Clusters and economic growth in Asia 9780857930095, 9780857930088, 0857930087, 0857930095

1. Cluster policies and entrepreneurial states in East Asia / Alexander Ebner -- 2. Information and communication techno

461 19 3MB

English Pages 208 pages [201] Year 2013

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

Clusters and economic growth in Asia
 9780857930095, 9780857930088, 0857930087, 0857930095

Table of contents :
1. Cluster policies and entrepreneurial states in East Asia / Alexander Ebner --
2. Information and communication technology and economic growth of four Asian industrialized economies / Yanfei Li and Wai-Mun Chia --
3. Industrial agglomeration of Taiwanese electronics firms in Dongguan, China : home effects and implications for industrial upgrading / Felix Haifeng Liao, Karen Zhihua Xu and Bin Liang --
4. The rise of the biomedical cluster in Wonju, Korea / Jun Koo and Jongmin Choi --
5. The global economic crisis as leverage for emerging regional growth paths? : differentiated evidence from China three years onwards / Daniel Schiller and Henning Kroll --
6. Technological intensity of FDI in Vietnam : implications for future economic development and emerging clusters / Curt Nestor --
7. The aircraft industry as a tool for economic and industrial development : the case of Indonesia / Sören Eriksson --
8. Foreign knowledge transfer in the development of aircraft industry clusters : the case of Chengdu, China / Sören Eriksson.

Citation preview

Clusters and Economic Growth in Asia

ERIKSSON 9780857930088 PRINT.indd i

29/01/2013 16:59

ERIKSSON 9780857930088 PRINT.indd ii

29/01/2013 16:59

Clusters and Economic Growth in Asia Edited by

Sören Eriksson Jönköping International Business School, Jönköping University, Sweden

Edward Elgar Cheltenham, UK • Northampton, MA, USA

ERIKSSON 9780857930088 PRINT.indd iii

29/01/2013 16:59

© Sören Eriksson 2013 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2012951751 This book is available electronically in the ElgarOnline.com Economics Subject Collection, E-ISBN 978 0 85793 009 5

ISBN 978 0 85793 008 8

03

Typeset by Servis Filmsetting Ltd, Stockport, Cheshire Printed and bound by MPG Books Group, UK

ERIKSSON 9780857930088 PRINT.indd iv

29/01/2013 16:59

Contents List of contributors Preface

vi ix

1

1

2

3

4 5

6

7

8

Cluster policies and entrepreneurial states in East Asia Alexander Ebner Information and communication technology and economic growth of four Asian industrialized economies Yanfei Li and Wai-Mun Chia Industrial agglomeration of Taiwanese electronics firms in Dongguan, China: home effects and implications for industrial upgrading Felix Haifeng Liao, Karen Zhihua Xu and Bin Liang The rise of the biomedical cluster in Wonju, Korea Jun Koo and Jongmin Choi The global economic crisis as leverage for emerging regional growth paths? Differentiated evidence from China – three years onwards Daniel Schiller and Henning Kroll Technological intensity of FDI in Vietnam – implications for future economic development and emerging clusters Curt Nestor The aircraft industry as a tool for economic and industrial development – the case of Indonesia Sören Eriksson Foreign knowledge transfer in the development of aircraft industry clusters – the case of Chengdu, China Sören Eriksson

Index

21

40 66

85

119

141

165

183

v

ERIKSSON 9780857930088 PRINT.indd v

29/01/2013 16:59

Contributors Wai-Mun Chia obtained her Bachelor’s degree in economics from the University of London and then pursued her Master’s degree at the London School of Economics (LSE) with scholarship. She received a PhD in economics from Nanyang Technological University (NTU) in Singapore and is currently Assistant Professor at NTU. She is an Assistant Editor for the Singapore Economic Review. Her research interests are international macroeconomics and economic integration in East Asia. She has numerous publications in international scientific journals. Jongmin Choi is a doctoral student in the Department of Public Policy at the University of North Carolina at Chapel Hill. He received a Master’s degree in public administration at Korea University. He has a keen interest in industry cluster, science and technology policy, and urban development. Alexander Ebner is Professor of Political Economy and Economic Sociology as well as Director of the Schumpeter Center for Clusters, Innovation and Public Policy at Goethe University in Frankfurt am Main, Germany. Previously, he was an Associate Professor of Political Economy at Jacobs University Bremen. Previous research affiliations include the Institute of Southeast Asian Studies in Singapore. Alexander Ebner’s research interests involve the matters of entrepreneurship, innovation, governance and regional economic development. Sören Eriksson is a Professor of Economic Geography at Jönköping International Business School, Sweden. His research activities focus on technology diffusion, globalization processes, logistics issues and regional economic development. He is an authority on East and Southeast Asia’s geography and an expert on the international aerospace industry. He has lectured, conducted seminars and been appointed as external reader and opponent of doctoral dissertations at a number of foreign universities and research establishments in Africa, Asia, Europe and North America. Jun Koo is an Associate Professor in the Department of Public Administration at Korea University. He holds a doctorate degree in city and regional planning from the University of North Carolina at Chapel Hill. Before joining Korea University he worked for the World Bank and

vi

ERIKSSON 9780857930088 PRINT.indd vi

29/01/2013 16:59

Contributors

vii

taught at Cleveland State University. He has diverse research interests including innovation and entrepreneurship, industry cluster and urban development. His research has appeared in many major international journals on regional science, management and public administration. Henning Kroll is an economic geographer and a researcher at the Fraunhofer Institute for Systems and Innovation Research in Karlsruhe, Germany. His research interests include the analysis of national and regional innovation systems in Germany, Europe and Asia, the development of regional innovation indicators, as well as the assessment of regional innovation and technology policies. Recently, he has been working on projects for the municipal governments in both Northern and Southern China. Yanfei Li received his BA in economics from Peking University, China, and a PhD in economics from Nanyang Technological University (NTU), Singapore. He is currently a Research Fellow at NTU, where he conducts research in economics of technological change and Asian economies, serving both academic and consulting constituents. He has several papers and book chapters published in Emerald and Elsevier journals and books and with Economic Research Institute for ASEAN and East Asia (ERIA). Bin Liang is currently a graduate student in the Department of Family and Consumer Studies at the University of Utah. Her research interests include globalization, housing, migration, urban planning, transportation and public health, with a regional focus on China and the United States. Felix Haifeng Liao is currently a PhD candidate in the Department of Geography at the University of Utah. His research interests include economic/urban geography, regional development, globalization, industrial location, GIS and spatial statistics, with a regional focus on China and the United States. Curt Nestor has a PhD in economic geography and is Senior Lecturer at the School of Business, Economics and Law at the University of Gothenburg, Sweden. His research interests and publications focus on Vietnamese economic development, trade and foreign investment flows, and regional economic integration. Daniel Schiller is an economic geographer and a researcher at the Lower Saxony Institute for Economic Research (NIW) in Hannover, Germany. His research interests include knowledge-based regional development, institutions and governance, higher education systems and public finance, with a regional focus on Europe and Asia (especially Thailand and China).

ERIKSSON 9780857930088 PRINT.indd vii

29/01/2013 16:59

viii

Clusters and economic growth in Asia

Recently, he has been working on projects about the spatial and organizational transition of the electronics industry in Hong Kong and the Pearl River Delta. Karen Zhihua Xu graduated from the Department of Urban Planning and Design of the University of Hong Kong. Her research interests include regional development in China, foreign investment, and the cooperation and interaction between Guangdong and Hong Kong. She is currently Assistant Director of the Advanced Institute for Contemporary China Studies, Hong Kong Baptist University.

ERIKSSON 9780857930088 PRINT.indd viii

29/01/2013 16:59

Preface The book is based on invited papers from scholars with the common research interest in economic growth and cluster development in East and Southeast Asia. These issues have attracted considerable attention in recent years, although compared with some other parts of the world there is a limited choice of literature dealing with Asia. A clear ambition was to invite authors who were able to contribute with new and interesting dimensions about clusters and economic growth. Hopefully this book will not only add to the existing literature, but also create new questions and thoughts about this increasingly important part of the world. The first chapter by Alexander Ebner deals with the increasing relevance of cluster policies and the need to understand them in the context of an ongoing institutional and structural transition of East Asian newly industrializing economies towards an innovation-driven pattern of development. In this context, the national institutional frameworks are subject to changes that involve the transformation of the model of the ‘developmental state’ towards specific kinds of ‘entrepreneurial states’. Chapter 2 by Yanfei Li and Wai-Mun Chia investigates the role of information and communication technology (ICT) in economic growth since the late 1990s. It follows the growth accounting model to analyse the role of ICT in economic growth in four Asian economies: Japan, Hong Kong, South Korea and Singapore. The study also implies the possibility that ICT development could be a source of potential convergence between Asian newly industrializing economies and economies such as the USA and Japan. In Chapter 3 Felix Haifeng Liao, Karen Zhihua Xu and Bin Liang explore the industrial agglomeration of Taiwanese electronics firms in 32 towns and districts in Dongguan, China. Over the past two decades, the industrial agglomeration of Taiwanese electronics investment in Mainland China has resulted in some electronics clusters. Based on firm-level interviews and statistics this chapter also has important policy implications for the upgrading of clusters in developing countries. The rise of the biomedical cluster in Wonju, South Korea, provides the subject for Jun Koo and Jongmin Choi in Chapter 4. This study aims at achieving two things. First, the authors try to unpack the cluster

ix

ERIKSSON 9780857930088 PRINT.indd ix

29/01/2013 16:59

x

Clusters and economic growth in Asia

development process to better understand factors that play a role in the take-off stage of an industry cluster. Second, the focus on the biomedical cluster was chosen as there has been a dearth of literature on such clusters in Asian countries. The most important finding is that the presence of successful venture firms in the early cluster development stage can play a pivotal role in the growth of clusters. The contribution by Henning Kroll and Daniel Schiller, in Chapter 5, presents a critical discussion and analysis regarding Chinese growth models. Quite evidently there have been a number of different sectoral and regional growth models in China before the slowdown in the world economy. The authors argue that we are in need of a differentiated understanding of the impact that the crisis had on different drivers of growth in China. At the end of the 1980s, Vietnam embarked on an ambitious economic reform programme with the aim of promoting economic development. The foreign-invested sector has made contributions to average GDP growth rates, exceeding 7 per cent over the period. In Chapter 6, Curt Nestor examines the technology intensity of FDI in Vietnam and the implications for economic growth and emerging clusters. Finally, proposals for future industrial cluster policies and economic development are discussed. For a number of reasons, an increasing number of developing countries have tried to build up an internationally competitive aircraft industry. During Suharto’s rule the establishment of a domestic aircraft manufacturing industry was the largest and most ambitious investment to promote technology development in Indonesia. Chapter 7 by Sören Eriksson investigates the main factors behind the long-term failure and discusses critical arguments against creating this kind of industry for the purpose of economic and industrial growth. Already in the 1980s statements were made that aircraft production would be an important industry in China’s new stages of economic and industrial growth. The government also expressed the interest in and ambition to develop aircraft-related clusters. In Chapter 8, Sören Eriksson investigates the origin and characteristics of foreign technology transfer into Chengdu, one of the country’s most important locations for the aircraft manufacturing industry. I would like to acknowledge my sincere appreciation to all authors who have contributed their knowledge, time and support to this book. Sören Eriksson Professor of Economic Geography Jönköping International Business School Sweden

ERIKSSON 9780857930088 PRINT.indd x

29/01/2013 16:59

1.

Cluster policies and entrepreneurial states in East Asia Alexander Ebner

INTRODUCTION Cluster policies aim to activate and sustain the competitive interaction of firms in local and regional business agglomerations. Policy instruments tend to augment market forces by providing distinct types of collective goods. As such, cluster policies differ markedly from traditional types of industrial policy that highlight the nationwide targeting of particular firms and industries by means of market intervention. Still, the logic of cluster policies is most convincingly derived from the persistent relevance of national institutional frameworks, most prominently involving nationstates, and their ongoing transformation in the process of economic development. This line of reasoning is most appropriately exemplified by the East Asian development experience. Indeed, it may be argued that the increasing relevance of cluster policies in East Asia parallels the advent of a new model of government–business relations that may be labelled ‘entrepreneurial state’. This concept suggests that entrepreneurial aspects of state activity, which were already prevalent within the East Asian developmental states, currently turn out as dominant policy features, thus changing the dominant rationale of government towards an entrepreneurial direction, implying a shift from the developmental assimilation of technological novelties in catch-up growth to their entrepreneurial creation in a setting that allows for technological leadership. The related policy rationale promotes innovation as the source of international competitiveness, framed by a multi-level architecture of governance that strengthens a regionalized type of industrial policies, which points to the formation of cluster policies. Therefore, in examining this relationship among clusters, cluster policies and the advent of the entrepreneurial state in East Asia, the following explorations proceed in three sections. First, the matter of cluster policies and the role of the state in the promotion of clusters are brought to the fore. The discussion highlights the Porterian cluster approach and its 1

ERIKSSON 9780857930088 PRINT.indd 1

29/01/2013 16:59

2

Clusters and economic growth in Asia

policy implications, underlining the impact of the national institutional framework on the actual orientation of cluster policies. The second section then takes on the transformation of the East Asian developmental states and their interventionist industrial policies. As the process of catch-up growth proceeds, new types of state functions arise that are well summarized under the label of entrepreneurial states. Corresponding changes in government–business relations allow for the promotion of cluster policy as a new kind of multi-scalar approach to industrial policy. Thus, cluster policies are an extension of the advent of the entrepreneurial state. The third section illustrates these arguments by pinpointing recent efforts in East Asian cluster policies.

CLUSTERS, CLUSTER POLICIES AND THE ROLE OF THE STATE The competitive advantages of firm-specific interactions within a particular regional setting of industries and institutions are usually addressed in terms of industrial clusters. It is a widely shared insight that industrial clusters serve as the backbone of regional competitiveness. This implies that related approaches to the analysis of clusters provide conceptually sound, empirically significant and politically viable research perspectives. To some, however, the concept of clusters is still controversial (Benneworth et al., 2003; Martin and Sunley, 2003; Benneworth and Henry, 2004). A paradigmatic definition by Michael Porter defines clusters as ‘a geographically proximate group of interconnected companies and associated institutions in a particular field, linked by commonalities and complementarities’ (Porter, 1998, p. 199). Cluster dynamics are shaped by the competitive conditions of firms, namely factor supply and demand profile conditions, and the industrial structure in related and supporting industries, as well as firm strategy and structure. The underlying relationships that form a distinct cluster within a national economy are either of the vertical type that links buyers and suppliers, or of the horizontal type that links common customers, technologies and distribution channels – while the interchange among industries in a cluster is best organized in geographical agglomerations (Porter, 1990, pp. 149, 157). This means, in the Porterian framework, that regional development with its comprehensive innovation, income and employment effects is driven by the dynamic constellations of industrial clusters (Porter, 2000, 2003). Yet Porter’s approach has been repeatedly criticized for its somewhat mechanistic, structurally oriented cluster concept, which essentially implies that as long as all actors deemed necessary are present in a region, a cluster with all associated benefits is

ERIKSSON 9780857930088 PRINT.indd 2

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia

3

likely to emerge. Accordingly, the most pressing research challenges to the Porterian approach to the ‘microeconomics of competitiveness’ focus on the institutional and structural match between company sophistication and the related business environment (Ketels, 2006). Still, Porter’s arguments are said to neglect the institutional substance of clusters, that is, their social structuration, their organizational outlook and the related logic of complementarity and coherence (Steinle et al., 2007). A more interaction-oriented perspective developed in parallel to Porter’s work, with authors mainly rooted in the preceding Marshallian tradition of industrial district research (Becattini, 1991). This has been complemented by research on the ‘innovative milieu’ of interconnected firms in dynamic regions (Crevoisier, 2004). An innovative milieu can be defined as ‘the set of relationships that occur within a given geographical area that bring unity to a production system, economic actors, and industrial culture, that generate a localized dynamic process of collective learning and that act as an uncertainty-reducing mechanism in the innovation process’ (Camagni, 1995, p. 320). In these views, local culture plays an important role in cluster formation, with a particular form of collaboration and competition being made possible by a common socialization and a common ideal of regional allegiance. Consequently, institutional networks and their impact on cluster dynamics have been assessed more prominently, for clusters contain inter-organizational networks that are indispensable for generating and disseminating knowledge and innovations (Bergmann et al., 2001; Visser, 2009). In this manner, clusters may be interpreted as structures of co-located industry insiders that engage in flexible modes of experimentation with distinct network arrangements within and among clusters. This implies that the organization of learning processes becomes a most decisive strategic aspect of economic development (Malmberg and Maskell, 2002; Maskell and Lorenzen, 2004). A delicate balance between competition and cooperation among firms is a necessary feature of this constellation, as the interlinking of cooperative partnerships is strategically important to capturing the benefits of learning-based competitiveness (Asheim, 2007). Thus, concepts such as the ‘learning region’ correspond with a Porterian cluster structure, which is augmented by the institutional architecture of regional coalitions for learning and innovation (Polenske, 2008). In this line of reasoning, the region is viewed as a geo-institutional set of socioeconomic resources and relations, involving components such as human capital and production routines. Spatial proximity matters, too. It enhances the competitiveness of firms by facilitating interpersonal processes of learning and innovation, which tend to reduce transaction costs by establishing common symbols and codes (Maskell and Malmberg, 1999). Crucially, then, the dynamics of

ERIKSSON 9780857930088 PRINT.indd 3

29/01/2013 16:59

4

Clusters and economic growth in Asia

economic development are determined by the structure of innovation networks with their systemic linkages among knowledge-producing organizations such as universities, intermediary organizations such as government agencies, and the regional set of industrial clusters with its profile of both small and large firms (Cooke, 1998; Cooke and Schienstock, 2000). In addition to that, the assessment of the developmental dynamics of clusters also requires a reconsideration of the external linkages of the involved firms and related organizations, quite in line with the overall developmental pattern of an increasing openness of clusters (Giuliani et al., 2005). The importance of non-regional networks is decisive for the absorption of new technologies and organizational practices. The scope of these strategic interactions contributes to various degrees of external economies and increasing returns in an evolving setting of organizational as well as territorial modularity (Whitford and Potter, 2007). Accordingly, the external linkages of cluster firms in learning regions serve as systemic carriers of knowledge transfers and learning effects. They support the systemic openness of clusters and thus tend to obstruct an institutional and technological lock-in of development trajectories by promoting adaptive flexibility, an aspect that becomes paramount when the cluster life cycle reaches maturity (De Martino et al., 2006; Zucchella, 2006; Menzel and Fornahl, 2010). Thus, the availability of external partners for innovation is paramount in furthering the openness of clusters. Apart from ‘local buzz’ and localized capabilities, the requirement for knowledge exchange leads to a reconsideration of ‘global pipelines’ in cross-cluster knowledge flows (Bathelt et al., 2004; Maskell et al., 2006). The underlying capability to integrate new knowledge into local routines depends on complementarities with established routines and skills, for pieces of knowledge originating in a context too far away from the recipient may be difficult to absorb (Loasby, 2001). In summary, these considerations on cluster dynamics acknowledge their multi-scalar structuration, which is reflected in the multi-level governance structures of internal and external cluster linkages. Such a perspective implies the need for a more elaborate differentiation of external linkages, thus transcending the simple dichotomy of the local versus the non-local by addressing issues such as network interactions on different levels and scales (Lagendijk and Oinas, 2005; Ebner and Schiele, 2012). Indeed, the evolution of the competitive capabilities of cluster firms and related organizations is subject to local, national and international interactions (Hassink, 2005; Whitley, 2007). In this context, the national level of the business environment still stands out in shaping the routines and practices of cluster firms (Gertler, 2001). At this point, the role of the state needs to be taken into account as an institutional force that shapes the economic dynamics of clusters by

ERIKSSON 9780857930088 PRINT.indd 4

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia

5

means of cluster policies. Indeed, the state matters first of all as a provider of regulatory standards and rules of the diverse national administrative and legal subsystems. Also, informal institutions such as social norms and cognitive models that are said to constitute a cultural setting are shaped by governmental activities. As Robert Wade put it during the heyday of the globalization controversy, ‘National boundaries demarcate the nationally specific systems of education, finance, corporate management and government that generate social conventions, norms, and laws and thereby pervasively influence investment in technology and entrepreneurship’ (Wade, 1996, p. 85). Accordingly, in the setting of local, national and global linkages, the institutional specificity of the national level may be assessed as a dominant factor in the external interaction of cluster firms – despite the fact that the national level is mainly absent in the established discourse on knowledge spillovers within and across cluster boundaries (Lundvall and Maskell, 2000; Maskell, 2001; Isaksen, 2009). This basic assessment is well reiterated in Porter’s notion of the ‘competitive advantage of nations’ that suggests that competitive industrial clusters mirror distinct advantages that are rooted in the historically evolving institutional and structural features of national economies. Porter addresses the persisting role of the national business environment as follows: ‘Competitive advantage is created and sustained through a highly localized process. Differences in national economic structures, values, cultures, institutions and histories contribute profoundly to competitive success’ (Porter, 1990, p. 19). The corresponding national innovative capacity with its interactions among firms, research institutes, universities and other innovation-oriented players reflects specialization patterns that are derived from interlinked factor conditions such as skilled human resources, adequate R&D endowments and an efficient financial system (Furman et al., 2002). In this context, Porterian cluster policy puts the private sector in the focus of proactive efforts in industrial upgrading. Corresponding prescriptions may be synthesized as follows: first, policy support should be available for all productive clusters, involving both domestic and foreign companies; second, existing clusters with their linkages and complementarities across industries should be the basis for further refinement and upgrading, while attempts to create entirely new clusters are problematic; third, cluster initiatives should be advanced by the private sector, while government serves as a facilitator; fourth, policy strategies should be designed in a bottom-up manner that allows for deliberation among all stakeholders on various policy levels, in particular the local level. It follows: ‘Enhancing cluster externalities and spillovers will increase productivity and prosperity in any cluster. Hence government should not choose among clusters but create policies that support

ERIKSSON 9780857930088 PRINT.indd 5

29/01/2013 16:59

6

Clusters and economic growth in Asia

upgrading in every cluster present in a location’ (Porter, 2007, p. 6). The latter argument emphasizes that the Porterian cluster approach underlines a dynamic understanding of competition as a positive sum game – with productivity as a key policy concern (Porter, 1998, pp. 248–9). This perspective differs markedly from those traditional types of industrial policy that highlight the national targeting of particular firms and industries, based on government interventions through subsidies, protective measures and related means, which alter the results of market competition (Woodward and Guimaraes, 2009). In summary, Porter’s concern with regional agglomerations of cluster firms mirrors both the multi-scalar and multi-level qualities of the innovation-driven process of economic development. Porter’s recent emphasis on the role of clusters as export-oriented agglomerations with distinct external linkages points in this direction (Simmie, 2008). This line of reasoning goes well together with the neo-Schumpeterian systems of innovation framework and its proposition that industrial structures, the institutional set-up of a national economy and its policy orientation stand out in determining the innovation performance of firms and industries, thus complementing regional and supranational constellations (Freeman, 2002; Lundvall et al., 2002). The national level matters with regard to learning and innovation, because the policies of national governments, national laws and a shared culture delineate an institutional arena that affects the intensity and direction of innovation (Nelson and Rosenberg, 1993; Lundvall, 1998). This persistent relevance of national institutional frameworks shapes the developmental trajectory of whole economies and thus plays a key role in the formation of policy approaches to support industrial clusters – as exemplified most appropriately by the East Asian development experience. The key question is whether the increasing relevance of cluster policies goes well together with an institutional differentiation of the state and the related national setting. The following section argues that cluster policies in East Asia parallel the advent of a new model of government–business relations that may be labelled ‘entrepreneurial state’.

TOWARDS ENTREPRENEURIAL STATES IN EAST ASIA The historically specific developmental impact of government–business relations in East Asian economies is subject to persistent discussions that have been most prominently fuelled by the World Bank’s 1993 policy research report on the ‘East Asian Miracle’. Capital accumulation,

ERIKSSON 9780857930088 PRINT.indd 6

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia

7

resource allocation and technological catch-up are identified as functions of economic growth, which have been promoted by a mixture of competitive market processes and supporting public policies (World Bank, 1993, pp. 10–11). Growth-promoting market interventions in the domain of industrial policy are addressed as components of these public policies with clear cost–benefit considerations and performance criteria (ibid., pp. 5–8). These considerations point to further discussions on the role of the state in East Asian economic development. A key issue has been the concept of the ‘developmental state’, as informed by Chalmers Johnson’s research on Japanese industrial policy. Johnson maintains that the regulatory function of states in Western economies that pioneered industrialization focuses on rules governing the economic process, whereas states in late industrializing economies, such as Japan, exhibit a developmental function, as they administratively guide industries and markets (Johnson, 1982, pp. 19–20). Yet, developmental states exhibit a transitory character, for the notion of the developmental state covers only a fraction of state functions. The functional priorities of states thus determine their institutional essence while following situational imperatives (ibid., pp. 305–7). The developmental imperative of catch-up growth refers to the role of the state in late industrialization, perceived as a process that is based on gradual upgrading and learning how to improve technology already in use abroad (Amsden, 1989, pp. 3–4). This process of state-guided adaptive technological learning in late industrialization may face stagnation as soon as the technology frontier is approached without the formation of local innovation capabilities (Amsden and Hikino, 1993, p. 259). Thus, the transitory character of the developmental state reflects its relative success in moving the economy towards the technological frontier. The notion of ‘governed markets’ addresses corresponding attempts in leading the market by political means, which then aim at stimulating innovation in the private sector. Governing markets thus requires both institutional capacity and shared knowledge (Wade, 1990, pp. 28–9). State capacity serves as the institutional basis for the corresponding policy strategies of developmental states, which foster entrepreneurial perspectives in the long term by promoting transformative investments and lowering associated risks. ‘Embedded autonomy’ then marks the internal organization of developmental states and their capacity for promoting industrial transformation (Evans, 1995, p. 12). However, the results of this transformation feed back upon the state itself, for the actors that emerge from the policyrelated state interventions tend to recreate the underlying conditions of their activity – which is most relevant in terms of the shifting balance of powers among the social forces and their political articulation. Successful industrial transformation makes industrial capital less dependent on the

ERIKSSON 9780857930088 PRINT.indd 7

29/01/2013 16:59

8

Clusters and economic growth in Asia

state, thus allowing for a reconfiguration of government–business networks. Therefore, the reconstructive self-transformation of the transitory developmental state mirrors an increasing complexity of the socioeconomic domain (ibid., pp. 226–34). This line of reasoning refocuses on the strategic interdependence between government and business. As exemplified by East Asian development, corresponding modes of governance involve a ‘catalytic state’, usually acting in cooperation with the private sector while exercising negotiated leadership in the coordination of policy networks that support technological upgrading and innovation (Weiss, 1998, p. 67). Transformative capacity then implies that government– business cooperation is subject to adaptation over time. Accordingly, the East Asian developmental state is subject to a country-specific transformation with state capacity approaching a less hierarchical mode of coordination that relates to ongoing changes in the economic setting (ibid., pp. 64–5). Thus, the developmental motive of catch-up growth is gradually replaced by a strategic concern with continuous technological upgrading in an internationalizing competitive setting (Weiss, 2000, p. 22). Echoing these concerns, more recent World Bank policy discussions on East Asian development highlight the promotion of innovation as means for enhancing productivity, based on strengthening public–private interactions, local coherence and international connectedness, while claiming that major policy challenges relate to how East Asian countries cultivate creativity within their economies (Yusuf et al., 2003, p. 29). Therefore, the articulation, intensity and content of entrepreneurial effort becomes ever more knowledge- and science-intensive in approaching the technological frontier, building on established capabilities that are embedded in nationspecific institutional frameworks (Lall, 2000, p. 14). In addressing these tendencies, the theory of the developmental state has become subject to various modifications. For instance, it is argued that the developmental state undergoes a transformation towards a new rationale in coping with staying ahead of or keeping up with international competitors, in particular by assisting in industrial restructuring. A more gradual and continuous mode of upgrading skills and technologies is at stake – as witnessed by the maturing of Japanese and Taiwanese industries whose restructuring is guided by strategic policies that resemble the rationale of a ‘transformative state’ (Weiss, 2000, pp. 27–9). Related arguments pinpoint the ideal type of ‘transitional developmental state’ that allows for a transition from interventionism to liberalization – which need not entail a retreat of the state but even its strengthening with regard to the enforcement of the market order (Wong and Ng, 2001, pp. 43–7). In associated terms, developmental and regulatory state functions are differentiated. The ‘neo-developmental state’ for high-tech industries copes with the promotion of competitive

ERIKSSON 9780857930088 PRINT.indd 8

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia

Table 1.1

9

A typology of state functions and industrial policies Type of State

Policy rationale Ideology Governance mode Innovation style Policy scale

Regulatory

Developmental

Resource coordination Market liberalism Rule-based Hierarchical Commercialization National

Factor mobilization Innovation Developmentalism Interventionist Hierarchical Assimilation National

Entrepreneurial

Entrepreneurialism Communicative Horizontal Creation Multi-scalar

economies of scale, the support of industrial R&D and employment creation in industrial change, complemented by the ‘regulatory state’ for liberalized services, governing competition and international openness (Amsden and Chu, 2003, pp. 167–72). Moreover, the rationale of government shifts towards locational policies in support of industrial networks and technology-intensive interactions – resembling a pattern of state-led networking (ibid., pp. 15–16). Clusters and cluster policy thus become key issues in this institutional transformation of the state and its policy concerns, which may be interpreted in terms of an ‘entrepreneurial state’. Indeed, the concern with entrepreneurship in the creation, modification and adaptation of technological and organizational innovations resembles a distinct set of state functions, which requires a conceptual framework of its own: the entrepreneurial state. The underlying reasoning suggests that entrepreneurial aspects of state activity that were already prevalent with the East Asian developmental states currently turn out as dominant policy features, thus changing the dominant rationale of government towards an entrepreneurial direction that implies a shift from the developmental assimilation of technological novelties to their entrepreneurial creation. The innovation capacity of the entrepreneurial state addresses the potential for exercising policies that promote innovation on an economy-wide scale, either by direct interventions in the economic process or by conditioning through institutional and physical infrastructures (Ebner, 2007, pp. 118–19). Ideal typically, it may be argued that three sets of state functions shape the process of economic development, as outlined in Table 1.1. They are simultaneously present, yet the overall outlook of the state will depend on the hegemonic type of function, which is subject to historically specific variation in the development process. The commercialization of technology resembles the operations of a regulatory state, typical for the model of liberal market economies. The

ERIKSSON 9780857930088 PRINT.indd 9

29/01/2013 16:59

10

Clusters and economic growth in Asia

policy rationale of regulatory states highlights resource coordination through an institutional enforcement of the market order. Accordingly, industrial policies of regulatory states emphasize the guidance of market forces in the innovation process, which may be termed as an innovation style of commercialization. In contrast to that, the developmental state, which has been prevalent in East Asia, combines its policy rationale of factor mobilization with long-run goals of national development. The concern with entrepreneurship in the generation of technological innovations resembles a distinct set of state functions, which is represented by the notion of the entrepreneurial state. It points to recent transformations of the state all over the OECD countries – and beyond. Its policy rationale promotes innovation as the source of international competitiveness, framed by a multi-level architecture of governance that strengthens a regionalized type of industrial policies, which involves distinct cluster policies (Ebner, 2009, pp. 382–3). Furthermore, the scalar policy dimension of the entrepreneurial state is more differentiated than that of the regulatory or developmental types. Indeed, the entrepreneurial state transforms the national policy range towards a multi-scalar setting that strengthens regional interactions and thus paves the way for distinct cluster policies. In summary, the notion of the entrepreneurial state entails the following set of preliminary propositions that may be subject to further scrutiny (ibid.): ●





The concern with the formation of entrepreneurial capacity in the generation and carrying out of innovations becomes a crucial feature of industrial policy. This involves both direct measures that include selective interventions in the market process as well as indirect measures involving the moulding of formal and informal institutions that shape innovation efforts of the private and public sectors alike. Policy efforts shift from catching up within an established technological paradigm to a rationale of paradigm creation that involves a potential for technological leadership in an uncertain environment. State capacity remains crucial for mediating between interest groups and for communicating broad developmental orientations. The reorientation of policy concerns towards self-sustaining knowledge-based interactions in promoting competitiveness coevolves with an institutional transformation of government and administration. Framed by a common discourse on entrepreneurship and innovation, governance structures evolve as complex policy networks.

ERIKSSON 9780857930088 PRINT.indd 10

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia ●





11

The institutional architecture of the entrepreneurial state underlines the role of knowledge transfers and communication in state–society relations that form reflexive policy modes. Experimentation, learning and innovation characterize the paradigmatic efforts in government and administration for problem-solving efforts. The policy rationale of entrepreneurial states reflects the diversity of structural, institutional and spatial patterns of globalization, highlighting the combination of global production networks and local clusters of innovation activities. The spatial dimension of innovation becomes a crucial component of industrial policy. The policy orientation of entrepreneurial states combines international competitiveness, capability-building and locational strategies that address the entrepreneurial orientation of both local and foreign firms. The formation of knowledge-based clusters becomes a key element of related policy efforts.

Thus, stated in terms of the Schumpeterian theory of economic development, the institutional dynamism of the entrepreneurial state reflects the co-evolution of state and market in the process of economic development (Ebner, 2006, pp. 511–12). In this line of reasoning, the logic of cluster policies reflects the transformation of East Asian newly industrialized economies towards an innovation-driven pattern of development.

CLUSTER POLICIES AND THE ‘EAST ASIAN RENAISSANCE’ Following the decades of high performance growth from the 1970s to the 1990s, the East Asian emerging economies have been witnessing the challenge of the Asian financial crisis of 1997 and its aftermath. While some observers had argued that this crisis would actually herald the end of East Asian catch-up growth, the empirical situation evolved in a different direction. Indeed, fuelled by the growth dynamics of the Chinese economy and supported by a reconstruction of transnational business networks within the East Asian division of labour, former ‘tiger economies’ such as South Korea, Taiwan and Singapore have regained most of their developmental strengths while undergoing changes in both their industrial and institutional settings. Despite its less convincing growth dynamics, Japan still serves as the regional centre of high value-added and knowledge-intensive manufacturing and service operations. All of this has amounted to the World Bank’s slogan of the ‘East Asian Renaissance’. Decisively, this notion of a resurgence of the East Asian development

ERIKSSON 9780857930088 PRINT.indd 11

29/01/2013 16:59

12

Clusters and economic growth in Asia

trajectory has been conceptually linked with specific forces of growth that involve economies of scale and agglomeration externalities, that is, key aspects in the competitive performance of firms in regional clusters (Gill and Kharas, 2007, pp. 12–16). This perception of a persistent growth and development process even holds in the aftermath of the global financial crisis that hit the US and most European economies especially hard and thus had a temporarily recessive impact on the export-oriented East Asian economies, yet without substantially altering their growth dynamics. Indeed, most of these economies are back on track when it comes to a sound economic performance (Asian Development Bank, 2011, pp. 3–5). In this context, it may be argued that the new emphasis on cluster policies and the advent of the entrepreneurial state are, all together, part of this ‘East Asian Renaissance’, as the World Bank has it, involving both the expansion of transnational business networks and localized government– business interactions. A key issue in this resurgence of the East Asian growth and development performance is the matter of innovation. According to the World Bank, the formation of innovative clusters is a key thrust in the corresponding policy recommendations. Yet right at the outset it is obvious that clusters in East Asia come in vastly different types and structures, involving a range of manufacturing and service industries with both different levels of sophistication and capabilities. The complexity of these clusters is well exemplified by the Pearl River Delta in China with its major light manufacturing industries. Also, it represents a cluster that is driven by a distinct combination of foreign direct investment and public sector enterprises. These kinds of regional clusters differ already in their extended territorial range from local high-tech clusters, such as Taipei/Hsinchu Park in Taiwan or Jurong in Singapore (Yusuf et al., 2003, pp. 234–6). Typically, East Asian clusters include the following components: an initial setting that combines market opportunities with governmental efforts in industry promotion; domestically or export-oriented industrial zones with their particular infrastructure facilities; institutions for capacity-building in human resources; and ‘anchor firms’, which play a key role in promoting the cluster interactions, accompanied by related firms that provide goods and services within the cluster structure (Kuchiki and Tsuji, 2008, pp. 5–6). This specific pattern of cluster formation goes together with structural changes in the East Asian clusters. In particular, firms have been upgrading their operations from basic manufacturing to higher value-added and innovative activities. This upgrading pattern exhibits a transnational dimension, which is shaped by the formation of an ‘Asian triangle’ of transnational production networks, most of which are nested in Japanese, Chinese and ASEAN clusters (Kuchiki and Tsuji, 2011, pp. 2–4).

ERIKSSON 9780857930088 PRINT.indd 12

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia

13

Cluster policies, however, have regularly taken the shape of urban planning programmes, infrastructure projects or financial assistance schemes for certain locations such as science parks and export processing zones. These substitute formats for distinct cluster policies have been largely insufficient in combining physical, knowledge and social capital to promote cluster interactions. According to the World Bank, a favoured pattern of cluster policies in East Asia involves the following aspects: first, network governance operations, which initiate networking dynamics and cooperation among firms and business associations; second, the provision of specific kinds of public goods involving the means for technological information and workforce training, both of which tend to be under-supplied by private firms; third, well-organized cluster management that prevents clusters from phenomena such as network closure and thus maintains their adaptive flexibility in turbulent world markets (Yusuf et al., 2003, pp. 249–54). Accordingly, East Asian cluster policies are set to mirror both domestic economic conditions and the actual position in the international division of labour that is moulded by the informal dynamics of East Asian regional economic integration (Suehiro, 2009). This array of distinct responses to the technological and organizational challenges of catch-up growth and late industrialization, with its new emphasis on industrial cluster strategies, reflects even more comprehensive institutional changes that herald the emergence of the East Asian entrepreneurial states. Yet this tendency does not imply a convergence towards a best practice model. Rather, it upholds a diversity of institutional frameworks and structural conditions that is even enlarged through the new strategic options for firms and governments, which are provided by the cluster perspective. The notion of a ‘modular economy’ illustrates this diversity of strategic options in the East Asian economies: The organisation of production obeys less and less a single predetermined model which would be a must for everyone, reducing the field of possible spaces. On the contrary it opens up this field. The agglomerations of enterprises, districts, clusters or poles of competitiveness can perfectly benefit from the variety of their systems of organisation. (Ganne and Lecler, 2009, p. 22)

Accordingly, East Asian production networks become part of multilayered ‘global networks of networks’, which combine diverse national models and their components – with clusters serving as network hubs in a complex setting of transnational flows of resources, goods and services (Ernst and Kim, 2002). These considerations apply first of all to Japan as the regional technology leader. Indeed, a restructuring of government and administration lies at the heart of the reorientation of the Japanese development

ERIKSSON 9780857930088 PRINT.indd 13

29/01/2013 16:59

14

Clusters and economic growth in Asia

pattern towards a more competitive and entrepreneurial setting (Aoki, 2002, p. 2). Japan’s economy has been gradually opening its competitive structures both domestically and internationally, thus expanding patterns of competitive pressures that were already prevalent in most of the internationally competitive industries (Porter and Sakakibara, 2004, pp. 35–6). Decisively, in the course of these policy reforms, the rationale of generating innovations through flexibilization, decentralization and competitive reorientation of governance structures has become prominent (Whittaker, 2003, pp. 80–81). Japan’s Ministry of Economy, Trade and Industry (METI), which has championed Japanese industrial policies on a national scale for decades, is currently spearheading a set of regionalized innovation and entrepreneurship strategies that address the cluster aspects of innovation, thus allowing for new spatial and institutional components in industrial policy (Ibata-Arens, 2004, pp. 4–5). Already since the late 1990s, the emphasis of industrial policy in Japan has refocused from the support of small business networks in the manufacturing sector to the restructuring of industries that face the challenge of international relocation, primarily to China and other East and Southeast Asian countries. These new types of cluster strategies in terms of distinct policies that aim to promote regional industrial agglomeration in order to raise competitiveness and innovation have played a key part in the formation of complex regional patterns of interaction between firms, universities, research institutes and related organizations (Kitagawa, 2007). During the 2000s, national government initiatives in the domain of cluster policies have been promoting diverse cluster projects and models that highlight cluster formation through regional networking among established firms and research organizations, as well as through entrepreneurial startups. Yet these top-down approaches are increasingly complemented by bottom-up initiatives, in particular in support of science-based clusters (Sanz-Menendez and Cruz-Castro, 2005). Both South Korea and Taiwan have evolved as major challengers to the Japanese leadership in technological innovation. South Korea is actually said to be challenged by a paradigm shift from an ‘industrial learning paradigm’ to a ‘technology creation paradigm’ – with policy-assisted innovation efforts in biotechnology as an outstanding example that points to the strategic focus on knowledge-based cluster policies (Wong et al., 2004, p. 46). Yet this transition towards industrial leadership and clusteroriented policies is differentiated with regard to product groups and market segments within the large Korean conglomerate firms; an aspect that adds to the diversity of cluster types and their distinct logics (Hobday et al., 2004, pp. 1455–6). Similar implications hold for Taiwan, where sustaining competitiveness implies continuous technological upgrading

ERIKSSON 9780857930088 PRINT.indd 14

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia

15

towards high-tech sectors – involving institutional change, industrial upscaling and agglomeration economies (Amsden and Chu, 2003). Taiwan has actually emerged as the most innovative economy among the East Asian newly industrialized economies, as measured in terms of R&D indicators. Its performance provides evidence for the argument that global production networks are also important as sources of knowledge for firms from late industrializing economies. This holds especially with regard to the globalized structures of knowledge-intensive and high-tech industries, which involve clusters of local capabilities that need close connections with global production networks and related operations of multinational firms (Hu and Mathews, 2005, p. 1347). The corresponding need for attracting globalized knowledge flows requires that local and global resources are adequately recombined. The Singaporean development model illustrates this case by promoting the vision of a local knowledge agglomeration in a global knowledge-based economy. In this setting, multinational enterprises introduce novelty into the local economic system; yet, also included in the sample of entrepreneurial agents are governmentlinked companies, as well as government boards, which may enforce and coordinate innovation-driven economic change. As Porterian cluster strategies have been put to use already since the 1990s, it is safe to argue that the globalizing local economy of Singapore actually pioneered the logic of cluster policies in an evolving entrepreneurial state (Ebner, 2004, pp. 56–9; Low, 2004).

CONCLUSION The increasing relevance of cluster policies in East Asia needs to be understood as a manifestation of an ongoing institutional and structural transition of the East Asian newly industrialized economies towards an innovation-driven pattern of development, involving both the expansion of transnational business networks and localized government–business interactions. In this context, the national institutional frameworks of the East Asian economies are subject to comprehensive changes that involve the transformation of the model of the ‘developmental state’ towards specific kinds of ‘entrepreneurial states’. Corresponding policy efforts have shifted from a rationale of catching up within an established technological paradigm to a rationale of paradigm creation that involves a potential for technological leadership on an international scale. This means that traditional types of industrial policy, which have targeted certain industries on the grounds of national development goals, are augmented and even replaced by industrial clusters policies, which put an emphasis on

ERIKSSON 9780857930088 PRINT.indd 15

29/01/2013 16:59

16

Clusters and economic growth in Asia

the competitive performance and innovation capacity of agglomerations of firms and industries. Entrepreneurial capacities in the generation and carrying out of innovations and the spatial dimension of production and innovation become crucial components of this new kind of industrial policy. Consequently, the cluster policy approach of East Asian entrepreneurial states reflects a diversity of structural, institutional and spatial patterns that highlight the adaptive recombination of global production networks and local industrial clusters. This strategic combination of international competitiveness, capability-building and locational strategies addresses both local and foreign firms. In this manner, it emphasizes the transnational connectedness of firms and clusters in East Asia – and thus also points to the transnational range of the related entrepreneurial states and their cluster policies.

REFERENCES Amsden, A. (1989), Asia’s Next Giant: South Korea and Late Industrialization, New York: Oxford University Press. Amsden, A. and W. Chu (2003), Beyond Late Development: Taiwan’s Upgrading Policies, Cambridge, MA: MIT Press. Amsden, A. and T. Hikino (1993), ‘Borrowing Technology or Innovating: An Exploration of the Two Paths to Industrial Development’, in. R. Thomson (ed.), Learning and Technological Change, New York: St. Martin’s Press, pp. 243–66. Aoki, M. (2002), ‘Japan in the Process of Institutional Change’, Miyakodayori, 31, 25 January, Tokyo: Research Institute for Economy, Trade, and Industry, accessed 1 November at www.rieti.go.jp/en/miyakodayori/031.html. Asheim, B. (2007), ‘Learning and Innovation in a Globalising Economy: The Role of Learning Regions’, in B. Gustavsen, B. Nyhan and R. Ennals (eds), Learning Together for Local Innovation: Promoting Learning Regions, Luxembourg: Office for Official Publications of the European Communities, pp. 218–34. Asian Development Bank (2011), Asia Economic Monitor, December 2011, Manila: Asian Development Bank. Bathelt, H., A. Malmberg and P. Maskell (2004), ‘Clusters and Knowledge: Local Buzz, Global Pipelines and the Process of Knowledge Creation’, Progress in Human Geography, 28(1), 31–56. Becattini, G. (1991), ‘Italian Industrial Districts: Problems and Perspectives’, International Studies of Management and Organization, 21(1), 83–90. Benneworth, P. and N. Henry (2004), ‘Where Is the Value Added in the Cluster Approach? Hermeneutic Theorising, Economic Geography and Clusters as a Multiperspectival Approach’, Urban Studies, 41(5–6), 1011–23. Benneworth, P., M. Danson and P. Raines (2003), ‘Confusing Cluster? Making Sense of the Cluster Approach in Theory and Practice’, European Planning Studies, 11(5), 511–20. Bergmann, E.M., D. Charles and P. den Hertog (2001), ‘In Pursuit of Innovative

ERIKSSON 9780857930088 PRINT.indd 16

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia

17

Clusters’, in OECD (ed.), Innovative Clusters: Drivers of National Innovation Systems, Paris: OECD, pp. 7–18. Camagni, R. (1995), ‘The Concept of Innovative Milieu and its Relevance for Public Policies in European Lagging Regions’, Papers in Regional Sciences, 74(4), 317–40. Cooke, P. (1998), ‘Introduction: Origins of the Concept’, in H.-J. Braczyk, P. Cooke and M. Heidenreich (eds), Regional Innovation Systems: The Role of Governance in a Globalized World, London: UCL Press, pp. 2–25. Cooke, P. and G. Schienstock (2000), ‘Structural Competitiveness and Learning Regions’, Enterprise and Innovation Management Studies, 1(3), 265–80. Crevoisier, O. (2004), ‘The Innovative Milieu Approach: Towards a Territorialized Understanding of the Economy?’ Economic Geography, 80(4), 367–79. De Martino, R., R.D. McHardy and S. Zygliodopoulos (2006), ‘Balancing Localization and Globalization: Exploring the Impact of Firm Internationalization on a Regional Cluster’, Entrepreneurship and Regional Development, 18(1), 1–24. Ebner, A. (2004), ‘Development Strategies and Innovation Policies in Globalisation: The Case of Singapore’, in S. Mani and H. Romijn (eds), Innovation, Learning and Technological Dynamism of Developing Countries, Tokyo: United Nations University Press, pp. 48–76. Ebner, A. (2006), ‘Institutions, Entrepreneurship and the Rationale of Government: An Outline of the Schumpeterian Theory of the State’, Journal of Economic Behavior and Organization, 59(4), 497–515. Ebner, A. (2007), ‘Public Policy, Governance, and Innovation: Entrepreneurial States in East Asian Economic Development’, International Journal of Technology and Globalisation, 3(1), 103–24. Ebner, A. (2009), ‘Entrepreneurial State: The Schumpeterian Theory of Industrial Policy and the East Asian “Miracle”’, in U. Cantner, J.-L. Gaffard and L. Nesta (eds), Schumpeterian Perspectives on Innovation, Competition, and Growth, Berlin: Springer, pp. 367–88. Ebner, A. and H. Schiele (2012), ‘The Role of Domestic and International Cluster Linkages Explored on the Example of Buyer–Supplier Relations in Learning Regions: A Cross-functional Assessment’, European Planning Studies, 20 (forthcoming). Ernst, D. and L. Kim (2002), ‘Global Production Networks, Knowledge Diffusion and Local Capability Formation’, Research Policy, 31(3), 1417–29. Evans, P. (1995), Embedded Autonomy: States and Industrial Transformation, Princeton, NJ: Princeton University Press. Freeman, C. (2002), ‘Continental, National and Sub-national Innovation Systems – Complementarity and Economic Growth’, Research Policy, 31(2), 191–211. Furman, J.F., M. Porter and S. Stern (2002), ‘The Determinants of National Innovative Capacity’, Research Policy, 31(6), 899–933. Ganne, B. and Y. Lecler (2009), ‘From Industrial Districts to Poles of Competitiveness’, in B. Ganne and Y. Lecler (eds), Asian Industrial Clusters, Global Competitiveness and New Policy Initiatives, Singapore: World Scientific, pp. 3–24. Gertler, M. (2001), ‘Best Practice? Geography, Learning and the Institutional Limits to Strong Convergence’, Journal of Economic Geography, 1(1), 5–26. Gill, I. and H. Kharas (2007), An East Asian Renaissance: Ideas for Economic Growth, Washington, DC: World Bank.

