Clusters and Economic Growth in Asia

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
Sren Eriksson
Jnkping International Business School, Jnkping
University, Sweden
Edward Elgar
Cheltenham, UK Northampton, MA, USA
ERIKSSON 9780857930088 PRINT.indd iii
29/01/2013 16:59
Sren 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
4
5
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
Sren Eriksson
Foreign knowledge transfer in the development of aircraft
industry clusters the case of Chengdu, China
Sren Eriksson
Index
21
40
66
85
119
141
165
183
ERIKSSON 9780857930088 PRINT.indd v
29/01/2013 16:59
Contributors
Wai-Mun Chia obtained her Bachelors degree in economics from the
University of London and then pursued her Masters 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 Masters
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 Ebners
research interests involve the matters of entrepreneurship, innovation,
governance and regional economic development.
Sren Eriksson is a Professor of Economic Geography at Jnkping
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 Asias
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
FraunhoferInstitute 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
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 Suhartos 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 Sren 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 Chinas new stages of economic and
industrial growth. The government also expressed the interest in and
ambition to develop aircraft-related clusters. In Chapter 8, Sren Eriksson
investigates the origin and characteristics of foreign technology transfer
into Chengdu, one of the countrys 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.
Sren Eriksson
Professor of Economic Geography
Jnkping 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 governmentbusiness 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
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
governmentbusiness 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 Porters 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
29/01/2013 16:59
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, Porters 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
Porters 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
29/01/2013 16:59
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
29/01/2013 16:59
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 Porters 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
29/01/2013 16:59
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. 2489). 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, Porters concern with regional agglomerations of cluster
firms mirrors both the multi-scalar and multi-level qualities of the
innovation-driven process of economic development. Porters 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 governmentbusiness relations that may be labelled
entrepreneurial state.
TOWARDS ENTREPRENEURIAL STATES IN EAST

ASIA
The historically specific developmental impact of governmentbusiness
relations in East Asian economies is subject to persistent discussions that
have been most prominently fuelled by the World Banks 1993 policy
research report on the East Asian Miracle. Capital accumulation,
29/01/2013 16:59
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. 1011). Growth-promoting market interventions in the domain of
industrial policy are addressed as components of these public policies with
clear costbenefit considerations and performance criteria (ibid., pp. 58).
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 Johnsons 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. 1920).
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. 3057). 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. 34). 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. 289). 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
29/01/2013 16:59
state, thus allowing for a reconfiguration of governmentbusiness networks. Therefore, the reconstructive self-transformation of the transitory
developmental state mirrors an increasing complexity of the socioeconomic domain (ibid., pp. 22634). 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. 645). 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 publicprivate 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. 279). 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. 437). In associated terms, developmental
and regulatory state functions are differentiated. The neo-developmental
state for high-tech industries copes with the promotion of competitive
29/01/2013 16:59
Table 1.1
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. 16772). 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. 1516). 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. 11819). 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
29/01/2013 16:59
10
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. 3823). 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.
29/01/2013 16:59
11
The institutional architecture of the entrepreneurial state underlines

the role of knowledge transfers and communication in statesociety
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. 51112). 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 Banks slogan of the East Asian Renaissance.
Decisively, this notion of a resurgence of the East Asian development
29/01/2013 16:59
12
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. 1216). 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. 35).
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. 2346). 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. 56).
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. 24).
29/01/2013 16:59
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. 24954). 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 arrayof
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
29/01/2013 16:59
14
pattern towards a more competitive and entrepreneurial setting (Aoki,

2002, p. 2). Japans 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. 356). 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. 8081). Japans 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. 45). 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
marketsegments 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. 14556). Similar implications hold for Taiwan, where
sustaining competitiveness implies continuous technological upgrading
29/01/2013 16:59
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. 569; 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 governmentbusiness
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
29/01/2013 16:59
16
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), Asias Next Giant: South Korea and Late Industrialization,
New York: Oxford University Press.
Amsden, A. and W. Chu (2003), Beyond Late Development: Taiwans 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. Martins Press,
pp. 24366.
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. 21834.
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), 3156.
Becattini, G. (1991), Italian Industrial Districts: Problems and Perspectives,
International Studies of Management and Organization, 21(1), 8390.
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(56), 101123.
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), 51120.
Bergmann, E.M., D. Charles and P. den Hertog (2001), In Pursuit of Innovative
29/01/2013 16:59
17
Clusters, in OECD (ed.), Innovative Clusters: Drivers of National Innovation

Systems, Paris: OECD, pp. 718.
Camagni, R. (1995), The Concept of Innovative Milieu and its Relevance for
Public Policies in European Lagging Regions, Papers in Regional Sciences,
74(4), 31740.
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. 225.
Cooke, P. and G. Schienstock (2000), Structural Competitiveness and Learning
Regions, Enterprise and Innovation Management Studies, 1(3), 26580.
Crevoisier, O. (2004), The Innovative Milieu Approach: Towards a Territorialized
Understanding of the Economy? Economic Geography, 80(4), 36779.
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), 124.
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. 4876.
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), 497515.
Ebner, A. (2007), Public Policy, Governance, and Innovation: Entrepreneurial
States in East Asian Economic Development, International Journal of
Technology and Globalisation, 3(1), 10324.
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. 36788.
Ebner, A. and H. Schiele (2012), The Role of Domestic and International
Cluster Linkages Explored on the Example of BuyerSupplier 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), 141729.
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), 191211.
Furman, J.F., M. Porter and S. Stern (2002), The Determinants of National
Innovative Capacity, Research Policy, 31(6), 899933.
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. 324.
Gertler, M. (2001), Best Practice? Geography, Learning and the Institutional
Limits to Strong Convergence, Journal of Economic Geography, 1(1), 526.
Gill, I. and H. Kharas (2007), An East Asian Renaissance: Ideas for Economic
Growth, Washington, DC: World Bank.
29/01/2013 16:59
18
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. 112.
Hassink, R. (2005), How to Unlock Regional Economies from Path Dependency?
From Learning Region to Learning Cluster, European Planning Studies, 13(4),
52135.
Hobday, M., H. Rush and J. Bessant (2004), Approaching the Innovation
Frontier in Korea: The Transition Phase to Leadership, Research Policy, 33(4),
143357.
Hu, M.-C. and J.A. Mathews (2005), National Innovative Capacity in East Asia,
Research Policy, 34(3), 132249.
Ibata-Arens, K. (2004), Japans 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), 115566.
Johnson, C. (1982), MITI and the Japanese Miracle: The Growth of Industrial
Policy, 19251975, Stanford: Stanford University Press.
Ketels, C.H.M. (2006), Michael Porters Competitiveness Framework Recent
Learnings and New Research Priorities, Journal of Industry, Competition and
Trade, 6(1), 11536.
Kitagawa, F. (2007), The Regionalization of Science and Innovation Governance
in Japan?, Regional Studies, 8(41), 1099114.
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. 317.
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. 112.
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. 120.
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. 1368.
Loasby, B.-J. (2001), Time, Knowledge and Evolutionary Dynamics: Why
Connections Matter, Journal of Evolutionary Economics, 11(4), 393412.
Low, L. (2004), Singapores Developmental State Between a Rock and a Hard
Place, in L. Low (ed.), Developmental States: Relevancy, Redundancy or
Reconfiguration?, New York: Nova Science, pp. 16177.
Lundvall, B.-A. (1998), Why Study National Systems and National Styles of
Innovation?, Technology Analysis and Strategic Management, 10(4), 40721.
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. 35372.
29/01/2013 16:59
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), 21331.
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), 42949.
Martin, R.A. and P. Sunley (2003), Deconstructing Clusters: Chaotic Concept or
Policy Panacea? Journal of Economic Geography, 3(1), 535.
Maskell, P. (2001), The Firm in Economic Geography, Economic Geography,
77(4), 32944.
Maskell, P. and M. Lorenzen (2004), The Cluster as Market Organisation, Urban
Studies, 41(56), 9911009.
Maskell, P. and A. Malmberg (1999), Localised Learning and Industrial
Competitiveness, Cambridge Journal of Economics, 23(2), 16785.
Maskell, P., H. Bathelt and A. Malmberg (2006), Building Global Knowledge
Pipelines: The Role of Temporary Clusters, European Planning Studies, 14(8),
9971013.
Menzel, M.-P. and D. Fornahl (2010), Cluster Life Cycles Dimensions and
Rationales of Cluster Evolution, Industrial and Corporate Change, 19(1),
20538.
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. 118.
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. 13349.
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. 25374.
Porter, M.E. (2003), The Economic Performance of Regions, Regional Studies,
37(5), 54978.
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), 2750.
Sanz-Menendez, L. and L. Cruz-Castro (2005), Explaining the Science and
Technology Policies of Regional Governments, Regional Studies, 39(3), 93954.
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
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), 23551.
Suehiro, A. (2009), From an Industrial Policy Approach to an Industrial Cluster
29/01/2013 16:59
20
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. 2567.
Visser, E.-J. (2009), The Complementary Dynamic Effects of Clusters and
Networks, Industry and Innovation, 16(2), 16795.
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. 6088.
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), 2155.
Whitford, J. and C. Potter (2007), Regional Economies, Open Networks and the
Spatial Fragmentation of Production, Socio-Economic Review, 5(3), 497526.
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. 5785.
Wong, J., U. Quach, H. Thorsteinsdttir, P.A. Singer and A.S. Daar (2004),
South Korean Biotechnology A Rising Industrial and Scientific Powerhouse,
Nature Biotechnology, 22 (Supplement) December, 427.
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. 154.
Woodward, D. and P. Guimaraes (2009), Porters Cluster Strategy and Industrial
Targeting, in S.J. Goetz, S. Deller and T. Harris (eds), Targeting Regional
Economic Development, London: Routledge, pp. 6883.
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), 2144.
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
29/01/2013 16:59
22
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
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
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
PA,t 2 PA,t21
PA,t21
bPA,t21
29/01/2013 16:59
24
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
29/01/2013 16:59
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 inputoutput 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.
29/01/2013 16:59
26

