVD) Electronics industries: Japan Some countries and industries show ambiguous tendencies either in the IVD index and/or export market shares (e.g., some industries did not change measurably in either IVD and/or export market shares during the two periods). These cases do not neatly fit into the categories and are therefore omitted from the following examination. The following sub-section of the paper examines the three cases, excluding case (b) which does not have any industries fitted in. 6.5.2. Absolute winners Absolute winner countries have gained in world export market share and captured a larger share of value-added in regional production networks. The gains in value added are composed of two elements: (1) in absolute terms, more value-added is generated by rising exports, and (2) in relative terms, a rising share of value-added is retained within the economy. Note that the inter-country value-added distribution is solely determined by the supply-side, namely, the industrial input– output structure. Because of this, we can say that more value-added is captured in the local supply base, relative to suppliers located outside the country. The IVD index can detect where the sector’s value-added is vested and distributed. In other words, the index is irrelevant to demand elements; it traces where the value-added is generated (e.g., locally or overseas) only. From the viewpoint of IVD, the question is the extent to which value-added is retained within the local economy. Economies that demonstrate a growing share of domestic IVD are those that became less dependent on overseas suppliers. What we cannot say is if this is explained by the growing participation of a more competent and competitive local supply base, the arrival of more foreign suppliers, or the

182 Inter-Country Value Distribution in East Asia

internalization of additional production within the vertical structure of TNC lead firms affiliates. What we can say is that more input value has been captured by the economy concerned.

6.5.3. Questionable winners This category is interesting from the view point of GVC upgrading theory. A country that gains an increasing share of the world export market is normally evaluated positively as a successful case of export-oriented industrialization. This may lead to an increase in the country’s welfare in terms of employment, tax revenue, and other benefits. However, it may not signal deep technological learning, entrepreneurship, or the creation of sustainable indigenous supply base. In fact, these latter aspects of development might be in decline, even as exports grow. Questionable winner countries may get trapped in low value segments of the chain as higher value-added activities leak to other countries. My estimation clearly identifies the industry/country pairs that have this type of cross-country input–output trend. As shown in Tables 6.3 and 6.4, the electronics industries in China, Korea, Malaysia, Taiwan, the Philippines, and Thailand can be considered questionable winners. The typical case is China’s electronics industry, whose share in the world export market increased rapidly from 1.02% in 1990 to 5.03% in 2000, while the domestic IVD of China’s household electrical equipment sector steadily decreased from 94.07% in 1990 to 85.09% in 2000. Other economies mentioned here show similar trends. Moreover, Malaysia, the Philippines, and Thailand have lost a considerable percentage of value-added within their local economies and allowed larger shares of value-added to outflow overseas. An extreme case is Thailand, which retained only less than 20% of the total value-added that its semiconductor sector created in 2000. The largest external beneficiary was Japan, implying an increase of intermediates imports from this country. A similar pattern exists for other ASEAN economies. Such economies also show a relatively lower percentage in the retention of value-added produced locally. Such value outflows have increased over time. Therefore, they are certainly questionable winners in the region’s production networks. These results stand in contrast to studies that characterize emergence of the electronics industry in ASEAN countries as an unmitigated success (Borrus et al. 2000; Hobday 2001). The present study shows that successful industrial development per se cannot be equated with output, employment, exports, or simple value-added. As was examined in

Hiroshi Oikawa

183

Section 6.1 and 6.2, the TNC-dependent industrial strategy of many of these countries may partly explain this outcome because it led to heavy dependence on external sources of technology and supply, as embodied in advanced components. However, it may also reflect the migration of large segments of the local industry from one country to another, as was the case with Taiwan’s PC industry in the early 2000s, as discussed in Chapter 1 of this volume. 6.5.4. Losers In this study, only the Japanese electronics industry appears to be a loser. Tables 6.3 and 6.4 clearly show that Japan’s shares of world electronics exports and domestic IVD have declined substantially. Such significant declines on the part of Japan can be partially explained by the higher growth rates of the industry in other economies in the region, but a more fundamental explanation may lie within the Japanese economy itself. The Japanese economy experienced an extremely difficult period after the Bubble burst in 1989, and the subsequent recession lasted throughout the 1990s. Companies could not afford additional domestic investments in production, R&D, human resources, and so on, and this resulted in a loss of international competitiveness in many industries. The result suggests that during this period, for the Japanese electronics industry, increasing hollowing out and dependence on imported inputs resulted in an outflow of value-added outside of the country. The 1990s are often called “Japan’s lost decade.” The result of this study clearly confirms Japan’s declining presence as a final assembler in the regional economy during the 1990s. 6.5.5. Other findings 6.5.5.1. Re-examination of the role of Japan and the United States in East Asia’s production networks? There has been a qualitative change in the role of Japan in East Asia’s production networks. The results of this study suggest that, in terms of exports, output, and value-creation, Japan’s presence in the East Asian electronics and automobile industries has clearly declined. However, it should also be noted that this does not necessarily mean that Japan’s role in the region’s production networks has become less important. Instead, its contribution has become more indirect, as Japanese production networks have shifted from a self-contained to an outwardoriented structure. As stated earlier, the export market share of Japan’s electronics and automobile products has steadily declined; at the same

184 Inter-Country Value Distribution in East Asia

time, the domestic IVD of these industries has also decreased. However, most East Asian economies, particularly the ASEAN countries, have become more dependent on intermediate imports from Japan. An extreme example is, as mentioned above, the Thai semiconductor industry, which retained less than 20% of the total value-added that the sector created in 2000. Japan gained 21.13% from this industry, more than all of the value added in Thailand. A similar tendency is reported in other ASEAN countries, particularly in the electronics sector. Japan’s high share of value-added in intermediate goods reveals its’ central role of supplying high value core components (both as direct and indirect inputs) to East Asia’s regional production networks (Akinwande et al. 2005). At the same time, the United States is becoming increasingly important in some East Asian economies. This is particularly the case in the semiconductor industries of China and Malaysia. During the 1990s, the United States’s IVD indices in the semiconductor industry increased from 1.45% to 5.84% for China and from 7.70% to 14.26% for Malaysia. In general, other sectors also show a trend of increasing US contributions to value-added. This implies that the United States has become an important supply base for East Asian semiconductor industries. And, it reveals the important role the US lead firms play in East Asia’s production networks. When American firms outsource production to East Asian contract manufacturers, they often specify the use of US produced components. 6.5.5.2. Differences between the electronics and automobile production networks? Table 6.3 clearly shows, in terms of value-added distribution, a general tendency for the automobile industry to be more nationally based than the electronics industry. For example, for most East Asian economies, the automobile sector enjoys retention of more than 80% of total value-added within the producing economies. Malaysia, Korea, Taiwan, the Philippines, and Thailand have lower percentages than this figure, but their auto sectors do have generally larger shares of local value added than their electronics sectors. The higher value of domestic IVD in automobile industry probably stems from higher transport costs and local content regulations (Sturgeon et al. 2008). This finding is consistent with the presumption in Section 6.3 that the automobile value chain is less fragmented than those in the electronics industry.

Hiroshi Oikawa

185

This point leads our attention to another consideration related to my sweeping classification of several East Asian economies as “questionable winners” in electronics GVCs. The outflows of value-added found in this study may be partly explained by this technological factor, that is, the technological characteristics of electronics and the standardized specifications and portability of components, which allows production to be more easily fragmented than the automobile sector. Therefore, the relatively large outflow of value in electronics sectors may be partly attributed to such a technical feature of these products in general.

6.6. Summary and discussion This study has examined the distribution of value in the regional production networks of East Asia. The view that East Asia’s economies, in general, have achieved successful export-oriented industrialization tends to overlook the unevenness of value allocation among individual economies in the region. In other words, we need to ask , “Which economy gains the most and which the least?” To answer this question, an international input–output method was employed to trace the creation and flow of value-added among ten Asia-Pacific economies, including the United States. Performance was evaluated by using both valueadded acquisition and market share in global exports. The findings are summarized as follows. First, the electronics industry of Indonesia and the automobile industries of Indonesia, Malaysia, Taiwan, and Thailand have been successful in reaping large benefits from the region’s production networks in terms of capturing more value-added and greater world export market share. Second, the electronics industries of six East Asian economies (China, Korea, Malaysia, Taiwan, Philippines, and Thailand) were questionable winners, in that their exports have expanded steadily without proportional gains in the IVD index. Third, the Japanese electronics industry shrank dramatically during the 1990s, allowing other East Asian economies to increase their shares of the global export market and regional value-added. As a general conclusion, the following points should be noted. First, this study suggests that the TNC-led industrialization path commonly pursued by ASEAN governments have been less than fully successful; hence their positions in this study appeared as “questionable winners.” These economies have enjoyed a steady expansion of world

186 Inter-Country Value Distribution in East Asia

manufacturing export share but a falling share of domestic value-added within regional production networks. This less than ideal outcome may be due to the industrialization policies of these economies, which have generally allowed TNCs to decide where to source value-added. Second, in comparison with the performance of ASEAN economies, the contract manufacturing and branded product strategy pursued by Korea and Taiwan has been relatively successful in terms of the ratio of value-added retained within their economies. Their relative success is understandable as their industrialization path is much less dependent on FDI than other developing countries in East Asia. While this strategy may seem to be a desirable one, it cannot be replicated easily in any latecomer economy. The success of this approach critically depends on the transfer of advanced capabilities through arms length sourcing relationships, and in heavy investment in key component technologies at the right time. Another consideration is the impact of the Asian financial crisis. Indonesia, Korea, Malaysia, and Thailand were the most seriously affected by the crisis. Afterwards, these economies raised their market share of the world’s exports both in electronics and automobiles, with some thanks due to currency devaluations. However, there was no significant change in the IVD index. While the shares of regional valueadded in the electronics sector fell for Korea and Malaysia after the crisis, those of Indonesia and Thailand appear to have remained relatively stable, or even increased. The automobile sectors are also relatively stable in the IVD index. Thus, the overall results are mixed. Some countries incurred losses while others did not. Finally, it is important to mention the role of the US economy in East Asia’s production networks. The economic resurgence of the United States during the late 1990s, particularly after the 1997 economic crisis, is confirmed by the study. This may seem somewhat surprising, since it is generally believed that large US companies have widely outsourced their peripheral activities to external agents, including contract manufacturers operating in East Asia. This should have provided wider opportunities for value diffusion over East Asia, but the results show the opposite to be the case. This implies that the largest value-added nodes remain within the United States while the less profitable segments have been outsourced externally (see Linden et al. 2007). The study suggests that the United States maintains its position that increasingly reaps the benefits from East Asia’s production networks.

Table 6.2

Sector classification and integration: Asian international input–output table 1990, 1995, and 2000 Sector classification 2000

25 sectors integrated Description

Agriculture, livestock, forestry, and fishery

Mining and quarrying

Original 76 sectors Code

Description

001 002 003

004

Sector classification 1990 and 1995 26 sectors integrated

Original 78 sectors Code

Description

Paddy

001

Paddy

Other grain

007A

Other grain

Food crops

002

Cassava

004

Sugarcane and beets

005

Oil palms and coconuts

007B

Other food crops

003

Natural rubber

006

Fiber crops

008

Other commercial crops

Non-food crops

Description

Agriculture, Livestock, Forestry, and Fishery

Livestock and poultry

009

Livestock and poultry

006

Forestry

010

Forestry

007

Fishery

011

Fishery

008

Crude petroleum and natural gas

012

Crude petroleum and natural gas

009

Iron ore

015A

Iron ore

010

Other metallic ore

013

Copper ore

014

Tin ore

015B

Other metallic ore

016

Non-metallic ore and quarrying

011

Non-metallic ore and quarrying

Mining and Quarrying

187

005

188

Table 6.2 (Continued) Sector classification 2000 25 sectors integrated Description

Original 76 sectors Code

Description

012

Milled grain and flour

26 sectors integrated Description

Original 78 sectors Code

Description

018

Milled rice

019

Other milled grains and flour

013

Fish products

021A

Fish products

014

Slaughtering, meat products and dairy products

021B

Slaughtering and meat products

Food, beverage, and tobacco 015

Textile, leather, and the products thereof

Sector classification 1990 and 1995

Food, beverage, and tobacco

Other food products

017

Oils and fats

020

Sugar

021C

Other food products

016

Beverages

022A

Beverage

017

Tobacco

022B

Tobacco

018

Spinning

023

Spinning

019

Weaving and dyeing

024

Weaving and dyeing

020

Knitting

025

Knitting

021

Wearing apparel

026

Wearing apparel

022

Other made-up textile products

027

Other made-up textile products

023

Leather and leather products

028

Leather and leather products

Textile, leather, and the products thereof

Other manufacturing products

024

Timber

029

Timber

025

Wooden furniture

030A

Furniture

026

Other wooden products

030B

Other wooden products

027

Pulp and paper

031

Pulp and paper

028

Printing and publishing

032

Printing and publishing

029

Synthetic resins and fiber

033A

Synthetic resins and fiber

030

Basic industrial chemicals

033B

Other basic industrial chemicals

031

Chemical fertilizers and pesticides

034

Chemical fertilizers and pesticides

032

Drugs and medicines

035A

Drugs and medicine

033

Other chemical products

035B

Other chemical products

034

Refined petroleum and its products

036

Refined petroleum and its products

035

Plastic products

050A

Plastic products

036

Tires and tubes

037

Tires and tubes

037

Other rubber products

038

Other rubber products

038

Cement and cement products

039

Cement and cement products

039

Glass and glass products

040

Glass and glass products

040

Other non-metallic mineral products

041

Other non-metallic mineral products

Other manufacturing products

189

Table 6.2 (Continued)

25 sectors integrated

Original 76 sectors

Sector classification 1990 and 1995 26 sectors integrated

Original 78 sectors

Description

Code

Description

Description

Iron and steel

041

Iron and steel

Iron and steel

042

Non-ferrous metal

042

Non-ferrous metal

Non-ferrous metal

043

Non-ferrous metal

Metal products

043

Metal products

Metal products

044

Metal products

Boilers, engines, and turbines

044

Boilers, engines, and turbines

Engines and turbines

045E

Engines and turbines

General machinery

045

General machinery

Ordinary industrial machinery

045C-2

Ordinary industrial machinery

Metal working machinery

046

Metal working machinery

Specialized industrial machinery

045B-1

Specialized industrial machinery

Ordinary industrial machinery

045C-2

Ordinary industrial machinery

Agricultural machinery

045A

Agricultural machinery

Specialized industrial machinery

045B-2

Specialized industrial machinery

Specialized machinery

047

Specialized machinery

Code

Description Iron and steel

Heavy electrical equipment

048

Heavy electrical equipment

Heavy electrical machinery

045D

Heavy electrical machinery

Television sets, radios, audio and

049

(continued) Communication equipment

Electronics and electronic products

046A

Electronics and electronic products

Electronic computing equipment

050

Electronic computing equipment

Semiconductors and integrated circuits

051

Semiconductors and integrated circuits

Other electronics and electronic products

052

Other electronics and electronic products

190

Sector classification 2000

Household electrical equipment, lighting, fixtures, batteries, etc.