ERIKSSON 9780857930088 PRINT.indd 17

29/01/2013 16:59

18

Clusters and economic growth in Asia

Giuliani, E., R. Rabellotti and M.P. van Dijk (2005), ‘Introduction’, in E. Giuliani, R. Rabellotti and M.P. van Dijk (eds), Clusters Facing Competition: The Importance of External Linkages, Aldershot: Ashgate, pp. 1–12. Hassink, R. (2005), ‘How to Unlock Regional Economies from Path Dependency? From Learning Region to Learning Cluster’, European Planning Studies, 13(4), 521–35. Hobday, M., H. Rush and J. Bessant (2004), ‘Approaching the Innovation Frontier in Korea: The Transition Phase to Leadership’, Research Policy, 33(4), 1433–57. Hu, M.-C. and J.A. Mathews (2005), ‘National Innovative Capacity in East Asia’, Research Policy, 34(3), 1322–49. Ibata-Arens, K. (2004), ‘Japan’s Quest for Entrepreneurialism: The Cluster Plan’, JPRI Working Paper No. 102, Encinitas, CA: Japan Policy Research Institute. Isaksen, A. (2009), ‘Innovation Dynamics of Global Competitive Regional Clusters: The Case of the Norwegian Centres of Expertise’, Regional Studies, 43(9), 1155–66. Johnson, C. (1982), MITI and the Japanese Miracle: The Growth of Industrial Policy, 1925–1975, Stanford: Stanford University Press. Ketels, C.H.M. (2006), ‘Michael Porter’s Competitiveness Framework – Recent Learnings and New Research Priorities’, Journal of Industry, Competition and Trade, 6(1), 115–36. Kitagawa, F. (2007), ‘The Regionalization of Science and Innovation Governance in Japan?’, Regional Studies, 8(41), 1099–114. Kuchiki, A. and M. Tsuji (2008), ‘Introduction’, in A. Kuchiki and M. Tsuji (eds), The Flowchart Approach to Industrial Cluster Policy, Basingstoke: Palgrave Macmillan, pp. 3–17. Kuchiki, A. and M. Tsuji (2011), ‘Introduction’, in A. Kuchiki and M. Tsuji (eds.), Industrial Clusters, Upgrading and Innovation in East Asia, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 1–12. Lagendijk, A. and P. Oinas (2005), ‘Proximity, External Relations, and Local Economic Development’, in A. Lagendijk and P. Oinas (eds), Proximity, Distance, and Diversity: Issues on Economic Interaction and Local Development, Aldershot: Ashgate, pp. 1–20. Lall, S. (2000), ‘Technological Change and Industrialization in the Asian Newly Industrializing Economies: Achievements and Challenges’, in L. Kim and R.  Nelson (eds), Technology, Learning, and Innovation: Experiences of Newly Industrializing Economies, Cambridge, UK: Cambridge University Press, pp. 13–68. Loasby, B.-J. (2001), ‘Time, Knowledge and Evolutionary Dynamics: Why Connections Matter’, Journal of Evolutionary Economics, 11(4), 393–412. Low, L. (2004), ‘Singapore’s Developmental State Between a Rock and a Hard Place’, in L. Low (ed.), Developmental States: Relevancy, Redundancy or Reconfiguration?, New York: Nova Science, pp. 161–77. Lundvall, B.-A. (1998), ‘Why Study National Systems and National Styles of Innovation?’, Technology Analysis and Strategic Management, 10(4), 407–21. Lundvall, B.-A. and P. Maskell (2000), ‘Nation States and Economic Development – From National Systems of Production to National Systems of Knowledge Creation and Learning’, in G.L. Clark, M.P. Feldman and M. Gertler (eds), The Oxford Handbook of Economic Geography, Oxford: Oxford University Press, pp. 353–72.

ERIKSSON 9780857930088 PRINT.indd 18

29/01/2013 16:59

Cluster policies and entrepreneurial states in East Asia

19

Lundvall, B.-A., B. Johnson, E.S. Andersen and B. Dalum (2002), ‘National Systems of Production, Innovation and Competence-building’, Research Policy, 31(2), 213–31. Malmberg, A. and P. Maskell (2002), ‘The Elusive Concept of Localization Economies: Towards a Knowledge-based Theory of Spatial Clustering’, Environment and Planning A, 34(3), 429–49. Martin, R.A. and P. Sunley (2003), ‘Deconstructing Clusters: Chaotic Concept or Policy Panacea?’ Journal of Economic Geography, 3(1), 5–35. Maskell, P. (2001), ‘The Firm in Economic Geography’, Economic Geography, 77(4), 329–44. Maskell, P. and M. Lorenzen (2004), ‘The Cluster as Market Organisation’, Urban Studies, 41(5–6), 991–1009. Maskell, P. and A. Malmberg (1999), ‘Localised Learning and Industrial Competitiveness’, Cambridge Journal of Economics, 23(2), 167–85. Maskell, P., H. Bathelt and A. Malmberg (2006), ‘Building Global Knowledge Pipelines: The Role of Temporary Clusters’, European Planning Studies, 14(8), 997–1013. Menzel, M.-P. and D. Fornahl (2010), ‘Cluster Life Cycles – Dimensions and Rationales of Cluster Evolution’, Industrial and Corporate Change, 19(1), 205–38. Nelson, R.R. and N. Rosenberg (1993), ‘Technical Innovation and National Systems’, in R. Nelson (ed.), National Innovation Systems: A Comparative Analysis, Oxford: Oxford University Press, pp. 1–18. Polenske, K. (2008), ‘Clustering in Space versus Dispersing over Space’, in C. Karlsson (ed.), Handbook of Research on Cluster Theory, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 133–49. Porter, M.E. (1990), The Competitive Advantage of Nations, New York: Free Press. Porter, M.E. (1998), On Competition, Cambridge, MA: Harvard Business Review Books. Porter, M.E. (2000), ‘Location, Clusters, and Company Strategy’, in G.L. Clark, M. Gertler and M.P. Feldman (eds), The Oxford Handbook of Economic Geography, Oxford: Oxford University Press, pp. 253–74. Porter, M.E. (2003), ‘The Economic Performance of Regions’, Regional Studies, 37(5), 549–78. Porter, M.E. (2007), Clusters and Economic Policy: Aligning Public Policy with the New Economics of Competition, Harvard Business School, Institute for Strategy and Competitiveness, ISC White Paper November 2007. Porter, M.E. and M. Sakakibara (2004), ‘Competition in Japan’, Journal of Economic Perspectives, 18(1), 27–50. Sanz-Menendez, L. and L. Cruz-Castro (2005), ‘Explaining the Science and Technology Policies of Regional Governments’, Regional Studies, 39(3), 939–54. Simmie, J. (2008), ‘The Contribution of Clustering to Innovation: From Porter I Agglomeration to Porter II Export Base Theories’, in C. Karlsson (ed.), Handbook of Research on Innovation and Clusters, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 19–32. Steinle, C., H. Schiele and K. Mietzner (2007), ‘Merging a Firm-centred and a Regional Policy Perspective for the Assessment of Regional Clusters: Concept and Application of a “Dual” Approach to a Medical Technology Cluster’, European Planning Studies, 15(2), 235–51. Suehiro, A. (2009), ‘From an Industrial Policy Approach to an Industrial Cluster

ERIKSSON 9780857930088 PRINT.indd 19

29/01/2013 16:59

20

Clusters and economic growth in Asia

Approach: Japan, East Asia and Silicon Valley’, in B. Ganne and Y. Lecler (eds), Asian Industrial Clusters, Global Competitiveness and New Policy Initiatives, Singapore: World Scientific, pp. 25–67. Visser, E.-J. (2009), ‘The Complementary Dynamic Effects of Clusters and Networks’, Industry and Innovation, 16(2), 167–95. Wade, R. (1990), Governing the Market: Economic Theory and the Role of Government in East Asian Industrialization, Princeton, NJ: Princeton University Press. Wade, R. (1996), ‘Globalization and its Limits: Reports of the Death of the National Economy are Greatly Exaggerated’, in S. Berger and R. Dore (eds), National Diversity and Global Capitalism, Ithaca, NY: Cornell University Press, pp. 60–88. Weiss, L. (1998), The Myth of the Powerless State: Governing the Economy in a Global Era, Cambridge, UK: Polity Press. Weiss, L. (2000), ‘Developmental States in Transition: Adapting, Dismantling, Innovating, not “Normalizing”’, The Pacific Review, 13(1), 21–55. Whitford, J. and C. Potter (2007), ‘Regional Economies, Open Networks and the Spatial Fragmentation of Production’, Socio-Economic Review, 5(3), 497–526. Whitley, R. (2007), Business Systems and Organizational Capabilities: The Institutional Structuring of Competitive Competences, Oxford: Oxford University Press. Whittaker, D.H. (2003), ‘Crisis and Innovation in Japan: A New Future Through Technoentrepreneurship?’, in W.W. Keller and R.J. Samuels (eds), Crisis and Innovation in Asian Technology, Cambridge, UK: Cambridge University Press, pp. 57–85. Wong, J., U. Quach, H. Thorsteinsdóttir, P.A. Singer and A.S. Daar (2004), ‘South Korean Biotechnology – A Rising Industrial and Scientific Powerhouse’, Nature Biotechnology, 22 (Supplement) December, 42–7. Wong, P.-K. and C.-N. Ng (2001), ‘Rethinking the Development Paradigm: Lessons from Japan and the Four Asian NIEs’, in P.-K. Wong and C.-N. Ng (eds), Industrial Policy, Innovation and Economic Growth: The Experience of Japan and the Asian NIEs, Singapore: Singapore University Press, pp. 1–54. Woodward, D. and P. Guimaraes (2009), ‘Porter’s Cluster Strategy and Industrial Targeting’, in S.J. Goetz, S. Deller and T. Harris (eds), Targeting Regional Economic Development, London: Routledge, pp. 68–83. World Bank (1993), The East Asian Miracle: Economic Growth and Public Policy, Oxford: Oxford University Press. Yusuf, S., M.A. Altaf, B. Eichengreen, S. Gooptu, K. Nabeshima, C. Kenny, D.H.  Perkins and M. Shotten (2003), Innovative East Asia: The Future of Growth, New York: Oxford University Press. Zucchella, A. (2006), ‘Local Cluster Dynamics: Trajectories of Mature Industrial Districts Between Decline and Multiple Embeddedness’, Journal of Institutional Economics, 2(1), 21–44.

ERIKSSON 9780857930088 PRINT.indd 20

29/01/2013 16:59

2.

Information and communication technology and economic growth of four Asian industrialized economies Yanfei Li and Wai-Mun Chia

INTRODUCTION There has been a widespread debate among economists about the role of information and communication technology (ICT) in economic growth since the late 1990s, especially in the progress of the New Economy in the USA. Despite the Solow Paradox,1 it is generally agreed that ICT production and application have been a major force of economic growth in the USA since 1995 (Jorgenson and Stiroh, 1999, 2000; Oliner and Sichel, 2000; Jorgenson, 2001). Additionally, much effort has been devoted to investigating why the European countries generally lagged behind in utilizing ICT to promote growth performance in terms of real GDP and labour productivity growth, as well as why ICT investment in the USA declined but labour productivity growth continued to accelerate after the year 2000 (Gordon, 2004). The literature has suggested that the promotion of growth performance by ICT does not happen automatically. Rather, it is conditional on many factors including organizational innovation/ investment (Brynjolffson and Hitt, 2000) and sequential complementary innovations for ICT as a general purpose technology (GPT) (Helpman and Trajtenberg, 1996; Basu et al., 2003), as well as sufficient high-skill labour to apply ICT (Basu et al., 2003). It is also found that those service industries (mainly wholesale trade, retail trade, finance and insurance) that invest heavily in ICT are the major non-ICT-producing industries that contributed to the late 1990s’ labour productivity acceleration in the USA (Jorgenson et al., 2002; Stiroh, 2002). While many studies have been found to focus on the contribution of ICT to economic growth in the USA and EU, only a few studies have been reported for Asian countries. Japan has been examined in studies covering OECD countries and in studies for specific cross-country comparison. Van Ark et al. (2002) and Jorgenson and Motohashi (2005) 21

ERIKSSON 9780857930088 PRINT.indd 21

29/01/2013 16:59

22

Clusters and economic growth in Asia

find that the ICT production industries in Japan enjoyed very similar efficiency gains to those in the USA. However, with respect to ICT application, ICT-using industries did not have efficiency gains commensurate with those in the USA and EU. Kanamori and Motohashi (2007) compare the contributions of ICT to economic growth in Japan and in South Korea. It is found for both countries that the importance of ICT capital service growth for economic growth has been increasing. The contribution of non-ICT capital service is much more significant in South Korea than in Japan. Total factor productivity (TFP) growth of the non-ICT sector of Japan is faster than that of South Korea. One may derive the implication from the above observations that ICT application in the non-ICT sector performs better in Japan than in South Korea. Accordingly, in terms of the contribution of ICT to economic growth, the ranking sequence is the USA and EU, followed by Japan and then South Korea. Convergence has been generally predicted in the literature (Basu et al., 2003). As Asian industrialized economies have very distinct social and economic structures (Young, 1992), the study of the pattern of ICT development in Japan and three newly industrialized economies, South Korea, Hong Kong and Singapore, could potentially provide more understanding about the constraints and preconditions of fully exploiting ICT advantages. Thus, the central contribution of this study is a comprehensive analysis of the contributions of ICT to economic growth of these Asian economies, using the growth accounting method. The study also implies the possibility that the ICT revolution could be a source of potential convergence2 between Asian newly industrialized economies (NIEs) and leading economies such as the USA and Japan. The chapter is organized as follows. The next section describes the growth accounting model. The third section describes the data sources and data estimation methodology. The fourth section reports the results from the growth accounting model. The fifth section concludes.

MODEL DESCRIPTION Following Oliner and Sichel (2000), a growth accounting model at the national level starts with the production function: P 3 Y 5 P 3 F (A,Kict,Knict,E) P is the aggregate price level of the economy and Y is the output level. Thus, P 3 Y gives nominal GDP. A is the technology term. Kict

ERIKSSON 9780857930088 PRINT.indd 22

29/01/2013 16:59

Information and communication technology and economic growth

23

is the ICT capital stock, Knict is the non-ICT capital, and E refers to employment. Total differentiation gives: P3

dY 0F dA 0F dKict 0F dKnict 0F dE 5P3a 1 1 1 b dt 0A dt 0Kict dt 0Knict dt 0E dt

Dividing both sides by P 3 Y and assuming Hicks-neutral technological change,3 the above equation becomes: dY 1 dt P 3 Y

P3

5

P 3 0F 3 Kict dKict 1 1 P 3 0F 3 A dA 1 1 a P3Y 0A dt A 0Kict dt Kict 1

P 3 0F 3 Knict dKnict 1 P 3 0F 3 E dE 1 1 b 0Knict dt Knict 0E dt E #

#

#

#

5 A 1 SKict 3 K ict1SKnict 3 Knict1SE 3E #

#

#

#

#

Y 5 A 1SKict 3 Kict1SKnict3 Knict 1SE 3E

(2.1)

Equation (2.1) shows the decomposition of the growth rate of the real # GDP. A is the growth rate of total factor productivity (TFP). SKict, SKnict and SE are the nominal income share # of ICT # capital stock, non-ICT capital stock and labour, respectively. Kict and Knict are the growth rates # of real ICT capital stock and real non-ICT capital stock, respectively. E is the growth rate of employment. The model can be easily extended to treat real ICT capital stock as composed of two subcategories, namely tangible ICT and software. To obtain the estimation of real capital stock, the perpetual inventory method is adopted: Kt 5 It 1 (1 2 d) Kt21, where I is the investment term and d the rate of depreciation rate. Capital service is calculated by multiplying the real capital stock by its rental price. The rental price of one unit of real capital stock is estimated as follows: Rk,t 5 art 1 d 2

ERIKSSON 9780857930088 PRINT.indd 23

PA,t 2 PA,t21 PA,t21

bPA,t21

29/01/2013 16:59

24

Clusters and economic growth in Asia

where r is the real net rate of return and PA is the asset price. Following Jorgenson (2001), the change in labour quality is defined as # # # # employment growth (E ) and q 5 E 2 L. That is, the# difference # between # # labour hour growth (L). Given E 5 L 1 q, the decomposition of real GDP growth is: #

#

#

#

#

#

Y 5 A 1SKict* Kict1 SKnict * Knict 1SE * (L 1q)

(2.2)

The decomposition of the average labour productivity (ALP) growth is: #

#

#

#

#

#

Y 2 L 5 A 1SKict* kict1 SKnict* knict1 SE * q

(2.3)

As already suggested by the previous literature, the accuracy of estimating the contribution of ICT to economic growth depends critically on the measure of the price index of ICT capital goods, as the speed of the decline in the ICT price index will decide the value of real ICT capital stock as well as the rental price, which determines the ICT capital service. Data Description The study is conducted at the national level. National account data, ICT price data and ICT fixed capital formation data, which cover the period from 1986 to 2006,4 are required. Since data for different economies were collected from various sources, the consistency of measurement is assured for all the four economies by using SNA93 statistical standards.5 National account data Specifically, nominal GDP, real GDP, GFCF (gross fixed capital formation) at constant price and GFCF at current price were collected. For Japan, these series were collected from the Statistics Bureau of Japan. Additionally, the JIP 2006 database of ESRI (Economic and Social Research Institute) provided a benchmark real net capital stock and capital service estimation. For South Korea, nominal and real GDP data were collected from the Bank of Korea. Nominal and real GFCF data and consumption of fixed capital were obtained from the Organisation for Co-operation and Development (OECD) STAN database. Information from the National Wealth Survey has been used to provide benchmark year fixed capital stock. The money market rate, which was obtained from the UN Statistical Yearbook, is used as the real net rate of return to general capital stock. For Hong Kong, national account data came from the CEIC database for global emerging and developed markets. Real rate of return is the

ERIKSSON 9780857930088 PRINT.indd 24

29/01/2013 16:59

Information and communication technology and economic growth

25

merge of interbank overnight rate from the CEIC database and discount window base rate from the Hong Kong Monetary Authority. Capital stock per worker data and residential housing stock data were obtained from the Penn World Table 5.6 to estimate the benchmark capital stock. Singapore national account data also mainly came from the CEIC database. Real net rate of return to capital stock is money market rate from the UN Statistical Yearbook. Corporate fixed capital stock data were from a government-conducted corporate sector survey. Together with residential housing stock data, they are used to estimate the benchmark capital stock (Tan and Ping, 2004). ICT capital stock and price index The JIP 2003 and JIP 2006 databases provide an estimation of real ICT capital stock and its capital service of Japan. JIP compiles ICT capital stock using the input–output tables of Japan as benchmarks of ICT fixed capital formation level (Nomura, 2004). Interpolation and extrapolation have been applied using additional information sources (Fukao et al., 2007). This is exactly the methodology that this chapter follows to develop ICT capital stock estimation for South Korea, Hong Kong and Singapore. Since the ICT capital definition in Japan is broad, the same definition hardly applies to the other three economies due to data availability. As a result, ICT capital goods are confined to include (1) office computing and accounting machinery; (2) computers and peripheral equipment; (3) communication equipment; and (4) custom software. As JIP 2006 provides detailed real net capital stock information by asset types, an estimation of this narrowly defined ICT capital stock can be extracted from the database. The JIP 2003 database, which is the early edition of JIP 2006, provides an estimation of the ICT price index up to 1998 excluding software (Fukao et al., 2002). It is used as a tangible ICT capital stock price index and extended using corresponding information from the chained corporate goods price index (CGPI) of Japan prepared by the Bank of Japan. The software price index is obtained from 1995 onwards from the CGPI. Labour cost information has been used to extend the time series back to 1986. As the price index of tangible ICT capital goods shows a completely different path pattern as compared with that of software, software is treated separately, with the rest of ICT capital being treated as tangible ICT capital. Figure 2.1 (a) presents the tangible ICT price, the growth of tangible ICT capital stock (GTICT), and the nominal share of tangible ICT capital service in nominal GDP of Japan, using 1986 as the base year. Starting around 1995, the price index dropped faster. Correspondingly, the growth rate of tangible ICT stock and the nominal share of tangible ICT surged sharply, but did not manage to maintain the same speed subsequently.

ERIKSSON 9780857930088 PRINT.indd 25

29/01/2013 16:59

26

Clusters and economic growth in Asia Tangible ICT-Japan 140 120 Index

100 80 60

TICT Share TICT Price GTICT

40 20

1997

1998

1999

2000

2001

2002

1998

1999

2000

2001

2002

1996

1995

1997

a)

1994

1993

1992

1991

1990

1989

1988

1987

1986

0

Year

Software Share Software Price

b)

1996

1995

1994

1993

1992

1991

1990

1989

1988

1987

GSW

1986

Index

Software-Japan 400 350 300 250 200 150 100 50 0

Year

Figure 2.1a, b Nominal share, price index and growth of real capital stock of Japan’s (a) tangible ICT and (b) software Figure 2.1 (b) presents the software price, the growth of software stock (GSW) and the nominal share of software service in nominal GDP of Japan. While software price showed a generally upward sloping trend, acceleration was even more obvious for the growth of software capital stock. Meanwhile the share of software service in nominal GDP increased sharply. The 1997 Asian financial crisis clearly halted these trends due to the overall cooling down of economic activities. Towards the end of the 1990s, the trends rallied, but around the beginning of the new century they were dampened again. For South Korea, tangible ICT investment is derived from its input– output tables from 1980 to 2003 provided by the Bank of Korea. Data of the years between the benchmark years were interpolated. To do that, data of the production of tangible ICT goods in South Korea were collected from the OECD STAN database and Korean government reports. The tangible ICT price index is estimated using the producer price index (PPI).

ERIKSSON 9780857930088 PRINT.indd 26

29/01/2013 16:59

Information and communication technology and economic growth

27

TICT Share TICT Price GTICT (1986 = 10)

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Index

Tangible ICT-Korea 200 180 160 140 120 100 80 60 40 20 0

a)

Year

Software Share Software Price GSW

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Index

Software-Korea 850 750 650 550 450 350 250 150 50

b)

Figure 2.2a, b

Year

Nominal share, price index and growth of real capital stock of South Korea’s (a) tangible ICT and (b) software

The software investment data from 2002 to 2006 were obtained from the Korea Association of Information and Telecommunication (KAIT) statistical report. The data from 1997 to 2001 were estimated according to the production growth rate of the software industry reported by KAIT. The data for 1986 onwards were derived using the growth rate of the tangible ICT investment, as software investment is supposed to be complementary to hardware investment. PPI for ‘computer-related services’ was used to estimate the software price index since 1995. Before 1995, the price index was estimated using data from labour cost and tangible ICT price. According to Figure 2.2, a prominent feature of the South Korea case is the sharp acceleration in the growth rate of stock of both tangible ICT and software. Drastic fluctuations are observed, and the 1997 Asian financial crisis clearly plays a role. The other feature is that the spikes of software stock growth clearly lagged behind the spikes of the tangible ICT stock growth. The trend of faster growth in software investment despite the

ERIKSSON 9780857930088 PRINT.indd 27

29/01/2013 16:59

28

Clusters and economic growth in Asia

TICT Share TICT Price GTICT

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Index

Tangible ICT-Hong Kong 450 400 350 300 250 200 150 100 50 0

a)

Year

Software-Hong Kong 600 500 Index

400

Software Share Software Price GSW

300 200 100 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

0

b)

Figure 2.3a, b

Year

Nominal share, price index and growth of real capital stock of Hong Kong’s (a) tangible ICT and (b) software

rising price of software implies that software investment is complementary to ICT hardware capital stock. The tangible ICT investment data for Hong Kong from 1998 to 2005 were obtained directly from the government reports. The tangible ICT goods trade statistics of Hong Kong were obtained from the UN Commodity Trade Statistics Database to extrapolate the tangible ICT investment.6 Based on the government PPI index from 1998 to 2006, the tangible ICT price index is estimated using tangible ICT price indices of the USA, Japan and South Korea, which are important trading partners of Hong Kong’s tangible ICT goods. Their trade volumes are used as weights. The estimation coincides with the government estimation from 1998 to 2000, with only some slight deviation after 2000. This evidence also shows that international trade price, to a large extent, decides the prices of tangible ICT goods in Hong Kong. The CEIC database provided Hong Kong software investment data both at current price and at base year constant price. Thus, implicit software investment price can be derived. As shown by Figure 2.3, the breakthrough of ICT around 1995 and

ERIKSSON 9780857930088 PRINT.indd 28

29/01/2013 16:59

Information and communication technology and economic growth

29

the 1997 Asian financial crisis both had a lagged impact on Hong Kong’s growth of the tangible ICT and software stock. Although the growth rates of both tangible ICT and software capital stock were not impressive, the surge in the nominal share of capital service of the two in nominal GDP was. For Singapore, input–output tables from 1983 to 2000 are available, from which the tangible ICT investment data can be retrieved for the benchmark years. For the years between the benchmarks, trade statistics of the tangible ICT goods were used to convey information about the fluctuation in tangible ICT investment. From 1986 onwards, Yearbook of Statistics Singapore records tangible ICT price index in its domestic supply price index. The software investment data were estimated according to a relevant occasional paper published by the Department of Statistics of Singapore. The paper provided the share of software investment in GFCF of Singapore from 2000 to 2004. The changes of the GDP share of the information and communication service industry provides a reference to adjust the share of software investment for the rest of the years. As Singapore does not prepare any service price index, the software price index is assumed to be the same as that of Hong Kong. As shown by Figure 2.4, acceleration in the growth rate of the tangible ICT investment has been moderate after 1995, with drastic fluctuation. The software investment, on the other hand, reacted strongly first to the ICT breakthrough in 1995 and later to the Asian financial crisis in 1997. The nominal share of software service in nominal GDP also improved significantly after 1995. During this process supportive policies by the Singapore government played an important role. Since the 1980s, the government started various programmes to promote ICT including the Civil Service Computerisation Programme (1981), National IT Plan (1986), IT2000 (1992), Singapore ONE (1996), Next Generation National Broadband Network and Wireless@SG programme. These policies affected the timing, structure and magnitude of the ICT investment in Singapore. Figure 2.5 compares the actual growth rates of the tangible ICT stock (a) and software stock (b) in all four economies. Generally, the three NIEs have higher ICT investment than Japan. Figure 2.6 compares the speed of change in the ICT service share in the four economies. Hong Kong and South Korea experienced relatively faster increase in service share of tangible ICT and software. The following growth accounting analysis combines all the above information about ICT investment in these economies to see to what extent ICT contributes to economic growth.

ERIKSSON 9780857930088 PRINT.indd 29

29/01/2013 16:59

30

Clusters and economic growth in Asia Tangible ICT-Singapore 300

TICT Share TICT Price GTICT (1986 = 10)

250 Index

200 150 100 50

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

0

a)

Year

Software-Singapore 320 270

Software Share Software Price GSW (1986 = 10)

Index

220 170 120 70

–80

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

20 –30 b)

Figure 2.4a, b

Year

Nominal share, price index and growth of real capital stock of Singapore’s (a) tangible ICT and (b) software

Labour input The JIP 2006 database provides labour hour, nominal labour cost and labour quality data for Japan. In the case of South Korea, employment data with hours worked and wage data are directly available from the Korea Statistical Office. Hong Kong employment, labour cost and labour hour data are from the Census and Statistics Department of Hong Kong. Singapore employment data are from the UN Statistical Yearbook database. Labour cost data are from CEIC database. Data on weekly hours worked were obtained from the Yearbook of Statistics Singapore.

RESULTS AND ANALYSIS To identify the contribution from various sources to real GDP growth and labour productivity growth, a decomposition analysis according to equations (2.2) and (2.3) is conducted. The former is decomposition of real GDP growth (RGDP) rate and the latter is decomposition of average labour productivity (ALP) growth. Since the detailed data of both

ERIKSSON 9780857930088 PRINT.indd 30

29/01/2013 16:59

Information and communication technology and economic growth

31

Tangible ICT Growth Japan Korea Hong Kong Singapore

0.5 0.4 Percentage

0.3 0.2 0.1

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

0 –0.1 a)

Year

Software Growth 0.6

Japan Korea Hong Kong Singapore

0.5 Percentage

0.4 0.3 0.2 0.1 0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

–0.1 b)

Figure 2.5a, b

Year

Cross-country comparison of the growth rates of the ICT capital stock

tangible ICT capital stock and software capital stock are available, the ICT capital stock could be further decomposed to these two components. Following the standard practice in literature, the decomposition is conducted for the average growth rate of a certain period, rather than for the growth rate of each year. Four periods are considered: 1986–90, 1991–95, 1996–2000 and 2001–06. In the case of Japan, due to lack of data, the last period is 2001–02. Table 2.1 and Table 2.2 present the decomposition results for the four economies for real GDP (RGDP) growth and ALP growth respectively. According to Table 2.1, the contribution of ICT to real GDP growth rose in the late 1990s, while that of non-ICT capital growth generally declined. According to Figure 2.7, in terms of relative contribution of ICT to real GDP growth, Japan is the leading economy. In fact, despite Japan’s declining real GDP growth, the contribution of ICT increased until 2000. Singapore followed Japan initially, but its acceleration of contribution of ICT after the ICT breakthrough in 1995 was weaker than that of South Korea and Hong Kong. Initially, South Korea and Hong Kong lagged behind Singapore in the late 1980s but managed to catch up and even

ERIKSSON 9780857930088 PRINT.indd 31

29/01/2013 16:59

32

Clusters and economic growth in Asia

Japan Korea Hong Kong Singapore

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Index

Tangible ICT Share 450 400 350 300 250 200 150 100 50 0

a)

Year

Japan Korea Hong Kong Singapore

1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006

Index

Software Share 900 800 700 600 500 400 300 200 100 0

b)

Figure 2.6a, b

Year

Service share (capital service/nominal GDP) of (a) tangible ICT and (b) software capital

outperform Singapore in the late 1990s. However, the contribution of ICT in Hong Kong dropped drastically after 2000. A prominent feature displayed by Table 2.1 is that even after the shock of the 1997 Asian financial crisis, with real GDP growth rate declining sharply, the relative contribution of ICT kept increasing in all four economies. After 2000, the contribution of ICT in absolute value declined in Japan, Hong Kong and Singapore but kept accelerating in South Korea. However, the relative contribution of ICT continued to increase in all economies except Hong Kong. The strong recovery of the Hong Kong economy after the Asian financial crisis seems to have been driven by extraneous factors other than the ICT capital growth and not by traditional contributing forces such as non-ICT capital growth, labour input growth and labour quality growth. For all four economies, the contribution of tangible ICT growth exceeded that of software, which is true for all sample periods. Yet in Japan and South Korea the contribution of software increased to catch up with that of tangible ICT. The catch-up in contribution between software and tangible ICT was prominent during 2001–06, which is also the period

ERIKSSON 9780857930088 PRINT.indd 32

29/01/2013 16:59

Information and communication technology and economic growth

Table 2.1

33

Decomposition of real GDP growth Percentage

Japan Growth rate of RGDP Contribution from: Non-ICT capital ICT capital* Tangible ICT capital Software Labour input Labour quality MFP growth

South Korea Growth rate of RGDP Contribution from: Non-ICT capital ICT capital* Tangible ICT capital Software Labour input Labour quality MFP growth Hong Kong Growth rate of RGDP Contribution from: Non-ICT capital ICT capital* Tangible ICT capital Software Labour input Labour quality MFP growth Singapore Growth rate of RGDP Contribution from: Non-ICT capital ICT capital* Tangible ICT capital Software Labour input Labour quality MFP growth

1986–90

1991–95

1996–2000

2001–02

2.14

1.52

1.44

–0.03

1.13 0.33 0.29 0.04 –0.24 0.46 0.46

1.05 0.29 0.25 0.04 –0.32 0.45 0.05

0.49 0.34 0.25 0.09 –0.33 0.40 0.54

0.17 0.16 0.08 0.08 –0.82 0.20 0.27

1986–90

1991–95

1996–2000

2001–06

8.05

7.81

4.56

4.63

4.66 0.23 0.21 0.02 1.45 0.35 1.35

5.02 0.25 0.22 0.03 1.49 0.38 0.70

3.03 0.52 0.42 0.10 0.06 0.52 0.40

1.26 0.67 0.38 0.29 0.49 0.76 1.47

5.33

5.6

3.57

4.78

1.01 0.24 0.16 0.08 0.29 0.00 3.78

1.03 0.25 0.17 0.08 0.40 0.09 3.84

2.17 0.40 0.29 0.11 0.87 –0.06 0.19

0.92 0.29 0.22 0.07 0.31 0.08 3.18

8.93

8.87

6.4

4.7

1.02 0.54 0.53 0.01 1.34 0.16 5.88

1.08 0.50 0.49 0.01 0.74 0.15 6.40

2.17 0.44 0.32 0.12 1.95 0.05 1.80

0.54 0.40 0.40 0.00 0.64 0.23 2.88

Note: * The contribution ICT capital is the sum of those of tangible ICT and software.

ERIKSSON 9780857930088 PRINT.indd 33

29/01/2013 16:59

34

Clusters and economic growth in Asia Relative contribution of ICT to GDP growth 25

Percent

20 15 10 5 0 1986–90

1991–95

1996–2000

2001–06

Period Japan

Figure 2.7

Korea

Hong Kong

Singapore

Relative contribution of ICT to real GDP growth: the share of contribution of ICT in total positive contribution from all sources

when the contribution of tangible ICT was weakening. The same does not apply to Hong Kong and Singapore. In these two economies, the contribution of tangible ICT continued to outweigh that of software. Table 2.2 shows the results from ALP decomposition for Japan, South Korea, Hong Kong and Singapore. According to Table 2.2, the contribution of ICT to ALP growth in South Korea and Hong Kong saw a sharp increase in absolute value in 1996–2000, despite the impact of the 1997 Asian financial crisis. For Japan, the increase in the contribution of ICT in absolute value was smoother over the same period. For Singapore, although the contribution of ICT decreased in absolute terms according to Figure 2.8, the contribution in relative terms increased quickly. In the late 1990s Hong Kong, South Korea and Singapore had a sharp increase in relative contribution of ICT to ALP growth. For South Korea and Singapore this trend of increasing relative contribution continued after 2000. Again, the relative contribution of ICT to ALP growth remained high in spite of the 1997 Asian financial crisis, which dampened ALP growth in all economies. After 2000, the relative contribution of ICT to ALP growth declined in Japan and Hong Kong. The decline was especially strong in Hong Kong. Before 2000, the contribution of software had been far less important than that of tangible ICT in all economies. After 2000, the contribution of software in both absolute and relative terms increased for Japan and South Korea, while that of tangible ICT declined. As a result, the contribution of software to ALP growth tended to catch up with that of tangible ICT. Similarly, Hong Kong and Singapore experienced a decline in relative

ERIKSSON 9780857930088 PRINT.indd 34

29/01/2013 16:59

Information and communication technology and economic growth

Table 2.2

35

Decomposition of ALP growth Percentage

Japan ALP growth Contribution from: Non-ICT capital ICT capital* Tangible ICT capital Software Labour quality MFP growth

South Korea ALP growth Contribution from: Non-ICT capital ICT capital* Tangible ICT capital Software Labour quality MFP growth Hong Kong ALP growth Contribution from: Non-ICT capital ICT capital* Tangible ICT capital Software Labour quality MFP Growth Singapore ALP growth Contribution from: Non-ICT capital ICT capital* Tangible ICT capital Software Labour quality MFP growth

1986–90

1991–95

1996–2000

2001–02

4.23

2.15

2.11

1.72

0.95 0.51 0.45 0.06 0.48 2.29

1.23 0.32 0.27 0.04 0.45 0.16

0.65 0.38 0.28 0.10 0.40 0.68

0.56 0.23 0.13 0.10 0.20 0.74

1986–90

1991–95

1996–00

2001–06

6.25

5.86

4.48

4.04

2.18 0.15 0.14 0.01 0.32 3.61

4.34 0.22 0.19 0.02 0.38 0.92

3.00 0.52 0.42 0.11 0.52 0.43

1.12 0.64 0.37 0.28 0.76 1.51

6.71

4.34

1.14

3.93

0.01 0.09 0.05 0.05 0.004 6.61

0.87 0.05 –0.02 0.07 0.09 3.33

1.60 0.30 0.21 0.10 –0.06 –0.70

0.68 0.30 0.24 0.06 0.08 2.87

4.34

7.15

2.21

3.44

–1.68 0.27 0.30 –0.03 0.15 5.60

0.90 0.50 0.49 0.003 0.14 5.61

0.65 0.40 0.31 0.09 0.05 1.12

0.22 0.67 0.67 –0.003 0.21 2.33

Note: * The contribution ICT capital is the sum of those of tangible ICT and software.

ERIKSSON 9780857930088 PRINT.indd 35

29/01/2013 16:59

36

Clusters and economic growth in Asia Relative contribution of ICT to ALP growth 25

Percent

20 15 10 5 0 1986–90

1991–95

1996–2000

2001–06

Period Japan

Figure 2.8

Korea

Hong Kong

Singapore

Relative contribution of ICT to ALP growth: the share of contribution of ICT in total positive contribution from all sources

contribution of tangible ICT. However, for these two economies both absolute and relative software contribution to ALP growth demonstrated a drastic downturn in 2001–06. In Singapore, the software contribution was even slightly negative during this period.

CONCLUSIONS This study follows the growth accounting model in the literature to analyse the role of ICT in economic growth of four Asian industrialized economies, Japan, Hong Kong, South Korea and Singapore. Based on the model, six general observations are documented. First, the contribution of ICT in economic growth (real GDP growth and ALP growth) experienced a strong acceleration since 1995 in all economies except Singapore. The relative contribution of ICT to real GDP growth and ALP growth increased for all four economies before 2000. After 2000, Hong Kong experienced a drastic decline in its relative contribution of ICT. Similarly, Japan also experienced a downturn in relative contribution of ICT to ALP growth. Second, the contribution of ICT in Japan grew steadily. The relative contribution of ICT was generally higher compared with the other three Asian economies. Third, the contribution of ICT in absolute terms was higher in Singapore than that in Hong Kong from 1986. However, in terms of relative contribution of ICT, Hong Kong managed to catch up with Singapore in

ERIKSSON 9780857930088 PRINT.indd 36

29/01/2013 16:59

Information and communication technology and economic growth

37

1996–2000, but dropped behind again after 2000. Singapore also saw a steadier increase in relative contribution of ICT as compared to Hong Kong. Fourth, in terms of the relative contribution of ICT to economic growth, South Korea started at a lower level than Japan, Singapore and Hong Kong. However, its absolute and relative contribution of ICT, especially of software, grew fast and constantly even during the period of 2001–06, when the contribution of ICT of all other economies experienced a downturn either in absolute or in relative terms. It is also noted that after 2000 South Korea surpassed Singapore and Hong Kong in its relative contribution of ICT to real GDP growth and ALP growth. Fifth, software investment is becoming an increasingly important driving force in contributing to the economic growth of Japan and South Korea. This could be due to the fact that information processing ability now depends more on improvement of software. However, Hong Kong and Singapore seem to follow along the old path of ICT investment. Sixth, the contribution of ICT in absolute terms has been a main convergence force for Singapore, South Korea and Hong Kong to catch up with Japan in real GDP and labour productivity since 1996. For Singapore, this is true with respect to real GDP as early as 1986–90. The speed of the catch-up effect through ICT for Hong Kong is the slowest among the three NIEs. Despite the fruitful lessons learned about the role of ICT in the four industrialized economies in Asia, there is a possible scope for exploring further. For example, the decomposition of multi-factor productivity (MFP) growth using industry-level data is useful in distinguishing the impact of ICT production from ICT application in these economies. A key issue to improve the reliability of the analysis is the availability of constant-quality price indices of ICT goods, which have not been generally developed except for the case of Japan. Internationally harmonized ICT price indices enabling more accurate cross-nation comparisons are desirable in future studies.

NOTES 1. Robert Solow, ‘We’d better watch out’, New York Times Book Review, 12 July 1987, p. 36. 2. On the one hand, the advancement of ICT as a general purpose technology revolution is supposed to offer more chances for technological catch-up. On the other hand, the connotation of this technology per se implies faster diffusion of knowledge and technology (Vu, 2007).

ERIKSSON 9780857930088 PRINT.indd 37

29/01/2013 16:59

38

Clusters and economic growth in Asia

3. The assumption is made for simplicity, without which the following is not true: A Y

#

0F 1 dA 1 3 0A 3 dA dt 3 A 5 dt 3 A 5 A.

4. Sample range for Japan covers the period from 1986 to 2002 only because of a data availability problem. 5. The shortcoming of data compiled using SNA93 is that only custom software is included as productive capital. As detailed data of software investment are not available, this study considers custom software investment as the fixed capital formation of software. 6. This practice is different from the methodology that applies to the other three economies in this chapter, due to the fact that the Hong Kong government does not prepare input– output tables.

REFERENCES Basu, S., J.G. Fernald, N. Oulton and S. Srinivasan (2003), ‘The Case of the Missing Productivity Growth, Or Does Information Technology Explain Why Productivity Accelerated in the United States But Not in the United Kingdom?’, NBER Macroeconomics Annual, 18(1), 9–71. Brynjolfsson, E. and L.M. Hitt (2000), ‘Beyond Computation: Information Technology, Organizational Transformation and Business Performance’, Journal of Economic Perspectives, 14(4), 23–48. Fukao, K., T. Inui, H. Kawai and T. Miyagawa (2002), ‘Sectoral Productivity and Economic Growth in Japan, 1970–98: An Empirical Analysis Based on the JIP Database’, paper prepared for the NBER Thirteenth Annual East Asian Seminar on Economics, 20–22 June, Melbourne, Australia. Fukao, K., S. Hamagata, T. Inui, K. Ito, H.U. Kwon, T. Makino, T. Miyagawa, Y. Nakanishi and J. Tokui (2007), ‘Estimation Procedures and TFP Analysis of the JIP Database 2006 (revised)’, RIETI Discussion Paper Series No. 07-E-003. Gordon, R.J. (2004), ‘Five Puzzles in the Behavior of Productivity, Investment, and Innovation’, NBER Working Paper No. 10660, Cambridge, MA: NBER. Helpman, E. and M. Trajtenberg (1996), ‘Diffusion of General Purpose Technologies’, NBER Working Paper No. 5773, Cambridge, MA: NBER. Jorgenson, D.W. (2001), ‘Information Technology and the U.S. Economy’, American Economic Review, 91(1), 1–32. Jorgenson, D.W. and K. Motohashi (2005), ‘Information Technology and the Japanese Economy’, NBER Working Paper No. W11801, Cambridge, MA: NBER. Jorgenson, D.W. and K.J. Stiroh (1999), ‘Information Technology and Growth’, American Economic Review, 89(2), 109–15. Jorgenson, D.W. and K.J. Stiroh (2000), ‘Raising the Speed Limit: U.S. Economic Growth in the Information Age’, Brookings Papers on Economic Activity, 2000(1), 125–210. Jorgenson D.W., M. Ho and K.J. Stiroh (2002), ‘Information Technology, Education, and the Source of Economic Growth across U.S. Industries’, working paper. Kanamori, T. and K. Motohashi (2007), ‘Information Technology and Economic

ERIKSSON 9780857930088 PRINT.indd 38

29/01/2013 16:59

Information and communication technology and economic growth

39

Growth: Comparison between Japan and Korea’, RIETI Discussion Paper Series, No. 07-E-009. Nomura, K. (2004), ‘Capitalizing Own Account Software in Japan’, Program on Technology and Economic Policy (PTEP), John F. Kennedy School of Government, Harvard University. Oliner, S.D. and D.E. Sichel (2000), ‘The Resurgence of Growth in the Late 1990s: Is Information Technology the Story?’, Journal of Economic Perspectives, 14(4), 3–22. Stiroh, K.J. (2002), ‘Information Technology and the U.S. Productivity Revival: What Do the Industry Data Say?’, American Economic Review, 92(5), 1559–76. Tan, K.E. and T.J. Ping (2004), ‘What Explains Private Non-residential Gross Fixed Capital Investment in Singapore?’, Economic Survey of Singapore, Third Quarter 2004, 56–74. van Ark, B., R. Inklaar and R.H. McGuckin (2002), ‘“Changing Gear”: Productivity, ICT and Service Industries: Europe and the United States’, paper presented at the ZEW Conference 2002 on Economics of Information and Communication Technologies, 24–25 June, Mannheim. Vu, K. (2007). ‘Determinants of Economic Growth in the Information Age’, paper presented at the Singapore Economic Review Conference, 2–4 August, Singapore. Young, A. (1992), ‘A Tale of Two Cities: Factor Accumulation and Technical Change in Hong Kong and Singapore’, NBER Macroeconomics Annual, 7, 3–54.

ERIKSSON 9780857930088 PRINT.indd 39

29/01/2013 16:59

3.

Industrial agglomeration of Taiwanese electronics firms in Dongguan, China: home effects and implications for industrial upgrading Felix Haifeng Liao, Karen Zhihua Xu and Bin Liang

INTRODUCTION Since the 1990s, Taiwanese electronics manufacturers, especially computer producers, have become one of the dominant exporters in the global personal computer (PC) market. However, the regional division of labour in the Taiwanese electronics industry has significantly changed since the late 1990s. In 2009, Taiwan’s PC manufacturing output value reached USD 107.83 billion, while only 0.6 per cent of the worldwide output of Taiwanese electronics firms in PC hardware manufacturing was realized within Taiwan. By contrast, the proportion of that from Mainland China rose to 95.1 per cent (Ministry of Economic Affairs of Taiwan, 2010). Largely depending on the influx of Taiwanese electronics investment, Mainland China has surpassed Taiwan and the USA to be the largest manufacturer of computer hardware since 2005 (Reed Electronics Research, 2007). Over the past two decades, the industrial agglomeration of Taiwanese electronics investment in Mainland China has resulted in some electronics clusters in the specific coastal regions, particularly in the Yangtze River Delta (YRD) and the Pearl River Delta (PRD).1 For instance, as a result of the concentration of Taiwanese electronics firms, Dongguan, a worldwide famous manufacturing centre in the PRD, has shifted its industrial structure from the manufacture of labour-intensive products such as clothes and footwear to, more recently, the production of electronic and computer-related products. In 2009, the share of the electronics industry in

40

ERIKSSON 9780857930088 PRINT.indd 40

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

41

Dongguan’s total industrial output rose to 37.5 per cent, while the number was just 14.3 per cent in 1990 (Dongguan Statistical Bureau, 1991–2010). The agglomeration of Taiwanese electronics firms has also increased the importance of Dongguan in the global PC industry. Notably nearly one out of three disk drives and one out of five scanners and mini-power switches in the global market are nowadays made in Dongguan, and almost 95 per cent of the components and modules in a PC can now be easily sourced in the Dongguan area (e.g., Yang, 2006). With an emphasis on the forces behind the agglomeration of Taiwanese electronics firms in Dongguan, increasing empirical studies have shown that the original inter-firm production linkages in Taiwan have played an important role in the geographical concentration of Taiwanese electronics firms in the region (e.g., Tong and Wang, 2002; Yang, 2007). For instance, in a comparative study of Taiwanese and Hong Kong electronics clusters in Dongguan, Yang and Liao (2010a) summarized that the specific inter-firm linkages of Taiwanese electronics firms in the cluster could be regarded as a coordinated industrial district characterized by a home-based exclusive network. It is worth noting that most of the empirical research on the agglomeration of Taiwanese electronics firms in China, Dongguan in particular, has primarily been based on qualitative analysis and case studies. Although there has been a general perception about the network-based production of Taiwanese electronics firms, little quantitative evidence has been provided to justify these home-based inter-firm linkages. The implications of the agglomeration of Taiwanese electronics firms for the upgrading of the local electronics industry in Dongguan are also unclear. Against this backdrop, this chapter aims to explore the industrial agglomeration of Taiwanese electronics firms in 32 towns or districts within the city of Dongguan using the Ellison and Glaeser index (EG index). We attempt to investigate whether the specific inter-firm production linkages of Taiwanese electronics firms can be justified, to some extent, in a quantitative way. Furthermore, the case of the agglomeration of Taiwanese electronics firms in Dongguan also adds to limited empirical analyses based on the EG index (e.g., Maurel and Sédillot, 1999; Bertinelli and Decrop, 2005; Alecke et al., 2006; Ge, 2009). In addition, through the case study of Taiwanese electronics firms in Dongguan, the chapter also sheds light on the role of transnational corporations (TNCs) in the upgrading in the clusters in developing countries (Wei et al., 2011). The chapter is organized as follows. In the next section, the agglomeration of Taiwanese electronics firms is discussed with a focus on their inter-firm production linkages. This is followed by a brief review of the techniques of measuring industrial agglomeration. And then, employing the EG index,

ERIKSSON 9780857930088 PRINT.indd 41

29/01/2013 16:59

42

Clusters and economic growth in Asia

the industrial agglomeration of Taiwanese electronics firms in Dongguan is analysed in terms of the extent of agglomeration, the ‘co-agglomeration’ of related sub-industries and firm size distribution within the agglomerations. The chapter ends with a discussion on the implications of the agglomeration of Taiwanese electronics firms for the industrial upgrading of the local electronics industry in Dongguan.

INTER-FIRM LINKAGES AND AGGLOMERATION OF TAIWANESE ELECTRONICS FIRMS Industrial agglomeration – the concentration of economic activities from related sectors in a geographical area – is a focal topic in economic geography (Martin and Sunley, 2003). The literature on industrial agglomeration is quite extensive and can be traced back to Marshall’s theory (Marshall [1890] 1920). In Marshall’s model of agglomeration economies, three factors, the pooling of markets for specialized skilled labour, the development of subsidiary trade and suppliers of intermediate inputs and the information within the community of firms, could drive industrial activities to locate together. Based on the strict econometric model, Krugman (1991) argued that increasing returns could be obtained through the pecuniary mechanisms. Glaeser et al. (1992) demonstrated that the information spillover positively facilitated the agglomeration of firms and the geographical proximity also strengthened technological learning and innovation within the cluster. Based on the social network model (Granovetter, 1985), recent literature highlighted that the agglomeration of firms played an important role in fostering interpersonal trust and informality, which in turn contributes to the local ‘embeddedness’ of the clustered firms (Amin and Thrift, 1994). The advances in the research on industrial agglomerations have also brought about some important insights into the analysis of the location of transnational corporations (Dunning, 1998; McCann and Mudambi, 2004). Echoing Markusen’s description of ‘sticky places in slippery space’ (Markusen, 1996), Dunning (1998) demonstrated the importance of agglomeration for the location of TNCs. He argued that, while the globalization force made the location of production activities more geographically dispersed because mutual benefits occur from shared access to localized support facilities, shared service centres, distribution networks, customized demand partners and specialized factor inputs at the regional level within the specific country, TNCs engaging in related activities have tended to agglomerate in limited geographical areas (Dunning, 1998). Specifically, since TNCs do come from different source regions and

ERIKSSON 9780857930088 PRINT.indd 42

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

43

have various cultural and institutional backgrounds, the agglomeration of TNCs may also significantly vary due to the so-called ‘home effects’ (Dunning, 1988, 1992). In association with the aforementioned relationship between inter-firm linkages and the industrial agglomeration of firms, recent studies have shown that the ‘home effects’, with an emphasis on the original inter-firm linkages of TNCs in the home regions, is often relevant to their location choices in the host regions (e.g., Leung, 1990; Kenney and Florida, 1992; He, 2003; Yang, 2009). As Ivarsson and Alvstam (2005) observed, in order to cope with ‘a growing demand for technological capabilities, reduced production costs and increased delivery precision, together with economies of scale in production and design’ (p. 1327), TNCs increasingly decide to stay with their established international suppliers. This gives rise to the co-location of TNCs and their following suppliers (see also UNCTAD, 2001; Ivarsson and Alvstam, 2005). For instance, when Japanese auto TNCs undertake their overseas operations in the USA, strong inter-firm supply linkages within Japanese auto business groups (keiretsu) are likely to be replicated and transplanted in the host regions, contributing to new industrial agglomerations in the USA (Kenney and Florida, 1992). In China, the influx of Taiwanese electronics investment has played an important role in the emergence of several electronics clusters in coastal regions, especially in the YRD and the PRD (Wang and Tong, 2005; Yang and Hsia, 2007; Yang, 2009). In particular, one of the salient characteristics of the agglomeration of Taiwanese electronics firms is the maintenance of the home-based inter-firm production linkages. For instance, Tong and Wang (2002) reported that the co-location of Taiwanese electronics firms in Dongguan, especially in specific towns like Shijie, was significantly driven by the home-based production linkages of Taiwanese firms. In a case study of the Taiwanese electronics firms in the YRD, Hsu (2006) also pointed out that the normal practices of Taiwanese lead firms are to move the whole subcontract system and all suppliers from Taiwan to the YRD under the strategy of ‘the hen brought little chickens together’ (ibid., p. 240). This is because the local environment could not offer available qualified local suppliers. Yang and Hsia (2007) investigated the transforming networking of Taiwanese electronics firms in the Greater Suzhou area and also found that the ‘network power’ of Taiwanese electronics firms has been strengthened by the governance mechanism of Taiwanese system manufacturers during the transplantation. Taiwanese electronics firms also tend to act as subcontractors for large manufacturing firms that are characterized by a ‘developmental’ network of subcontractors of small and medium-sized enterprises (Yeung, 2004). Continuing with such ongoing discussion on the inter-firm production linkages of Taiwanese

ERIKSSON 9780857930088 PRINT.indd 43

29/01/2013 16:59

44

Clusters and economic growth in Asia

electronics firms and their impacts on the spatial pattern of Taiwanese electronics firms, this chapter intends to justify these impacts with substantial quantitative evidence based on a town-level analysis of the distribution of Taiwanese electronics firms in Dongguan. In addition, scholars are also concerned about the impacts of clustered TNCs on the industrial upgrading in developing countries (UNCTAD, 2001; Humphrey and Schmitz, 2002; Ivarsson and Alvstam, 2005). Drawing on the case of the Taiwanese electronics industry in Dongguan, the implications of the agglomeration of Taiwanese firms for the upgrading of local electronics industry are also discussed at the end of the chapter.