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
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 Japans (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).
29/01/2013 16:59
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 Koreas (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
29/01/2013 16:59
28
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
b)
Figure 2.3a, b
Year

of Hong Kongs (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 Kongs 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
29/01/2013 16:59
29
the 1997 Asian financial crisis both had a lagged impact on Hong Kongs
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, inputoutput 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.
29/01/2013 16:59
30

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
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

of Singapores (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
29/01/2013 16:59
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: 198690, 199195,
19962000 and 200106. In the case of Japan, due to lack of data, the last
period is 200102. 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 Japans
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
29/01/2013 16:59
32
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 200106, which is also the period
29/01/2013 16:59
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*
Software
Labour input
Labour quality
MFP growth
Hong Kong
Growth rate of RGDP
Contribution from:
Non-ICT capital
ICT capital*
Software
Labour input
Labour quality
MFP growth
Singapore
Growth rate of RGDP
Contribution from:
Non-ICT capital
ICT capital*
Software
Labour input
Labour quality
MFP growth
198690
199195
19962000
200102
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
198690
199195
19962000
200106
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.
29/01/2013 16:59
34

Relative contribution of ICT to GDP growth
25
Percent
20
15
10
5
0
198690
199195
19962000
200106
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 19962000, 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
29/01/2013 16:59
Table 2.2
35
Decomposition of ALP growth

Percentage
Japan
ALP growth
Contribution from:
Non-ICT capital
ICT capital*
Software
Labour quality
MFP growth
South Korea
ALP growth
Contribution from:
Non-ICT capital
ICT capital*
Software
Labour quality
MFP growth
Hong Kong
ALP growth
Contribution from:
Non-ICT capital
ICT capital*
Software
Labour quality
MFP Growth
Singapore
ALP growth
Contribution from:
Non-ICT capital
ICT capital*
Software
Labour quality
MFP growth
198690
199195
19962000
200102
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
198690
199195
199600
200106
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.
29/01/2013 16:59
36

Relative contribution of ICT to ALP growth
25
Percent
20
15
10
5
0
198690
199195
19962000
200106
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 200106. 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
29/01/2013 16:59
37
19962000, 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
200106, 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 198690. 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, Wed 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).
29/01/2013 16:59
38
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), 971.
Brynjolfsson, E. and L.M. Hitt (2000), Beyond Computation: Information
Technology, Organizational Transformation and Business Performance,
Journal of Economic Perspectives, 14(4), 2348.
Fukao, K., T. Inui, H. Kawai and T. Miyagawa (2002), Sectoral Productivity
and Economic Growth in Japan, 197098: An Empirical Analysis Based on the
JIP Database, paper prepared for the NBER Thirteenth Annual East Asian
Seminar on Economics, 2022 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), 132.
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), 10915.
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), 125210.
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
29/01/2013 16:59
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),
322.
Stiroh, K.J. (2002), Information Technology and the U.S. Productivity Revival:
What Do the Industry Data Say?, American Economic Review, 92(5), 155976.
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, 5674.
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, 2425 June, Mannheim.
Vu, K. (2007). Determinants of Economic Growth in the Information Age,
paper presented at the Singapore Economic Review Conference, 24 August,
Singapore.
Young, A. (1992), A Tale of Two Cities: Factor Accumulation and Technical
Change in Hong Kong and Singapore, NBER Macroeconomics Annual, 7, 354.
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, Taiwans 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
29/01/2013 16:59
Industrial agglomeration of Taiwanese electronics firms
41
Dongguans total industrial output rose to 37.5 per cent, while the number
was just 14.3 per cent in 1990 (Dongguan Statistical Bureau, 19912010).
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 Sdillot, 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,
29/01/2013 16:59
42
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 Marshalls theory
(Marshall [1890] 1920). In Marshalls 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 Markusens 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
29/01/2013 16:59
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
29/01/2013 16:59
44
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 81km2 and
there is a small city centre of 14km2 in the northern part (see Figure3.1).
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)
29/01/2013 16:59
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
Dongguans 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
29/01/2013 16:59
46
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
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 is 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
29/01/2013 16:59
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 Sdillots study of the distribution of manufacturing activities in France (Maurel and Sdillot, 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 inputoutput relations
29/01/2013 16:59
48
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 ODonoghue and Gleave, 2004). In this chapter, due to the
lack of inputoutput 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 theenterprises 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.
29/01/2013 16:59
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 Chinas 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 Chinas 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
29/01/2013 16:59
50
4.5
Kilometres
18
USD (million)
0.0029.85
29.8676.35
76.36159.68
159.69429.20
429.21675.41
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 Changan, 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
29/01/2013 16:59
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 EGs 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
29/01/2013 16:59
52
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
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).
29/01/2013 16:59
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
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 Sdillot, 1999).
29/01/2013 16:59
54
rC 5
2 H 2 a rjw2j (1 2 Hj)
j51
12 ax
i
2
i
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 jrjwj
(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 Table3.3.
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
29/01/2013 16:59
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
0.031
0.463
0.023
0.253
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
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
29/01/2013 16:59
56
Table 3.5
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
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)
29/01/2013 16:59
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.
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.
29/01/2013 16:59
58
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 Dongguans 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, 19902009 (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
Dongguans (%) 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 (19912010); Guangdong Statistical Bureau (1990
2006); National Bureau of Statistics of China (19902006).
29/01/2013 16:59
Table 3.8
59
Industrial output of local and foreign firms in Dongguans