053

Household electrical equipment

Other electrical machinery and appliances

046B

Other electrical machinery and appliances

054

Lighting fixtures, batteries, wiring, and others

Motor vehicles

055

Motorcycles

056

Motor vehicles

Motor vehicles

047A

Motor vehicles

Motorcycles

Motorcycles and bicycles

047B-1

Motorcycles and bicycles

Shipbuilding Other transport equipment

057

Shipbuilding

Shipbuilding

048B

Shipbuilding

058

Other transport equipment

Motorcycles and bicycles

047B-2

Motorcycles and bicycles

Aircrafts

048A

Aircrafts

Other transport equipment

048C

Other transport equipment

Precision machines

059

Precision machines

Precision machines

049

Precision machines

Other manufacturing products

060

Other manufacturing products

Other manufacturing products

050B

Other manufacturing products

191

192

Table 6.2 (Continued) Sector classification 2000 25 sectors integrated Description

Electricity, gas, water supply, construction, trade, transportation, and other public services

Original 76 sectors Code

Description

061

Electricity and gas

062

Water supply

063 064

Sector classification 1990 and 1995 26 sectors integrated Description

Original 78 sectors Code

Description

051

Electricity, gas and water supply

Building construction

052A

Building construction

Other construction

052B

Other construction

065

Wholesale and retail trade

053A

Wholesale and retail trade

066

Transportation

053B

Transportation

067

Telephone and telecommunication

054A

Telephone and telecommunications

068

Finance and insurance

054B

Finance and insurance

069

Real estate

054D-1

Other services

070

Education and research

054C

Education and research

071

Medical and health service

054D-2

Other services

072

Restaurants

054D-3

Other services

073

Hotels

054D-4

Other services

074

Other services

054D-5

Other services

076

Unclassified

056

Unclassified

075

Public administration

055

Public administration

Electricity, gas, water supply, construction, trade, transportation, and other public services

Source: Originally from Asian International Input–Output Tables (IDE 1990, 1995, 2000) and modified by the author.