METHODS AND DATA Town-level Analysis As Kloosterman and Lambregts (2001) explain, the local context, especially the structure of the urban region in which the industrial activities take place, is crucial in the process of spatial concentration of economic activities and affects ‘the level(s) of spatial agglomeration on which these clusters of economic activities manifest themselves’ (p. 720). In this chapter, with respect to the spatial breakdown of the quantitative analysis of Taiwanese electronics clusters, it is important to note that unlike most of the prefecture-level cities in China, Dongguan does not have any countylevel administrative jurisdictions. The original Dongguan municipality is divided into 32 towns or districts with an average land area of 81 km2 and there is a small city centre of 14 km2 in the northern part (see Figure 3.1). Notably, most of these towns were originally agricultural towns before the reform in 1979. Since the early 1980s, the decentralization of the upperlevel governments has provided the town-level governments in Dongguan with a great deal of incentives and high autonomy to attract foreign investment (Wu, 1997), which to some extent constitutes the so-called town-based economies in the city of Dongguan (Dongguan Bureau of Planning, 2006). Therefore, in order to illustrate the spatial pattern of the Taiwanese electronics cluster in the local context of Dongguan, a ‘townlevel analysis’ is advocated (Yang and Liao, 2010b). The Measure of Agglomeration A number of techniques measuring industrial agglomerations have been developed in the past decade. Based on a comprehensive review of the indices measuring industrial clustering, Duranton and Overman (2005)

ERIKSSON 9780857930088 PRINT.indd 44

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

45

Guangzhou

± Gaobu

Hong Kong

Shijie Shipai Qishi

Chashan Wanjiang hengli

Machong

Qiaotou

Liaobu

Daojiao

Changping

Houjie

Dalingshan

Dalang

Xiegang

Huizhou

Zhangmoutou

Shatian Qingxi

Huangjiang Humen

Tangxia Changan

Shenzhen

Fenggang

Highway Railway Boundary of Dongguan

Kilometres 0 2.5 5

Figure 3.1

10

Dongguan’s location in Guangdong Province and administrative units of Dongguan (2010)

proposed five criteria of the ideal index for the measurement of an industrial agglomeration. They suggest such an index (1) is comparable across industries, (2) controls for the overall agglomeration of manufacturing, (3) controls for industrial concentration, (4) is unbiased with respect to scale and agglomeration and across administrative boundaries and (5) assesses the statistical significance. According to these five requirements, the indices measuring industrial agglomeration can be distinguished into three types: indices fulfilling the first and second requirements, indices satisfying the first three requirements and indices fulfilling all criteria (Bertinelli and Decrop, 2005). First, some indices satisfy the first and second requirements such as the typical Gini coefficient. Owing to its ease of calculation and the limited data requirements, the Gini coefficient has been widely used in the research on geographical concentration of industrial activities in various countries (Krugman, 1991; Sukkoo, 1995; Audretsch and Feldman, 1996). Using the Gini coefficient, He et al. (2008) studied the geographical concentration of industrial locations in China and found that the sectors with a large share of foreign firms are more concentrated. Resembling the Gini

ERIKSSON 9780857930088 PRINT.indd 45

29/01/2013 16:59

46

Clusters and economic growth in Asia

coefficient, several other indices have been put forth to explore industrial agglomerations in different ways, such as the Herfindahl index, the Herfindahl-Hirschman index, the dissimilarity index and the coefficient of variance (CV) index. Based on the Herfindahl index and location quotient indices, Fan and Scott (2003) verified the positive relationship between the clustering of industrial activities and regional growth in China. Employing the coefficient of variance (CV), Wang and Xu (2004) found that TNCs in the electronics industry are more concentrated over time, but the TNCs in traditional sectors have evolved to be more dispersed. Using the location quotient index and dissimilarity index, He (2003) identified that TNCs from Hong Kong, Taiwan, Japan and the United States display different patterns of spatial distribution within China at the city level, and these differences also exist in various sectors. The above indices have to some extent successfully fulfilled the first and second criteria proposed by Duranton and Overman (2005); however, most of them do not control the effects of industrial concentration and fail to satisfy the third requirement. Ellison and Glaeser (1997) advanced the work of measuring industrial agglomeration by proposing an EG index that fulfils the first three criteria simultaneously. First, the equation of the EG index is:

ri 5

Gi 2 a1 2 a x2C bHi C (3.1) a1 2 a x b (1 2 Hi) 2 C C

where Gi 5 a (SiC 2 xC) 2    Hi 5 a Zij2 c

j

The index of Gi is defined as an index of so-called ‘raw geographical concentration’, that is, the degree of concentration without the consideration of firm size distribution within the agglomerations, where sic refers to the share of an industry i’s investment or employment in region c and xc is the share of total manufacturing employment or investment in region c. The definition of the Hi (Herfindahl-Hirschman) index is to measure the effect of industrial concentration, which is calculated as the sum of squared firmsize shares in terms of investment or employment by industry i, where j 5 1. . .. . .N; N is the number of firms. Hi is a function of the number and size distribution of firms in industry i (Bertinelli and Decrop, 2005). The value of Hi is generally high for industries with a small number of plants and

ERIKSSON 9780857930088 PRINT.indd 46

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

47

an uneven size distribution. If the inverse of the Hi is one, the firms in the industry are all of the same size measured by employment or investment. According to the EG index equation, the contribution of the EG index is mainly its combination of the index measuring the raw geographical concentration by the Gi and the index of industrial concentration, that is, the Hi index. Further, in contrast to the previous indices with a lack of absolute value in terms of the standard criteria of industrial clustering or agglomeration, Ellison and Glaeser (1997) propose some explicit benchmarks measuring the comparable extent of industrial agglomeration. They assume that if the EG index is ,0.02, it would indicate that the industry is not very concentrated, while if the value is between 0.02 and 0.05, the industry could be regarded as relatively concentrated and, more importantly, if the EG index is larger than 0.05, it is suggested that the industry is highly concentrated. Following the work of Ellison and Glaeser, many empirical studies have been conducted in various countries, and the overall cut-off value put forward by Ellison and Glaeser is also tested across a number of regions and countries and covers different scales of areas. Two of the cited cases are Maurel and Sédillot’s study of the distribution of manufacturing activities in France (Maurel and Sédillot, 1999) and the case study in Germany conducted by Alecke et al. (2006). Interestingly, the EG index is also applicable to some small countries like Belgium (Bertinelli and Decrop, 2005). However, in China, although a great deal of techniques have been used to examine the concentration of industrial activities, few empirical studies have been based primarily on the EG index (Ge, 2009; Yang and Liao, 2010b), and most of the research using the EG index tends to focus on the large spatial aggregates such as provinces. For example, Ge (2009) applied the EG index to the industrial agglomeration at the provincial level and found that there has been a substantial increase in the degree of industrial agglomeration for most manufacturing industries. However, the industrial agglomeration of firms down to the township level in China has rarely been studied using the EG index. The EG index, although powerful, suffers from some drawbacks. For instance, the EG index aims to purge the industrial concentration from the raw geographical concentration. However, some empirical studies have demonstrated that the value of the EG index may still be biased towards the location of large-scale firms (Holmes and Stevens, 2002; Bertinelli and Decrop, 2005). In a case study of the manufacturing activities in Belgium, Bertinelli and Decrop (2005) explicitly demonstrated that the EG indices would by and large reflect the location of large-scale firms rather than small- and medium-scale enterprises (SMEs) (ibid., p. 577). In addition, since the EG index fails to explore the actual input–output relations

ERIKSSON 9780857930088 PRINT.indd 47

29/01/2013 16:59

48

Clusters and economic growth in Asia

among a set of related industries, some scholars also employ an input– output analysis to identify the functional inter-industry agglomeration at the aggregated level, such as the national level. But it is unfortunate that since the input and output data across different sectors are often unavailable at the regional and local level, this approach tends to be applied on an aggregated spatial scale, such as the national scale (Feser and Bergman, 2000; see also O’Donoghue and Gleave, 2004). In this chapter, due to the lack of input–output data of the manufacturing industries within such small geographical areas as Dongguan, the EG index is still preferred because of its applicability in various countries and clear criteria that have been tested in many empirical cases. Firm-level Data and Interviews In the existing literature, data availability has restricted the disaggregated investigation of the industrial agglomerations in China. The firm-level database used in this chapter provides a unique dataset of the foreign enterprises in Dongguan. The dataset was collected from the Dongguan Bureau of Foreign Trade and Economic Cooperation in 2006. The firm-level database offers detailed statistics of every foreign enterprise across 32 towns and districts in Dongguan until the year 2005, including their establishment dates, source regions, total amount of investment, locations, two-digit and three-digit industry codes, entry mode and major products. The database is believed to be reliable and best satisfies the purpose of this research. First, the database is extracted directly from the official department monitoring all the foreign enterprises in Dongguan. Second, in comparison with most existing statistical material in China, the database successfully distinguishes Taiwanese enterprises from the enterprises with funds from Hong Kong and Taiwan.2 Third, unlike  the data used in the previous quantitative analysis of foreign investment in China, which is mostly collected from the surveys or censuses, the database used in this chapter is more reliable and helps to overcome the biased evaluation during the sampling. Moreover, the chapter adopts a broader definition of electronics industry, which generally encompasses two two-digit sectors, that is, the manufacture of communications, computers and other electronic equipment manufacturing (code 40) and the manufacture of electrical machinery and equipment manufacturing (code 39). The three-digit industrial codes also allow us to measure the ‘co-agglomeration effects’ between the sub-industries within the same two-digit industries. In order to supplement our quantitative analysis of the firm-level data, we also conducted a dozen follow-up interviews in the summer of 2009.

ERIKSSON 9780857930088 PRINT.indd 48

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

49

These interviews were facilitated by the local government officials and provided us with rich details about the formation and evolution of the Taiwanese electronics cluster in Dongguan.

INDUSTRIAL AGGLOMERATION OF TAIWANESE ELECTRONICS FIRMS IN DONGGUAN Formation of the Taiwanese Electronics Cluster in Dongguan Taiwanese electronics firms began relocating to Dongguan in the mid1990s. During this time, Taiwanese electronics firms faced growing competition from Southeast Asian electronics firms and the challenges of rising production costs in Taiwan. The early 1990s also witnessed the socialist market reform in China, which accelerated China’s integration into the global economy (Gu et al., 2001). This is specifically relevant to the PRD and, in particular, Dongguan, known for their vanguard roles in China’s reform (Lin, 1997). Based on the clustering of Taiwanese firms in the production of periphery equipment, the PRD and Dongguan have emerged as one of the largest manufacturing bases of keyboards, mice and power switches in the world. In fact, our interviews in Dongguan indicate that one of the most important reasons for the cross-border production of Taiwanese electronics firms in Dongguan is also the lower production and land costs. Moreover, Taiwanese electronics firms in Dongguan have also kept a very close relationship with the suppliers of raw materials and machines and traders in Taiwan. Spatial Distribution of Taiwanese Electronics Firms in Dongguan Figure 3.2 shows the general spatial distribution of Taiwanese electronics firms in Dongguan. Most of the early influx of Taiwanese investment specialized in labour-intensive manufacturing activities such as footwear and furniture and was generally concentrated along the major transportation routes in the city. For instance, Taiwanese investors specializing in the production of furniture tend to be co-located in Dalingshan town and Houjie town, which are both located along the most important highway crossing the Dongguan region, the No. 107 National Highway (Guangdong Provincial Government, 2005).3 In contrast to the spatial distribution of Taiwanese investment in the labour-intensive sectors, as shown in Figure 3.2, Taiwanese electronics firms have tended to concentrate in Qingxi, Shijie, Tangxia and Huangjiang. According to the firm-level statistics, within the 32 towns or districts in Dongguan, only

ERIKSSON 9780857930088 PRINT.indd 49

29/01/2013 16:59

50

Clusters and economic growth in Asia

±

0

4.5

9

Kilometres 18

USD (million) 0.00–29.85 29.86–76.35 76.36–159.68 159.69–429.20 429.21–675.41

Source: Compiled from firm-level database of foreign enterprises in Dongguan, Dongguan Bureau of Foreign Trade and Economic Cooperation (2006).

Figure 3.2

Distribution of Taiwanese electronics investment in Dongguan, 2005

five towns, Qingxi, Shijie, Huangjiang, Tangxia and Chang’an, have accounted for 58 per cent of total Taiwanese electronics investment in Dongguan, reflecting the fact that the general distribution of Taiwanese electronics investment within the city of Dongguan is unbalanced and geographically concentrated in several specific towns. Specifically, the three towns with the largest amount of Taiwanese electronics investment, Shijie, Qingxi and Huangjiang, are all not located near the important highways or railways in Dongguan. The concentration of Taiwanese electronics manufacturers in these towns is greatly attributed to the establishment of some famous leading PC manufacturing TNCs from Taiwan, such as Delta Electronics in Shijie, GVC Corporation in Qingxi and GPM and Gigabyte Electronics in Huangjiang. For example, in Huangjiang town, the concentration of Taiwanese investment is based on a large industrial park established by Taiwanese investors in the early 2000s. The industrial park holds a number of famous Taiwanese electronics firms such as Gigabyte and GPM, which accounted for 40 per cent of

ERIKSSON 9780857930088 PRINT.indd 50

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

51

total foreign investment in the town and over 50 per cent of Taiwanese investment in Huangjiang (Dongguan Bureau of Foreign Trade and Economic Cooperation, 2006). Extent of Industrial Agglomeration Employing the EG index, the industrial agglomeration of Taiwanese electronics investment is examined more precisely in the following three subsections. To begin with, based on the two-digit industrial breakdown, Table 3.1 shows that in 25 two-digit industries, 24 industries have positive values of the EG indices. This result indicates that Taiwanese firms in general tend to be geographically concentrated in specific areas (towns) within Dongguan, especially when compared with the pure random model proposed by Ellison and Glaeser (1997).4 More precisely, according to the EG’s simple dartboard model without any spillovers and natural advantages, the firms of an industry prefer to choose their location in a random manner. When observing the distribution of firms in an industry, a natural first step is to test whether the observed raw geographical concentration represented by the G index is statistically different from the value of G0 based on pure random location choice. Using the paired sample t-test, the differences between the theoretical value of G0 and the empirical value of G in 25 two-digit industries are statistically significant. Specifically, Taiwanese firms in the two-digit sectors of communications, computers and other electronic equipment (40) and electrical machinery and equipment (39) are more related to the agglomeration effects rather than the random location choices. Second, if one takes 0.05 and 0.02 as upper and lower benchmarks, the industrial agglomeration of Taiwanese electronics firms is evident. As shown in Table 3.1, in the sectors of communications, computers and other electronic equipment (40) and electrical machinery and equipment (39), the values of the EG indices are all higher than 0.02, showing that the spatial distribution of Taiwanese electronics firms is moderately concentrated in specific towns in Dongguan. In order to further explore the spatial distribution of Taiwanese electronics firms, the EG indices based on the three-digit sectorial breakdown were also computed (see Table 3.2). Using the upper benchmark of the EG index of 0.05 again, among nine sub-industries within the electronics industry, seven of them are significantly concentrated at the town level. In particular the EG index of the largest sub-industry, computer-related products (404), is geographically agglomerated (EG index is 0.051). It is worth noting that in comparison with the two-digit industries, the threedigit industries tend to be more concentrated. This result is consistent with

ERIKSSON 9780857930088 PRINT.indd 51

29/01/2013 16:59

52

Clusters and economic growth in Asia

Table 3.1

Industrial agglomeration of Taiwanese firms in Dongguan (two-digit)

Twodigit

Sector

13

Processing of food from agricultural products Foods Textile Textile wearing apparel, footwear and caps Leather, fur, feather and its products Timbers, manufacture of wood, bamboo, rattan, palm and straw products Furniture Paper and paper products Printing, reproduction of recording media Articles for culture, education and sports activity (toys) Chemical raw material and chemical products Chemical fibres Rubber Plastic products Non-metallic mineral products Processing of ferrous metals Processing of non-ferrous metals Metal products General purpose machinery Special purpose machinery Transport equipment Electrical machinery and equipment Communications, computers and other electronic equipment Measuring instrument and machinery for cultural activity and office work Artwork, other manufacture n.e.c.

14 17 18 19 20

21 22 23 24 26 28 29 30 31 32 33 34 35 36 37 39 40 41 42

Percentage of Total Investment

Number of Firms

EG Index

1.09

2

15.595a

0.85 1.99 11.72

13 121 423

0.712a 0.040 0.045

0.92 0.78

95 29

0.035 0.059

4.40 1.45 0.70

197 93 41

0.158a 0.009a 0.053

3.38

216

0.059

1.87

92

0.002a

0.10 1.04 8.48 1.93 0.03 0.26 5.53 1.07 4.64 0.87 6.79 38.33

4 70 549 79 3 14 387 94 242 39 467 1058

0.96

49

0.052

0.81

87

0.012a

–0.378ab 0.008a 0.005a 0.028 0.717a 0.020 0.018 0.030 0.048 0.019 0.041 0.022

Notes: a. Refers to not significant at the 5% level. b. The negative value of the EG index is highlighted. Source: Dongguan Bureau of Foreign Trade and Economic Cooperation (2006).

ERIKSSON 9780857930088 PRINT.indd 52

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

Table 3.2

Industrial agglomeration of Taiwanese electronics firms in Dongguan (three-digit)

Three-digit

Sub-industry

404 406 393

Computers Electronic components Wires, cables, optical fibres and electrical materials Electricity distribution and control equipment Illuminating appliances Domestic TV sets and radio receivers Domestic electrical appliances Communication equipment Electromotors

392 397 407 395 401 391

53

Number of Firms

EG Index

G Index

H Index

308 668 162

0.051 0.023 0.056

0.071 0.038 0.068

0.025 0.017 0.018

84

0.061

0.121

0.071

134 61

0.056 0.010

0.078 0.063

0.029 0.058

79 21 8

0.094 0.096 0.093

0.137 0.557 0.219

0.057 0.549 0.154

Source: Dongguan Bureau of Foreign Trade and Economic Cooperation (2006).

He et al.’s (2007) study and highlights that the lower-digit (more disaggregated) industries usually have stronger intra-sector production linkages, resulting in greater geographical concentration. Co-agglomeration Effects It is important to note that industrial agglomeration of manufacturing activities may not just result from the spillover or interdependence within the specific sub-industries. One of the important aspects of industrial agglomeration also includes the co-location of related sub-industries within the same industrial groups. With respect to the electronics industry, the socalled ‘co-agglomeration effects’ are related to the ‘inter-firm production linkages’ in the same industry group, such as the vertical linkages between the sub-industries of electronic components and assembly manufacturing. In order to measure this common effect derived from the contribution of interindustry agglomeration rather than intra-industry concentration, Ellison and Glaeser (1997) proposed an index to quantify the ‘co-agglomeration’ effects to ascertain whether agglomerated sub-industries within the same industry groups tend to locate together or separately. According to the equation of this co-agglomeration index of gc below, the index could be regarded as a weighted estimator of the combined effects of inter-industry and intra-industry agglomeration (Maurel and Sédillot, 1999).

ERIKSSON 9780857930088 PRINT.indd 53

29/01/2013 16:59

54

Clusters and economic growth in Asia

£ rC 5

G

r

§ 2 H 2 a rjw2j (1 2 Hj) j51 12 ax i 2 i

r

1 2 a w2j j51

(3.2)

In the equation of the gc index above, wj is the share of industry j in the two-digit industry group and xi is the share of the region i in the total employment or investment in the whole area. In general, gc mainly reflects the correlation among the sub-industries at the more disaggregated levels (Ellison and Glaeser, 1997, p. 917). If gc 5 0, it would indicate there is no correlation between the sub-industries and hence no more agglomeration in these sub-industries than that simply resulting from the concentration of their intra-industry agglomeration. Since gc is defined as a combined EG index, its criteria could also be based on the benchmarks of 0.02 and 0.05, which indicate the modest co-agglomeration effect and significant co-agglomeration effect, respectively. Ellison and Glaeser (1997) also proposed a descriptive way to quantify the relative strength of the subindustry-driven and group-specific agglomeration across different two-digit groups. As shown in equation (3.3) below, they defined a l index written as the fraction of the simple weighted EG indices of the sub-industries: l;

gc a jrˆjwj

(3.3)

Based on the rescaling by the l index, it is argued that the value of the l index could be compared across the different industry groups, and if the value of the l index is close to 1, it indicates that spillovers across the subindustries are perfectly correlated across the three-digit industries in the same two-digit industry group. The results of the gc index and the l index for the two industry groups within the electronics industry, that is, the two-digit industry of communication, computers and other electronic equipment (40) and the two-digit industry of electrical machinery and equipment (39), are outlined in Table 3.3. According to Table 3.3, in general the gc indices in the electronics industry are all positive, and if the criteria of 0.02 and 0.05 are employed again, some co-agglomeration effects of the electronics industry should be noted (the values of gc are 0.031 and 0.023, respectively). In consideration of the l index, the strength of such co-agglomeration effects is not low.5 This result implies that Taiwanese electronics firms specializing in related

ERIKSSON 9780857930088 PRINT.indd 54

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

55

Table 3.3

Co-agglomeration effects of Taiwanese electronics firms (gc)

Two-digit

Sector

39

Electrical machinery and equipment Communications, computers and other electronic equipment

40

Number of Sub-industries

gc Index

l Index

5

0.031

0.463

4

0.023

0.253

Source: Dongguan Bureau of Foreign Trade and Economic Cooperation (2006).

Table 3.4

Selected towns of aggregated Taiwanese electronics firms in Dongguan

Sub-industry

End products Computer-related products Computer-related products Computer-related products Sub-total Components Electronic components Electronic components Electronic components Sub-total

Town

Number Average Share in Total the Subof Firm Investment Firms Scale industry by (USD (USD Investment million) (%) million)

Huangjiang

477.81

26

18.38

26

Qingxi

207.04

37

5.60

11

Shijie

472.69

30

15.76

25

1157.55

93

Huangjiang

192.12

70

2.74

11

Qingxi

397.41

112

3.55

23

Shijie

212.00

67

3.16

12

801.53

249

62

46

Source: Dongguan Bureau of Foreign Trade and Economic Cooperation (2006).

industrial activities in the electronics industry, such as the sub-industry of electronics components and the sub-industry of computer manufacturing, are likely to be co-located with each other. Table 3.4 further illustrates the co-agglomeration effects in the three

ERIKSSON 9780857930088 PRINT.indd 55

29/01/2013 16:59

56

Table 3.5

Clusters and economic growth in Asia

Co-location of Taiwanese electronics firms by firm size

Firm Size (Z) by Investment (USD million)

Share by Number of Firms

Z.510 55,Z,10 Z,5 Correlation coefficient of SMEs and largescale firms Significance (two-tailed)

7.28% 5.70% 87.02% 0.730 0.000

Source: Dongguan Bureau of Foreign Trade and Economic Cooperation (2006).

aforementioned towns where Taiwanese electronics firms are most concentrated. It shows that Taiwanese firms in the two sub-industries of end products of computer and electronic components are similarly and geographically concentrated in Shijie, Huangjiang and Qingxi, which consistently echoes the previous qualitative analysis of these famous ‘electronics towns’ in Dongguan (Tong and Wang, 2002; Yang, 2007). This result also confirms that the co-agglomeration effects of Taiwanese component suppliers and the end-product producers are evident, and the strong inter-firm production linkages of Taiwanese electronics firms can also be justified to some extent in a quantitative manner. Firm Size Distribution and Agglomeration The major contribution of the EG index, compared with the previous indices of geographical concentration, is to purge the raw geographic concentration from the industrial concentration through the combination of G and H indices. However, as Holmes and Stevens (2002) noted, even after controlling for industrial concentration (H index), the results of the EG indices may still obviously vary with the size of firms. To be specific, in terms of the manufacturing industries, the EG index is probably biased towards the locations of large-scale firms but underestimates the co-location of large firms and SMEs (Bertinelli and Decrop, 2005). Table 3.5 shows that the general patterns regarding the firm size distribution of Taiwanese electronics firms is biased towards the large-scale firms. Even though the large-scale firms with total investment of over USD 10 million only account for 7.28 per cent by number, they have occupied nearly half of the total amount of Taiwanese electronics investment in Dongguan. With reference to the benchmark of USD 10 million, we compute the simple correlation coefficient (Pearson correlation index)

ERIKSSON 9780857930088 PRINT.indd 56

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

Table 3.6

EG index of Taiwanese electronics firms by firm size

Threedigit

Sector

392

Electricity distribution and control equipment Wires, cables, optical fibres and electrical materials Domestic electrical appliances Computers Electronic components Domestic TV sets and radio receivers

393 395 404 406 407

57

SMEs’ EG Index

Large-scale All Firms’ Firms’ EG Index EG Index

0.012

0.138

0.061

0.045

0.080

0.056

0.032

0.221

0.094

0.018 0.034 0.025

0.083 0.053 –0.085

0.051 0.023 0.010

Note: Not all sub-industries in the electronics industry group have firms with a total amount of investment over USD 10 million. Source: Dongguan Bureau of Foreign Trade and Economic Cooperation (2006).

between the numbers of the ‘total investment’ of Taiwanese electronics SMEs and large-scale Taiwanese electronics firms. In other words, the statistical analysis is used to roughly examine whether Taiwanese electronics SMEs, which are normally the suppliers, prefer to be co-located near Taiwanese large-size firms, which are mostly leading assembly firms. Table 3.5 depicts that the location decisions of Taiwanese SMEs in the electronics industry are significantly related to the locations of large-scale Taiwanese electronics TNCs (the value of the correlation coefficient is close to 0.73). Therefore, the industrial agglomeration of Taiwanese electronics firms is more likely to be driven by the co-location of both SMEs and large-scale TNCs rather than the individual location choices of largescale firms. Employing the standard of USD 10 million again, we further computed the EG indices for these two groups of Taiwanese electronics firms (see Table 3.6). It is obvious that the values of the EG indices increase when the calculations are limited to large-scale Taiwanese firms. In other words, the industrial agglomeration of Taiwanese electronics investment in Dongguan is more related to the spatial distribution of large-scale Taiwanese firms rather than Taiwanese SMEs. Moreover, this result is consistent with the work of Bertinelli and Decrop (2005) in Belgium, noting that the results of the EG indices may be biased towards the locations of large-scale firms.

ERIKSSON 9780857930088 PRINT.indd 57

29/01/2013 16:59

58

Clusters and economic growth in Asia

IMPLICATIONS FOR THE UPGRADING OF LOCAL ELECTRONICS INDUSTRY The previous sections have identified the spatial pattern of Taiwanese electronics firms in relation to their unique inter-firm linkages. We also find that the agglomeration of Taiwanese firms at the town level is sensitive to industrial scales (two-digit and three-digit) and firm sizes. The subsequent analysis further discusses the impacts of the agglomeration of Taiwanese electronics firms on the development of the local electronics industry in the city of Dongguan. As illustrated in Table 3.7, the clustering of Taiwanese electronics firms has significantly contributed to the growth of the output value of Dongguan’s electronics industry and increased its shares in both the Guangdong Province and the whole nation. Nevertheless, Table 3.8 demonstrates that the contribution of domestic Chinese firms in the electronics industry in Dongguan remains modest, occupying less than 15 per cent of the total output in the sector. Further, our interviews reflected that most of the interviewed Taiwanese firms would  like to cooperate with other Taiwanese firms rather than local Chinese firms. First, in the specific context of Dongguan, the local economy is relatively weak and the local suppliers are less competitive Table 3.7

Industrial output of electronics industry in Dongguan and its shares in Guangdong and China, 1990–2009 (in RMB million)

Dongguan

1990 1995 2000 2005 2006 2007 2008 2009

8.28 60.61 342.95 1627.23 1979.60 2345.24 2505.20 2279.17

Guangdong

291.44 1522.46 3630.80 13 619.01 16 726.64 19 620.77 22 519.00 23 087.20

China

1381.28 5124.78 12 384.26 40 895.67 51 243.10 63 242.84 74 331.66 78 320.62

Dongguan’s (%) Share in: Guangdong

China

2.84 3.98 9.45 11.95 11.84 11.95 11.12 9.87

0.60 1.18 2.77 3.98 3.98 3.71 3.37 2.91

Notes: a. The output value of the electronics industry combines the numbers of the sector of electrical machinery and equipment (C39) and the sector of communication equipment, computers and other electronic equipment (C40). b. The numbers refer to the state-owned enterprises and the non-state-owned enterprises above the designated size. Sources: Dongguan Statistical Bureau (1991–2010); Guangdong Statistical Bureau (1990– 2006); National Bureau of Statistics of China (1990–2006).

ERIKSSON 9780857930088 PRINT.indd 58

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

Table 3.8

59

Industrial output of local and foreign firms in Dongguan’s electronics industry, 2001–09

Year

Output Value of Local Firms (yuan million)

Output Value of Foreign Firms (yuan million)

Share of Local Firms (%)

Share of Foreign Firms (%)

2001 2002 2003 2004 2005 2006 2007 2008 2009

3 518.61 8 701.88 11 979.83 9 981.06 12 358.07 20 969.87 23 659.16 23 303.75 34 231.64

31 539.9 41 632.33 66 490.68 85 126.28 104 907.21 176 990.53 210 864.55 230 543.53 193 685.51

10.04 17.29 15.27 10.49 10.54 10.54 11.22 10.11 17.67

89.96 82.71 84.73 89.51 89.46 89.41 89.91 90.82 84.98

Notes: a. The output value of the electronics industry combines the numbers of the sector of electrical machinery and equipment (39) and the sector of communication equipment, computers and other electronic equipment (40). b. The output value is limited to the output of industrial enterprises above the designated size. c. Non-local firms here refers to the industrial enterprises with funds from Hong Kong, Macao, Taiwan and other foreign countries or regions. Source:

Dongguan Statistical Yearbook (1991–2010).

when compared to Taiwanese suppliers. Second, the home effect, as evident in the previous quantitative analysis, also restricts the opportunities provided for local firms to enter the supply network of Taiwanese firms (Yang and Liao, 2010a). Third, the institutions in China to some extent also play a role in the formation of the exclusive network of Taiwanese electronics firms. As our interviewees indicated, Taiwanese firms are mostly reluctant to cooperate with local suppliers since intellectual property is poorly protected in China. They all complained about how local firms may threaten their competitiveness in the future due to potential technological leakage. These opinions are consistent with the recent research on Taiwanese electronics firms in other regions such as the YRD in China, where Taiwanese firms also tend to network among themselves but the linkages between TNCs and local firms are weak (Wei, 2010; Wei et al., 2011). As a result of the network-based model of Taiwanese electronics firms, the potential technological spillover from Taiwanese electronics firms to local firms might be quite limited (Yang and Liao, 2010a). Our interviews in Dongguan in 2009 also provided details about the

ERIKSSON 9780857930088 PRINT.indd 59

29/01/2013 16:59

60

Clusters and economic growth in Asia

ongoing redistribution and transformation of Taiwanese electronics firms in China. The interviewees reported that the global financial crisis has substantially affected their revenue from the international market, while the domestic market in China has played a more important role in the marketing of Taiwanese firms. Taiwanese firms have also expanded their local marketing and tried to cooperate with local giant firms to tap into the domestic Chinese market (Li et al., 2011). However, their business operations, especially in the PRD, have faced more challenges in recent years due to the rise of labour costs and land prices in the PRD (Yang, 2012). Some of the largest Taiwanese electronics firms such as Foxconn have started expanding or relocating their production sites from the PRD to the inland provinces such as Hehan Province, where land and labour resources are abundant and cheaper than those in the PRD. In order to keep their business relationship with these largest lead firms, the interviewed Taiwanese firms in Dongguan have also considered relocating to these inland provinces. Therefore, the network-based redistribution of Taiwanese firms in China has been undergoing a new round of geographical redistribution beyond the PRD and the coastal areas in China (Yang, 2009).

CONCLUSION Using Ellison and Glaeser’s (1997) concentration index, the primary objective of this chapter is to explore quantitatively the industrial agglomerations of Taiwanese electronics firms across 32 towns and districts in Dongguan. Based on firm-level interviews and statistics, this chapter also discusses the implications of the agglomeration of Taiwanese firms for the industrial upgrading of the local electronics industry. First, the chapter offers important quantitative evidence for the proliferating empirical studies of the Taiwanese electronics cluster in Dongguan (Yang and Hsia, 2005; Yang, 2007; Yang and Liao, 2010a, 2010b). It shows that the underlying entities of the world-famous PC production cluster are de facto some aggregated towns of Taiwanese electronics firms. Employing the EG index, the study further demonstrates the coagglomeration effect of related industries in the two-digit industry groups and the collocation of SMEs and large-scale firms. We conclude that the inter-firm supply linkages of Taiwanese electronics firms led by large assembly firms can be verified by rigorous quantitative analyses. Further, since few empirical studies of the industrial agglomerations in China have been based on the EG index, the quantitative analysis of the agglomeration of Taiwanese electronics firms in this chapter contributes to the existing limited empirical research (Ge, 2009; Yang and Liao, 2010b).

ERIKSSON 9780857930088 PRINT.indd 60

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

61

The application of the EG index to the case of Taiwanese electronics firms in Dongguan also shows the advantages of the EG index with its combination of industrial concentration and geographical concentration. We also highlight that both industrial scales (two-digit and three-digit) and firm size (SMEs or large-scale firms) matter when analysing patterns of industrial agglomeration in China and Dongguan in particular (He et al., 2007). However, some drawbacks of the EG index such as the biased results towards large-scale firms are observable and they need to be noticed in the future application of the index. In addition, the dataset used in this research has some limitations. First, it cannot provide the analyses with the employment data. Consequently firm size is measured by the total amount of investment rather than the number of employees. Second, some two-digit industries such as the machinery industry, which may also be related to the manufacture of electronics products, are neglected because it is difficult to distinguish the firms only serving the electronics manufacturers from all the firms in these industries. In this sense, under the notion of ‘electronics industry’, the analysis of the co-location of the sub-industries is confined to the nine three-digit sub-industries within the  aforementioned two two-digit industries (39 and 40). This may affect the results in measuring ‘co-agglomeration effects’. Our research has also important empirical and policy implications, with an emphasis on the upgrading of clusters in developing countries. First, although Dongguan has successfully become a hotspot of Taiwanese electronics investment in China, Taiwanese electronics firms have not brought about extensive linkages with local firms, which results in the slow growth of the local electronics industry. The development experience of the Taiwanese electronics cluster in Dongguan is thus generally contrary to the successful experience in other Asian newly industrialized economies where TNCs have established beneficial linkages with local firms and contributed to the upgrading of the electronics industry in the host regions or countries (Yeung, 2007). The case study of Taiwanese firms in Dongguan clearly highlights the challenges that many industrial clusters in the developing countries have been facing under globalization. Second, the findings can inspire the policy-makers in Dongguan, who aim to embed the Taiwanese electronics firms and promote local development through coupling with TNCs (Yang, 2009). It is recommended that local government and policy-makers should be aware of the specific inter-firm linkage characteristics of Taiwanese electronics firms. In particular, more attention should be paid to the future corporate strategies and redistribution of Taiwanese lead firms (ibid.). Last, further studies should be conducted to determine how to provide incentives for Taiwanese firms to cooperate with domestic Chinese firms.

ERIKSSON 9780857930088 PRINT.indd 61

29/01/2013 16:59

62

Clusters and economic growth in Asia

NOTES 1. In this chapter, the terms ‘agglomeration’, ‘geographical concentration’ and ‘clustering’ are used interchangeably. 2. According to the registration system of the foreign enterprises in China, the enterprises with funds from Hong Kong and Taiwan belong to the same category without proper differentiations, that is, the enterprises with funds from Hong Kong, Macao and Taiwan. 3. Dalingshan town has also become the town known for furniture production across the Asia-Pacific region due to the influx of Taiwanese furniture investment (Chen, 2005, p. 198). 4. In the industry of chemical fibre manufacturing, the value of the EG index is negative. However, investment in the industry just accounts for 0.1 per cent of total Taiwanese investment in Dongguan. 5. According to Ellison and Glaeser’s study and the following empirical studies based on the EG index, if the value of l is more than 0.5, the spillover across the sub-industries is quite significant in comparison with the so-called intra-industry agglomerations.

REFERENCES Alecke, B., C. Alsleben, F. Scharr and G. Untiedt (2006), ‘Are There Really High-tech Clusters? The Geographic Concentration of German Manufacturing Industries and its Determinants’, Annals of Regional Science, 40 (1), 19–42. Amin, A. and N. Thrift (1994), Globalization, Institutions, and Regional Development in Europe, New York and Oxford: Oxford University Press. Audretsch, D.B. and M.P. Feldman (1996), ‘R&D Spillovers and the Geography of Innovation and Production’, American Economic Review, 86(3), 630–40. Bertinelli, L. and J. Decrop (2005), ‘Geographical Agglomeration: Ellison and Glaeser’s Index Applied to the Case of Belgian Manufacturing Industry’, Regional Studies, 39 (5), 567–83. Chen, G.M. (2005), Force of Sustainable Development: The Path of Industrial Upgrading in Dongguan, Guangzhou: Guangdong People’s Press [in Chinese]. Dongguan Bureau of Foreign Trade and Economic Cooperation (2006), Firmlevel Database of Foreign-invested Enterprises of Dongguan (internal document). Dongguan Bureau of Planning (2006), Master Planning of Dongguan City, 2005– 2020, Dongguan, China (internal document). Dongguan Statistical Bureau (1991–2010), Dongguan Statistical Yearbook (various annual issues), Beijing, China: China Statistical Press. Dunning, J.H. (1988), Explaining International Production, London: Unwin Hyman. Dunning, J.H. (1992), Multinational Enterprises and the Global Economy, Wokingham: Addison-Wesley. Dunning, J.H. (1998), ‘Location and the Multinational Enterprise: A Neglected Factor?’, Journal of International Business Studies, 29 (1), 45–66. Duranton, G. and H.G. Overman (2005), ‘Testing for Localization Using Microgeographic data’, The Review of Economic Studies, 72(4), 1077–106. Ellison, G. and E.L. Glaeser (1997), ‘Geographic Concentration in U.S. Manufacturing Industries: A Dartboard Approach’, Journal of Political Economy, 105(5), 889–927. Fan, C.C. and A.J. Scott (2003), ‘Industrial Agglomeration and Development:

ERIKSSON 9780857930088 PRINT.indd 62

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

63

A Survey of Spatial Economic Issues in East Asia and a Statistical Analysis of Chinese Regions’, Economic Geography, 79(3), 295–319. Feser, E.J. and E.M. Bergman (2000) ‘National Industry Cluster Templates: A Framework for Applied Regional Cluster Analysis’, Regional Studies, 34(1), 1–19. Ge, Y. (2009), ‘Globalization and Industry Agglomeration in China’, World Development, 37 (3), 550–59. Glaeser, E.L., H.D. Kallal, J.A. Scheinkman and A. Shleifer (1992), ‘Growth in Cities’, Journal of Political Economy, 100(6), 1126–52. Granovetter, M. (1985), ‘Economic Action and Social Structure: The Problem of Embeddedness’, American Journal of Sociology, 91 (3), 481–510. Gu, C.L., J.F. Shen, K.Y. Wong and F. Zhen (2001), ‘Regional Polarization under the Socialist-market System Since 1978: A Case Study of Guangdong Province in South China’, Environment and Planning A, 33(1), 97–119. Guangdong Provincial Government (2005), Specialized Towns in Guangdong [in Chinese], Guangzhou, China: Press of Yangcheng Evening News. Guangdong Statistical Bureau (1990–2006), Guangdong Statistical Yearbook, 1990–2006 (various annual issues), Beijing, China: China Statistical Press. He, C. (2003), ‘Location of Foreign Manufacturers in China: Agglomeration Economies and Country of Origin Effects’, Papers in Regional Science, 82(3), 351–72. He, C., Y.H.D. Wei and F. Pan (2007), ‘Geographical Concentration of Manufacturing Industries in China: The Importance of Spatial and Industrial Scales’, Eurasian Geography and Economics, 48(5), 603–25. He, C., Y.H.D. Wei and X. Xie (2008), ‘Globalization, Institutional Change, and Industrial Location: Economic Transition and Industrial Concentration in China’, Regional Studies, 42(7), 923–45. Holmes, T.J. and J.J. Stevens (2002), ‘Geographic Concentration and Establishment Scale’, Review of Economics and Statistics, 84(4), 682–90. Hsu, J.Y. (2006), ‘The Dynamic Firm-territory Nexus of Taiwanese Informatics Industry Investments in China’, Growth and Change, 37(2), 230–54. Humphrey, J. and H. Schmitz (2002), ‘How does insertion in global value chains affect upgrading in industrial clusters?’, Regional Studies, 36(9), 1017–27. Ivarsson, I. and C.G. Alvstam (2005), ‘Technology Transfer from TNCs to Local Suppliers in Developing Countries: A Study of AB Volvo’s Truck and Bus Plants in Brazil, China, India, and Mexico’, World Development, 33(8), 1325–44. Kenney, M. and R. Florida (1992), ‘The Japanese Transplants: Production Organization and Regional Development’, Journal of the American Planning Association, 58(1), 21–49. Kloosterman, R.C. and B. Lambregts (2001), ‘Clustering of Economic Activities in Polycentric Urban Regions: The Case of the Randstad’, Urban Studies, 38(4), 717–32. Krugman, P. (1991), ‘Increasing Returns and Economic Geography’, Journal of Political Economy, 99(3), 483–99. Leung, C.K. (1990), ‘Locational Characteristics of Foreign Equity Joint Venture Investment in China, 1979–85’, Professional Geographer, 42(4), 403–21. Li, C.S.J., J. Henley, J. Ansell and T.P. Dong (2011), ‘Location-specific Factors, Localisation Strategy, and Firm Performance: A Case Study of Taiwanese Manufacturing MNE Subsidiaries Investing in China’, Tijdschrift voor economische en sociale geografie, 102(4), 426–40.

ERIKSSON 9780857930088 PRINT.indd 63

29/01/2013 16:59

64

Clusters and economic growth in Asia

Lin, G.C.S. (1997), Red Capitalism in South China: Growth and Development of the Pearl River Delta, Vancouver: University of British Columbia Press. Markusen, A. (1996), ‘Sticky Places in Slippery Space: A Typology of Industrial Districts’, Economic Geography, 72(3), 293–313. Marshall, A. ([1890] 1920), Principles of Economics: An Introductory Volume (8th edition), London: Macmillan. Martin, R. and P. Sunley (2003), ‘Deconstructing Clusters: Chaotic Concept or Policy Panacea?’, Journal of Economic Geography, 3(1), 5–35. Maurel, F. and B. Sédillot (1999), ‘A Measure of the Geographic Concentration in French Manufacturing Industries’, Regional Science and Urban Economics, 29(5), 575–604. McCann, P. and R. Mudambi (2004), ‘The Location Behavior of the Multinational Enterprise: Some Analytical Issues’, Growth and Change, 35(4), 491–524. Ministry of Economic Affairs of Taiwan (2010), 2010 Yearbook of China’s Information Industry [in Chinese], Taipei, Taiwan: Ministry of Economic Affairs. National Bureau of Statistics of China (1990–2006), China Statistical Yearbook, 1990–2006 (various annual issues), Beijing, China: China Statistical Press. O’Donoghue, D. and B. Gleave (2004), ‘A Note on Methods for Measuring Industrial Agglomeration’, Regional Studies, 38(4), 419–27. Reed Electronics Research (2007), The Yearbook of World Electronics Data 2007, Oxon, UK: Reed Electronics Research. Sukkoo, K. (1995), ‘Expansion of Markets and the Geographic Distribution of Economic Activities: The Trends in U.S. Regional Manufacturing Structure, 1860–1987’, Quarterly Journal of Economics, 110(4), 881–908. Tong, X. and J. Wang (2002), ‘Global–Local Networking of PC Manufacturing in Dongguan, China’, in Roger Hayter and Richard Le Heron (eds), Knowledge, Industry and Environment: Institutions and Innovation in Territorial Perspective, Burlington, VT: Ashgate, pp. 67–86. United Nations Conference on Trade and Development (UNCTAD) (2001), World Investment Report: Promoting Linkages, New York and Geneva: United Nations. Wang, J. and X. Tong (2005), ‘Industrial Clusters in China: Embedded and Disembedded?’, in C.G. Alvstam and E.W. Schamp (eds), Linking Industries Across the World: Process of Global Networking, Aldershot: Ashgate, pp. 223–42. Wang, J. and K.N. Xu (2004), ‘Regional Agglomeration and Spatial Evolution of FDI’, China Industrial Economy [in Chinese], 12, 61–7. Wei, Y.H.D. (2010), ‘Beyond New Regionalism, Beyond Global Production Networks: Remaking the Sunan Model, China’, Environment and Planning C: Government and Policy, 28(1), 72–96. Wei, Y.H.D., I. Liefner and C.H. Miao (2011), ‘Network Configurations and R&D Activities of the ICT Industry in Suzhou Municipality, China’, Geoforum, 42(4), 484–95. Wu, J.M. (1997), ‘Strange Bedfellows: Dynamics of Government–Business Relations Between Chinese Local Authorities and Taiwanese Investors’, Journal of Contemporary China, 6(15), 319–46. Yang, C. (2006), ‘Overseas Chinese Investments in Transition: The Case of Dongguan’, Eurasian Geography and Economics, 47(5), 604–21. Yang, C. (2007), ‘Divergent Hybrid Capitalisms in China: Hong Kong and

ERIKSSON 9780857930088 PRINT.indd 64

29/01/2013 16:59

Industrial agglomeration of Taiwanese electronics firms

65

Taiwanese Electronics Clusters in Dongguan’, Economic Geography, 83(4), 395–420. Yang, C. (2009), ‘Strategic Coupling of Regional Development in Global Production Networks: Redistribution of Taiwanese Personal Computer Investment from the Pearl River Delta to the Yangtze River Delta, China’, Regional Studies, 43(3), 385–407. Yang, C. (2012), ‘Restructuring the Export-oriented Industrialization in the Pearl River Delta, China: Institutional Evolution and Emerging Tension’, Applied Geography, 32(1), 143–57. Yang, Y.R. and C.J. Hsia (2005), ‘The Local Embeddedness of the Transborder Production Networks and the Evolution of Local Institution: A Case Study of Greater Dongguan Area’s IT Cluster’, Journal of City and Planning [in Chinese], 32(3), 275–99. Yang, Y.R. and C.J. Hsia (2007), ‘Spatial Clustering and Organizational Dynamics of Transborder Production Networks: A Case Study of Taiwanese Informationtechnology Companies in the Greater Suzhou Area, China’, Environment and Planning A, 39(6), 1346–63. Yang, C. and H. Liao (2010a), ‘Backward Linkages of Cross-border Production Networks of Taiwanese PC Investment in the Pearl River Delta, China’, Tijdschrift voor economische en sociale geografie, 101(2), 199–217. Yang, C. and H. Liao (2010b), ‘Industrial Agglomeration of Hong Kong and Taiwanese Manufacturing Investment in China: A Town-level Analysis in Dongguan’, Annals of Regional Science, 45(3), 487–517. Yeung, H.W.C. (2004), Chinese Capitalism in a Global Era: Towards Hybrid Capitalism, London: Routledge. Yeung, H.W.C. (2007), ‘From Followers to Market Leaders: Asian Electronics Firms in the Global Economy’, Asia Pacific Viewpoint, 48(1), 1–25.

ERIKSSON 9780857930088 PRINT.indd 65

29/01/2013 16:59

4.

The rise of the biomedical cluster in Wonju, Korea* Jun Koo and Jongmin Choi

INTRODUCTION Since Porter’s (1990) popular book, The Competitive Advantage of Nations, the concept of industry cluster has been a mantra for many scholars as well as practitioners. In particular, despite the neoclassic economic theories that ignore the role of geography in the economic space, the geographic concentration of economic activities is the most striking empirical feature characterizing the industry cluster. The presence of fast-growing cities is strong evidence for the agglomeration phenomenon that absorbs and concentrates significant resources in one place. Cases such as Silicon Valley or the Research Triangle Park are show cases demonstrating that industry cluster-driven cities are the most important foci of national growth (Scott and Storper, 2003). Consequently, the last decade has witnessed a surge in cities that have introduced a wide variety of strategic efforts to become the next Silicon Valley or Research Triangle Park. In particular, the biomedical cluster has attracted significant attention recently. According to a recent survey of 77 state and local economic development agencies in the USA, some 83 per cent responded that the biotech-related industries are main development targets (Grudkova, 2001). This trend is well observed in the relative share of venture capital investment in US biomedical start-ups. The share of venture capital deals and investment going to biomedical start-ups increased from 10 per cent to 26 per cent and 8 per cent to 35 per cent, respectively, over the last decade (Shane, 2008). Despite such heightened interest in the industry cluster, particularly in the biomedical field, only a handful of studies have paid some attention to the evolution of a successful industry cluster. How to ignite the cluster engine is a million dollar policy question. In particular, the biomedical field has a reason to draw significant attention in the cluster policy discourse because, as many existing studies argued, there is a relatively more visible player in the development process, namely research universities. 66

ERIKSSON 9780857930088 PRINT.indd 66

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

67

However, few studies offer an answer to such a question. Several studies suggest that the modus operandi of well-functioning established clusters is different from that of emerging ones (Bresnahan et al., 2001; Koo et al., 2009). This means that well-known success factors, such as research universities and human capital, may serve only as necessary conditions for the development of the industry cluster. In addition, regional heterogeneity across different countries may vary substantially, but there is a dearth of research on this topic for Asian countries, which have actively adopted cluster strategies to build their industrial competitiveness. Considering a heightened interest in Asian economies and their recent successes, this is a significant gap in the cluster literature. Against this background, this study investigates the aforementioned policy question. To understand how to spark the engine of the biomedical cluster one needs to investigate the evolutionary process of the industry cluster. We particularly chose a recent biomedical cluster development case in Wonju, Korea. Korea is one of the leading industrialized countries in Asia, which has experienced a substantial growth in the biomedical field based on the cluster strategy. In particular, the Wonju region can offer an interesting success story with a unique perspective. The driving forces behind Wonju’s success have been prosperous local firms, which we call ‘star ventures’, active intermediary organizations and supportive local government. This differentiates the Wonju case from business-driven US clusters such as the San Jose region as well as government-driven Japanese clusters such as the Kansai region. An analysis of Wonju in a comparative context can, therefore, provide important insights to find the answer for the proposed question. The rest of the chapter proceeds as follows. In the next section, we briefly review the theoretical foundations for the cluster concept. The third section describes a rise of the biomedical cluster in Wonju. The fourth analyses the key elements that sparked the Wonju’s biomedical cluster. The last section summarizes the findings and discusses their implications in a comparative context.