electronics industry, 200109
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 (19912010).
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
29/01/2013 16:59
60
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 Glaesers (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).
29/01/2013 16:59
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.
29/01/2013 16:59
62
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 Glaesers 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), 1942.
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), 63040.
Bertinelli, L. and J. Decrop (2005), Geographical Agglomeration: Ellison and
Glaesers Index Applied to the Case of Belgian Manufacturing Industry,
Regional Studies, 39 (5), 56783.
Chen, G.M. (2005), Force of Sustainable Development: The Path of Industrial
Upgrading in Dongguan, Guangzhou: Guangdong Peoples 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 (19912010), 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), 4566.
Duranton, G. and H.G. Overman (2005), Testing for Localization Using Microgeographic data, The Review of Economic Studies, 72(4), 1077106.
Ellison, G. and E.L. Glaeser (1997), Geographic Concentration in U.S.
Manufacturing Industries: A Dartboard Approach, Journal of Political
Economy, 105(5), 889927.
Fan, C.C. and A.J. Scott (2003), Industrial Agglomeration and Development:
29/01/2013 16:59
63
A Survey of Spatial Economic Issues in East Asia and a Statistical Analysis of

Chinese Regions, Economic Geography, 79(3), 295319.
Feser, E.J. and E.M. Bergman (2000) National Industry Cluster Templates: A
Framework for Applied Regional Cluster Analysis, Regional Studies, 34(1),
119.
Ge, Y. (2009), Globalization and Industry Agglomeration in China, World
Development, 37 (3), 55059.
Glaeser, E.L., H.D. Kallal, J.A. Scheinkman and A. Shleifer (1992), Growth in
Cities, Journal of Political Economy, 100(6), 112652.
Granovetter, M. (1985), Economic Action and Social Structure: The Problem of
Embeddedness, American Journal of Sociology, 91 (3), 481510.
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), 97119.
Guangdong Provincial Government (2005), Specialized Towns in Guangdong
[inChinese], Guangzhou, China: Press of Yangcheng Evening News.
Guangdong Statistical Bureau (19902006), Guangdong Statistical Yearbook,
19902006 (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),
35172.
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), 60325.
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), 92345.
Holmes, T.J. and J.J. Stevens (2002), Geographic Concentration and
Establishment Scale, Review of Economics and Statistics, 84(4), 68290.
Hsu, J.Y. (2006), The Dynamic Firm-territory Nexus of Taiwanese Informatics
Industry Investments in China, Growth and Change, 37(2), 23054.
Humphrey, J. and H. Schmitz (2002), How does insertion in global value chains
affect upgrading in industrial clusters?, Regional Studies, 36(9), 101727.
Ivarsson, I. and C.G. Alvstam (2005), Technology Transfer from TNCs to Local
Suppliers in Developing Countries: A Study of AB Volvos Truck and Bus
Plants in Brazil, China, India, and Mexico, World Development, 33(8), 132544.
Kenney, M. and R. Florida (1992), The Japanese Transplants: Production
Organization and Regional Development, Journal of the American Planning
Association, 58(1), 2149.
Kloosterman, R.C. and B. Lambregts (2001), Clustering of Economic Activities
in Polycentric Urban Regions: The Case of the Randstad, Urban Studies, 38(4),
71732.
Krugman, P. (1991), Increasing Returns and Economic Geography, Journal of
Political Economy, 99(3), 48399.
Leung, C.K. (1990), Locational Characteristics of Foreign Equity Joint Venture
Investment in China, 197985, Professional Geographer, 42(4), 40321.
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), 42640.
29/01/2013 16:59
64
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), 293313.
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), 535.
Maurel, F. and B. Sdillot (1999), A Measure of the Geographic Concentration
in French Manufacturing Industries, Regional Science and Urban Economics,
29(5), 575604.
McCann, P. and R. Mudambi (2004), The Location Behavior of the
Multinational Enterprise: Some Analytical Issues, Growth and Change, 35(4),
491524.
Ministry of Economic Affairs of Taiwan (2010), 2010 Yearbook of Chinas
Information Industry [in Chinese], Taipei, Taiwan: Ministry of Economic
Affairs.
National Bureau of Statistics of China (19902006), China Statistical Yearbook,
19902006 (various annual issues), Beijing, China: China Statistical Press.
ODonoghue, D. and B. Gleave (2004), A Note on Methods for Measuring
Industrial Agglomeration, Regional Studies, 38(4), 41927.
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,
18601987, Quarterly Journal of Economics, 110(4), 881908.
Tong, X. and J. Wang (2002), GlobalLocal 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. 6786.
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. 22342.
Wang, J. and K.N. Xu (2004), Regional Agglomeration and Spatial Evolution of
FDI, China Industrial Economy [in Chinese], 12, 617.
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), 7296.
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), 48495.
Wu, J.M. (1997), Strange Bedfellows: Dynamics of GovernmentBusiness
Relations Between Chinese Local Authorities and Taiwanese Investors, Journal
of Contemporary China, 6(15), 31946.
Yang, C. (2006), Overseas Chinese Investments in Transition: The Case of
Dongguan, Eurasian Geography and Economics, 47(5), 60421.
Yang, C. (2007), Divergent Hybrid Capitalisms in China: Hong Kong and
29/01/2013 16:59
65
Taiwanese Electronics Clusters in Dongguan, Economic Geography, 83(4),

395420.
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), 385407.
Yang, C. (2012), Restructuring the Export-oriented Industrialization in the Pearl
River Delta, China: Institutional Evolution and Emerging Tension, Applied
Geography, 32(1), 14357.
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 Areas IT Cluster, Journal of City and Planning [in Chinese],
32(3), 27599.
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), 134663.
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), 199217.
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), 487517.
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), 125.
29/01/2013 16:59
4.
The rise of the biomedical cluster in

Wonju, Korea*
Jun Koo and Jongmin Choi
INTRODUCTION
Since Porters (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
29/01/2013 16:59
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 Wonjus 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 Wonjus 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
29/01/2013 16:59
68
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 firms 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
29/01/2013 16:59
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
29/01/2013 16:59
70
(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
29/01/2013 16:59
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 citys 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 citys 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
29/01/2013 16:59
72
Table 4.1
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
managing the industryuniversity 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 citys
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
29/01/2013 16:59
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 countrys 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
29/01/2013 16:59
74
Table 4.3
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 1998May 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 1998May 2000
DongSeo Hi-Tech
Odicine
MeeRe Engineering
Biotron
Medisco
DongYang Medical
Cals Medical
Mediana
Ahone Information &
Communication
June 1998May 2000

June 1998Sep 1999
May 1998Sep 1999
July 1998Oct 1999
July 1998Oct 1999
May 1998Apr 1999