193 Table 6.3 Index of International Value Distribution China

2000

China

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

67.66%

59.18%

76.41%

77.86%

85.09%

85.16%

0.30%

0.45%

0.36%

0.44%

0.36%

0.23%

Japan

6.18%

7.33%

4.56%

5.27%

3.35%

4.65%

Korea

2.71%

3.15%

2.14%

2.09%

1.39%

1.11%

Malaysia

0.82%

1.20%

0.58%

0.42%

0.24%

0.15%

Taiwan

2.98%

3.87%

2.40%

2.12%

1.46%

1.09%

Philippines

0.36%

0.55%

0.22%

0.13%

0.07%

0.03%

Singapore

0.65%

1.17%

0.46%

0.31%

0.17%

0.10%

Thailand

0.26%

0.71%

0.24%

0.23%

0.15%

0.13%

USA

4.19%

5.84%

2.86%

2.38%

1.73%

1.30%

Freight & Insurance

1.16%

1.63%

1.06%

0.96%

0.47%

0.28%

Imports from HKG

4.61%

5.47%

2.96%

1.97%

0.70%

0.14%

Imports from EU

3.64%

3.12%

2.07%

2.03%

1.11%

2.05%

Imports from ROW

1.89%

3.15%

1.73%

1.94%

2.70%

1.24%

Duties & Taxes

2.60%

3.17%

1.95%

1.86%

1.01%

2.34%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00%

China

1995

Electronics computing equipment

Indonesia

Total

1990

Television & audios∗1

75.05%

91.68%

84.58%

Indonesia

0.11%

0.18%

0.09%

Japan

7.43%

2.60%

3.74%

Korea

1.46%

0.55%

0.61%

Malaysia

0.34%

0.25%

0.12%

Taiwan

0.47%

0.58%

0.47%

Philippines

0.04%

0.04%

0.02%

Singapore

0.47%

0.15%

0.11%

Thailand

0.10%

0.06%

0.06%

USA

1.67%

0.98%

1.34%

Freight & Insurance

0.85%

0.17%

0.31%

Imports from HKG

6.14%

0.96%

0.30%

Imports from ROW

5.08%

1.69%

7.57%

Duties & Taxes

0.78%

0.10%

0.69%

Total

100.00%

100.00%

100.00%

China

79.18%

94.07%

87.19%

Indonesia

0.17%

0.24%

0.16%

Japan

4.94%

1.93%

3.94%

Korea

0.53%

0.09%

0.10%

Malaysia

0.22%

0.25%

0.16%

Taiwan

0.59%

0.34%

0.23%

Philippines

0.02%

0.04%

0.02%

Singapore

0.09%

0.06%

0.05%

Thailand

0.05%

0.07%

0.07%

USA

1.45%

1.08%

1.25%

Freight & Insurance

0.15%

0.05%

0.17%

Imports from HKG

10.11%

1.04%

0.34%

Imports from ROW

0.96%

0.55%

4.84%

Duties & Taxes

1.52%

0.20%

1.50%

100.00%

100.00%

100.00%

Total

194 Table 6.3 (Continued) Indonesia

China

2000

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

1.10%

1.27%

0.44%

0.34%

1.35%

0.66%

74.16%

85.22%

93.77%

73.27%

80.32%

Japan

4.11%

4.84%

1.52%

1.01%

3.61%

7.05%

Korea

1.07%

1.25%

0.42%

0.30%

1.50%

0.48%

Malaysia

0.55%

0.61%

0.25%

0.23%

1.08%

0.33%

Taiwan

0.60%

0.71%

0.23%

0.15%

0.70%

0.39%

Philippines

0.05%

0.06%

0.02%

0.01%

0.05%

0.07%

Singapore

0.63%

0.74%

0.22%

0.14%

0.56%

0.22%

Thailand

0.41%

0.47%

0.17%

0.14%

0.69%

0.34%

USA

2.16%

2.53%

0.81%

0.56%

2.22%

1.30%

Freight & Insurance

0.71%

0.90%

0.16%

0.00%

1.24%

0.33%

Imports from HKG

0.17%

0.21%

0.04%

0.00%

0.27%

0.04%

Imports from EU

3.29%

4.18%

0.75%

0.03%

3.30%

0.98%

Imports from ROW

5.47%

6.00%

9.39%

3.30%

8.50%

3.38%

Duties & Taxes

1.63%

2.06%

0.37%

0.02%

1.67%

4.13%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00%

Total China

0.71%

1.24%

0.66%

64.03%

74.89%

69.54%

Japan

4.30%

3.44%

13.86%

Korea

0.91%

1.33%

0.83%

Malaysia

0.42%

0.34%

0.20%

Taiwan

0.89%

0.86%

0.58%

Philippines

0.05%

0.04%

0.05%

Singapore

1.11%

0.59%

0.36%

Thailand

0.19%

0.28%

0.10%

USA

3.15%

1.89%

1.58%

Freight & Insurance

0.79%

0.76%

0.84%

Imports from HKG

0.17%

0.17%

0.05%

Imports from ROW

20.87%

12.30%

8.53%

2.41%

1.89%

2.83%

100.00%

100.00%

100.00%

Indonesia

1995

Electronics computing equipment

78.05%

Indonesia

Duties & Taxes Total China

1990

Television & audios∗1

0.46%

0.42%

0.32%

Indonesia

63.49%

75.99%

74.34%

Japan

10.51%

5.63%

13.78%

Korea

1.22%

0.96%

0.48%

Malaysia

0.43%

0.39%

0.19%

Taiwan

1.96%

1.25%

0.37%

Philippines

0.08%

0.09%

0.03%

Singapore

1.13%

1.08%

0.25%

Thailand

0.21%

0.25%

0.07%

USA

3.56%

3.91%

0.94%

Freight & Insurance

0.55%

0.63%

0.82%

Imports from HKG

1.48%

0.56%

0.11%

Imports from ROW

12.98%

7.27%

5.10%

1.94%

1.57%

3.20%

100.00%

100.00%

100.00%

Duties & Taxes Total

195

2000

Japan

1995

Electronics computing equipment

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

China

0.88%

0.68%

0.40%

0.58%

0.82%

Indonesia

0.34%

0.35%

0.22%

0.32%

0.42%

0.30%

Japan

89.88%

84.81%

95.78%

91.92%

92.25%

96.09%

Korea

0.89%

1.59%

0.35%

0.59%

0.32%

0.24%

Malaysia

0.42%

0.74%

0.21%

0.32%

0.24%

0.15%

Taiwan

1.10%

1.93%

0.38%

0.76%

0.31%

0.23%

Philippines

0.24%

0.47%

0.11%

0.19%

0.11%

0.08%

Singapore

0.22%

0.49%

0.09%

0.16%

0.07%

0.04%

Thailand

0.17%

0.18%

0.08%

0.15%

0.13%

0.29%

USA

2.91%

4.13%

1.36%

2.25%

1.92%

1.28%

Freight & Insurance

1.10%

2.22%

0.27%

0.85%

0.24%

0.07%

Imports from HKG

0.27%

0.19%

0.04%

0.36%

0.10%

0.02%

Imports from EU

0.45%

0.74%

0.21%

0.42%

0.89%

0.26%

Imports from ROW

0.66%

0.77%

0.38%

0.79%

1.87%

0.34%

Duties & Taxes

0.46%

0.73%

0.13%

0.33%

0.32%

0.10%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00% 0.33%

Total

0.52%

China

0.64%

0.56%

Indonesia

0.14%

0.19%

0.13%

Japan

92.15%

94.11%

97.22%

Korea

1.01%

0.37%

0.19%

Malaysia

0.35%

0.13%

0.08%

Taiwan

0.56%

0.23%

0.14%

Philippines

0.13%

0.04%

0.04%

Singapore

0.38%

0.09%

0.04%

Thailand

0.25%

0.10%

0.06%

USA

2.84%

1.78%

0.93%

Freight & Insurance

0.05%

0.04%

0.02%

Imports from HKG

0.14%

0.03%

0.01%

Imports from ROW

1.16%

2.19%

0.75%

Duties & Taxes

0.21%

0.15%

0.05%

100.00%

100.00%

100.00%

China

0.23%

0.28%

0.21%

Indonesia

0.25%

0.37%

0.29%

Japan

94.74%

93.75%

96.72%

Korea

0.39%

0.21%

0.18%

Malaysia

0.17%

0.14%

0.14%

Taiwan

0.38%

0.20%

0.14%

Philippines

0.08%

0.09%

0.07%

Singapore

0.09%

0.06%

0.03%

Thailand

0.08%

0.07%

0.05%

USA

2.29%

1.72%

1.30%

Freight & Insurance

0.05%

0.04%

0.03%

Imports from HKG

0.04%

0.01%

0.00%

Imports from ROW

1.10%

2.90%

0.79%

Duties & Taxes

0.11%

0.15%

0.04%

100.00%

100.00%

100.00%

Total

1990

Television & audios∗1

Total

196 Table 6.3 (Continued)

2000

Korea

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

1.84%

2.30%

1.47%

1.64%

1.48%

1.31%

Indonesia

0.48%

0.55%

0.30%

0.73%

0.76%

0.75%

Japan

11.95%

12.81%

11.04%

10.36%

7.17%

6.17%

Korea

56.78%

52.80%

53.95%

65.51%

78.76%

83.49%

Malaysia

1.00%

1.37%

1.08%

0.84%

0.45%

0.32%

Taiwan

1.72%

2.21%

2.06%

1.10%

0.48%

0.29%

Philippines

0.49%

0.70%

0.64%

0.32%

0.12%

0.05%

Singapore

0.76%

1.37%

0.95%

0.41%

0.26%

0.14%

Thailand

0.28%

0.54%

0.23%

0.32%

0.14%

0.12%

11.62%

11.49%

14.22%

5.57%

4.07%

3.50%

Freight & Insurance

4.17%

5.54%

6.18%

2.99%

0.72%

0.22%

Imports from HKG

1.71%

1.94%

1.01%

2.25%

0.51%

0.11%

Imports from EU

2.93%

2.65%

4.27%

2.15%

1.49%

1.55%

Imports from ROW

3.69%

3.14%

2.51%

5.17%

3.04%

1.54%

Duties & Taxes

0.60%

0.59%

0.09%

0.64%

0.54%

0.45%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00%

Total

1995

Electronics computing equipment

China

USA

China

1.12%

1.23%

0.96%

Indonesia

0.20%

0.25%

0.18%

Japan

12.04%

5.83%

6.67%

Korea

68.40%

83.85%

83.73%

Malaysia

0.85%

0.31%

0.18%

Taiwan

1.13%

0.36%

0.22%

Philippines

0.22%

0.12%

0.04%

Singapore

1.00%

0.29%

0.12%

Thailand

0.24%

0.10%

0.07%

USA

7.47%

3.40%

2.91%

Freight & Insurance

1.08%

0.37%

0.41%

Imports from HKG

0.77%

0.14%

0.02%

Imports from ROW

3.90%

2.72%

3.50%

Duties & Taxes

1.58%

1.01%

0.98%

100.00%

100.00%

100.00%

China

0.08%

0.05%

0.04%

Indonesia

0.36%

0.45%

0.31%

Japan

19.06%

8.59%

8.00%

Korea

63.66%

82.07%

83.72%

Malaysia

0.72%

0.58%

0.33%

Taiwan

0.82%

0.36%

0.25%

Philippines

0.18%

0.10%

0.05%

Singapore

0.41%

0.15%

0.08%

Thailand

0.13%

0.08%

0.05%

USA

8.78%

4.63%

3.80%

Freight & Insurance

0.44%

0.21%

0.36%

Imports from HKG

1.06%

0.14%

0.03%

Imports from ROW

2.87%

1.89%

1.92%

Duties & Taxes

1.42%

0.72%

1.05%

100.00%

100.00%

100.00%

Total

1990

Television & audios∗1

Total

197 Malaysia

Other electronics products

Household electrical equipment & others

Automobiles

2.12%

2.92%

1.82%

1.88%

2.26%

1.29%

1.31%

0.98%

0.99%

1.21%

0.94%

17.45%

14.10%

17.50%

17.75%

15.23%

18.66%

1.44%

2.57%

3.01%

2.95%

2.78%

1.87%

1.34%

32.78%

27.23%

36.80%

33.78%

50.35%

60.96%

Taiwan

2.82%

4.47%

3.07%

2.94%

1.77%

1.41%

Philippines

1.11%

1.12%

1.37%

1.30%

0.48%

0.15%

Singapore

4.34%

6.62%

4.88%

4.64%

3.28%

1.55%

Thailand

1.06%

4.45%

0.88%

0.90%

0.91%

0.85%

16.96%

16.60%

14.26%

16.17%

8.58%

3.64%

Freight & Insurance

1.10%

1.43%

1.17%

1.20%

0.82%

0.59%

Imports from HKG

1.81%

4.94%

2.03%

2.06%

2.30%

0.42%

Imports from EU

7.37%

4.81%

5.86%

6.98%

5.45%

4.52%

Imports from ROW

6.09%

6.31%

5.78%

5.99%

4.65%

2.38%

Duties & Taxes

1.13%

0.67%

0.65%

0.66%

0.85%

1.15%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00% 0.31%

Malaysia

2000

Semiconductors & ICs

Indonesia Korea

USA

Total China

1.00%

1.02%

Indonesia

0.42%

0.42%

0.24%

13.59%

13.97%

27.67%

Japan Korea

1.87%

1.92%

0.39%

54.03%

45.16%

50.23%

Taiwan

1.79%

1.84%

0.34%

Philippines

0.27%

0.27%

0.07%

Singapore

3.74%

3.86%

0.38%

Thailand

0.79%

0.81%

0.31%

USA

6.64%

6.86%

1.20%

Freight & Insurance

1.73%

2.65%

1.88%

Imports from HKG

1.19%

1.82%

0.07%

Imports from ROW

12.39%

18.61%

7.16%

0.56%

0.80%

9.75%

100.00%

100.00%

100.00%

China

0.61%

0.62%

0.21%

Indonesia

0.51%

0.53%

0.33%

12.16%

12.92%

27.68%

Malaysia

1995

Electronics computing equipment

China Japan

Duties & Taxes Total

Japan Korea

1.02%

1.09%

0.17%

53.60%

50.06%

54.04%

Taiwan

1.63%

1.74%

0.29%

Philippines

0.30%

0.32%

0.03%

Singapore

3.81%

4.08%

0.30%

Thailand

0.40%

0.42%

0.13%

USA

7.70%

8.26%

1.35%

Freight & Insurance

1.04%

1.14%

1.56%

Imports from HKG

0.80%

0.88%

0.03%

Imports from ROW

15.86%

17.38%

7.66%

Duties & Taxes

0.56%

0.57%

6.22%

100.00%

100.00%

100.00%

Malaysia

1990

Television & audios∗1

Total

198 Table 6.3 (Continued) Taiwan

2000

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

2.09%

2.05%

1.29%

1.95%

2.82%

Indonesia

0.63%

0.76%

0.60%

0.90%

1.09%

0.48%

Japan

16.96%

20.81%

13.92%

15.80%

10.97%

10.65%

Korea

4.05%

5.14%

3.06%

3.15%

1.30%

0.92%

Malaysia

1.56%

2.02%

1.97%

1.10%

0.61%

0.28%

48.69%

42.89%

52.66%

55.87%

64.56%

73.35%

Philippines

0.92%

1.53%

1.42%

0.59%

0.15%

0.11%

Singapore

1.63%

2.15%

1.63%

0.79%

0.33%

0.15%

Thailand

0.46%

1.12%

0.44%

0.46%

0.40%

0.19%

USA

8.81%

10.84%

11.13%

6.69%

3.88%

2.66%

Freight & Insurance

0.42%

0.34%

0.35%

0.42%

0.30%

0.20%

Imports from HKG

1.61%

1.74%

1.07%

1.57%

0.97%

0.15%

Imports from EU

5.24%

3.78%

4.77%

3.48%

2.32%

3.95%

Imports from ROW

5.95%

4.53%

5.51%

6.86%

9.61%

3.02%

Duties & Taxes

0.96%

0.30%

0.16%

0.37%

0.69%

2.87%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00%

Total

1.03%

China

1.15%

1.88%

0.99%

Indonesia

0.25%

0.29%

0.20%

Japan

15.51%

8.74%

9.42%

Korea

2.51%

0.71%

0.65%

Malaysia

1.65%

0.25%

0.19%

53.05%

57.56%

72.88%

Philippines

0.40%

0.08%

0.05%

Singapore

1.59%

0.30%

0.18%

Thailand

0.53%

0.16%

0.10%

USA

6.98%

3.46%

4.08%

Freight & Insurance

2.25%

1.04%

0.68%

Imports from HKG

1.97%

0.58%

0.09%

Imports from ROW

9.87%

23.38%

5.48%

Duties & Taxes

2.28%

1.56%

5.02%

100.00%

100.00%

100.00%

China

0.10%

0.08%

0.05%

Indonesia

0.35%

0.53%

0.26%

Japan

19.69%

12.56%

8.87%

Korea

1.14%

0.57%

0.61%

Taiwan 1995

Electronics computing equipment

China

Taiwan

Total

Malaysia

0.89%

0.36%

0.21%

54.79%

71.49%

72.52%

Philippines

0.30%

0.22%

0.07%

Singapore

0.81%

0.22%

0.10%

Thailand

0.22%

0.12%

0.06%

USA

10.74%

5.90%

5.43%

Freight & Insurance

0.58%

0.29%

0.29%

Imports from HKG

1.78%

0.61%

0.12%

Imports from ROW

7.01%

5.72%

6.81%

Duties & Taxes

1.60%

1.34%

4.60%

100.00%

100.00%

100.00%

Taiwan 1990

Television & audios∗1

Total

199 Philippines

2000

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

2.07%

1.73%

0.75%

1.25%

1.91%

Indonesia

1.51%

0.72%

0.29%

0.55%

2.12%

2.56%

Japan

16.57%

27.60%

18.69%

22.01%

9.41%

10.74%

Korea

2.85%

1.73%

3.63%

2.24%

2.76%

2.76%

Malaysia

3.83%

1.88%

0.97%

1.40%

1.64%

1.20%

Taiwan

1.37%

4.70%

1.71%

3.28%

2.27%

1.99%

38.84%

36.71%

33.86%

40.95%

56.71%

59.84%

Singapore

4.67%

4.44%

1.89%

3.17%

1.11%

0.92%

Thailand

0.46%

3.29%

0.48%

2.04%

0.78%

0.85%

USA

7.73%

7.75%

16.04%

9.96%

4.26%

3.08%

Freight & Insurance

2.73%

0.30%

2.82%

1.13%

2.54%

1.82%

Imports from HKG

6.22%

3.34%

3.12%

2.94%

1.39%

0.84%

Imports from EU

4.76%

2.43%

11.17%

5.20%

2.85%

1.71%

Imports from ROW

2.39%

2.20%

2.60%

2.56%

7.66%

7.25%

Duties & Taxes

3.99%

1.17%

1.99%

1.34%

2.59%

2.41%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00% 0.90%

Total

1995

Electronics computing equipment

China

Philippines

2.03%

China

0.49%

0.74%

Indonesia

0.19%

0.26%

0.16%

Japan

11.05%

6.32%

29.09%

Korea

1.84%

3.11%

1.99%

Malaysia

0.92%

0.51%

0.54%

Taiwan

1.10%

1.73%

0.74%

48.99%

63.13%

46.91%

Singapore

2.40%

1.16%

0.19%

Thailand

0.30%

0.20%

0.54%

USA

12.13%

7.24%

1.30%

Freight & Insurance

1.99%

3.54%

2.21%

Imports from HKG

5.00%

1.61%

0.09%

Imports from ROW

3.25%

4.58%

2.67%

10.34%

5.87%

12.68%

Philippines

Duties & Taxes Total

1990

Television & audios∗1

100.00%

100.00%

100.00%

China

0.18%

0.23%

0.14%

Indonesia

0.19%

0.26%

0.19%

Japan

20.71%

8.59%

23.87%

Korea

1.11%

2.94%

2.00%

Malaysia

0.69%

0.65%

0.34%

Taiwan

1.40%

2.13%

0.58%

48.53%

65.18%

62.45%

Singapore

2.43%

0.76%

0.16%

Thailand

0.20%

0.12%

0.18%

12.20%

8.44%

1.55%

Freight & Insurance

1.29%

1.16%

2.18%

Imports from HKG

2.63%

0.82%

0.04%

Imports from ROW

2.05%

4.43%

1.60%

Duties & Taxes

6.40%

4.29%

4.73%

100.00%

100.00%

100.00%

Philippines

USA

Total

200 Table 6.3 (Continued)

2000