INDUSTRY CLUSTER: THEORY AND PRACTICE Industry cluster has been a mantra for many scholars as well as policymakers. The concept has gained significant popularity recently, and it is difficult to find any region without some kind of cluster strategies. However, the origin of the cluster concept dates back to as early as Marshall ([1890] 1920) and Weber ([1909] 1929). Since their seminal research, many economists, regional scientists, and planners have studied

ERIKSSON 9780857930088 PRINT.indd 67

29/01/2013 16:59

68

Clusters and economic growth in Asia

the geographic concentration of economic activities. Inspired by these early studies, a number of recent studies have shown that geographic ‘clustering’ or ‘agglomeration’ of economic activities improves productivity and creates the externalities that lead to increasing returns to scale of production (Carlino, 1979; Beeson, 1987; Fogarty and Garofalo, 1988; Feser, 2001; Henderson, 2003; Porter, 2003; Wheeler, 2004; Andersson et al., 2007). This early concept of geographic concentration of economic activities was often called industrial district and was later repackaged as the industry cluster by Porter (1990, 1998). According to Porter as well as Marshall, the sources of the industry cluster that create productivity-enhancing externalities are threefold. First, firms may co-locate looking for highquality suppliers (Abdel-Rahman, 1988). A location close to high-quality suppliers may lower the firm’s transaction costs, as well as transportation costs. A network of intermediate input suppliers located in proximity may facilitate the exchange of information and ideas that likely improve the quality of products as well as production efficiency. A good example is the Bluegrass Auto Manufacturers Association, which was initiated and supported by Toyota Corp. located in Kentucky, USA (Feser and Koo, 2001). Toyota sends engineers to its nearby parts suppliers to exchange ideas and engage in joint problem-solving. Second, the geographic colocation may benefit firms by tapping into a large labour pool (Helsley and Strange, 1990). Second, firms can lower the search cost for skilled workers if co-located firms share the local labour market for workers of similar calibre. The advantage of a large engineer pool in Silicon Valley is a case in point. Saxenian (1991, 1994) pointed out that the competitiveness of the region can be attributed in large part to the depth and flexibility of the local labour market. Firms can easily find high-calibre engineers in the region, and workers can easily move from one firm to another with their knowledge and experience intact. Such a vibrant local labour market may attract more firms and workers, which can increase the geographic agglomeration of economic activities. Finally, knowledge spillovers between co-located firms may facilitate innovations and can accordingly result in a productivity increase for firms (Ogawa and Fujita, 1980; Fujita and Ogawa, 1982). The presence of localized knowledge spillovers among co-located firms is considered a particularly important determinant of regional competitiveness in the twenty-first-century knowledge economy. This is because industries that characterize the knowledge economy (such as biotech and IT) depend for their success heavily on innovations, which can be substantially influenced by free flows of information and ideas among individuals and firms. The

ERIKSSON 9780857930088 PRINT.indd 68

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

69

industry cluster, where highly skilled workers operate in close proximity, often creates the perfect environment that stimulates innovations through intense face-to-face interactions. In addition, the localization of knowledge spillovers can further strengthen the agglomeration of firms (Koo, 2005). Against this background, the next logical question regarding the biomedical cluster is what elements characterize a region with a proven record of successful cluster development. A great number of studies have addressed this question, and findings can be summarized as follows (Koo et al., 2009). First, the development of knowledge-intensive clusters, such as biotech, hinges on strong science-based knowledge creation activities. Accordingly, major research universities, as knowledge creators, are considered one of the most important necessary conditions for developing a successful biomedical cluster (Feldman, 2000; Prevezer, 2001). For instance, Audretsch and Stephan (1996) showed that biotech firms recruit knowledge workers heavily from local research universities. In addition, Zucker and Darby (1996) and Zucker et al. (1998) found that biotech firms tend to locate in close proximity to star scientists. However, the presence of world-class research universities is not a sufficient condition for biomedical cluster development. Baltimore, the home of one of the most prestigious research universities and hospitals in the world (Johns Hopkins University and its affiliated hospitals), is not considered a hotbed for biomedical clusters. Second, entrepreneurial activities are another critical element for the development of a knowledge-intensive industry cluster. Newly created knowledge becomes economically meaningful only after it has been brought to the market. The role of entrepreneurs is this link that connects new ideas to the market (Schumpeter, 1934; Acs et al., 2003). This is particularly the case in the biomedical sector. Well-known entrepreneurial scientists, such as Walter Gilbert from Harvard (founder of Biogen) and Herbert Boyer from the University of California at San Francisco (founder of Genentech), are at the centre of the success of the US biotech industry. Along the same line of research, recent studies have focused on how the formation of clusters is stimulated by local entrepreneurship (Feldman and Francis, 2003, 2004). The importance of entrepreneurship implies that venture capital can play a crucial role in cluster development. For instance, throughout the history of US biotech development, venture capitalists have bridged the gap between scientists and entrepreneurs, serving as the driving force behind many successful early biotech firms (Prevezer, 2001). Third, previous studies have found that prior experience and knowledge in the related field plays an important role in high-tech ventures

ERIKSSON 9780857930088 PRINT.indd 69

29/01/2013 16:59

70

Clusters and economic growth in Asia

(Shane and Venkataraman, 2000; Feldman, 2001; Feldman and Francis, 2004). This suggests that the presence of strong related industries can be conducive for cluster development. Empirical evidence for this line of argument is prevalent. A significant number of biotech entrepreneurs in the Washington DC area are former employees of the National Institute of Health (NIH) or other biotech-related established firms (Feldman, 2001). In addition, Orsenigo (2001) and Zeller (2001) showed that the location of successful biotech firms is often associated with a strong presence of the chemical industry. A brief review of the state of the art in the cluster theory shows that the current literature is lacking in two ways. As mentioned earlier, few studies have investigated what factors spark the cluster machine. The early stage of cluster formation is still considered a black box and needs to be unpacked. In addition, most existing cluster studies have investigated cases in the USA and Europe. Despite the strong economic performance of East Asian economies over the last several decades, relatively little attention has been paid to the industry cluster in Asia. Only a handful of studies have investigated cluster development and its role in the growth processes of China and Japan (Sonobe et al., 2002; Fan and Scott, 2003; Otsuka, 2006; Collins, 2008).

WONJU BIOMEDICAL CLUSTER IN KOREA Asia is a rising powerhouse in biomedical research. Major biomedical companies, particularly pharmaceuticals like GlaxoSmithKline, Pfizer and Novartis, have chosen Asian countries as their research and development locations as R&D productivity decreases and the cost of new drug development rises. Countries such as Japan, China, Singapore and Korea offer a well-trained R&D workforce and favourable regulatory environment and are being transformed into biomedical R&D hubs. In addition, many expatriates working for major biotech companies in the USA and Europe often bring their ideas and money back to their home countries. Accordingly, a favourable environment for biomedical cluster development has recently been formed, and there have been a few very successful cases, such as the Biopolis in Singapore (Parayil, 2005; Waldby, 2009). As a result, according to estimations by Global Seed Capital,1 the second largest biomedical market, after North America, emerged in Asia with estimated total revenues of $100 billion in 2010. However, there has been a dearth of research on the biomedical cluster in Asia. In particular, few studies explore cases in Korea, which has the potential to become one of the leading players in the biomedical field in the near future. The rise of Wonju is a particularly

ERIKSSON 9780857930088 PRINT.indd 70

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

71

interesting case because of its somewhat unexpected emergence as a promising biomedical cluster from a former military base. To investigate how Wonju ignited its cluster engine, we conducted seven in-depth interviews with local key players in universities and public organizations as well as 13 local firms that witnessed the process closely. Secondary data from the public domain were then combined with the interviews for the interpretation. Past and Present Until the early 1990s, Wonju had been well known as a military base, and there was a significant lack of manufacturing activities. The presence of the military base in Wonju created a negative impression of the city, which made most firms reluctant to locate their offices in the city despite the geographic proximity to the capital region. In the mid-1990s, the city government launched an initiative to improve the negative impression of Wonju and to transform the city’s fate as a lagging periphery. The focus of this initiative was a bold plan to develop a biomedical cluster in the region. This idea originally came from a local university faculty, the well-known biomedical engineering department at Yonsei University, which was a close partner of the city government in the early stage of cluster development. As part of the effort to develop a biomedical cluster, the city government in partnership with Kangwon Province and Yonsei University applied for the TechnoPark Development Project in 1997. The purpose of the Techno-Park Development Project, which was fully funded by the central government, was to build technology-based local industrial clusters across the country. The central government, however, disapproved the city’s application to develop a biomedical device cluster in the region.2 Despite the initial failure, Wonju made independent efforts and established a business incubation centre as its first step towards a successful biomedical cluster development. The beginning of the centre was modest. About 11 start-up firms were originally housed in the centre. However, without any noticeable support from the central government, which had been considered a necessary ingredient for success in Korea (Rowen, 2007), Wonju brought about some achievement. The city consistently strengthened its manufacturing infrastructure in the biomedical sector. For instance, it procured the Tea-Jang agricultural industrial complex in 1999 and built a new biomedical industrial complex. Additionally, a new industrial complex was built on another site in 2003. The city also established an intermediary organization, the Wonju Medical Industry Techno Valley (MITV), which played a pivotal role in coordinating and

ERIKSSON 9780857930088 PRINT.indd 71

29/01/2013 16:59

72

Table 4.1

Clusters and economic growth in Asia

Biomedical cluster in Wonju

Firms Sales ($ billion) Employment

2003

2005

2007

2009

50 0.4 350

60 0.6 693

79 2.0 1259

106 3.1 2430

Source: Wonju Medical Industry Techno Valley.

Table 4.2

Biomedical activities in major regions in Korea

2006 Sales (% of national total) Export (% of national total)

Southern Wonju Kyunggi

Seoul Chungnam Pusan

DaeguKyungbuk

33.9

15.9

12.8

7.4

5.9

6.2

29.7

29.5

5.2

13.8

1.4

5.0

Source: Wonju Medical Industry Techno Valley.

managing the industry–university network. The main roles of Wonju MITV were the management of specialized facilities as well as education and networking among universities, research institutes, businesses and the local government. In addition, the newly established organization expanded its horizon to include business recruitment and marketing. This movement was a collaborative effort of the city government, university researchers and local business community leaders. As a result of such early efforts, Wonju biomedical cluster was designated as one of the National Innovative Cluster Complexes by the central government in 2004 and received a substantial amount of government funding. In 2009, some 106 biomedical-related firms were located in the Wonju area with estimated sales of 3.1 billion dollars, employing around 2500 workers (see Table 4.1). These numbers are not impressive in comparison with other established successful biomedical clusters, such as San Jose or RaleighDurham. However, the number of biomedical firms and employment in Wonju has increased fivefold since 2002, which is substantially higher than the national average (about 5 per cent annual growth). Although the city’s biomedical cluster is still in an early stage of development, the performance of the cluster over the last ten years indicates that the region has a strong potential to become a leading player in the biomedical industry. Table 4.2 shows regional shares of sales and export of the biotech sector

ERIKSSON 9780857930088 PRINT.indd 72

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

73

in Korea. One interesting point is that Wonju accounts for 30 per cent of total exports, although the local firms claim only about 16 per cent of the domestic market. This finding indicates that firms in the Wonju cluster are more active in the international market, which also suggests that these firms likely have strong competitiveness and growth potential. According to interviewees from the Wonju MITV, the firms in the Wonju cluster strengthen their competitiveness by competing in overseas markets. Accordingly, they argued, these firms may grow bigger and faster than other Korean firms, which mostly serve a smaller and more restricted domestic market.

DRIVERS OF CLUSTER FORMATION Role of the University as a Knowledge Producer and Innovative Entrepreneur Although the region still has a long way to go in comparison with other competitive bio-clusters, Wonju seems to be going through a successful take-off stage of cluster formation. Local universities lie at the centre of this process. This is not surprising given the strong science bases of the biomedical sector. A strong role of local research universities is typically found in other major biomedical clusters across the globe such as San Jose, Boston, Tokyo, Munich and Cambridge. As a matter of fact, scientists from eminent research universities established the early successful bio-ventures (Prevezer, 2001). For instance, the early successful bio-ventures, Cetus, Genentech, Biogen and Hybritech, are cases in point. A close relationship between academia and industry in the early stages of biomedical development is inevitable. In the case of Wonju, this is a particularly intriguing point given the country’s unique development history. Korea has a long history of state-driven development strategy. The central government made important policy decisions to coordinate the allocation of limited national resources. The current economic status of Korea, a leading developing country with advanced technologies, can be attributed to this paradigm from the 1960s to the 1980s. Although state-driven development strategies were considered effective during the early development stage in Korea, as the economy has matured and the complexity of the economic system deepened, such an approach has faced a wide variety of issues. The central government is no longer capable of taking into account increasingly complex needs of individuals, firms and local communities. As the role of the central government shrinks, the

ERIKSSON 9780857930088 PRINT.indd 73

29/01/2013 16:59

74

Table 4.3

Clusters and economic growth in Asia

Early start-ups nurtured in the Wonju Business Incubator

Start-up Name

Main Products

Nurturing Period

Korea Optical Telecom ChoongWae Medical

Blood gas analyser

May 1998–May 2000

Infant incubator, patient monitor Oximeter Low frequency therapy Electroencephalogram tester Volumetric infusion pump Measuring instrument for animal experiments, electronic stimulator Automatic film processor Multi-mode infusion pump Jaundice treatment, patient monitor Medical information software

May 1998–May 2000

DongSeo Hi-Tech Odicine MeeRe Engineering Biotron Medisco

DongYang Medical Cals Medical Mediana Ahone Information & Communication

June 1998–May 2000 June 1998–Sep 1999 May 1998–Sep 1999 July 1998–Oct 1999 July 1998–Oct 1999

May 1998–Apr 1999 May 1998–Oct 1999 May 1998–Dec 1999 May 1998–Aug 1999

Source: Wonju Medical Industry Techno Valley.

private sector is increasingly playing an important role in the development process. Wonju’s success epitomizes this trend. According to key interviewees who have played a significant role in forming the Wonju biomedical cluster, the aforementioned failure of receiving support from the central government through the Techno-Park Development Project opened new opportunities to the region. Many interviewees criticized the inefficiency of the national-level cluster programmes. According to them, the central government pays little attention to unique needs of local communities and often provides one-size-fits-all support, which is not relevant given the development stage of the region (for example, buying unnecessary equipment). When Wonju failed to be selected for the Techno-Park Development Project funded by the central government, the temporary vacuum was filled by the city government, local universities and entrepreneurs. In particular, Yonsei University played a pivotal role in the early stage of cluster development. Researchers at Yonsei University pushed an initiative to launch local efforts to build a biomedical cluster in the region. The most important outcome of such efforts was the births of the 11 start-up companies founded by Yonsei University graduates (see Table 4.3). These

ERIKSSON 9780857930088 PRINT.indd 74

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

75

new ventures were housed in the Wonju Business Incubator, which was co-sponsored by the city and the university. New technologies developed in university labs were brought to the market through the coordinated efforts of university researchers and the city. In addition, many of these start-ups have grown to become major players in the biomedical field in Korea. For instance, Mediana exported over $17 million worth of products all over the world in 2006. ChoongWae Medical, a producer of infant incubators, successfully entered foreign markets after receiving approval for its products from the US Food and Drug Administration. The success of these early ventures was critical for the growth of the biomedical cluster in the region. Role of the Local Government as a Resource Allocator and Coordinator Although we argued, against a popular belief, that the role of the central government was limited in the development of the Wonju cluster, that does not mean there has been no room for the government to act. In particular, the city government was the catalyst that coordinated the joint efforts to build a biomedical cluster in the region. Such a unique role of the local government as a catalyst in the early stage of cluster development is also found in other successful biomedical clusters, such as the Research Triangle Park (Koo et al., 2009). Although the later growth stage is dominated by market forces, the early cluster development stage requires extra efforts of an impartial player to orchestrate a wide range of local efforts. The early efforts of the Wonju city government led to subsequent support from the central government as well. Wonju was selected for the Regional Research Centre Programme, funded by the Ministry of Science and Technology in 1999. This programme was designed to provide funding for small technology firms to purchase R&D facilities. The timing was perfect. When 11 early-stage start-ups housed in the Wonju Business Incubator had some trouble because of the lack of funding for research facilities in the late 1990s, this programme provided timely financial support for these technology-intensive start-ups. In addition, Wonju attracted the Technology Innovation Centre Programmes in 2004, which infused $20 million into the region for research. That is, critical financial support from the central government arrived in the region right on time due to the coordinated efforts of the local government and the community. Wonju’s experience provides interesting implications regarding the division of labour between the central and local governments in the cluster development process. Direct support from the central government has often taken the one-size-fits-all approach, which does not reflect actual needs of the region. Therefore, the local government is in a better position

ERIKSSON 9780857930088 PRINT.indd 75

29/01/2013 16:59

76

Clusters and economic growth in Asia

to understand local needs and can provide customized support for local businesses. This also implies that the support from the central government may be more effective when it is indirect and subtle. This point is a key difference between Wonju and leading biomedical clusters in Japan, such as Tokyo and Osaka. Similar to Korea, Japan is often characterized as a state-driven economy. In addition, Japan is well known for its structured biomedical cluster development programme. Recognizing the importance of the biomedical field for future growth, Japan launched the Biotechnology Strategy Council comprising 12 members including the Prime Minister, the Chief Cabinet Secretary and the Minister for Science and Technology Policy. Local governments at the prefectural level are also important drivers of the biomedical cluster strategy in Japan, but their roles are somewhat limited and orchestrated by the central government. The central government still has a strong grip based on a significant biomedical research budget distributed through four government ministries: the Ministry of Economy, Trade, and Industry, the Ministry of Education, Culture, Sports, Science and Technology, the Ministry of Agriculture, Forestry and Fisheries and the Ministry of Health, Labour and Welfare.3 Role of the Intermediary Organization as a Network Catalyst and Management Consultant Many interviewees who were involved in forming the Wonju cluster put the Wonju MITV, a major intermediary organization in the region, at the centre of the cluster development process. The intermediary organization is a semi-governmental institution that connects public and private areas. Wonju MITV created a network environment among local researchers, entrepreneurs, public officials and the community. Figure 4.1 illustrates the network of primary players in the Wonju biomedical cluster. Many nearby universities4 supplied a well-trained labour force and shared cutting-edge technology with local firms through Wonju MITV, which served as a network catalyst that linked key stakeholders in the region. MITV also collaborated with the city government to secure funding to purchase research facilities. On the other hand, local firms often supported university research, which was applied to their product development. Such close relationships between firms and universities were indispensible to the formation of the cluster in Wonju. One of the difficulties that many early-stage start-up firms encounter is how to manage their businesses to survive and eventually grow. This is because most founders of biomedical ventures are from university labs and usually lack management skills. In order to overcome such limitations,

ERIKSSON 9780857930088 PRINT.indd 76

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

University Educating human resource Supporting research

77

Enterprise Increasing export and employment Developing local industry

Wonju Medical Industry TechnoValley Building and managing networks among firms, university and governments

Central and local government Providing research infra and financial support

Figure 4.1

Inter-organizational network in the Wonju biomedical cluster

Wonju MITV provided entrepreneur management programmes for start-up firms. The purpose of the programme was to provide consulting services for early-stage firms, ranging from market research and product planning to overseas market development. Our interviewees underscored the importance of Wonju MITV as such a management consultant. Many previous studies reported findings about the importance of management skills and know-how to entrepreneurial success (Feser and Koo, 2001; Koo et al., 2009). These studies argued that management know-how and excellent managers play a significant role in achieving entrepreneurial success. In advanced countries, such as the USA, venture capital firms mainly play the role of management consultants. However, in many developing countries, where venture capital activities are relatively weak, the intermediary organization often fills the gap. Although Korea is often categorized as an advanced country, this may differentiate the Wonju case from cities in other advanced countries. The role of intermediary organizations is not as significant in US biotech hotspots such as San Jose and the Research Triangle Park. Given that Wonju MITV was established through collaborative efforts of the city government and major players in the private sector, the Wonju cluster can epitomize another prototype of cluster development, in between the business-driven US type and the governmentdriven Japanese type.

ERIKSSON 9780857930088 PRINT.indd 77

29/01/2013 16:59

78

Clusters and economic growth in Asia

The Role of Star Ventures as Catalysts of Cluster Engines Universities, governments, entrepreneurs, and venture capitalists are common players in cluster development that have been scrutinized in many early cluster studies (Audretsch, 2001; Bresnahan et al., 2001; Feldman and Francis, 2003, 2004; Feldman et al., 2005; Koo et al., 2009). In addition to these, the intermediary organization is also considered as an important factor that facilitates cluster development in some studies (Kodama, 2008). In the case of Asian countries such as Korea, where the public sector traditionally exercises strong influence on the market, the role of intermediary organizations is often more visible. However, the presence of these factors does not necessarily lead to successful cluster development. A close examination of the Wonju cluster reveals another factor that served as a catalyst for the cluster engine. According to our interviewees who led Wonju’s biomedical cluster in the nascent stage, the city’s efforts to attract biomedical firms to the region were futile in the beginning because most firms were concerned about the region’s unattractive attributes, such as the lack of infrastructure, the negative impression as a military city, and the distance from Seoul. However, such concerns slowly dissipated as the success stories of several local biomedical ventures became known to the market. We would like to call them star ventures. These star ventures, started in the Business Incubator Centre in Wonju, have earned multi-million dollar revenues and attained success in overseas markets in a relatively short period. Fairytale success stories of star ventures spread through local networks and slowly changed the image of the city. The best-known examples of star ventures are the cases of Mediana and Human-Tech. These companies received prestigious government awards in 2003 in recognition of their outstanding performance in domestic as well as overseas markets. In particular, Human-Tech succeeded in commercializing its lab product in 1999 and earned more than $5 million from exports in 2005 alone. These stories and the changed image of the city sent a positive signal to the market. The ‘I can do it like them’ spirit led many local latent entrepreneurs to the market place with new ideas. In addition, firms in other regions started paying attention to the city as their potential site. Successful bio-ventures such as I-Sense and Nuga Medical are cases in point that have located facilities in Wonju recently. The results of such changes are well reflected in the trends of biomedical firms and sales in Table 4.1. The number of firms and sales increased by over 100 per cent and 900 per cent respectively during the 2003–09 period. According to our interviewees, the status of Wonju as a hub for biomedical activities seems to

ERIKSSON 9780857930088 PRINT.indd 78

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

79

have been established, and firms started moving into the city without significant recruitment efforts. The most important finding of this study is that the presence of successful venture firms in the early cluster development stage can play a pivotal role in the subsequent growth of the cluster. Successful venture firms as catalysts of cluster engines stimulate potential local entrepreneurs as well as non-local firms to locate in the region. This phenomenon is similar to a halo effect. In this study, we would like to call it the ‘star venture effect’. The star venture effect is not a unique phenomenon observed only in Wonju. In fact, the effect is often observed among innovative clusters in other countries. For instance, well-known high-tech clusters such as Silicon Valley and Austin in the USA have successful venture stories such as those of HP and Dell. In a similar vein, the Research Triangle Park in the USA also provides important circumstantial evidence for similar halo effects that transformed the image of the region. The Research Triangle Park, established in the 1950s, showed no progress for over ten years. The region took off after IBM and the then US Department of Health, Education and Welfare (now the Department of Health and Human Services) located their research facilities in the region. The location decision of these major institutions created a positive halo effect for the region and changed the region’s negative image based on the tobacco and farming industries. The rest is history. A number of spin-offs were created by former workers of these two institutions and universities nearby, and many high-tech firms established their research facilities and management offices in the region (Link and Scott, 2003). In order to verify the star venture effects in the Wonju biomedical cluster, we carried out short interviews with 13 local biomedical venture firms regarding their location decision factors. The question posed to the firms was what factors made them locate in the Wonju area.5 Nine out of 13 venture firms answered that the benefits from the city government were the most important factor. Six firms selected the expectation and confidence gained from other successful local venture firms as a major location determinant. Two firms responded that knowledge acquisition, convenient traffic and the labour pool were attractive factors. Although it was not an extensive survey, our interviews provide circumstantial evidence to support the star venture effects. Government support also seems to have played a role in biomedical venture firms locating in Wonju. This partially confirms the role of government as a necessary condition for cluster development. A mix of universities, research institutes and entrepreneurs in combination with effective government support may create a hotbed for the emergence of star ventures. In addition, the advent of star ventures may increase the propensity of new firm formation by latent

ERIKSSON 9780857930088 PRINT.indd 79

29/01/2013 16:59

80

Clusters and economic growth in Asia

Necessary conditions for cluster development

Government Resource allocation Acceleration of cluster development Intermediary organization Network management and consulting

University Knowledge creation

Birth of venture firms

Emergence of star ventures

Increase in local entrepreneurship

Research centres Knowledge commercialization

Figure 4.2

Cluster formation process

local entrepreneurs, which accelerates the cluster development process. That is, star ventures may play a role as a key determinant that pushes the cluster from the nascent stage to the take-off stage. Figure 4.2 illustrates this process.

CONCLUSION For many urban scholars and practitioners the cluster strategy has been considered a one-size-fits-all solution for regional development. A great number of regions have adopted the cluster strategy for regional development. However, many of them simply tried to copy what other successful regions, such as Silicon Valley, did. This is a formula for policy failure. Success factors identified in well-functioning clusters are only necessary conditions for successful cluster development. Besides, regional heterogeneity makes it even more difficult to apply the same formula to different places. Against this background, this study aimed to achieve two things. First, we tried to unpack the cluster development process to better understand factors that play a role in the take-off stage of an industry cluster. Second, we focused on the biomedical cluster case of Wonju, Korea since there has been a dearth of research on clusters in Asian countries. Findings of this study can be summarized as follows.

ERIKSSON 9780857930088 PRINT.indd 80

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

81

First, this study confirms the importance of knowledge creators, such as universities and research institutes. Key players in the Wonju biomedical cluster were also local universities, their research subsidiaries and spin-off venture firms. Second, universities and research institutes are only necessary conditions for building knowledge clusters. The presence of world-class hospitals in Baltimore (Johns Hopkins Hospital) and Cleveland (Cleveland Clinic) has not led to successful biomedical cluster development in those regions. These necessary conditions need to be combined with other factors, such as an effective government as a resource allocator and an intermediary organization as a network catalyst, for local cluster seeds to sprout. In particular, the role of intermediary organizations is uniquely strong in Wonju, Korea. Wonju MITV was at the centre of the biomedical cluster development. On the other hand, contrary to a popular belief, the role of the central government was limited. This seems to suggest reduced influence of the central government in the development process as the country moves from the developing country stage to a more advanced and mature country stage. However, unlike US clusters, the cluster development process was not business- or market-driven in Korea. Reflecting the country’s unique development history and characteristics, the influence of the public domain still remains in the form of coordinated efforts of the public and private sectors in intermediary organizations. For many developing countries that have adopted the cluster strategy for regional development, Wonju’s experience may offer an important implication. The immaturity of the venture capital market may hinder the development of knowledge clusters in developing countries. However, this can be overcome by active intermediary organizations and indirect government support through them. Lastly, the most interesting finding of this study was the role of star ventures. The presence of star ventures served as a catalyst for the cluster engine. Local and non-local firms’ location decisions were heavily affected by the presence of local star ventures. In other words, a positive feedback mechanism of cluster growth can be ignited by a strong presence of local star ventures. This indicates that the cluster development strategy may have to adopt a more comprehensive perspective. In addition to traditional strategies to support university research, entrepreneurship policies, such as how to find and nurture promising local ventures and how to help them grow, should be considered together in a comprehensive cluster strategy framework. In addition, the region may have to pay some attention to marketing as well to inform success stories of local ventures more effectively.

ERIKSSON 9780857930088 PRINT.indd 81

29/01/2013 16:59

82

Clusters and economic growth in Asia

NOTES * 1. 2. 3.

4. 5.

This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (NRF-2010-332-B00587). Global Seed Capital LLC is a Boston-based seed venture company with investment interest in media, IT and bio-companies. A significant amount of seed money was provided to the selected cities for the project. Although the central government still exercises significant influence on the cluster-based regional development strategy in Japan, it recently adopted a new framework that gives more attention to the regional context. Kobe is a case in point that illustrates the coordinated efforts of central and local governments to build the biomedical cluster (Collins, 2008). While Yonsei University is at the centre of the network, nine other universities nearby are also involved in the Wonju biomedical cluster development project. Firms were allowed to choose multiple factors.

REFERENCES Abdel-Rahman, H. (1988), ‘Product Differentiation, Monopolistic Competition and City Size’, Regional Science and Urban Economics, 18(1), 69–86. Acs, Z.J., D.B. Audretsch and B. Carlsson (2003), ‘The Knowledge Filter and Entrepreneurship in Endogenous Growth’, paper presented at the North American Regional Science Council, Philadelphia, PA. Andersson, F., S. Burgess and J. Lane (2007), ‘Cities, Matching and the Productivity Gains of Agglomeration’, Journal of Urban Economics, 61(1), 112–28. Audretsch, D.B. (2001), ‘The Role of Small Firms in U.S. Biotechnology Clusters’, Small Business Economics, 17(1–2), 3–15. Audretsch, D.B. and P.E. Stephan (1996), ‘Company–Scientist Locational Links: The Case of Biotechnology’, American Economic Review, 86(3), 641–52. Beeson, P. (1987), ‘Total Factor Productivity Growth and Agglomeration Economies in Manufacturing, 1959–73’, Journal of Regional Science, 27(2), 183–99. Bresnahan, T., A. Gambardella and A. Saxenian (2001) ‘“Old Economy” Inputs for “New Economy” Outcomes: Cluster Formation’, Industrial and Corporate Change, 10(4), 835–60. Carlino, G.A. (1979), ‘Increasing Returns to Scale in Metropolitan Manufacturing’, Journal of Regional Science, 19(3), 363–72. Collins, S. (2008), ‘Knowledge Clusters and the Revitalization of Regional Economies in Japan: A Case Study of the Biomedical Industry in Kobe’, Prometheus, 26(1), 111–22. Fan, C.C. and A.J. Scott (2003), ‘Industrial Agglomeration and Development: A Survey of Spatial Economic Issues in East Asia and a Statistical Analysis of Chinese Regions’, Economic Geography, 79(3), 295–319. Feldman, M.P. (2000), ‘Where Science Comes to Life: University Bioscience, Commercial Spin-offs, and Regional Economic Development’, Journal of Comparative Policy Analysis, 2(3), 345–61. Feldman, M.P. (2001), ‘The Entrepreneurial Event Revisited: Firm Formation in a Regional Context’, Industrial and Corporate Change, 10(4), 861–91. Feldman, M.P. and J.L. Francis (2003), ‘Fortune Favours the Prepared Region:

ERIKSSON 9780857930088 PRINT.indd 82

29/01/2013 16:59

The rise of the biomedical cluster in Wonju, Korea

83

The Case of Entrepreneurship and the Capitol Region Biotechnology Cluster’, European Planning Studies, 11(7), 765–88. Feldman, M.P. and J.L. Francis (2004), ‘Homegrown Solutions: Fostering Cluster Formation’, Economic Development Quarterly, 18(2), 127–37. Feldman, M.P., J. Francis and J. Bercovitz (2005), ‘Creating a Cluster While Building a Firm: Entrepreneurs and the Formation of Industrial Clusters’, Regional Studies, 39(1), 129–41. Feser, E.J. (2001), ‘A Flexible Test for Agglomeration Economies in Two US Manufacturing Industries’, Regional Science and Urban Economics, 31(1), 1–19. Feser, E.J. and J. Koo (2001), Kentucky Clusters: Industrial Interdependence and Economic Competitiveness, Lexington, KY: Kentucky Science and Technology Corporation. Fogarty, M.S. and G.A. Garofalo (1988), ‘Urban Spatial Structure and Productivity Growth in the Manufacturing Sector of Cities’, Journal of Urban Economics, 23(1), 60–70. Fujita, M. and H. Ogawa (1982), ‘Multiple Equilibrium and Structural Transition of Non-monocentric Urban Configurations’, Regional Science and Urban Economics, 12(2), 161–96. Grudkova, V. (2001), The Technology Economy: Why Do Tech Companies Go Where They Go?, Washington, DC: EDA National Forum. Helsley, R.W. and W.C. Strange (1990), ‘Matching and Agglomeration in a System of Cities’, Regional Science and Urban Economics, 20(2), 189–212. Henderson, J.V. (2003), ‘Marshall’s Scale Economies’, Journal of Urban Economics, 53(1), 1–28. Kodama, T. (2008), ‘The Role of Intermediation and Absorptive Capacity in Facilitating University–Industry Linkages: An Empirical Study of TAMA in Japan’, Research Policy, 37(8), 1224–40. Koo, J. (2005), ‘Technology Spillovers, Agglomeration, and Regional Economic Development’, Journal of Planning Literature, 20(2), 99–115. Koo, J., J. Bae and D. Kim (2009), ‘What Does it Take to Become a Biotech Hot Spot?’, Environment and Planning C: Government and Policy, 27(4), 665–83. Link, A.N. and J.T. Scott (2003), ‘The Growth of Research Triangle Park’, Small Business Economics, 20(2), 167–75. Marshall, A. ([1890] 1920), Principles of Economics (8th edition), London: Macmillan. Ogawa, H. and M. Fujita (1980), ‘Equilibrium Land Use Patterns in a Nonmonocentric City’, Journal of Regional Science, 20(4), 455–75. Orsenigo, L. (2001), ‘The (Failed) Development of a Biotechnology Cluster: The Case of Lombardy’, Small Business Economics, 17(1–2), 77–92. Otsuka, K. (2006), ‘Cluster-based Industrial Development: A View from East Asia’, Japanese Economic Review, 57(3), 361–76. Parayil, G. (2005), ‘From “Silicon Island” to “Biopolis of Asia”: Innovation Policy and Shifting Competitive Strategy in Singapore’, California Management Review, 47(2), 50–73. Porter, M.E. (1990), The Competitive Advantage of Nations, London: Macmillan. Porter, M.E. (1998), ‘Clusters and the New Economics of Competition’, Harvard Business Review, Nov–Dec, 77–90. Porter, M. (2003), ‘The Economic Performance of Regions’, Regional Studies, 37(6–7), 545–6.

ERIKSSON 9780857930088 PRINT.indd 83

29/01/2013 16:59

84

Clusters and economic growth in Asia

Prevezer, M. (2001), ‘Ingredients in the Early Development of the U.S. Biotechnology Industry’, Small Business Economics, 17(1–2), 17–29. Rowen, H. (2007), ‘How Governments Shaped their IT Industries’, in Henry S. Rowen, Marguerite Gong Hancock and William F. Miller (eds), Making IT: The Rise of Asia in High Tech, Stanford, CA: Stanford University Press. Saxenian, A. (1991), ‘The Origins and Dynamics of Production Networks in Silicon Valley’, Research Policy, 20(5), 423–37. Saxenian, A. (1994), Regional Advantage: Culture and Competition in Silicon Valley and Route 128, Cambridge, MA: Harvard University Press. Schumpeter, J.A. (1934), The Theory of Economic Development, Cambridge, MA: Harvard University Press. Scott, A. and M. Storper (2003), ‘Regions, Globalization, Development’, Regional Studies, 37(6/7), 549–78. Shane, S.A. (2008), Fool’s Gold? The Truth Behind Angel Investing in America, New York: Oxford University Press. Shane, S. and S. Venkataraman (2000), ‘The Promise of Entrepreneurship as a Field of Research’, Academy of Management Review, 25(1), 217–26. Sonobe, T., D. Hu and K. Otsuka (2002), ‘Process of Cluster Formation in China: A Case Study of a Garment Town’, Journal of Development Studies, 39(1), 118–39. Waldby, C. (2009), ‘Singapore Biopolis: Bare Life in the City-State’, East Asian Science, Technology and Society: An International Journal, 3(2–3), 367–83. Weber, A. ([1909] 1929), Theory of the Location of Industries, Chicago, IL: University of Chicago Press. Wheeler, C.H. (2004), ‘Productivity and the Geographic Concentration of Industry: The Role of Plant Scale’, Working Paper No. 2004-024A, St. Louis, MO: Federal Reserve Bank of St. Louis. Zeller, C. (2001), ‘Clustering Biotech: A Recipe for Success? Spatial Patterns of Growth of Biotechnology in Munich, Rhineland and Hamburg’, Small Business Economics, 17(1–2), 123–41. Zucker, L.G. and M.R. Darby (1996), ‘Star Scientists and Institutional Transformation: Patterns of Invention and Innovation in the Formation of the Biotechnology Industry’, Proceedings of the National Academy of Sciences of the United States of America, 93(23), 12709–16. Zucker, L.G., M.R. Darby and M.B. Brewer (1998), ‘Intellectual Human Capital and the Birth of U.S. Biotechnology Enterprises’, American Economic Review, 88(1), 290–306.

ERIKSSON 9780857930088 PRINT.indd 84

29/01/2013 16:59

5.

The global economic crisis as leverage for emerging regional growth paths? Differentiated evidence from China – three years onwards Daniel Schiller and Henning Kroll

INTRODUCTION The Chinese economy has realized high growth rates and a remarkable speed of transition for a period of about 30 years and led scholars assessing its development trajectory to postulate the emergence of a new model of development (Stiglitz, 2008). It suggests that the ongoing transformation of the economy will in the future be based on the breakthrough of innovation in addition to production (Altenburg et al., 2008) and the advent of a knowledge-based society by the year 2020 as envisaged by the long-term planning of the Chinese government (Kroll and Schiller, 2010). The major building blocks of this new growth model can be summarized as (1) sectorial structural change from traditional to modern industrial branches, (2) re-focusing of the export growth model towards the domestic market and (3) upgrading of technological and organizational capabilities of firms. These industry-level determinants are complemented by regional determinants that are shaped to a large degree by different layers of the state and by the national and respective regional innovation systems. Despite these optimistic appraisals, reports of economic turmoil in China during the global economic crisis in 2008–09 seemed to suggest a need for a reconsideration of the optimistic assumptions made so far. Particularly for the highly export-oriented provinces in the coastal regions of China, the slackening of global demand seemed to forebode dire socioeconomic implications. In the end, however, the Chinese economy picked up with unanticipated momentum, became the world’s largest exporter at the end of 2009 and was back at its pre-crisis strength in late 2010 (OECD, 2010). Quite evidently, this impressive recovery proved those wrong who 85

ERIKSSON 9780857930088 PRINT.indd 85

29/01/2013 16:59

86

Clusters and economic growth in Asia

had predicted a failure of the Chinese growth model under distress conditions. Apparently China had been more resilient than expected and was back on track. Precisely therefore, however, there is a certain danger that analyses of the course and wake of the crisis will remain reduced to a bird’s-eye perspective that might miss some far-reaching structural implications that the crisis may have had on the Chinese economy. As economic geographers, the authors are critical regarding the assumption of ‘a’, that is, one singular, Chinese growth model. Quite evidently there have been a number of different sectorial and regional growth models in China before the crisis – and most likely will be in its wake. This chapter, therefore, will argue that precisely because of the newly gained momentum we are in need of a differentiated understanding of the impact that the crisis has had on different drivers of growth in China. Arguably such a sectorial, regional and firm-level perspective is instructive to anticipate the implications of the crisis for the overall future growth path at the national level. Overall, the considerations put forward in this chapter will be guided by two partly conflicting hypotheses: (1) due to the massive public stimulus package and the competitiveness of Chinese firms, the economy is emerging from the crisis ‘just as before’, on its old growth path, with similar drivers and little structural change; (2) the crisis has been an opportunity for comprehensive structural change and the Chinese economy is thus emerging from the crisis on a different growth path, driven by new sectors and internally transformed firms. As outlined above, it is the aim of this study to reflect on these hypotheses based on disaggregated data on a sectorial and a regional level. The authors will do so by analysing gross industrial output as well as patent application data from different sources. Thus they aim to establish whether there is evidence of structural changes that have occurred as a result of the challenges posed and the coping strategies developed as a result of the economic crisis. Further, the study data are complemented by primary data from a company survey in the Pearl River Delta that was carried out just after the outbreak of the crisis in late 2009. In particular, these data will be used to confirm the entrepreneurial reorientation of firms within a sector with respect to technological upgrading and market orientation. The remainder of the chapter is structured as follows. The second section identifies determinants and related outcomes of the economic crisis in Chinese regions by taking the regional resilience framework of Simmie and Martin (2010) as a starting point. This section will develop guiding assumptions about the influence of industry-level and regional determinants on the outcome of the crisis. The third section applies sectorially and regionally disaggregated secondary data to test the hypotheses at the

ERIKSSON 9780857930088 PRINT.indd 86

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

87

macro level. Additionally, data from a company survey in the Pearl River Delta are used to illustrate the outcomes of the crisis at the micro level. Subsequently, the fourth section discusses the findings with a view to the guiding assumptions developed in the conceptual chapter, and the final section summarizes, concludes and ends with a brief outlook.

DETERMINANTS FOR PATHS OF RESILIENCE, RECOVERY AND FURTHER DEVELOPMENT The identification of determinants of the crisis and their impact on regional economies is related to the recent debate about regional resilience from an evolutionary perspective. This general conceptual issue is closely interconnected with the literature on economic growth and technological catch-up if the focus is explicitly on emerging regions. Based on these conceptual considerations, the pre-crisis patterns and post-crisis outlook are discussed for the case of China. The section concludes with the deduction of hypotheses about the effect of determinants at the industry and regional level on post-crisis development. Resilience and Evolution of Regional Growth and Catch-up Paths During Crisis Simmie and Martin (2010) presented a four-stage adaptive cycle as a conceptual framework to understand regional resilience from an evolutionary perspective. Their framework moves beyond ‘equilibrist’ approaches that simply ask whether a regional economy returns to the existing equilibrium state or to a new one after a major economic shock. Their definition of regional resilience refers to the capacity of regional economies to constantly adapt themselves to changing economic conditions. The adaptive cycle starts with a structural reorganization of the economic system, commences with the exploitation of this new structure, enters a phase of conservation and finally runs into a state of decline that requires the release of the present lock-in. Regional resilience is expected to be high in the early stages of the adaptive cycle and low in the later ones. Based on the adaptive capacity, three stylized responses of regional economies to major economic shocks are distinguished by Simmie and Martin (ibid.). First, if the present growth path is still competitive, a region will simply return to it after a shock has occurred as predicted by equilibrium theorists. Second, if a region has already entered the phase of structural reorganization of the economy to a substantial degree, the shock will provide an opportunity for the breakthrough of a higher growth part.

ERIKSSON 9780857930088 PRINT.indd 87

29/01/2013 16:59

88

Clusters and economic growth in Asia

Finally, if insufficient steps have been taken before the crisis to move away from conserved or declining growth paths, it is quite likely that the regional economy will return to an inferior growth path after the crisis. The impact of the crisis is expected to differ between developed and developing countries. Naudé (2009) discusses six reasons why the outlook for developing countries is rather optimistic. The resilience of certain regions in emerging economies might be much higher than in developed countries and these regions might recover much sooner. China is mainly affected by the reduction in exports to developed countries whereas the other two major transmission mechanisms of the crisis identified by Naudé (ibid.), reduction in bank lending and financial flows to developing countries, are less relevant due to vast domestic fiscal resources. Thus it seems reasonable to expect that the crisis has a leverage effect that puts at least some regional economies within China on a higher growth path than before the crisis. The general patterns of adaptation and resilience are closely connected with the existing understanding of economic development and technological catch-up in emerging economies. Regional economies in these countries aim to improve their technological capabilities and market shares to catch up with incumbent regions and to move from quantitative to qualitative growth (Lall, 1992; Lee and Lim, 2001). The design of their innovation system fosters accelerated learning to quickly move from imitation and adaptation to innovation (Kim, 1997). In addition to a mere replication of existing growth paths, for example by replicating the product life cycle of developed countries, late-comers might also skip some stages or even create their own growth path and thus possess the potential to leapfrog incumbent firms or regions (Lee and Lim, 2001). In a similar vein, Wong (1999) distinguishes generic strategies and growth paths followed by firms and industries in emerging regions. A first set of strategies tries to reverse either the product life cycle by an increased speed of structural change or the value chain position by moving quickly from original equipment manufacturing towards own design and branding within an industry. A second set of strategies instead comprises efforts to become a specialist or even pioneer in terms of innovative products, processes or applications. Based on the conceptual considerations, the following firm and industrylevel determinants are used to characterize the development model of the three coastal regions in China: ●

Sectorial structural change. Based on the literature cited above, the accelerated structural reorganization of the economy is a general feature of the adaptive cycle and a specific feature of economic catch-up.

ERIKSSON 9780857930088 PRINT.indd 88

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth ●



89

Technological capability building. Within each industrial sector, also in traditional ones, there is considerable leeway for upgrading, for example from low-tech parts and assembly activities to high-tech components and design activities. Development of market orientation. At first, firms from emerging nations are often positioned at the low value-added end of global value chains. As soon as buying power on the national market develops, they become able to launch own solutions domestically.

These three dimensions will structure the following sections of this chapter, in that they are considered the baseline of transformations caused by entrepreneurially induced reactions to the external shock of the crisis. With a view on our core hypothesis laid out above, however, the public contribution through stimulus packages and procurement has to be included in our explanatory framework: ●

The role of the state. The Chinese business system cannot be fully understood without taking into account the role of the state in designing the institutional environment and the influence of stateowned and state-controlled business.

The following sections will briefly analyse the influence of each determinant on pre-crisis patterns and the post-crisis outlook of the growth paths in Chinese regions. Pre-crisis Patterns and Post-crisis Outlook for Selected Determinants of Regional Growth Paths in China Sectorial structural change The redirection of production factors from sectors with low productivity to those with higher productivity was an important source of growth in China and its importance has increased in particular during the last few years. While capital accumulation is still responsible for about 60 per cent of total growth, sectorial shifts already explained another quarter of the total growth performance from 2003 to 2008 (OECD, 2010). The general pre-crisis trend was to shift production factors in the coastal provinces from labour- and resource-intensive sectors towards capital- and human capital-intensive sectors. This move was underpinned by the definition of pillar industries and the targeting of so-called high-tech industries at the national level and a specific focus on the development of the service sector in the 11th Five-Year Plan. At the provincial level, the promotion of sectorial change was implemented in a similar way. In Guangdong

ERIKSSON 9780857930088 PRINT.indd 89

29/01/2013 16:59

90

Clusters and economic growth in Asia

Province, for example, a policy to ‘empty the cage to let the new bird move in’ has been implemented with the aim to relocate so-called ‘three high and one low’ industries, that is, high pollution, high use of energy, high use of resources and low value-added (HKCPU, 2010). Some of the restrictive policies for low-tech sectors have been eased during the recent economic crisis to prevent a loss of labour in traditional sectors at a time when modern sectors have not been able to procure sufficient additional employment. However, for the post-crisis period it is clearly expected that the direction of sectorial change will be unchanged and that the politically favoured high-tech industries will be in an advantageous position. Technological capability building The process of technological capability building at the firm or industry level differs significantly from sectorial change. A high level of technological capabilities can be reached by so-called low-tech industries like textiles and garments as well as by firms in a high-tech sector like telecommunications. In the 1980s and 1990s, the globalization of manufacturing activities led to separation of production and innovation capabilities (Bell and Pavitt, 1995), with China’s coastal regions entering globalization with a focus on production. Nonetheless, a foundation for the upgrading of technological capabilities has gradually emerged (Altenburg et al., 2008) and, more recently, many Chinese companies have at least developed some ambitions in the field of medium-tech innovation (Kroll and Schiller, 2010). For the post-crisis period it is expected that even more political priority will be given to innovation activities and upgrading within the value chain by providing incentives for investments in R&D and by further restricting processing trade activities. The recent stimulus package has already earmarked 10 per cent for investments in innovation (OECD, 2010). However, a trade-off might exist between the promotion of hightech sectors and technological upgrading. In the electronics industry, for example, it might be easier for domestic firms to develop technological capabilities in more traditional branches like household appliances than in leading-edge technological fields such as the latest IC design capabilities. While for a firm embarking on an innovation strategy for the first time the requirements of the global markets for cutting-edge technologies will prove difficult to meet, it may be well equipped to provide medium-tech solutions in the still price-sensitive, but increasingly quality-oriented domestic arena. The post-crisis outlook is expected to be characterized by an ongoing emphasis on the development of a knowledge-based economy (Kroll and Schiller, 2010).