May 1998Oct 1999
May 1998Dec 1999
May 1998Aug 1999
private sector is increasingly playing an important role in the development

process.
Wonjus 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
29/01/2013 16:59
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.
Wonjus 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
29/01/2013 16:59
76
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,
29/01/2013 16:59
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.
29/01/2013 16:59
78
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,
therole 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 Wonjus biomedical cluster in the nascent stage, the citys efforts to
attract biomedical firms to the region were futile in the beginning because
most firms were concerned about the regions 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 200309 period. According to our interviewees, the status of Wonju as a hub for biomedical activities seems to
29/01/2013 16:59
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 regions 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
29/01/2013 16:59
80
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.
29/01/2013 16:59
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 countrys 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, Wonjus 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.
29/01/2013 16:59
82
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), 6986.
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), 11228.
Audretsch, D.B. (2001), The Role of Small Firms in U.S. Biotechnology Clusters,
Small Business Economics, 17(12), 315.
Audretsch, D.B. and P.E. Stephan (1996), CompanyScientist Locational Links:
The Case of Biotechnology, American Economic Review, 86(3), 64152.
Beeson, P. (1987), Total Factor Productivity Growth and Agglomeration
Economies in Manufacturing, 195973, Journal of Regional Science, 27(2),
18399.
Bresnahan, T., A. Gambardella and A. Saxenian (2001) Old Economy Inputs
for New Economy Outcomes: Cluster Formation, Industrial and Corporate
Change, 10(4), 83560.
Carlino, G.A. (1979), Increasing Returns to Scale in Metropolitan
Manufacturing, Journal of Regional Science, 19(3), 36372.
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), 11122.
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), 295319.
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), 34561.
Feldman, M.P. (2001), The Entrepreneurial Event Revisited: Firm Formation in a
Regional Context, Industrial and Corporate Change, 10(4), 86191.
Feldman, M.P. and J.L. Francis (2003), Fortune Favours the Prepared Region:
29/01/2013 16:59
83
The Case of Entrepreneurship and the Capitol Region Biotechnology Cluster,

European Planning Studies, 11(7), 76588.
Feldman, M.P. and J.L. Francis (2004), Homegrown Solutions: Fostering Cluster
Formation, Economic Development Quarterly, 18(2), 12737.
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), 12941.
Feser, E.J. (2001), A Flexible Test for Agglomeration Economies in Two US
Manufacturing Industries, Regional Science and Urban Economics, 31(1),
119.
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), 6070.
Fujita, M. and H. Ogawa (1982), Multiple Equilibrium and Structural Transition
of Non-monocentric Urban Configurations, Regional Science and Urban
Economics, 12(2), 16196.
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), 189212.
Henderson, J.V. (2003), Marshalls Scale Economies, Journal of Urban
Economics, 53(1), 128.
Kodama, T. (2008), The Role of Intermediation and Absorptive Capacity in
Facilitating UniversityIndustry Linkages: An Empirical Study of TAMA in
Japan, Research Policy, 37(8), 122440.
Koo, J. (2005), Technology Spillovers, Agglomeration, and Regional Economic
Development, Journal of Planning Literature, 20(2), 99115.
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), 66583.
Link, A.N. and J.T. Scott (2003), The Growth of Research Triangle Park, Small
Business Economics, 20(2), 16775.
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), 45575.
Orsenigo, L. (2001), The (Failed) Development of a Biotechnology Cluster: The
Case of Lombardy, Small Business Economics, 17(12), 7792.
Otsuka, K. (2006), Cluster-based Industrial Development: A View from East
Asia, Japanese Economic Review, 57(3), 36176.
Parayil, G. (2005), From Silicon Island to Biopolis of Asia: Innovation
Policy and Shifting Competitive Strategy in Singapore, California Management
Review, 47(2), 5073.
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, NovDec, 7790.
Porter, M. (2003), The Economic Performance of Regions, Regional Studies,
37(67), 5456.
29/01/2013 16:59
84
Prevezer, M. (2001), Ingredients in the Early Development of the U.S.

Biotechnology Industry, Small Business Economics, 17(12), 1729.
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), 42337.
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), 54978.
Shane, S.A. (2008), Fools 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), 21726.
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),
11839.
Waldby, C. (2009), Singapore Biopolis: Bare Life in the City-State, East Asian
Science, Technology and Society: An International Journal, 3(23), 36783.
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(12), 12341.
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), 1270916.
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), 290306.
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 200809 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 worlds 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
29/01/2013 16:59
86
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 birds-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
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.
29/01/2013 16:59
88
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. Naud (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
Naud (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.
29/01/2013 16:59
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
29/01/2013 16:59
90
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 Chinas 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).
29/01/2013 16:59
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 Chinas 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
29/01/2013 16:59
92
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 (Schller
and Schler-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
29/01/2013 16:59
93
Market orientation
Mid-tohigh
Global
++
Low-tobasic
Technological level
Domestic
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 Chinas 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.
29/01/2013 16:59
94
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
regions 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.
29/01/2013 16:59
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
29/01/2013 16:59
96
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 years 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 Chinas 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
29/01/2013 16:59
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
29/01/2013 16:59
98
Table 5.2
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
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
29/01/2013 16:59
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
Figure 5.3

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,
29/01/2013 16:59
100
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
29/01/2013 16:59
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
Resourcebased and domestic marketoriented
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 countrys development and generate
more than 40 per cent of Chinas GDP.
29/01/2013 16:59
102
Table 5.3
Basic economic characteristics of Chinas 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
29/01/2013 16:59
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 governments
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 areas 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 areas 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,
29/01/2013 16:59
104
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 Zhejiangs 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
29/01/2013 16:59
Table 5.4
105

different provinces
Average
Average
Average
Average
Average
02/0806/08 07/0806/09 07/0912/09 01/1012/10 01/1109/11
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
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 countrys 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
29/01/2013 16:59
106
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
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,
29/01/2013 16:59
107
80 000
70 000
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
thecourse 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
29/01/2013 16:59
108
140 000
Midhightech applications
Leadingedgetech applications
120 000
Lowtech 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 5.95.11, the Bohai Bay areas 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.Figures5.95.11).
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
29/01/2013 16:59
109
12 000
Yangtze River Delta
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.
SIPO raw data.
Figure 5.9
Chinese leading-edge patent applications at the SIPO in recent

years, by region
on the regions 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
29/01/2013 16:59
110
30 000
25 000
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.
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
29/01/2013 16:59
111
40 000
35 000
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.
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
29/01/2013 16:59
112
Table 5.5
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.
29/01/2013 16:59
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
29/01/2013 16:59
114
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
29/01/2013 16:59
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.
29/01/2013 16:59
116
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
29/01/2013 16:59
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 countrys 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? Chinas and
Indias Transition from Production to Innovation, World Development, 36(2),
32544.
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. 69101.
Chen, L. and P. De Lombaerde (2010), The Crisis in the U.S. and the Future of
29/01/2013 16:59
118
East Asian Production Sharing, Global Journal of Emerging Market Economies,

2(1), 91108.
Guan, J. and K. Chen (2010), Measuring the Innovation Production Process: A
Cross-region Empirical Study of Chinas High-tech Innovations, Technovation,
30(56), 34858.
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. 25881
HKCPU (2010), Hong Kongs 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 Koreas 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), 11729.
Lall, S. (1992), Technological Capabilities and Industrialization, World
Development, 20(2), 16586.
Lee, K. and C. Lim (2001), Technological Regimes, Catching-up and
Leapfrogging: Findings from the Korean Industries, Research Policy, 30(3),
45983.
Naud, 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), 2134.
Schller, M. and Y. Schler-Zhou (2009), Chinas Economic Policy in the Time
of the Global Financial Crisis: Which Way Out?, Journal of Current Chinese
Affairs, 38(3), 16581.
Simmie, J. and R. Martin (2010), The Economic Resilience of Regions: Towards
an Evolutionary Approach, Cambridge Journal of Regions, Economy and
Society, 3(1), 2743.
Stiglitz, J. (2008), China: Towards a New Model of Development, China
Economic Journal, 1(1), 3352.
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,
912 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), 111334.
Zhou, Y. (2008), Synchronizing Export Orientation with Import Substitution:
Creating Competitive Indigenous High-tech Companies in China, World
Development, 36(11), 235370.
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
29/01/2013 16:59
120
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 Marshalls ([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.
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 201012 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 ASEANChina Free
Trade Agreement phased in from 2010. Vietnam is currently experiencing
29/01/2013 16:59
122
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
(prel.)
Source: GSO (2011) and FIA (2012).
Figure 6.1
Vietnam inward FDI flows 19952011
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 Vietnams 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
29/01/2013 16:59
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.
29/01/2013 16:59
124
Table 6.1
Industrial zone developments 19912011 (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
12
65
131
267
283
300
2360
12 066
25 206
71 614
76 000
155
743
2120
3980
4113
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 Vietnams 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
29/01/2013 16:59
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
29/01/2013 16:59
126
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 19882000 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
19882000. 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
1522, 36, 37
23, 2528, 351
24, 2935 (excluding 351)
Note: See UN (1990) for a complete list of ISIC Rev. 3 codes with explanatory notes.
Source: UNIDO (2011a).
29/01/2013 16:59
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 19882000.
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
29/01/2013 16:59
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 19882000.
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
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 19952010. 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 19952010, 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
29/01/2013 16:59
130
Table 6.6
Year
1995
2000
2005
2010
Vietnam gross value of industrial output in manufacturing