Singapore

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

3.55%

3.19%

2.19%

2.63%

2.77%

Indonesia

0.74%

0.88%

0.96%

0.60%

1.26%

0.98%

Japan

18.19%

15.90%

20.18%

13.59%

12.17%

12.02%

Korea

2.32%

1.99%

2.38%

1.96%

1.50%

1.12%

Malaysia

4.45%

6.01%

3.80%

4.20%

3.61%

2.84%

Taiwan

1.97%

2.04%

2.10%

1.48%

1.34%

0.70%

Philippines

0.28%

0.36%

0.28%

0.21%

0.17%

0.11%

Singapore

47.18%

39.14%

39.61%

52.18%

51.20%

52.06%

Thailand

0.90%

1.87%

0.71%

0.73%

0.97%

0.63%

USA

9.89%

10.04%

13.19%

7.60%

8.32%

5.57%

Freight & Insurance

1.85%

2.70%

4.23%

1.46%

1.85%

2.56%

Imports from HKG

0.66%

0.69%

1.16%

0.51%

0.69%

1.41%

Imports from EU

2.09%

2.33%

4.09%

2.21%

4.13%

6.49%

Imports from ROW

5.93%

12.85%

5.12%

10.65%

10.02%

11.22%

Total

1995

Electronics computing equipment

China

Duties & Taxes

2.25%

0.00%

0.00%

0.01%

0.00%

0.00%

0.03%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00% 1.10%

China

1.37%

1.70%

Indonesia

0.82%

1.10%

1.21%

Japan

17.30%

11.25%

14.04%

Korea

3.75%

1.82%

0.71%

Malaysia

3.97%

4.09%

1.38%

Taiwan

1.80%

1.34%

0.75%

Philippines

0.67%

0.29%

0.10%

Singapore

48.32%

55.34%

53.34%

Thailand

2.80%

0.95%

0.67%

USA

7.94%

4.75%

5.08%

Freight & Insurance

3.36%

2.05%

2.06%

Imports from HKG

1.46%

1.67%

0.54%

Imports from ROW

6.44%

13.65%

19.00%

Duties & Taxes

0.00%

0.00%

0.01%

100.00%

100.00%

100.00%

China

0.72%

1.24%

0.76%

Indonesia

0.53%

1.18%

1.22%

Japan

24.69%

18.69%

23.52%

Korea

2.05%

1.16%

0.38%

Malaysia

3.87%

2.36%

2.17%

Taiwan

2.22%

1.54%

0.68%

Philippines

0.65%

0.20%

0.07%

39.37%

52.32%

46.59%

Total

1990

Television & audios∗1

Singapore Thailand

1.05%

0.47%

0.75%

14.89%

9.03%

8.57%

Freight & Insurance

1.33%

1.43%

2.05%

Imports from HKG

2.39%

0.71%

0.30%

Imports from ROW

6.22%

9.65%

12.89%

USA

Duties & Taxes Total

0.01%

0.01%

0.04%

100.00%

100.00%

100.00%

201

2000

Thailand

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

4.00%

2.89%

4.00%

2.89%

3.50%

Indonesia

1.32%

0.92%

1.32%

0.92%

1.10%

0.67%

Japan

21.13%

10.32%

21.13%

10.32%

15.61%

21.72%

Korea

3.97%

1.75%

3.97%

1.75%

1.09%

0.94%

Malaysia

2.83%

1.92%

2.83%

1.92%

1.70%

0.78%

Taiwan

1.90%

2.68%

1.90%

2.68%

1.92%

0.95%

Philippines

0.73%

0.66%

0.73%

0.66%

0.65%

0.62%

Singapore

2.48%

2.82%

2.48%

2.82%

0.81%

0.39%

Thailand

19.81%

53.75%

19.81%

53.75%

46.61%

51.10%

USA

1.27%

14.25%

8.67%

14.25%

8.67%

4.99%

2.99%

Freight & Insurance

8.90%

5.91%

8.90%

5.91%

2.85%

1.86%

Imports from HKG

1.95%

1.03%

1.95%

1.03%

0.84%

0.17%

11.10%

2.83%

11.10%

2.83%

5.91%

5.37%

Imports from ROW

3.66%

2.32%

3.66%

2.32%

9.40%

2.68%

Duties & Taxes

1.97%

1.52%

1.97%

1.52%

3.03%

8.50%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00% 1.01%

Total

1995

Electronics computing equipment

China

Imports from EU

China

1.10%

1.33%

Indonesia

0.22%

0.45%

0.17%

Japan

15.08%

9.65%

17.84%

Korea

1.63%

0.98%

1.00%

Malaysia

2.67%

0.86%

0.41%

Taiwan

2.42%

1.67%

0.73%

Philippines

1.01%

0.39%

0.21%

Singapore

3.74%

0.99%

0.55%

Thailand

45.62%

55.01%

52.98%

USA

9.65%

4.05%

2.12%

Freight & Insurance

2.70%

1.59%

1.66%

Imports from HKG

1.17%

0.42%

0.14%

Imports from ROW

8.00%

17.03%

9.23%

Duties & Taxes

4.98%

5.55%

11.97%

Total

1990

Television & audios∗1

100.00%

100.00%

100.00%

China

0.54%

1.04%

1.03%

Indonesia

0.23%

0.76%

0.26%

Japan

19.83%

11.30%

22.02%

Korea

1.46%

0.63%

0.58%

Malaysia

1.33%

0.92%

0.69%

Taiwan

1.98%

1.56%

0.68%

Philippines

0.23%

0.15%

0.10%

Singapore

4.04%

0.83%

0.47%

Thailand

37.78%

52.27%

49.41%

USA

16.97%

4.18%

2.60%

Freight & Insurance

1.30%

1.28%

1.51%

Imports from HKG

0.94%

0.31%

0.13%

Imports from ROW

6.37%

17.66%

8.34%

Duties & Taxes

6.99%

7.12%

12.17%

100.00%

100.00%

100.00%

Total

202 Table 6.3 (Continued)

2000

USA

Semiconductors & ICs

Other electronics products

Household electrical equipment & others

Automobiles

0.68%

1.18%

0.21%

0.64%

0.60%

0.62%

Indonesia

0.10%

0.21%

0.04%

0.10%

0.07%

0.09%

Japan

3.43%

5.39%

1.34%

3.48%

1.51%

3.07%

Korea

1.13%

1.53%

0.34%

1.05%

0.37%

0.39%

Malaysia

0.44%

0.96%

0.16%

0.44%

0.13%

0.15%

Taiwan

0.96%

1.41%

0.27%

0.99%

0.43%

0.40%

Philippines

0.31%

0.51%

0.12%

0.31%

0.12%

0.11%

Singapore

0.35%

1.15%

0.15%

0.39%

0.08%

0.09%

Thailand

0.14%

0.43%

0.05%

0.16%

0.09%

0.10%

84.06%

78.12%

94.12%

85.17%

89.70%

86.23%

Freight & Insurance

0.19%

0.37%

0.05%

0.19%

0.14%

0.15%

Imports from HKG

0.71%

1.13%

0.18%

0.83%

0.26%

0.09%

Imports from EU

1.28%

1.62%

0.80%

1.58%

1.59%

1.54%

Imports from ROW

6.18%

5.97%

2.16%

4.61%

4.81%

6.83%

Duties & Taxes

0.05%

0.03%

0.02%

0.06%

0.11%

0.14%

100.00%

100.00%

100.00%

100.00%

100.00%

100.00% 0.36%

Total

1995

Electronics computing equipment

China

USA

China

0.51%

0.50%

Indonesia

0.09%

0.06%

0.07%

Japan

5.67%

2.12%

2.83%

Korea

1.48%

0.44%

0.37%

Malaysia

0.98%

0.26%

0.23%

Taiwan

1.05%

0.48%

0.35%

Philippines

0.22%

0.11%

0.08%

Singapore

1.14%

0.19%

0.14%

Thailand

0.34%

0.11%

0.09%

82.84%

89.60%

86.05%

Freight & Insurance

0.20%

0.13%

0.16%

Imports from HKG

0.33%

0.16%

0.07%

Imports from ROW

4.96%

5.64%

8.96%

Duties & Taxes

0.18%

0.20%

0.24%

100.00%

100.00%

100.00%

China

0.12%

0.12%

0.14%

Indonesia

0.04%

0.04%

0.05%

Japan

3.44%

1.30%

3.25%

Korea

0.43%

0.19%

0.26%

Malaysia

0.26%

0.08%

0.10%

Taiwan

0.54%

0.26%

0.33%

Philippines

0.08%

0.03%

0.03%

Singapore

0.38%

0.06%

0.06%

Thailand

0.09%

0.04%

0.04%

90.01%

94.03%

87.07%

Freight & Insurance

0.13%

0.09%

0.15%

Imports from HKG

0.24%

0.10%

0.04%

Imports from ROW

4.04%

3.50%

8.24%

Duties & Taxes

0.19%

0.17%

0.23%

100.00%

100.00%

100.00%

USA

Total

1990

Television & audios∗1

USA

Total

Note: The “Television and Audio” sector includes radios and communication equipment, “Household electrical equipment” includes lighting fixtures, batteries, wiring and others, and “Automobile” includes auto components. Source: Calculated by the author.

Table 6.4

World market share of nine East Asian economies and the USA (electronics and electrical exports) in percentages (%) 1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

China Indonesia Japan Korea Malaysia Taiwan Philippines Singapore Thailand USA

1.02 0.05 20.67 4.03 2.21 4.72 0.23 5.22 1.02 16.77

1.18 0.10 21.37 4.27 2.68 5.10 0.54 5.53 1.26 16.93

2.19 0.27 NA 5.16 3.80 6.19 0.40 7.27 1.79 20.45

1.97 0.29 20.38 4.26 3.66 5.47 0.37 6.68 1.58 16.47

2.40 0.36 18.68 4.54 4.09 5.26 0.40 7.99 1.78 15.94

2.75 0.36 16.89 5.11 4.32 5.35 0.43 8.14 1.82 15.41

3.00 0.45 14.33 4.61 4.43 5.63 1.26 8.25 1.91 15.78

3.46 0.40 13.45 4.24 4.25 5.71 1.64 7.63 1.88 16.52

3.96 0.34 12.12 3.89 4.10 5.68 2.15 6.75 1.88 15.88

4.31 0.39 11.97 4.67 4.68 5.99 2.51 6.49 1.86 15.80

5.03 0.73 11.96 5.26 4.63 6.66 2.22 6.62 1.90 16.04

Total share

55.95

58.96

47.52

61.12

61.45

60.57

59.66

59.17

56.77

58.67

61.04

Note: Electronics and electrical products refer to the items covered by codes 75, 76, and 77 in the SITC Rev.2 classification. Taiwan’s trade data are obtained from “Other Asia, NES.” in the UN Comtrade Database Category. Source: Calculated from UN Comtrade Database.

203

204 Table 6.5 World market share of nine East Asian economies and the USA (automobile and auto component exports) in percentages (%) 1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

China Indonesia Japan Korea Malaysia Taiwan Philippines Singapore Thailand USA

1.23 0.01 21.13 1.08 0.04 0.73 0.02 0.18 0.07 9.83

1.61 0.02 22.37 1.15 0.05 0.89 0.02 0.19 0.09 10.51

0.43 0.07 NA 1.61 0.10 1.00 0.02 0.27 0.12 13.97

0.37 0.10 24.09 1.73 0.12 1.02 0.03 0.29 0.19 12.41

0.47 0.08 21.30 1.76 0.10 0.95 0.05 0.26 0.23 12.28

0.63 0.09 18.22 2.36 0.11 0.96 0.06 0.28 0.17 11.49

0.57 0.08 16.20 2.73 0.11 0.84 0.07 0.29 0.17 11.43

0.60 0.07 16.63 2.67 0.11 0.82 0.09 0.27 0.24 12.05

0.71 0.06 15.53 2.34 0.10 0.75 0.08 0.18 0.26 11.24

0.75 0.08 15.78 2.56 0.09 0.76 0.10 0.16 0.38 10.81

1.21 0.09 16.23 2.84 0.08 0.83 0.12 0.17 0.46 11.03

Total share

34.32

36.90

17.58

40.35

37.48

34.36

32.49

33.55

31.27

31.47

33.04

Note: Automobile and auto components are code 78. Taiwan’s trade data are obtained from “Other Asia, NES.” in the UN Comtrade Database Category. Source: Calculated from UN Comtrade Database.

Hiroshi Oikawa

205

References Akinwande, Akintunde, Douglas Fuller, and Charles Sodini. 2005. “Leading, Following or Cooked Goose: Explaining Innovation Successes and Failures in Taiwan’s Electronics Industry.” In Global Taiwan: Building Competitive Strengths in a New International Economy. ed. Suzanne Berger and Richard Lester, paper 3. Armonk, NY: ME Sharpe. Borrus, Michael, Dieter Ernst, and Stephan Haggard, eds. 2000. International Production Networks in Asia: Rivalry or Riches? London and New York: Routledge. Ernst, Dieter and Linsu Kim. 2002. “Global Production Networks, Knowledge Diffusion, and Local Capability Formation.” Research Policy 31(8–9): 1417–1429. Fröbel, Folker, Heinrichs Jürgen, and Otto, Kreye. 1980. The New International Division of Labour. Cambridge: Cambridge University Press. Fujikawa, Kiyoshi, Mitsuru Shimoda, and Takatoshi Watanabe. 2005. “The Structure of International Division of Labor in the Asia-Pacific Region.” Konan Ronshu 46(3): 1–34. Feenstra, Robert C. and Gary H. Hamilton. 2006. Emergent Economies, Divergent Paths: Economic Organization and International Trade in South Korea and Taiwan. Cambridge, UK: Cambridge University Press. Gereffi, Gary. 1999. “International Trade and Industrial Upgrading in the Apparel Commodity Chain.” Journal of International Economics 48 (1): 37–70. Gereffi, Gary, and Raphael Kaplinsky. eds. 2001. The Value of Value Chains: Spreading the Gains from Globalisation. IDS Bulletin 32(3). Brighton: Institute of Development Studies. Gereffi, Gary and Miguel Korzeniewicz. eds. 1994. Commodity Chains and Global Capitalism. Westport: Praeger. Henderson, Jeffrey. 1998. “Danger and Opportunity in the Asia-Pacific.” In Economic Dynamism in the Asia-Pacific, ed. G. Thompson. London and New York: Routledge: 356–384. Hobday, Mike. 1995. “East Asian Latecomer Firms: Learning the Technology of Electronics.” World Development 23(7): 1171–1193. Hobday, Mike. 2001. “The Electronics Industries of the Asia-Pacific: Exploiting International Production Networks for Economic Development.” Asian Pacific Economic Literature 15(1): 13–29. Kaplinsky, Raphael. 2000. “Globalisation and Unequalisation: What Can Be Learned from Value Chain Analysis?” Journal of Development Studies 37(2): 117–146. Kaplinsky, Raphael. 2005. Globalization, Poverty and Inequality. Polity Press. Kawai, Masahiro. 2005. “East Asian Economic Regionalism: Progress and Challenges.” Journal of Asian Economics 16(1): 29–55. Kimura, Fukunari, and Mitsuyo Ando. 2003. “Fragmentation and Agglomeration Matter: Japanese Multinationals in Latin America and East Asia.” North American Journal of Economics and Finance 14(3): 287–317. Kimura, Fukunari and Mitsuyo Ando. 2005. “Two-Dimensional Fragmentation in East Asia: Conceptual Framework and Empirics.” International Review of Economics and Finance 14(3): 317–48. Lall, Sanjaya. 1995. “Industrial Strategy and Policies on Foreign Direct Investment in East Asia.” Transnational Corporations 4(3): 1–26.

206 Inter-Country Value Distribution in East Asia Lall, Sanjaya, Manuel Albaladejo, and Jinkang Zhang. 2004. “Mapping Fragmentation: Electronics and Automobiles in East Asia and Latin America.” Oxford Development Studies 32(3): 407–432. Linden, Greg, Kenneth Kraemer, and Jason Dedrick. 2007. “Who Captures Value in a Global Innovation System? The Case of Apple’s iPod.” University of California at Irvine, Personal Computing Industry Center (PCIC) working paper, June. Matsumura, Fumitake, and Kiyoshi. Fujikawa. 1998. Economic Analysis of Home Production (in Japanese) Tokyo: Iwanami Shoten. Ng, Francis and Alexander Yeats. 1999. Production Sharing in East Asia: Who Does What for Whom, and Why? World Bank. Oikawa, Hiroshi Michael. 2004. Is Proton Car Really Made by Malaysians? mimeo. Petri, Peter A. 2006. “Is East Asia Becoming More Interdependent?” Journal of Asian Economics 17(3): 381–394. Schmitz, Hubert (ed.). 2004. Local Enterprises in the Global Economy—Issues of Governance and Upgrading. Cheltenham: Edward Elgar. Sohn, Byeong Hae. 2002. “Regionalization of Trade and Investment in East Asia and Prospects for Further Regional Integration.” Journal of the Asia Pacific Economy 7(2): 160–181. Song, Lei. 2007. “Modularization, modularity traps and competitiveness: towards an architecture analysis of China’s AV industry.” ITEC workshop presentation, 24 October, Kyoto, Japan. Sturgeon, Timothy. 2007. “How Globalization Drives Institutional Diversity: The Japanese Electronics Industry’s Response to Value Chain Modularity.” Journal of East Asian Studies 7(1): 1–34. Sturgeon, Timothy J., Johannes V. Biesebroeck, and G. Gereffi. 2008. “Value Chains, Networks and Clusters: Reframing the Global Automotive Industry.” Journal of Economic Geography 8(3): 297–321. Sturgeon, Timothy J., and Richard Lester. 2004. “The New Global Supplybase: New Challenges for Local Suppliers in East Asia.” In Global Production Networking and Technological Change in East Asia, ed. S. Yusuf, A. Altaf and K. Nabeshima. Oxford University Press: 35–87. Sturgeon, Timothy, and Olga Memedovic (forthcoming). “Measuring Global Value Chains: Intermediate Goods Trade and Structural Change in the World Economy.” UNIDO Working Paper. United National Industrial Development Organization, Vienna. Urata, Shujiro, Chia S. Yue, and Fukunari Kimura, eds. 2006. Multinationals and Economic Growth in East Asia. New York: Routledge. Whittaker, Hugh, Tianbiao Zhu, Timothy Sturgeon, Mon Han Tsai, and Toshie Okita. 2010. “Compressed Development” Studies in Comparative International Development. October.