ERIKSSON 9780857930088 PRINT.indd 90

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

91

Market orientation Stiglitz (2008) points out that to move away from an export-led growth model towards one that balances exports and domestic consumption is the most critical feature of China’s new model of development. He underlines that while the country already possesses competitive domestic firms, the development of an independent innovation system can only be pursued successfully when firms are able to upgrade and to innovate with the support of external actors – a position that the Chinese government in the meantime has haltingly acknowledged. On the other hand, recent studies have shown that even the south of China is less export-dependent than often suspected (OECD, 2010). In many sectors, domestic firms are well prepared to meet the demands of domestic customers who are more interested in reliable products at competitive prices than in latest technologies and highest quality (Zhou, 2008). As one result, even five years ago, Chinese brands were supplying two-thirds of all personal computers in the domestic market (Zhou, 2005). Arguably, domestic actors might best exploit their potential if they combine export orientation with a strong presence in the domestic market, which is less subject to global business cycles. Additionally the position of domestically oriented firms is improved by the fact that the stimulus package explicitly tries to strengthen domestic consumption in peripheral areas (OECD, 2010), which do not possess the purchasing power to buy foreign products. It is therefore expected that firms with a strong position in the domestic market will recover more quickly from the crisis than firms that predominantly export their outputs to developed markets with a less convincing growth outlook – a situation that in both Europe and the United States has worsened rather than improved in the wake of the crisis. The basis for the argument put forward by Chen and De Lombaerde (2010) – that Chinese firms should increase rather than decrease their presence in global value chains – has with all likelihood changed. While Chen and De Lombaerde are right in stating that the focus of any increase in presence should include other emerging markets, their 2009/10 optimism about the growth potential of traditional value chains – those based on European and US markets – must now be considered much more limited. While it is certainly true that the export orientation of many Chinese firms cannot and should not be given up entirely, a new balance between the ailing economies of the West and the so far still dynamic growth of the Chinese middle class will have to be found. Institutional evolution and the role of the state With regard to the role of the state Overholt (2010) points out the need for a new economic strategy, refocusing from the current combination of

ERIKSSON 9780857930088 PRINT.indd 91

29/01/2013 16:59

92

Clusters and economic growth in Asia

state-led infrastructure development and private export-based exploitation of cheap labour towards a model mainly resting on upgraded low-end manufacturing in the private sector. Likewise, the latest OECD report (OECD, 2010) calls for a further liberalization of the Chinese market to sustain the growth after the immediate recovery that is still fuelled by huge public investments under the stimulus package. While it is acknowledged in the OECD report that the performance of state-owned and statecontrolled firms has improved over time, they are still well behind the private sector in terms of productivity and growth. While the immediate effects of the stimulus package are expected to favour the state-owned sector due to public procurement in the construction and transport sectors and increased credits made available by state-owned banks, the long-term post-crisis outlook favours the private sector. The report suggests that political levers of control on the technologyrelated sectors should be loosened. The three coastal regions selected for the empirical analysis are the best performers in terms of R&D and commercialization efficiency in China (Guan and Chen, 2010), but they differ in terms of the structure of their industrial sectors. A further point worth noting is that just before the crisis, in early 2008, the Chinese government took several measures to slow down the overheating economy due to raising fears of inflationary tendencies (Schüller and Schüler-Zhou, 2009). The effect of these measures is expected to be stronger in those regions whose economy is controlled by the state to a larger degree like Beijing and may have aggravated the slowdown that resulted from the global crisis in late 2008 and early 2009. In a similar vein, such measures were again taken in some regions of the Yangtze River Delta in 2011. Hypotheses about the Impact of the Determinants on the Post-crisis Growth Paths of Regions in China Based on the conceptual considerations outlined above, it is expected that regions that depend on global markets, are specialized in traditional sectors, and possess limited technological capabilities are most vulnerable to the effects of the crisis and will more likely be less successful in recovering from its impact during a continued slowdown of the global economy. On the contrary, firms that are oriented towards the domestic market might yet survive in traditional sectors as some of their products are still in demand in rural China and their price competitiveness might still match the high price elasticity on the domestic markets. On the other hand, the rise of an increasingly discerning middle class with significant buying power suggests that other more ambitious strategies may well be

ERIKSSON 9780857930088 PRINT.indd 92

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

93

Market orientation

Mid-tohigh

Global

++

+

+

0

Low-tobasic

Technological level

Domestic

+

0

0



Modern

Traditional

Modern

Traditional

Industrial sector

Figure 5.1

Industrial sector

Expected impact of sector, technological level and market orientation on post-crisis recovery of firms in China

more successful and more sustainable, even on the domestic market. In the global market, Chinese firms will arguably have little other option to sustain their competitiveness than by focusing on lines of business in ‘modern sectors’ or by upgrading their technological capabilities in others. In conclusion, the determinants of China’s post-crisis growth trajectory identified in our reflections above can be summarized in a three-dimensional conceptual framework (Figure 5.1). In more detail, this framework implies the following guiding assumptions: Assumption 1a: Firms in modern sectors will, ceteris paribus, more successfully recover: while firms in traditional basic needs-oriented sectors may be affected less by the crisis they will also profit from the ensuing recovery to a lesser degree. Firms in modern sectors, on the contrary, are expected to benefit more during the recovery than those in traditional sectors because structural change is accelerated by the crisis. Assumption 1b: Upgrading firms or firms already active in the mediumtechnology field will, ceteris paribus, more successfully recover: the exception could be domestic firms that upgrade their low-tech assembly activities in traditional sectors towards higher-value-added activities and will to a higher degree be able to benefit from the recovery better than those that do not change their business model. In general, this holds true for both the low-tech and the high-tech sector. While the firms may not be able to provide cuttingedge products for the global market, they have ample opportunities to move from a mere assembly orientation towards design.

ERIKSSON 9780857930088 PRINT.indd 93

29/01/2013 16:59

94

Clusters and economic growth in Asia

Assumption 1c: Firms targeting the growing domestic market (with nonlow-tech strategies) will, ceteris paribus, more successfully recover: firms with a domestic market-oriented sales strategy will recover more quickly in the short run as the dynamic Chinese domestic market resting on a fast-growing middle class can be considered more resilient than the global markets, which appear set to remain in turmoil for years to come. Consequently, both technologically upgraded activities in traditional sectors and low-tech activities in high-tech sectors will pay off better when at least in part targeting the domestic market than when being wholly dependent on global customers. However, the importance of the fiscal stimulus package for bringing the Chinese economy back to the forefront of the recovery must not be underestimated. As stated above, the core hypotheses of this chapter are only partially conflicting. Even if some structural change has been driven by entrepreneurial decisions, it may in many contexts have been superimposed by the effects of fiscal stimulus packages and macro management of the economy: Assumption 2a: Effects of the state-induced rise in public demand superimpose structural changes based on decisions and strategies of individual business people affected by the crisis: the fiscal stimulus package has provided a substantial impetus, helping the Chinese economy to return to the pre-crisis growth path. However, these short-term measures have concealed the actual structural impact of the crisis on the different sectors, which only becomes visible over time. Assumption 2b: The aforementioned effects of the fiscal stimulus package have been very different from sector to sector and from region to region: The effects of the stimulus package have differed substantially between sectors and regions depending on the nature of the sector and the structure of each region’s industries. With regard to this, ownership structure will play an important role since foreign firms are to a far lesser extent able to benefit from public procurement. As the above section has illustrated, it is obvious that a mere macro perspective is indeed insufficient to even describe the shifts that have occurred in the past two or three years. Consequently, they will be analysed in the following with reference to the concept developed in Figure 5.1. Guided by its three key dimensions, the empirical section of this chapter will explore whether the main hypotheses of structural change based on entrepreneurial decision can be upheld.

ERIKSSON 9780857930088 PRINT.indd 94

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

95

REGIONAL ANALYSIS OF THREE COASTAL REGIONS IN CHINA General Findings A first general impression regarding the Chinese reaction to the global economic crisis can be gained from the quarterly business and entrepreneurial climate surveys conducted in China. Figure 5.2 illustrates that from mid-2008 onwards, Chinese business people were acutely aware of the upcoming difficulties. Unsurprisingly, therefore, the sharpest decrease of business confidence occurred between the third and fourth quarters of 2008. Interestingly, in the first quarter of 2009, when business pessimism culminated, entrepreneurial hopes had already significantly picked up, thus indicating that the individual business actors in the Chinese industrial sector had made new plans for the road to recovery before the recovery itself began to gain momentum. With reference to our hypotheses, this suggests that the behaviour of business actors was not merely reactive, but that opportunity-based reactions may have played a part. On the other hand, entrepreneurial confidence in the construction sector was continuously 150

140

Index

130

120

110

100

Business climate index ECI industry

2011-03

2011-02

2011-01

2010-04

2010-03

2010-02

2010-01

2009-04

2009-03

2009-02

2009-01

2008-04

2008-03

2008-02

2008-01

2007-04

2007-03

2007-02

2007-01

90

Entrepreneurial confidence index (ECI) ECI construction

Source: Own graph based on data from the National Bureau of Statistics of China.

Figure 5.2

Business climate and entrepreneurial confidence in China during and after the crisis

ERIKSSON 9780857930088 PRINT.indd 95

29/01/2013 16:59

96

Clusters and economic growth in Asia

higher than in the industrial sector. This suggests that infrastructure construction has indeed played a role in the process of recovery and that some opportunities may result from public spending. The observed pattern would thus be consistent with a situation in which different processes are at work. Additionally, Figure 5.2 illustrates that pre-crisis levels of business climate and entrepreneurial confidence have not yet been reached again. On the contrary, a renewed downturn seems to be underway since early 2011. In line with the above findings, the effects of the crisis started to exert influence on the growth rate of overall industrial output value generated in the Chinese economy as early as mid-2008. Following the worldwide financial turmoil in September 2008 the decrease in growth rates sharpened, prompting a first low in overall growth rates in November and December 2008. After a brief recovery in January and February, atypical during spring festival times, growth dropped again in March and April 2009 until a continuous process of recovery began in May/June 2009, bringing the economy back to its pre-crisis growth level by November 2009. In early 2010, however, a renewed disruption occurred. The impact of that year’s spring festival on industrial growth was more severe than usual, until growth picked up again in March and April. Following that, growth gradually declined to slightly below pre-crisis levels (around 12 per cent). Despite the sluggish world economy, growth of industrial output value reached a new high in 2011. The ensuing smooth decrease in the course of the year, furthermore, appears like evidence of dampening macro management. Undoubtedly, the crisis as such is over. Impacts and Recovery by Industrial Sector Table 5.2 and Figure 5.3 illustrate that the double minimum in the development of value-added growth can be explained through the distinct development paths of different sectors. In the following, this will be illustrated based on data for selected industries that can be distinguished with sufficient clarity according to some of the main dimensions of the conceptual model detailed in Figure 5.1. Sectors that mostly grow according to different logics and public regulation, in contrast, have been excluded from our analysis (for instance state-controlled sectors like mining and the public supply of utilities). In total, the selected sectors listed and described in Table 5.1 produce half of China’s gross industrial output value. In the low-tech field, relatively few sectors, like the locally oriented food sector, were only insignificantly affected during the crisis and continued to grow at a more or less stable rate. In general, on the contrary, low-tech, export-oriented sectors like textile production appear structurally affected

ERIKSSON 9780857930088 PRINT.indd 96

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

Table 5.1

97

Basic characteristics of main economic sectors in China Share of Market Nat. Orientation Total (%)

Total M of foods M of textiles M of ferrous metals M of chemical prod. M of transport equipment M of medicines M of general purpose machinery M of electrical machinery M of ICT equipment M of measurement and office machinery

Share of Firms in Gross Output Value (%) State- Private Foreign owned

1.5 4.0 7.1 6.2 9.9 1.3 4.9

– Domestic Export Domestic Domestic Domestic Domestic Mixed

32 10 3 50 26 43 20 19

35 44 70 32 42 17 37 54

33 46 27 18 33 41 43 28

5.2 9.0 0.9

Mixed Export Export

12 9 12

43 07 31

45 84 57

Note: As exports are reported by good rather than by sector, only approximate statements can be made. Source: Own compilation and calculation based on data from the National Bureau of Statistics of China.

and unable to return to their pre-crisis growth path – even though in the course of the downturn they suffered less than others. Moreover, as export-oriented low-tech sectors they did not participate strongly in the boosts of recovery during early and late 2009. Most heavy industries such as the ferrous metals and the chemical products sector, in contrast, were hit early and severely in late 2008, but became main growth agents in 2009. Afterwards, however, they dropped back to average growth paths. Those with a larger private element like the chemical sector grew from a higher basis and maintained a higher growth path. Likewise, the more technology-oriented transport equipment sector became the strongest agent of growth in the course of late 2009 and was able to maintain a higher growth longer than others before its growth dropped back below their pre-crisis levels in the course of 2011. Several mid-tech sectors (such as pharma, electrical engineering and machine building) display a substantial recovery earlier than the heavy industry in early 2009, before their growth rates drop back to late 2008 levels or even below, until mid-2009 (partially concealed in Table 5.2 but

ERIKSSON 9780857930088 PRINT.indd 97

29/01/2013 16:59

98

Table 5.2

Clusters and economic growth in Asia

Growth rates in percentage of industrial value-added in different sectors Average Average Average Average Average 02/08– 07/08– 07/09– 01/10– 01/11– 06/08 06/09 12/09 12/10 09/11

Total 9.40 Low-tech domestic (private & foreign) M of foods 18.34 13.88 Low-tech export (private) M of textiles 12.68 9.15 Heavy industry domestic (strong state influence) M of ferrous metals 14.50 2.24 M of chemical prod 14.10 7.42 Mid-tech domestic (state & foreign) M of transport equip 21.86 10.40 Mid-tech domestic (private & foreign) M of medicines 18.56 16.23 M of gen. purp. mach. 21.68 11.15 M of electrical mach. 21.06 14.65 Mid-high-tech mixed (foreign) M of ICT equipment 16.72 4.85 M of measuring & off. mach. 17.28 6.13

15.13

14.59

14.11

15.73

14.40

16.92

10.00

9.72

7.65

18.93 21.87

10.86 14.76

10.50 14.54

28.17

21.09

11.69

16.32 14.70 13.48

13.71 19.84 16.56

17.39 18.87 14.97

9.63 4.68

15.43 16.65

15.02 17.32

Source: Own compilation and calculation based on data from the National Bureau of Statistics of China.

obvious in monthly data). All three, however, maintain a higher growth path in the course of both 2010 and 2011, thus standing out against the transport sector. In late 2011 they are among the lead sectors of the economy, but slightly below their pre-crisis level. High-tech sectors with a strong foreign ownership and export orientation  (such as instruments, office machinery, electronics and communication) display another, distinct path of development. Their growth rates in the course of 2009 remain clearly below those of all other main sectors until October, before they pick up significantly. Apparently, only parts of the instruments sector share the early 2009 boost. In the course of 2010 and 2011, however, they were able to realize sustainable growth rates and are back on their before-crisis growth path. Our data thus suggest that the 2009 boost in growth is indeed connected with the Chinese fiscal recovery package as it becomes particularly visible in those sectors receptive to increased public domestic demand. This interpretation appears particularly obvious since the strongest increase is to be

ERIKSSON 9780857930088 PRINT.indd 98

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

99

35% 30% 25% 20% 15% 10% 5%

–5%

2008-02 2008-03 2008-04 2008-05 2008-06 2008-07 2008-08 2008-09 2008-10 2008-11 2008-12 2009-01 2009-02 2009-03 2009-04 2009-05 2009-06 2009-07 2009-08 2009-09 2009-10 2009-11 2009-12 2010-01 2010-02 2010-03 2010-04 2010-05 2010-06 2010-07 2010-08 2010-09 2010-10 2010-11 2010-12 2011-01 2011-02 2011-03 2011-04 2011-05 2011-06 2011-07 2011-08 2011-09

0%

Total Manufacture of textile Manufacture of general purpose machinery Manufacture of communication equipment, computer and other electronic equipment Manufacture of transport equipment Manufacture of foods Manufacture and processing of ferrous metals

Source: Own compilation and calculation based on data from the National Bureau of Statistics of China.

Figure 5.3

Growth rates in percentage of industrial value-added in different sectors

witnessed in the transport sector, a field particularly likely to profit from public investment in traffic infrastructure. Moreover, it gains in plausibility as the growth levels in many sectors dropped back significantly with the beginning of 2010 and as sectors with a comparatively high level of foreign ownership did not display any high-momentum recovery in the first place (for example, information and communication equipment). In short, the momentum of economic recovery in the course of 2009 can be attributed to four interdependent but distinct forces of recovery – suggesting that more than one differentiating factor is at play. First, a brief, non-lasting increase of growth in the mid-tech sectors in the first half of 2009. Second, a more substantial boost in the heavy industry and the transportation equipment sector in late 2009 stretching well into 2011. Third, a lasting recovery of the mid-tech sectors in the course of late 2009 and early 2010 to slightly below pre-crisis growth paths. Finally, a recovery of the foreign-oriented mid- to high-tech fields in the instrument,

ERIKSSON 9780857930088 PRINT.indd 99

29/01/2013 16:59

100

Clusters and economic growth in Asia

BOX 5.1 INDUSTRY GROUPINGS Globalized industries Manufacture of Communication Equipment, Computer and Other Electronic Equipment, Manufacture of Measuring Instruments and Machinery for Cultural Activity and Office Work, Manufacture of Textiles, Manufacture of Textile Wearing Apparel, Footwear, and Caps, Manufacture of Leather, Fur, Feather and its Products Resource-based, domestic market-oriented industries Processing of Petroleum, Coking, Processing of Nuclear Fuel, Manufacture of Non-metallic Mineral Products, Manufacture and Processing of Ferrous Metals, Manufacture and Processing of Non-ferrous Metals Protected and favourite industries Manufacture of Foods, Manufacture of Beverages, Manufacture of Tobacco, Manufacture of Medicines Least protected Manufacture of Chemical Fibres, Manufacture of Rubber, Manufacture of Plastics, Manufacture of Electrical Machinery and Equipment Source:

Own summary, based on He (2009, pp. 265, 267, 269, 270).

office machinery, electronics and communication sectors to their pre-crisis growth paths in the course of late 2009 and early 2010. With regard to the latter, it is important to remember that while some of these sectors are labelled high-tech, the intrinsic technological ambition of many (assembly) firms within them is not necessarily very high. In conclusion, additional interesting insights can be gained when the developments are analysed based on a grouping of industries used in a recent World Bank report (2009), as it is thus possible to broaden the basis of the analysis beyond the sectors selected above. The report distinguishes between ‘globalized industries’, ‘resource-based and domestic market-oriented’, ‘protected and favourite industries’ and ‘least protected industries’ as follows in Box 5.1. In line with the above findings, the resource-based and domestic marketoriented industries peak in early 2010 to drop below their usual growth

ERIKSSON 9780857930088 PRINT.indd 100

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

101

30%

25%

20%

15%

10%

5%

–5%

2008-02 2008-03 2008-04 2008-05 2008-06 2008-07 2008-08 2008-09 2008-10 2008-11 2008-12 2009-01 2009-02 2009-03 2009-04 2009-05 2009-06 2009-07 2009-08 2009-09 2009-10 2009-11 2009-12 2010-01 2010-02 2010-03 2010-04 2010-05 2010-06 2010-07 2010-08 2010-09 2010-10 2010-11 2010-12 2011-01 2011-02 2011-03 2011-04 2011-05 2011-06 2011-07 2011-08 2011-09

0%

Globalized industries

Resource‐based and domestic market‐oriented

Protected and favourite industries

Least protected industries

Total

Note: Classification according to the World Bank (2009); unweighted averages of development in respective technological fields; own calculation.

Figure 5.4

Growth rates of industrial value-added by sectorial groups according to the World Bank

path in late 2010 and only recover in 2011 – much like the heavy industries selected above. Moreover, the finding that some domestic low-tech industries like foods have weathered the storm is confirmed by the aggregate development of those industries designated protected and favourite. Finally, there are two more distinct paths of recovery: of least protected industries and globalized industries with the latter recovering later, but no less sustainable, despite their quite different nature. The core of the aforementioned conclusions is thus confirmed on a broader basis in Figure 5.4. Impacts and Recovery by Province and Region In the following, the impacts of the crisis will be studied broken down by selected regions. The three main economic areas covered by this analysis constitute the main motors of the country’s development and generate more than 40 per cent of China’s GDP.

ERIKSSON 9780857930088 PRINT.indd 101

29/01/2013 16:59

102

Table 5.3

Clusters and economic growth in Asia

Basic economic characteristics of China’s three main economic regions

2009

Share of National GDP

Export Quota

3.3 2.1 4.7 10.1

27.2 27.1 6.2

50.8 41.4 34.8 40.8

6.3 15.0 45.2 26.6

42.9 43.6 20.0 32.7

4.1 9.4 6.3 19.8

64.3 39.4 39.5

34.3 12.7 16.0 18.2

10.9 39.8 53.1 37.3

54.7 47.5 30.9 44.5

10.8

62.0

18.0

20.8

61.2

Beijing Tianjin Hebei Bohai Bay (Jing-Jin-Ji) Shanghai Jiangsu Zhejiang Yantgze River Delta Guangdong

Gross Output Value State-owned Private ForeignChinese firms Chinese firms invested firms

Source: Own calculations, based on China Statistical Yearbook, 2010 and oanda.com.

Studying regions is of relevance to this chapter as the growth engines of the Chinese economy are not only located in certain sectors, but also in certain regional economic systems with distinct market orientations and ownership structures. Greater Beijing, Tianjin and Shanghai are to a greater extent focused on foreign-invested firms, state-owned companies and joint ventures, while the industrial sector of Jiangsu and particularly Zhejiang is characterized by private domestic firms and a certain amount of foreign investment. Hebei in turn is characterized by state-owned and private firms, whereas foreign investment is quite low. In the PRD, finally, most industrial output is generated by export-oriented foreign-invested firms (Zhou, 2005) while private domestic firms are less prevalent (cf. Table 5.3). Correspondingly, the export quotas of Shanghai and Guangdong are highest, above 60 per cent, followed by Zhejiang and Jiangsu with around 40 per cent each. While Beijing and Tianjin do generate substantial amounts of exports and foreign investment in the regions is high, their quotas of just below 30 per cent are only mid-range in comparison with other Chinese provinces. Hebei, finally, remains an almost entirely domestic-oriented economy (cf. Table 5.3). As expected, these regions display differing trajectories of development during the crisis. Figure 5.5 and Table 5.3 illustrate that the Bohai Bay area was the first region to experience a substantial downturn of economic

ERIKSSON 9780857930088 PRINT.indd 102

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

103

activities in mid-2008. Partially, this may be attributed to the completion of the substantial construction efforts related to the 2008 Olympic Games as well as the reduction of industrial production in preparation for the Olympics. Part of it may also have been due to the government’s efforts to avoid an overheating of the economy in early 2008. Nonetheless, both effects were certainly reinforced beyond the envisaged extent by the dawning economic crisis (cf. Figure 5.5). Remarkably, however, the Bohai Bay area’s growth rates remained high throughout the height of the crisis in the last quarter of 2008, when the decline in many other regions had only just begun. The increased momentum lasted until April 2009, when the short-term early 2009 boost of growth in many state-dominated sectors came to an end. At that time, the Bohai Bay area saw a brief renewed decrease in growth rates. Of the Bohai Bay area’s provinces, Beijing is the one that most clearly displays the substantial increases in value-added following May 2009 that we suspect to be based on the public economic recovery packages. This level of growth, however, was only maintained until mid-2010. Later, Beijing returned to a below pre-crisis growth path. The effect of the crisis on the Bohai Bay area as a whole was only mitigated by the continuous expansion of industrial activities in Tianjin generating stable growth of above 20 per cent, as well as domestically oriented Hebei, which has returned almost to its pre-crisis growth levels (cf. Table 5.4). In the Yangtze River Delta, in contrast, growth rates continuously decreased since the first clear signs of impending problems in mid-2008. At the beginning of the crisis, the Yangtze River Delta was hit hardest of all major Chinese regions, in particular due to the situation in Shanghai. In December 2008 the growth rates dropped below zero for the overall region and down to about 10 per cent for municipal Shanghai. In contrast to Shanghai, Jiangsu Province was able to maintain above 9 per cent growth rates throughout the crisis. Zhejiang was affected, although nowhere as gravely (cf. Table 5.4). All provinces of the Delta, however, very clearly participated in the substantial short-term boost of growth rates following March/April 2009. Given that the three provinces are to very different degrees dominated by state-owned firms and oriented towards the domestic market, it appears noteworthy that all of them display a fairly similar path of recovery, particularly as each of them had been affected by the crisis to a different extent. While the Yangtze River region seemed to have recovered to an average growth path in the course of 2010, it returned to a below pre-crisis growth path and showed no sign of improvement as of late 2011 (cf. Figure 5.5). In 2010, it seemed to be Shanghai that recovered better than the other two provinces in the region, a situation that, however, only lasted until early 2011, when the three provinces equalled out and it became Zhejiang that fell behind the two others. Part of this,

ERIKSSON 9780857930088 PRINT.indd 103

29/01/2013 16:59

104

Clusters and economic growth in Asia

25%

20%

15%

10%

5%

2008-02 2008-03 2008-04 2008-05 2008-06 2008-07 2008-08 2008-09 2008-10 2008-11 2008-12 2009-01 2009-02 2009-03 2009-04 2009-05 2009-06 2009-07 2009-08 2009-09 2009-10 2009-11 2009-12 2010-01 2010-02 2010-03 2010-04 2010-05 2010-06 2010-07 2010-08 2010-09 2010-10 2010-11 2010-12 2011-01 2011-02 2011-03 2011-04 2011-05 2011-06 2011-07 2011-08 2011-09

0%

National Total Yangtze River Delta

Bohai 3 (Jing-Jin-Ji) Guangdong

Note: Weighted averages based on 2008 GDP figures. Source: Own graph based on data from the National Bureau of Statistics of China.

Figure 5.5

Growth rates of industrial value-added by major economic area in the course and wake of the crisis

however, may be due to official action announced and probably in the meantime taken to keep Zhejiang’s economy from overheating.1 The province of Guangdong, finally, displays a path of development distinct from almost any other province in China. In the Pearl River Delta, the effects of the crisis were felt both later (around September 2008) and far less intensely than in other provinces. In Guangdong, the rate of valueadded growth never dropped below its 3.5 per cent low in April 2009, the only month when it went below 5 per cent at all (cf. Figure 5.5). In line with the development of the locally important export-oriented ICT sectors, the low point with regard to growth was reached four months later in the Yangtze River Delta. Guangdong thus reached its absolute low latest of all provinces, pointing to an above-average resilience of the local economy but also to a below-average participation in the mid- to late 2009 nationallevel boost in growth rates. In the course of 2010, growth again increased to above 15 per cent, thus more than exceeding its pre-crisis average of around 13 per cent. Even in late 2011 it continued to grow above 13 per

ERIKSSON 9780857930088 PRINT.indd 104

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

Table 5.4

105

Growth rates in percentage of industrial value-added in different provinces Average Average Average Average Average 02/08–06/08 07/08–06/09 07/09–12/09 01/10–12/10 01/11–09/11

National total Beijing Tianjin Hebei Shanghai Jiangsu Zhejiang Guangdong

16.18 8.86 21.18 17.36 11.18 16.22 11.58 13.00

9.40 –2.49 21.04 9.82 –0.48 13.82 4.00 9.03

15.13 20.25 23.48 19.95 10.62 16.92 11.78 12.57

14.69 13.42 24.28 15.70 18.60 14.49 14.89 16.08

14.11 6.53 20.87 15.92 7.93 13.91 11.53 13.13

Source: Own compilation and calculation based on data from the National Bureau of Statistics of China.

cent, close to at par with Jiangsu, despite the continued downturn in the global economy (cf. Table 5.4). In summary, the analysis by provinces underlines the above finding that there are different, yet equally successful trajectories of recovery, in particular with a view to market orientation. In view of the global downturn, it appears remarkable that the most export-oriented province has at first proven most resilient and later very sustainably recovered. Nonetheless, recovery is no less substantial in Jiangsu despite the more domestic ownership structure of its industry and even in Hebei, which is very weakly connected with the global economy in any respect. Apparently, none of the country’s major growth models have been entirely shattered. Impacts on Technological Upgrading The impact of the crisis on technological upgrading is a bit more difficult to assess in detail, as in China with its extensive assembly and re-export activities, many industrial activities in mid- to high-tech sectors are in fact not very technologically ambitious. Consequently, our overall approach will be twofold. First, we will examine the development of patent applications. Second, we will draw on first-hand evidence data obtained from a survey of electronics firms in Guangdong Province in late 2009. Unfortunately, both approaches by and large force us to gauge reactions during the crisis in 2009 rather than in its wake in 2010 and 2011. On an aggregate level, however, some recent data

ERIKSSON 9780857930088 PRINT.indd 105

29/01/2013 16:59

106

Clusters and economic growth in Asia

60 000 Total 50 000

Total 5 month average

Domestic

Domestic 5 month average

Foreign

Foreign 5 month average

40 000 30 000 20 000 10 000

2006-01 2006-03 2006-05 2006-07 2006-09 2006-11 2007-01 2007-03 2007-05 2007-07 2007-09 2007-11 2008-01 2008-03 2008-05 2008-07 2008-09 2008-11 2009-01 2009-03 2009-05 2009-07 2009-09 2009-11 2010-01 2010-03 2010-05 2010-07 2010-09 2010-11 2011-01 2011-03 2011-05 2011-07 2011-09

0

Source: Own graph and calculations based on data from the State Intellectual Property Office.

Figure 5.6

Development of patent applications at the State Intellectual Property Office (SIPO) in the course and wake of the crisis

are now available, suggesting that 2009 data may in fact be most relevant to understand the transformatory momentum of the crisis. Patent applications, main findings With regard to the innovative performance of the Chinese economy, Figure 5.6 shows that the crisis has not generally led to a slump in patent applications by Chinese nationals in 2009, although it did cause a temporary decrease in their growth rate. Foreign patent applications in China, however, decreased by about 10 per cent compared to their 2008 level but picked up again in 2010. While that seems to suggest that the technological engagement of foreign firms diminished during the crisis itself, applications have now reached a pre-crisis level. Apparently, intellectual property rights are seen as important assets on a Chinese domestic market – one that even for them has developed beyond price-based competition. Moreover, monthly figures illustrate how patent activities of, in particular, domestic actors have continued to surge in the course of 2010 and 2011 – far more rapidly than foreign applications. Given the still relatively limited exposure of Chinese applicants to the global technology market, this seems to suggest a general increase in the domestic importance of intellectual property rights, fragile as many of those may still be. Apparently,

ERIKSSON 9780857930088 PRINT.indd 106

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

107

80 000 70 000

Bohai 3 (Jing-Jin-Ji)

Yangtze River Delta

Guangdong

Shanghai

60 000 50 000 40 000 30 000 20 000 10 000 0 2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Note: By priority year. Due to the usual time lag of publication, figures for 2009 remain estimates. Source: Own calculations based on the European Patent Office (EPO) Worldwide Patent Statistical Database and SIPO raw data.

Figure 5.7

Development of Chinese patent applications at the SIPO in recent years, by region

the Chinese growth model continues to focus on technological upgrading, even during the crisis. Patent applications, specific findings Interestingly, a parallel dampening effect of the crisis is felt on Chinese patent applications in provinces with pronounced foreign orientation such as Guangdong but also Beijing and Shanghai (cf. Figure 5.7). While growth in no case turns negative, the growth rate decreases. In other provinces like Zhejiang and Jiangsu, in contrast, the number of applications grew unabatedly, indicating a broad-based process of upgrading. For the Bohai Bay area and Guangdong, in contrast, this appears less clear in the course of 2009. Possibly an additional perspective can help to better understand the situation at hand and to further differentiate the seemingly similar impact of the crisis that was felt in the structurally quite different regions of the Bohai Bay area and the Pearl River Delta. With a view to our guiding assumptions, it appears noteworthy that the strongest relative decreases in applications occurred in the leading-edge technology fields, while in

ERIKSSON 9780857930088 PRINT.indd 107

29/01/2013 16:59

108

Clusters and economic growth in Asia

140 000 Mid‐high‐tech applications Leading‐edge‐tech applications

120 000

Low‐tech applications 100 000 80 000 60 000 40 000 20 000 0 2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Note: By priority year. Fraunhofer ISI High-tech classification. Due to the usual time lag of publication, figures for 2009 remain estimates. Source: Own calculations based on the EPO Worldwide Patent Statistical Database and SIPO raw data.

Figure 5.8

Development of Chinese patent applications at the SIPO in recent years, by field

mid- to low-tech fields they were more limited (Figure 5.8). Arguably most technological activities that continued during the crisis were focused on R&D areas of close relevance to the domestic market. Long-term investments in leading-edge fields, in contrast, appear to have been somewhat reduced during that period. As illustrated in Figures 5.9–5.11, the Bohai Bay area’s technological capabilities as measurable by the number of domestic patent applications have been quite notably affected by the crisis. Although local patent output grew strongest among the three major economic regions in 2008, the setback in 2009 was similarly pronounced, particularly in the field of leading-edge but also in the fields of mid- to high-tech and lowtech patenting. In the Bohai Bay area, there is thus little evidence that the crisis has prompted a strong shift towards technological upgrading (cf. Figures 5.9–5.11). While affected even more substantially in the field of leading-edge technologies, Guangdong maintained a similar momentum of applications in the high-tech and a distinctly stronger one in the low-tech field. To a degree this can be attributed to the effect that the global crisis has exerted

ERIKSSON 9780857930088 PRINT.indd 108

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

109

12 000 Yangtze River Delta

Bohai 3 (Jing-Jin-Ji)

Guangdong

Shanghai

10 000

8000

6000

4000

2000

0 2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Note: By priority year. Fraunhofer ISI High-tech classification. 2009 data, estimate. Source: Own calculations based on the EPO Worldwide Patent Statistical Database and SIPO raw data.

Figure 5.9

Chinese leading-edge patent applications at the SIPO in recent years, by region

on the region’s leading technology firms. On the other hand, it seems to suggest that some sort of reorientation of technological enterprises may have taken place instead of the general, broad-based setback visible in the Bohai Bay area and, in particular, Beijing. Apparently, many local actors are strengthening their technological upgrading efforts with direct relevance to the domestic market. To explore this phenomenon somewhat closer, the following subsection will further investigate the issue of technological upgrading based on firm-level data from Guangdong. Firm-level survey of the electronics industry in Guangdong The analyses presented in the following are based on a survey of electronics companies that was carried out in the Pearl River Delta (PRD) in Guangdong during late 2009. In total 422 electronics companies were interviewed in the period of September to November 2009. The firms were selected from company directories and from participant lists of leading electronics fairs in the PRD. Questionnaires were distributed to 793 firms (433 via post and telephone follow-ups, 360 via fair visits and follow-ups at the fair). The overall response rate was 53 per cent (38 per cent for postings, 72 per cent for fair visits). Of the firms who answered the questionnaire, 167

ERIKSSON 9780857930088 PRINT.indd 109

29/01/2013 16:59

110

Clusters and economic growth in Asia

30 000

25 000

Bohai 3 (Jing-Jin-Ji)

Yangtze River Delta

Guangdong

Shanghai

20 000

15 000

10 000

5000

0 2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Note: By priority year. Fraunhofer ISI High-tech classification. 2009 data: estimate. Source: Own calculations based on the EPO Worldwide Patent Statistical Database and SIPO raw data.

Figure 5.10

Chinese mid- to high-tech patent applications at the SIPO in recent years, by region

were located in the city of Shenzhen, 177 in Dongguan, 67 in Huizhou and 11 in Heyuan. Shenzhen and Dongguan are the most important locations of the electronics industry in the PRD. The company survey was addressed to the CEOs or senior executives of electronics companies and consisted of five sections: (1) market and strategy, (2) organization and marketing, (3) product and process development, (4) human resources, (5) external contacts. Many questions in the five sections attempted to compare the situation in 2007 with that of 2009 in order to gauge the impact of the crisis on upgrading strategies and business models insofar it could be clearly identified. Table 5.5 illustrates the impact of the crisis that the surveyed firms had experienced by late 2009. The first part of the table compares high-tech or modern sectors, for example integrated circuits or telecommunication equipment, with traditional sectors such as household appliances. Firms in modern sectors have performed systematically better for all indicators, but the differences are only significant for employment growth. Since being in a high-tech sector does not necessarily imply a higher technological level of production, the sample was differentiated by the amount of money spent for new product development. Again, firms that spent more than 30 per cent of sales for new product development

ERIKSSON 9780857930088 PRINT.indd 110

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

111

40 000 35 000

Bohai 3 (Jing-Jin-Ji)

Yangtze River Delta

Guangdong

Shanghai

30 000 25 000 20 000 15 000 10 000 5000 0 2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Note: By priority year. Fraunhofer ISI High-tech classification. 2009 data: estimate. Source: Own calculations based on the EPO Worldwide Patent Statistical Database and SIPO raw data.

Figure 5.11

Chinese low-tech patent applications at the SIPO in recent years, by region

performed significantly better than firms with less innovation expenditures in terms of sales and profits in early 2009 and their growth rates also dropped to a lesser degree between 2007 and early 2009. Finally, the market orientation was analysed by identifying the main market of each firm as well as its main owner. Surprisingly, the differences among firms oriented towards domestic and foreign markets proved insignificant. Instead, firms with mainly Chinese ownership performed systematically better than foreign-owned firms, even though only differences for employment growth were slightly significant. Finally, Table 5.6 organizes the survey data according to the dimensions illustrated in Figure 5.1. First, it underlines that the largest momentum of entrepreneurial action has been documented in the course of the crisis in those firms investing 30 per cent and more of sales volume in new product development. In next to any dimension, those firms have been affected less, grown faster and remained more profitable than their technologically less ambitious counterparts. Second, the table confirms that business dynamics and resilience have been generally higher in technologically more advanced electronics sectors (‘high-tech’) than in more traditional ones (‘other’). In particular, the

ERIKSSON 9780857930088 PRINT.indd 111

29/01/2013 16:59

112

Table 5.5

Clusters and economic growth in Asia

Impact of the crisis on electronics firms in the Pearl River Delta by characteristic Profit Sales EmployGrowth Growth ment Growth (first half (first half (2007 to of 2009) of 2009) first half of 2009)

Technological High-tech level of the Other sector F-test sign.

Change Change in Sales in Profit Growth Growth (2007 to (2007 to first half first half of 2009) of 2009)

53.1% 17.9% 7.414 ***

30.8% 18.0% 2.161 n.s.

17.3% 11.7% 1.088 n.s.

–2.3% –10.0% 0.618 n.s.

–3.1% –7.0% 0.327 n.s.

Expenditure on product development in % of sales

.30% ,530% F-test sign.

28.5% 19.0% 1.076 n.s.

28.6% 14.2% 5.667 **

19.8% 7.8% 11.013 ***

–3.8% –12.3% 1.653 n.s.

–1.4% –9.7% 3.335 *

Main market

Domestic Abroad F-test sign.

24.4% 19.9% 0.248 n.s.

19.8% 19.5% 0.003 n.s.

10.9% 14.8% 1.114 n.s.

–8.6% –9.8% 0.032 n.s.

–7.6% –4.8% 0.394 n.s.

Main owner

Chinese Foreign F-test sign.

30.6% 13.2% 3.826 *

22.4% 16.4% 1.020 n.s.

13.5% 11.0% 0.491 n.s.

–7.5% –11.0% 0.286 n.s.

–6.2% –6.9% 0.030 n.s.

Note: Significance: ***: 1% level, **: 5% level, *: 10% level, n.s.: not significant. Source: Own survey.

analysis underlines that while there still was a resilient and profitable development of non-innovating firms in the high-tech assembly segment (normal double lines), non-innovating firms in more basic electronics sectors proved vulnerable with respect to profits and developed least dynamically (saw-toothed double lines). With respect to market orientation, finally, the findings are somewhat less unambiguous. While the employment dynamics are higher in domestically oriented firms, sales dynamics, profits and resilience are higher for export-oriented firms. Nonetheless, both share one characteristic: innovative firms are more profitable and often grow more dynamically than non-innovating firms. With respect to the sectorial focus, however, the situation is less clear.

ERIKSSON 9780857930088 PRINT.indd 112

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

Table 5.6

113

Development in PRD electronics firms according to conceptual logic

Performance Indicator

Change in staff: 2007 thr. 2009 %

Annual growth rate first half 2009: sales Annual growth rate first half 2009: net profit

Expenditure on Product Development in % of Sales

All Firms

Market Orientation Domestic

High- Other tech

Abroad

High- Other High- Other tech tech

More than 30%

85%

17%

92%

21%

71%

12%

30% and less

24%

18%

33%

16%

3%

22%

More than 30%

41%

27%

34%

25%

52%

29%

30% and less

25%

13%

30%

14%

7%

11%

More than 30%

20%

20%

21%

14%

17%

27%

30% and less

16%

7%

18%

7%

8%

6%

9%

–5%

9% –10%

9%

0%

Change of annual growth rate 07 thr. 09: sales

more than 30% 30% and less

–9%

–13%

–11%

–9%

–3%

–19%

Change of annual growth rate 07 thr. 09: net profit

more than 30%

–4%

–1%

–2%

–5%

–6%

4%

30% and less

–3%

–11%

–4% –10%

2%

–12%

Source:

Own survey.

Apparently, different successful market orientations co-exist: while noninnovative production in less ambitious fields expands strongly when it is focused on the global market, it has proven most vulnerable to the crisis and become the least profitable business model (dotted single lines). The non-innovative production of high-end goods for the domestic market, in

ERIKSSON 9780857930088 PRINT.indd 113

29/01/2013 16:59

114

Clusters and economic growth in Asia

contrast, was more profitable and grew quite dynamically (normal double lines). With respect to innovative firms, the production of less advanced goods appears the most profitable export strategy, while domestically oriented firms profited from a focus on high-end products (normal single lines). A strong growth in sales, however, indicates that high-end goods may become more relevant for export in the future (saw-toothed single lines). In summary, the findings from the empirical study thus unambiguously underline that upgrading is underway and from an entrepreneurial point of view has become more and more crucial to keep different business models resilient and profitable. Moreover, the observed momentum is discernibly stronger in advanced than in technologically less ambitious fields of the electronics sector. In general, technologically ambitious activities in high-tech fields grow more dynamically than expected. Nonetheless, Table 5.6 also illustrates that the ‘old’ export-based growth model tends to prevail to a relevant extent, thus confirming the secondary empirical findings of a specific development trajectory in Guangdong.

DISCUSSION Assumption 1a: successfully.

Firms in modern sectors will ceteris paribus recover more

Official statistical figures show that low-tech and heavy industries have suffered the most during the economic crisis. While industries targeting the national market with medium-tech solutions (e.g., general purpose machinery, electric machinery) have felt a similar impact, they have recovered far more dynamically. In particular, export-oriented low-tech industries like the textile sector have not been able to return to their pre-crisis growth rates, and the heavy industries were unable to maintain not only the publicly generated 2009/10 boost in growth but in some cases also to return to their pre-crisis growth rates. Assumption 1b: Firms upgrading or firms already active in the mediumtechnology field will ceteris paribus recover more successfully. Our analysis shows that technological activities of Chinese applicants in particular continue to surge. Evidently, leading-edge patenting has been affected by the crisis, while more market-oriented medium-tech patenting has not. A substantial technological momentum was maintained, especially in the mid-tech field, and different sectors for which upgrading

ERIKSSON 9780857930088 PRINT.indd 114

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

115

plays a role have recovered sustainably and successfully including the socalled high-tech sectors. By means of an example, the analysis of survey data corroborates that upgrading pays off in the electronics industry in general but is absolutely crucial in traditional subsectors, where the profitability of non-innovative business models seems to fade. Put differently, the example of the sector surveyed confirms that during the crisis a stronger momentum of growth was developed in those firms that decided to upgrade. Assumption 1c: Firms targeting the growing domestic market (with nonlow-tech strategies) will ceteris paribus recover more successfully. The analysis based on official statistical figures suggests that the most dynamic recovery has taken place in at least in part domestically oriented sectors. In traditional low-tech sectors, domestic orientation is the only option to maintain a stable growth path at all. Among the mid-tech sectors, the highest momentum can arguably be identified among those that are at least not dominantly export-oriented. Moreover, strong technological and economic dynamics are documented for Zhejiang and Jiangsu, provinces with a mixed market orientation. Additionally, Hebei provides evidence of the dynamics that can emerge from a domestic orientation. Nonetheless, findings are less clear than in the aforementioned cases as the successful recovery of Guangdong verifies that the export-oriented growth model has remained operational, even after a crisis likely to have noticeably diminished global demand in many fields, a finding confirmed by survey data for Guangdong. Even in this context, however, the data suggest that in some cases domestically oriented strategies have, in the course of the crisis, paid off somewhat better than others, in particular for less innovative firms at the more ambitious end of the electronics field. Assumption 2a: Effects of the state-induced rise in public demand superimpose structural changes based on decisions and strategies of individual business people affected by the crisis. Almost all value-added indicators clearly show unanimous evidence of a first boost in early 2009 and a general boost throughout mid- to late 2009. In early 2010, in contrast, growth levels have dropped back significantly and the impact of the spring festival on the economy seems to have been greater than in the preceding years, including 2009. In the long-term perspective it becomes even more evident that ‘something happened’ in late 2009, in particular in the heavy industries, that did not last beyond mid-2010.

ERIKSSON 9780857930088 PRINT.indd 115

29/01/2013 16:59

116

Clusters and economic growth in Asia

Assumption 2b: The aforementioned effects of the fiscal stimulus package have been very different from sector to sector and from region to region. As our analysis has clearly shown, the mid- to late 2009 recovery effect was particularly strong in the more state-dominated environments of the Bohai Bay area and municipal Shanghai. The more privately oriented provinces of Zhejiang and Jiangsu shared it to a degree, but less substantially. In the export-oriented private economy of Guangdong Province finally the effects remained relatively unremarkable. In the mid-term perspective, however, this development was not maintained and growth rates levelled out in the course of 2010. In contrast to the Bohai Bay area and Guangdong, growth in the Yangtze River Delta seems to have diminished in the course of 2011, even if arguably due to conscious limitations.

CONCLUSION In summary, our study has found evidence for most of the guiding assumptions proposed in the conceptual section. What can be said with certainty is that the Chinese economy, after a substantial setback in late 2008/early 2009, has seen a speedy recovery prompted by massive state intervention as well as a flexible reaction of individual players in the firm sector. As expected, no clear-cut support can be given to the main hypotheses stated at the outset of this chapter. Undoubtedly a number of transformations have taken place, as suggested by official statistics and, in a specific case, confirmed by survey data. Nonetheless, the same secondary data underline that the growth of the Chinese economy continues to be fuelled by very different sectorial and regional growth engines. The different intensity with which the impact was felt among sectors and regions has clearly revealed the substantial scope of persisting structural disparities in the Chinese industrial sector. Overall, this chapter has identified four distinct but overlapping trajectories of resilience, recovery and further development. Likewise, it has documented the survival of different growth models from Hebei to Zhejiang to Guangdong. In summary, this suggests that the different trajectories of recovery cannot one-dimensionally be explained as the result of sectors and regions profiting to different degrees from publicly triggered demand. On the contrary, it appears conclusive to assume that the crisis has prompted various shifts in the strategies of various firms with regard to both technological development and market orientation. Even within one specific sector, the Guangdong firm-level data reveal complex processes of adaptation as well as co-existing, similarly dynamic

ERIKSSON 9780857930088 PRINT.indd 116

29/01/2013 16:59

The global economic crisis as leverage for emerging regional growth

117

business models. One finding, however, is unambiguous: technological upgrading has been underway – during the crisis and most certainly in its wake. In this respect, patent data confirm that, in line with our assumptions, the most resilient momentum was that in the field of closeto-market technologies. Apparently, leading-edge activities were more dispensable than low- to mid-level upgrading strategies. In line with that, the Guangdong survey results seem to suggest that, even with an orientation towards the domestic market, technological ambitions are rising. In summary, the Chinese economy had to – and has to – weather a slowdown in the world economy that at first seemed to question at least the export-based component of its growth model. Apparently, however, neither a partial collapse nor a full-fledged reorientation of that model has occurred. Instead, different players have reacted in different ways and different growth models continue to exist in different places. In the end, both the domestically oriented Bohai Bay area and export-oriented Guangdong have returned to specific, stable growth paths. If anything, further differentiation has ensued. Only the country’s major cities and certain low-tech export sectors seem to have shifted to a lower growth path as a result of the 2008/09 external shock, arguably for the better as the extent of agglomeration in those fields and regions is already quite high and their growth rates remain well above 5 per cent. By means of an outlook, the recent political decisions regarding the artificial slowdown of development in Zhejiang suggest that the Chinese problem had in early 2011 again become one of cooling down an overheating economy rather than combating a downturn. At the time of writing these conclusions, however, the next global crisis may well be dawning on the horizon. Unaffected by all this, technological upgrading in China continues.

NOTE 1. http://english.eastday.com/e/110211/u1a5716521.html; accessed 6 November 2012.

REFERENCES Altenburg, T., H. Schmitz and A. Stamm (2008), ‘Breakthrough? China’s and India’s Transition from Production to Innovation’, World Development, 36(2), 325–44. Bell, M. and K. Pavitt (1995), ‘The Development of Technological Capabilities’, in Irfan ul Haque (ed.), Trade, Technology and International Competitiveness, Washington, DC: World Bank, pp. 69–101. Chen, L. and P. De Lombaerde (2010), ‘The Crisis in the U.S. and the Future of

ERIKSSON 9780857930088 PRINT.indd 117

29/01/2013 16:59

118

Clusters and economic growth in Asia

East Asian Production Sharing’, Global Journal of Emerging Market Economies, 2(1), 91–108. Guan, J. and K. Chen (2010), ‘Measuring the Innovation Production Process: A Cross-region Empirical Study of China’s High-tech Innovations’, Technovation, 30(5–6), 348–58. He, C. (2009), ‘Industrial Agglomeration and Economic Performance in Transitional China’, Chapter 16, in World Bank (ed.), World Development Report 2009, Reshaping Economic Geography, Washington, DC: The World Bank, pp. 258–81 HKCPU (2010), ‘Hong Kong’s Role in the Development of the Mainland’, Paper Ref: CSD/1/2010, Hong Kong: Central Policy Unit. Kim, L. (1997), Imitation to Innovation: The Dynamics of Korea’s Technological Learning, Boston, MA: Harvard Business School Press. Kroll, H. and D. Schiller (2010), ‘Establishing an Interface Between Public Sector Applied Research and the Chinese Enterprise Sector: Preparing for 2020’, Technovation, 30(2), 117–29. Lall, S. (1992), ‘Technological Capabilities and Industrialization’, World Development, 20(2), 165–86. Lee, K. and C. Lim (2001), ‘Technological Regimes, Catching-up and Leapfrogging: Findings from the Korean Industries’, Research Policy, 30(3), 459–83. Naudé, W. (2009), ‘The Financial Crisis of 2008 and the Developing Countries’, UNU-WIDER Discussion Paper No. 2009/01, Helsinki: UNU-WIDER. OECD (2010), Economic Survey of China 2010, Paris: Organisation for Economic Co-operation and Development. Overholt, W.H. (2010), ‘China in the Global Financial Crisis: Rising Influence, Rising Challenges’, The Washington Quarterly, 33(1), 21–34. Schüller, M. and Y. Schüler-Zhou (2009), ‘China’s Economic Policy in the Time of the Global Financial Crisis: Which Way Out?’, Journal of Current Chinese Affairs, 38(3), 165–81. Simmie, J. and R. Martin (2010), ‘The Economic Resilience of Regions: Towards an Evolutionary Approach’, Cambridge Journal of Regions, Economy and Society, 3(1), 27–43. Stiglitz, J. (2008), ‘China: Towards a New Model of Development’, China Economic Journal, 1(1), 33–52. Wong, P.K. (1999), ‘National Innovation Systems for Rapid Technological Catch-up: An Analytical Framework and a Comparative Analysis of Korea, Taiwan and Singapore’, paper presented at the DRUID Summer Conference, 9–12 June 1999, Rebild, Denmark. World Bank (2009), World Development Report 2009. Reshaping Economic Geography, Washington, DC: The World Bank. Zhou, Y. (2005), ‘The Making of an Innovative Region from a Centrally Planned Economy: Institutional Evolution in Zhongguancun Science Park in Beijing’, Environment and Planning A, 37(6), 1113–34. Zhou, Y. (2008), ‘Synchronizing Export Orientation with Import Substitution: Creating Competitive Indigenous High-tech Companies in China’, World Development, 36(11), 2353–70.