19952010: technological intensity by ownership (constant
1994 VND billion and %)
Technological Intensity
Total
Low technology
Total
Low technology
Total
Low technology
Total
Low 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 (19952000) and ISIC
Rev. 4 converted to ISIC Rev.3 (200510). 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 countrys 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).
29/01/2013 16:59
Table 6.7
Year
131
Manufacturing value-added in Vietnam 19952010

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 Vietnams 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 VIETNAMS

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 Vietnams foreign trade is examined here.
Vietnams 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
29/01/2013 16:59
132
Table 6.8
Vietnam manufacturing trade 19962010, 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 (%)
Unclassified
Exports (annual average)

19962000
200105
200610
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)

19962000
200105
200610
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 (19962009) 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 19962010 is shown in
Table6.8. The technological content of Vietnams foreign trade is based
on UNCTADSTATs (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
29/01/2013 16:59
133
43.7 per cent in 19962000 to 56.6 per cent in 20062010. Though still
of the utmost importance for Vietnams 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 19962000 to 18.2 per cent in 20062010.
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 Vietnams
total manufactured imports were sourced from China, including similar
proportions of goods of medium and high skill and technological intensity.
Vietnams 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 19952010,
reaching an average of USD 10.0 billion during the period 200610.
29/01/2013 16:59
134
Table 6.9
Vietnam manufacturing trade balance 19962010, by skill and

technological intensity (annual average, USD billion)
Year
Total trade balance
Manufactured goods trade balance
Trade balance by degree of manufacturing:
19962000
200105
200610
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 200610.
In sum, Vietnams 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 Vietnams 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 Vietnams 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
29/01/2013 16:59
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 governments 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).
29/01/2013 16:59
136
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 governments 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 Vietnams 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
29/01/2013 16:59
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
29/01/2013 16:59
138
governments 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 Vietnams
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 199798. 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), 55370.
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,
29/01/2013 16:59
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. 17896.
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), 183201.
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. 27110.
MPI (2009), Vietnams IPs, EPZs and EZs Ideal Places for Manufacturing
Base. A Guide to Investing in Vietnams 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 19912011 [Conference Proceedings: 20 Years
of Developing Industrial Zones, Export Processing Zones, and Economic Zones
in Vietnam 19912011], Hanoi: Ministry of Planning and Investment [in
Vietnamese].
MUTRAP (2009), Analyzing Viet Nams 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
19882000 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 19872007 Reviewing and Looking Forward, Hanoi:
Knowledge Publishing House, pp. 12936.
OECD (2009), Policy Framework for Investment Assessment of Vietnam, Geneva:
OECD.
29/01/2013 16:59
140
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 20102012, 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.
29/01/2013 16:59
7.
The aircraft industry as a tool

for economic and industrial
development the case of Indonesia
Sren 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
29/01/2013 16:59
142
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
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 todays 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 worlds large aerospace companies are
located in advanced economies, but to an increasing degree in developing and newly industrializing economies as well (Eriksson, 1995; Vrtesy,
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 worlds 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
29/01/2013 16:59
144
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 Suhartos 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, Indonesias 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 Indonesias aircraft manufacturing industry has had difficulties
succeeding in the long term and to identify the main factors behind this
failure.
AN OVERVIEW OF INDONESIAS ECONOMIC

POLICIES
Following independence, Indonesias economic policy was shaped by
a strong sense of nationalism flavoured with anti-colonial sentiment.
Indonesias 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. Indonesias 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, OPECs 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 Banks (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, Indonesias per capita income
placed it among the lower middle income economies, but the nation was
severely hit by the Asian financial crisis in 199798. 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
29/01/2013 16:59
145
(199899). During Suhartos 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
Indonesias 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
Indonesias present export of natural resources and textiles, clothing and
footwear only had a limited lifespan, Habibie saw Indonesias 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).
Habibies 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 Habibies 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).
29/01/2013 16:59
146
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 countrys
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.
Indonesias 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 Indonesias 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-Blkow-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
29/01/2013 16:59
147
Habibie at HFB. Even before the arrival of engineers from Indonesia

at HFB, Habibie had recruited several Indonesians living in Europe.
Through Habibies 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
Indonesias 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 Companys Long-term Goals

Apart from developing and building aircraft, IPTN led Indonesias 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.
29/01/2013 16:59
148
International
utilization of business
Increasing company assets
Financial profit
Employee prosperity
IPTNs role
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 (Habibies

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;
togenerate 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.
IPTNs 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.
29/01/2013 16:59
149
Indonesias 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 Indonesias 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, 1925 June 1996).
The Influence of Habibie
Indonesias 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 IMFs intervention in 1998 and after he lost the presidential elections in 1999 and was
succeeded by Abdurrahman Wahid. Some features of Habibies 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 Habibies direct access to the
President. (Hill, 1998, p. 44)
Another feature of Habibies 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).
Habibies model of development rejected orthodox economic
approacheswith 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. Habibies approach
largely depends on guiding principles such as a strong education base and
29/01/2013 16:59
150
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.
Habibies 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 nations 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.
Its not a matter of decision, its not a matter of capital, its 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 IPTNs 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 IPTNs
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).
29/01/2013 16:59
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
Arospatiale 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.
29/01/2013 16:59
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
412) 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.)
NSA-330 and NAS-332 (Cooperation with Aropatiale,
France)
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

100
153
CASA
IPTN
90
Work load (%)
80
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

Americanhelicopter, the Bell 412. In September 1995, the production of
this helicopter was halted after a joint audit of IPTNs Bandung plant
by the US Federal Aviation Administration (FAA) and Indonesias
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, 612 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)
IPTNs 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
29/01/2013 16:59
154
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 CASAs (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,
29/01/2013 16:59
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,
17 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
Fatarudins 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 (IPTNs 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 5054seater N-250 regional turbo-prop. This was IPTNs first indigenously
developed aircraft. In June 1993, Habibie abandoned the 50-seater design
in favour of a stretched version, the 6468-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 countrys reforestation budget (Flight International,
1319 July 1994, p. 20). This sum was then equivalent to almost half of the
funds spent on the project (Borsuk, 1995).
29/01/2013 16:59
156
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 worlds dominating aircraft manufacturing country, it lacked a manufacturer in the 6468-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 IPTNs 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, 28 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, 1319 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 latters 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 Indonesias domestic airlines, Merpati and Bouraq, for 127
aircraft (Flight International, 2228 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.
29/01/2013 16:59
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, 915 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, 814 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, 1925 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 Republics 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 August4 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
29/01/2013 16:59
158
130-seater, the N-2130. In an interview, Habibie claimed that the ultimate

aim was to become the Toyota of Aerospace (Flight International, 2228
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 Presidents 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, 511 March 1997, p. 9).
A part of Habibies 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 Bachelors and Masters 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 IPTNs 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.
29/01/2013 16:59
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. 1678). 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
29/01/2013 16:59
160
(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 Indonesias 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.
29/01/2013 16:59
161
A main factor for the long-term failure is the way to look at the companys 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 Europes 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,
Its not a matter of capital, its 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 dtre,
apart from the political motivation of national prestige. The economic
and political crises of the late 1990s worsened the situation for IPTN/
29/01/2013 16:59
162
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), Habibies Grand Design, Flight International, 1925 February,
5155.
Borsuk, R. (1995), Suharto Pushes Plan to Build Jet in Indonesia, Asian Wall
Street Journal, 1112 August.
Davidson, B. (1981), PT Nurtanio: The Rising Star of the Orient, Interavia No.
12, 12369.
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, 48
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) 65364.
Eriksson, S. (2006), Cluster Creation and Innovation Within an Emerging
29/01/2013 16:59
163
Taiwanese High-tech Sector, International Journal of Technology Transfer and