7 Learning and Earning in Global Value Chains: Lessons in Supplier Competence Building in East Asia Timothy J. Sturgeon and Greg Linden

7.1. Introduction The research collected in this volume represents years of challenging fieldwork and careful documentation on the part of the authors. Instead of making armchair assumptions about causes and outcomes of development and technological learning based on the thin gruel provided by official statistics, the authors have spent many months in the field, visiting companies, interviewing entrepreneurs, talking to government officials, collecting official and unofficial statistics, administering questionnaires, and learning about the key technologies and firms in specific industries. The richness of this approach is obvious from the preceding chapters. As in prior field-based global value chain (GVC) research, however, the richness and variety of the findings can make it difficult to reach generalizations needed to form the basis of sound strategic or policy advice. Nevertheless, a central theme stands out. Global value chain analysis highlights both the opportunities and limits that come with participation in cross-border production networks. The analysis centers on the spatial separation of business activities that are sequenced along the value chain (e.g., design, manufacturing, marketing, distribution), and especially the division of labor between developed country lead firms and developing country suppliers. Because of the segmentation and spatial distance between value chain functions, local suppliers can become specialized in segments of the GVC with low profitability and limited opportunities for learning. By and large, the research collected in this volume reinforces this interpretation. However, because the research methods are so in-depth, the results also show that it is 207

208 Learning and Earning in Global Value Chains

sometimes possible for local firms to learn over time, upgrade their capabilities, and gain more control over their own fate. It is this combination of findings—confirming the barriers to upgrading found in prior GVC research and uncovering ways in which local firms have managed to overcome these barriers—that makes this volume so valuable for industry practitioners, policy-makers, and GVC theorists alike. In this final chapter we highlight the issues related to value chain governance and industrial upgrading that appear prominently in this volume. We focus on the barriers to upgrading and the ways in which developing country suppliers have managed to enhance their capabilities through engagement in GVCs. This summary analysis is accomplished through an exploration of six main theoretical themes that run throughout the book: asymmetry in GVCs, supplier strategies, platform leadership, market power, standards, and lead firm nationality. We provide some examples from our own research and end with some concluding comments about the lessons learned and about the policies needed to support supplier learning in GVCs.

7.2. The contributions of this volume The contributions of this volume in the realm of GVC theory building are significant. This volume shifts the central question from “What are the archetypical patterns of GVC governance and what determines them?” to “What determines supplier competence in the context of late developing economies?” To help answer this, the authors draw on insights from within GVC governance theory (complexity and codifiability remain as important explanatory variables) and from the other two pillars of GVC theory building (power and institutions; the latter, while underemphasized in some chapters, do play a role in many of the cases). The authors have also adapted other, adjacent theory-building streams, such as the firm capability approach (Teece and Pisano 1994; Teece et al. 1997), the technological learning approach (Bell and Pavitt 1992, 1995, 1997; Lall 1992), and the entrepreneurial orientation framework (Covin and Slevin 1989, 1991) in ways that are complementary to GVC governance theory. By shifting the research question to focus on competence building and entrepreneurialism, and enriching the field of causation and explanation, the authors have improved upon previous attempts to examine the processes of GVC-engaged supplier upgrading. They explicitly link their research questions to GVC governance theory and forge bridges to complementary theories. As we have stated elsewhere (Sturgeon 2009), it is

Timothy J. Sturgeon and Greg Linden

209

just this sort of articulated, “modular” theory building that is needed to deepen our understanding of the relationship between economic development and globalization. With these theoretical tools in hand, we can begin to explain why some places have been able to profit from GVCengagement while other places have foundered, and why in some industries local suppliers have begun to wield substantial power within GVCs, while in others they remain subservient to lead firms from developed economies, stuck in low value-added segments of GVCs, and find their competence-building strategies blocked rather than enabled by their linkages to GVCs. In the remainder of this section, we explore six theoretical themes that emerge from the research presented in this volume.

7.2.1. Asymmetry in GVCs A core argument of GVC governance theory is that different governance forms come with different requirements for network entry, and that these requirements help to structure the opportunities, costs, and risks that come with participation in GVCs. Because latecomer firms in developing countries begin with relatively meager technological and managerial competencies, they tend to have a narrower range of choices than suppliers based in advanced economies. Although the process is uneven and contradictory examples exist, the trend since the mid-1990s has been for lead firms in GVCs to ask more from suppliers in terms of responsiveness, quality, use of information technology, and compliance with environmental and labor standards (Dolan and Humphrey 2000; Sturgeon and Lester 2004). Since these requirements typically exceed those that prevail in developing countries, it follows that local suppliers should seek to learn through business linkages to lead firms in GVCs, especially in ways that maximize their own power and autonomy (i.e., via modular and relational linkages). However, this can be very difficult to achieve at the outset, or at all, and suppliers that begin with more asymmetric linkages (i.e., captive) can be locked into low value activities and segments of the value chain for very long periods of time. The spatial structure of GVCs also affects the prospects for supplier upgrading. As Humphrey and Schmitz (2002) point out, and as the Chapters 3 and 4 of this volume demonstrate, entering into a captive supply relationship with the local affiliate of a powerful multinational corporation (MNC) can be an effective route to learning. However, a key theme, repeated throughout this volume, is that information, knowledge, and value capture are unequally distributed within GVCs. Learning prospects for local firms can be severely truncated

210 Learning and Earning in Global Value Chains

when innovative, profitable, symmetrical, and relational GVC linkages (based on the exchange of uncodified, tacit information and knowledge) are located elsewhere. While opportunities for supplier upgrading in captive supply relationships might be significant at first, they can be limited over time when information, knowledge, and value capture remain partitioned, out of reach, in the global headquarters and regional technology centers of lead firms, global suppliers, and platform leaders. The very fact that GVCs fragment intellectual and physical business processes, and that these fragments can form an international division of labor, means that local suppliers can toil for decades on low value-added functions without the opportunity to engage in, or even observe, higher-level functions such as design and marketing. Evidence can be found for this in the “questionable winners” of China, Korea, Malaysia, Taiwan, the Philippines, and Thailand in terms of their contribution to regional value added in electronics over time, as depicted in Chapter 6 by Hiroshi Oikawa. When GVC engagement is limited in this way, managers at local firms may be only partially aware—or even completely unaware—of the product and technology roadmaps and long-term strategies of their immediate customers (lead firm MNC affiliates or global buyers), or the evolving preferences of the ultimate end-users of what they produce. This barrier to upgrading is evident in all chapters in this volume. Consider the case of the Malaysian suppliers to the local affiliate of Company X (a Japanese consumer electronics company), discussed by Hiroshi Oikawa in Chapter 5. The location of Company X’s product design and component selection activities in Japan and Singapore exclude local suppliers from these activities. Higher value components, such as electrical and electro-mechanical parts, were also sourced in Singapore and Japan. As a result, local purchasing by Company X’s Malaysian affiliate, while gradually increasing, remained concentrated in basic activities and inputs such as assembly, metal and rubber parts, and packaging material. Here we see learning opportunities provided by captive linkages to local MNC affiliates limited because higher value functions were concentrated in the MNC’s home country and in the most developed economies in East Asia. The modular linkages between branded PC companies and contract manufacturers in Taiwan discussed in Chapter 1 by Momoko Kawakami provide a contrasting case. These linkages did not come with as much hands-on local assistance as the captive linkages to MNC affiliates described in Chapters 3, 4, and 5, but they appear to

Timothy J. Sturgeon and Greg Linden

211

have afforded Taiwanese firms with much greater scope for long-term learning and autonomy. 7.2.2. Supplier strategies matter . . . to a point While the GVC governance variables of complexity and codifiability (including, prominently, de facto standards such as the counterfeit 1950s-era Honda Cub mentioned in Chapters 3 and 4 and Intel PC chip sets in Chapter 1) make central appearances in this volume, the chapters break new ground by probing deeply into the third key variable in the GVC governance framework: supplier competence. Where does supplier competence come from? This is certainly a fertile area for GVC research. In the GVC governance model supplied by Gereffi et al. (2005), supplier competence is an independent variable and conceived of in relative terms. In this sense it is quite different from the variables of complexity or codifiability, which seek to objectively characterize the link between activities in GVCs. Supplier competences lie on only one side of the transaction and are relative; what might be adequate or superior competence in one setting may fall short in another. As Kaplinsky et al. (2002) point out, it is competence asymmetries, more than anything else, that drive supplier upgrading in GVCs since firms with disparate levels of competence seek to forge functional business relationships. In the context of GVCs, less competent suppliers are forced to learn or exit. The research collected in this volume examines supplier strategies in depth and in situ, and as a result, the power asymmetries in the GVCs under examination come to the foreground. Clearly, suppliers in latecomer economies inevitably operate within constraints. The fact that the bulk of strategic agency tends to lie elsewhere in GVCs, specifically with the managers of lead firms and platform leaders from advanced countries, is powerfully evident in the research results. How can suppliers cope in the face of these power asymmetries, technological deficiencies, and deep, long-term dependencies? One way is for them to focus on domestic value chains and markets, either as suppliers serving local lead firms or by developing and selling their own products, as is discussed in Chapters 2, 3, and 5. At the very least, local markets can allow suppliers to develop baseline capabilities that can help them to win initial business from lead firms in GVCs on more favorable terms than would otherwise be the case. In addition, the focus on supplier strategy highlights the role of managerial agency. The theory of GVC governance developed by Gereffi et al. (2005) is meant to be operational—that is, assigning high or low values

212 Learning and Earning in Global Value Chains

to key variables of complexity, codifiability, and supplier competence yields five distinct forms of GVC governance, the model’s dependent variable. However, as the chapters in this volume clearly show, suppliers do not respond automatically to technical requirements; these requirements pass through the entrepreneurial and strategic filters of suppliers’ founders and managers (see especially Chapter 5). The various supplier strategies specified in the volume—flexible substitution of labor and capital, mixing of local and export markets, and moving between intermediate and final goods production—take us a long way toward dispelling the misconception of GVC governance as a mechanical, technologically determined process (see especially Chapters 2, 3, and 4).

7.2.3. Platform leadership matters For developing country firms involved in product innovation, the solution in technologically intensive product areas such as electronics is to purchase highly modular design solutions from “platform leaders” (Gawer and Cusumano 2002). This allows fast market entry, but also leads into several traps. First, as we have already mentioned and will discuss in more depth later, there are the high costs associated with acquiring highly functional components and subsystems, as well as the royalties that must be paid, directly or indirectly, to the platform leaders and other standard setters in an industry. Second, there is the “modularity trap” as identified by Chesbrough and Kusunoki (2001), where the highly integrated off-the-shelf components and subsystems provided by platform leaders reduce product distinctiveness. Firms selling products based on highly integrated platforms are forced, as the local Chinese mobile handset makers discussed in Chapter 2 have been, to target the low end of the market and rely on cosmetic attributes to attract customers. While, in the case of China, the size of the local market allowed this strategy to be viable for a time, rapid increases in customer sophistication quickly led to trouble, and without the internal technological resources to develop truly innovative products, local handset producers and contract design firms were forced to adopt—and pay for—the highly integrated, more “completely modular” chipset platforms offered by Mediatek and others. At the same time as this modularity trap deepened, MNC lead firms such as Motorola and Nokia recognized the importance of the low end of the market and of places within China that had so far been underserved by their distribution networks, and adjusted their strategies accordingly.

Timothy J. Sturgeon and Greg Linden

213

Clearly, firms with deep cash reserves and technological capabilities will have an easier time moving from high to low end products than firms with limited financial and technical resources will have moving in the opposite direction.1 In Chapter 1, Momoko Kawakami shows that Taiwan’s ODM notebook PC contract manufacturers (e.g., Quanta), despite producing more than 90% of the world’s notebook PCs and providing a host of sophisticated design, procurement, manufacturing, and logistics management services, have far lower profitability than the lead firms that sell the machines (e.g., Dell) and platform leaders (e.g., Intel) that provide their core components. Through this research, we gain a better understanding of why a set of firms that have, on the face of it, captured a dominant market share of a huge and rapidly growing product segment find it difficult to earn significant rents or gain any real market power. Using the language of Baldwin and Clark (2000), we can say that Intel, as a platform leader, has the technological capability and market power to unilaterally change the location of key “pinch points” in the personal computer value chain. In other words, Intel can decide how to bundle tacit activities and where to locate the points in the chain where codified handoffs can occur. In Chapter 1, we saw how Intel “encapsulated” a bundle of hard-to-integrate PC functions and sold them as a standard chipset, radically devaluing one of the critical “intrinsic cores” of lead firms in the notebook PC industry, especially Japanese lead firms, which had developed and relied on system integration competencies more than US firms. As a result, Taiwan’s notebook PC producers have taken on many new tasks, and gained many new competencies, but with Intel products lying at the center of nearly all their products, their competencies have remained more or less interchangeable, undermining their market power and profitability. In most industries, however, lead firms, not component suppliers, define system architecture. The PC and mobile phone handset cases, while important and well-known cases, are quite unusual. The motorcycle case studies by Fujita and Sato in Chapters 3 and 4 clearly reveal the dominance of the lead firms that create and control the system architecture. Nevertheless, the sustained consideration of platform leaders in this volume provides a major step forward in GVC research. The notebook PC and mobile phone handset cases show why the term “lead firm” does not necessarily imply that these GVC actors are the most dominant, important, or profitable actors in the chain. From a GVC perspective, lead firms begin or “lead” the GVC’s actions by placing orders in their supply chain. Order placing gives them market power,

214 Learning and Earning in Global Value Chains

certainly, as does their close linkage to end-users. This role is earned, if not by technological leadership, then by the financial risk taken on in the period between placing orders and selling products (a window that lead firms are ever trying to shorten). However, in some industries, such as PCs, mobile phones, and bicycles (Galvin and Morkel 2001; Fixon and Park 2008), platform leaders do dominate. Going forward, GVC research should follow this volume’s lead in examining and theorizing the role of platform leaders, if they exist in the industry under examination. 7.2.4. Market power matters To expand on Ken Imai and Jing Ming Shiu’s discussion in Chapter 2 of how China’s cellular handset market is structured by the global market leaders in the mobile handset industry, Table 7.1 shows how, in contrast to China, where market share tends to change rapidly from year to year, the global market has been fought over in recent years by the same five firms, whose combined share of the total has grown from 76.5 to 81.5%. The oligopolistic market power of these firms provides them with advantages in terms of negotiating for parts availability and price and generating the cash needed to fund R&D and strengthen their brand identity through advertising. But the dominance of some of these firms runs even deeper than profits and returns to scale from high market Table 7.1 Top 5 cellular handset vendors worldwide, 2005 and 2008 (millions of units) Rank

2005 Vendor

1 2 3 4 5

Shipments

2008 Market Vendor share (%)

Nokia Motorola Samsung LG Sony-Ericsson Others

264.9 146.0 102.9 54.9 51.1 190.7

32.7 18.0 12.7 6.8 6.3 23.5

Total

810.5

100.0

Nokia Samsung LG Motorola Sony-Ericsson Others Total

Shipments

Market share (%)

468.4 196.7 100.7 100.1 96.6 218.5

39.7 16.7 8.5 8.5 8.2 18.5

1,180.9

100.0

Source: 2008 data from IDC, compiled by Digitimes, February 5, 2009; 2005 data from Strategy Analytics press release, January 30, 2006.