ERIKSSON 9780857930088 PRINT.indd 118

29/01/2013 16:59

6.

Technological intensity of FDI in Vietnam – implications for future economic development and emerging clusters Curt Nestor

INTRODUCTION Strategically located in Southeast Asia, in the midst of a most dynamic economic region, Vietnam embarked on an ambitious economic reform programme at the end of the 1980s. The reforms included foreign direct investment (FDI) policies with the aim of promoting economic development through technological transfer and employment generation. Over the past two decades, the country has achieved a most remarkable economic development that has radically transformed the economic landscape. The foreign invested sector has made substantial contributions to the average GDP growth rates, which exceeded 7 per cent during the period. The economic growth has been broad-based and has led to improved living standards for large parts of the Vietnamese population of 89 million. The nominal GDP per capita was recorded at less than USD 100 in the early 1990s, but Vietnam reached the status of a (lower) middle-income country with a GDP per capita of USD 1000 in 2008. The Vietnamese government aims to achieve the status of a modern and industrialized country by 2020. This chapter first provides a brief outline of the theoretical underpinnings of industrial agglomerations and clusters and the potential role of FDI in the transfer of technology in developing countries. The next section presents an overview of the current status of FDI in Vietnam and highlights the role of industrial enterprises in industrial zones (IZs) in promoting manufacturing activities over time, followed by a detailed examination of the technological content of FDI production during the 1990s. The subsequent sections examine changes in the size and structure of manufacturing output and value-added, and an appraisal of the technological sophistication embodied in the foreign trade of manufactured

119

ERIKSSON 9780857930088 PRINT.indd 119

29/01/2013 16:59

120

Clusters and economic growth in Asia

products to date. The chapter concludes with a discussion on the extent to which the objectives of FDI-related transfer of technology and employment generation have been achieved in Vietnam over the period.

INDUSTRIAL CLUSTERS AND FDI-INDUCED TRANSFER OF TECHNOLOGY The literature on the importance of industrial agglomerations and clusters and the role of FDI in the diffusion of technology driving economic development has expanded vastly in recent decades, offering new insights into the phenomena.1 For the purpose of this chapter, the notion of industrial agglomeration refers to the geographical concentration of economic activities in the manufacturing sector, that is, co-location of industries. Based on Marshall’s ([1890] 1920) concepts from the late nineteenth century, these industries primarily benefit from agglomeration economies, for example, reduced transport costs and labour market pooling. However, enterprises largely operate independently from each other, with limited interaction and collaboration. The concept of an industrial cluster here denotes industries that, apart from being located in close geographical proximity, have also developed functional relationships in the form of deeper interaction between companies in related and supporting activities, producing synergies within the cluster (Porter, 1998). The potential of FDI-linked transfer of technology in emerging industrial  agglomerations and clusters in developing countries is widely recognized in the literature. Anwar and Nguyen (2010) and Le and Pomfret (2011) recently examined the issue of FDI and the transfer of technology in Vietnam and, in the interest of space, the theoretical framework applied in this chapter is based on a summary of their findings. The diffusion of technology from foreign invested enterprises (FIEs) may induce various spillover effects to domestic enterprises, primarily in the form of horizontal and vertical spillovers. Horizontal spillovers occur within the same industry and refer to demonstration effects of technology and management methods applied by FIEs and imitated by domestic enterprises. Technology can also be disseminated by labour turnover, that is, labour trained and employed by FIEs moves to enterprises in the domestic sector. FDI production is also likely to lead to increased competitive pressure on domestic enterprises, which are compelled to upgrade their technologies in order to meet the competition. Vertical spillovers take place within the same industry and may produce positive externalities in the form of backward or forward linkages.

ERIKSSON 9780857930088 PRINT.indd 120

29/01/2013 16:59

Technological intensity of FDI in Vietnam

121

Backward linkages refer to the transfer of technology resulting from the interaction between FIEs and domestic suppliers. FIEs may provide direct technological assistance and training in close cooperation with domestic enterprises in order to secure the supply of locally produced inputs of the required quality. Alternatively, FIEs may indirectly compel domestic suppliers to upgrade their existing technology and production methods in order to meet the cost and quality requirements of the FIEs. In the process, the technological capability of domestic enterprises is improved. Technology may also be disseminated through various forward linkages, for example, domestic enterprises may obtain inputs of superior quality at competitive costs produced by FIEs, leading to improved productivity among the enterprises in the domestic sector. The theoretical aspects of industrial agglomerations and clusters, and the transfer of technology from FIEs to domestic enterprises in the Vietnamese context, are further examined in the concluding discussion of this chapter.

FDI IN VIETNAM – AN OVERVIEW Vietnam has been extremely successful in attracting foreign capital since 1988, when legislation encouraging FDI was introduced. The Vietnamese investment environment has improved over time and has recently been examined in detail, for example by UNCTAD (2008) and OECD (2009). Vietnam gained three positions and was ranked eighth among the top priority host economies for FDI for the period 2010–12 in the most recent edition of the World Investment Prospects Survey (UNCTAD, 2010). The Vietnam Foreign Investment Agency (FIA, 2012) reports that 13 664 FIEs had been established in Vietnam by the end of 2011. In total, these enterprises account for USD 198 billion in registered approved investment capital,2 whereas the actual disbursement is lagging and accounts for about 45 per cent to date.3 Following the removal of the US trade and investment embargo in 1994 and Vietnam joining the Association of Southeast Asian Nations (ASEAN) in 1995, the prospects of the growing Vietnamese domestic market unleashed a major wave of FDI in the mid-1990s. In the aftermath of the Asian financial crisis, FDI inflows temporarily contracted towards the end of the decade. However, renewed momentum has been building in recent years, spurred on by the anticipated implications of the bilateral trade agreement with the USA in 2001, the admission of Vietnam to the World Trade Organization (WTO) in 2007 and the ASEAN–China Free Trade Agreement phased in from 2010. Vietnam is currently experiencing

ERIKSSON 9780857930088 PRINT.indd 121

29/01/2013 16:59

122

Clusters and economic growth in Asia

80

1800 Approved capital Disbursed capital Number of projects

70 USD Billion

60

1600 1400 1200

50

1000

40

800

30

600

20

400

10

200 0

19 95 19 96 19 97 19 98 19 99 20 00 20 01 20 02 20 03 20 04 20 05 20 06 20 07 20 08 20 09 20 10 20 11

0

(prel.) Source: GSO (2011) and FIA (2012).

Figure 6.1

Vietnam inward FDI flows 1995–2011

an investment boom with high and sustained levels of GDP growth and unprecedented FDI capital inflows (Figure 6.1). About three-quarters of the total approved FDI capital and almost one-half of the total number of FIEs have been approved and established since Vietnam’s entry to the WTO. Foreign investors participate in all sectors of the Vietnamese economy, although to different extents. The overwhelming majority of FIEs (7987 projects) and approved investment capital (USD 93 billion) has been directed to manufacturing activities. Foreign investors have also made substantial contributions to the development of the service sector in terms of real estate developments and the hospitality business. Furthermore, FDI has been instrumental in the development of the oil industry, whereas activities in the agricultural sector, for example, have been modest. The vast majority of FIEs in Vietnam have been established in the form of wholly foreign-owned enterprises (10 592 projects), whereas the joint venture mode of entry accounts for 2644 projects. Manufacturing FDI was often subject to the mandatory joint venture form of investment in the 1990s; however, such restrictions have since been relaxed and manufacturing activities are today typically performed by wholly foreign-owned enterprises. FDI has been sourced from more than 90 countries, though most foreign investors and FDI capital originate from countries in East and Southeast Asia, notably Singapore, Japan, South Korea, Taiwan and Hong Kong; there have also been rapidly increasing investments from China in recent years. However, FDI capital from source countries in

ERIKSSON 9780857930088 PRINT.indd 122

29/01/2013 16:59

Technological intensity of FDI in Vietnam

123

Europe, North America and Australia has grown in importance during the last decade (FIA, 2012). A number of FDI activities in Vietnam are conducted via foreign subsidiaries located in Hong Kong and Singapore or via tax haven jurisdictions, which distorts the true geographical origin of FDI, suggesting that the capital inflows from the USA and EU member countries, for example, are actually larger than indicated by the official data (Nestor, 2007, pp. 156f). FDI has been dispersed across the whole country over time, albeit with major concentrations in three designated key economic areas (KEAs). In essence, the KEAs represent a growth pole strategy introduced in the 1990s based on urban networks centred on the three major municipal areas of Hanoi, Da Nang and Ho Chi Minh City and the adjacent provinces, in the northern, central and southern parts of the country, respectively (ibid., pp. 113ff).

INDUSTRIAL ZONE DEVELOPMENTS OVER TIME An IZ is a developed site for industrial enterprises offering standard factory accommodation with associated services and other utilities, for example, an ensured power and water supply and telecommunication systems, provided to the tenants. The tenants are generally composed of small- or medium-scale enterprises attracted to the IZ by special incentives and preferential treatment in terms of simplified administrative procedures. The zones were established with the objectives of attracting FDI, creating employment and promoting exports and foreign exchange earnings, and encouraging the use of modern technologies and management. The development of IZs evolved rapidly over time and the zones play a vital role in the foreign as well as the domestic manufacturing sector in Vietnam today. Since the first IZ was established in Ho Chi Minh City in 1991, the number of zones has mushroomed. By 2011, the total number of approved IZs reached 283, including 31 IZs established by foreign developers.4 A total of 180 IZs are currently in operation and the remaining 103 IZs are under construction (MPI, 2012, pp. 7, 10). Encouraged by various government policies, the location of foreign invested as well as domestic manufacturing activities, including both newly established enterprises and relocated industries, is increasingly concentrated on IZs. The 180 IZs in operation by the end of 2011 contained almost 8800 foreign and domestic invested enterprises (Table 6.1). The zones were initially primarily intended to accommodate FIEs, however, the number of domestic enterprises located in IZs has increased rapidly over the last decade and has superseded the number of foreign controlled firms since 2004.

ERIKSSON 9780857930088 PRINT.indd 123

29/01/2013 16:59

124

Table 6.1

Clusters and economic growth in Asia

Industrial zone developments 1991–2011 (cumulative data) 1991

Number of industrial zones Total area (ha) Enterprises by ownership: Foreign invested enterprises Domestic enterprises Registered investment capital: FDI (USD billion) Domestic capital (VND billion)

1995

2000

2005

2010

2011

1

12

65

131

267

283

300

2360

12 066

25 206

71 614

76 000

0

155

743

2120

3980

4113

0

50

500

2370

4380

4681

0 0

1.5 8.7 16.8 53.6 59.6 200 35 200 115 200 334 000 420 000

Source: Compiled by author based on MPI (2012).

The total stock of approved FDI capital of the 4113 FIEs currently located in IZs amounts to USD 59.6 billion,5 of which about 27.0 billion (45 per cent) has been realized. The intended investment capital of domestic enterprises (4681) totals VND 420 000 billion (equivalent to approximately USD 20 billion), of which an estimated VND 200 000 billion (47.6 per cent, equivalent to USD 9.5 billion) has been disbursed. This yields an average capital value of FIEs located in IZs more than three times larger than the capital size of domestic enterprises (MPI, 2012, pp. 8ff). In 2010, IZ enterprises provided employment for more than 1.6 million workers, and recorded a turnover of almost USD 34 billion, of which the turnover of FIEs comprised USD 31.5 billion. Most of the USD 19 billion worth of goods exported from IZs is also attributed to FIEs as well as the USD 18.5 billion worth of imports (World Bank, 2011, pp. 59f). The IZ developments form an extensive network of manufacturing agglomerations located in 58 of Vietnam’s 63 provinces6 with a major concentration of IZs in the three KEAs, where 199 of the 267 approved IZs were located in 2010 (Table 6.2). The three KEAs are economically the most developed areas of Vietnam, offering favourable conditions in terms of major urban areas providing access to labour, consumer markets and sea freight trade. About three-quarters of the national total gross value of industrial output (GVIO) in 2010 was derived from the three KEAs, including important contributions to the gross value of manufacturing output of foreign (and domestic) enterprises located in IZs within these areas. In fact, the economic performance of the northern and southern

ERIKSSON 9780857930088 PRINT.indd 124

29/01/2013 16:59

Technological intensity of FDI in Vietnam

Table 6.2

125

Geographic distribution of IZs and total GVIO by key economic area, 2010

Geographical Area Total Key economic areas (KEAs) of which Northern KEA Central KEA Southern KEA Outside KEAs

Number of Industrial Zones

GVIO (1994 constant VND, %)

267 199

808 745 billion 74.2

52 23 124 68

24.0 5.4 44.8 25.8

Note: Total gross value of industrial output (GVIO) includes mining, manufacturing and utilities. Source: MPI (2012) and GSO (2011) (calculations by author).

KEAs, the manufacturing hubs of Hanoi and Ho Chi Minh City with the adjacent provinces, respectively, is outstanding and accounted for about two-thirds of the total number of IZs and the total GVIO in 2010. However, the development of IZs comes at a cost. In the fierce competition for the use of scarce land, larger tracts of agricultural land in the vicinity of major urban centres have been converted to industrial use. The zones have rapidly increased the local demand for electricity, water and other utilities, and caused environmental concerns. The employment opportunities in the zones have generated new migratory flows, requiring improved social and physical infrastructure. The total area reserved for IZ development is currently 76 000 ha, yielding an average size of 268 ha per IZ. The IZ land area is planned to increase to 130 000 ha by 2015 and to expand further to 200 000 ha by 2020 in order to provide space for a total of around 500 approved and proposed IZ developments. In fact, many of the currently planned IZ developments are not likely to materialize, as this may lead to an oversupply of IZs similar to the situation during the latter half of the 1990s in the aftermath of the Asian financial crisis (Nestor, 2007, pp. 229ff). The current average occupancy rate of prepared land for the industrial use of IZs in operation is about 65 per cent (MPI, 2012, p. 9), however, the ratio of leased areas to total areas reserved for IZs is well below 50 per cent, suggesting that there are large tracts of idle land yet to be developed, let alone tenants to be found. In light of this, the Vietnamese Prime Minister recently issued instructions for a temporary moratorium on the construction of new IZ developments during 2012 pending a review of the situation of the existing IZs.7

ERIKSSON 9780857930088 PRINT.indd 125

29/01/2013 16:59

126

Clusters and economic growth in Asia

TECHNOLOGICAL CONTENTS OF MANUFACTURING FDI The Vietnamese government introduced legislation permitting FDI in 1988, aiming inter alia to advance the national level of technological know-how. In support of this objective, a number of legislative acts have been issued by the government over the years in order to promote and facilitate the transfer of technology. The extent to which this major objective was achieved during the 1990s is assessed here based on data derived from detailed records of all the licensed FDI projects in Vietnam during the period 1988–2000 compiled by the author (Nestor, 2007). The analyses of the technological content embodied in manufacturing FDI result in classifications of low technology (LT), medium-low technology (MLT) and medium-high technology (MHT) projects. The categories were formed by grouping the 2-digit International Standard Industrial Classification (ISIC) Rev. 3 code (UN 1990) associated with each manufacturing FIE based on a classification schema applied by UNIDO (2011a) (Table 6.3). The degree of technological intensity applied by FIEs provides an indication of their inherent capability and potential for transfer of technology and is here assessed in terms of location and source country. A total of 2231 manufacturing FDI projects comprising USD 21.3 billion in registered investment capital were approved during the period 1988–2000. Foreign investors’ preferences for locating projects of different technological sophistication in one of the 53 IZs in operation in 2000 rather than at alternative sites elsewhere are examined in Table 6.4. In total, more than one-third of manufacturing FIEs (759) and approved capital (USD 7.3 billion) were located in IZs. Most projects and the largest capital commitments in manufacturing activities were thus located outside IZs.8 The general results indicate an investment pattern dominated Table 6.3

Technological classification of manufacturing value-added, ISIC Rev. 3

Type of Activity

ISIC Division

Low-technology (LT) manufacturing Medium-low technology (MLT) manufacturing Medium- and high-technology (MHT) manufacturing

15–22, 36, 37 23, 25–28, 351 24, 29–35 (excluding 351)

Note: See UN (1990) for a complete list of ISIC Rev. 3 codes with explanatory notes. Source: UNIDO (2011a).

ERIKSSON 9780857930088 PRINT.indd 126

29/01/2013 16:59

Technological intensity of FDI in Vietnam

Table 6.4

127

Approved capital in manufacturing FDI by technological intensity and location, end of 2000 (USD million and %)

Technological Intensity Total Low technology Medium-low technology Medium and high technology

Projects Approvals Projects Approvals Projects Approvals Projects Approvals

Total

IZ

Non-IZ

2231 21 342 1186 8902 463 7034 582 5405

759 34.0 346 36.3 154 19.2 259 49.4

1472 66.0 840 63.7 309 80.8 323 50.6

Note: Data refer to the total number of manufacturing FIEs (ISIC D) licensed during the period 1988–2000. Source: MPI extended database (calculations by author).

by labour-intensive activities with low technological (LT) content, such as food and beverages and textiles. The LT group comprises more than one-half of the projects (1186), including 346 projects in IZs, and over twofifths of the total approved capital (USD 8.9 billion) in manufacturing. The production of most FIEs involved in LT activities was export-oriented. The MLT type of projects numbered 463, with committed capital of USD 7.0 billion in primarily more capital-intensive production of cement and other construction materials. The MHT type of investments mainly refers to the production of chemicals, motor vehicles, electric machinery and electronic equipment. This group attracted a comparatively large number of projects (582), although the FDI capital was limited to USD 5.4 billion. The majority of both MLT and MHT activities was aimed at the domestic market and located outside IZs. Foreign investors from different source countries exhibited different patterns in terms of the technological intensity applied in their investment undertakings (Table 6.5). Only investors from Japan and North America (mainly the USA) allocated the majority of the committed FDI capital to MHT activities. Japanese investors registered the largest amount of approved FDI capital, USD 1.7 billion in 98 projects, yielding an average size of USD 17.3 million per project. The number of projects of North American investors was limited to 34, with an average size of USD 14.7 million. By comparison, investors from Western Europe and South Korea registered more investment capital in MLT activities than the MHT group. The combined FDI of the MLT and MHT groups from the ASEAN group of countries (mainly Singapore) exceeded the FDI capital dedicated to

ERIKSSON 9780857930088 PRINT.indd 127

29/01/2013 16:59

128

ERIKSSON 9780857930088 PRINT.indd 128

29/01/2013 16:59

238 270 615 149 330 213 103 313 2231

3046 2536 3853 953 3744 2562 1035 3612 21 342

Approvals 94 158 359 110 163 115 43 144 1186

Projects 377 1145 2263 758 1704 1096 305 1253 8902

Approvals

Low Technology

46 56 111 18 74 38 26 94 463

Projects 971 737 1110 92 1166 825 229 1905 7034

Approvals

Medium-low Technology

Source: MPI extended database (calculations by author).

Note: Data refer to the total number of manufacturing FIEs (ISIC D) licensed during the period 1988–2000.

Japan South Korea Taiwan Hong Kong ASEAN West Europe North America Tax havens and others Total

Projects

Total

98 56 145 21 93 60 34 75 582

Projects

1699 654 480 103 874 642 500 454 5405

Approvals

Medium and High Technology

Approved capital in manufacturing FDI by technology intensity and major source countries, end 2000 (USD million)

Country/Country Groups

Table 6.5

Technological intensity of FDI in Vietnam

129

the LT group. Conversely, the majority of Taiwanese and Hong Kong investments in terms of the number of projects and capital were aimed at LT activities. Investors from Japan and South Korea together with the ASEAN countries and Western Europe also made important capital commitments to a large number of projects involving LT activities. In sum, the FDI activities during the 1990s were primarily concentrated on labour-intensive activities with limited but increasing technological sophistication over time. Thus, the main objective of the Vietnamese government, to promote technological development, was mainly achieved in terms of labour-intensive low technology activities providing employment to some 250 000 workers in the year 2000.

MANUFACTURING GROSS OUTPUT AND VALUE-ADDED In order to highlight further the impact of FDI on the Vietnamese manufacturing sector in terms of technological content over time, the manufacturing gross output and value-added are here examined in detail for the period 1995–2010. Data on the gross value of industrial output (GVIO) in constant 1994 VND are derived from the Vietnamese General Statistics Office (GSO, 2004 and 2011) with the same classification of technological content as applied above, with some modifications.9 The manufacturing GVIO roughly doubled every five years from VND 83 261 billion in 1995 to VND 722 222 billion in 2010 (Table 6.6). In 1995, the state-owned enterprises (SOEs) dominated the manufacturing sector with over one-half of the total GVIO. Over the next 15 years, the GVIO structure changed with rapidly decreasing relative importance of SOEs and a corresponding rise in the private domestic sector (PDS) and especially in FDI. Whereas the GVIO of SOEs increased more than threefold in absolute numbers over the period 1995–2010, the SOE relative share in the total GVIO decreased from 52.1 per cent in 1995 to 18.8 per cent in 2010. The contribution of private domestic enterprises increased by 8.6 percentage points to 38.4 per cent and the recorded share of the foreign invested sector more than doubled from 18.1 per cent to 42.8 per cent of the total manufacturing GVIO during the same period. The structure of the technology applied in manufacturing activities also changed over the period, with a major decrease in the GVIO share in LT activities and corresponding increases in the MLT and MHT categories. Manufacturing FDI in Vietnam today accounts for two-thirds of MHT activities in the country. A large share of especially LT manufacturing by foreign investors is concentrated in IZs, together with increasing shares of

ERIKSSON 9780857930088 PRINT.indd 129

29/01/2013 16:59

130

Clusters and economic growth in Asia

Table 6.6

Year

1995

2000

2005

2010

Vietnam gross value of industrial output in manufacturing 1995–2010: technological intensity by ownership (constant 1994 VND billion and %)

Technological Intensity

Total Low technology Medium-low technology Medium and high technology Total Low technology Medium-low technology Medium and high technology Total Low technology Medium-low technology Medium and high technology Total Low technology Medium-low technology Medium and high technology

Total

83 261 52 520 19 467 11 273 158 098 88 200 43 043 26 856 351 685 184 236 104 270 63 179 722 222 351 552 238 614 132 056

Domestic Sector SOE

PDS

52.1 49.9 53.1 60.8 42.7 44.7 39.8 40.6 30.7 31.1 31.5 28.1 18.8 14.7 26.0 16.4

29.8 33.3 30.4 12.1 27.2 32.7 26.5 10.3 33.2 39.3 33.9 14.2 38.4 48.1 36.4 16.3

FDI

18.1 16.8 16.4 27.2 30.1 22.5 33.7 49.2 36.1 29.5 34.6 57.7 42.8 37.2 37.5 67.3

Note: Classification of technological intensity based on ISIC Rev. 3 (1995–2000) and ISIC Rev. 4 converted to ISIC Rev.3 (2005–10). See text for qualifications of data. SOE 5 stateowned enterprise; PDS 5 private domestic sector; FDI 5 foreign direct investment. Source: GSO (2004 and 2011) (calculations by author).

MLT and MHT activities. Furthermore, the GVIO data indicate a general shift over time from LT manufacturing to MLT and MHT activities in the FDI sector and among SOEs as well, a trend that is less pronounced among companies of the PDS. Data on the manufacturing value-added (MVA), that is, the sum of the gross output less the value of the intermediate inputs used in the production of goods, are only available at the aggregate value of the entire manufacturing sector expressed in constant 2000 USD (World Bank, 2012). The MVA is a measure of the national industrial capacity, a basic indicator of a country’s level of industrialization. The Vietnamese MVA developed in unison with economic growth and recorded an impressive and consistent increase during the period, from USD 3.5 billion in 1995 to USD 14.6 billion in 2010. Adjusted for population size, the Vietnamese per capita MVA increased from a mere USD 47 in 1995 to USD 180 in 2010 (Table 6.7).

ERIKSSON 9780857930088 PRINT.indd 130

29/01/2013 16:59

Technological intensity of FDI in Vietnam

Table 6.7 Year

131

Manufacturing value-added in Vietnam 1995–2010 Manufacturing Value-added (MVA), Constant 2000 USD Total MVA (USD billion)

Per capita MVA (USD)

3.4 5.8 10.0 15.6

47 75 122 180

1995 2000 2005 2010 Source: World Bank (2012).

UNIDO publishes a Competitive Industrial Index benchmarking 118 countries based on a composite index measuring several dimensions of national industrial performance. Vietnam was ranked fifty-eighth in 2009, gaining 14 positions since the previous ranking in 2005 (UNIDO, 2011a). This is a major achievement indicating significant improvements in Vietnam’s relative competitive position among regional economies in general, as well as on the global manufacturing scene. Nevertheless, in the regional context, Vietnam still lags behind most other countries, suggesting that the country still has some distance to cover before reaching the level of competitiveness of the more developed members of ASEAN and China, let alone Taiwan and South Korea (UNIDO, 2011b, pp. 27f).

TECHNOLOGICAL CONTENT OF VIETNAM’S FOREIGN TRADE As yet another piece of evidence of the important contributions made by FIEs to the Vietnamese economic development over the last two decades, the technological content of Vietnam’s foreign trade is examined here. Vietnam’s foreign merchandise trade followed a tremendous trajectory from the early 1990s with a rapidly changing structure in terms of the value and composition of goods as well as its trade partners. The total trade turnover was recorded at a mere USD 13.6 billion in 1995 and reached a new record level exceeding USD 200 billion in 2011; the exports were valued at USD 96.9 billion and the imports amounted to USD 106.7 billion. FDI-related exports accounted for 53.4 per cent of the Vietnamese total exports (excluding crude oil) in 2011 and the share of imports was estimated at 45.7 per cent (Vietnam Customs, 2012). FDI thus accounts for a significant share of the non-oil exports, and especially the manufactured exports, as well as an important share of the imports, that is, raw

ERIKSSON 9780857930088 PRINT.indd 131

29/01/2013 16:59

132

Table 6.8

Clusters and economic growth in Asia

Vietnam manufacturing trade 1996–2010, by skill and technological intensity

Year

Total (USD billion) Manufactured goods (USD billion) By degree of manufacturing (%) Labour-intensive and resource-based Low skill and technology intensity Medium skill and technology intensity High skill and technology intensity Unclassified Year

Total (USD billion) Manufactured goods (USD billion) By degree of manufacturing (%) Labour-intensive and resource-based Low skill and technology intensity Medium skill and technology intensity High skill and technology intensity Unclassified

Exports (annual average) 1996–2000

2001–05

2006–10

10.4 4.5

22.2 11.1

55.8 31.0

73.1 2.5 9.5 12.5 2.4

69.7 5.1 11.0 11.7 2.5

60.3 7.1 13.7 14.5 4.4

Imports (annual average) 1996–2000

2001–05

2006–10

12.3 9.3

26.0 19.0

68.4 48.0

25.0 15.6 24.9 33.3 1.3

23.1 16.7 27.7 31.0 1.6

18.2 18.1 29.6 32.7 1.4

Note: Degree of manufacturing refers to SITC 5 to 8 less 667 and 68. See UNCTADSTAT (2011) for details of the composition of commodity groups. Reporter: Vietnam (1996–2009) and UNCTAD estimates based on trade partner mirror statistics (2010). Source: UNCTADSTAT (2012) (calculations by author).

materials, parts and components used in FDI assembly of final goods for exports or for sale on the domestic market. The development of trade over the period 1996–2010 is shown in Table 6.8. The technological content of Vietnam’s foreign trade is based on UNCTADSTAT’s (2011) application of the Standard International Trade Classification (SITC) of goods in four categories by degree of manufacturing in terms of skill and technological intensity, namely labourintensive and resource-based manufacturing and manufactures with low/ medium/high skill and technology intensity. Manufactured exports grew faster than the total exports and the share of manufactured goods in the total exports increased from an average of

ERIKSSON 9780857930088 PRINT.indd 132

29/01/2013 16:59

Technological intensity of FDI in Vietnam

133

43.7 per cent in 1996–2000 to 56.6 per cent in 2006–2010. Though still of the utmost importance for Vietnam’s exports, the share of labourintensive and resource-based manufactured exports recorded a decrease from 73.1 per cent in the late 1990s to 60.3 per cent during 2006 to 2010. Exports of goods involving different skill and technological intensity levels increased over the period, during which the largest earnings were derived from the export of manufactured goods of medium and high skill and technology intensity. Europe was the main export market for Vietnamese exports of manufactured goods during the latter half of the 1990s together with Japan, Taiwan and ASEAN member countries. The export structure in terms of the main trade partners changed over the next decade. The relative importance of the EU diminished, whereas exports to the USA expanded rapidly as a result of the bilateral trade agreement signed in 2001. By 2010, the USA absorbed more than 30 per cent of Vietnamese manufactured exports, including 43 per cent of labour-intensive and resource-based products. Manufactured imports grew at a slightly slower rate compared with the total imports and the share of manufactured goods in the total imports decreased from 75.3 per cent to 70.2 per cent during the period examined. The structure of imports in terms of skills and technological intensity is more equally distributed among the four categories. Similarly to exports, the labour-intensive and resource-based manufactured imports decreased over time, from 25.0 per cent in 1996–2000 to 18.2 per cent in 2006–2010. Manufacturing with low and medium skill and technological intensity increased over time. The largest expenses were generated by imports of manufactured goods characterized by high skill and technological intensity, that is, the kind of goods that to a large extent is not produced in Vietnam. The share of this category remained at the same level of about one-third of the total imports over the period. Japan and the EU were the main source countries of medium and high skill and technological intensity merchandise in the late 1990s, together with important contributions of imports from South Korea, Taiwan and Singapore. However, the import structure has changed during the last decade, with a marked decline in the import share from the EU and a rapid increase in imports from China. In 2010, over 30 per cent of Vietnam’s total manufactured imports were sourced from China, including similar proportions of goods of medium and high skill and technological intensity. Vietnam’s trade balance was negative and has increased over time (Table 6.9). The trade surplus in labour-intensive and resource-based manufacturing trade increased tenfold during the period 1995–2010, reaching an average of USD 10.0 billion during the period 2006–10.

ERIKSSON 9780857930088 PRINT.indd 133

29/01/2013 16:59

134

Table 6.9

Clusters and economic growth in Asia

Vietnam manufacturing trade balance 1996–2010, by skill and technological intensity (annual average, USD billion)

Year Total trade balance Manufactured goods trade balance Trade balance by degree of manufacturing: Labour-intensive and resource-based Low skill and technology intensity Medium skill and technology intensity High skill and technology intensity

1996–2000

2001–05

2006–10

–2.0 –4.8

–3.8 –7.9

–12.6 –17.0

1.0 –1.3 –1.9 –2.5

3.4 –2.6 –4.0 –4.6

10.0 –6.5 –9.9 –11.2

Note: See notes for Table 6.8. Source: UNCTADSTAT (2012) (calculations by author).

However, this surplus did not cover the deficit in trade of manufactured goods of a higher order of skills and technological intensity, producing an average annual deficit of USD 17 billion over the period 2006–10. In sum, Vietnam’s foreign trade is characterized by two major developments over the last decade: the rise of two new major trade partners – the USA in terms of a key export partner and China as the largest supplier of imported manufactured products including higher order technological goods. Vietnam recorded a surplus in trade of labour-intensive and resource-based manufactured goods, of which a large share was exported to the USA. Many of Vietnam’s imports of manufactured goods of different levels of skill and technological sophistication previously sourced from Europe, Japan and the emerging economies in East Asia are now increasingly supplied by China. The expanded trade with China has resulted in a rapidly increasing trade deficit. An important share of the manufactured imports consists of various inputs such as parts and components used in FDI assembly of final goods for exports or for sale on the domestic market, indicating the incorporation of Vietnam by FIEs into regional production networks. This type of production has rapidly developed in recent years, since Vietnam’s entry to the WTO, and is likely to continue to increase in the future.

CONCLUDING DISCUSSION The FDI-driven economic development of Vietnam has transformed the country from a predominantly agrarian-based society to an increasingly

ERIKSSON 9780857930088 PRINT.indd 134

29/01/2013 16:59

Technological intensity of FDI in Vietnam

135

competitive exporter of low-cost manufactured products integrated into regional production networks and the global economy over the last two decades. Stimulated by WTO membership in 2007 and other trade agreements, Vietnam has in recent years successfully attracted renewed attention from the international community of foreign investors as a plausible alternative to China for the location of labour-intensive investments in manufacturing activities as well as the production of higher technological content. The establishment of an extensive nationwide network of IZs has played an instrumental role in this process and a substantial part of foreign- and domestic-controlled manufacturing activities is today concentrated in such zones. The zones constitute potential centres for the transfer of technology and employment generation, two major objectives of the Vietnamese government’s FDI policy. The concept of an industrial cluster in the sense of closely integrated and mutually supporting enterprises in similar activities is relatively new in Vietnam and is often confused with that of IZs (Ketels et al., 2010, p. 91). Even though many IZs have explicit sectoral profiles, that is, IZs reserved for enterprises in specific sectors, in practice the current tenants of most IZs represent a collection of enterprises engaged in rather disparate types of activities. Though some of these IZs may form the embryos of future cluster formations, most zones currently rather constitute industrial agglomerations characterized by co-location of enterprises in diverse activities with limited functional ties and cooperation. Foreign investors are mainly engaged in manufacturing activities based on the abundant supply of low-cost labour, that is, labour-intensive production with limited technological content. Relatively modern technologies have been transferred to FIE operations in Vietnam primarily in the telecommunications, electronics and motor vehicle industries. However, there are also numerous reports of foreign investors importing dated, even obsolete, machinery and equipment in their labour-intensive light industry manufacturing in Vietnam, especially during the 1990s (Pham, 2004, pp. 71ff). More recent surveys of FDI production in Vietnam confirm the limited use of more advanced technologies. For example, a 2008 survey of 429 FIEs located in IZs in Ho Chi Minh City revealed that only a handful of enterprises attained a high-technology level in production and over one-half of the surveyed companies applied below-average technologies (Ketels et al., 2010, p. 45). Furthermore, Vietnamese FDI-related exports are heavily reliant on imports of intermediate inputs assembled in Vietnam into final products for exporting with limited domestic technological content. The import content in Vietnamese-manufactured exports is difficult to estimate but may reach two-thirds of the export value (MUTRAP, 2009, p. 17).

ERIKSSON 9780857930088 PRINT.indd 135

29/01/2013 16:59

136

Clusters and economic growth in Asia

A  recent survey covering 1970 FIEs, of which 866 are located in IZs, indicates the limited links to local suppliers. Only about 40 per cent of the intermediate inputs were sourced from the domestic market, mainly from other FIEs and SOEs (VNCI, 2012, p. 49). A growing body of studies examining the impact of FDI on the Vietnamese economic development over the last decade confirms the positive but limited contribution of FDI to technology transfer.10 The positive spillover effects are mainly confined to backward vertical linkages, that is, FDI-induced transfer of technology through interaction with domestic suppliers. In terms of horizontal spillover effects, positive demonstration effects are more than offset by the increased competitive pressures on domestic enterprises exercised by FIEs. Spillover effects of the transfer of technology from FIEs have been hampered by the limited absorptive capacity of domestic enterprises (Anwar and Nguyen, 2010, pp. 564f; Le and Pomfret, 2011, p. 198). This indicates that the efforts made by the Vietnamese government over the years to encourage and facilitate the transfer of technology from FIEs to domestic enterprises, for example by providing fiscal incentives, have not produced the desired results. However, the importance of the apparently limited levels of technology transferred by FIEs should not be underrated. Even though the most recent technology has generally not been applied in FIE undertakings in Vietnam, the machinery and equipment used are often of a more recent vintage than the existing stock in Vietnamese domestic companies (Nestor, 2007, p. 127). In addition, foreign invested manufacturing has generated a large number of employment opportunities whereby IZs provide direct and indirect employment to millions of Vietnamese workers exposed to the different production methods and comparatively modern machinery and equipment of FIEs. In sum, the Vietnamese government’s objective of promoting economic development by means of FDI has mainly been achieved in terms of employment generation and only partially fulfilled as regards the transfer of technology. The manufacturing sector has been at the core of Vietnam’s impressive economic growth over the last two decades and will be of the utmost importance for future economic development. To date, Vietnam has largely been used by foreign investors as a location for low-cost export-oriented manufacturing. In order to sustain the current degree of economic growth and reach the next level of economic development, Vietnam needs to develop higher value-added manufacturing. The current comparative advantage of an abundant supply of low-cost labour is being eroded over time as Vietnam meets increasing competition from other developing countries in the region. Despite the progress made in recent

ERIKSSON 9780857930088 PRINT.indd 136

29/01/2013 16:59

Technological intensity of FDI in Vietnam

137

years, there is no room for complacency as the country is facing several key challenges that need to be addressed urgently. As detailed by Ketels et al. (2010) and UNIDO (2011b), to reach the next level of economic development, the Vietnamese industrial policy needs to be coupled with reforms of the educational system in order to provide the Vietnamese labour force with appropriate skills. Furthermore, additional important improvement of the physical infrastructure in terms of transport, communication and energy is required in order to facilitate the transfer to a higher value-added economy in Vietnam. In this process, the development of industrial agglomerations and clusters in Vietnam is of paramount importance. The network of IZs constitutes the core of such industrial agglomerations. Some IZs have attracted the location of major FDI flagship or anchor firms, that is, major transnational corporations such as Intel, Samsung, Canon and Foxconn. However, leveraging the role of FDI anchor firms to promote the development of industrial clusters in Vietnam has to date proved less successful (Ketels et al., 2010, p. 91). Developing a local supply chain is often a protracted process, and due to the lack of supporting and related industries, major FIEs tend to rely on their existing production networks, leading to imports of intermediary inputs produced by established affiliates located in other countries. The FDI anchor firms’ efforts to build local supplier networks may eventually develop into the formation of clusters of related activities characterized by close and dynamic collaboration. A report commissioned recently by UNIDO issues recommendations on how to promote cluster formation in Vietnam. The report emphasizes that clusters cannot be created artificially through public policy (Coniglio et al., 2011, p. 32). Initially, the existing agglomerations of industrial activities need to be analysed and there have been two recent attempts to examine the spatial concentration of industrial activities in similar sectors (McCarty et al., 2005; UNIDO, 2010). Both studies indicate the occurrence of distinct patterns of provincial specialization and agglomeration forces within similar industrial sectors. These studies need to be complemented with detailed examinations in order to identify those agglomerations with the highest future potential to form a competitive cluster. An industrial cluster policy needs to include measures to facilitate and promote the development of domestic supporting industries in the targeted industrial agglomerations. Improved productivity and upgraded competitiveness of domestic supporting industries lead to expanded local supplier networks for FIEs that are likely to attract additional FDI to the emerging cluster. Such a policy would also entail the possibility for domestic enterprises to tap into regional and global production networks of FIEs. Provided there is proper implementation of the proposed policy measures, the Vietnamese

ERIKSSON 9780857930088 PRINT.indd 137

29/01/2013 16:59

138

Clusters and economic growth in Asia

government’s intention to achieve the status of a modern and industrialized country by 2020 may be realized.

NOTES 1.

See, for example, Kuchiki and Tsuji (2005), Yusuf et al. (2008) and Ganne and Lecler (2009) for an overview of the development of FDI and industrial agglomerations and clusters in Asia in general, and Kuroiwa and Toh (2008) and Kuroiwa (2009) for a discussion regarding Southeast Asia and Vietnam specifically. 2. This amount excludes USD 38 billion in registered capital of 2 048 FDI projects that have been revoked or have expired over time. 3. FDI capital values in Vietnam typically represent an all-inclusive concept that deviates from the IMF/OECD standard definition based on a balance of payment concept (see Freeman and Nestor, 2004 and Nestor, 2008 for a detailed discussion of Vietnamese FDI data sources and discrepancies). 4. The total number of IZs includes three export processing zones (EPZs) and two hi-tech zones (HTZs). Since the uniform investment law was introduced in 2005 and Vietnam’s admission to the WTO, there is today little difference between IZs and EPZs. Both HTZs are under development and have relatively few tenants, especially FIEs. All three types of zones are here referred to as IZs. 5. FIEs located in IZs represented 51.5 per cent of the total number of FIEs in manufacturing and 64.0 per cent of the total registered capital in manufacturing activities in Vietnam by 2011. 6. See MPI (2009) for maps indicating the location of approved IZs. 7. Prime Minister instruction 07/CT-TTg, dated 2 March 2012. 8. The distributional pattern of FIEs in IZ/non-IZ locations needs to be considered in the concurrent context: IZs were a relatively new phenomenon at the time and did not turn into an immediate success in terms of locational attractiveness. Furthermore, many IZs only became operational during the latter half of the decade, that is, during and after the Asian financial crisis in 1997–98. Most importantly, FDI in a number of manufacturing activities was at the time only permitted in the form of joint ventures with domestic companies. Considering the nascent nature of the private domestic sector in Vietnam during the 1990s, foreign investors were in practice compelled to form joint ventures with state-owned enterprises (SOEs) and typically located production in the existing premises of the SOEs, that is, outside IZs (Nestor, 2007, pp. 289f). 9. GVIO data are only available at the ISIC 2-digit level. The entire ISIC 35 division is here conservatively classified in the MLT group rather than the MHT group. 10. See UNIDO/MPI (forthcoming) for a comprehensive review of such studies.

REFERENCES Anwar, S. and P.L. Nguyen (2010), ‘Absorptive Capacity, Foreign Direct Investment-linked Spillovers and Economic Growth in Vietnam’, Asian Business & Management, 9(4), 553–70. Coniglio, N., F. Rota and G. Viesti (2011), Promoting Industrial Clusters in Vietnam, Bari (Italy): CERPEM/UNIDO. FIA (2012), Tinh hinh dau tu truc tiep nuoc ngoai 12 thang nam 2011 [The Situation of Foreign Direct Investment in 2011], Foreign Investment Agency,

ERIKSSON 9780857930088 PRINT.indd 138

29/01/2013 16:59

Technological intensity of FDI in Vietnam

139

5 January 2012, accessed 31 March 2012 at http://fia.mpi.gov.vn/News. aspx?ctl5newsdetail&p5&aID51128 [in Vietnamese]. Freeman, N. and C. Nestor (2004), ‘Rethinking Foreign Investment in Vietnam: Fuzzy Figures and Sentiment Swings’, in D. McCargo (ed.), Rethinking Vietnam, Rethinking Southeast Asia series, London: Routledge, pp. 178–96. Ganne, B. and Y. Lecler (eds) (2009), Asian Industrial Clusters, Global Competitiveness and New Policy Initiatives, Singapore: World Scientific. GSO (2004), Nien giam Thong ke Viet Nam 2003 – Vietnam Statistical Yearbook 2003, Hanoi: General Statistics Office [bilingual]. GSO (2011), Nien giam Thong ke Viet Nam 2010 – Vietnam Statistical Yearbook 2010, Hanoi: General Statistics Office [bilingual]. Ketels, C., D.C. Nguyen, T.T. Nguyen and H.H. Do (2010), Vietnam Competitiveness Report, Hanoi: CIEM. Kuchiki, A. and M. Tsuji (eds) (2005), Industrial Clusters in Asia: Analyses of Their Competition and Cooperation, Basingstoke: Palgrave MacMillan/IDE-JETRO. Kuroiwa, I. (ed.) (2009), Plugging into Production Networks – Industrialization Strategy in Less Developed Southeast Asian Countries, Singapore: IDE-JETRO and ISEAS. Kuroiwa, I. and M.H. Toh (eds) (2008), Production Networks and Industrial Clusters – Integrating Economies in Southeast Asia, Singapore: IDE-JETRO and ISEAS. Le, Q.H. and R. Pomfret (2011), ‘Technology Spillovers from Foreign Direct Investment in Vietnam: Horizontal or Vertical Spillovers?’, Journal of the Asia Pacific Economy, 16(2), 183–201. Marshall, A. ([1890] 1920), Principles of Economics (8th edition), London: Macmillan. McCarty, A., R. Record and J. Riedel (2005), ‘Part I: Industrial Clusters in Vietnam. Chapter 2: Competition and Cooperation: Vietnam’, in A. Kuchiki and M. Tsuji (eds), Industrial Clusters in Asia: Analyses of their Competition and Cooperation, Basingstoke: Palgrave MacMillan/IDE-JETRO, pp. 27–110. MPI (2009), Vietnam’s IPs, EPZs and EZs – Ideal Places for Manufacturing Base. A Guide to Investing in Vietnam’s IPs, EPZs and EZs, Hanoi: Ministry of Planning and Investment. MPI (2012), Ky yeu 20 nam Xay dung va Phat trien Khu Cong nghiep, Khu Che xuat, Khu Kinh te o Viet Nam 1991–2011 [Conference Proceedings: 20 Years of Developing Industrial Zones, Export Processing Zones, and Economic Zones in Vietnam 1991–2011], Hanoi: Ministry of Planning and Investment [in Vietnamese]. MUTRAP (2009), Analyzing Viet Nam’s Trade Deficit and the Balance of Payments Provisions of the WTO, Hanoi: EU Multilateral Trade Assistance Project. Nestor, C. (2007), ‘Foreign Direct Investment in the Socialist Republic of Vietnam 1988–2000 – Geographical Perspectives’, Doctoral dissertation, Department of Human and Economic Geography, School of Business, Economics and Law at University of Gothenburg, Series B, No. 112. Nestor, C. (2008), ‘A Note on FDI Statistical Issues in Vietnam’, in Foreign Investment Agency and Ministry of Planning and Investment, 20 Years of Foreign Investment 1987–2007 – Reviewing and Looking Forward, Hanoi: Knowledge Publishing House, pp. 129–36. OECD (2009), Policy Framework for Investment Assessment of Vietnam, Geneva: OECD.

ERIKSSON 9780857930088 PRINT.indd 139

29/01/2013 16:59

140

Clusters and economic growth in Asia

Pham, H.M. (2004), FDI and Development in Vietnam: Policy Implications, Singapore: ISEAS. Porter, M.E. (1998), On Competition, Boston, MA: Harvard Business School Press. UN (1990), International Standard Industrial Classification of All Economic Activities, Rev 3, Statistical Office of the United Nations, New York: United Nations. UNCTAD (2008), Investment Policy Review of Vietnam, Geneva: UNCTAD. UNCTAD (2010), World Investment Prospects Survey 2010–2012, New York and Geneva: UNCTAD. UNCTADSTAT (2011), Classifications Update (September), Geneva: UNCTAD. UNCTADSTAT (2012), Beyond 20/20, International Trade by Partner and Category/Product (online database), available at: http://unctadstat.unctad.org/. UNIDO (2010), ‘Identification of the Main Manufacturing Industry Clusters in Vietnam through a Statistical Approach’, Hanoi: UNIDO (mimeo). UNIDO (2011a), Industrial Development Report 2011, Vienna: UNIDO. UNIDO (2011b), Vietnam Industrial Competitiveness Report 2011, Hanoi: UNIDO. UNIDO/MPI (forthcoming), Vietnam Industrial Investment Report 2011, Hanoi: UNIDO/MPI. Vietnam Customs (2012), Statistics of Main Exports/Imports by Month, December 2011, Tables 14B and 15B, Hanoi: General Department of Customs. VNCI (2012), Vietnam Competitiveness Initiative 2011, Hanoi: USAID/VCCI. World Bank (2011), Vietnam Development Report 2012: Market Economy for a Middle-income Vietnam, Hanoi: World Bank. World Bank (2012), World Development Indicators 2011 Database (online), accessed 31 March 2012 at http://databank.worldbank.org/ddp/home.do. Yusuf, S., K. Nabeshima and S. Yamashita (eds) (2008), Growing Industrial Clusters in Asia: Serendipity and Science, Washington, DC: World Bank.

ERIKSSON 9780857930088 PRINT.indd 140

29/01/2013 16:59

7.

The aircraft industry as a tool for economic and industrial development – the case of Indonesia Sören Eriksson

INTRODUCTION Discussions occur regularly on the possible advantages of different kinds of industries and their influence or importance as creators of employment and economic development. Included in the question at hand are the possible spillover effects in specific industries, mostly high-tech industries (Eriksson, 2000). High-technology sectors are frequently cited objectives of regional development policy. High-technology industries are both misunderstood and overrated, although at the same time they are the most probable source of innovations, of successful entrepreneurs, of new firms and of new industries (Malecki, 1997). Pavitt (1990) argues that distinct modes of innovation can be observed across four sectors: science-based, scale-intensive, information-intensive and specialized supplier-dominated. Nelson and Rosenberg (1993) point out the differences between complex systems and other commodities such as chemicals and bulk commodities such as steel. In contrast to commodity goods, complex product systems are large customized engineering goods that are seldom, if ever, mass produced (Miller et al., 1995). Examples include aeroplanes, flexible manufacturing systems, flight simulators, telecommunication systems, chemical process plants and nuclear power plants. The product characteristics of complex system industries differ substantially to mass production goods, implying distinctive forms of innovation and organization. They embody at least three general characteristics: first, they are made up of many interconnected, often customized, elements (including control units, sub-systems and components), usually organized in a hierarchical way; second, complex systems exhibit non-linear and continuously emerging properties, where small changes in one part of the system can lead to large alterations in other parts of the system; and 141

ERIKSSON 9780857930088 PRINT.indd 141

29/01/2013 16:59

142

Clusters and economic growth in Asia

third, there is a high degree of user involvement in the innovation process (ibid.). One of the most systematic approaches to developing technological capability is the followers’ strategy for technological development. This emphasizes the need for human resources to allow an economy or a region to ‘shift’ from labour-intensive operations, such as those found in the early stages of the product life cycle, to more skilled-intensive activities at higher levels in the international division of labour (Sen, 1979). Japan was the first country to follow this strategy and it was later followed by South Korea and Taiwan. In the initial stage, the implementation of imported foreign technology and dependence on foreign experts prevail. The second stage, the assimilation of the technology, permits product diversification based on indigenous capabilities. Sometimes a local components industry develops, too. The third stage comprises the improvement of technology to enhance the competitiveness of both product and processes in international markets. Tied to this phase is the development of local scientific and engineering talent. The fourth stage emphasizes the development of an independent innovative capability. The stages in industrial development that generally correspond to the notion of technological learning are found in the sequences of industrial development in several Asian countries. Four tiers of industries correspond to successively higher capital/labour ratios and higher levels of technological sophistication (UNCTAD, 1995): ● ● ●



tier 1: labour-intensive light industries (toys, clothing, footwear, sporting goods); tier 2: scale-intensive heavy and chemical industries (steel, metals, fertilizers, basic chemicals); tier 3: assembly-based industries where product differentiation and both scale and scope economies dominate (motor vehicles, televisions and other consumer durables); tier 4: innovation-intensive ‘Schumpeterian’ industries in which R&D and close customer interaction are key inputs (aircraft, computers, pharmaceuticals).