Commercialisation, 5(3), 20836.
Eriksson, S. (2010), Chinas Aircraft Industry: Collaboration and Technology
Transfer The Case of Airbus, International Journal of Technology Transfer
and Commercialisation, 9(4), 30625.
Far Eastern Economic Review (1987), 11 June.
Far Eastern Economic Review (1993), 26 August.
Flight International (1992), 2228 January.
Flight International (1994), 28 March.
Flight International (1994), 1319 July.
Flight International (1995), 17 November.
Flight International (1995), 612 December.
Flight International (1995), 1319 December.
Flight International (1996), 1925 June.
Flight International (1996), 915 October.
Flight International (1997), 814 January.
Flight International (2000), 29 August4 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. 6784.
Fromhold-Eisebith, M. and G. Eisebith (2002), The Indonesian Technology
Region of Bandung, International Development Planning Review, 24(1),
4157.
Hill, H. (1998), Introduction, in Hal Hill and Thee Kian Wie (eds), Indonesias
Technological Challenge, Singapore and Canberra: Institute of Southeast Asian
Studies and Australian National University, pp. 152.
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), 3966.
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), 363400.
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. 322.
Pavitt, K. (1990), What We Know about the Strategic Management of
Technology, California Management Review, 32 (3), 1726.
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), Indonesias Technological
29/01/2013 16:59
164
Challenge, Canberra and Singapore: Australian National University and

Institute of Southeast Asian Studies, pp. 18598.
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), 13349.
Sen, A. (1979), Followers Strategy for Technological Development, The
Developing Economies, 17(4), 50628.
Simanjuntak, P. (1988), IPTN: Carrying Indonesia Toward High Technology,
Jakarta: CV Panggal.
Smith, S.L. (1998), Batam Island and Indonesias High-technology Strategy,
in Hal Hill and Thee Kian Wie (eds), Indonesias Technological Challenge,
Singapore and Canberra: Institute of Southeast Asian Studies and Australian
National University, pp. 34263.
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), 38895.
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), 6093.
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.
Vrtesy, 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, 3644.
World Bank (1993), The East Asian Miracle: Economic Growth and Public Policy,
New York: Oxford University Press.
29/01/2013 16:59
8.
Foreign knowledge transfer in the

development of aircraft industry
clusters the case of Chengdu,
China
Sren Eriksson
INTRODUCTION
The Peoples Republic of China came into being in 1949 and after nearly
three decades of self-reliance, Deng Xiaoping emerged as leader following
Maos death in 1976. It was under Dengs 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.
Chinas 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
29/01/2013 16:59
166
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
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. Taiwans industry did lack experience from the commercial aircraft industry and industries with this kind of characteristics.
Taiwans 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, Taiwans 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
29/01/2013 16:59
168
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 Chengdus 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 Chinas 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.
CHINAS 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
Soviets 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 Zedongs 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).
29/01/2013 16:59
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 Chinas economy, particularly its
tourism sector.
The seventh Five-Year Plan, which covered the period 198690, singled
out improvements to transport, particularly air transport, as a national
priority. This fact naturally stimulated the nations aircraft industry. The
establishment of a satisfactory air transport system was, and still is, vital
for the development of Chinas economy, and apart from Chinas 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 worlds 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 worlds 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.
29/01/2013 16:59
170
The Aircraft Industry and Open-door Policy

The open-door policy that started under Deng Xiaopings 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).
Chinas 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 Chinas military aircraft production), while AVIC II gave priority to
smaller aircraft and helicopters. In October 2008, AVIC I and AVIC II
29/01/2013 16:59
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 Chinas 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 nations 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 Chinas 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).
29/01/2013 16:59
172
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 Ukraines 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 Airbuss 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
29/01/2013 16:59
into service in 2016. This aircraft forms Chinas long-term goal to break
Airbus and Boeings duopoly in the medium-sized jet aircraft market.
Chinas aerospace industry has advanced over recent few decades.
While some of this progress can be attributed to rapidly growing governmental support for Chinas aerospace sector, aerospace capabilities have
largely benefited from the increasing participation of its aircraft industry
in the international supply chains of the worlds leading aerospace firms
(Eriksson, 1995, 2010, 2011a).
A study of Chinas aircraft industry, based on inputoutput data, concluded that Chinas 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.
29/01/2013 16:59
174
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 SinoSoviet 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 (196676), Chinas 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,
Chengdus 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 SinoWestern 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).
29/01/2013 16:59
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 Russias Mikoyan Aero-Science Production Group provided assistance in design as well as installing the engine systems (Flight
International, 2430 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 (28 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 Janes that Chinas Chengdu
J-10 fighter aircraft benefited from significant, direct input from Israels
Lavi programme including access to the Israel Aircraft Industries (IAI)
Lavi aircraft itself. In a number of interviews Janes 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
29/01/2013 16:59
176
modelling, wind-tunnel testing and advanced aerodynamic design input. Senior

Russian engineers who spoke to Janes recalled their many visits to Chengdu,
and elsewhere in China, some of which began in the 1980s. Janes 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, Chinas 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 latters 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
29/01/2013 16:59
(information from US aerospace engineers at Asian Aerospace 2007

Congress, Hong Kong, 45 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 CACs 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
29/01/2013 16:59
178
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. Russias 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 governments 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 CACs
period as a subcontractor to the US company. This transfer was also
29/01/2013 16:59
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 Chengdus 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 Chinas 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 inputoutput data and various kinds of qualitative studies to
look at the real activities taking place.
29/01/2013 16:59
180
NOTES
1. For an overview of the production figures and locations of the aircraft manufacturing
industries until the mid-1990s, see Eriksson (1995, pp. 12543).
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,
18April.
Bailey, J. (1993), Chasing the Genie, Flight International, 2329 June, 5660.
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), 3156.
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),
283305.
Breschi, S. and F. Lissoni (2001), Knowledge Spillovers and Local Innovation
Systems: A Critical Survey, Industrial and Corporate Change, 10(4), 9751005.
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), 47180.
De Bresson, C. (1989), Breeding Innovation Clusters: A Source of Dynamic
Development, World Development, 17(1), 116.
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), 65364.
Eriksson, S. (2003), Technology Policies in East and Southeast Asia: Creating a New
Global Structure in the Aircraft Industry, Asian Geographer, 22(12), 16985.
Eriksson, S. (2006), Cluster Creation and Innovation Within an Emerging
Taiwanese High-tech Sector, International Journal of Technology Transfer and
Commercialisation, 5(3), 20836.
Eriksson, S. (2010), Chinas Aircraft Industry: Collaboration and Technology
Transfer The Case of Airbus, International Journal of Technology Transfer
and Commercialisation, 9(4), 30625.
29/01/2013 16:59