Timothy J. Sturgeon and Greg Linden

215

share. First movers develop well-honed organizational capabilities that are difficult to replicate, and as the global economy has become better integrated, what were national and regional advantages can become global advantages. For example, Nokia, Motorola, and Ericsson have ranked in the top five mobile handset producers since the early 1990s, and are among the key holders of cellular technology patents. Each technology generation has required extensive development, coordination, and negotiation with other firms and organizations. In return for this effort, and because of their long-term market success, these firms can earn significant royalty streams from licensing their intellectual property to following firms. Typical 2.5G royalties run about $5 per phone and third-generation (3G) royalties can be higher. Chinese firms selling domestically pay lower rates in some cases, and doubtless many of the “unauthorized” mobile phone producers mentioned by Imai and Shiu in Chapter 2 do not pay royalties at all, but Chinese companies have had to pay hundreds of millions of dollars in royalties to the industry’s leading patent holders. China’s domestic mobile phone lead firms (Lenovo, Ningbo Bird, TCL, etc.) have started to develop their own R&D capability, but, as Imai and Shiu point out, they are still largely reliant on the platforms (chipsets and software) provided by companies like Mediatek, Texas Instruments, Qualcomm, and others, which makes it difficult for them to differentiate their products and gain market share. Our assessment is that China’s firms remain five to ten years away from participating in large-scale efforts to develop innovative technologies for world markets. This will require not only basic R&D capability, but also a commitment of resources for participation in standard-setting bodies, including basic and applied research in the most advanced realms of wireless communication.

7.2.5. Standards matter As already mentioned, chains of value-adding activity are commonly divided at the “pinch points” where information is codified (Baldwin and Clark 2000), creating opportunities for both outsourcing and offshoring. Because they set the rules of the game and define the pinch points that bracket, or “encapsulate,” complex technologies and bundles of tacit activities, standards play an important role in determining the structure and trajectory of GVCs. Moreover, as we have already discussed, firms that dominate the standard setting processes accrue significant advantages in terms of value capture.

216 Learning and Earning in Global Value Chains

The deep technological competence and powerful market positions of the lead firms (e.g., Honda, Nokia, Motorola, Dell) and platform leaders (e.g., Intel, Qualcomm) that have appeared as important actors in this volume allow them to set de facto standards or at least play dominant roles in de jure industry standard setting. Industry standard setting bodies, and the standards themselves, form institutions that in turn set the parameters for less powerful actors in GVCs, and for the governance of the chain as a whole. In other words, they come to comprise an industry’s “conventions” (Ponte and Gibbon 2005). As a result, standards that enable complex product and process designs to be codified are not set or given. Nor do they arise from an institutional, rule setting nexus that exists apart from corporate strategy. They are instead key arenas for strategic action for the lead firms and platform leaders that dominate these industries. They arise as much from competitive battles and outcomes—competitive success generates both financial power and market power through path dependency and network effects—as from objective and dispassionate government rule setting or even pragmatic compromise among industry players. Consider the complex intellectual property ownership patterns behind the current generation of 3G mobile communications technology, which has been recently introduced in China. Development began in the 1990s in individual firms and also in consortia such as the European-dominated 3rd Generation Partnership Project (3GPP) and the Qualcomm-dominated 3GPP2. In 1999, a United Nations organization, the International Telecommunication Union, approved several 3G wireless standards. One of the competing 3G standards is called Wideband Code Division Multiple Access (WCDMA), a follow-on technology to 2G called Global System for Mobile Communications (GSM). WCDMA was developed in part by the Japanese wireless carrier NTT DoCoMo, and launched in Tokyo in 2001. The underlying patents are held by a broad array of companies, particularly Nokia, Ericsson, Qualcomm, and Motorola. About 20% of the patents are owned by a dozen companies including NTT, Panasonic, France Telecom, Siemens, and SK Telecom, who pool their licensing through a group called 3G Licensing. The American firm Qualcomm, which had also developed the 2G technology CDMA, principally developed a competing 3G standard, CDMA 2000 (also known as IMT Multi-Carrier (IMT-MC)). The other main holders of essential CDMA patents are Nokia (Finland), Motorola (USA), and NTT DoCoMo (Japan). In the China context, an important 3G technology is Time Division Synchronous Code Division Multiple Access (TD-SCDMA), which

Timothy J. Sturgeon and Greg Linden

217

Chinese organizations developed based on an initial technology proposal from Germany’s Siemens. The goal was to promote Chinese technological independence, but it is likely that TD-SCDMA equipment makers will still pay royalties to the Western companies that own almost half the relevant patents. In January 2009, China licensed its carriers to build 3G networks, dividing them among the different technologies: TD-SCDMA for China Mobile, CDMA2000 for China Telecom, and WCDMA for China Unicom. Patent holders for cellular technology are in a position to crosslicense with each other, which lowers their royalty expenses relative to non-patent-holding rivals. This system gave the technology leaders for 2G—Motorola, Nokia, and Ericsson—a built-in cost advantage over rival handset makers and helped to keep them among the global leaders. The introduction of TD-SCDMA is the first time Chinese companies are in a position to benefit from intellectual property ownership in cellular technologies, but at this time it seems unlikely that the technology will be important in the global market. In this context, each of the three groups of local firms described by Ken Imai and Jing Ming Shiu in Chapter 2, the independent design firms, baseband chipset vendors, and integrated handset producers, all have to labor under the burden of technological catch up. Not only do they have to keep abreast of rapidly moving technologies and standards, they have to pay hefty royalties to incumbent firms—including Motorola, Nokia, and Sony-Ericsson which are direct competitors in the Chinese market—regardless of which standard they embrace. Furthermore, there are additional layers of technology licensing, and additional royalties, to be paid. When Chinese handset makers adopt an integrated handset platform offered by Mediatek or Spreadtrum, for example, or try to create their own platform from available modules, they must pay, directly or indirectly, for the proprietary technology embedded in blocks of “semiconductor intellectual property” offered by companies such as ARM (UK) and CEVA (USA). 7.2.6. Lead firm nationality matters As Borrus (1995) has argued, GVC-related strategies tend to differ according to the nationality of lead firms. Lead firm strategies can and do differ greatly from firm to firm, and from sector to sector, but national patterns can be quite clear (see Sturgeon 2002 and 2007; Dupuy et al. 2010). One of the most striking examples of this is how Japanese firms tend to control their value chains more tightly than lead firms from other countries. On the other side of the spectrum are lead firms

218 Learning and Earning in Global Value Chains

based in the United States, which tend to outsource more broadly and dynamically. Key requirements for market entry are also important. For suppliers seeking to develop a relationship with a Japanese lead firm, a willingness to learn and pay close attention to customer requirements (especially for quality) are attributes that may, after a lengthy period of examination, lead to the establishment of a long-term relationship, even if the supplier subsequently makes mistakes. For American lead firms, price is more important, qualification periods are shorter, little assistance is offered, and as a result supplier capabilities must be further developed from the beginning. For local suppliers these differences create a pair of double-edged swords. Working for Japanese lead firms can mean more help in the form of technical assistance and long-term financing, but it can also limit opportunities for moving beyond basic tasks to the more design- and technologically intensive functions carried out by Japaneseowned suppliers and by the lead firms themselves, in part because of the long-term commitments made by lead firms to affiliated suppliers and internal employees. The research presented in Chapters 3, 4, and 5 reinforce the view that Japanese MNC affiliates provide excellent learning opportunities for the local suppliers that are ultimately selected. Suppliers to Japanese lead firms were supported, especially, in their efforts to meet quality requirements. On the other hand, because Japanese lead firms tend to prefer long-term, captive relationships, such opportunities were few and far between, and local suppliers were largely walled off from higher value segments of the value chain. Working for American and European companies can provide broader learning opportunities, as shown in Chapter 1, but managers’ willingness to switch suppliers means that mistakes can be followed by a sudden loss of business. If suppliers fail to provide adequate services, buyers are more likely to move on without helping suppliers find remedies. In sum, and in general, the learning opportunities for local suppliers, though still limited, appear to be more dynamic with American and European lead firms as customers. Also, in Chapter 4, Mai Fujita shows that the Vietnamese-Chinese chains had market and modular linkages to the local supply base (supported by the de facto standard of counterfeit product designs coming from China), which offered more flexibility and autonomy but low potential for learning and profitability, while Japanese-led chains had captive or semi-captive linkages, which offered a steeper gradient for initial technological learning and more potential for earning, but less autonomy.

Timothy J. Sturgeon and Greg Linden

219

Since the research on MNC affiliate sourcing in East Asia presented in this volume has focused on Japanese lead firm examples, we provide some balance here with a case from Singapore that traces one local supplier’s development, largely in service of American and European MNC affiliates in the region. In 1981 two Singaporean engineers decided to start their own company after they were laid off from the Singaporean subsidiary of the German camera manufacturer Rollei. Seeing that the local tool and die business in Singapore was underdeveloped—because most foreign firms tended to bring in their own tooling—the two set up their own tool and die shop on a chicken farm owned by one of the founders’ parents. From their experience at Rollei they knew that advanced lathes for precision metal cutting spun very fast but could be stopped quickly to make rapid set-up changes. The two retrofitted some inexpensive lathes with motorcycle brakes to achieve the same effect. The company, which was originally called Uraco, generated $700,000 in revenues during its first year of operation, mostly by supplying precision metal parts to American disk drive producers, which were investing heavily in manufacturing in Singapore and Malaysia at the time (Business Times, 1995). As Uraco grew, it began to supply a wider range of products to the disk drive industry, including precision metal stampings and assembled electronic circuit boards. Most of the company’s business was with Seagate, the leading American disk drive manufacturer, but the company also exported precision parts to Hitachi’s disk drive operations in the Philippines. Because of the extreme volatility in disk drive and PC markets, in 1987 managers began the first of many efforts to diversify Uraco’s customer base by distributing electronic components, eventually winning distributorships from Motorola, Harris Semiconductor, and Siemens.2 In the mid-1990s the company tried a more autonomous route, leveraging its experience with electronic components, contract manufacturing, and warehouse management to manufacture and sell products of its own design, including connectors, crystals, automated warehouse vehicles, electronic ballasts for fluorescent lamps, light bulbs, and telecommunication-related products. Ultimately, these attempts were not successful, and the bulk of Uraco’s business remained in providing contract manufacturing services and precision-engineered metal parts to foreign firms operating in the South East Asian region. As traditional distribution networks in the region matured, the need for the company’s distribution services waned as well. Nevertheless, in 1995 the company underwent a successful initial public offering on the Singaporean stock exchange. In 1996, as revenues

220 Learning and Earning in Global Value Chains

were approaching $53 million, Uraco won an important contract to manufacture flatbed scanners for Hewlett Packard (Business Times 1996a, 1996b, and 1997). The company’s troubles were not over, however, and flagging profitability led to a management reshuffle in 2000 and a name change, to Beyonics, in 2001. The company returned to profitability in 2001, when it generated nearly $300 million in revenues, with 62% coming from contract manufacturing services, 29% from precision engineering, and 9% from distribution (Geocities 2004). The company’s current product and service offerings are electronics manufacturing services (i.e., contract manufacturing), medical and consumer plastic injection molding and assembly, precision engineering services, precision metal stampings, and precision tooling design and fabrication services. This is a highly focused and complementary product and services portfolio, covering many of the processes and a few of the basic products required to produce a wide variety of electronics and closely related goods. The company has followed the rest of the electronics contract manufacturing industry toward the bundling of services to enable the production of complete products through its acquisitions of precision plastic moldings suppliers Techplas (in 2000) and Pacific Plastics (in 2002). In 2003 the company merged with a similar Singaporean contract manufacturer, Flairis Technology Corporation, to achieve additional economies of scale and scope (company website). The company’s distribution activities and attempts at selling its own branded products have been dropped entirely. With this tighter focus the company has expanded dramatically. According to the market research newsletter Manufacturing Market Insider, Beyonics revenues of $1.57B (with a razor-thin net profit of 0.3%) ranked the company 13th on a list of world’s largest electronics contract manufacturers in 2008 (http://www.mfgmkt.com/, accessed August 12, 2009). Through a combination of internal expansion and acquisition, Beyonics has developed a solid regional manufacturing footprint, most notably by establishing “vertically integrated” electronics contract manufacturing campuses in Kulai, Malaysia, in 2005; Suzhou, China, in 2006; and Batam, Indonesia, in 2007. In all, the company currently operates 16 facilities: 3 in Singapore, 6 in Malaysia, 3 in China, 2 in Thailand, and 2 in Indonesia. While Beyonics may have grown much larger than most local firms in East Asia that started as suppliers to MNCs, there are several lessons to be drawn from this “best” case. First, Beyonics’ managers demonstrated the use of dynamic capabilities (Teece 2009) for sensing opportunities, seizing them, and transforming the company as needed. Second, they

Timothy J. Sturgeon and Greg Linden

221

stumbled by trying to diversify and develop their own products, which required end-user marketing competences they had not yet developed, but recovered when they refocused on providing producer services to MNCs in the region. Third, like most large electronics contract manufacturers, Beyonics has struggled to remain profitable, even as the company has grown rapidly. Fourth, as the company expanded it chose a variety of lower cost locations within East Asia, balancing its investments in China with locations in Malaysia, Thailand, and Indonesia. What the Beyonics case illustrates most dramatically, however, is how, with enough time (a 28-year span in this case), local firms with extremely humble roots have been able to grow, master advanced technologies, and set up multiple locations in East Asia, largely by serving American MNC affiliates in the region. The shifting strategies that guided the development of Beyonics over the years were dynamically aligned with the evolving outsourcing strategies of its customers. This is a perfect case of “dynamic external fit,” described by Momoko Kawakami in Chapter 1, citing Kimura (2007), where “a latecomer firm operating as a supplier in GVCs identifies the domain of activities that align with the sourcing strategies of lead firms and focus on the activities that create a dynamic platform for learning and upgrading.” At the same time, it has to be noted that Beyonics has, so far, only been successful by growing within its expanding but unprofitable GVC niche of contract manufacturing.