The unrelenting pace of technology change and fierce capitalist competition pose great dilemmas for those who have ambitions to enter global markets. While some industries and products are somewhat sheltered from ‘continuous products innovation’ (Storper, 1992), no sector is immune from the technology-derived standards of quality and price that are set by world-class firms (Malecki, 1997). Advanced R&D is generally seen as activities dominated by advanced

ERIKSSON 9780857930088 PRINT.indd 142

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

143

industrialized nations. Catching up and overtaking established technological leaders poses formidable problems for imitators and aspirants of leadership, since they must aim at a moving target. It is no use simply importing today’s technology from leading countries, for by the time it has been introduced and assimilated the leaders have moved on (Freeman, 1988, p. 73). The aircraft industry is characterized by complex high value-added products in relatively small quantities, produced by relatively few players. Products have long development periods, that is, extremely long product life cycles and high development costs. The industry is also characterized by volatile markets with orders affected by a variety of financial and political factors. These high technology requirements necessitate a high level of R&D. In no other industry is there more of an interdependence and cross-fertilization of advanced technology than in the aerospace sector. Consequently, most of the world’s large aerospace companies are located in advanced economies, but to an increasing degree in developing and newly industrializing economies as well (Eriksson, 1995; Vértesy, 2011). The extremely high developments costs and high technology requirements of advanced aircraft has also forced leading Western companies to cooperate in alliances and partnerships, that is, risk/revenue sharing (Eriksson, 2000). Developing countries entering this kind of industry have an extremely demanding task of trying to compete with the leading companies in the international arena because of a lack of technical and technological competence, industrial infrastructure as well as financial resources. The technology used in modern aircraft is extremely demanding because of the high levels of functional performance, reliability, safety and efficiency required at a system level. Much of the expenditure on developing a new aeroplane is spent on integrating numerous technologies and systems with origins from various fields and industries such as metallurgy, composites, electronics, hydraulics and petroleum. These extremely high technological requirements, rising development costs and too many system integrators have, in recent decades, escalated merger activities and a weeding out of companies in the aircraft industry sector (Eriksson, 2010). For economic, technological, political and prestige reasons, many developing countries have tried to build up an internationally competitive aircraft industry (Eriksson, 1995), but very few have succeeded (Eriksson, 2006). One exception is the Brazilian manufacturer Embraer, which has developed into one of the world’s main commercial aircraft manufacturers. From being a main producer of turboprop aircraft, it has now developed into one of the leading manufacturers of regional jets in the 50- to 110-passenger capacity range. Most other developing economies have not

ERIKSSON 9780857930088 PRINT.indd 143

29/01/2013 16:59

144

Clusters and economic growth in Asia

yet been able to develop into successful fully-fledged aircraft producers because of the very complex and demanding nature of this industry. Indonesia started in the 1970s with heavy investment in the commercial aircraft industry. During Suharto’s rule, the establishment of IPTN was the largest and most ambitious investment by the Indonesian government to promote technology development in the country. Despite huge investments in engineering and production facilities, Indonesia’s emerging aircraft industry really never ‘took off’ and it has had very limited commercial success. The overall objective of this chapter is thus to investigate why Indonesia’s aircraft manufacturing industry has had difficulties succeeding in the long term and to identify the main factors behind this failure.

AN OVERVIEW OF INDONESIA’S ECONOMIC POLICIES Following independence, Indonesia’s economic policy was shaped by a strong sense of nationalism flavoured with anti-colonial sentiment. Indonesia’s second president, Suharto, introduced the ‘New Order’, based on a series of five-year plans designed to support the strong development of industrial projects and large-scale as well as smaller consumer- and exportoriented processing industries. Indonesia’s principal mineral resources are petroleum and natural gas, which have been the most important export earners. A drop in world oil prices led the government to postpone several large-scale projects in the middle of the 1980s and it has since tried to encourage the growth of non-petroleum sectors. Indonesia, OPEC’s only member in Southeast Asia, left this organization in January 2009. A number of market-friendly policies including deregulation and privatization allowed Indonesia to become one of the World Bank’s (1993) Highly Performing Asian Economies despite the preponderance of stateowned firms. From the late 1980s until 1997, foreign investment flowed into Indonesia, particularly into the rapidly developing export-oriented industrial sectors. Between 1993 and 1997, Indonesia’s per capita income placed it among the lower middle income economies, but the nation was severely hit by the Asian financial crisis in 1997–98. This led to the collapse of the rupiah against major currencies, rampant price inflation and a double-digit drop in real economic activity, along with increased unemployment and social unrest. Suharto was forced from power in 1998 and replaced by Dr B.J. Habibie, the former State Minister for Research and Technology, who also was the founder of IPTN. He was the third shortest-serving president of Indonesia

ERIKSSON 9780857930088 PRINT.indd 144

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

145

(1998–99). During Suharto’s rule, the Indonesian government was characterized by two distinct and competing groups of economic advisers: the technocrats and a diverse group of economic nationalists. While the technocrats were strongly committed to markets and competition, the nationalists had reservations about free market ideology and pressed for active government intervention in market behaviour as well as regulation. The most forceful exponent of the nationalist view was the architect of Indonesia’s high-technology strategy, namely Habibie. According to Habibie, Indonesia could never catch up economically with industrialized nations without a strong government-led push to support a self-sustaining high-tech manufacturing base. Believing that Indonesia’s present export of natural resources and textiles, clothing and footwear only had a limited lifespan, Habibie saw Indonesia’s future competitive advantage in value-added high technology and in the upgrading of human resources. He believed that Indonesia must focus on the ‘competitive advantages’ that only technology can provide rather than relying on its traditional and ‘comparative advantages’ of abundant land and labour (Smith, 1998). Habibie’s focus on high technology earned him plenty of criticism at home and abroad. The major arguments against his ideas concerned their limited effects on the economy as a whole. The technocrats questioned the economic validity of high-technology production in a labour surplus economy. There were also fears about the establishment of economic and technological enclaves with very limited links to the rest of society. Other critics argued that Habibie’s approach was costly for the economy, implying that the high-technology strategy drained money that could have been used for more productive purposes (Kompas, 4 March 1993). The majority of Indonesian state-owned industries operate in government-protected markets. The performance of these industries has been low, which is particularly alarming since many of them are labelled as strategic industries (Soedarsono et al., 1998). The return on assets of stateowned industries under the Agency of Strategic Industry Management was only 1.7 per cent (SWA Sembada, 1993). In the Indonesian periodical Warta Ekonomi (1994), a study showed that these strategic industries had a low performance compared with other state-owned industries. This general poor performance has continued to characterize stateowned industries. The Ministry of State-owned Enterprises (2006) reported that only 74 of 158 such enterprises in 2004 generated a profit and were able to provide a dividend. Between 1992 and 2004, return on assets averaged only 2 per cent and return of equity 8 per cent, implying the poor management of government-owned companies (Djajanto and Rosdaniah, 2006).

ERIKSSON 9780857930088 PRINT.indd 145

29/01/2013 16:59

146

Clusters and economic growth in Asia

Indonesia has a tradition of mega-projects bordering on the grandiose. These have often enjoyed considerable political and community support, in part because they are seen as a convincing illustration of the country’s capacity to overcome its technological backwardness and short-circuit the long and arduous process of technological development (Hill, 1998). Stewart and Nihei (1987, p. 151) discuss the general Indonesian situation by stating that ‘the top three priorities for expanding absorptive capacity are management, management, and management’. Indonesia’s manufacturing capacity, including the supply and quality of skilled labour, access to skilled management and existence of local suppliers, was the lowest among a number of Asian countries studied by Roessner et al. (1996). Since the late 1970s, ten industries have been targeted as strategic industries, among them aerospace, car assembly, shipbuilding, railroads, telecommunications, electronics, steel and machine goods.

ESTABLISHMENT OF IPTN The city of Bandung, about 200  km southeast of Jakarta, is the capital of Indonesia’s emerging aircraft industry. In Bandung lie the headquarters and facilities of PT Dirgantara/Indonesian Aerospace (IAe). The company was established in 1976 under the name PT Pesawat Terbang Nurtanio (Nurtanio Aircraft Industry Ltd) but changed its name in 1985 to Industri Pesawat Terbang Nusantara, or IPTN. In August 2000, its name was once again changed to PT Industri Dirgantara for domestic use and Indonesian Aerospace (IAe) as its global identity. The name IPTN is used in this study. The company is one of the indigenous aerospace companies in Asia with core competence in aircraft design and the development and manufacturing of commuter aircraft. The creation of IPTN was, to a large extent, the work of Habibie. For about ten years from 1965, Habibie worked for Hamburger Flugzeugbau (HFB) and Messerschmitt-Bölkow-Blohm (MBB). His last position there was as Vice-President and Director of the application of technology, with some 7000 employees under his direction. It is obvious that Habibie, many years before the creation of IPTN, had started thinking of creating an aircraft industry in Indonesia. In the 1960s he had already started to invite Indonesian engineers to work with him at HFB. When Habibie went to Indonesia for a sales mission in 1969, he arranged an informal meeting at Hotel Indonesia in Jakarta to discuss the creation of an aircraft industry. Around 20 engineers, all graduates of European schools, came to this meeting. Four of these joined

ERIKSSON 9780857930088 PRINT.indd 146

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

147

Habibie at HFB. Even before the arrival of engineers from Indonesia at HFB, Habibie had recruited several Indonesians living in Europe. Through Habibie’s initiative, HFB/MBB at one time employed some 30 Indonesians (Simanjuntak, 1988). Habibie and a small team returned to Indonesia in 1974 and established the Advanced Technology and Aviation Technology Divisions in the state-owned oil company Pertamina, which provided them with a working environment to get started. The Pertamina divisions and stateowned Lembaga Industri Penerbangan Nurtanio merged in 1976 and the result was the Nurtanio Aircraft Industry. In April 1976, President Suharto issued a decree giving Habibie the managerial responsibility for the company, which was established in August 1976. He held his position at MBB until 1978. For two years, he commuted between Germany and Indonesia. In 1978, he was also appointed Minister of State for Research and Technology. The close personal relationship between Habibie and President Suharto was without doubt a favourable circumstance for starting this project. Habibie always had strong support from Suharto. In many ways, it was like a father/son relationship (personal communication with persons knowing Habibie). It seems that Habibie was given carte blanche by Suharto to do anything he wanted, short of starting a revolution. A newspaper report (Pura, 1985) gave an impression of the birth of the company. During a meeting between President Suharto and Habibie one evening in January 1974: Mr Habibie recalls that the president spoke of his long-range vision for Indonesia’s economy, focusing on his goal of achieving a ‘take-off’ into industry by the mid-1990s. Mr Suharto said he needed someone to advise him on getting and applying modern technology. I have decided that you are the man to do that. When Mr. Habibie suggested he start by using his expertise to build planes, Mr Suharto gave him his blessing.

The Company’s Long-term Goals Apart from developing and building aircraft, IPTN led Indonesia’s drive to broaden its technological base and expand its level of industrialization over the next few decades. The company thus played a role in national development (Figure 7.1). In an interview with Interavia (Davidson, 1981, p. 1236), Habibie said about the future of aviation in Indonesia: Look at the spread of Indonesia, nearly 14 000 islands extending over a distance equal to that between Paris and New York, and equal in area to the whole of Europe, and a population of 163 million which is dependent on air transport.

ERIKSSON 9780857930088 PRINT.indd 147

29/01/2013 16:59

148

Clusters and economic growth in Asia

1

International utilization of business • Increasing company assets • Financial profit • Employee prosperity

IPTN’s role

2

Utilization of the development of technology

Influencing the national development

Preparing the industrial era

• Mastering technology • Increasing national capability 3

Utilization on national economy • Creating added value • Making other industries grow (state owned company or private)

Source: Information from IPTN during a visit in August 1990, published in Eriksson (1995).

Figure 7.1

The role of IPTN in the national development (Habibie’s perspectives)

We are condemned to aviation – there is no other way. As we have to use aircraft for essential communications, why not build them on our own rather than buy them.

From an advertisement in Fortune, 3 September 1984: During the roll-out ceremonies for the CN-235, President Suharto gave five reasons why the aircraft industry is ‘absolutely important’ for his country: to further integrate and unify the archipelagic nation; for security and defence; to generate job opportunities; to develop new technologies; and, he concluded, ‘to increase the confidence of other countries and the world in our ability to apply modern technology’.

IPTN’s long-range goals were as follows: 1. 2. 3. 4. 5.

to become self-reliant in the design and manufacture of aircraft and aerospace products; to become competitive in the international market; to support national defence and security; to foster the development of other domestic technologies and industries; to establish R&D in advanced technologies and new products.

ERIKSSON 9780857930088 PRINT.indd 148

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

149

Indonesia’s high-technology programmes, especially within aerospace, have been the subject of considerable controversy over the years. Critics have questioned whether this kind of industry is an appropriate one for a country of Indonesia’s economic and technological stage of development. The investment in IPTN is the largest made in a single industrial project. According to Far Eastern Economic Review (11 June 1987), the estimated cumulative investment was about $900 million until mid-1987. A later figure was that government investment in IPTN had reached $1.6 billion (Flight International, 19–25 June 1996). The Influence of Habibie Indonesia’s technology projects were long decided and influenced by Habibie because of his position as State Minister for Research and Technology. His influence started to diminish after the IMF’s intervention in 1998 and after he lost the presidential elections in 1999 and was succeeded by Abdurrahman Wahid. Some features of Habibie’s approach stand out: One is the extremely personalized manner in which projects have been undertaken. With the exception of the N-2130 project, still in its infancy, all the initiatives have been directly under his control: no major decision can be taken without his approval; no credible financial performance statements have ever been released; none of the usual checks and balances (such as scrutiny by the Department of Finance) is present; and not even the most powerful ‘technocrat’ in the cabinet has been able to challenge Habibie’s direct access to the President. (Hill, 1998, p. 44)

Another feature of Habibie’s approach relates to its intellectual foundations. Rice (1998) considers that Habibie was one of the few members of Indonesian cabinets with a coherent and internally consistent approach to his portfolio. According to Rice, some of his basic tenets were intellectually respectable, including the promotion of human resources development, the role of S&T institutions, market failure in the operation of technology markets and a ‘stages’ approach to technological development. Habibie also shared some of the views expressed by the Harvard professor Michael Porter (1990). Habibie’s model of development rejected orthodox economic approaches with their emphasis on cost/benefit analysis and comparative advantage. Instead, he proposed a measure defined as national performance productivity, which corresponds to a ratio of value-added to the value of intermediate inputs used in the production process. Habibie’s approach largely depends on guiding principles such as a strong education base and

ERIKSSON 9780857930088 PRINT.indd 149

29/01/2013 16:59

150

Clusters and economic growth in Asia

the development of national R&D capacities as a tool to minimize reliance on imported technologies. An important part in this ‘national strategy’ was creating ‘vehicle industries’ as a way to develop ‘hands-on’ scientific competence. Habibie’s model implicitly operated within a rather closed economy in the sense that it aimed for self-sufficiency rather than an interactive capacity in increasingly competitive technology markets. It is obvious that the main goal is technology itself and the development of Indonesia, not the business that has been the priority. This can be exemplified by an interview (Elson, 1983, p. 15) in Aviation Week & Space Technology: ‘Future programs will be selected largely for their ability to meet domestic needs and further the nation’s industrialization and technology transfer objectives’. In another interview (Bailey, 1992, p. 51), Habibie said: The approach to making an aircraft industry in Indonesia is not new. There was a minister for aircraft industry under Soekarno, but none of them ever succeeded. The reason is very simple: no real preparation and background. It’s not a matter of decision, it’s not a matter of capital, it’s a matter of know-how.

A danger with this is that it is easy to forget that managerial capability and managing the whole project is an indispensable ingredient in technological development: The neglect of managerial know-how results in part from a bias at work in the way policy makers in most countries (not only Indonesia) think about ‘innovation’, ‘technology transfer’, and ‘technological development’. Most are preoccupied with the tangible indicators of technological advance – the number of scientists and engineers, licensing agreements, new industries established – rather than coupling them with considerations about what it takes to be commercially viable, even in the long run. As a consequence, the ‘softer’ and less glamorous managerial skills associated with coordination, marketing, after-sales service, personal management, pricing, scheduling and inventory control are neglected. (McKendrick, 1992, p. 65)

During a visit to IPTN’s facilities in August 1993, the author was told that Habibie could decide on his own what kinds of aircraft were to be bought by various Indonesian airlines. Additional policy tools to assist IPTN’s development included an import ban on competing aircraft, exemption from government policies directing state enterprises to buy domestic inputs and considerable discretionary authority granted to Habibie (McKendrick, 1992, p. 42).

ERIKSSON 9780857930088 PRINT.indd 150

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

151

Implementation of Phase 1 To fulfil the goals set up and to develop skills in aircraft production, emerging countries are dependent on technology transfer from foreign aircraft manufacturing companies. There are many ways to carry out a technology transfer programme. The development of an indigenous aircraft industry must be based on a gradual approach. In this way, it will be possible to establish different measures for learning and abilities, which, in the most successful cases, will lead to an ability to develop and manufacture aircraft indigenously. Instead of starting a technology learning process with a gradual approach, through component manufacture and different levels of maintenance competence, Indonesia directly went into the full assembly and manufacture of aircraft. In its efforts to establish an indigenous aircraft industry with a far-reaching ability, with the intentions to design, develop and manufacture whole aircraft, IPTN followed a four-phase scheme for the transfer of aircraft manufacture technology (Figure 7.2). Licence Programme (Phase 1) Initially, IPTN adopted the so-called Progressive Manufacturing Programme (PMP), which is characterized by two concepts: first that the transfer of technology concerning a certain type of aircraft made by IPTN under licence is not planned according to the duration of the licence agreement, but to the number of aircraft built, and second that the process by which IPTN manufactures the aircraft starts from the end and finishes at the beginning. The PMP acts in the following way: it starts with a finished aircraft from the manufacturer (as an example CASA) and then disassembles it to see how it works, reassembles and then flies it. The next aircraft from the manufacturer is taken in an unassembled condition and then put together. The assembly process is continued with subsequent aircraft with the incorporation of simple parts at first, then gradually increasing the complexity and use of locally produced components (Figure 7.3). The first licence agreements were the Spanish CASA 212 Aviocar, a STOL light transport aircraft, and the German MBB Bo 105 helicopter. Work on these two aircraft started in 1976. Later, Nurtanio/IPTN obtained a licence to build a number of Aérospatiale SA 330 Puma helicopters for the Indonesian market. The production of this functioned as a learning programme for the bigger, more modern and more complicated SA 332 Super Puma.

ERIKSSON 9780857930088 PRINT.indd 151

29/01/2013 16:59

152

ERIKSSON 9780857930088 PRINT.indd 152

29/01/2013 16:59

1) Design and production of CN-235 (IPTN-CASA Venture)

Phase 2 Technology Integration (Joint-venture programme)

1) Material development • Composite • Conventional metal • New metal 2) Development of the N-250

Phase 3 Development of New Technology (Development programme)

Implementation of technology at IPTN

1) Centre of Science and Technology 2) Laboratory for construction test 3) Laboratory for aerodynamics and vibration 4–12) More Laboratories and research facilities * Development of the transonic 130-seater, N-2130 aircraft

Phase 4 Implementation of Research & Development of future technology

The Programme Of The Aircraft Industry

Based on information from a visit to IPTN, August 1990, published in Eriksson (1995).

Figure 7.2

Source:

1) Assembly/Manufacture of NBO-105 (Cooperation with MBB, W.-Germ.) 2) Assembly/Manufacture of NSA-330 and NAS-332 (Cooperation with Aéropatiale, France) 3) Assembly/Manufacture of NBELL-412 (Cooperation with Bell Helicopter USA) 4) Assembly/Manufacture of NC-212 (Cooperation with CASA, Spain)

Phase 1 Utilization of Existing Technology (Licence Programme)

1) Politics 2) Economy & Technology

National Motivation For Aircraft Manufacturing

The aircraft industry as a tool for economic and industrial development 100

CASA IPTN

90 80 Work load (%)

153

70 60 50 40 30 20 10 0 1978

Source:

1981

1983

1987

Based on information from IPTN, August 1990, published in Eriksson (1995).

Figure 7.3

Increase of local produced components in the NC-212, PMP

The next agreement, signed in 1982, was for the production of an American helicopter, the Bell 412. In September 1995, the production of this helicopter was halted after a joint audit of IPTN’s Bandung plant by the US Federal Aviation Administration (FAA) and Indonesia’s Directorate General of Air Communications because IPTN had failed to conform to FAA regulations. The parties revealed that some Bell 412 design data were missing or outdated (Flight International, 6–12 December 1995, p. 30). The inspection also uncovered the misuse of manufacturing manuals and poor communication between IPTN and Bell Fort Worth (ibid.). In the middle of the 1980s, IPTN received permission to build the MBB/ Kawasaki BK-117 helicopter, but this programme was not successful and was terminated after only a few had been built. Joint Venture Programme (Phase 2) IPTN’s scheme was not arranged in such a way that one phase must be complete before the next step begins. The company entered the second phase when it was just in its third year of operation. This phase was the integration of existing technology through the realization of co-design and manufacturing programmes with CASA for the CN-235. In October 1979, the announcement came to develop the CN-235 commuter/transport aircraft in collaboration with CASA of Spain. This was aimed at the regional airliner market, although the CN-235 very much looked like a compromise between commercial and military needs. The

ERIKSSON 9780857930088 PRINT.indd 153

29/01/2013 16:59

154

Clusters and economic growth in Asia

reason for this was that the design had a rear ventral door and sponsonmounted main undercarriage and shoulder-mounted wings. The aircraft was also planned to carry four LD-3 containers or a pair of 88-inch cargo pallets. A single management company, Air Technology Industries, was set up in Madrid in 1979, with Habibie as president. Design studies started in 1980 and prototype construction the year after. Simultaneous roll-outs were made in September 1983, and the first IPTN delivery was in 1986. Assembly lines were set up in Bandung and Spain. IPTN entered into a new kind of relationship with CASA, as the CN-235 was a new risk-sharing joint venture. This programme was critical to IPTN as success implied that the company could be looked upon as an internationally recognized aircraft manufacturer. From a comparative perspective, the company had in a relatively short period moved from kit assembly to be engaged, on a joint basis, in the design of a larger and much more advanced aircraft. The work-sharing agreement between the two companies was as follows: IPTN was responsible for the production of the main components of the tail unit (horizontal stabilizer, vertical stabilizer and rudder) and outer wing, outboard flap, aileron and door. CASA has responsibility for the production of the centre wing and power plant, inner flap, main and nose landing gear and nose fuselage. As for the production and assembly of the fuselage, the centre and rear fuselage was the responsibility of each of the companies. Until 1995, a major stumbling block for sales of the IPTN-built version of the CN-235 was its lack of an internationally recognized certificate of airworthiness. The company was finally granted Joint Aviation Rules Part 25 by the European Joint Aviation Authorities in 1995, but not before CASA had virtually cleaned up on the export front with its own FAAcertified CN-235. Until 2001, about 35 CN-235s were produced in Indonesia. During the next ten years, up to early 2011, 22 more were manufactured, implying a total of 57 aircraft. This is a very small number of aircraft produced from an international perspective and far from break even. Around 250 CN-235s were manufactured, most at CASA’s (now EADS) Spanish assembly line in Seville. Fifty aircraft were manufactured under licence by Turkish Aerospace Industries (TAI) under an agreement signed between CASA and TAI. These aircraft were produced and delivered to the Turkish Air Force between 1991 and 1998. One way to increase Indonesian orders for the CN-235 was to force a number of these aircraft onto the Indonesian Merpati Airlines, but to be able to carry through this task some political measures had to be taken. During the autumn of 1995, the Indonesian transportation minister,

ERIKSSON 9780857930088 PRINT.indd 154

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

155

Haryanto Dhanutirto, ordered the removal of Merpati Airlines President Ridwan Fatarudin, after he refused to lease 16 new CN-235-220 planes. The first eight aircraft were to be delivered in 1996/97 (Flight International, 1–7 November, 1995, p. 9). Fatarudin had refused to take the aircraft, complaining that the $110 000 a month being demanded by leasing company PT Arthasaka Nusaphala was too expensive for the financially struggling domestic carrier. Afterwards, the government denied any connection between Fatarudin’s sacking and the lease of the CN-235s. His opposition to the deal, however, was viewed as politically dangerous given the position of Habibie as the head of IPTN and Minister of Research and Technology. It is worth mentioning that the leasing company, PT Arthasaka Nusaphala, was also run by Humotomo Mandala Putra, son of the former President Suharto. A sales hurdle for the company has been its lack of export credit and thus it has instead resorted to barter trade, such as taking Proton cars from Malaysia in exchange for six CN-235s (Far Eastern Economic Review, 26 August 1993, p. 54; letter from IPTN 5 October 1994) or the deliveries of two CN-235s to the Thai Ministry of Agriculture and Cooperatives based on a counter-purchase agreement in which Indonesia purchased 110 000 tons of sticky rice from Thailand (IPTN’s home page 1 July 2000). The company has not been able to export any version of the aircraft to the civil market on pure commercial grounds. Development of New Technology (Phase 3) This phase was the application of the acquired technology for the indigenous design and manufacture of entirely new products. It came into being with the announcement, in June 1989, to launch the fly-by-wire 50–54seater N-250 regional turbo-prop. This was IPTN’s first indigenously developed aircraft. In June 1993, Habibie abandoned the 50-seater design in favour of a stretched version, the 64–68-seater N-270, redesignated N-250-100. Apart from a fly-by-wire flight control system, full-authority digital engine control, composite propellers, engine-indication and crewalerting system the aircraft was equipped with other state-of-the-art technology such as a Doppler turbulence weather radar and a collisionavoidance system. The roll-out took place in November 1994 in Bandung, with flight testing starting in 1995. One way to finance the development of the N-250 was to divert $185 million from the country’s reforestation budget (Flight International, 13–19 July 1994, p. 20). This sum was then equivalent to almost half of the funds spent on the project (Borsuk, 1995).

ERIKSSON 9780857930088 PRINT.indd 155

29/01/2013 16:59

156

Clusters and economic growth in Asia

From a very early phase of the project, Habibie had already started to talk about establishing overseas production, that is, a second production line, for the N-250. This would be one way to stimulate sales in the large US market. Although the US is the world’s dominating aircraft manufacturing country, it lacked a manufacturer in the 64–68-seater range: IPTN chairman Dr. Bacharrudin Habibie is to visit the USA in May to look at potential sites in three states – Arkansas, Arizona, and Utah. He says that a US operation would be necessary to supply the North American market. Habibie says that overseas production of the N-250 is part of IPTN’s long-term strategic vision, stretching over 15 to 20 years, and would be in addition to domestic production for the ‘captive market’ of up to 400 N-250 sales which are forecast within Indonesia. (Flight International, 2–8 March 1994, p. 16)

In 1995, IPTN was to decide on the location of its probable US plant. The alternatives were the former Pacific Aircraft Maintenance site at Portland International Airport as well as sites outside Mobile in Alabama and Phoenix, Arizona (letter from IPTN). Later that year, the company was proceeding with plans to assemble the N-250 in Mobile, Alabama, from 1997 onwards. IPTN intended to hold a 40 per cent stake in American Regional Aircraft Industries and aimed to attract local investors to take the remaining 60 per cent stake in the $100 million joint venture (Flight International, 13–19 December 1995, p. 11). At the Asian Aerospace exhibition at Changi, Singapore, in 1990, Habibie told the press that IPTN had a letter of intent with the Swedish aircraft leasing firm FFV for the latter’s purchase of 24 N-250s (Press Conference, 15 February, 1990). In fact, FFV was a large maintenance and repair company and had not been involved in aircraft leasing, although there were plans to start such activities. Thus, this information was a pure lie in front of the press. IPTN also released false and fictitious information on many occasions, such as claiming that the N-250 had been launched with letters of intent from two of Indonesia’s domestic airlines, Merpati and Bouraq, for 127 aircraft (Flight International, 22–28 January 1992, p. 5). At the Asian Aerospace exhibition in February 1996, IPTN announced that the company had received 204 orders and options for the N-250 (Press Release from IPTN). This covered 100 units placed by Merpati, 16 units by Sempati, 64 units by Bouraq, ten units by Gulfstream from the USA and four units by Colombia, with the rest by European leasing companies. This huge number for Merpati was mentioned even though it lost 130 billion rupiah ($53.5 million) in 1996, on top of its losses of 133 billion rupiah in 1995. At the end of 1996, the airline was saddled with 600 billion rupiah of debts.

ERIKSSON 9780857930088 PRINT.indd 156

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

157

The second prototype of the N-250 (first certification prototype) was set to fly in May 1996, but as a result of component documents falling below FAA requirements, the maiden flight of this aircraft had to be delayed (Flight International, 9–15 October 1996, p. 10). The FAA refused to accept the aircraft until IPTN brought its vendor record system in line with international standards (ibid.). The maiden flight for the second prototype took place on 19 December 1996, thus eight months behind schedule (Flight International, 8–14 January 1997, p. 10). In early 1997, IPTN enlisted a team of European aerospace consultants to try to help secure Joint Airworthiness Authorities (JAA) type certification of this aircraft. Jakarta-based consultancy Bramadi Pratama recruited a group of former British Aerospace employees, ex-JAA officials and test pilots to assist IPTN and the Indonesian Directorate General of Air Communications with certification (Flight International, 19–25 March 1997, p. 17). In 1998, during the Asian financial crisis, flight testing was frozen because of financial restrictions as the development programme ran into trouble when the IMF blocked further government support for IPTN: When the managing director of the International Monetary Fund, Michel Camdessus, told a press conference in Jakarta in mid-January that one of the conditions of the $43 billion IMF-rescue of Indonesia was the ending of government subsidies to Industri Pesawat Terbang Nusantara (IPTN) there was a ripple of applause among the journalists. For the Indonesian establishment – the family, cronies and courtiers of President Suharto – IPTN is a gleaming symbol of the Republic’s prestige. For the intellectuals and hard-nosed economists, who want to restore the economy of the potentially rich nation, IPTN is a symbol of all that went wrong. (Interavia, February 1998, p. 32)

As this project more or less came into a standstill, IPTN contacted a number of foreign companies in Asia and Europe to find industrial partners for the N-250 programme. In 1999, there were discussions between the Indonesian government and China about this project (Flight International, 29 August–4 September, 2000, p. 16). IPTN was looking for an investor to supply $90 million to complete the certification of the aircraft. For a few more years, two N-250 prototypes were used in flight testing, but the whole project was then closed down. Large-scale R&D Programme (Phase 4) The last phase was in development for several years and thus overlapped the others. It can be described as the implementation and R&D of future technology. This phase included a plan for launching a transonic turbofan

ERIKSSON 9780857930088 PRINT.indd 157

29/01/2013 16:59

158

Clusters and economic growth in Asia

130-seater, the N-2130. In an interview, Habibie claimed that the ultimate aim was to become the ‘Toyota of Aerospace’ (Flight International, 22–28 January 1992, p. 4). The decision to start with the N-2130 was not taken as a result of successful sales in domestic or international aircraft markets. In fact, only a few CN-235s had been exported and IPTN had no orders for the CN-250 even though the company used all its efforts to give that impression. However, it was not surprising when the decision to develop the N-2130 was taken: At the first flight of the N-250 last August, Dr. B.J. Habibie, Minister of State for Research and Technology and Chairman, President and CEO of IPTN announced the Indonesian President’s approval for development of a jet to be known as N-2130. (Asian Airlines & Aerospace, January 1996, p. 36)

And during the LIMA ’95 aerospace exhibition at Langkawi, Habibie announced that he had received government approval to develop the new jet when the N-250 made its first flight (Aviation Week & Space Technology, 11 December 1995, p. 37). IPTN had intended building three different sizes of N-2130, seating 80, 100 and 130 passengers. Subsequent consultations with airlines, however, revealed little support for either an 80-seater or a five-abreast cabin crosssection (Flight International, 5–11 March 1997, p. 9). A part of Habibie’s strategy was to ensure that the family connection with the company continued: Indonesian Minister for Research and Technology, Mr Jusuf Habibie, was the proud father yesterday, as he talked about his 32-year-old son, the aircraft designer. The young Habibie, Ilham Akbar Habibie, is chief designer of the newest aircraft of Industri Pesawat Terbang Nusantara (IPTN), the 100-seater N2130. He has Bachelor’s and Master’s degrees and a Ph.D. in aircraft design from the Technical University of Munich, and he is executive vice-president for the N2130 programme, his father said at an IPTN briefing. He is chairman, director and chief executive officer of it. (Strait Times, 8 February, 1996)

Turning to business, Habibie said that the N2130, which was scheduled to enter the market in the year 2006, would be founded totally by Indonesia. There were no worries about demand he said. Indonesia, with a population of nearly 200 million, has a big domestic market (ibid.). A special company, PT Dua Satu Tiga Puluh, was founded to function as IPTN’s fund-raising agency for the N-2130. The project continued for some years, including wind-tunnel testing and some work on a prototype, but in late 1999 it was officially dismissed and the project closed down.

ERIKSSON 9780857930088 PRINT.indd 158

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

159

Subcontracting and Other Activities Besides the main objective to become a major international aircraft producer, IPTN also had a goal to develop as an international supplier of aircraft components and parts. Throughout the years, the company has carried out subcontracting work for international aircraft companies such as Boeing (B737 leading edge flap, B767 flaps), Fokker (speed brakes, lift dumpers, wing to fuselage fairings, pedestal frames), General Dynamics/ Lockheed (F16: forward engine access door, wing flapperon, fuel pylon, weapon pylon, main landing gear door, graphite epoxy skin vertical fin) and Mitsubishi (B767 keel beams) (Eriksson, 1995, pp. 167–8). After 1995, the company also started to produce components for the Airbus consortium. A main business has been the continued production of airframe sections for the Spanish-built CN-235. According to the company (information received during a visit in 2001), it had delivered 159 ship-sets of CN-235 components to CASA, worth $ 77 million. In 2012, IPTN carried out international subcontracting as follows: ● ● ● ●

Airbus 380 (various components); Airbus 320/321 (leading edge skin, pylons and D-nose skin); A340/A350 (few components); Boeing B747-8 (seal retainers).

Eurocopter chose the Indonesian company as its main supplier for the latest version of the Super Puma family of helicopters (EC725/EC225). This includes main airframe parts, including 125 sets of the tail boom and fuselage. Deliveries will take place between 2011 and 2020. In the 1980s, IPTN signed an agreement with the US engine manufacturer General Electric for assistance to establish a maintenance centre for repairing and overhauling aero-engines. The company also had ambitions to become a producer of aero-engine parts, but these hopes never came to fruition. The company is also engaged in various maintenance, repair and overhaul services, both for its own produced aircraft and for different kinds of aircraft for various domestic and international operators. One way to create additional revenues is to carry out various kinds of engineering services to non-aeronautical companies. Between August 1990 and September 2001, a large number of discussions were carried out with staff at IPTN, researchers at the Institut Teknologi Bandung, local administrators and business people about the possible spillover effects from IPTN/IAe activities. The overall impression was that local effects were very limited. This is supported by Sutanto

ERIKSSON 9780857930088 PRINT.indd 159

29/01/2013 16:59

160

Clusters and economic growth in Asia

(1996), which states that IPTN neglected to ‘breed’ advanced suppliers in local and regional levels. Material imports accounted for 97 per cent of the entire value of its inputs in 1994 (ibid.). Instead, local private technology firms were forced to contribute financially to aeroplane development activities (Fromhold-Eisebith and Eisebith, 2002). In general, Bandung lacks qualitative dynamics and it shows few signs of technological learning or ‘catching up’ (ibid.)

CONCLUSION The creation of IPTN is an example of how one influential person, in conjunction with the state, can play an important role in the establishment and location of high-tech aircraft production in a developing country. This gigantic project, the largest and most ambitious investment made by the Indonesian government to promote technology development in the country, was largely the work of Habibie, although with strong and direct support by the former President Suharto. In the early phase of the development process, the project was justified by its role of spearhead, leading Indonesia’s drive to broaden its technological base and expand its level of industrialization over the next few decades. Without doubt, one important reason behind the establishment of IPTN, although never put into words, was national prestige. From the very beginning, a strategy for the development of technology transfer and manufacturing was initiated through the PMP. It seems that phase 1 of the PMP worked rather well as a learning process and as a way to establish appropriate technology at an early stage of the development. A more complicated task was phase 2, namely the transfer and build-up of aircraft technology. The company entered this phase when it was just in its third year of operation. This was the CN-235 joint venture programme with CASA, which from an Indonesian perspective was unsuccessful because of the limited number of aircraft manufactured in Bandung. A major hurdle was acquiring an internationally recognized certificate of airworthiness. The Indonesian manufacturer had big problems entering the international market. In a few cases they succeeded with the help of barter trade. However, the company lacked experience in sales and marketing such advanced products. It seems that the business side of the coin was never carefully prepared. An underlying thought was perhaps that the large domestic market, and an extrapolation of the long-term growth rates of the Indonesian economy, supported with government orders, were sufficient to support the Indonesian aerospace industry.

ERIKSSON 9780857930088 PRINT.indd 160

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

161

A main factor for the long-term failure is the way to look at the company’s development, focusing only on technology itself, with very little thought on the less glamorous business, management and marketing side. An overall lack of management has been a major hurdle, because managerial competence has to interact with technological competence. Another obvious conclusion is that after the second phase of technology development, the company decided to go ahead with phase 3 without enough orders and with problems with certification. The decision to start the third stage seemed to be taken only for reasons of national and personal prestige. IPTN had no real customers despite claims of more than 200 aircraft on order and options. Altogether, 180 of these were orders and options from domestic carriers with no real possibility to receive, or finance, such a large number of aircraft. The first two steps were built on technology transfer from established foreign companies. During the second phase, the company could rely on cooperation with CASA, thus sharing costs and technology support from one of Europe’s main aerospace companies. When moving into the third phase, the company decided to go alone, which must have been rather complicated. This strategy is in contrast to the rest of the international aerospace industry, which has increased co-financing, cooperation and work sharing in the development and manufacturing of aircraft. For instance, the company decided to go alone on the N-250 but not until the projects entered a technological and financial crisis did they try to get foreign investors into the project. In a statement by Habibie in 1992 about the creation of an aircraft industry in Indonesia, he says, ‘It’s not a matter of capital, it’s matter of know-how’. In fact, as shown in this case, capital is a very important matter, along with many other elements if successful development should take place. Owing to the circumstances, the fourth phase of technology development was never entered into. One obvious failure, from a long-term perspective, is that IPTN failed to receive any ‘launch orders’ from established airlines abroad to secure production and valuable business practices. Although the company has focused on technology, it has not had the resources to be a technology leader and it has had the demanding task of trying to compete with leading companies in the international arena, that is, to aim at moving targets. Despite its know-how and impressive technology resources, the company is struggling to right itself as state funding has been forced away and it still faces the challenge of defining its own raison d’être, apart from the political motivation of national prestige. The economic and political crises of the late 1990s worsened the situation for IPTN/

ERIKSSON 9780857930088 PRINT.indd 161

29/01/2013 16:59

162

Clusters and economic growth in Asia

IAe. Since then, the company has struggled to stay afloat mainly by keeping its lines from licence production and international subcontracting work going. What can be learnt from the Indonesia case? The development of an international competitive aircraft industry is much more complicated than anticipated and the creation of this company seems more of a ‘technonationalist dream’ than one based on realistic economic and science-based foundations and theories. The experiences from Indonesia are a failing argument to create this kind of industry for the purpose of economic and industrial development. The country cannot support this kind of industry. The general level of the technological, human knowledge and R&D resources needed for the advanced aircraft industry was not met because of a lack of business practices. A weak industrial base meant problems providing components and other services. A huge supporting ‘infrastructure’ of suppliers, service firms, and strong universities is needed as well as large and solid financial resources. The entry barriers in this industry are extremely high and they have increased in the past few decades. Furthermore, it seems that IPTN/ Indonesian Aerospace has become an ‘industrial enclave’ with little, if any, connections to the rest of the economy.

REFERENCES Asian Airlines & Aerospace (1996), January. Aviation Week & Space Technology (1995), 11 December. Bailey, J. (1992), ‘Habibie’s Grand Design’, Flight International, 19–25 February, 51–55. Borsuk, R. (1995), ‘Suharto Pushes Plan to Build Jet in Indonesia’, Asian Wall Street Journal, 11–12 August. Davidson, B. (1981), ‘PT Nurtanio: The Rising Star of the Orient’, Interavia No. 12, 1236–9. Djajanto, P. and Rosdaniah, S. (2006), ‘The State-owned Enterprises Reform in Indonesia,’ paper presented at a seminar on ‘Improving Governance of Public Sector Enterprise’, Asian Development Bank Institute, Tokyo, 4–8 September. Elson, B.M. (1983), ‘Nurtanio Leads Industrial Modernization’, Aviation Week & Space Technology, 26 December. Eriksson, S. (1995), Global Shift in the Aircraft Industry: A Study of Airframe Manufacturing with Special Reference to the Asian NIEs, publication edited by  the Departments of Geography, University of Gothenburg, Series B, No. 86. Eriksson, S. (2000), ‘Technology Spillover from the Aircraft Industry: The Case of Volvo Aero’, Technovation, 20(12) 653–64. Eriksson, S. (2006), ‘Cluster Creation and Innovation Within an Emerging

ERIKSSON 9780857930088 PRINT.indd 162

29/01/2013 16:59

The aircraft industry as a tool for economic and industrial development

163

Taiwanese High-tech Sector’, International Journal of Technology Transfer and Commercialisation, 5(3), 208–36. Eriksson, S. (2010), ‘China’s Aircraft Industry: Collaboration and Technology Transfer – The Case of Airbus’, International Journal of Technology Transfer and Commercialisation, 9(4), 306–25. Far Eastern Economic Review (1987), 11 June. Far Eastern Economic Review (1993), 26 August. Flight International (1992), 22–28 January. Flight International (1994), 2–8 March. Flight International (1994), 13–19 July. Flight International (1995), 1–7 November. Flight International (1995), 6–12 December. Flight International (1995), 13–19 December. Flight International (1996), 19–25 June. Flight International (1996), 9–15 October. Flight International (1997), 8–14 January. Flight International (1997), 5–11 March. Flight International (1997), 19–25 March. Flight International (2000), 29 August–4 September. Fortune (1984), 3 September. Freeman, C. (1988), ‘Technology Gaps, International Trade and the Problems of Smaller and Less Developed Economies’, in Christopher Freeman and BengtAke Lundvall (eds), Small Countries Facing the Technological Revolution, London: Pinter Publishers, pp. 67–84. Fromhold-Eisebith, M. and G. Eisebith (2002), ‘The Indonesian Technology Region of Bandung’, International Development Planning Review, 24(1), 41–57. Hill, H. (1998), ‘Introduction’, in Hal Hill and Thee Kian Wie (eds), Indonesia’s Technological Challenge, Singapore and Canberra: Institute of Southeast Asian Studies and Australian National University, pp. 1–52. Interavia (1998), February. Kompas (1993), 4 March. Malecki, E.J. (1997), Technology & Economic Development: The Dynamics of Local, Regional and National Competitiveness (2nd edition), London: Longman. McKendrick, D. (1992), ‘Obstacles to “catch-up”: The Case of the Indonesian Aircraft Industry’, Bulletin of Indonesian Economic Studies, 28(1), 39–66. Miller, R., M. Hobday, T. Leroux-Demers and X. Olleros (1995), ‘Innovation in Complex Systems Industries: The Case of Flight Simulation’, Industrial and Corporate Change, 4(2), 363–400. Ministry of State-owned Enterprises (2006), A Ministerial Report 2006, Jakarta. Nelson, R.R. and N. Rosenberg (1993), ‘Technical Innovation and National Systems’, in Richard R. Nelson (ed.), National Innovation Systems: A Comparative Analysis, New York: Oxford University Press, pp. 3–22. Pavitt, K. (1990), ‘What We Know about the Strategic Management of Technology’, California Management Review, 32 (3), 17–26. Porter, M. (1990), The Competitive Advantage of Nations, London: Macmillan. Pura, R. (1985), ‘Indonesia Pursues Costly Aircraft Plan’, Asian Wall Street Journal, 2 September. Rice, R.C. (1998), ‘The Habibie Approach to Science, Technology and National Development’, in Hal Hill and Thee Kian Wie (eds), Indonesia’s Technological

ERIKSSON 9780857930088 PRINT.indd 163

29/01/2013 16:59

164

Clusters and economic growth in Asia

Challenge, Canberra and Singapore: Australian National University and Institute of Southeast Asian Studies, pp. 185–98. Roessner, J.D., A.L. Porter, N. Newman and D. Cauffiel (1996), ‘Anticipating the Future High-tech Competitiveness of Nations: Indicators for Twenty-eight Countries’, Technological Forecasting and Social Change, 51(2), 133–49. Sen, A. (1979), ‘Followers’ Strategy for Technological Development’, The Developing Economies, 17(4), 506–28. Simanjuntak, P. (1988), IPTN: Carrying Indonesia Toward High Technology, Jakarta: CV Panggal. Smith, S.L. (1998), ‘Batam Island and Indonesia’s High-technology Strategy’, in Hal Hill and Thee Kian Wie (eds), Indonesia’s Technological Challenge, Singapore and Canberra: Institute of Southeast Asian Studies and Australian National University, pp. 342–63. Soedarsono, A.A., S.L. Murray and Y. Omurtag (1998), ‘Productivity Improvement at a High-tech State-owned Industry – An Indonesian Case Study of Employee Motivation’, IEEE Transactions of Engineering Management, 45 (4), 388–95. Stewart, C.T. and Y. Nihei (1987), Technology Transfer and Human Factors, Lexington, MA: Lexington Books. Storper, M. (1992), ‘The Limits to Globalization: Technology Districts and International Trade’, Economic Geography, 68(1), 60–93. Strait Times (1996), 8 February. Sutanto, A. (1996), ‘Penilaian Dempak Ekonomi PT: IPTN Terhadap Perekonomian Jawa Barat dan Nasional Dengan Menggunakan Metoda InputOutput’ (unpublished thesis), Bandung, Department of Regional and City Planning, Institut Teknologi Bandung. SWA Sembada (1993), ‘Rottens gnaw state-owned companies’, 6(9), September. UNCTAD (1995), World Investment Report 1995: Transnational Corporations and Competitiveness, New York: United Nations. Vértesy, D. (2011), ‘Interrupted Innovation: Emerging Economies in the Structure of the Global Aerospace Industry’, Maastricht: University of Maastricht (Doctoral dissertation). Warta Ekonomi (1994), VI/25(9)9, 36–44. World Bank (1993), The East Asian Miracle: Economic Growth and Public Policy, New York: Oxford University Press.

ERIKSSON 9780857930088 PRINT.indd 164

29/01/2013 16:59

8.

Foreign knowledge transfer in the development of aircraft industry clusters – the case of Chengdu, China Sören Eriksson

INTRODUCTION The People’s Republic of China came into being in 1949 and after nearly three decades of self-reliance, Deng Xiaoping emerged as leader following Mao’s death in 1976. It was under Deng’s leadership that China began to jettison the self-reliance policy of the previous 30 years and to make links with the world economy. The pivotal year was 1979, when China began its ‘open-door policy’ based upon a carefully controlled trade and inward investment strategy. This was set within the so-called ‘four modernizations’ focusing on agriculture, industry, education and science/defence. Since then, China has gone through three more decades of enormous economic and industrial development. In many industries, especially those that are labour-intensive, China is the dominant global producer force in low-tech manufacturing, but it is actively moving into areas where technology plays an important role and where labour is not the dominant cost factor. China’s impressive industrial development during the past 30 years has been dependent on the foreign transfer of technology, management skills and other kinds of knowledge. From an adoption perspective, that is, absorb, generate and disperse technological competence, two main types of theories can be identified (Nelson and Pack, 1999): neoclassical ‘accumulation theories’ that focus on the role of physical and capital investments and ‘assimilation’ theories that use more evolutionary views and explicitly stress learning in adapting and operating foreign technologies. Learning is characterized by externalities, spillovers and exchanges of information and skills by persons within a firm, organization or other networks (Lall, 2000).

165

ERIKSSON 9780857930088 PRINT.indd 165

29/01/2013 16:59

166

Clusters and economic growth in Asia

In the 1980s and 1990s, the Chinese government had already stated that commercial aircraft production would be an important industry in the new stages of economic and industrial growth (Eriksson, 1995). The government also expressed its interest and ambition to develop aircraft-related clusters based on ‘old’ locations as well as to create new ones. In general, most clusters are the outcome of historical processes of cumulative path-dependent growth processes, while some are creations of regional or national polices in which governments have taken a prominent role in promoting economic growth and the development of industries. However, although most of the cluster literature maintains the importance of local knowledge and spillovers as key drivers of cluster development, globalization processes and the international dispersion of industrial activities has led to the need to focus more on the role and importance of external linkages (Breschi and Lissoni, 2001). Bathelt et al. (2004) question the view that tacit knowledge transfer is confined to local milieus, whereas codified knowledge may roam the globe almost frictionless. They highlight the conditions under which both tacit and codified knowledge can be exchanged locally and globally. This implies that successful clusters are those that are able to build and maintain a variety of channels, labelled ‘global pipelines’, for the exchange of knowledge among various industries and locations around the world. However, the role of external linkages is not new. Sweeney (1987) states that the level of innovativeness in an area depends on the degree to which firms are linked to both local networks of suppliers and external global markets. The combination of the sophisticated needs of customers and technical expertise by suppliers leads to mutually supportive interactions (De Bresson, 1989). In general, there is a need to broaden the research agenda regarding clusters by looking into dimensions such as changes over time, technology transfer and supply chains (Eriksson, 2011b). Owing to their local and global connections, some local firms are able to absorb non-local knowledge and transmit it into firms or clusters (Owen-Smith and Powell, 2004). This seems to be the case in some hightechnology industries such as the biotechnology and aerospace industries (Biggiero and Sammarra, 2010). The structure seems to conceal globalization and regionalization through a small number of connections, that is, the local dimension is connected globally through bridge firms. Eriksson (2000) shows that within Volvo Aero, the Swedish aero-engine producer, the global supply chain acts as a source of spillover and innovation that creates direct and indirect technology diffusion internally as well as externally. The latter aim at networks with universities, international collaborative partners and so on. Niosi and Zhegu (2005) conclude that supply chain management is the vehicle of knowledge spillover in the

ERIKSSON 9780857930088 PRINT.indd 166

29/01/2013 16:59

Foreign knowledge transfer in the development of aircraft industry clusters 167

aerospace industry. This includes technology as well as management skills, working methods, logistics knowledge and compliance with international rules and regulations of airworthiness and safety (Eriksson, 2003, 2010), which also implies greater functional integration in development, design and manufacturing (Eriksson, 2011a). Another kind of knowledge transfer is the movement of educated scientists and specialists. When, for instance, the Taiwanese government, together with private companies, made large-scale efforts to develop a commercial aerospace industry during the 1990s, also implying the build-up of an aircraft industry cluster, one kind of knowledge transfer was brain-gain, that is, foreign-educated Taiwanese citizens returned home to take part in the development. Most of these individuals were educated in the USA and had academic or industry positions in the aerospace sector (Eriksson, 1995). In a follow-up study of this Taiwanese development (Eriksson, 2006), it was concluded that the development of a successful commercial aircraft industry cluster was much more difficult than expected for a number of reasons, including the following: ●





The aircraft industry is characterized by the manufacturing of extremely high-value products produced in relatively small quantities by relatively few players. Products in this industry have long development cycles, very high development costs and low company turnover. Taiwan’s industry did lack experience from the commercial aircraft industry and industries with this kind of characteristics. Taiwan’s high-tech industry has profound experience in the electronics and computer industries with many competitors, short product life cycles and rapid company turnover. Most Taiwanese firms are SMEs, while the large system integrators in the world aircraft market are firms such as Airbus and Boeing. The aerospace industry is extremely complex and one of the most demanding in terms of system integration. A large number of technological subsystems must work in an integrated way. This differs a lot from modularization and disintegration, which are used by Taiwanese companies as a way to lower the entry barriers for their semiconductor industry. Thus, complex product systems are large customized engineering goods of which most Taiwanese firms lack experience and knowledge.