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), 7083.
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
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 (2005), 2430 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 Janes,
5 October, accessed 16 November 2012 at http://www.janes.com/products/janes/
defence-security-report.aspx?id51065926403.
Jackson, P. (2010), Janes All the Worlds Aircraft 20102011, Coulsdon: IHS
Janes.
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), 41636.
Niosi, J. and M. Zhegu (2005), Aerospace Clusters: Local or Global Knowledge
Spillovers?, Industry and Innovation, 12(1), 529.
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), 521.
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 Martins Press.
Swire Group (2012), accessed 15 February 2012 at http://www.swire.com/eng/
activities/aviation_company.php?company5TaikooSichuanAircraftEngineering
ServicesCompanyLimited.
29/01/2013 16:59
29/01/2013 16:59
Index
agglomerations
importance of industrial 120121,
137
regional 1, 2, 6, 14
theoretical aspects 1378
see also industrial agglomeration
study
Airbus 159, 1723, 176, 177
aircraft industry
characteristics 143
China 16873
Chengdu 17380
government support for 166
as complex product system 1412
importance of external linkages 179
Indonesian investment in 144
as innovation-intensive industry 142
internationally competitive 1434
IPTN (Industri Pesawat Terbang
Nusantara) 1467, 160162
development of new technology
1557
implementation 151, 152
influence of Habibie 14950
Joint Venture Programme 1535
large-scale R&D programme
1578
licence programme 151, 153
long-term goals 1479
subcontracting and other activities
15960
Taiwanese 1678
technological requirements 143
Amsden, A. 7, 9, 15
anchor firms 12, 137
Anwar, S. 120, 136
ARJ21 regional aircraft 1712, 1767
ASEAN countries 121, 1279, 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, 3031, 347
Aviation Industry Corporation of
China (AVIC) 170171, 172, 173
Bailey, J. 150, 169
Bandung, Indonesia 146, 1535,
15960
Belgium 47, 57
bio-ventures 73, 78
biomedical clusters 667, 8082
drivers of cluster formation
intermediary organizations 767
local government 756, 81
star ventures 7880
universities 735
theory and practice of industry
cluster 6770
Wonju, South Korea 7073
Biopolis, Singapore 70
Bluegrass Auto Manufacturers
Association 68
Boeing 159, 169, 1723, 176
Bohai Bay, China 1023, 1079, 116,
117
business incubation centres 71, 745,
78
C919 narrow-body jet-airliner 1723
capability-building
strategies 11, 16
technological 89, 90, 142
CASA, Spain 1514, 159, 160, 161
catch-up growth 12, 7, 8, 11, 13, 879
Chengdu Aircraft Industries Group
(CAC) 1734, 175, 1767
Chengdu, China 1738
China
aircraft industry 17780
Chengdu 1738
183
29/01/2013 16:59
184
commercial aircraft production

166
early years 1689
and open-door policy 165, 169,
170171
in twenty-first century 1713
concentration of industrial activities
456, 47
Cultural Revolution 169, 174
economic and industrial
development 165
imports to Vietnam 1334
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 7880
cluster policies 12
and clusters 24
East Asian 1116
industrial 137
and role of the state 46
clusters 26
aircraft industry 1668, 173, 179
biomedical, in South Korea 6682
in coastal regions 40, 43
components of East Asian 12
established versus emerging 67
formation of 3, 7380
importance of development in
Vietnam 137
industrial, and FDI-induced transfer
of technology 120121
industrial, theory and practice
6770
study
CN-235 aircraft 148, 1535, 158, 159,
160
co-agglomeration effects 536
coefficient of variance (CV) index 46
competition
amongst electronics firms 49
balance with cooperation 3
for low-cost labour 136
and Porterian cluster approach 56

for use of scarce land 125
competitive advantage of nations 5,
66
competitiveness 23
in China 923
international 10, 11, 16
and local labour market 68
technological improvement 142
threats to 59
in Wonju 73
coordinators 756
culture, local, and cluster formation 3
Dassault, France 177, 178
developmental state concept 7, 810,
15
dissimilarity index 46
Dongguan, China 110
study
Dunning, J.H. 423
East Asia
East Asian Miracle 67
entrepreneurial states 611
performance of economies 70
Renaissance cluster policies 1116
Ebner, A. 4, 9, 10, 11, 15
electronics industry
firm-level survey in Guangdong
10914, 11617
Taiwanese, in Dongguan 4062
Ellison and Glaeser (EG) index 41,
468, 513, 567, 6061
embedded autonomy 7
Embraer 143, 172
entrepreneurial states 1, 611
entrepreneurs
biotech 70
confidence in China 956
innovative 69, 735
and management know-how 77
Eriksson, S. 141, 143, 162, 163, 16673,
176, 180
ERJ-145 regional jets 172
ESI Group, France 177, 178
Europe
investment in technological activities
in manufacturing FDI 1279
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 11920, 1213, 1348
industrial clusters and FDI-induced
transfer of technology 120121
industrial zone developments 1235
manufacturing gross output and
value-added 12931
technological content of foreign
trade 1314
Francis, J.L. 69, 70, 78
general purpose technology (GPT) 21
Gini coefficient 456
Glaeser, E.L. 42
see also Ellison and Glaeser index
(EG) index
global economic crisis in China
determinants for development
impact on growth paths 924, 117
institutional evolution and role of
state 912
market orientation 91
resilience, regional growth and
catch-up paths 879
sectorial structural change 8990
technological capability building
90
discussion 11416
economic recovery 856, 116
hypotheses on 867, 924, 116
regional analysis
firm-level survey of electronics
industry 10914, 11617
general findings 956
impacts and recovery by industrial
sector 96101
impacts and recovery by province
and region 1015
impacts on technological
upgrading 1056, 117
patent applications 1069
Global Seed Capital 70, 82
globalization 5, 11, 42, 90
governed markets 7
185
government-business relations 1, 2, 6,
8, 12, 15
gross output, manufacturing 12931
gross value of industrial output
(GVIO) 12430, 138
growth accounting model
algorithms 224
data description
2530
labour input 30
national account data 245
growth paths 86, 879, 97100, 103,
117
Guangdong, China
Dongguans location in 45
economic characteristics 102
firm-level survey of electronics
industry 10914, 11617
growth rates in industrial valueadded 1045
output shares 58
patent applications 107, 1089
sectorial structural change 8990
successful recovery of 115, 117
Habibie, B.J. 14450, 1546, 158, 161
Hamburger Flugzeugbau (HFB)
1467
Hanoi, Vietnam 123, 125
He, C. 43, 45, 46, 53, 61, 100
Hebei, China 102, 103, 105, 115
Herfindahl-Hirschman index 467
Herfindahl index 46
Ho Chi Minh City, Vietnam 123, 125,
135
Hong Kong
enterprises with funds from 48, 59,
62
ICT and economic growth 317
289
labour input 30
Hsia, C.J. 43, 60
Huangjiang, Dongguan, China 4951,
556
Human-Tech 78
29/01/2013 16:59
186