7.3. Concluding remarks The overarching question asked by this volume is “Where do supplier capabilities come from?” The answer, we learn, is complex. On the one hand, lead firms in GVCs offer learning opportunities. But on the other hand the nature of these opportunities is neither uniform nor static. They vary from firm to firm, over time, by industry, by location, by position in the value chain, and to some measure, by lead firm nationality. What this volume makes clear is how difficult and tenuous this path can be. Given the potential for economic development and the significant challenges, several questions come to the fore. What learning opportunities are present? How can they be leveraged? How can supplier achievements be measured? What are the patterns, best practices, and lessons to be learned from supplier efforts to learn and earn in GVCs? This volume has provided a rich set of answers to these questions. Most importantly, perhaps, we have learned that suppliers can use a variety of strategies to try to overcome the challenges posed by GVC

222 Learning and Earning in Global Value Chains

participation. If any single point stands out from the cumulative learning offered by this volume, it is that developing country suppliers have agency, and that their strategic approach to supplying lead firms in GVCs matter, to a point. The chapters take us far away from the simplistic view that supplier upgrading in GVCs is entirely a function of lead firm strategy, on one hand, or at worst, of supplier malfeasance as they seek to appropriate proprietary knowledge from customers to meet their own goals. Suppliers, if sufficiently entrepreneurial, highly motivated, and provided access to functional financial markets (e.g., Beyonics and Quanta), can maximize learning opportunities and leverage what they have learned to expand their markets, without necessarily alienating their initial customers. Supplier learning, in turn, changes conditions on the ground. Higher-level competencies might attract more business from existing customers and open up new sets of customers, sometimes in local markets or in different industries. But these opportunities have limits and come with huge challenges. Learning and earning in GVCs is neither easy nor automatic. This volume powerfully demonstrates how GVC governance patterns can change with extreme rapidity as firms jockey for position in fast moving markets by encapsulating product functionality into new platforms and core modules. New actors, technologies, standards, and products can appear on the scene to change markets’ characteristics and the critical factors for success. In developing countries, markets can explode from very small bases, and while one might logically expect these rising tides to lift all boats, Chapters 2–5 of this volume clearly show that that this overall growth can contain, and sometimes mask, extreme market share volatility. Such environments provide opportunities, obviously, but they can also prove to be extremely challenging, and even threaten the survival of local suppliers when conditions suddenly change, as they did for the Chinese mobile phone handset design houses discussed in Chapter 2. While the dynamic and variable character of technologies and standards supporting value chain modularity may provide new entry points, major advantages accrue to those actors that actively participate in the rule-setting process, which favors the established actors and locations where the standard setting process is hammered out (Sturgeon 2003). What can governments and non-governmental institutions do in developing countries to try to level the playing field for suppliers without creating barriers to GVC engagement? State and other institutional actors in developing countries can help to fill information and capability gaps by monitoring prevailing trends in information technology

Timothy J. Sturgeon and Greg Linden

223

in general and in supply-chain inter-connect standards specifically, and by developing policies that help local firms implement them in appropriate ways. Policies that focus on capability development in the realm of information technology and standards for supply-chain coordination have several advantages over traditional industrial and technology development policies. First, it is likely to be more effective to characterize, monitor, and help disseminate world-standard information technology than to attempt to foster the basic research and scientific discovery that will lead to endogenous new market and industry creation, which has been a goal of more than a few technology and innovation policies in developing countries. Second, ITC systems for electronic design and supply-chain coordination are relatively industry-independent. Firms from a variety of sectors may use similar systems, although differences certainly exist. This mitigates the practical and political difficulties associated with “picking winners.” Support for better systems for supplychain coordination can benefit a wide range of economic actors in an economy, from farmers to manufacturers to distributors to retailers and other service providers. Third, better coordination in supply chains can provide benefits at the local level, even if global linkages are yet to be made or fail to materialize. An efficiently functional local supply base can be nurtured without the need for international linkages, and once it is in place, it will be more likely to attract orders from global buyers and lead firms. Most important, perhaps, are programs that provide basic support and extend credit to small and medium-sized companies as they struggle to adapt to new conditions and rebound from severe market shocks. GVC theory, as it is presently constituted, will require years of iterative field-testing and refinement before truly robust theoretical tools are in place. In this, metrics are critical, and work is proceeding on several fronts, including the quantification of value creation and capture in specific GVCs (Oikawa, Chapter 6, this volume; Linden et al., 2009; and Dedrick et al., 2010), the collection of a broad range of economic data, such as employment, sourcing locations, and job characteristics according to an exclusive, exhaustive, parsimonious, and generic list of “business functions” (Brown 2008; Nielsen 2008; Sturgeon and Gereffi 2009), and a reexamination of intermediate goods trade (Sturgeon and Memedovic, forthcoming). But equally important, as we have seen in this volume, are the theoretical insights to be gained from detailed field-based research on GVCs in specific industries in specific places.

224 Learning and Earning in Global Value Chains

Notes 1. While this statement is logical enough, there are ample cases of it being difficult for certain firms to simplify products. Japanese electronics companies, for example, have had trouble designing certain products (mobile phone handsets and notebook PCs, specifically) for developing markets, and have turned to contract design companies to fill out the low end of their product lines (Sturgeon 2007). 2. The opportunity for electronic component distribution in Singapore and Malaysia stemmed from the lack of an adequate conduit to connect local chip assembly and test operations with the growing sub-assembly and productlevel manufacturing that foreign firms were doing in the region. Offshore affiliates of both semiconductor and product-level firms had increased their Asian operations, and Uraco’s new distribution arm helped to connect the dots.

References Baldwin, Carliss, and Clark, Kim. 2000. Design Rules, Cambridge, MA: MIT Press. Bell, Martin, and Pavitt, Keith. 1992. “Accumulating Technological Capability in Developing Countries.” Proceedings of the World Bank Annual Conference on Development Economics: pp. 257–281. Bell, Martin, and Pavitt, Keith. 1995. “The Development of Technological Capabilities.” Trade, Technology and International Competitiveness, pp. 69–101. Bell, Martin, and Pavitt, Keith. 1997. “Technological Accumulation and Industrial Growth: Contrasts Between Developed and Developing Countries.” In D. Archingugi and J. Mitchie (eds.) Technology, Globalisation and Economic Performance. Cambridge: Cambridge University Press. Borrus, Michael. 1995. Left for Dead: Asian Production Networks and the Revival of US Electronics. University of California at Berkeley: Berkeley Roundtable on the International Economy Working Paper, p. 100. Brown, Sharon. 2008. “Business processes and business functions; a new way of looking at employment,” Monthly Labor Review (December): 51–70. Business Times (On-line). 1995. “Precision Engineering Group Uraco Plans Expansion in Jahor.” July 14, p. 3. http://www.businesstimes.com.sg Business Times (On-line). 1996a. “Uraco Aims to Boost Revenue to US$400m.” April 22. http://www.businesstimes.com.sg. Business Times (On-line). 1996b. “Uraco Tops Actives with 16.7m Shares Traded.” December 12. http://www.businesstimes.com.sg Business Times (On-line). 1997. “Uraco Managing Director Quits; 3 Executives Replace Him.” January 23. http://www.businesstimes.com.sg. Chesbrough, Henry, and Kusunoki, Ken. 2001. “The Modularity Trap: Innovation, Technology Phase Shifts, and the Resulting Limits of Virtual Organizations.” In Nonaka, I., and Teece, D.J. (eds.) Managing Industrial Knowledge Creation, Transfer, and Utilization. London: Sage, pp. 202–230. Covin, Jeffrey, and D. P. Slevin, Dennis. 1989. “Strategic Management of Small Firms in Hostile and Benign Environments.” Strategic Management Journal 10: 75–87.

Timothy J. Sturgeon and Greg Linden

225

Covin, Jeffrey, and D. P. Slevin, Dennis. 1991. “A Conceptual Model of Entrepreneurship as Firm Behavior.” Entrepreneurship: Theory and Practice 16: 7–24. Dedrick, Jason, Kraemer, Kenneth L., and Linden, Greg. 2010. “Who Profits From Innovation in Global Value Chains?: A Study of the iPod And Notebook PCs,” Industrial and Corporate Change 19(1): 81–116. Dolan, Catherine, and Humphrey, John. 2000. “Governance and trade in fresh vegetables: the impact of UK supermarkets on the African horticulture industry.” Journal of Development Studies 37(2): 147–176. Dupuy, C., Lavigne, S., and Nicet-Chenaf, D. 2010. “Does Geography Still Matter? Evidence on the Portfolio Turnover of Large Equity Investors and Varieties of Capitalism.” Economic Geography 86(1): 75–98. Fixon, Sebastian, and Park, Jin-Kyu. 2008. “The Power of Integrality: Linkages Between Innovation, and Industry Structure.” Research Policy 37: 1296–1316. Galvin, Peter and Morkel, Andre. 2001. ‘The Effect of Product Modularity on Industry Structure: The Case of the World Bicycle Industry.’ Industry and Innovation 8(1): 31–47. Gawer, Annabelle, and Cusumano, Michael. 2002. Platform Leadership: How Intel, Microsoft, and Cisco Drive Industry Innovation. Cambridge, MA: Harvard Business School Press. Geocities, 2004. “Beyonics Technology.” Company profile. http://www.geocities. com/fa_book/Benyonics_181201.html (accessed, January 23, 2004). Gereffi, Gary, Humphrey, John, and Sturgeon, Timothy. 2005. “The Governance of Global Value Chains.” Review of International Political Economy 12(1): 78–104. Humphrey, John and Schmitz, Hubert. 2002. “How Does Insertion in Global Value Chains Affect Upgrading in Industrial Clusters?” Regional Studies 36(9): 1017–1027. Kaplinsky, Raphael, Morris, Mike, and Readman, Jeff. 2002. “The Globalization of Product Markets and Immiserizing Growth: Lessons from the South African Furniture Industry.” World Development 30(7): 1159–1177. Kimura, Seishi. 2007. The Challenges of Late Industrialization: The Global Economy and the Japanese Commercial Aircraft Industry. Basingstoke and New York: Palgrave Macmillan, p. 222. Lall, Sanjaya. 1992. “Technological Capabilities and Industrialization.” World Development 20(2): 165–186. Linden, Greg, Kraemer, Kenneth L., and Dedrick, Jason. 2010. “Who Captures Value in a Global Innovation System? The Case of Apple’s iPod,” Communications of the ACM 52(3): 140–144. Nielsen, Peter (ed.). 2008. International Sourcing – Moving Business Functions Abroad, Statistics Denmark, www.dst.dk/publ/InterSourcing. Perrow, Charles. 2009. “Modeling Firms in the Global Economy.” Theory and Society 38(3): 217–243. Pietrobelli, Carlo, and Rabellotti, Roberta. 2007. Upgrading to Compete: Global Value Chains, Clusters and SMEs in Latin America. Cambridge, MA: Harvard University Press. Sturgeon, Timothy. 2002. “Modular Production Networks. A New American Model of Industrial Organization.” Industrial and Corporate Change 11(3): 451–496.

226 Learning and Earning in Global Value Chains Sturgeon, Timothy. 2003. “What Really Goes on in Silicon Valley? Spatial Clustering Dispersal in Modular Production Networks.” Journal of Economic Geography. April, v.3. Sturgeon, Timothy. 2007. “How Globalization Drives Institutional Diversity: the Japanese Electronics Industry’s Response to Value Chain Modularity.” Journal of East Asian Studies 7(1): 1–34. Sturgeon, Timothy. 2009. “From Commodity Chains to Value Chains: Interdisciplinary Theory Building in an Age of Globalization.” In Bair, Jennifer (ed.) Frontiers of Commodity Chain Research. Palo Alto, CA: Stanford University Press. Sturgeon, Timothy and Gereffi, Gary. 2009. “Measuring success in the global economy: international trade, industrial upgrading, and business function outsourcing in global value chains.” An essay in memory of Sanjaya Lall. Transnational Corporations 18(2): 1–36 (August). Sturgeon, Timothy and Lester, Richard. 2004. “The New Global Supply-base: New Challenges for Local Suppliers in East Asia.” In Shahid Yusuf, Anjum Altaf and Kaoru Nabeshima (eds), Global Production Networking and Technological Change in East Asia. Oxford: Oxford University Press. Chapter 2. Sturgeon, Timothy and Memedovic, Olga. 2011. “Measuring Global Value Chains: Intermediate Goods Trade, Structural Change and Compressed Development,” UNIDO Working Paper, Vienna. Teece, David. 2009. Dynamic Capabilities and Strategic Management: Organizing for Innovation and Growth. New York: Oxford University Press. Teece, David and Gary Pisano. 1994. “The Dynamic Capabilities of Firms: An Introduction.” Industrial and Corporate Change 3(3): 537–556. Teece, David, Gary Pisano, and Amy Shuen. 1997. “Dynamic Capabilities and Strategic Management.” Strategic Management Journal 18(7): 509–533.