Furthermore, Taiwan’s industrial capacity had limited innovation capability in this industry and lacked the industrial infrastructure needed to be able to create an internationally competitive commercial aircraft

ERIKSSON 9780857930088 PRINT.indd 167

29/01/2013 16:59

168

Clusters and economic growth in Asia

industry. It was also highly dependent on foreign technology transfer and knowledge. This chapter focuses on one of the major aircraft manufacturing locations in China, Chengdu in the Sichuan Province, which is one of the oldest aircraft manufacturing establishments in the country. The main research questions are: What are the main origins of foreign technological transfer linkages in the development of Chengdu’s aircraft industry ‘cluster’? What are the main characteristics of the foreign transfer of knowledge and technology? The next section gives an account of the background of the development of China’s aircraft industry, mainly with reference to historic and recent development processes. Then follows the empirical section, which focuses on Chengdu and the development of foreign technology transfer. The chapter finishes with conclusions.

CHINA’S AIRCRAFT INDUSTRY The Early Years Before the 1949 Communist Revolution there was almost no aircraft industry in China, except for the assembly of a few foreign aeroplanes from ready-made parts. After 1949, the Communist government decided to develop an aircraft industry, mainly for defence. As the West had imposed an economic embargo in the wake of the revolution, China imported foreign technology from the Soviet Union. Starting in 1953, agreements were signed to manufacture, on licence, a variety of mainly military types such as the MI-4 helicopter, An-2 utility biplane and Yak-18 trainer (Eriksson, 1995). At the end of the 1950s, relations between China and the Soviet Union began to deteriorate as ideological differences emerged. At the time of the Soviet’s abrupt withdrawal from China in 1960, China had just begun to manufacture on licence the next generation of Soviet aircraft, such as the MiG-19 fighter and Tu-16 bomber. The imports of components and raw material ceased, and a period of reorientation towards technical and industrial independence began. Owing to the increased tensions between China and the Soviet Union, but also to the concern that the government in Taiwan might persuade the USA to attack the mainland, Mao Zedong’s third-front strategy was implemented, implying a large-scale relocation of factories in its southwestern interior, where it would be strategically secure in the event of a war (China Today, 1989).

ERIKSSON 9780857930088 PRINT.indd 168

29/01/2013 16:59

Foreign knowledge transfer in the development of aircraft industry clusters 169

At the beginning of the 1960s China set out to design its own aircraft, necessarily based on available Soviet designs. This task, along with the development of transport aircraft, became difficult with the onset of the anti-technological Cultural Revolution in 1966. Only well-established aircraft programmes were able to continue through the ten years of turmoil. The main strategy was to reverse-engineer foreign-developed aircraft and give them Chinese designations (Eriksson, 1995). With the end of the Cultural Revolution and the emergence of the new leadership, China changed direction and adopted an open-door policy in 1979. During the time of the open-door policy, military aircraft production was cut back and the development of civil air transport emphasized supporting the expected growth of China’s economy, particularly its tourism sector. The seventh Five-Year Plan, which covered the period 1986–90, singled out improvements to transport, particularly air transport, as a national priority. This fact naturally stimulated the nation’s aircraft industry. The establishment of a satisfactory air transport system was, and still is, vital for the development of China’s economy, and apart from China’s own industry, it gave great opportunities for foreign producers of aircraft and equipment. China represented a great potential market with its large territory and enormous population (Eriksson, 1995). This can be exemplified with the following quotation: In the USA, the world’s most mature market, there were 1.7 airline trips per head of population in 1990. The comparable figure for China is just 0.001 trips per head. Boeing predicts that China will need US dollar 41 billion worth of new aircraft by 2010, which would make it the world’s second largest market. (Bailey, 1993)

One of the great problems for the Chinese industry, not only the aerospace industry, during the 1980s was the lack of the most sophisticated technology. Another important problem was the lack of management skills and methods. Until the late 1980s, China had manufactured thousands of aircraft, mainly military but also civil, most of them based on old Soviet designs.1 A special case was when, in 1970, the local government of Shanghai and the Ministry of the Aerospace Industry decided to launch the Y-10 commercial aircraft programme. This closely resembled the old US Boeing 707 and similarly was equipped with two Pratt & Whitney JT3D turbofans. Only a few prototypes were built and it is nowadays considered to be a pure reverse-engineered project that became a failure.

ERIKSSON 9780857930088 PRINT.indd 169

29/01/2013 16:59

170

Clusters and economic growth in Asia

The Aircraft Industry and Open-door Policy The open-door policy that started under Deng Xiaoping’s leadership in 1979 brought about changes in the aircraft industry. In April 1985, an agreement was signed between the Shanghai Aviation Industrial Group of China and McDonnell Douglas (USA) to start co-production of the MD-82 airliner. It was an offset agreement for the sale of McDonnell Aircraft to China and the assembly of 25 MD-82s, which started in 1986. An extension was granted for ten more MD-82s with a final delivery in 1994. This assembly project was the first modern airliner ever built in China and thus it became an important learning experience for the emerging Chinese aerospace industry (Eriksson, 1995, 2010). The gradually improving political relations with Western countries, not least with the USA, opened possibilities for China to access advanced Western technology. One important obstacle disappeared when CoCom2 restrictions on the sale of military technology to China were eased in 1985. From the early3 1980s until the mid-1990s, numerous agreements with Western companies were signed, allowing increased access to modern technology and manufacturing methods for the Chinese industry. From early on the Chinese government used the strategic tool of offset (Eriksson, 1995, 2010). When Chinese airlines ordered foreign-manufactured aircraft they demanded that some portion of production took place within China. In other words, they demanded technology transfer in exchange for market access. Later, this was complemented by the need to lower production costs by outsourcing the production of components and parts to China. In the early stages, the main jobs were ‘simple’ parts such as fairings and small doors, while later Chinese subcontractors increasingly became more involved in advanced components, systems, materials and technologies (Eriksson, 1995, 2010, 2011a). China’s aerospace industries were restructured in June 1993. The former Ministry of the Aerospace Industry was divided into two new profitmaking bodies. The aircraft production was then headed by the Aviation Industry Corporation of China (AVIC) whose aim was to reduce the complicated bureaucracy of the old body and to unify control of the aerospace industry. This structure was once again changed in the late 1990s. A new organization originated and AVIC was split into AVIC 1 and AVIC II – both under government control. AVIC 1 was established in July 1999 to develop and manufacture military and commercial aviation equipment. It focused on medium-sized and large aircraft (i.e., the bulk of China’s military aircraft production), while AVIC II gave priority to smaller aircraft and helicopters. In October 2008, AVIC I and AVIC II

ERIKSSON 9780857930088 PRINT.indd 170

29/01/2013 16:59

Foreign knowledge transfer in the development of aircraft industry clusters 171

merged because the previous separation resulted in a split of resources and led to redundant projects. After the merger, the new AVIC consisted of ten business units: aeroengines, avionics, defence, general aviation aircraft, helicopters, transport aircraft, aviation R&D, flight test, trade & logistics and asset management. The major focus of AVIC was to develop indigenous military technologies as well as commercial aircraft to compete in domestic and international markets. AVIC remains one of the largest state-owned enterprises managed by the central government and employs around 400 000 people. Commercial Aircraft Projects in the Twenty-first Century China is now establishing a modern commercial aircraft industry based on domestic efforts but with important reliance on agreements with foreign partners and foreign technology input. In 2000, the Commission of Science, Technology and Industry for National Defense acknowledged that China’s aircraft industry lacked the capability to develop and manufacture modern medium-sized and large aircraft. Owing to the large demand for new aircraft in the Chinese market, now and in the foreseeable future, decision-makers considered that the nation’s aviation sector would be incomplete without developing its own civil aircraft. There are now several ongoing projects, domestic as well as joint ventures with foreign partners. The ARJ21 regional aircraft was unveiled at the 2001 Beijing Air Show, representing China’s most comprehensive effort to build a modern indigenous aircraft, although with a number of foreign partners and suppliers and technical assistance from large US and European companies. The first flight took place on 28 November 2008, although entry into commercial service has been delayed a number of times (Eriksson, 2010). Although the Chinese advertising of the aircraft refers to the ARJ21 as a completely independent design with completely independent intellectual property rights, all its main technologies are Western-based such as avionics, engines and the fly-by-wire system. Indeed, more than 20 American and European contractors supply a large number of critical materials and technical systems and parts. Included among the foreign supplying companies are CFM International (France/USA), Eaton (USA), General Electric (USA), Honeywell (USA), Goodrich (USA), Hamilton Sundstrand (USA), Moog (USA), Parker Aerospace (USA), Rockwell Collins (USA), Liebherr Aerospace (Germany/France) and SAFRAN (France) (based on information received in a seminar at Asian Aerospace 2007 Congress, Hong Kong, 5 September 2007).

ERIKSSON 9780857930088 PRINT.indd 171

29/01/2013 16:59

172

Clusters and economic growth in Asia

In fact, the ARJ21 has the same cabin cross-section, nose profile and tail as the US MD-80/90 aircraft (earlier licensed and manufactured in Shanghai) that ceased production in 1999. Another technology input, not widely known outside the core project team, is that Ukraine’s Antonov supplied the ARJ21 with a new super-critical wing as well as integral analysis of the construction strength. It also performed additional wind tunnel testing (Antonov News, 2007). In May 2000, the Brazilian aircraft manufacturing company Embraer opened an office in Beijing, making it a base for increased cooperation and sales in China. In 2000, Embraer received the first Chinese orders from Sichuan Airlines (five ERJ-145 regional jets), followed by orders from China Southern Airlines (20 orders plus five options of the ERJ-145) and Wuhan Airlines (10 ERJ-145). The Brazilian president involved himself in the deal, which included discussions with the Chinese president Jiang Zemin, without being able to solve the situation (Aviation Daily, 18 April 2001). In March 2002, Embraer set up a spare parts centre in Beijing in order to maintain its growing fleet of aircraft. During the Asian Aerospace Exhibition in Singapore (February 2002), it was revealed that Embraer was planning to set up a final assembly plant in Harbin to produce Embraer aircraft for the Chinese market. In April 2002, the Chinese government increased import tax on foreign-produced aircraft from 5 per cent to 23 per cent, thus giving favourable conditions for domestically produced aircraft. This increase could also be seen as pressure from the Chinese government to produce these Brazilian aircraft in China instead of importing them from Brazil. Embraer and its Chinese partner hold a 51 per cent and a 49 per cent share, respectively. The agreement also includes a commitment towards technology transfer and management education. In June 2004, the first Chinamade Embraer delivery was made to China Southern Airlines. In 2006, the European aircraft consortium Airbus decided to build an aircraft assembly plant in Tianjin. The production site is a joint venture between Airbus and a Chinese consortium of the Tianjin Free Trade Zone, AVIC 1 and AVIC 2. It is the first Airbus final assembly plant outside Europe, and it was a strategic decision to strengthen Airbus’s position in China relative to its main competitor Boeing (Eriksson, 2010) in order to be able to sell more aircraft to Chinese airlines. In March 2007, AVIC 1 announced that it would be starting to manufacture large domestically developed commercial aircraft aiming at entering the market in 2020. This project is underpinned by strong economic growth, technological advances and considerable ‘techno nationalism’. The first ‘drawings’ of the project were made available in March 2007. Another project is the C919 narrow-body jet-airliner, planned to enter

ERIKSSON 9780857930088 PRINT.indd 172

29/01/2013 16:59

Foreign knowledge transfer in the development of aircraft industry clusters 173

into service in 2016. This aircraft forms China’s long-term goal to break Airbus and Boeing’s duopoly in the medium-sized jet aircraft market. China’s aerospace industry has advanced over recent few decades. While some of this progress can be attributed to rapidly growing governmental support for China’s aerospace sector, aerospace capabilities have largely benefited from the increasing participation of its aircraft industry in the international supply chains of the world’s leading aerospace firms (Eriksson, 1995, 2010, 2011a). A study of China’s aircraft industry, based on input–output data, concluded that China’s aircraft clusters were inefficient and disadvantaged in comparison with those in the USA and that they have rather few domestic connections to other industrial subgroups (Chu et al., 2010).

CHENGDU Introduction Chengdu in the Sichuan Province is the location of the Chengdu Aircraft Industries Group (CAC), a conglomerate that manufactures aircraft as well as components and parts. CAC was a key enterprise under AVIC I before the merger of AVIC 1 and AVIC 2 in 2008. It was founded in 1958 as the Chengdu State Aircraft Factory (No. 132) and completed in 1964 with the goal of supplying military aircraft to Chinese military forces. CAC consists of three main facilities: the airframe plant, the engine company and the aircraft design institute. The aircraft and engine company employs more than 15 000 people, while the Chengdu Aircraft Design Institute employs around 1800. The aircraft complex is supported by the Chinese Academy of Engineering and the Ministry of Science and Technology through the ‘National High-Tech R&D Program’, and it focuses on advanced material technology and the aerospace industries. The construction of a high-tech industrial park that will feature space and aviation technology is now underway. According to the Guide to Invest in the Aviation and Aerospace Industry of Chengdu (2008), more than 30 organizations and companies are involved in various aviation and aerospace activities such as manufacturing and service activities, but this is not supported by any reliable information. In and around Chengdu, some well-known companies are involved in various aircraft and aerospace manufacturing and service activities including the Haite Group, an enterprise specializing in the repair and inspection of airline appliances, mechanical accessories and small engines as well as the development of airline IT and associated electronic equipment.

ERIKSSON 9780857930088 PRINT.indd 173

29/01/2013 16:59

174

Clusters and economic growth in Asia

The former Chengdu Aviation Instruments Corporation is now integrated into the Sichuan Chengfei Integration Technology Company, which is involved in the development and manufacturing of mouldings and the application of computer integration technology (Sichuan Chengfei Integration Technology Company, 2011). In the commercial maintenance sector, the Chinese actors include the Chengdu Hot Aviation Technology Company. The first aircraft built by CAC were based on Soviet technology, and the firm became a production site for the Chinese-built version of the Soviet MiG-17, known as J-5. Originally, the first J-5s were built in Shenyang from 1957 onwards, but a second production was established in Chengdu. A two-seat trainer (JJ-5) and an export version (F-5) were later developed. The production of the J-5 ended in 1969/70, while the two-seater was produced until 1986. In spite of the Sino–Soviet split in 1960, China did, in 1962, receive the more advanced MiG-21s in kits along with parts and technical documents and set about reverse-engineering the aircraft under the designation J-7. As with the J-5, early production took place in Shenyang, but then all production was transferred to Chengdu. A number of variants for domestic use or export (F-series) were developed, and production started in 1964 and ended in 2006. During the Cultural Revolution (1966–76), China’s aviation industry was heavily damaged and management brought to chaos, but with the end of that period a new ‘dawn’ was seen because of the ambitions to reorganize the Chinese aircraft industry. In 1979, the No. 132 factory became known as the Chengdu Aircraft Company. With the opening up of China, Chengdu’s aircraft manufacturing facilities, as well as other factories, were able to access foreign knowledge and equipment to improve their existing products and develop new ones. In the early 1980s, the Chinese aviation industry started to reorganize and diversify into other products. In Chengdu, this led to the manufacture of light vehicles, windows, motorcycles and dry-cleaning machines. The Development from the 1980s During the honeymoon period of Sino–Western relations in the early 1980s, the Chinese Ministry of Aeronautics decided to import 100 sets of the British GEC-Marconi avionics to upgrade the existing J-7 fleet. The Chengdu Aircraft Manufacturing Factory was responsible for the integration of the British avionics with the J-7 (II) airframe. The project was later cancelled due to changing requirements but Chengdu continued the upgrade without the involvement of the British company (SinoDefence, 2012).

ERIKSSON 9780857930088 PRINT.indd 174

29/01/2013 16:59

Foreign knowledge transfer in the development of aircraft industry clusters 175

A few years later, a joint agreement was signed between the Chinese government and the US aerospace company Grumman to upgrade the J-7 fighters. It was known as the Super-7 fighter programme and some of the upgrading included changes to the nose and air intakes as well as upgrading the fighter with Western-made avionics and engines (Flight International, 26 November 1988). Owing to the increased political tensions between the USA and China, the programme was cancelled in 1990. Anyhow, advanced technologies were transferred to China for a few years. The CAC and Chengdu Aircraft Design Institute (No. 611) continued to work on the project for several years, but China and Pakistan then signed a development and production deal for the FC-1 in July 1999, which was for a rebranded Super J-7. The Russian Klimov RD-93 engine provided the power, while Russia’s Mikoyan Aero-Science Production Group provided assistance in design as well as installing the engine systems (Flight International, 24–30 May 2005). The Pentagon discovered another Chinese aircraft project when an American surveillance satellite orbiting over China spotted several new fighter planes on the runway of a Chinese airbase traditionally used for the test and evaluation of prototype aircraft. This discovery was revealed by the aerospace weekly Flight International (2–8 November 1994) and it created a lot of attention, as the aircraft resembled the cancelled Israel Aircraft Industries (IAI) Lavi fighter. The Lavi was an advanced project with financial and technology support from the USA. Until it was cancelled in 1987, much of the technological development was paid for by the USA. Both Chinese and Israeli officials refuted any purported links between the new aircraft, dubbed the J-10, and the Lavi. A few years later, US officials confirmed that Israel had helped China develop an advanced combat aircraft. However, Israel was not able to transfer the US Pratt & Whitney 1120 turbofan engine and as neither China nor Israel was capable of developing the propulsion system required by the J-10, in 1991 China acquired the AI31F turbofan engine from Russia for incorporation into the J-10 fighter. According to Hewson (2008): Russian aerospace engineers have confirmed to Jane’s that China’s Chengdu J-10 fighter aircraft benefited from significant, direct input from Israel’s Lavi programme – including access to the Israel Aircraft Industries (IAI) Lavi aircraft itself. In a number of interviews Jane’s has talked at length with several engineers, designers and technical specialists – some of whom have been working with their Chinese counterparts for decades and have had firsthand experience on Chinese military projects. They have provided detailed accounts of the assistance given to various Chinese manufacturers and their military aircraft projects. This has included extensive design and performance

ERIKSSON 9780857930088 PRINT.indd 175

29/01/2013 16:59

176

Clusters and economic growth in Asia

modelling, wind-tunnel testing and advanced aerodynamic design input. Senior Russian engineers who spoke to Jane’s recalled their many visits to Chengdu, and elsewhere in China, some of which began in the 1980s. Jane’s was told how Chengdu officials of the highest level stated how they had one of the IAI Lavi prototypes in their facilities.

In March 2007, the Russian Mil Moscow Helicopter Plant JSC set up a joint venture with the Sichuan Lantian Helicopter Company in Chengdu to repair and manufacture the Mi-17 series helicopters for both Chinese and international customers. Production started in May 2008 (SinoDefence, 2011). Yet another Chengdu project is the J-20, as unveiled by a Chinese nongovernmental website in January 2011. Indeed, China’s first known stealth aircraft has recently emerged from a secret development programme (Sweetman, 2011), and thus little is known about its technology. International Subcontracting, Technology Transfer and Investments Other important input into the Chengdu aerospace conglomerate, namely technology linkages, has been subcontracting production to foreign aircraft manufacturers. The first subcontracting work performed was the nose section of the McDonnell Douglas MD-80, with the first delivery in 1991. Chengdu continued to manufacture the MD-80/90 sections for the main US assembly line until 1999. Later, the subcontracting work expanded and CAC produced the Boeing 737 forward entry doors and over-wing exit doors as well as the B747-8 ailerons/spoilers. CAC also produced the empennage of the Boeing 757 (ceased production in 2004). A few years ago, a new agreement was signed with Boeing for the production of the composite rudder of the new Boeing 787 ‘Dreamliner’ (Eriksson, 2011a). The company also manufactures the rear doors and nose cones of the Airbus 320 (Eriksson, 2010). In 2010, an agreement was signed to produce carbon composite wing spoilers and centre hinge fittings for the new Airbus A350. In 2001, CAC became a risk-sharing partner in the ARJ21 regional jet programme. In June 2007, it announced a partnership with US firm Vistagy, in which the latter’s FiberSIM software will be used in the design and development of composite products and parts (Jackson, 2010). The factory also manufactures the nose section for the ARJ21 regional aircraft. The tooling used was originally provided by McDonnell Douglas for the licensed production of the MD-80/90 in China. This implies that the ARJ21 nose section manufactured in Chengdu is based on the same tooling as the company used when working as a subcontractor on the MD-80/90 nose section for McDonnell Douglas between 1991 and 1999

ERIKSSON 9780857930088 PRINT.indd 176

29/01/2013 16:59

Foreign knowledge transfer in the development of aircraft industry clusters 177

(information from US aerospace engineers at Asian Aerospace 2007 Congress, Hong Kong, 4–5 September 2007). In 2007, a strategic collaboration was also signed between the French ESI Group, a world leader in virtual prototyping simulation, and CAC for aeronautic composites manufacturing engineering (ESI Group, 2007). In the same year, the French aerospace company Dassault announced that CAC would be using CATIA (Computer-Aided Three-dimensional Interactive Application) for composite design to develop aircraft parts. This software allows a virtual aircraft to be assembled, in simulation, to check for interferences and to verify the proper fit of the many thousands of parts, thus reducing costly rework. It is widely used by aircraft manufacturing companies worldwide. Finally, other aircraft components have also been manufactured in Chengdu for other international customers. For a number of years, CAC has followed a business strategy labelled a ‘Main Body With Two Wings’, where the development and production of military aircraft is the ‘main body,’ and the subcontracted production of commercial aircraft parts/components and the production of other items, such mechanical and electric articles, are the ‘two wings’. In August 2007, Chengdu Commercial Aircraft was created as a separate company to be responsible for CAC’s civil aviation activities (Jackson, 2010). In recent years, foreign companies have invested in the commercial MRO (maintenance, repair and overhaul) sector. For example, the Sichuan Services Aero Engine Maintenance Company (SSAMC), which is a 60/40 joint venture between Air China and the French aircraft and rocket engine manufacturer SNECMA, based at the Chengdu-Shuangliu airport, offers a wide range of MRO services for CFM engines (SNECMA, 2012). Established in 2008 and commencing service in 2010, the Taikoo Sichuan Aircraft Engineering Services Company is a joint venture between HAECO4 (40 per cent), the Sichuan Airlines Group (42 per cent), Taikoo (Xiamen) Aircraft Engineering (9 per cent) and the Sichuan Haite HighTech Company (9 per cent). It provides heavy maintenance, aircraft conversion, line maintenance, fleet technical management, inventory technical management and other engineering services for Airbus aircraft (Swire Group, 2012).

CONCLUSIONS Chengdu has one of the oldest aircraft facilities in China. Originally founded in the late 1950s with the goal of supplying military aircraft to Chinese military forces, the aircraft built were based on Soviet technology

ERIKSSON 9780857930088 PRINT.indd 177

29/01/2013 16:59

178

Clusters and economic growth in Asia

and reverse-engineered until the early 1980s when the Chinese aviation industry started to reorganize as a result of the political changes taking place in China. The gradually improving political relations with the West, not least the USA, opened the possibilities for China to access advanced Western technology. One important obstacle disappeared when COCOM restrictions on the sale of military technology to China were eased in 1985. In Chengdu, this was reflected in the joint agreement with Grumman for upgrading the J-7 fighters (Super 7). Thus, a leading US aerospace company became involved in technology transfer regarding a military fighter aircraft completely based on Soviet technology. Owing to the increased political tensions between the USA and China, however, the programme was cancelled in 1990. Anyhow, advanced military aerospace technologies were transferred to China during these years. As China and Pakistan did not have a suitable engine or the necessary advanced engine systems, the Russian Klimov Company supplied the engines and technology systems needed. Russia’s Mikoyan Aero-Science Production Group also provided assistance in design as well as installing the engine systems. From the early 1990s, Israel transferred major technologies based on the cancelled Lavi aircraft, which had been developed with financial and technology support from the USA. Israel was not able to transfer the US Pratt & Whitney 1120 turbofan engine and as neither China nor Israel was capable of developing the propulsion system required by the J-10, Russia also became the engine supplier for this aircraft. Owing to the government’s increased emphasis on the commercial aircraft industry, CAC has since the early 1990s been a subcontractor to foreign commercial aircraft manufacturers, implying increased access to foreign knowledge and technology linkages. This started with the work performed for McDonnell Douglas, but was followed by subcontracting works for Boeing and Airbus. The increased emphasis on commercial industrial activities led to the creation of Chengdu Commercial Aircraft as a separate company within CAC in 2007. The same year an agreement was signed with Vistagy to use its FiberSIM software, while strategic collaborations were agreed with ESI and Dassault of France. There was also the transfer of aircraft technology systems from other parts of China, although with foreign origin. The early production of the J-5 and J-7 took place in Shenyang but then all production was transferred to Chengdu. The original source was the Soviet Union. The ARJ nose section, argued to be an independent design, was actually an updated McDonnell Douglas design (MD-80/90) derived from CAC’s period as a subcontractor to the US company. This transfer was also

ERIKSSON 9780857930088 PRINT.indd 178

29/01/2013 16:59

Foreign knowledge transfer in the development of aircraft industry clusters 179

linked to Shanghai where the licensed assembly of MD-82s took place. The Chengdu agglomeration/cluster has in recent years also witnessed the establishment of foreign actors within the commercial MRO sector, which can be attributed to Chengdu’s increased importance as a commercial aviation hub. On one hand, this study focused on foreign companies as major sources of knowledge and technology transfer. It is well known that foreign companies are important sources of technology transfer in emerging economies and that supply chains are major tools of knowledge and technology transfer in the aircraft industry. On the other hand, nothing has been investigated regarding any localized processes or the connectedness of knowledge creation and technology development. The processes of knowledge creation and innovation consist of a complex set of networks and processes operating within and across various geographical scales: global, national and regional down to local. Neither has any kind of social or economic interconnectedness or dynamics at the local level been looked at. However, a previous study of China’s aircraft industry, based on input– output data, concluded that its aircraft clusters were inefficient with few domestic connections to other industrial subgroups. From a domestic perspective, China has a number of institutions that supply the industry with engineers, researchers and technicians. It is known that foreign-educated engineers and researchers have returned to China after being educated abroad and thus brought competence and knowledge back to China, although we do not know to what extent that has occurred in this industry and in various locations, such as Chengdu. Nevertheless, foreign technology transfer has been of major importance to develop aircraft manufacturing in Chengdu, as there is still limited capacity in the Chinese aircraft industry to generate new indigenous technology. From an adoption perspective, there are signs of both neoclassical ‘accumulation theories’ and ‘assimilation’ theories that use more evolutionary views and explicitly stress learning to adapt and operate foreign technologies. In an advanced high-technology sector such as the aircraft industry, external linkages, that is, the foreign technology input of major technology systems, are more important than in most other industries. This also raises questions about the local, regional and national innovation systems within this high-tech industry. To be able to gain a deeper knowledge and understanding of the connectedness, innovation and knowledge creation in this and other aircraft/ aerospace clusters there is a need to carry out both more in-depth analysis based on input–output data and various kinds of qualitative studies to look at the real activities taking place.

ERIKSSON 9780857930088 PRINT.indd 179

29/01/2013 16:59

180

Clusters and economic growth in Asia

NOTES 1. For an overview of the production figures and locations of the aircraft manufacturing industries until the mid-1990s, see Eriksson (1995, pp. 125–43). 2. Coordinating Committee for Multilateral Export Controls. 3. The first foreign subcontracting work awarded to China was the offset deal in 1979 from McDonnell Douglas to produce MD-80 landing gear doors in Shanghai (Eriksson, 1995). 4. HAECO (Hong Kong Aircraft Engineering Company) is a subsidiary of the Swire Group.

REFERENCES Antonov News (2007), accessed 29 October 2011 at www.antonov.com/news/index. xml?news5antk-main/news_20071224.xml. Aviation Daily (2001), ‘Embraer Sells 40 Jets to China in $1 Billion Order’, 18 April. Bailey, J. (1993), ‘Chasing the Genie’, Flight International, 23–29 June, 56–60. Bathelt, H., A. Malmberg and P. Maskell (2004), ‘Clusters and Knowledge: Local Buzz, Global Pipelines and the Process of Knowledge Creation’, Progress in Human Geography, 28(1), 31–56. Biggiero, L. and A. Sammarra (2010), ‘Does Geographical Proximity Enhance Knowledge Exchange? The Case of the Aerospace Industrial Cluster of Centre Italy’, International Journal of Technology Transfer and Commercialisation, 9(4), 283–305. Breschi, S. and F. Lissoni (2001), ‘Knowledge Spillovers and Local Innovation Systems: A Critical Survey’, Industrial and Corporate Change, 10(4), 975–1005. China Today: Aviation Industry (1989), Beijing: Ministry of Aero-Space Industry (MAS), China Aviation Industry Press. Chu, B. H. Zhang and F. Jin (2010), ‘Identification and Comparison of Aircraft Industry Clusters in China and United States’, Chinese Geographical Science, 20(5), 471–80. De Bresson, C. (1989), ‘Breeding Innovation Clusters: A Source of Dynamic Development’, World Development, 17(1), 1–16. Eriksson, S. (1995), ‘Global Shift in the Aircraft Industry: A Study of Airframe Manufacturing with Reference to the Asian NIEs’, publication edited by the Department of Geography, University of Gothenburg, Series B, No. 86 (Doctoral dissertation). Eriksson, S. (2000), ‘Technology Spillover from the Aircraft Industry: The Case of Volvo Aero’, Technovation, 20(12), 653–64. Eriksson, S. (2003), ‘Technology Policies in East and Southeast Asia: Creating a New Global Structure in the Aircraft Industry’, Asian Geographer, 22(1–2), 169–85. Eriksson, S. (2006), ‘Cluster Creation and Innovation Within an Emerging Taiwanese High-tech Sector’, International Journal of Technology Transfer and Commercialisation, 5(3), 208–36. Eriksson, S. (2010), ‘China’s Aircraft Industry: Collaboration and Technology Transfer – The Case of Airbus’, International Journal of Technology Transfer and Commercialisation, 9(4), 306–25.

ERIKSSON 9780857930088 PRINT.indd 180

29/01/2013 16:59

Foreign knowledge transfer in the development of aircraft industry clusters 181 Eriksson, S. (2011a), ‘Globalisation and Changes of Aircraft Manufacturing Production/Supply-chains – The Case of China’, International Journal of Logistics Economics and Globalisation, 3(1), 70–83. Eriksson, S. (2011b), ‘Promotion of Company and Local Economic Growth Through Clusters’, in Charlie Karlsson and Robert G. Picard (eds), Media Clusters: Spatial Agglomeration and Content Capabilities, Cheltenham, UK and Northampton, MA, USA: Edward Elgar, pp. 30–43. ESI Group (2007), accessed 20 November 2011 at http://www.esi-group.com/ corporate/news-media/press-releases/2007-english-pr/chengdu-aircraft-industryand-esi-group-strategic-collaboration-for-aeronautic-composites-manufacturingengineering. Flight International (1988), 26 November. Flight International (2004), 2–8 November. Flight International (2005), 24–30 May. Guide to Invest in the Aviation and Aerospace Industry of Chengdu (2008), accessed 28 October 2011 at www.ebn.eu/assets/assets/pdf/events/chengdu per cent20 per cent20aviation per cent20and per cent20aerospace per cent20industry.pdf. Hewson, R. (2008), ‘Chinese J-10 “Benefited from the Lavi Project”’, IHS Jane’s, 5 October, accessed 16 November 2012 at http://www.janes.com/products/janes/ defence-security-report.aspx?id51065926403. Jackson, P. (2010), Jane’s All the World’s Aircraft 2010–2011, Coulsdon: IHS Jane’s. Lall, S. (2000), ‘Technological Change and Industrialization in the Asian Newly Industrializing Economies: Achievements and Challenges’, in L. Kim and R.R. Nelsson (eds), Technology, Learning & Innovation, Experiences of Newly Industrializing Economies, Cambridge, Cambridge University Press. Nelson, R.R. and H. Pack (1999), ‘The Asian Miracle and Modern Growth Theory’, The Economic Journal, 109(457), 416–36. Niosi, J. and M. Zhegu (2005), ‘Aerospace Clusters: Local or Global Knowledge Spillovers?’, Industry and Innovation, 12(1), 5–29. Owen-Smith, J. and W.W. Powell (2004), ‘Knowledge Networks as Channels and Conduits: The Effects of Spillovers in the Boston Biotechnology Community’, Organization Science, 15(1), 5–21. Sichuan Chengfei Integration Technology Company (2011), accessed 12 December 2011 at http://en.cac-citc.cn/newEbiz1/EbizPortalFG/portal/html/index.html. SinoDefence (2011), accessed 20 October 2011 at http://www.sinodefence.com/ airforce/helicopter/mi17.asp. SinoDefence (2012), accessed 12 January 2012 at http://www.sinodefence.com/ airforce/fighter/j7.asp. SNECMA (2012), accessed 12 February 2012 at http://www.snecma.com/chengdu-.html?lang5en. Sweetman, B. (2011), ‘Chinese J-20 Stealth Fighter in Taxi Tests’, Aviation Week & Space Technology, 3 January. Sweeney, G.P. (1987), Innovation, Entrepreneurs, and Regional Development, New York: St Martin’s Press. Swire Group (2012), accessed 15 February 2012 at http://www.swire.com/eng/ activities/aviation_company.php?company5TaikooSichuanAircraftEngineering ServicesCompanyLimited.

ERIKSSON 9780857930088 PRINT.indd 181

29/01/2013 16:59

ERIKSSON 9780857930088 PRINT.indd 182

29/01/2013 16:59

Index agglomerations importance of industrial 120–121, 137 regional 1, 2, 6, 14 theoretical aspects 137–8 see also industrial agglomeration study Airbus 159, 172–3, 176, 177 aircraft industry characteristics 143 China 168–73 Chengdu 173–80 government support for 166 as complex product system 141–2 importance of external linkages 179 Indonesian investment in 144 as innovation-intensive industry 142 internationally competitive 143–4 IPTN (Industri Pesawat Terbang Nusantara) 146–7, 160–162 development of new technology 155–7 implementation 151, 152 influence of Habibie 149–50 Joint Venture Programme 153–5 large-scale R&D programme 157–8 licence programme 151, 153 long-term goals 147–9 subcontracting and other activities 159–60 Taiwanese 167–8 technological requirements 143 Amsden, A. 7, 9, 15 anchor firms 12, 137 Anwar, S. 120, 136 ARJ21 regional aircraft 171–2, 176–7 ASEAN countries 121, 127–9, 133 Asian financial crisis 11, 26, 27, 29, 32, 34, 121, 125, 144, 157 Asian triangle 12

Audretsch, D.B. 45, 69, 78 average labour productivity (ALP) growth 24, 30–31, 34–7 Aviation Industry Corporation of China (AVIC) 170–171, 172, 173 Bailey, J. 150, 169 Bandung, Indonesia 146, 153–5, 159–60 Belgium 47, 57 bio-ventures 73, 78 biomedical clusters 66–7, 80–82 drivers of cluster formation intermediary organizations 76–7 local government 75–6, 81 star ventures 78–80 universities 73–5 theory and practice of industry cluster 67–70 Wonju, South Korea 70–73 Biopolis, Singapore 70 Bluegrass Auto Manufacturers Association 68 Boeing 159, 169, 172–3, 176 Bohai Bay, China 102–3, 107–9, 116, 117 business incubation centres 71, 74–5, 78 C919 narrow-body jet-airliner 172–3 capability-building strategies 11, 16 technological 89, 90, 142 CASA, Spain 151–4, 159, 160, 161 catch-up growth 1–2, 7, 8, 11, 13, 87–9 Chengdu Aircraft Industries Group (CAC) 173–4, 175, 176–7 Chengdu, China 173–8 China aircraft industry 177–80 Chengdu 173–8

183

ERIKSSON 9780857930088 PRINT.indd 183

29/01/2013 16:59

184

Clusters and economic growth in Asia

commercial aircraft production 166 early years 168–9 and open-door policy 165, 169, 170–171 in twenty-first century 171–3 concentration of industrial activities 45–6, 47 Cultural Revolution 169, 174 economic and industrial development 165 imports to Vietnam 133–4 PC hardware manufacturing 40 protection of intellectual property 59 socialist market reform 49 transformation into biomedical R&D hubs 70 see also global economic crisis in China; industrial agglomeration study ChoongWae Medical 74, 75 cluster engine catalysts 78–80 cluster policies 1–2 and clusters 2–4 East Asian 11–16 industrial 137 and role of the state 4–6 clusters 2–6 aircraft industry 166–8, 173, 179 biomedical, in South Korea 66–82 in coastal regions 40, 43 components of East Asian 12 established versus emerging 67 formation of 3, 73–80 importance of development in Vietnam 137 industrial, and FDI-induced transfer of technology 120–121 industrial, theory and practice 67–70 see also industrial agglomeration study CN-235 aircraft 148, 153–5, 158, 159, 160 co-agglomeration effects 53–6 coefficient of variance (CV) index 46 competition amongst electronics firms 49 balance with cooperation 3 for low-cost labour 136

ERIKSSON 9780857930088 PRINT.indd 184

and Porterian cluster approach 5–6 for use of scarce land 125 ‘competitive advantage of nations’ 5, 66 competitiveness 2–3 in China 92–3 international 10, 11, 16 and local labour market 68 technological improvement 142 threats to 59 in Wonju 73 coordinators 75–6 culture, local, and cluster formation 3 Dassault, France 177, 178 developmental state concept 7, 8–10, 15 dissimilarity index 46 Dongguan, China 110 see also industrial agglomeration study Dunning, J.H. 42–3 East Asia ‘East Asian Miracle’ 6–7 entrepreneurial states 6–11 performance of economies 70 ‘Renaissance’ cluster policies 11–16 Ebner, A. 4, 9, 10, 11, 15 electronics industry firm-level survey in Guangdong 109–14, 116–17 Taiwanese, in Dongguan 40–62 Ellison and Glaeser (EG) index 41, 46–8, 51–3, 56–7, 60–61 embedded autonomy 7 Embraer 143, 172 entrepreneurial states 1, 6–11 entrepreneurs biotech 70 confidence in China 95–6 innovative 69, 73–5 and management know-how 77 Eriksson, S. 141, 143, 162, 163, 166–73, 176, 180 ERJ-145 regional jets 172 ESI Group, France 177, 178 Europe investment in technological activities in manufacturing FDI 127–9

29/01/2013 16:59

Index supplying companies 171 and Vietnamese exports 133 external linkages 4, 166, 179 Feldman, M.P. 45, 69, 70, 78 Feser, E.J. 48, 68, 77 foreign direct investment (FDI) in Vietnam 119–20, 121–3, 134–8 industrial clusters and FDI-induced transfer of technology 120–121 industrial zone developments 123–5 manufacturing gross output and value-added 129–31 technological content of foreign trade 131–4 Francis, J.L. 69, 70, 78 general purpose technology (GPT) 21 Gini coefficient 45–6 Glaeser, E.L. 42 see also Ellison and Glaeser index (EG) index global economic crisis in China determinants for development impact on growth paths 92–4, 117 institutional evolution and role of state 91–2 market orientation 91 resilience, regional growth and catch-up paths 87–9 sectorial structural change 89–90 technological capability building 90 discussion 114–16 economic recovery 85–6, 116 hypotheses on 86–7, 92–4, 116 regional analysis firm-level survey of electronics industry 109–14, 116–17 general findings 95–6 impacts and recovery by industrial sector 96–101 impacts and recovery by province and region 101–5 impacts on technological upgrading 105–6, 117 patent applications 106–9 Global Seed Capital 70, 82 globalization 5, 11, 42, 90 governed markets 7

ERIKSSON 9780857930088 PRINT.indd 185

185

government-business relations 1, 2, 6, 8, 12, 15 gross output, manufacturing 129–31 gross value of industrial output (GVIO) 124–30, 138 growth accounting model algorithms 22–4 data description ICT capital stock and price index 25–30 labour input 30 national account data 24–5 growth paths 86, 87–9, 97–100, 103, 117 Guangdong, China Dongguan’s location in 45 economic characteristics 102 firm-level survey of electronics industry 109–14, 116–17 growth rates in industrial valueadded 104–5 output shares 58 patent applications 107, 108–9 sectorial structural change 89–90 successful recovery of 115, 117 Habibie, B.J. 144–50, 154–6, 158, 161 Hamburger Flugzeugbau (HFB) 146–7 Hanoi, Vietnam 123, 125 He, C. 43, 45, 46, 53, 61, 100 Hebei, China 102, 103, 105, 115 Herfindahl-Hirschman index 46–7 Herfindahl index 46 Ho Chi Minh City, Vietnam 123, 125, 135 Hong Kong enterprises with funds from 48, 59, 62 ICT and economic growth 31–7 ICT capital stock and price index 28–9 investment in technological activities in manufacturing FDI 128–9 labour input 30 national account data 24–5 Hsia, C.J. 43, 60 Huangjiang, Dongguan, China 49–51, 55–6 Human-Tech 78

29/01/2013 16:59

186

Clusters and economic growth in Asia

ICT capital stock and price index 25–30 ICT (information and communication technology) study model description 22–30 results and analysis 30–36 role in economic growth 21–2, 36–7 Indonesia aircraft industry 160–162 economic policies 144–6 establishment of IPTN 146–60 industrial agglomeration study 40–42, 60–62 co-agglomeration effects 53–6 extent of agglomeration 51–3 and firm size distribution 56–7 formation of electronics cluster 49 ‘home effects’ 43 implications for upgrading 58–60 and inter-firm linkages 42–4 methods and data firm-level data and interviews 48–9 measure of agglomeration 44–8 town-level analysis 44 spatial distribution of electronics firms 49–51 sub-industries 51, 53–6, 61 three-digit industries 48, 51, 53–4, 58, 61 two-digit industries 48, 51–2, 54–5, 58, 60–61 industrial clusters 120–121 industrial policies and state functions 9–10 industrial zones (IZs) 123–5, 126–7, 135–8 industry sector recovery in China 96–101 industry tiers and groupings 100, 142–3 innovation 6, 9–11, 12, 14–16, 141–2 innovative entrepreneurs 69, 73–5 innovative milieu 3 institutional evolution and role of state 91–2 inter-firm linkages 41, 42–4, 53, 61 intermediary organizations 76–7, 81 IPTN (Industri Pesawat Terbang Nusantara) see aircraft industry Israel 175, 178

ERIKSSON 9780857930088 PRINT.indd 186

J-5 aircraft 174, 178 J-7 aircraft 174–5, 178 J-10 aircraft 175–6, 178 J-20 aircraft 176 Japan auto TNCs 43 biomedical clusters 76, 82 ICT and economic growth 21–2, 31–7 ICT capital stock and price index 25–6 investment in technological activities in manufacturing FDI 127–9 labour input 30 as late industrializing economy 7 move to skilled-intense activities 142 national account data 24 as regional technology leader 13–14 Jiangsu, China 102, 103, 105, 107, 115, 116 Johnson, Chalmers 7 joint venture programme 153–5 Jorgenson, D.W. 21, 24 key economic areas (KEAs) 123–5 Kim, L. 13, 88 Klimov Company, Russia 175, 178 knowledge, globalized 15 knowledge producers 69, 73–5, 81 knowledge spillovers 68–9 knowledge transfer 11, 165–8, 179 Koo, J. 67, 68, 69, 75, 77, 78 Kroll, H. 85, 90 Kuchiki, A. 12, 138 labour productivity growth 21, 24, 30–31, 34–7 Lall, S. 8, 88, 165 Lavi fighter 175–6, 178 learning 10–11, 165 learning regions 3, 4 Li, Y. 60 Liao, H. 41, 44, 47, 59, 60 linkages backward 121, 136 external 166, 179 forward 121 inter-firm 41, 42–4, 53, 61 technology 176, 178 local government role 75–6, 81

29/01/2013 16:59

Index locational strategies 11, 16 low technology (LT) activities 126–30 Lundvall, B.-A. 5, 6 management consultants 76–7 manufacturing gross output and value-added 129–31 technological contents of 126–9 in Vietnam 119–25, 132–6 see also industrial agglomeration study manufacturing value-added (MVA) 130–131 market orientation 89, 91, 93, 97, 111–13 Marshall, A. 3, 42, 67, 120 Martin, R. 2, 42, 86, 87 Maskell, P. 3, 4, 5 McDonnell Douglas 170, 176, 178 MD-80 aircraft 172, 176, 178, 180 MD-82 airliner 170, 179 Mediana 74, 78 Medical Industry Techno Valley (MITV) 71–3, 76–7, 81 medium-low technology (MLT) activities 126–30 Messerschmitt-Bölkow-Blohm (MBB) 146–7, 151, 153 Mikoyan Aero-Science Production Group 175, 178 modular economy concept 13 N-250 turbo-prop 155–8, 161 N-2130 aircraft 149, 158 Naudé, W. 88 Nelson, R.R. 6, 141, 165 Nestor, C. 123, 125, 126, 136, 138 network catalysts 76–7 networks 3–4, 13–16 Nguyen, P.L. 120, 136 OECD (Organisation for Economic Co-operation and Development) 85, 89, 90, 91–2, 121 open-door policy 165, 169, 170–71 paradigm creation 10, 15 patent applications 106–9

ERIKSSON 9780857930088 PRINT.indd 187

187

Pearl River Delta (PRD) 12, 40, 49, 60, 102, 104, 107, 109–13 Porter, M.E. 2–3, 5–6, 14, 66, 68, 120, 149 Porterian cluster approach 2–3, 5–6, 15 Pratt & Whitney turbofans 169, 175, 178 private domestic sector (PDS), Vietnam 129–30 Progressive Manufacturing Programme (PMP) 151, 153, 160 provincial recovery in China 101–5 proximity, geographical 3, 42, 68–9, 71, 120 Qingxi, China 49–50, 55–6 R&D (research and development) advanced 142–3 large-scale programme 157–8 locations 70 real GDP growth (RGDP) 23–4, 30–34, 36–7 regional growth evolution 87–9 regional recovery in China 101–5 regulatory state functions and policies 8–10 Research Triangle Park 66, 75, 77, 79 resilience 87–8, 111–12 resource allocators 75–6 Rosenberg, N. 6, 141 Russia 175–6, 178 San Jose 67, 72, 73, 77 Schiller, D. 85, 90 Schumpeterian theory 11, 142 sectorial structural change 88, 89–90 Shane, S. 66, 70 Shijie, China 43, 49–50, 55–6 Silicon Valley 66, 68, 79, 80 Simmie, J. 86, 87 Singapore 11, 12 cluster development 15, 70 ICT and economic growth 31–7 ICT capital stock and price index 29–30 investment in technological activities in manufacturing FDI 127–9 labour input 30 national account data 25

29/01/2013 16:59

188

Clusters and economic growth in Asia

SinoDefence 174, 176 software investment 27–9, 37 Solow Paradox 21, 37 South Korea 11, 14 biomedical clusters 66–82 ICT and economic growth 22, 31–7 ICT capital stock and price index 25–9 investment in technological activities in manufacturing FDI 127–9 labour input 30 national account data 24 transformation into biomedical R&D hubs 70 star ventures 67, 78–80, 81 state-owned enterprises (SOEs) 129–30, 139 state, role of 4–6, 89, 91–2 statistical standards 24 Stiglitz, J. 85, 91 subcontracting 159–60, 176–7 Suharto, President 144–5, 147, 148, 160 Sunley, P. 2, 42 Taiwan 11, 12, 14–15 inter-firm linkages 41 investment in technological activities in manufacturing FDI 127–9 PC manufacturing output 40 relations with China 168 rising production costs 49 Taiwanese electronics firms see industrial agglomeration study Tangxia, China 49–50 Techno-Park Development Project 71, 74 technological capability building 89, 90, 142 technological content of foreign trade 131–4 of manufacturing 126–9 technological innovation 14–15 technological learning 7, 42, 142 technological upgrading 8, 105–7, 109, 117 technology commercialization of 9–10 and Habibie 145, 147, 149–50, 160 levels of 126–30 and manufacturing FDI 126–9

ERIKSSON 9780857930088 PRINT.indd 188

in modern aircraft 143–4 new, development of (phase in establishment of IPTN) 155–7 start-ups 75–6 transfer 120–121, 136, 151, 160–161, 176–9 see also ICT (information and communication technology) study total factor productivity (TFP) 22, 23 Toyota Corporation 68 transformative capacity 8 transnational corporations (TNCs) 42–4, 46, 57, 59, 61 Tsuji, M. 12, 138 UNCTAD (United Nations Conference on Trade and Development) 43, 44, 121, 132, 134, 142 UNIDO (United Nations Industrial Development Organization) 126, 131, 137 universities 69, 73–5, 81 USA airline trips 169 and Chinese aircraft project 175, 178 contractors to China 171 ICT and economic growth 21–2 investment in biomedical start-ups 66 investment in technological activities in manufacturing FDI 127–9 and Vietnamese exports 133, 134 value-added growth, development 96, 98–101, 104–5 manufacturing 126, 129–31 Vietnam challenges faced by 137 economic reform programme 119 exports 132–3, 135–6 FDI-driven economic development 134–5, 136 FDI overview 121–3 impact of FDI 129–31 imports 132, 133, 135–6 industrial clusters 120–121, 135, 137–8

29/01/2013 16:59

Index industrial zones (IZs) 123–5, 135, 137 manufacturing sector 136–7 technological content of foreign trade 131–4 technological intensity 132 technological transfer 120–121, 136 trade balance 133–4 trade with China 133–4 Wade, Robert 5, 7 Weber, A. 67 Wei, Y.H.D. 41, 59 Wong, P.K. 8, 88

ERIKSSON 9780857930088 PRINT.indd 189

189

Wonju, South Korea 67, 70–79, 81 World Bank 6–8, 11–13, 100–101, 130–131 Yang, C. 41, 43, 44, 47, 56, 59, 60, 61 Yang, Y.R. 43, 60 Yangtze River Delta (YRD) 40, 43, 59 Yeung, H.W.C. 43, 61 Yonsei University 71, 74 Zhejiang, China 102, 103–5, 107, 115, 116, 117 Zhou, Y. 91, 102

29/01/2013 16:59

ERIKSSON 9780857930088 PRINT.indd 197

29/01/2013 16:59

ERIKSSON 9780857930088 PRINT.indd 198

29/01/2013 16:59