2530
ICT (information and communication
technology) study
model description 2230
results and analysis 3036
role in economic growth 212, 367
Indonesia
aircraft industry 160162
economic policies 1446
establishment of IPTN 14660
industrial agglomeration study 4042,
6062
co-agglomeration effects 536
extent of agglomeration 513
and firm size distribution 567
formation of electronics cluster 49
home effects 43
implications for upgrading 5860
and inter-firm linkages 424
methods and data
firm-level data and interviews
489
measure of agglomeration 448
town-level analysis 44
spatial distribution of electronics
firms 4951
sub-industries 51, 536, 61
three-digit industries 48, 51, 534,
58, 61
two-digit industries 48, 512, 545,
58, 6061
industrial clusters 120121
industrial policies and state functions
910
industrial zones (IZs) 1235, 1267,
1358
industry sector recovery in China
96101
industry tiers and groupings 100, 1423
innovation 6, 911, 12, 1416, 1412
innovative entrepreneurs 69, 735
innovative milieu 3
institutional evolution and role of state
912
inter-firm linkages 41, 424, 53, 61
intermediary organizations 767, 81
IPTN (Industri Pesawat Terbang
Nusantara) see aircraft industry
Israel 175, 178
J-5 aircraft 174, 178

J-7 aircraft 1745, 178
J-10 aircraft 1756, 178
J-20 aircraft 176
Japan
auto TNCs 43
biomedical clusters 76, 82
ICT and economic growth 212,
317
256
labour input 30
as late industrializing economy 7
move to skilled-intense activities 142
as regional technology leader 1314
Jiangsu, China 102, 103, 105, 107, 115,
116
Johnson, Chalmers 7
joint venture programme 1535
Jorgenson, D.W. 21, 24
key economic areas (KEAs) 1235
Kim, L. 13, 88
Klimov Company, Russia 175, 178
knowledge, globalized 15
knowledge producers 69, 735, 81
knowledge spillovers 689
knowledge transfer 11, 1658, 179
Koo, J. 67, 68, 69, 75, 77, 78
Kroll, H. 85, 90
Kuchiki, A. 12, 138
labour productivity growth 21, 24,
3031, 347
Lall, S. 8, 88, 165
Lavi fighter 1756, 178
learning 1011, 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, 424, 53, 61
technology 176, 178
local government role 756, 81
29/01/2013 16:59
Index
locational strategies 11, 16
low technology (LT) activities
12630
Lundvall, B.-A. 5, 6
management consultants 767
manufacturing
gross output and value-added
12931
technological contents of 1269
in Vietnam 11925, 1326
study
manufacturing value-added (MVA)
130131
market orientation 89, 91, 93, 97,
11113
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) 713, 767, 81
medium-low technology (MLT)
activities 12630
Messerschmitt-Blkow-Blohm (MBB)
1467, 151, 153
Mikoyan Aero-Science Production
Group 175, 178
modular economy concept 13
N-250 turbo-prop 1558, 161
N-2130 aircraft 149, 158
Naud, W. 88
Nelson, R.R. 6, 141, 165
Nestor, C. 123, 125, 126, 136, 138
network catalysts 767
networks 34, 1316
Nguyen, P.L. 120, 136
OECD (Organisation for Economic
Co-operation and Development)
85, 89, 90, 912, 121
open-door policy 165, 169, 17071
paradigm creation 10, 15
patent applications 1069
187
Pearl River Delta (PRD) 12, 40, 49, 60,

102, 104, 107, 10913
Porter, M.E. 23, 56, 14, 66, 68, 120,
149
Porterian cluster approach 23, 56, 15
Pratt & Whitney turbofans 169, 175,
178
private domestic sector (PDS),
Vietnam 12930
Progressive Manufacturing Programme
(PMP) 151, 153, 160
provincial recovery in China 1015
proximity, geographical 3, 42, 689,
71, 120
Qingxi, China 4950, 556
R&D (research and development)
advanced 1423
large-scale programme 1578
locations 70
real GDP growth (RGDP) 234, 3034,
367
regional growth evolution 879
regional recovery in China 1015
regulatory state functions and policies
810
Research Triangle Park 66, 75, 77, 79
resilience 878, 11112
resource allocators 756
Rosenberg, N. 6, 141
Russia 1756, 178
San Jose 67, 72, 73, 77
Schiller, D. 85, 90
Schumpeterian theory 11, 142
sectorial structural change 88, 8990
Shane, S. 66, 70
Shijie, China 43, 4950, 556
Silicon Valley 66, 68, 79, 80
Simmie, J. 86, 87
Singapore 11, 12
cluster development 15, 70
2930
labour input 30
29/01/2013 16:59
188
SinoDefence 174, 176

software investment 279, 37
Solow Paradox 21, 37
South Korea 11, 14
biomedical clusters 6682
ICT and economic growth 22, 317
259
labour input 30
transformation into biomedical
R&D hubs 70
star ventures 67, 7880, 81
state-owned enterprises (SOEs)
12930, 139
state, role of 46, 89, 912
statistical standards 24
Stiglitz, J. 85, 91
subcontracting 15960, 1767
Suharto, President 1445, 147, 148, 160
Sunley, P. 2, 42
Taiwan 11, 12, 1415
inter-firm linkages 41
PC manufacturing output 40
relations with China 168
rising production costs 49
Taiwanese electronics firms see
industrial agglomeration study
Tangxia, China 4950
Techno-Park Development Project 71,
74
technological capability building 89,
90, 142
technological content
of foreign trade 1314
of manufacturing 1269
technological innovation 1415
technological learning 7, 42, 142
technological upgrading 8, 1057, 109,
117
technology
commercialization of 910
and Habibie 145, 147, 14950, 160
levels of 12630
and manufacturing FDI 1269
in modern aircraft 1434

new, development of (phase in
establishment of IPTN) 1557
start-ups 756
transfer 120121, 136, 151, 160161,
1769
see also ICT (information and
communication technology)
study
total factor productivity (TFP) 22, 23
Toyota Corporation 68
transformative capacity 8
transnational corporations (TNCs)
424, 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, 735, 81
USA
airline trips 169
and Chinese aircraft project 175, 178
contractors to China 171
investment in biomedical start-ups
66
and Vietnamese exports 133, 134
value-added
growth, development 96, 98101,
1045
manufacturing 126, 12931
Vietnam
challenges faced by 137
economic reform programme 119
exports 1323, 1356
FDI-driven economic development
1345, 136
FDI overview 1213
impact of FDI 12931
imports 132, 133, 1356
industrial clusters 120121, 135,
1378
29/01/2013 16:59
Index
industrial zones (IZs) 1235, 135,
137
manufacturing sector 1367
technological content of foreign
trade 1314
technological intensity 132
technological transfer 120121, 136
trade balance 1334
trade with China 1334
Wade, Robert 5, 7
Weber, A. 67
Wei, Y.H.D. 41, 59
Wong, P.K. 8, 88
189
Wonju, South Korea 67, 7079, 81

World Bank 68, 1113, 100101,
130131
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, 1035, 107, 115,
116, 117
Zhou, Y. 91, 102
29/01/2013 16:59
29/01/2013 16:59
29/01/2013 16:59

Clusters and Economic Growth in Asia - PDF - Yeal

Uploaded by

Copyright:

Available Formats

Clusters and Economic Growth in Asia - PDF - Yeal

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Clusters and Economic Growth in Asia - PDF - Yeal

Uploaded by

Copyright:

Available Formats

ERIKSSON 9780857930088 PRINT.indd i

ERIKSSON 9780857930088 PRINT.indd ii

ERIKSSON 9780857930088 PRINT.indd iii

Sren Eriksson 2013

A catalogue record for this book

ISBN 978 0 85793 008 8

Typeset by Servis Filmsetting Ltd, Stockport, Cheshire

ERIKSSON 9780857930088 PRINT.indd iv

Cluster policies and entrepreneurial states in East Asia

ERIKSSON 9780857930088 PRINT.indd v

ERIKSSON 9780857930088 PRINT.indd vi

taught at Cleveland State University. He has diverse research interests

ERIKSSON 9780857930088 PRINT.indd vii