Concluding Remarks Momoko Kawakami

Firm-level research is important because, in any society, firms are critical agents of learning and repositories of knowledge. Of course, actions and knowledge are, with only a few exceptions, the domain of human beings, and we need to be careful not to anthropomorphize firms or any other sort of organization. Nevertheless, corporate routines tend to be remarkably durable, even as individuals come and go. What we can say with confidence is that firm-level outcomes—employment and earning, skill development, organizational learning, innovation, value capture, and investment—drive the process of economic development. In this volume, we have explored the development of a handful of local firms in East Asia, with special focus on how their capability development has co-evolved with their engagement with global value chains (GVCs). In so doing, we have tried to keep some complex and even contradictory truths present at all times: that managers and entrepreneurs have the range of choices open to them, that firms in GVCs engage in complex games of cooperation and opportunism that both enable and limit their actions, and that firms do not exist independently, but are embedded in the societies and global industries in which they operate. For local firms based in developing countries, the character of relationships with powerful firms from developed economies is of high strategic importance. Interactions with the lead firms and platform leaders open new market opportunities and provide access to knowledge and opportunities for learning and capability development. But these relationships can also create barriers to learning over time if linkages lock suppliers into low-value segments of the chain. To explore this process in full, we extended the GVC perspective to highlight the co-evolution of the learning processes of latecomer firms and the outsourcing practices of lead firms. 227

228 Concluding Remarks

More specifically, we highlighted the agency of latecomer firms—the critical importance of their strategies and intentional actions—a subject that has so far been neglected in GVC theory. Our field research in East Asia revealed that the strategies of latecomer firms are a main driver of capability development in global industries. Some local firms in East Asia have become serious contenders on the world’s competitive landscape by exploiting the learning opportunities in cross-border transactions. Some participate in multiple value chains and pursue trade with multiple customers to tap into and leverage learning opportunities across different chains, technologies, and product areas. Even for firms operating in the same chains and based in the same country, different strategies to mix market segments and customers, leverage local and global value chains, and intentionally invest in learning have resulted in a diversity of competitive outcomes. This shows that latecomer firms’ purposeful and entrepreneurial actions are among the most important variables determining development outcomes in GVCs. Nevertheless, local firms are constrained by various factors in choosing and pursuing their learning strategies. These factors include subjective variables, such as the technological attributes of transactions; the institutional environment and local setting surrounding the firms; and what are at times stark power asymmetries vis-à-vis lead firms and platform leaders. GVCs are an arena where powerful firms from developed countries and latecomer firms based in developing countries cooperate and compete with each other to create and appropriate the value. Despite the clear power asymmetry between the two parties, our case studies reveal that the strategic and entrepreneurial decisions and actions of latecomer firms can enhance the prospects for capability development in the global value chains, within limits. We hope that the studies compiled in this volume will stimulate new research to understand and codify the strategies that can lead to dynamic learning in GVCs.

Index

Note: The letters ‘f’, ‘n’ and ‘t’ following the locators refer to figures, notes and tables respectively. 5Ss, 115, 121, 133 n.13 Analog Devices (ADI), 51 Apple, 16, 28 ARM, 61, 66 n.19, 217 Asian financial crisis, 107, 112, 140, 186 Asustek, 23, 38–9, 39 n.4 baseband chipset, 45–6, 50–1, 61 Beyonics, 220–2 Bird, 53, 215 brand-carrying firms, 18, 23–4, 32 brand-specific dedicated factories, 122–3 build-to-order systems, 27 business units (of Taiwanese ODMs), 31–2, 35 capabilities adaptive level, 71, 89, 92–4, 116, 118–19, 123–6, 132 n.4 assimilative level, 71, 92, 115–19, 125, 132 n.4, 133 n.14 building, 7, 47, 89–94, 168 deepening, 116, 121 definition of, 5 depth of, 103, 114, 132 n.4 equipment-related, 70, 71, 106, 114–15, 116–19, 125, 133 n.14 formation, 1–13, 17, 102, 106, 113, 118, 125 innovative level, 71, 132 n.4 managerial, 136, 160, 168 marketing, 71, 93, 116 matrix, 11, 71, 72, 73t, 85, 101, 103, 104t, 105t, 113–14, 115f, 121,

123, 125–6, 127t–31t, 132 n.4, n.5 operational level, 71, 116, 118, 132 n.4 planning, 71, 93, 94, 116, 124–5 production, 22, 26, 31, 37, 70, 90, 92–3, 94–6, 103, 114, 126, 161 production management, 71, 90, 92, 106, 114–16, 118–21, 125–6 see also technological capabilities central processing unit (CPU), 16, 25, 36, 61 CEVA, 61, 66 n.19, 217 China Telecom, 51, 53, 217 China Unicom, 51, 53, 217 codifiability of transactions, 4–9, 45, 70, 139–40, 208, 211–12 of technologies, 51 Compal, 23, 29–30 Compaq, 26 complexity of transactions, 4, 6–9, 45, 69, 75, 139–40, 208, 211–12 component consignment, 36 configuration-to-order systems, 27 contract manufacturers, 173, 184, 186, 210, 213, 220–1 electronics manufacturing services, 220 Taiwanese notebook PC, 16, 20, 23, 26, 28, 31 Taiwanese mobile phone handset, 51–4 Dell, 26, 35f dies, molds, jigs and tools (DMJT), 116–18, 126, 133 n.14 digital signal processor (DSP), 61, 66 n.19

229

230 Index direct shipment, 28 diversified customer base, 28–9 economic integration, 1, 167–8 education, 156 electronic design automation (EDA), 61 entrepreneurial alertness, 160 function, 145 orientation (E.O.), 12, 138, 144, 148, 157t, 162, 208 entrepreneurs, 11–12, 65, 138, 140, 144–8, 154, 155t, 156–63, 164n, 207, 227 market-oriented, 145 entrepreneurship, 9, 12, 137–8, 140, 144–7, 164 n.2, 174, 182 explicit coordination, 5, 45, 47, 70, 100–1 export processing zone (EPZ), 142, 162 foreign direct investment (FDI), 1, 74, 78, 80, 136–8, 140–1, 164n, 168–9, 171–2, 186 free trade zone (FTZ), 141, 147, 173 functions non-production, 102–3, 106, 114, 124, 126 production, 103, 113–14, 121 width of, 104t, 105t, 115f, 121–2 gains from trade, 170 global buyers, 43, 101, 167–8, 170, 210, 223 global commodity chain (GCC) perspective, 3, 13 n.1 global system for mobile communication (GSM), 50–2, 63, 216 global value chains approach/perspective, 2, 13 n.2, 17–18, 69–70, 75, 96 n.3, 101–2, 152, 170–1, 213, 227 captive linkages in, 4, 5, 101–2, 125–6, 147, 162–3, 210, 218 definition of, 1, 3 formation, 16

governance, 3–10, 17–19, 38, 44–5, 94–5, 100–2, 139–40, 208–9, 211–12, 216, 222 insertion into, 1, 138–9 and lead firm nationality, 208, 217, 221 modular linkages in, 210, 218 power asymmetry in, 5, 13 n.5, 18, 45, 100–2, 107, 147, 170, 208–9, 211, 228 relational linkages in, 4–5, 13 n.4, 100–1, 209 theory of, 12, 45, 64, 65 n.2, 123, 208, 211, 223, 228 see also value chains Honda, 74–7, 96n.6, 106, 109–10, 111t, 113, 122f, 132 n.6, n.8, 211, 216 Vietnam (HVN), 78–81, 83, 87f, 88, 90, 94–6, 97 n.11, n.18 Hewlett-Packard (HP), 28 human capital, 156 import tariffs, 78, 83 independent design houses (IDHs), 44, 51–2, 54, 56–65, 66 n.12, n.14, n.20 industrial base, 59, 64–5 industrialization export-oriented, 136, 169, 182, 185 TNC-led, 136, 173, 185 information pool, 34–7 input-output method, 169, 175, 185 tables, 12, 169, 175, 177, 179–80, 187t, 192t institutional environment and local settings, 6, 8–9, 228 Intel, 18, 20, 23–6, 28, 30, 32 intermediate imports, 184 intermediate inputs, 174 international division of labor, 17, 168, 170, 178, 210 international value distribution (IVD), 175, 178–86 Inventec, 23, 29 Japan’s lost decade, 183

Index Kanzen, 109f, 110, 122f knowledge flow, 16, 19, 27, 33 latecomer firms, 2, 5, 7–9, 209 lead firms definition of, 2, 4, 19 Japanese, 24, 26, 106–7, 112–13, 115–19, 123, 125–6, 132 n.8, 133 n.9, 213, 218–19 strategies, 4, 6–8, 101, 112–13, 217 learning opportunuities, 2, 7–11, 16, 18, 23, 26, 37–8, 144, 168, 210, 218, 221–2, 228 strategies, 6–9, 11, 19, 28–9, 95 Leontief’s inverse matrix, 176 LG, 55, 56, 214t linkages creation, 137–40, 143–4 formation, 136–7, 139, 140, 153, 163 local content requirement, 150, 184 rules, 74, 78, 80, 82–3 MediaTek, 25, 60–2, 215, 217 Microsoft, 18 Ministry of Information Industry (MII), China, 49, 52 modularity of the product, 25–6, 34 of transactions, 47, 54, 61–2 of value chains, 102, 222 trap, 212 motorcycle Chinese-brand (in Indonesia), 109f, 110, 132 n.7 Chinese (in Vietnam), 68, 76–7, 82–3, 97 n.10 Japanese brand, 68, 74–7, 106–7, 109–10, 112–13 Super Cub, 74, 97 n.18, 106 Wave Alpha, 78–80, 83, 88, 90 motorcycle assemblers, 68, 69, 82–4, 87f, 88, 93–4, 110 Vietnamese, 68, 69, 82–4, 87f, 88, 93–4, 110

231

motorcycle manufacturers Chinese, 75–6, 96 n.7 Japanese, 68–9, 74–9, 93–4, 106, 110, 112–13, 112t, 124–5 motorcycle value chains Japanese/Japanese-led, 79–81, 85, 86t, 88, 90–3, 113, 117, 124–6, 218 Vietnamese-Chinese, 79, 82–5, 86t, 88, 91t, 93–4 Motorola, 51–2, 55, 56f, 66 n.12, 141, 212, 214t, 215–17, 219 Nokia, 51–2, 55, 56f, 59–60, 212, 214t, 215–17 on-the-job training (OJT), 156 original design manufacturers (ODMs) 173 notebook PC, 16, 18, 20, 26–8, 32–4 mobile phone handset, 51–4, 57–8, 63–4 original equipment manufactures (OEM), 20, 22t, 127t–8t, 130t, 147, 173 outsourcing, 20, 26–7, 30, 52–3, 57, 59, 121, 137, 148, 149t, 169, 172, 215, 221 strategies, 6–9, 137, 148, 149t, 153, 164 n.4, 221 performance matrix, 178, 179t platform Centrino, 30 leader, 2, 18, 23, 26, 30, 38, 210–14, 216, 227–8 leadership, 20, 23, 208, 212 Mediatek’s, 60 product, 59 strategy, 37, 59 technology, 46–8, 51–4, 56–7, 59–65 turnkey technology, 61–2, 64 vendors (in mobile phone handsets), 47–8, 51, 53–4, 57, 60–2 policy Bumiputra, 156 FDI, 169 government, 52, 63, 137, 150, 172 industrial, 52, 172–3

232 Index product design architecture, 95 production management, 27, 35, 90, 92, 94, 106, 113–16, 118–19, 122, 125–6 capabilities, 71, 90, 92, 94, 106, 113, 115–16, 118–19, 125–6 production networks, 1, 17, 19, 43, 167–9, 171–2, 174, 178–86, 207 protocol stack, 46, 51 purchasing, 18, 62, 138, 142, 147–54, 156, 163, 173, 210

qualified, 150, 160 replacement, 153–4 second-tier, 80, 88, 92, 113 strategies, 6–7, 84, 95, 100, 102, 106, 114, 121–4, 126, 137, 208, 211–12 supply base, 4, 8–9, 30, 40 n.12, 45, 68–9, 94, 110, 114, 119, 125, 137, 140, 144, 181–2, 184, 218, 223 Suzuki, 74, 76, 87f, 109–10, 111t, 113, 122f, 132 n.6, n.8

Qualcomm, 215–16 quality control (QC), 90, 93, 115, 119, 133 n.13, 150 quality, cost and delivery (QCD), 74–5, 80, 85, 90, 112 Quanta, 23, 28–32, 35f, 39 n.6, 213, 222

TCL, 53, 215 technological capabilities, 5, 11, 29, 46–7, 52, 69, 101, 170, 173–4, 213 see also capabilities Texas Instruments (TI), 51, 60, 66 n.16, 215 Toshiba, 16, 18–19, 26, 40 n.11, n.15, 141 trade theory, 170, 178 transnational corporations (TNCs), 12, 43–4, 46–8, 52–5, 57, 59, 63, 65, 65 n.4, 136–44, 148, 156, 160–1, 163, 164n, 167–8, 170–4, 178, 186 turn-key suppliers, 28

reference designs, 28, 57 repatriation of profits 171 research and development (R&D), 26–7, 31, 33, 90, 93, 96 n.6, 123, 151, 157, 161, 169, 183, 214–15 risk-taking, 145, 158, 161 road map product, 24, 34 technology, 210 Samsung, 55, 56f, 214t Sony Ericsson, 55, 56f, 214t, 217 special economic zone, 139 Spreadtrum, 60–3, 217 standards de facto, 51, 75, 83–4, 95, 106–7, 211, 216, 218 second-generation mobile phone, 47, 50–2, 65 n.9 third-generation mobile phone, 216 supplier capabilities, 4, 6–7, 11, 29, 31, 37, 89–94, 100–1, 106, 124, 126, 218, 221 development, 7, 37, 124 suppliers Chinese, 83–4, 88 first-tier, 38, 80, 88, 90, 92, 94, 110, 111t, 113, 117 Japanese, 74–5, 79–80, 107, 153

upgrading functional, 11, 102–3, 121, 132 n.1 process, 103, 132 n.1 product, 102, 132 n.1 technological, 157, 159–61, 169, 172 value-added, 169–70 distribution, 181, 184 outflow of, 171, 178, 183, 185 value chains captive, 4–5, 11, 70, 75, 79, 94–5, 100–2, 106–7, 113–16, 118, 121, 123–6 dynamics, 5, 9, 11, 38, 45, 50, 63–4, 79, 84 function, 25, 85, 93, 103, 207 governance, 5t, 6, 8, 13 n.4, 13 n.7, 38, 95, 100–2, 132 n.2, 208 local, 64 market-based, 75, 79

Index

233

modular, 13 n.8, 34, 100, 124

Wistron, 23, 39 n.4, 40 n.13

multiple, 8, 95, 228

Yamaha, 74, 76, 77f, 92–3, 97 n.11, 109–10, 111t, 113, 122f, 132 n.6, n.8 Vietnam, 76, 77f, 87f, 92–3, 97 n.11

participation, 85, 88–9, 95 relational, 4, 13 n.4, 34, 70, 75, 100–2 strategies, 62 see also global value chains

ZTE, 57, 66 n.12